5SS^^53.522>SS5S2
Northeastern University
School of Law
Library
77V^-v^^vWVv\^^^vVx^-W^V^
INVESTIGATION OF CONCENTRATION
OF ECONOMIC POWER
HEARINGS
BEFORE THE
TEMPORARY NATIONAL ECONOMIC COMMITTEE
CONGRESS OF THE UNITED STATES
SEVENTY-SIXTH CONGRESS
THIRD SESSION
PURSUANT TO
Public Resolution No. 113
(Seventy-fifth Congress)
AUTHORIZING AND DIRECTING A SELECT COMMITTEE 10
MAKE A FULL AND COMPLETE STUDY AND INVESTIGA-
TION WITH RESPECT TO THE CONCENTRATION OF
ECONOMIC POWER IN, AND FINANCIAL CONTROL
OVER, PRODUCTION AND DISTRIBUTION
OF GOODS AND "services
PART 26
IRON AND STEEL INDUSTRY
UNITED STATES STEEL CORPORATION STUDIES
PRICES AND COSTS
JANUARY 23, 24, AND 25, 1940
Printed for the use of the Temporary National Economic Committee
UNITED STATES
GOVERNMENT PRINTING OFFICE
WASHINGTON : 1940
RTHEASTERN UNiVF.RSiTY SCHOOlof LAWDB'
TEMPORARY NATIONAL ECONOMIC COMMITTEE
(Created pursuant to Public Res. 113, 75th Cong.)
JOSEPH C. O'MAHONEY, Senator from Wyoming, Chairman
HATTON W. SUMNERS, Representative from Texas, Vice Chairman
WILLIAM H. KING, Senator from Utah
CLYDE WILLIAMS, Representative from Missouri
B. CARROLL REECE, Representative from Tennessee
THURMAN W. ARNOLD, Assistant Attorney General
•WENDELL BERGE, Special Assistant to the AttorD«s:'<-*eneral
Representing the Department of Justice
JEROME N. FRANK. Chairman
•LEON HENDERSON, Commissioner
Representing the Securities and Exchange Commission
GARLAND S. FERGUSON, Commissioner • ^-*-^
•EWIN L. DAVIS, Commissioner
Representing the Federal Trade Commission -_*
ISADOR LUBIN, Commissioner of Labor Statistics CD
•A. FORD HINRICHS,. Chief Economist, Bureau of Labor Statistics (jQ
Representing the Department of Labor
JOSEPH J. O'CONNELL, JR., Special Assistant to the General Counsel
Representing the Department of the Treasury
SUMNER T. PIKE, Business Adviser to the Secretary
Representing the Department of Commerce
JAMES R. BRACKETT, Executive Secretary
CZi
•Alternates.
REPRINTED
BY
WILLIAM S HEIN & CO , INC
BUFFALO. N. Y.
1968
CONTENTS
Testimony of — Page
Appert, Richard H., attorney, Rutherford, N. J 13650
Bean, Dr. Louis, Economic Adviser, Department of Agriculture,
Washington, D. C 13719-13732
deChazeau, Dr. Melvin G., professor at the University of Virginia,
Charlottesville, Va 13617-13648, 13671-13675
Ezekiel, Dr. MordecaijEconomic Adviser to the Secretary, Depart-
ment of Agriculture, Washington, D. C 13676-13694
Fairless, Benjamin F., president, United States Steel Corporation,
New York City .. 13585-13586
Lewis, Harold Gregg, instructor in economics. University of Chicago,
Chicago, lU 13650, 13738
Taitel, Martin, Senior Consulting Economist, Work Projects Adminis-
. tration, Washington, D. C 13694-13709
Yntenta, Theodore Otte, professor of statistics. University of Chicago,
Chicago, lU 13587-13616, 13650-13671, 13710, 13718, 13732-13741
Summary of United States Steel Corporation studies 1 3587
Effects of price reductions 13597
Cash costs 13600
The Corporation's analysis of cost in relation to volume 13617
Significance of the cost study in pricing policy 13625
United States Steel Corporation's analysis of demand for steel 13632
Significance of concept of demand for industrial price policy 13634
IN MEMORIAM SENATOR WILLIAM E. BORAH 13648
Discussion of United States Steel Corporation studies 13650
Relationship between prices, demand, and costs 13655
Analysis of operating costs 13666
Analysis of Dr. Yntema's statement concerning prices, volume, costs and
profits 13677
Failure of high prices to promote business 13686
Question of concerted action to expand production 13688
United States Steel Corporation analysis of costs in relation to price de-
cision-making 13695
.Examination of United States Steel Corporation analyses 13720
Schedule and summary of exhibits v
Tuesday, January 23, 1940---. 13585
Wednesday, January 24, 1940 13649
Thursday, January 25, 1940 13719
Appendix , 13743
Supplemental data - 14127
Index I
in
SCHEDULE OF EXHIBITS
Number and summary of exhibits
1409. Book of charts and tables, submitted by U. S. Steel Cor-
poration, as follows: financial, costs, prices, capacity
and production, labor, miscellaneous
1410. Pamphlet, submitted by U. S. Steel Corporation, entitled
"Some Factors in the Prici;ig of Steel"
1411. Pamphlet, submitted by U. S. Steel Corporation, entitled
"A Statistical Analysis of the Demand for Steel,
1919-1938" .
1412. Pamphlet, submitted by U. S. Steel Corporation, entitled
"An Analysis of Changes in the Demand for Steel and
in Steel Prices, 193fr-1939"
1413. Pamphlet, submitted by U. S. Steel Corporation, entitled
"An Analysis of the Demand for Steel in the Automo-
bile Industry"
1414. Pamphlet, submitted bj^ U. S. Steel Corporation, entitled
"An Analysis of the Demand for Steel in the Railroad
Industry"
1415. Pamphlet, submitted by U. S. Steel Corporation, entitled
"An Analysis of the Demand for Steel in the Container
Industry"
1416. Pamphlet, submitted by U. S. Steel Corporation, entitled
"An Analysis of Steel Prices, Volume and Costs — Con-
trolling Limitations on Price Reductions"
1417. Pamphlet, submitted by U. S. Steel Corporation, entitled
"An Analysis of Steel Prices, Volume and Costs Con-
trolling Limitations on Price Reductions"
1418. Appears in Hearings, Part 27.
2180. Pamphlet, submitted by U. S. Steel Corporation, entitled
"The Distribution of Steel to Major Consuming In-
dustries"
2181. Pamphlet, submitted by U. S. Steel Corporation, entitled
"Indexes of Mill-Net Yields on Products Shipped by
United States Steel Corporation Subsidiaries"
2182. Pamphlet, submitted by U. S. Steel Corporation, entitled
"Improved Quality of Steel as a Price Reduction"
2183. Chart: Relation of industrial production, excluding iroo
and steel, to steel sales
2184. Chart: Contrast in production-profit computations
2185. Table: Reconciliation of total costs before bond interest
and inter-company items in "Analysis" and registra-
tion statement, 1935-37 — ^U. S. Steel Corporation.
Table: Comparison of break -down of lumped costs in the
"Analysis" and in registration statement, 1935-37- —
U. S. Steel Corporation.
Table: Additions to reserves charged to cost of goods
sold, etc., 1935-37— U. S. Steel Corporation.
Table: Taxes other than Federal income and Social Secu-
rity taxes, 1927-38— U. S. Steel Corporation.
Table: Taxes other than Federal income and Social Secu-
rity taxes, 1927-38 — Recomputed "fixed" and "vari-
able" costs — U. S. Steel Corporation.
Table: Maintenance and repairs, '1927-38 — U. S. Steel
Corporation.
Table: Stripping and development expenses, 1927-38 —
U. S. St^el Corporation
Intro-
duced at
page—
(')
(0
(')
(')
(')
(')
n)
0)
(')
13586
13586
13586
13680
13695
Appears
13743
13893
13913
13942
13981
13999
14016
14032
14082
14095
14101
14109
14119
14120
13701 I 14121
See Hearings, Part I
See also p. 13.585, infra.
VI
CONTENTS
SCHEDULE OF EXHIBITS— Continued
Number and sumn^ary of exhibits
Intro-
duced at
page-
Appears
on page—
2186. Chart: Price of tin plate and of canned goods, 1923-1938.
2187 Chart" The net regression of volume on price
13726
13729
13731
14124
14125
2188. Chart: Indexes of wholesale prices of iron and steel and
other goods, 1919-1938
14126
SUPPLEMENTAL DATA
Unnumbered. Telegrams, dated October 10, 1939, from Harry
Moreland, vice president. Great Lakes Pipe
Line Co., to James Brackett, secretary of the
Committee, authenticating certain statements
made by Mr. Eugene Orvis in his prepared
statement, admitted to the record as "Exhibit
No. 1293" and included in Hearings, Part 16,
appendix, p. 9330
14127
INVESTIGATION OF CONCENTRATION OF ECONOMIC POWER
TUESDAY, JANUARY 23, 1940
United States Senate,
Temporary National Economic Committee,
Washington, D. C.
The committee met at 10:35 a. m., pursuant to call of the chair-
man, in the Caucus Room, Senate Office Building, Senator William
H. King, Utah, presiding.
Present: Senator lOng (acting chairman); Messrs. Hiarichs,
O'Connell, and Brackett.
Present also: Willis Ballinger and Walter B. Wooden, represent-
ing the Federal Trade Commission; John V. W. Reynders, represent-
ing the Department of Commerce; A. H. Feller, representing the
Department of Justice; Dr. Theodore J. Kreps, economic adviser to
the committee.
Acting Chairman King. Dr. Kreps, are you ready to proceed?
Dr. Kreps. Yes, sir.
Acting Chairman King. Mr. Fairless, come forward, please. You
were sworn here before as a witness. You may proceed, Dr. Kreps.
TESTIMONY OF BENJAMIN F. FAIRLESS, PRESIDENT, UNITED
STATES STEEL CORPORATION, NEW YORK CITY
Dr. Kreps. Mr. Fairless, will you make your statement, please?
Mr. Fairless. Mr. Chairman, at the hearing held on November 8,
1939, I submitted to the committee certain charts and pamphlets
which were marked "Exhibits Nos. 1409 to 1418," inclusive.^ I am
informed that these exhibits may now form a part of the record of
the committee, and, accordingly, I offer these exhibits for that pur-
pose, together with 3 additional pamphlets and 13 additional charts,
which have been, heretofore given to Dr. Kreps.
("Exhibits Nos. 1409 to 1417," inclusive, are included in the ap-
pendix on pp. 13743-14082. "Exhibit No. 1418" appears in Hearings,
Part 27, appendix, p. 14:i9.)
These additional pamphlets are entitled "The Distribution of Steel
to Major Consuming Industries," "Indexes of Mill-Net Yields on
Products Shipped by United States Steel Corporation Subsidiaries,"
and "Improved Quality of Steel as a Price Reduction." The 13
additional charts should be included in the book of charts, designated
as "Exhibit No. 1409." ^
Dr. Kreps. Mr. Chairman, I suggest that these exhibits be placed
into the record as indicated by Mr. lairless.
' See Hearings, Part 20, p. 10803.
' The 13 additional charts are included in "Exhibit No. 1409," appendix, p. 13743.
13586 CONCENTRATION OF ECONOMIC POWER
Acting Chairman King. They may be so received and placed in
the record.
(The three pamphlets referred to were marked "Exhibits Nos.
2080, 2181, and 2182," respectively, and are included in the appendix
on pp. 14095, 14101 and 14109.)
Mr. Fairless. I should like, Mr. Chairman, to say a few words
about these various pamphlets and charts which have been submitted
to the committee by the Steel Corporation. From the outset, we have
understood that it was the pm-pose of the T. N. E. C. to obtain the
material facts about the steel industry as a part of the committee's
objective inquiry into American business. The Steel Corporation has
tried to the best of its ability to cooperate with the committee to this
end. Accordingly, we organized a special T. N. E. C. group, consist-
ing of 30 or more individuals. Some were employees of the Corpora-
tion; some were economists and graduate students in economics em-
ployed by the Corporation for this purpose; and some were lawyers
assigned to this work by Governor Miller and Mr. Olds, two of our
directors, from their respective law offices. This group over a period
of more than a year and a half has conducted various studies, the re-
sults of which are contained in these papers. The work, of an economic
nature, was imder the direction of Dr. Theodore O. Yntema, of the
University of Chicago, who is here today and is prepared to explain
to the committee the various studies made mider his direction.
On behalf of the United States Steel Corporation, I should like to
express our appreciation of your admission into the record of these
various papers, which I hope wiU be of aid to the committee in its
consideration of the topics discussed therein.
Mr. Ballinger. Mr. Chairman, could the Federal Trade Commis-
sion make the request that the pamphlet on the basing-point system ^
not be released to the press until our reply to it accompanies it? ^
Acting Chairman King. What are your views about that, Dr.
Kreps?
Dr. Kreps. I think that would be fair.
Acting Chairman King. Do yoii have any objection to that, Mr.
Fairless?
Mr. Fairless. No objection.
Acting Chairman King. Granted.
Dr. Kreps. That is all. Mr. Fairless will be called back later in
the hearing.
Acting Chairman King. The committee may not be able to read
those records during this hearing, I mean today, but I think we shall
before we' conclude our labors.
Dr. Kreps. These have all been submitted to the members of the
committee as well as to the members of the staff.
I would like next to call Prof. Theodore Ottc Yntema to take the
stand.
Acting Chairman King. Doctor, will you come forward? Will you
hold up your right hand and be sworn?
Do you .solemnly swear that the testimony you will give in this
hearing shall be the truth, the whole truth, and nothing but the truth,
so help you God?
1 "Exhibit No. 1418."
' Mr. Ballinper refers to "An Analysis of the Basing Point System of pelivered Prices as Presented by
United State* Steel Corporation in *E.\hibits Nos. 1110 and 1418' by Walter B. Wooden, Assistant Chief
Counsel, and Hugh E. Wbitt , Examiner Tederal Trade Commission," admitted to the record as "Exhibit
No. 2242" and appearing in Hearings, Part 27.
CONGENTKATION OF EOONOMIC POWER 13587
Dr. Yntema. I do.
Acting Chairman King. You may proceed.
TESTIMONY OF DR. THEODORE OTTE YNTEMA, PROFESSOR OF
STATISTICS, UNIVERSITY OF CHICAGO, CHICAGO, ILL.
Dr. Keeps. Dr. Yntema, for the pm-pose of the record, will you
state your full name and address, please?
Dr. Yntema. My name is Theodore Otte Yntema. My address is
1154 East Fifty-sixth Street, Chicago, 111.
t)Ti Kreps. You are professor of statistics in the School of Business
of the University of Chicago, is that correct?
Dr. Yntema. That is correct.
'Dr. Kreps. And director of research in the Cowles Commission for
Research in Economics, an institution affihated with the University
of Chicago?
Dr. Yntema. That is correct.
Dr. 'Kreps. You are also a certified public accountant?
Dr. Yntema. That is correct,
Dr. Kreps. You have taught accounting for several years at the
University of Chicago?
Dr. Yntema. Yes.
Dr. Keeps. Since what period have you been associated with the
United States Steel Corporation as a consulting economist?
Dr. Yntema. Since July 1938.
Dr. Kreps. How much of your time have you devoted to the
studies which have been submitted here by Mr. Fairless?
Dr. Yntema. I have devoted over half of my time to the preparation
of these studies.
Dr. Kreps. Have you had assistance?
Dr. Yntema. Yes. Under my supervision I had a special research
section, consisting of economists, graduate students and others.
Dr. Keeps. A number of these had already made intensive studies
of the steel industry before they worked on your staff?
Dr. Yntema. One of these members had.
Dr. Kreps. I believe you have a brief summarj^ of your analysis
which you would like to present to the committee entitled "Factors
Affecting the Demand for Steel and the Relation of Steel Prices to
Costs"?
Dr. Yntema. That is correct.
Dr. Keeps. You may proceed.
Acting Chairman King. Proceed, Doctor.
summaey of united states steel coepoeation studies
Dr. Yntema. In summarizing these studies, it will be necessary for
me to pass over many interesting and important details, to confine
myself to a rather cursory statement of the facts and the inferences to
be drawn from them, and to omit at many points the qualifications
which would be desirable if there were time for them.
This committee has evidenced a deep interest in the relation of
steel prices to production and employment in the steel industry.
Recognizing the importance of this problem and the committee's
interest in it, the United States Steel Corporation has prepared and
13588 CONCENTRATION OF ECONOMIC POWER
submitted to the committee a number of studies dealing with this
subject. It is our hope that the members of the committee may find
them helpful in their deliberations. We welcome criticism of these
studies and hope that out of the discussion there may come a better un-
derstanding of the important problems to which they relate.
These studies do not, of course, answer all the questions relating to
price flexibility in durable goods, but they do, we believe, present
factual evidence illuminating some aspects of the problem.
The basic questions to which our studies were addressed are these:
1. To what extent will the production and sale of steel respond to
changes in the price of steel?
2. To what extent do costs vary with volume of production?
3. How far, if at all, is it feasible for the steel industry to achieve
additional sales, production and employment in depression by reduc-
tion of prices?
In other words, is it possible for the steel industry to achieve fuller
utihzation of its productive facOities and thus greater employment by
means of price reductions in periods of low demand?
An analysis of the evidence available to us leads to these conclusions:
1. The quantity of steel that can be sold is relatively unresponsive
to changes in the level of steel prices. In other words, the demand for
steel is inelastic. A reduction in the price of steel, therefore, will
bring only a small increase in its consumption. The fluctuations in
the production of steel have been due primarily to shifts in demand
caused by changes in general business activity, consumers' income
and industrial profits. In comparison with these factors, the price
of steel has been a minor influence on the quantity of steel sold.
2. The reduction in average costs resulting from increased output
is much less than the reduction in prices which is necessary to induce
such increase in output. All but a small percentage of the costs of
producing steel, in good times and bad, are out-of-pocket expenditures.
Unless wages and other costs could have been further reduced in de-
pression, a substantially lower price level for steel during the past 10
years would have brought general bankruptcy in the industry.
In view of these facts, full production and employment cannot be
maintained in the steel industry during depression by means of
reduction in steel prices.
These conclusions are based on the assumption of a reduction only
in the price of steel. It has often been pointed out that the inelas-
ticity of demand for individual durable goods does not afford an
adequate basis for demonstrating the inelasticity of demand for dur-
able goods in the aggregate. This we recognize. What would happen
to production and employment if there were greater cyclical flexibility
in the prices of all durable goods is a most difficult and perplexing
question. Although we have given the matter much study, we have
not been able to reach conclusions which we can establish beyond
reasonable doubt. We have found that others better qualified to
deal with this problem have had the same experience, and that there
is on the part of many economists honest doubt as to the efficacy of
price flexibility as a cure for depressions. There is, however, general
recognition that the existent inflexibilty in costs, particularly in wages,
taxes, and transportation charges, aU of which are subject in piM: r
or less degree to Government regulation or influence, is so great as
to preclude any considerable increase in the flexibility of the prices of
CONCENTRATION OF ECONOMIC POWER 13589
finished products. Unless the costs of producing durable goods are
flexible, it is idle to talk of flexibility in their prices.
In analyzing the demand for steel we approached the problem in
two ways. First, we undertook to study the demand for steel from
major consuming industries. On the basis of rough estimates of the
elasticity of demand for the products of these industries and the
relative proportion of steel cost to the prices of their products, it was
possible to discover approximately the effect of changes in steel prices
upon the prices and consumption of the finished products made from
steel and thus upon the consumption of steel in these industries. In
the second place, we made a statistical analysis of the. data over the
past 20 years to discover the relation of steel production to steel
prices and other factors determining demand, and thereby to ascer-
tain the relative importance of price as a factor influencing the
quantity of steel produced and sold.
During the last 15 years the automobile, railroad, and container
industries have consumed almost 40 percent of the steel produced in
this country. These industries represent three different types of
steel consmners, one using steel as a raw material in the manufacture
of a consumer's dm-able good, another using steel in the form of plant
and equipment, and the third using steel as a raw material in the
manufactm-e of a consumer's perishable good.
Acting Chairman King. Pardon me. Doctor, but have you broken
down that 40 percent to which you have just referred, to determine
what percentage of it was required by the railroads?
Dr. Yntema. I haven't the percentage figure handy. Senator,
but we have just submitted to the committee a special study, "The
Distribution of Steel to Major Consuming Industries." ^
Acting Chairman King. And that would show the amount which
was consumed by the railroads, would it?
Dr. Yntema. Yes, sir; that shows, to the best of our abiUty to make
such an estimate, the amount consumed by the railroads.
Acting Chairman King. Proceed.
Dr. Yntema. The automobile industry: The automobile industry
has been the largest single consumer of steel for 5 of the last 6 years,
taking between one-fourth and one-sixth of the total of all hot-rolled
steel products.
Acting Chairman King. They took between one-fourth and one-
sixth of the 40 percent which you have attributed to the automobile
and the railroad industries and left the residue, of about 30 percent
or more^ to the railroads?
Dr. \ NTEMA. There are three industries whose consumption com-
prises the 40 percent of the total production: The automobile, railroad,
and container industries.
Acting Chairman King. Oh, yes.
Dr. Yntema. Although it has had a long-term upward trend, auto-
mobile production has been subject to severe cyclical fluctuations. In
1929, approximately 5.6 milUon cars were produced. In 1932, pro-
duction slumped to about 1.4 milUon, only 25 percent of the 1929
production. By 1937, production had risen to approximately 5,000,-
030 cars, more than three times the volume in 1932.
The reasons for these wide fluctuations in automobile production
were carefully analyzed by C. F. Roos and Victor von Szehski in a
: "Exhibit No. 2180," appendix, p. 14095.
13590 CONCENTllATION OF ECM3NOMIC POWER
study made for General Motors Corporation. They found that the
number of new cars sold in any year was dependent on (1) the national
income; (2) the number of cars in operation; (3) the age distribution
of cars in operation; (4) the scrapping rate; (5) the price; and (6)
other factors, including used-car allowance, financing terms, operating
cost, and dealers' used-car stocks. Taking all these factors into
account, they showed that a 1 -percent reduction in the price of new
passenger cars would cause approximately a 1.5-percent increase in
the number of cars sold. They concluded that the effect of price
changes on the number of cars sold was overshadowed by the in-
fluence of changes in national income, and that the changes in price
could not level out the sharp fluctations in production of automobiles.
The railroad industry: For many years the railroad industry
ranked first as a consumer of steel. In 1926, railroads consunried
approximately 7.6 million tons of hot-rolled finished products, which
represented about 21.6 percent of that year's total production. During
the last 10 years, this industry's purchases of steel have decHned abso-
lutely and relatively.
The cyclical fluctuations in railroad purchases of steel are particu-
larly marked. In 1932, the railroads took approximately 1,000,000
tons of steel, while in 1937, a relatively good year, they purchased
4.1 million tons, still much less than their predepression consumption.
The serious plight of the railroads is common knowledge. They
have suffered both from a downward trend in operation and from the
severity of the recent depression. As a consequence, the need for
new equipment has declined and the revenues in>many cases have been
inadequate for maintenance and replacement of existing facilities •
The cyclical fluctuations in railroad traffic have been closely related
to changes in the national income and in industrial production, while
the downward trend has been due primarily to competition from
alternative means of transportation.
Acting Chairman Kino. You refer to trucks, I suppose, and water
transportation?
Dr. Yntema. Yes, trucks and water transportation; also pipe-line
transportation.
Capital expenditures for rolling stock and other equipment requiring
steel are ultimately dependent upon the demand for rail transportation
but in the short run they are determined by the pressure of current
traffic on existing facilities and by the funds available for capital
outlays. Hence, when the demand for railroad services declines, and
only a part of the rolling stock and other equipment is needed^ to
furnish all the services required, there is obviously less need for capital
expenditures.
Tluis, capital expenditures for equipment dropped from about
§328,000,000 in 1930 to about $15,000,000 in 19^3, a decline of about
95 percent. Purchases of steel for maintenance purposes-have, how-
ever, been more closely related to the volume of traffic currently
handled by the railroads, since some degree of maintenance must
continue even in depression. Nonetheless, declining traffic and
lower revenues have drastically reduced total steel purchases by the
railroads.
The container industry: Consumption of steel by the container
industry (whose principal product is the tin can, a consumers' per-
ishable good) has. shown a substantial upward trend since 1923. In
CONCENTRATION OF ECONOMIC POWER 13591
that year the container industry took 3.6 percent of the total finished
rolled steel, but since 1932 it has taken on the average more than 8
percent of the total output. In 1938, it ranked third among consum-
ing industries, accounting for 9.9 percent of the total output of steel.
While the annual average consumption of steel by the contamer indus-
try was 1.4 million gross tons during the period from 1923 to 1929, in
the period from 1932 to 1938 its annual consumption averaged about
1.9 million gross tons, which is rouglily an increase of 36 percent.
This relative stability of the container industry during depression peri-
ods is further shown by the fact that tin-plate production ranged
from about 60 to 90 percent of capacity during the depression, while
total steel production varied from about 15 percent to 60 percent of
capacity.
Although the demand for products packed in tin cans is largely de-
pendent upon consumers' income, the relatively greater stability of
tin-plate production in depression periods is due primarily to the fact
that the majority of containers made from tin plate are used for food
products. Being a perishable necessity, food must be purchased even
in depression times, whereas purchases of more durable products may
be postponed. The comparative stability of tin-plate production
arises from this fact.
The decline in consum.ption of steel by the container industry from
1929 to 1932 was further reduced by the underlying upward trend in
tin-can consumption, due largely to the increasing use of tin cans to
pack additional kinds of foods and other products.
Other investigations have shown that the demand for various agri-
cultural products is inelastic. This is to say that a given percentage
price reduction does not produce a corresponding percentage increase
in consumption. The available data also indicate that fluctuations
in the total consumption of canned food products have had little net
relation to fluctuations in canned food prices or to fluctuations in the
ratio of canned food prices to other food prices. From these facts,
it is reasonable to infer that the demand for canned food products has
a low elasticity.
Acting Chairman King. You have no figures showing the curve of
upward or downward steel consumption in building operations, have
you?
Dr. Yntema. We have that evidence in this pamphlet which we
submitted to the committee this morning, "The Distribution of Steel
to Major Consuming Industries." *
Acting Chairman King. All right; proceed,
Dr. Yntema. Other steel-consuming industries': We have not
examined in detail the demand for steel in other major steel-consuming
industries. In nearly all cases the products of these industries are
durable goods subject to great cyclical fluctuations in demand.
Many of them are producers' goods, vt^hich are utilized in the pro-
duction of other goods and services. In such cases, the cost of the
product made from steel is not usually a large proportion of the value
of the goods and services produced by the industries using these
products made from steel. Consequently, there is good reason to
believe that the demand for the products of these industries is generally
not very elastic and in many cases is inelastic.
1 "Exhibit No. 2180," appendix, p. 14095.
13592 CONCENTRATION OF E<X>NOMIC t»OWER
The relations of the cost of steel to the price of the finished product:
The demand for steel is derived from the demand for the services
rendered by steel products, or, more directly, from the demand for the
finished products themselves. A reduction in the price of steel, if
passed on, willl reduce the price of the finished product. In greater
or less degree, this will increase the consumption of the product and,
thus, the consumption of steel used in its manufacture. Furthermore,
a reduction in the price of steel may perhaps increase the use of steel
per miit of finished product. In each of these cases, however, the
critical question is, how much?
The percentage decrease in the price of a finished product made
possible by a reduction in the price of steel depends upon the propor-
tion of the cost of steel to the value of the finished product. What is
this proportion?
In the case of low-priced automobiles, the cost of steel is about 10
percent of the delivered price. This percentage would be lower for a
more expensive automobile. For a representative list of canned food
products, the cost of tin plate per can varied from 3.4 to 13.9
percent of the retail price of such food products. The cost of steel
consumed by the railroads is estimated to average only about 5
percent of the value of transportation services furnished by them.
In the construction industry, steel costs range from 4 percent of the
total cost of a frame house to as much as 30 percent of the total cost
of a steel bridge. For a modern automatic packaging machine, the
steel cost cc Tiponent was found to be less than 2 percent of the selling
price. Ext .'eme examples may be cited showing a very high or very
low ratio of the cost of steel to the price of the finished product, but
10 percent appears to be a reasonably typical proportion.
On this basis, a 10-percent reduction in the price of steel would
correspond to a 1-percent reduction in the price of the finished product
made from steel.
Since the elasticity of demand for the finished products of most
steel-consuming industries is low, probably less than 1 or 2, a 1 -percent
decrease in the price of the product would not increase the quantity
sold by more than 1 or 2 percent. If other conditions affecting demand
and costs remain the same, a 10-percent reduction in the price of
steel Would not increase the consumption of steel by more than 1 or 2
percent through its effect upon the price of the finished product.
Substitution of steel for other materials: In the industries studied,
price is generally not an important factor in the substitution of steel
for other products. The physical characteristics of steel, especially
its great tensile strength and durability in comparison with other
materials, sharply limit the possible uses of substitutes. In the case
of tin cans, there is some degree of substitutability between containers
made of tin plate and those made of glass. Even in this case however
numerous factors limit the possibility of substitution in response to
price changes.
The amount of steel used in the finished product: In the automobile
industry, there was for a number of years an increasing use of steel
per car due to the growing popularity of closed and heavier models,
and the changes in construction in the interests of safety. These
developments, however, cannot be attributed to steel prices. In
most cases, technical considerations determine, within rather rigid
limits, the qiiantity of steel employed in any particular product.
CONCENTRATION OF ECONOMIC POWER 13593
Although some shght increase in the weight of steel used per unit of
product may result from a reduction in the price of steel, this effect
is certainly not of substantial proportions.
The elasticity of the demand for steel: From the discussion thus far,
it is apparent that the quantity of steel sold is not very responsive to
changes in the level of steel prices.
Let m.e interpolate here, to point out that I am talking about the
responsiveness of the total quantity of steei sold by the industry, in
response to the changes in steel prices.
To ro.ake an estimate of the elasticity of demand for steel in the
various consum.ing industries, we must take into account, first, the
elasticity of demand for the products made from steel; second, the
proportion of steel cost to the price of the finished product; third, the
substitutability of steel for other materials; and, fourth, the possibility
of increasing the amount of steel in the finished product. In apprais-
ing these factors, we have found that the elasticity of dem.and for the
products made from steel is generally rather low, in m.ost instances,
probably not greater than 1 or 2 ; that the proportion of steel cost to
the price of the finished product is, on the average, in the neighborhood
of 10 percent.
Acting Chairman King. Say that again. I didn't get it.
Dr. Yntema. And that the proportion of steel cost to the price of
the finished product is, on the average, in the neighborhood of 10
percent.
Acting Chairman King. Going down?
Dr. Yntema. That is a very rough approximation.
Acting Chairman King. But that is going down to 10 percent?
Dr. Yntema. It ranges usually from 3 to 30 percent but 10 per-
cent is a roughly typical figure that we might take for discussion.
To continue: The possibilities of substitution of steel for other mate-
rials and of increasing the amount of steel in the finished product are
of relatively minor importance.
From the evidence, it is safe to conclude that the demand for steel
is inelastic, that is, that a given percentage reduction in price wjll not
bring about as large a percentage increase in the quantity sold. Al-
though any such estimate is subject to a wide margin of error it
seems probable that the elasticity of demand for steel is not greater
than 0.3 or 0.4, that is, that a 10 percent reduction in price would
not increase the quantity of steel sold by more than 3 or 4 percent.
In concluding this part of our discussion, it should be pointed out
that these estimates are based upon the assumption that other prices
and other factors affecting the demand for steel remain the same.
Mr. Ballinger. Dr. Yntema, how many industries did you
analyze that use steel? I mean in testing this elasticity of demand?
Dr. Yntema. We analyzed in some detail these three to which I
have referred.
Mr. Ballinger. Representing about 40 percent; the other 60 per-
cent, you didn't look into?
Dr. Yntema. Yes; we investigated briefly the construction in-
dustry and found the problems there were so complicated that we
were not prepared to submit a study of them to this committee.
Statistical analysis of the elasticity of demand for steel: The fore-
going conclusions as to the elasticity of demand were tested by a
separate statistical analysis of anrual sciies of relevant economic data
13594 CONCEKTKATION OF ECONOJMIC POWEii
from 1919 to 1938. Production, shipments and bookings were res-
pectively correlated with the factors deemed to exert a significant
influence on the quantity of steel demanded. These factors were:
(1) The price of steel.
(2) Industrial production.
(3) Consumers' income.
(4) Industrial profits.
(5) A time-trend variable.
From four such correlations involving difl"erent combinations of
these factors, it was found that a 1 percent decrease in the price of
steel (other factors remaining the same) would induce less than a 1
percent increase hi steel sales. Subsequent calculations in which
mill net yield? on shipments of steel, not available at the time of the
original study, were used in lieu of published prices, confirmed these
results. Although subject' to considerable error, the best estimate
of the elasticity of demand for steel indicated by this analysis is
thought to be approximately 0.3 or 0.4.
These findings confirm our other estimates based on the study of
the demand for steel b^^ consuming industries, and indicate that
changes in the level of steel prices cause smaller percentage changes
in the opposite direct'ion in the qu*'ntity of ;->tcei sold.
Mr. O'CoNNELL. May I nsk u question? Why did you use the
mill net yield instead of anotlicr basis for determining?
Dr. Yntema. In relating shipments to price, the appropriate price
measure to use is the price obtained on the shipment, and the mill net
yields constitute that particuhit price.
Mr. O'CoNNELL. Well, tl;e mill net yield does not represent :.he
cost to the purchaser oi steel, does it?
Dr. Yntema. No; but the mill net yield fluctuates in the cych in
very close correspondence to the price paid by the customer. If 'ou
had a price series of the actual average prices i^aid by the buyers of
steel, and then had a price series of mill net yields, the percentage
changes in both series would be almost identical.
Mr O'CoNNELL. You mean there would be a closer correlation
than there would be to make the comparison with the published prices?
Dr. Yntema. No; there are two points that I am making here.
One is that if you are studying the relation of shipments to some price,
then it must be the price paid for those shipments. The second point
that I am making is that because we did not have the actual prices
paid by customers, since the data were not available hi convenient
form, we used a series which paralleled that very closely, namely, the
average mill net yield. If you plotted those two series, you would
see that the relative fluctuations in them were almost identical. We
do not have the two series, but the evidence which will come out in a
later discussion, I think, will indicate this beyond any reasonable
doubt.
m the j)eriod covered by the analysis, changes in the level of steel
prices were a comparatively minor influence in determining changes
in the volume of steel sales. Even if fiuctuations in steel prices had
been considerably greatei* than they wore, ne\ertheles-; other factors
aflecting the demand for steel, such as consvuners' income, industrial
profits, and general business activity, were found to bo of such cou-
trollmg importance that they would still h;ive had far greater weight
than changes in steel prices.
CONCENTRATION OP EXX>NOMIC POWER 13595
Relation of cost to volume of output: Our second problem was to
determine how costs varied with output. To accomplish this, a
study was made of the. relation between cost and volume for the
United States Steel Corporation subsidiaries during the period from
1927 to 1938. In computing cost, intercompany items were excluded,
as were Federal income taxes and costs connected with extraneous
nonoperating transactions. Since the purpose of the analysis was to
ascertain the changes in costs which would result from changes in
volume of production, the effects of other factors had to be excluded.
This necessitated the adjustment of the cost figures for each of the
years to the levels of material prices, wages, interest, and tax rates, and
pension paympnts prevailing in 1938, and also an adjustment for
increases in efficiency which took place during this period. Having
removed the effects of changes resulting from these factors, the ad-
justed costs could then properly be related to volume of output.
Because of variation in the proportions of low and high cost prod-
ucts, the simple aggregate tonnage of these products was not a
satisfactory measure of output for our purpose. To eliminate the
effect of these variations in the product mix, proper weights were
assigned 'to low and high cost products and an annual weighted total
of products shipped was thus obtained.*
When the average relationship of adjusted total cost to weighted
volume was obtained, it showed that within a range of operation
from 18 percent to 90 percent of ingot capacity, the total costs of the
corporation and its subsidiaries under 1938 conditions amounted to
$55.73 per weighted ton plus $182,100,000.
It should be noted here that while the costs mentioned are exclusive
of all nonoperating income and expense, they cover all operations of
the Corporation and, hence, do not represent merely the cost of pro-
ducing steel. Furthermore, even weighted tonnages shipped do not
reflect the full volume of business, since some goods and services are
sold by the Corporation which are not measured in tons. Nevertheless
other operations rise and fall with increases and decreases in shipments
of products to a sufficient extent that the total costs maintain approxi-
mately the relationship to shipments just described. Since in 1938,
89 percent of the total revenues came from the sale of steel, presum-
ably about 89 percent of the above costs represent costs directly
related to steel production,
Mr. Reynders. May I ask a question? In other words, when
you arrive at this price, you include such items as fabricated steel,
going into buildings and ships?
Dr. Yntema. Yes. We have included all the steel products, all
the hot rolled products, shipped by the Steel Corporation as such.
Mr. Reynders. But only the rolled products? '
Dr. Yntema. I should like to correct that statement. This includes
all the products sold in the form they are sold by the Steel Corporation
subsidiaries, whatever form that may be.
^ •. Reynders. Then you Lssign, for instance, to a fabricated steel
a liigher relative tonnage rate.
Dr. Yntema. Yes; -that is correct.
Mr. Ballinger. In that range from 18 percent to 90 percent, is
this the cost, $55.73, or would it vary between that range?
' To obtain as large a 'coverage as possible, this weighted total included also products other than steel sold
on a tonnage basis by tne Corporation subsidiaries, weighted in a similar manner.
124491— 41— pt. 26 2
13596
CONC5ENTRATION OF ECONOMIC POWER
Dr. Yntema. No, sir; the cost is composed of two parts, as I shall
show in a few minutes, a total of $182,100,000 plus an additional cost
that varies with output, of $55.73 per weighted ton. I want to make
clear that that is not strictly the additional cost per weighted ton of
steel. If you could allocate these additional costs on the basis of
revenues obtained from steel and other products, i. e., if you multi-
plied thenx by 89 percent you would have a rough working estimate
of the additional cost of steel per ton.
This $182,100,000 represents the portion of the costs imder 1938
conditions which remained the same, independent of variations in
production within the above-stated capacities — provided, of course,
that other factors affecting costs stayed constant. This "fixed cost"
included not only interest and pensions, but also the portions of all
other costs which did not vary with output. The $55.73 per weighted
ton represents the additional cost of all operations per additional ton
of product. This additional cost per ton remains constant through-
out the range of operations covered by the data. The average cost
per ton,- of course, decreases as volume rises.
^The elements coniposing these additional and fixed costs follow in
table 1 :
Table 1. — Elements of total costs, 1938 conditions, United States Steel Corporation
and subsidiaries
Additional
cost for
Cost that
each addi-
Item
does not
tional
vary with
weighted
production
ton of
product
shipped
Interest _
$8, 300, 000-
7, 700, 000
24, 200, 000
0
$1.43
Payroll _. -....
62,100,000-
2,500,000
29.10
Social Security taxes
1.16
47, 800, 000
21.67
Total cash costs
152,600,000
53.36
29.500,000
2.37
Total costs --
182,100,000
55 73
Acting Chairman King. It may be included in your testimony.
Dr. Yntema. I should like to point out one or two characteristics
of this table. In the first place, that of the $55.73 additional cost
per weighted ton of products shipped. The additional pay roll ac-
counts for $29 and the additional other cash expenses, made up
largely of materials and services purchased from others, account for
$21. The depreciation and depletion is a relatively small item of
the total.
Of the total costs, $53.36 of the additional costs and $152,600,000
of the fixed costs are cash outlays, and the respective remainders
consist of depreciation and depletion.
I might interpolate there with one comment, that even in the case
of depreciation and depletion you can't avoid entirely some of the
expenditures necessary for replacements and additions to equipment
during depression. Those are costs which in the long run also must
be met in cash.
GONCENTRATIOISt Oli' EOONOMIC POWER 13597
Mr. HiNRiCHs. May I ask one question in this connection? In
these costs that do not vary with production you show a total of
$62,100,000 for labor.
Dr. Yntema. Yes, sir.
Mr. HiNRicHS. And you then show variable pay-roll costs of $29.10
a ton.
Dr. Yntema. Yes.
Mr. HiNRicHS. For additional labor per ton for whatever volume of
production there may be. That indicates an absolutely linear
relationship as far as labor costs are concerned, yet fixity in the
bottom end, perfect variability within the range of 18 to 90 percent
of capacity production. Have you tested out other types of relation-
ship as well as the straight linear relationship? Are you convinced
on the basis of numerous other tests that that relationship is abso-
lutely linear, because that is a difiFerent concept than I think the steel
industry itself has had in the past.
Dr. Yntema. In "Exhibit No. 1416," p. 28,^ there is a chart which
mdicates the relation of pay rolls to millions of weighted tons of all
tonnage products shipped. The dots cluster very close to a straight
line.
Mr. HiNRiCHS. I am sorry, my memory for numbers isn't good.
Dr. Yntema. It is "Exhibit No. 1416," entitled "An Analysis -of
Steel Prices Volmne and Costs."
Mr. HiNRiCHs. Summary?
Dr. Yntema. No, the larger pamphlet.
Mr. HiNRiCHS. Page 28,. you say? ^
Dr. Yntema. Yes, sir.
Dr. Kreps. We shall have something to say about that linear rela-
tionship when Dr. Louis Bean takes the stand.
Mr. HiNRicHs. My question was essentially whether you had tried
other types of relationship as well and concluded that this was the most
significant.
Dr. Yntema. No, we hayen't tried other types of relationship.
Mr. HiNRiCHS. This is based on the inspection
Dr. Yntema (interposing). It is based on inspection. I should be
reasonably certain, however, that with inspection you wouldn't obtam
a better fit from some other function. That is supported by oth.er stud-
ies we have made. We have analyzed the relationship of man-hours to
output by months and also by subdivisions of the Corporation, and in
most cases we get very nearly a straight-line relationship, although in
some instances, surprisingly enough, the tendency is for the line not
to rise throughout at a constant rate but, at the higher rates of ppera-
tion, to fail to rise at the same rate as it does in the lower range of the
curve. I should say, however, that you get a vdry good description
of the behavior of labor costs in relation to output — when you have
adjusted for differences of the average hourly earnings at different
points — by fitting a straight line to the data.
Acting Chairman King. Proceed.
EFFECTS OF PRICE REDUCTIONS
Dr. Yntema. From the relationship between costs and volume it
is possible to determine the increa-se in volume necessary to compen-
> Appendix, p. 14053.
13598 CONCENTRATION OF ECONOMIC POWER
sate for a given price reduction. Although our estimates of the
elasticity of demand for steel are less than 1, it will be assumed in
the following calculations that a given percentage reduction in price
will cause an equal relative increase in the volume of steel sold, so that
the dollar amount of sales will remain the same. In other words, the
elasticity of demand will be assumed equal to 1.
The. sales and revenues of United States Steel Corporation subsid-
iaries ill 1938 amounted to $77.66 per weighted ton of products shipped.
Of this amount $71.86 represented the amount received from the sale
of steel and other products, and $5.80 represented income from
transportation and miscellaneous operations.
On the assumption of unitary elasticity of demand and no increase
in transportation and miscellaneous operating revenues, a 10 percent
decrease from the average price level in 1938 would require an increase
of'48.8 percent in volume of shipments to avoid loss from price reduc-
tion. But the maximum increase in volume to be expected from the
price reduction, on the assumption made, is only 1 1 percent. Thus it
is clear that a price decrease would induce only a small proportion of
the tonnage increase which would be necessary to compensate for it.
In table 2 this relationship is shown for price reductions ranging from
1 to 20 percent,
I should like to insert table 2 in the record and defer discussion until
a later time when I shall show a chart which perhaps will bring out
more clearly the significance of this material.
Table 2. — Percentage increases in volume needed to offset various "percentage reduc-
tions from average 19S8 prices and effect of price reductions on losses — United
States Steel Corporation and subsidiaries
Percentage
Estimated
Percent-
age re-
duction
in price
incre^e in
volume
needed to
compensate
for price
reduction
Percentage
increase,
assuming
elasticity of
1
Estimated
additional
loss, assuming
elasticity of 1
additional
loss, if no
increase in
volume
resulted from
price reduction
1
3.3?
1.0
$3,900,000
$5,600,000
2
7.01
2.0
7,900.000
11,200,000
10.91
3.1
12.000,000
16.800,000
U.OO
4.2
16, 200, 000
22,400,000
IQ.fiO
5.3
20,500,000
28,000.000
24.48
6.4
24.900,000
33, 600, 000
29.77
7.5
2^,300,000
39, 200, 000
35. 54
8.7
33,900,000
44,800,000
41.83
9.9
38, 500, oon
50. 400, 000
10
48.75
11.1
43, 300, 000
56, 100, 000
11
56.38
12.4
48, 100, 000
61,700,000
12
64.82
13.6
53, 100, 000
67, 300, 000
13
74.24
14.9
58, 200, 000
72,900.000
U
84.78
16.3
63, 400, 000
78. 500, 000
15
96.70
17.7
08, 700, 000
84.100,000
20
190. 26
25.0
97, 400, 0(X)
112,100,000
In 1938 the subsidiaries of thj United States Steel Corporation
shipped 7,800,000 weighted tons, while in 1937 they shipped
13,200,000 tons. To bring the 1938 weighted tonnage up to the 1937
level, a 69.23 percent increase would have been necessary. On the
assumption of a unitary elasticity of demand, this would have required
a price decrease of 40.9 percent. After such a price reduction, revenue
per ^veighted ton would have been $48.26, or $5.10 less than the
additional cost per ton of products shipped. On the assumption
(contrary to our previous findings) that the price reduction of 40 9
CONOENTRATION OF ECONOMIC POWER 13599
percent would have been sufficient to restore the 1937 vohime, 13,-
200,000 weighted tons would have been sold. The Corporation and
its subsidiaries would then have had a cash loss of $152,600,000 out-
of-pocket fixed costs plus a further loss of $5.10 per ton, or a total cash
loss of $219,920,000. If depreciation and depletion of assets at this
rate of operations, amounting to $60,784,000, were added to the cash
loss, the total loss would have been $280,704,000. In 1 year this
would have wiped out more than half the current assets of the Corpo-
ration.
The 1938 price level used in the foregoing calculations is the average
price in effect both before and after the June 1938 reduction of approxi-
mately 10 percent in the published prices. The relationship between
annual sales and revenues and annual costs at various levels oi pro-
duction has also been computed on the basis of prices prevailing during
the second half of 1938. At this lower price level the break-even
point (under 1938 cost conditions, without any allowance for di^ndends
on preferred stock) would have been at about 10,500,000 weighted
tons, which is equivalent to an operating rate of 50 to 55 percent of
capacity. A 10-percent reduction in prices from this level would
have raised the break-even point to about 90 percent of capacity.
If the break-even point were this high, the Corporation would have
to operate at the impossible annual rate of 130 percent of capacity
to earn a return before income taxes of only 5 percent on its invest-
ment in tangible assets.
At this point, if it please the committee, I should like to show this
material in cha,rt form. I think it may be somewhat clearer.
Acting Chairman King. Proceed.
Dr. Yntema. This is chart B-9 in "Exhibit No. 1409."^ It is
entitled "Unadjusted Costs and Volume of Business Compared With
Estimated Costs for Corresponding Volumes Under 1938 Conditions."
For the time being I should like to neglect this dotted line, we have
drawn through the chart; I will come back and show the significance
of that later on.
We have plotted here the millions of dollars of costs, the actual
unadjusted costs," neglecting, however, intercompany items and non-
operating transactions which are extraneous to the production process
of the Corporation. We have plotted here the millions of weighted
tons of all tonnage products shipped; that is, weighting the lower-cost
products and the higher-cost products in such a way as to obtain as
satisfactory an index of volume of production as possible.
These dots represent the costs and the volume of production in the
respective years. For example, in 1929 the cost of production was
represented by the height at this point above the base line, and the
distance from the left-hand side of the chart over to the point repre-
sents the tons of products shipped.
These are the data with which we began our cost analysis. They
are the actual unadjusted costs and the weighted tonnages of products
shipped. You can see that even without any adjustment they are
distributed roughly along o, straight line.
I should like to show now, by another chart, the effect of our adjust-
ments upon these points.
Dr. Kreps. What number is this next chart?
» Appendix, p. 13789.
13600 CONOBNTRATION OP EJOONOMIC POWER
Dr. Yntema. This is No. B-1 in "Exhibit No. 1409," ' entitled
"Relationship Between Total Costs of Operation and Volume of
Business — 1938 Conditions." Along the horizontal base line we are
still plotting the millions of- weighted tons of all tonnage products
shipped. There has been no adjustment in that for each of the years,
but we have in each year adjusted these total costs as I have described
in the statement just subm.itted to the committee. In each case we
have taken the components of cost, the pay rolls, the taxes, and other
costs, and adjusted them to the wage rate, tax, and material price
conditions existing in 1938 insofar as those adjustments were possible.
We don't claim perfection for such adjustments. There are many
obstacles in the way of achieving perfection, but we think these
represent reasonably satisfactory adjustments. I understand there
will be som.e discussion of Ihis question later before the committee.
There may be more elaboration of these adjustments at that time.
Even a casual inspection of the chart will show that these points
lie very closely upon a straight line, and we have therefore fitted such
a line to these observations. This line purports to show, then, how
the total costs of the Steel Corporation would vary with variation in
volume, removing, however, the effects of changes in wage rates,
material prices, tax rates; it shows how total costs would respond to
changes in one factor alone, changes in volume.
We don't particularly want to call attention to the extrapolation
of this line beyond the range of the data. We are not interested in
that. Extrapolating the line does give, however, a convenient
method of describing how the , costs behave. This amount of
$182,100,000 we have determined as fixed cost. It represents in a
sense the amount of cost which does not vary with output. The total
costs rise by a constant amount per ton as the volume increases.
That increase in total costs, per ton increase in volume, is the addi-
tional cost, $55.73, which I have described in the statement just
read to you.
This next chart is No. B-2 in "Exhibit No. 1409." ^ It is entitled
"Composition of Total Costs of Operation in Relation to Volume of
Business." It shows total costs in relation to output, exactly the
same line that was just presented to you in chart B-1. We have
taken the line of total costs from chart B-1 and placed it here. Then
we made an analysis of the behavior of the individual components
of cost; we took the components of the total cost and studied how they
were related to volume. This chart gives the result of those studies.
They are. described in somewhat greater detail in "Exhibit No. 1416"
entitled "Analysis of Steel Prices, Volume and Costs."
CASH COSTS
Dr. Yntema. There are two characteristics of this chart which
I think are intercbting. • One of them is the relatively small proportion
of noncash outlays. The goods and services pm-chased from others
must be paid for in cash from year to year. Social Security taxes
must be paid for in cash from year to year. The pay rell must be
paid for in cash from year to year. The depreciation and depletion
to some extent, to a very considerable extent, may be deferred, but
1 Appendix, p. 13773.
• Appendix, pi 13776.
CONCENTRATION OF ECONOMIC POWER 13601
even that cannot be completely deferred in depression; what I mean
is that there is even in depression necessity for making some expendi-
tm-e for replacement of equipment. The taxes other than Federal
income and Social Secm-ity taxes must be met in cash from year to
year. You notice, therefore, that of the total costs, the total variable
costs, so-called, practically all of them are cash outlays; only a very
small proportion represents noncash outlays. Of the so-called costs
which we have termed "fixed costs" the goods and services purchased
from others must be paid for in cash; the Social Security taxes must be
paid in cash; the pay roU of course must be met by cash expenditure;
the depreciation and depletion does not represent entirely immediate
cash outlays. The taxes other than Federal income and Social
Security taxes must be met in cash, and pensions, and interest of
course represent cash payments.
In summary, therefore, the proportion of noncash overhead is
indeed very small. You will notice, furthermore, that we have not
included in this chart any provision for dividends or profits.
Mr. Wooden. Does the item of goods and services purchased
from others include only raw materials used in the manufacture and
production of steel?
Dr. Yntema. It includes raw materials purchased from others.
Mr. Wooden. For use only in the manufacture of steel?
Dr. Yntema. For use by the Steel Corporation in all its operations.
We have tried to bring together all operations in this picture. The
reason for doing that is not that we did not want to analyze the
components, but that any analysis of the component operations would
involve questions of accounting allocation which of necessity would be
arbitrary.
Mr. Wooden. Do you have any break-down between the cost of
goods the cost of services purchased from others?
Dr. Yntema. It wasn't possible for us to obtain that break-down
without a tremendous amount of work. If it had been available we
would have used it, but we found it was impossible to get that in-
formation because of the particular way in which the records of the
Corporation were kept, each subsidiary corporation's records being
kept as a separate entity.
Mr. HiNRiCHS. When you speak of pay roll you are referring to
all payments for services, including your salaried workers, payments
to ofiicials as well as wage earners?
Dr. Yntema. Yes, sir; but not for all services, it doesn't include
some types of professional service obtained from others, but it includes
all the salaried pay roll to the corporation employees.
Mr. Reynders. They are in that lower range of salaried pay roll
shown on this chart, aren't they?
Dr. Yntema. I can't tell you what the distribution is. A good
deal of the salaried employees I think wou'd appear in this pay roll
included among "fixed costs" at the bottom of the chart, but even
there it may not be entirely true. I just can't tell you the extent to
which this item is composed of wage earners and the extent to which
it is composed of salaried workers.
Mr. Reynders. The straight-line characteristic of your labor is a
fortuitous circumstance, isn't it, because the higher operating rate
you have, the higher are the rates of labor.
13602 CONCENTRATION OF ECONOMIC POWER
Dr. Yntema. No; we have eliminated the effects of differences of
average hourly earnings for all employees. We found by study that
the average hourly earnings of all employees, wage earners and others,
are approximately the same, independent of output, so long as there
are not wage rate changes. We have a chart here which we will be
glad to put in evidence if you would like to see it, showing that parti-
cular point. I think later on in the discussion we should like to pre-
sent it to the committee.
In this next chart. No. B-3, in "Exhibit No. 1409," ^ entitled "Rela-
tionship Between Sales and Costs — Effect of Reduction from Average
1938 Prices," again we have taken this line representing total costs
determined in the earlier chart, No. B-1, and we have superimposed
upon this chart two other lines, one — the light double line — represent-
ing total sales and revenues at 1938 average prices. This light double
line represents what average sales and revenues at the various volumes
would be under 1938 prices. In other words, the height of this line
at any giv^n point of volume represents what the total sales and reve-
nues would have been at 1938 prices if this volume had been sold.
The point at which the total sales and revenues are equal to the
total costs, sometimes called the break-even point, appears in this
chart at approximately 8,300,000 weighted tons of all tonnage prod-
ucts shipped, on the basis of average 1938 prices. If the price had
been reduced, if the 1938 average prices' had been 10 percent lower,
the sales and revenues line would have been represented by this heavy
broken line on this chart. You see that for any given volume of
product the height of this line is IQ percent less than that of the light
double line above it on this chart.
The break-even point in that latter case would have come at ap-
proximately 12,500,000 tons of aU tonnage products shipped, or
roughly the break-even point would have been in the neighborhood
of 70 percent of capacity. This shows the effect of a 10-percent reduc-
tion from the 1938 average price level upon the break-even point.
Mr. HiNRiCHS. You just said 70 percent; in your earlier testimony,
if, I remember correctly, you said a 10-percent reduction in prices
would have raised the break-even point to about 90 percent of ca-
pacity.
Dr. Yntema. No; that applies to the level of prices prevailing in
the second half of 1938, and I will show in just a moment a chart
corresponding to that testimony.
Mr. Wooden. Is it a composite base price or a composite mill-net
yield that you have used there?
Dr. Yntema. That is the composite of actual receipts.
Mr. Wooden. Mill-net yield?
Dr. Yntema. This is mill-net yields for aU steel products and net
vields to the Corporation from all other nonsteel products as well.
This next chart is No. B-4 in "Exhibit No. 1409." =^ It is entitled
"Increases in Volume Needed to Compensate for Various Decreases
in 1938 Prices Compared to Probable Resulting Increases in Volume."
You will remember in the statements which I made a few minutes
ago that WG assumed in our calculations an elasticity of 1 for demand
for steel, although we thought the elasticity was considerably lower
than that. We have plotted along the base line the percentage de-
• Appendix, p. 13777.
' Appendix, p. 13779.
CONCENTRATION OF ECONOMIC POWER 13603
creases from the average 1938 prices, and this black line labeled "Prob-
able Resulting Increases in Volume" shows what the volume increase
would have been if the elasticity of demand were unity.
The height of the bars in the chart shows what the increase in
volume would have had to be if the price decrease had not brought a
financial loss to the corporation. You can see very readily that in
all cases the needed increase in volume required to offset the decrease
in price was far greater than any increase in volume which could reason-
ably be expected to eventuate from the price reduction.
This next chart is numbered B-5 in "Exhibit No. 1409." It is
entitled "Estimated Additions to 1938 Deficit. How Deficit Would
Have Increased if Prices Had Been Reduced and Volume Had
Increased to Same Relative Extent." In this chart we have plotted
along the base line various percentage reductions in the 1938 average
price. The black part of the bar in each case represents what the 1 938
deficit of the Steel Corporation actually was. The dotted part of the
bar above the black portion represents the addition to the deficit if
prices had been reduced as indicated by these respective percentages
on the base line. Thus, for example, if there had been a 10 percent
reduction in price the additional deficit would have been indicated by
the large dotted portion of the bar above the lower black portion over
the figure 10 on the base line.
Dr. Keeps. Would you carefully state for the committee what the
assumption is upon which you base these estimates?
Dr. Yntema. Yes. We have assumed, first, an elasticity of
demand of one for steel.
Dr. Keeps. By demand you mean market demand rather than
demand in what is called the schedule sense?
Dr. Yntema. We use elasticity»in this sense, that if the price were
reduced 1 percent there would be a 1 percent increase in quantity of
steel which would be bought.
Dr. Keeps. You mean you are, therefore, allowing for no effect in
the reduction of the price of steel upon general business activity, con-
sumer income, and industrial profits.
Dr. Yntema. Yes. We are neglecting for the time any such indirect
effects; it seems to me beyond possible doubt that the indirect effects
increasing the possible elasticity would be far less than the error we
have made by assuming as high an elasticity as one. I state that
merely as my own personal opinion, that we have erred on the upper
side. Even if we took the definition which you have suggested as an
alternative I think the elasticity still would not be greater than one.
Dr. Keeps. This does not depend so much upon the elasticity of
. demand as upon the nature of yoiir cost curve, does it not?
Dr. Yntema. I should say that it depended upon both, that if you
made large changes in either one you would get somewhat different
results, but if you made moderate changes in the cost function or
moderate changes in the elasticity of demand for steel, either one, you
would still get substantially the same order of results.
Dr. Keeps. You agree that general business activity, consumer
income, and industrial profits affect the volume of steel sold, do you
not?
Dr. Yntema. Ob, that is correct, they are much more important
than price.
13604 CONCENTRATION OF ECONOMIC POWER
Dr. Kreps. You do not mean to say that the volume of steel
sold, the amount of employment in. the steel industry, steel being
roughly $1 out of 14, in the economy, has no effect upon industrial
activity, upon consumer income, and upon industrial profits?
Dr. Yntema. No; I would never want to be quoted to that effect —
that the amount of steel sold has no effect.
Dr. Kreps. Yet in your charts you have made no such allowance,
have you?
Dr. Yntema. Let me make clear what we have in the charts. We
have assumed with reference to the elasticity of demand a figure which
we think is high, perhaps twice or three times as high as we believe to
be the best estimate. We have tried to be conservative in that
respect. We would recognize that if you allow for the secondary-
repercussions there would be perhaps some increase in that elasticity.
I personally regard those effects as much smaller than the margin of
error which we have introduced by assuming an elasticity of 1 instead
of 0.3 or 0.4.
Dr. Kreps. That is your own personal estimate. You have made
no calculations?
Dr. Yntema. No; I don't see how it would be possible to make
satisfactory calculations on that point.
Dr. Kreps. You have made no endeavor to show to what extent
the activity in the steel industry and the leadership furnished by
steel executives and by steel price policies tends to influence if not
affect and determine general industrial activity and the general level
of industrial prices?
Dr. Yntema. We are carrying on research, not in the Steel Corpora-
tion but in the Cowles Commission at the University of Chicago, on
the demand for steel, and if you or anyone else has any other sugges-
tions as to how to proceed in such an investigation we shall be more
than delighted to receive them.
Dr. Kreps. We have some suggestions in that regard.
I want merely to make clear for the record that the assumption
upon which these charts are based is contrary to economic fact.
Mr. Ballinger. Dr. Yntema, I have been very much impressed
with your logic, but I have one question I want to ask. Why does
the Steel Corporation ever reduce prices then, since it seems to be a
ruinous policy? I mean they are going to sell the same amount of
steel for less revenue. That doesn't seem to make sense.
Dr. Yntema. I don't like to get into a long discussion. I think
the answer is competition.
Mr. Ballinger. If there were none?
Dr. Yntema. If there were no competition in the steel industry the
prices would certainly be much higher than they are and have been
In short, I think it is due to competition in the steel industry.
Mr. Ballinger. Assuming there was no competition, is there any
way you can explain it except there is elasticity to the demand for
steel? Suppose we settle that, then how could you explain a reduction
in the price of steel except on the condition that the demand for steel
was elastic?
Dr. Yntema. If there were no competition — that is a condition
which I think is contrary to fact — in the steel industry
Mr. Ballinger (interposing). Price competition.
CONCENTRATION OF ECONOMIC POWER 13605
Dr. Yntema. No price competition in the steel industry, no
effective price competition in the steel industry? I think you could
explain these facts only by the goodheartedness of the steel industry.
Acting Chairman King. Is it not a fact that in many industries
during this depression and even at other times there have been losses,
deficits, which have been made too often by invading capital and im-
pairing utimately the economic structure or stability of the corporation
Dr. Yntema. I didn't get one part of the question — by doing what
to capital?
Acting Chairman King. Invading capital.
Dr. Yntema. By invading capital?
Acting Chairman King. That is making it less valuable.
Dr. Yntema. Of course, it is true that if an industry does not main-
tain its productive facilities at a level with current technological
advances, consumers suffer thereby. I think it is possible to name
industries of that sort. I am no expert in the railroad industry, but
I think the railroads have suffered so severe a depression that that
probably has been true. I can't offer that as expert opinion, but
only as a lay observer.
Acting Chairman King. From your studies of our economic and
industrial situation have you discovered that many industries continue
to function as best they may, even though they are sustaining annually
very severe losses?
Dr. Yntema. I think that there are many industries which, in spite
of the losses they sustain, stUl maintain a fairly high level of technical
efficiency. I think there is no doubt but that in depression technical
advances may sometimes be slowed up due to the fact you cite. It is
hard to generalize in as complicated a question as that. I doubt if
that has been a major factor retarding technical progress. I don't
think it has in the steel industry.
Acting Chairman King. I didn't refer particularly so much to
technical progress, but merely to disposition of men in business
whether small business or large business, whether in the agricultural
business, the production of agricultural commodities, sugar for
instance, or in the mining industry in the production of copper, lead,
zinc, and other ores, is it not a fact that in many of those cases there
have been losses for protracted periods and yet the mines couldn't
afford to close down, the agriculturists couldn't afford not to plow their
fields, they had to raise something, and they have met their losses
with a good deal of courage and sometimes they haven't met them
and have gone into bankruptcy, but have continued oftentimes for
indefinite periods, hoping that the clouds would be dissipated and
the sun would again shine.
Dr. Yntema. I think that there are many stockholders in the
steel industry and many managers in the steel industry who would
have a sympathetic response to that suggestion.
Dr. Keeps. And laborers you would add, wouldn't you?
Dr. Yntema. Yss; quite so.
Acting Chairman King. When corporations shut down or close
their business or farmers don't produce, when mines shut down, the
laborers suffer of course, so obviously it doesn't need any explanation.
Mr. Feller. Assuming that the rate of operation of the Corporation
is in the neighborhood of 80 percent today, does it foUow from your
13606 CONCENTRATION OF ECONOMIC POWER
charts that if the Corporation were to double its price today and be
content with a rate of operation of 40 percent they would make just as
much money?
If you take half the production at twice the price and make just the
same profit?
Dr. Yntema. I think you would make a larger profit. If you ob-
tained the same total income and reduced your costs you would make
a larger profit than you did before.
Mr. Feller. Your costs would be reduced?
Dr. Yntena. Yes; if you reduced your operations from 80 to 40
percent.
Mr. Feller. On your calculation, you mean the average cost per
ton of steel would go down?
Dr. Yntema. No; the average cost per ton of steel would go up.
Mr. Feller. If you reduced your operation to 40 percent?
Dr. Yntema. Yes; but the margin would be so much greater that
the total profit would greatly exceed the current profit.
* Mr. Feller. Then isn't the industry very foolish in attempting to
get all this business? Why don't they go after less business?
Dr. Yntema. It is true that if any business were perfectly free to set
its prices anywhere it pleased, most businesses would set their prices
higher and most businesses would make larger profits, but that is not
the way business has to operate. Most business today is competitive,
and the limits upon the prices which any particular business can set
are determined by competition. It is not possible for the Steel Cor-
poration or for any concern in the steel industry to set its prices at
any level that it pleases. These phenomena would reveal that fact
if they were approached with some understanding of the degree of
competition which does exist in the steel industry.
Mr. Feller. Let's assume a constant price, then. Assuming that
the price is not raised, the price remains at the present level. On the
cost curve that you showed us a while back wouldn't it follow that the
increment to profit which results from increases in the rate of opera-
tions would progressively get smaller and smaller?
Dr. Yntema. Perhaps I can deal with that question in terms of this
chart. This is chart B-3 in "Exhibit No. 1409" \ entitled "Relation-
ship Between Sales and Costs. Effects of Reduction from Average
1938 Prices." The solid black line in this chart represents the total
cost under 1938 price, wage rate, tax rate, and other conditions at
various volumes of production. The double line represents what sales
and revenues would have been at various rates of operations with 1938
prices. Now the second question which you put to me might be
answered in this way. If you took, for example, the rate of operation
at 14,000,000 of weighted tons of all products shipped, the difference
between this double line and the solid black line would represent the
profit under 1938 conditions, with the exclusion of such items as we
have recognized. If we drop do^vn to 7,000,000 tons, the income would
then have been less than the cost, so a reduction in rate of operations
from 14,000,000 weighted tons to 7,000,000 weighted tons would have
converted a profit into a loss. That is based upon the assumption that
there is no change in prices.
' Appendix, p. 13777.
CONCENTRATION OP ECONOMIC POWER 13607
Dr. Keeps. And on the additional assumption that the costs which
you show for the United States Steel Corporation are costs that apply
to every other corporation in the industry.
Dr. Yntema. Yes; that is quite right. My remarks here don't
apply to the ether corporations in the industry. I am talking now
about the relation between price levels and costs for the Steel Cor-
poration.
(Mr. O'Connell assumed the Chair.)
Mr. Feller. The reason you have those two lines is this, isn't it:
That there are $200,000,000 approxiniately of fixed costs, $200,000,000
of costs which never vary, which are invariable. Your variable costs,
however, you have shown go up in a straight hne. Your sales line
Dr. Yntema (interposing). Pardon me, the variable costs per unit
remain constant. The aggregate of all variable costs goes up in a
straight line.
Mr. Feller. The variable costs per unit remain constant. There-
fore, the only possibility of making an increased profit on increased
rate of operations, the price remaining the same, is in that $200,000,000,
fixed cost. That is right, isn't it?
Dr. Yntema. There are many ways of saying this. One way of
putting it is as follows: If the price is above the initial cost the larger
the number of imits you sell at that price the larger the profit. That is
apparent, because if you increase your volume. the income increases
more rapidly than the costs for producing and supplying the product.
If the price were to be brought below the variable costs, the losses
would go up as the volume increased. If the price were exactly equal
to the variable cost, the total loss would be equal to the overhead,
no matter what the volume.
I should like to make one comment parenthetically, if I may. Any
statistician of any competence would know that you cannot, by
fitting a straight hne to a series of points, show that that is precisely
the true functional relation between one variable and the other.
But you can say, if the points approach very closely to a straight line,
that it represents to a reasonable degree of approximation, the func-
tional relation in question. I don't want to be quoted as saying that
the additional costs are precisely $55.73, and that they stay precisely
constant. I would say that the evidence we have surveyed indicates
that that is a reasonable approximation to the facts.
Dr. Kreps. To the facts for the U. S. Steel Corporation?
Dr. Yntema. To the facts for the U. S. Steel Corporation.
Dr. Keeps. Which you are applying to the entire industry?
Dr. Yntema. No. Now, let's make clear two "types of things.
What I am suggesting is this: Not that for the entire industry the
cost fimction would be identical with this, but that the operations of
producing steel are somewhat similar in the other steel corporations.
Dr. Kreps. But the results are dissimilar as shown by the proved
records of the various corporations, are they not?
Dr. Yntema. If I may continue, the operations are similar, and
if you were to construct a cost function for the other corporations
in the industry, you would find, as a matter of fact, somewhat the
same type ©f pattern. Where the profit margins are larger, you would
find, of course, that the cost line would be lower in relation to the
13608 CONCENTRATION OF ECONOMIC POWER
income line; where the proj&t margins are narrower, you would find
that the cost line was higher in relation to the price and income line.
Dr. Kreps. Do you have
Dr. Yntema (interposing). Just a minute. If the committee would
like to see a chart which was prepared, which hasn't yet come from
the drafting office, we will be glad to submit this afternoon an exhibit
which shows the unadjusted cost figures for a number of concerns in
the steel industry. Now, those costs have not been adjusted. They
are not comparable strictly to what we have had presented to the com-
mittee, but I think they wiU throw some light on this problem and
we will be glad to offer them. They are not our own figures.
Dr. Keeps. I want to pursue one point just a little further.
This cost line is the cost line for the United States Steel Corporation
is it not?
Dr. Yntema. That is correct; under IP'"^ conditions.
Dr. Keeps. Quite. You make no . nptions what it would be
for other competitors, let's say, in the steel industry?
Dr. Yntema. No, we make no assumptions because we have said
nothing about that except implicitly to this extent: I have made no
detailed study other than what appears in this chart. We haven't
access to the figures of our competitors. I wish we did. It would
be a very interesting study to make, to compare the cost functions of
other companies in the ind stry. [Laughter. j
My judgment is this, t at because the operations are somewhat
similar and because the companies have to compete in the same
markets generally for the services and products which are employed,
the cost functions would be somewhat similar. Now, there would
be a difference due to the degree of integration. The lower the degree
of integration in the companies, the smaller would be the fixed or
overhead cost and the higher would be the variable cost. So that I
should suspect, if anything, that the other — that many of the. other
smaller companies in the industry would have a comparatively smaller
proportion of overhead costs. But I am merely inferring that, and
I can't say it with any certainty.
Mr. Fellee. Dr. Yntema, I just want to get the exact significance
of your testimony with relation to costs. Did I understand this to
be correct, that looking at the chart, which I think is B-2, the largest
segment in the sloping part, the variable cost part, is taken up by
pay-roll?
Dr. Yntema. That is correct.
Mr. Feller. Now, your testimony is to this effect, that at aU rates
of operation, once the Corporation has passed this $200,000,000 fixed
cost, at all rates of operation, the unit cost of labor per ton of steel is
constant; is that correct?
Dr Yntema. Well, the additional cost, of additional labor per
ton of all products is approximately constant, and that is borne out
by other studies we have made. I think that is a very reasonable
approximation of fact.
Mr. Feller. I just wanted to get that clear.
Dr. Yntema. It is conceivable that some conditions might exist
under which that might not be true, but that, I think, is the fair
approximation to a description of the facts.
CONCENTRATION OF ECONOMIC POWER 13609
This next chart is numbered B-6 in "Exhibit No. 1409," ' and is
entitled "Relationship Between Sales and Costs. Effect of Reduction
from 2nd Half 1938 Prices." It is similar to chart B-3 in "Exhibit
No. 1409" ^ except for the fact that the sales and revenue line is
based upon the price levels prevailing in the second half of 1938 instead
of the full year of 1938.
In the middle of 1938 there occurred a substantial reduction in
steel prices, and this gives effect to that reduction in the price of steel.
As a consequence of that reduction the break-even point before any
provision for return on preferred or common stocks, rose to approxi-
mately 50 or 55 percent. If the price were further reduced 10 percent
from the level prevailing in the second half of 1938, and if the other
conditions in the year 1938 had prevailed, the break-even point would
have risen to 90 percent; in other words, under the wage rates, tax
rates and price levels prevailing in 1938, a price level 10 percent
lower than the price level prevailing in the second half of that year
would have necessitated an operating rate for the Steel Corporation of
90 percent, merely to cover its costs without any allowance for return
to the stockholders. If there had been included a provision of 5 per-
cent of return on the total investment in the Steel Corporation, that is,
the total assets less the current liabilities, the break-even point would
have had to be 130 percent.
That provides some evidence as to the possibilities further of
reductions in prices from the level prevailing in the second half of that
year.
Since that time, there have not been major changes in the published
prices of steel products.
Mr. Wooden, I take it that means, however, that it would not be
necessary to have a 130 percent rate of operation necessarily for
other units or other members of the industry?
Dr. Yntema. No, it would not. For some it probably would be
necessary to have higher rates than that and for others lower rates.
Mr. Wooden. Have you gone into that question?
Dr. Yntema. No, we havie not. As I said, in order to make a
complete study, I think this should be done for all the operating units
in the steel industry, but that, of course, is not within our province.
We made a preliminary study and I think in all likelihc od when you
completed such a study, you wouldn't find greatly different results
from those we have shown here, but that again is a matter of opinion
based upon the Examination only of limited published evidence.
This next chart is numbered B-7 in "Exhibit No. 1409," ^ It
shows the increases in volume needed to compensate for various
decreases in second-half 1938 prices, compared with the probable
resulting increases in volume. There is no need to describe the chart
in detail. It is similar to the chart preceding.
What it shows is that if there were to be decreases from the price
level prevailing in the second half of 1938, the increases in volume
necessary to compensate for the price decrease and leave no further
loss, would have been still greater than in the chart which you have
just seen.
' Appendix, p. 13783.
' Appendix, p. 13777.
'Appendix, p. 13785.
13610 CONCENTRATION OF ECONOMU^ POWER
This next chart is numbered B-9 inj "Exhibit^ No. 1409" ^ and is
entitled "Unadjusted Costs and Volume^of Business Compared with
Estimated Costs for Corresponding Volumes under 1938 Conditions."
We have inserted in tliis chart a line showinp the relation between
adjusted cost, that is, adjusted to 1938 conditions, and volume of
production. We have also inserted in this chart the actual costs,
not the adjusted costs, but the actual costs in these respective years.
Some interesting observations might be made from this chart.
One is that the level of cost prevailing in 1938 at various volumes of
output is this line, which represents what the cost would have been
under the 1938 wage rate, tax rate and price conditions at various
volumes of output for the United States Steel Corporation. The cost
level of output under 1938 conditions was substantially higher than it
had been in the years preceding; that is, the net effect of the price
changes and increases in wage rates, has been to increase substantially
the cost of production of steel, at least insofar as the evidence to be
obtained from the operations of the Steel Corporation is concerned.
Now, it is possible, as I will show in the chart to follow, to compute
an index by taking the ratio of the actual costs in 1929 to what the
costs would have been under 1938 conditions for that volume, the ratio
of the costs in 1930' to what the costs would have been under 1938
conditions for that volume, and thus to get an index of what the
actual costs were to what they would have been under 1938 con-
ditions. Let me present that in the following chart.
This chart is numbered C-25 in "Exhibit No. 1409." ^ It is entitled,
"Composite Mill Net Yield and Cost per Weighted Ton Shipped.
United States Steel Corporation and Subsidiaries." The lines repre-
sent index numbers. The dotted Hne depicts an index of the average
actual cost per weighted ton of products shipped. The actual cost
per ton — average actual cost per ton — reflects the effect of changes
in wage rates, changes in material prices and prices of services, changes
in tax rates and also changes in the volume of production." The
average costs tend to go up — do go up — as the volume of production
declines, because the fixed costs are then spread over a smaller number
of units of output.
Dr. Kreps. Dr. Yntema, would you explain the relationship
between these curves and profits? You do not mean to imply by the
fact that your index of composite mill net yield is below the index of
costs, from 1929 on, that there has been no year in which the Corpor-
ation has made a profit, do you?
Dr. Yntema. No; Dr. Kreps. ,1 was coming to that point in just
a moment. I should like to take these curves and explain what each
of them is and then come to your question. That is a point which
should be made clear in the discussion of this chart.
This second double line represents, on the basis of the chart we
have just shown, an estimate of what the costs would have been in
these various years under the wage rate conditions, the tax conditions,
and the material price conditions in those various years, but with the
volume of production which had existed in 1926.
Now, this is to some extent arbitrary, but I think it is none the less
a rather useful separation of the effects of two factors — the cflects of
' Appendix, p. 13789.
> Appendix, p. 13835.
CONCENTRATION OF ECONOMIC POWER 13611
changes in volume upon average cost, and the effects of all factors
taken together.
Dr. Keeps. Wouldn't you like to add to that, if you were an execu-
tive of the Steel Corporation, you might be looking around to see
whether you had maintained your efficiency?
Dr. Yntema. Yes; we should add
Dr. Kreps (interposing). Wouldn't this sort of a chart, sort of
arrest the attention of executives and make them wonder whether the
efficiency of their operations was quite as great as it had been a few
years previous?
Dr. Yntema. Well, I am not competent to appraise in any way the
efficiency of operations in the steel industry.
Dr. Kreps. At any rate, you have made no correction of any kind
for efficiency, have you?
Dr. Yntema. This is true : Yes ; we have in some of our studies made
a correction for efficiency, and there has been a definite dovvTiward
trend in costs in the Steel Corporation which I think can be attributable
only to increases in efficiency, and that is shown in the material pre-
sented in "Exhibit No. 1416," ^ entitled "An Analysis of Steel Prices,
Volume, and Costs — Controlhng Limitations on Price Reductions."
There is a downward trend in costs over the period studied, which does
reflect the increases in efficiency.
But to come back to this — this actual cost per ton of products
shipped would have been higher at this point if there had been no
increases in efficiency. This represents the net effect of all factors,
and if there had been no increase in eflLciency, this cost line, starting
from 100 here (this is an index number), would not have been as low
at the end as it actually was. This does represent a considerable
increase in efficiency, in comparison with the effept of other factors.
Dr. Keeps. An efficiency which did not, however, actuaUy lower
actual costs?
Dr. Yntema. No. The point is this, that the increase in efficiency
was a component tending to bring down average costs. The increase
in efficiency was not as great as the increase in other costs, particularly
wage rateSi
I think it is beyond doubt true that the wage rates in the Steel Cor-
poration have gone up faster than the efficiency of production, and
that is the primary reason why the costs per unit of product are higher
in these latter years than they were in the earlier years.
'Dr. Kreps. You have charts, I take it, that show the break-down
of costs and show the increasing efficiency?
Dr. Yntema. That is, as I just said, presented in a summary fashion
m "Exhibit No. 1416." We have not made extensive studies with
reference to that. I personally am not competent to pass on the
subject, because that would require
Dr. Keeps (interposing). I missed the detailed study. I did see the
assertion.
Dr., Yntema. Yes. The group with which I worked simply is not
competent to engage in a study of that sort, and if you wish to request
someone, Mr. Fairless- or someone from the Steel Corporation, who is
competent to respond to that question, I am sure he would be glad to
do so. But I don't feel competent to answer the question.
' Appendix, p. 14032.
124491— U—pt. 20 3
13612 CONCENTRATION OF ECONOMIC POWER
If I may come back to chart C-25 in "Exhibit No. 1409" ' the double
line represents, then, what the costs would have been, allowing for
changes in wage rates, tax rates, and efficiency, but eliminating the
eflFect of changes in volume. Now, it is subject to deficiencies, but
on the other hand, it is as satisfactory a result as we could obtain,
and we have shown you how we have obtained it.
We have also plotted here in the third fine the composite mill net
yield to the Corporation. This represents, as satisfactorily as we have
been able to obtain it, a measure of the price levels of the products sold
by the Corporation,
Now, the point was made earlier that the price level paid by those
who buy the products may be slightly different in its behavior from
the price level of the prices obtained by the Corporation. Those
differences, in my opinion, are negligible. It is the differences in the
fluctuations of those series, I think, that are negligible, and I believe
this is a satisfactory picture.
Mr. Feller. May I ask a question?
Dr. Yntema. May I go on one moment to complete this so we
won't break the structure of the argument?
Mr. Feller. I would like to interrupt you, if I may, to get on
what basis those two lines were plotted. You said the products sold
by the Corporation — you mean the steel products, or all products?
Dr. Yntema. The composite mill net yield covers only the steel
products. On the other hand, "only the steel products" includes
the great bulk, all but a relatively small percent, of the goods and
services sold by the Corporation.
Mr. Feller. That line which is above, designated actual cost
Dr. Yntema (interposing). That represents all costs of all prod-
ucts. The reason we have not
Mr. Feller (interposing). And of all subsidiaries?
Dr. Yntema. And of all subsidiaries.
Mr. Feller. Including transportation systems?
Dr. Yntema. Including all subsidiaries.
Mr. Feller. That's all. I just wanted to get that clear.
Dr. Yntema. The reason for handling the problem this way 's that
any separation would involve an arbitrary allocation of costs. It
would also involve a tremendous amount of work. I don't think that
if you made the separation the results would be substantially different
from these.
Mr. Wooden. Dr. Yntema, did you have access to any unit costs of
production, or did you make any study of the changes in unit costs as
compared to the labor costs per unit of production?
Dr. Yntema. What do you mean by unit costs of production?
Mr. Wooden. Per ton cost of production.
Dr. Yntema. You mean the average costs of production per ton?
Mr. Wooden. Yes.
Dr. Yntema. Yes, I have seem some of those. In such a case, in
case of any manufacturing operations, where there are multiple prod-
ucts produced, it is possible to make some fairly satisfactory study of
the costs immediately associated with each process, but when the
oveihead is allocated, the allocation becomes arbitrary. We were
inter, sted in this study not just in the direct costs, but in all the costs
of operation, so although I have seen some of the unit costs and I
' Appendix, p. 13835.
CONCENTRATION OF ECONOMIC POWER 13613
have studied the behavior of mill costs to some extent, those were not
appropriate for this particular problem and we have not presented
them to the committee. . In fact, I don't think that they are in form
to be particularly illuminating on the problems which concern the
committee and we have chosen from our material that which we
thought would be most useful to you.
If I may continue, this third line is an index of the composite mill
net yield of steel products. It is the best index we have been able to
present, showing the price level of the prices obtained by the Corpora-
tion. This chart is important because of the light it throws on price
flexibility. When you talk of price flexibility, there are, of course,
many criteria which you might use. You might talk of price flexi-
bility with respect to what the prices have been or were at some later
tima, or you might talk of price flexibility in relation to costs.
This, in a sense, is a picture of price flexibility in relation to costs.
As we proceed from 1929 to 1932, we see that the actual average cost
per ton went up sharply, and that the average miU net yield declined.
We find, however, that the cost prices, that an index of the cost prices,
if I may use that term, did not drop as much as the prices which the
Corporation got for its products. In other words, the Corporation,
during the depression, dropped the prices for its products more than
the prices which it paid for the goods and services which it used in
the production of those products.
It suffered, therefore, on two counts: First, because the price level,
or the prices of products and services which it sold, dropped more
than the price level of the products and services which it bought;
and second, because of the decrease in volume, which meant that the
overhead or fixed cost was spread over a smaller number of units. "
To come back to an earlier question which we should make perfectly
clear in this connection, these do not represent prices or cost per unit.
It is not possible in the chart in this case to represent those, for the
reason that the units are not the same. We took 1926 as a base.
In 1926, the Steel Corporation realized 6.2 percent return on its invest-
ment. We have not taken, therefore, I think, an unreasonable base
for the chart.
If this black line and this dotted line had remained at the same
level, and allowing for any possible imperfections in this index number,
since we do not claim it to be absolutely perfect, the return on invest-
ment would have stayed approximately at 6.2 percent. The return
ofi investment dropped and became negative because costs went up
in relation to prices in this period.
In 1937 there was a profit available, again because the ayerage
prices rose up to meet the average costs. I think this does afford some
interesting evidence, at least, on the general problems of price flexi-
bility.
Acting Chairman O'Connell. Dr. Yntema, if I may interrupt you,
would this be a convenient time to recess? Are you through with
this particular chart?
Dr. Yntema. This is almost the end of my statement.
Acting Chairman O'Connell. Then would you care to conclude?
Dr. Yntema. Yes. In brief, our studies show that the demand for
steel is determined primarily by general business activity, consumers'
income- and industrial profits, and to Only a minor extent by the price
of steel. The elasticity of demand for steel is so low that a reduction
13614 CONCENTRATION OF ECONOMIC POWER
in steel prices does not provide an effective means of increasing pro-
duction and employment in the industry. Because of this inelastic
demand and the character of costs in the industry, a moderate decrease
in price results in a great decrease in profits or increase in losses. Since
margins of profit in the steel industry during the past 10 years have
been and still are extremely low, no substantial reduction in steel
prices could have been borne or could now be borne by the industry
without corresponding reductions in costs. This could not be effected
without great reductions in wage rates.
Dr. Kreps. You should say by the Steel Corporation, shouldn't
you?
Dr. Yntema. No; I should say by the steel industry. According
to the figures which are submitted in your own record, the average
earnings in the steel industry have been extremely low in the last 10
years.
Dr. Kreps. For all the plants?
Dr Yntema. Not for all the individual plants but the average for
all of them is extremely low. I think that is a mild statement.
Mr. Ballinger. Very high perhaps from 1901 on to 1930?
Dr. Yntema. I am not talking about that; I am talking about
things as they are in this particular statement.
Dr. Kreps. Your evidence, in other words, does not show that this
is true for the industry ; it does show that it is true for the Corporation?
Dr. Yntema. No ; we are referring to figures submitted in the record
by your own research group on this particular point. We didn't think
it necessary to supplement that.
Dr. Kreps. But your comment concerns what you believe has been
inserted in the record about other corporations?
Dr. Yntema. It is my comment on what has been inserted in the
record, and also my comment as ^to what I believe to be the facts of
the case on the basis of such evidence as is available to me
A substantial reduction in prices could not be effected without
great reductions in wage rates.
Mr. Ballinger. Dr. Yntema, do you know how many industries in
the United States there are in which steel is a factor in cost of produc-
tion? I mean, steel is sold to them as their fabricating product and
the purchase of steel is a facior in production.
Dr. Yntema. I should say there are practically no industries in the
United States in which steel is not used, either as a raw material or in
the form of machinery, but I think it ought to be added immediately
that the cost of the steel, in proportion to other costs in niost of those
industries, is negligible, and even in the case of the major steel
consuming industries, the proportion of the cost of steel to the other
costs is very low; it is very low, as we have pointed out.
Acting Chairman O'Connell. Dr. Kreps, do you intend to recall
Dr. Yntema?
Dr. Kreps. Yes; I should like to recall Dr. Yntema after we have
heard certain other witnesses. Before dismissing, Dr. Yntema, I
should like, on behalf of all of 'us who have examined the data, to
express high tribute, not only to the United States Steel Corporation
but to Dr. Yntema personally, for' the many new item's of information
which they have given to the committee and to economists and
businessmen througho it the country.
CONCENTRATION OF ECONOMIC POWER 13615
If I may, I should like to summarize some of these for the committee.
New information has been made available in three fields, those of
costs, prices and labor. As you have noticed, the material on costs
not only presents break-downs which are entirely new, but even some
that are probably unknown to most firms in the industry at least in
the form in which they have been presented. Some of the items shown
are, first, total costs of the Corporation after the elimination of results
of intercompany transactions; break-down of the costs into a number
of components, that is, taxes, wages, material coste and depreciation,
and a measurement of the year-to-year changes in these cost factors.
Closely associated with this new break-down of cost figures is the
additional information furnished on income from operations; also
adjusted for inter-company transactions. In the field of prices, the
outstanding contribution in our judgment is the making available of
data on actual mill nets, hitherto, of course, regarded as roughly
corresponding to prices. There has been a long, and I think rather
sterile, controversy concerning the degree to which these pubhshed
prices actually represented prices paid. The new materials on the
miU nets received by subsidiaries of the 'United States Steel Corpora
tion give this information not only for steel in general, but for various
types of steel. This goes far beyond any previously published data
and should make unnecessary extended continuation of much of the
dispute concerning the significance of published prices.
Similarly in the field of labor, there are new break-downs of wages
and employment, and new information extending over a number of
years on hours, wage rates and weekly earnings of employees.
I should like to have it distinctly understood that what we now
propose is nothing else than a cooperative examination and an explora-
tion of this new and vital information. Whatever differences there
are, are largely differences of evaluation. They can be classified into
three groups: First, we question, or rather, we want to examine the
adequacy of some of the assumptions. Part of that has already come
out in this morning's discussion. Our next witness. Dr. deChazeau,
will dwell on that topic.
It is important constantly to keep in mind that the mathematical
techniques used depend upon and are determined by the economic
hypotheses, or economic assumptions, which one makes. For
mathematics is just a hopper which grinds out more or less finely what
one puts in We beheve that if one accepts Dr. Yntema's arbitrary
and limitr'l economic assumptions as being adequate and valid for
the commiitee, that is, if the committee does not consider any other
assumptions but those, then little substantial modification is possible
in the presentation by Dr. Yntema as given to the committee.
Secondly, we shall want to examine various aspects of certain of
the cost items. Finally, there are a few things that should be pointed
out about the correlation techniques. But I hope that nothing that
shall be presented by, the witnesses who are to come will in any way
detract from this tribute which we collectively wish to make to the
Steel Corporation and to Dr. Yntema. We will call Dr. Melvin
deChazeau this afternoon.
Mr. Wooden. I have a few questions I would like to ask of Dr.
Yntema. May that be postponed until the first thing this afternoon?
Acting Chairman O'Connell. Dr. Elreps, had you intended recall-
ing Dr. Yntema?^
13616 CONCENTRATION OF ECONOMIC POWER
Dr. Keeps. Dr. Yntema will be recalled, because it is our proposal,
Mr.- Chairman, after Dr. deChazeau has presented his statement, to
ask Dr. Yntema to make comments. In fact, it is our proposal that
such a procedure be followed for each of the witnesses. This is a
cooperative exploration. There are large areas of debatability and
legitimate differences of opinion. It is difficult to appraise new infor-
mation. Pioneering efforts are always subject to reevaluation. We
would really like to know what it is we have on our hands.
Acting Chairman O'Connell. Mr. Wooden, would you prefer to
have Dr. Yntema recalled immediately after lunch for your questions?
Mr. Wooden. It makes no difference.
Acting Chairman O'Connell, Then since we are going to have Dr.
Yntema a little later on, let's recess and hear Dr. deChazeau after
lunch. We stand in recess Imtil 2:30.
(Whereupon, at 12:40 p. m., a recess was taken until 2:30 p. m. of
the same day.)
AFTERNOON SESSION
(The hearing was resumed at 2:35 p. m., on the expiration of the
recess.)
Acting Chairman O'Connell. The committee will please be in
order.
Dr. KJreps,
Dr. Kreps. Mr. Chairman, before we resume the steel hearings, I
should like to finish the presentation of the cartel hearings, because due
to the death of Senator Borah, the senior senatorial member of the
Temporary National Economic Committee, its cartel hearings \v"rr
ended before it was possible to place the last witness. Dr. Rudolf
Callmann on the stand. Dr. Callmann, now residing at 23 Hammond
Street, Cambridge, Mass., is an internationally recognized authority
on cartel problems.
For the 10-year period just prior to 1936, when he was attracted to
the United States, he was Rechtsanwalt am Landgericht in Cologne,
Germany, engaged in legal consulting practice for a variety of German
catiels. Prior to that time, for a period of 4 years, he was managing
director of the firm, Rollmann & Mayer, a shoe manufacturing concern
at Cologne. In addition, he is the author of a number of authoritative
writings on cartel and related problems, most important of which are
his volumes on unfair competition, entitled "Der Unlautere Wett-
bewcrg (J. Bensheimer, Mannheim-Berlin-Leipzig, 1932), 670 pages,
and his -treatise on German cartel law entitled, "Das Deutsche
Kartelh-ect" (Philo Verlag und Buchhandlung GMBG, Berlin, 1934),
721 pages.
I should like to submit his statement, prepared and sworn to* by
him, for the record.
Acting Chairman O'Connell.. It will be included in the record of
the cartel hearings, witlbout objection.
(Dr. Callmann's prepared statement on "Cartels" appears in Hear-
ings, Part 25, p. 13347 et seq.).
Dr. Kreps. The first witness for the T. N. E. C. portion of the
hearings,' whom I should like to summon, is Dr. deChazeau.
' Hearings on the steel industry, included in Hearings, Parts 26 and 27) were presented fcr 'he staff of the
Temporary National Economic Committee, by the Department of Justice, and by the Federal Trade
Commission.
CONCENTRATION OF ECONOMIC POWER 13617
(Senator King assumed the Chair.)
Acting Chairman King. Have you been sworn?
Dr. deChazeau. Yes, sir.
TESTIMONY OF DR. MELVIN G. deCHAZEAU, PROFESSOR AT THE
UNIVERSITY OF VIRGINIA, CHARLOTTESVILLE, VA.— Resumed '
Dr. Kreps. Dr. deChazeau, for the purpose of the record, will you
state your full name, please?
Dr. deChazeau. Melvin G. deChazeau.
Dr. Keeps. And you are now on the staff of the University of
Virginia?
Dr. deChazeau. Yes; I am a professor there.
Dr. Kreps. How long have you been studying problems of the steel
industry?
Dr. deChazeau. Beginning in 1934, at which time I joined the
staff of the Bureau of Business Research of the University of Pitts-
burgh for a study of the steel industry.
Dr. Kreps. You are co-author of two volumes entitled "Economics
of the Iron and Steel Industry," published by McGraw-HUl Book Co.?
Dr. deChazeau. I am.
Dr. Kreps. In doing the research required, did you have a staff
at your disposal.
Dr. deChazeau. Yes.
Dr. Kreps. Did you make field trips and did you go through
plants and did you consult wi'th a number of steel executives?
Dr. deChazeau. We did.
Dr. Kreps, That, I take it, is stated in fuU in this rather lengthy
preface under the heading, "Obligations to Members of the Steel
Industry"?
Dr. deChazeau. Yes.
Dr. Kreps. You have a statement which you have prepared for
the Temporary National Economic Committee on the material
submitted this morning by the United States Steel Corporation?
Dr. deChazeau. I have.
Dr. Kreps. Will you please give that statement?
Dr. deChazeau. I would like to preface my remarks with a general
observation. I do not consider myself qualified as a statistician to
criticize in detaU the technique employed by Professor Yntema.
That wiU be reserved for Dr. Bean and for Dr. Ezekiel. My remarks,
therefore, are in the nature of — if I may presume to say so — attempt-
ing to present a point of view for the evaluation and significance of
these results.
In the course of my paper, I refer to some possible technical objec-
tions wliich will be developed more fully later.
the corporation's analysis of cost in relation to volume
Dr. deChazeau. The United States Steel Corporation through its
subsidiaries is the most highly integrated steel corporation in the
industry. It has always been supposed that the more thoroughly
integrated any organization became — that is, the more completely it
1 Dr. deOhazeau's previous testimony on the iron and steel industry appears ip Hearings, Part 19.
13618 CONCENTRATION OF ECONOMIC POWER
insulated the various stages of production and distribution from the
mar-ket — the larger would be its relatively fixed costs in comparison
with its variable costs.
This result was expected to follow not only because of heavy capital
investment and the necessary size of the corporate structure but also
because an integrated structure precluded the transformation of fixed
into variable costs which open market trading at each level of pro-
duction provides. The Corporation's analysis of costs and volume
is startlmg, therefore, in its apparent demonstration, both that
marginal or differential costs are constant over all observable rates
of output up to practical capacity and that variable costs per ton are
so high relative to prices received that the possibility of price re-
duction without out-of-pocket loss is of negligible significance.
I should add here that I am not so much surprised that the variable
costs are found constant in the steel industry as I am surprised at the
level of those variable costs. That is, in an industry in which increases
in output take place not through varying the rate at which equipment
is used but rather in bringing new items of equipment into use, one
would expect a certain uniformity in variable costs with increases in
output. The thing that does startle me, and the thing which is crucial,
it seems to me, in this analysis, is the level of the variable cost with
relation to the fixed cost.
This conclusion is so important for pubUc pohcy that it must be
examined very critically and exhaustively before it can be accepted
at its face value.
Acting Chairman King. You 4on't mean, do you Professor, that
business organizations, whether they are integrated, or rather, decen-
tralized, don't know what their costs are?
Dr. deChazeau. Oh, no.
Acting Chairman King. They have their balance sheets; they know
what the cost of materials are which they purchase?
Dr. deChazeau. Yes.
Actmg Chairman King. They know what they have received for
their sales?
Dr. deChazeau. Yes.
Acting Chairman King. They know what their wages are, and at
the end of the year, a balance sheet is prepared and they know their
losses or their gains. Is that right?
Dr. deChazeaq-. That is right.
Acting Chairman King. And in periods of depression or ups and
downs in the business world, obviously there must be ridges and
mountains and valleys in the activities and the costs and in the profits
and losses of corporations largely integrated? That is true, isn't it?
Dr. deChazeau. Yes. The vital tiling here. Senator, is not so much
the amoimt of cost but the variations of cost, that is, the distinction
between the overhead cost or fixed cost, and those costs which vary
with output — that is {he thing which is significant for price policy.
If the level of the curves shown for total costs were lower — that is,
less steep — the total variable cost would be smaller and the elasticity
of demand which would make profitable a reduction in price could be
much less. Taking the costs as they are given and the figures as
given, one would have to have a demand elasticity of close to four
before it would be profitable to reduce prices on the basis of the cost
discussion ^vren
CONCENTRATION OF ECONOMIC POWER 13619
An elasticity of four, as you know, means that with a given price
reduction the quantity taken would be four times what it was before.
Such an elasticity is beyond that conceived of by any students of
the steel industry, that or perhaps any other industry.
For that reason, it is vital. That is the point to which I refer.
Acting Chairman King. I had always supposed that businessmen,
whether they are running merchandising establishments or steel com-
panies or railroads or automobile plants, knew what their profits were
and knew almost daily what their expenditures were, loiew whether
they were making money or whether they were losing money, and then
with their bookkeeping systems plus the experts wliich they had, busi-
nessmen of abihty, who had been with many of the institutions for
years, would be enabled to evaluate their activities and determine a
proper allocation of all of the factors that were involved in expendi-
tures, profits, and losses.
Dr. deChazeau. Except for the word "allocation," I should agree
with you.
Mr. HiNRicHS. Well, the word "allocation" is very vital there, is
it not?
Dr. deChazeau. Yes.
Mr. HiNRiCHS. That is, you don't question the fact that the figures
of cost as given are accurate with reference to total figures?
Dr. deChazeau. No.
Mr. HiNRicHS. But the point is that no business organization can
know with absolute precision what the distinction between variable
and fixed costs is? Those could and will be established through
accounting conventions, in which judgment is a very important factor,
and it is on that question of allocation that you are raising questions,
not with reference to the aggregate figures of expenditure or the aggre-
gate figures of profit and loss; is that correct?
Dr. deChazeau. That is right, with one addition, if I may add it,
Mr. Hinrichs. One presumably has a variation of cost here with
changes in rate of utihzation of capacity, which is derived from an
historical study of total costs through the application of the correla-
tion method. In the application of that method, there may be
difficulties which thi'ow doubt on the significance of the result from
the point of view of how costs do in fact vary with actual changes in
utihzation.
Acting Chairman' King. Proceed.
Dr. deChazeau. ' The statistical method employed in this analysis
of fixed and variable components of total costs involves (a) a classi-
fication of total expenses, (6) an adjustment of each category of
expense to 1938 conditions, and (c) an adjustment of volume of out-
put by weighting the tonnage of each class of products sold by the
ratio of its average mill costs to the average mill costs of all rolled and
finished steel products during the period of 1933 through 1937.
With the exception of pay-roll and "other expense" items which were
further adjusted for time trend to correct for changes in efficiency, the
fixed and variable components in each expense category were then
ascertained by plotting annual adjusted expense against weighted
tonnage sold in a scatter diagram, fitting a regression line, and extra-
polating that fine to the base line.
There is always possible error in the projection of a total expense
function derived fcom an analysis of historical cost data. The shape
13620 CONCENTRATION OF ECONOMIC POWER
of the cost function at levels of output below those actually experienced
may be different from that within the range of observations. This is
of slight importance here since the low rate was 18 percent of capacity,
very close to the absolute minimum. In the present case, although
the range of observations is wide, the number of observations, especially
in the medium range of output (i. e., in the general region of 8 to 10
million weighted tons), which is crucial to the analysis, is very limited
or absent — I say crucial because the break-even point is around 8)^
million tons. This fact gravely restricts the reliability of the con-
clusion that the total cost function is linear and renders the probable
error in an extrapolation of the regression hne to determine the
fixed component of costs extremely high. Fm-thermore, a relatively
shght change in the slope of the regression line can make a substantial
change in the apparent size of fixed and variable costs. It is possible,
therefore, that the actual overhead expense of the Corporation is
greater than that calculated by the statistical method employed.
Neglecting this possibility, however, it is apparent that the character
of the total cost function and the relative magnitude of fixed and
variable components of cost depend on (a) the dependence of actual
expenses in a given year on the volume of sales in that year; (6) the
reasonableness of the adjustment to 1938 conditions; and (c) the
adequacy of the weights employed to obtain a homogeneous single
output series. Finally, the significance of the result for pricing policy'
depends on the applicability of this method of cost analysis to a
situation in which multiple plants are employed, multiple products
manufactured, and dynamic conditions of technology and capacity
obtain.
The data made available in this monograph will not permit an ex-
haustive or conclusive evaluation of the results from the points of
view just enumerated. Criticisms of the data analyzed, the adjust-
ment of data, and especially the weighting of tons are important pri-
marily because of their cumulative rather than their individual effect.
Because of the limited number of observations a relatively slight
change in the location of points in the scatter diagrams might render
the cost function curvilinear rather than linear. The most important
limitation on this study, however, is the narrow significance that may
rightly be accorded it for the purposes of pricing policy.
Mr. HiNRiCHS. May I interrupt just a second there? If I under-
stand what you have just been saying, it is that on the curves that we
saw this morning, what you call the regression line here was that dotted
line that passed more or less through the points?
Dr. deChazeau. That is right.
Mr. HiNRiCHS. And that in order to arrive at a figure of $162,000,-
000 as the total of fixed costs
Dr. deChazeau. One hundred and eighty-two million dollars.
Mr. HiNRicHS. It was necessary to extend that line back to the
theoretical point of what costs woidd be even if there were no ])roduc-
tion. Mr. Yntema made reservations this morning and wouldn't say
that those costs actually >./ould prevail at zero production, but that
for practical purposes it was necessoiy to extend that line back to the
idea of a zero base line; that Mr. Yntema this morning said, if I re-
member, correctly, that he didn't attach much importance to the
extrap'.lation, but in fact it has a very real significance to the points
that he was making because it is only when it is extended back that
CONCENTRATION OF E<^ONOMIC POWER 13621
that figure of $182,000,000 comes out of his figures. Now anything
that happens to the slope of that hne, even if it involves a relatively
small shift one way or another, might very well make that figure
$200,000,000 or $150,000,000, and that you are directing your atten-
tion, therefore, to the question of comparatively minor differences,
possibly within the range of observed fact, but trying to see how vari-
ations in this area of observation would affect the guess that you have
to make as to what the total volume of fixed costs is back there on the
line of zero production. Is that correct?
Dr. deChazeau. That is true so far as the point with relation to
extrapolation is concerned.
Mr. HiNRiCHS. And if you were to have, instead of a straight line,
a line which curved, it would be characteristic of a curve that it would
drop away very fast as it came back to that base line and flatten off
as you went out at the top. That is also correct?
Dr. deChazeau. Yes; and of course if your total costs described a
curve instead of a straight line, that would affect your conclusion that
the variable costs were uniform.
Acting Chairman King. Is it your contention that when the books
show a deficit there during a period of $182,000,000, the rnine owners
didn't know what they were doing, didn't know anything about it,
didn't know what it represented, and what caused it?
Dr. deChazeau. Senator, my point has nothing to do with profits
or losses. What I am interested in is whether the variable costs are
in fact uniform, and, second, how large they are with relation to the
total costs.
Actmg Chairman King. When you speak of variable costs, do you
mean, taking the industry now under consideration, the cost of iron
ore?
Dr. deChazeau. I mean the additional costs which are associated
with additions to output, which is the variable costs as Dr. Yntema
discussed them this morning.
Acting Chairman King. Well, the costs of operations vary Jrom
day to day and from month to month and from year to year ji any
industry, don't they?
Dr. deChazeau. Quite right.
Acting Chairman King. The cost of your ore one year may be
considerably different from the costs of ore for a preceding year, your
labor costs vary. Those are variable, aren't they?
Dr. deChazeau.' Yes, sir.^
Acting Chairman King. And there are many factots incident to
the determiiiation of your outgo and your income that are variable
market conditions, labor conditions, cost of raw materials, and the
prices of finished products all go into the sum total of your conditions
with a view to ascertaining just what the situation is and what the
condition of your business is.
£)r. deChazeau. It is all of those conditions which Dr. Yntema
v/as analyzing, eliminating variations in price by adjusting to 1938
conditions.
Acting Chairman King. You are not attempting to show that
those figures which were given as to the costs, losses, and so on, were
inaccurate?
Dr. deChazeau. In no sense.
13622 CONCENTRATION OF ECONOMIC POWER
Acting Chairman King. Yours is a sort of scientific technological
discussion of things which the practical man doesn't know anything
about.
Dr. deChazeau. I hope that that is not true, Senator.
Dr. Kreps. May I point out, Senator, that this line which sum-
marizes the experience of United States Steel does not extend below
slightly over 4,000,000 tons. There is nothing in the experience of
the Corporation which indicates what their fixed costs are ^elow that
level. Therefore, the $182,000,000 figure which they arrive at is a
guess, and a guess arrived at by extrapolation. That guess represents
no common sense experience. Yet if fixed costs are not relatively
small, and in particular if variable costs, instead of rising steeply,
rise slowly, then the whole of Dr. Yntema's further analysis falls by
the wayside.
Acting Chairman King. Is that testimony that you are giving now,
Doctor, or comment upon the testiinony?^
Dr. Kreps. I want to clarify the issue in this case.
Acting Chairman King. It may be clarified, but I haven't been
clarified. Ptoceed.
Dr beChazeau. Senator, I wonder if I could clarify the situation
by merely indicating this. If the points in the scatter diagram from
which the regression line was derived, were shifted slightly to the
right by change in the weighting in the latter years and slightly to the
left in the earlier years, that would have the effect of requiring a fitted
line to those points which would rise much less rapidly and which
would cut the base line at a higher fixed cost level. The result would
be that your figure for variable costs would be less than it is here
estimated. Hence the position of the points in the scatter diagram
is very important for the analysis, because as the variable costs fall
the elasticity of demand required to make it profitable to reduce price
is much less. That is the only point here considered. I am in no
way challenging the figures used by the Steel Corporation.
The relation of recorded expense to volume of sales may reflect
managerial policy rather than actual cost and thereby exaggerate the
apparent magnitude of variable costs.
The arbitrary character of accounting cost allocations with particu-
lar reference to this analj^sis will be discussed before this committee
by Mr. Martin Taitel and need not be examined in detail bj^ me.
The tendency to allocate costs co years in which there are receipts
suflScient to cover them is well known. To the extent that these
charges are excessive in good years and less than "true" costs in bad
years, a bias will be given to any total cost function derived from an
historical series which tends to overstate marginal with relation to
fixed costs and may impart an erroneous linearity to the function
itself. Possible examples are bonuses to executives associated with
with changes in the level of sales; depreciation (as well as debits to
other reserves) ; the purchase of supplies, materials for repair, and so
forth. The point is important primarily because of the size of the
pay roll and the "other expense" items. These two items consti-
tuted, during the period analyzed, almost 84 percent of aggregate
unadjusted expenses before income taxes, while "other expenses"
alone (a lump item without any breakdown in the analysis), accounted
for roughly 38 percent of the aggregate. These items so dominate
CONCENTRATION OF ECONOMIC POWER 13623
the final results that the con elusions can hardly be admitted without
more detailed analysis of the components which went into them.
Probable error in the adjustment of pay rolls and "other expense"
to 1938 conditions, because of the dominating importance of these
items, is sufficient to throw doubt on both the Imear character of the
total cost fimction and the magnitude of the variable costs.
Two adjustments were made in pay roll data: Average hourly
earnings were raised in each year of the period to the average for
1938 ($0,902) aud the adjusted pay roll was then corrected for time
trend to give effect to increasing productivity. But the first adjust-
ment assumes that the same composition of skills is used a't every scale
of output. The proportion of skilled persons employed, however, is
likely to be greater at lower rates of output. This may be the reason
why the Corporation shows approximately the same average hourly
earnings in 1931 as in earlier years despite reduced rates in 1931 and
lower average hourly earnings m. 1933 than in 1932 despite higher rates.
The importance of this factor cannot be ascertained a priori. Since
most skilled workers in a steel plant work on piecework with a guaran-
teed hourly minimum, the effect of variations in the composition of
the staff on average hourly earnings is modified by the variation in
earnings of skdled workers with variations in output.
With reference to this point, information which the Steel Corpora-
tion made available just last night, in which they tested hourly earn-
ings with relation to output over a period from April 1937 through
the current month of 39, indicates that there is not much variation in
hourly earnings with changes in output. I am not at this time
prepared either to criticize or to accept without reservation those
particular results. The method of adjustment employed, however,
tends to increase the slope of the regression line and therefore to raise
the apparent variable costs.
The adjustment for time trend (table 20 and chart 8 in "Exhibit
No. 1416"^) seems particularly unconvincing. Instead of a single
line of regression, three appear to be mdicated. From 1927 through
1929, although total pay roll remained about the same, output
increased, and, contrary to expectations, if variable costs were in fact
constant, the "trend in residuals" showed an increase in efficiency of
about 15 percent. From 1930 through 1033, a separate regression
curve is indicated. Whether by reason of faulty. wage adjustment for
changes in the composition of the labor force or because of techno-
logical changes, an increose of efficiency of over 10 percent is shown.
Fmally, a thu-d luae from 1934 through 1937, while output was
expanding and technological changes were being made, suggests
another gain of over 15 percent. These gains were not cumulative
throughout the period. In the absence of a break-down in pay-roll
figures and of a continuous trend of residuals during the period, the
conclusion that pay-roll data conform to a linear function is far from
certain. ' .
The adjustment of "other expenses," without knowledge of its com-
ponent i^ems, is even more suspect. Since its' magnitude is com-
parable to that for pay rolls, the indicated increase in such adjusted
expenses o\'er the period by 10.47 percent of average implies that the
gains of labor-saving technology (equivalent to 14.41 percent of
average pay roU) were largely nullified by such adjusted "costs."
' Appendix, pp. 14051 and 14052.
13624 CONCENTRATION OP ECONOMIC I'OWER
The assumptions that must be made to justify the weighted tons
employed in the analysis are so improbable as to throw doubt on the
conclusions derived.
Co^ts, for the purpose of this analysis (except for nonoperating
income and expense), cover all operations of the Corporation's sub-
sidiaries of whatever nature while sales are represented by weighted
tons of all products shipped (except cement and certain liquid and
gaseous coke-oven byproducts) — somewhat less than the full volume
of business represented in costs by the amount of goods and services
sold which are not measured in tons. This conglomerate is reduced to
a "homogeneous" number of equivalent average mill-cost units
called "weighted tons" by correcting the actual tons of each product
shipped in each year by the ratio of its average mill costs to the
average mill costs of all rolled and finished steel products over the
period 1933-37. No data are presented which would enable one to
appraise (a) the comparability of the cost items included in mill costs
for different products; (b) the representativeness of the average for
each product in terms of the range of mill costs at a given plant over
time or between plants at a given time ; or (c) the stability of the stand-
ard adopted, that is, the average of mill, costs for all rolled and finished
products at all plants over the 5-year period.
Assuming the propriety of the mill-cost averages, however, it is
necessary to assume also thr t the ratio of the average mill cost of each
product to the average mill ;ost of all rolled and finished steel products
during the sample period 1 33-37, was constant^throughout the period
analysed, 1927-38, inclusive. This is equivalent to an assumption
that no technological improvements took place in one department or
geographic area that did not take place in all departments or geo-
graphic areas. That this was not true as between steel and nonsteel
products may be discounted because of the relatively small ratio of
weighted tons of nonsteel products in total weighted tons. But the
same can hardly be said for hot and cold rolled light steel products
like strip, sheet, and tin plate which constituted a substantial and
apparently increasing percentage of the total tonnage of rolled and
finished steel shipped during the period. If, as seems likely, there
has been a downward trend in the mill costs of such light flat-rolled
products relative to the average of all rolled and finished steel products
over the period studied, the weights actually employed would reduce
the weighted tons in earlier years.
The range of actual adjustment is fairly large — the spread of the
correction factors being about 10 percent. Practically all of the high
volume years are corrected downward while the low volume years are
corrected upward. This results from the relatively greater proportion
of high^alue products like sheet, strip, and tin plate shipped in low
volume years. Were correction made for a downward trend in mill
costs for flat-rolled products the points in the scatter diagram would
be shifted to the right especially for the earlier years of the period.
The net effect on the slope or on the linear character of the regression
line cannot be determined for lack of data. Admittedly, however, a
substantial change in weights would be required to make a significant
change in results.
CONCENTRATION OF ECONOMIC POWER 13625
SIGNIFICANCE OF THE COST STUDY IN PRICING POLICY
Dr. deChazeau. The cost relations traced in this statement are
primarily static costs and cannot be used to measure the change in
costs that might be expected as the Corporation moves from one level
of output to another, the essential factor in pricing policy.
This, I think, is one of the most important criticisms. By correcting
cost items to 1938 levels, it was apparently desired to estimate from
an historical cost series the variation of costs with changes in the rate
of utilization of capacity at a given time. But despite substantial
changes in capacity over the period studied, no correction was made
in the weights (and therefore the output measurements) for variations
in the percent of capacity operated. The cost-output relation sought
was one applying to a specified set of capacities while the cost figures
used apply to different sets of capacities at various times. Much of
the so-called variable costs in this analysis may be found, on closer
examination, to reflect changes in capacity to produce.
Although there is reason to believe that costs behave differently
depending on the direction, and the rate of change in output level as
well as the preexisting level of output, no consideration was given to
the effect of such factors. Suffice it to say that there is some internal
evidence that such was the case with relation to pay rolls.
It was tacitly assumed that long-run and short-run costs for a given
output are the same, that is, that adaptation takes place at an ignite
rate. This result, of course, is implicit in the use of annual rather than
montlily or quarterly data. The use of annual data, undoubtedly was
forced on the Corporation by the character of the cost information
available.
The cost curve finally developed can be admitted as a limitation on
pricing policy only on assumptions which the industry has rightfully
denied and which falsify the dynamic character of costs.
The procedure employed in this study is most appropriate for a
single plant producing a single product. For many plants producing
a wide variety of products, having no inherent homogeneity, it can
provide only the roughest kind of approximation at best and it is
downright misleading at worst.
The steel industry has long taken the position (and rightly so in my
judgment) that —
(a) Costs are not comparable between plants and areas since they
vary not only with the prices of the factors but also with the com-
bination of products manufactured.
(b) Cost even for a given product at a given mill will vary from time
to time with the combination of specifications rolled.
(c) The steel industry produces a multitude of special tailor-made
steel products with widely diverse costs.
We are now asked to believe that an analysis of the total historical
expenses of the United States Steel Corporation, covering some 50,000
steel products and a multiple of nonsteel products and services ranging
from cement, coal, and iron ore to byproducts, transportation services
and construction, may be assumed to represent the cost situation that
confronts the Corporation in making a price for a given steel product.
13626 CONCENTRATION OF ECONOMIC POWER
At least I take it this is the imphcation of Dr. Yntema's remarks.
Acting Chairman King. You would expect, would you not, in fixing
the prices for their various commodities, that they would take into
account the general activities in which they are engaged?
Dr. deChazeau. Quite.
Acting Chairman King. It would be impossible to segregate one
little strand, so to speak, of the great fabric and say the cost for that
little one strand out of perhaps thousands of strands shall be so and so.
Dr. deChazeau. That is right.
Acting Chairman King. You have to take into account the entire
pattern.
Dr. deChazeau. That is right.
Acting Chairman King. And you have to take into account the
cost of the ore, the cost of sliipping it, you have, to take into account
the repair to the plants, the labor costs which change from day to day
and from month to month, the fluctuations of the market, the changes
in demand, and a multitude of factors which every businessman re-
luctantly or otherwise is compelled to meet and to adjust himself and
adapt himself and his business to those changing conditions.
Dr. deChazeau. That is right. I did assume, however. Senator,
that Dr. Yntema's remarks bad reference to pricing policy. If they
have merely reference to the fact that the Corporation makes losses,
at certain levels of output, at certain levels of price, that is a different
matter and restricts its significance considerably, but with relation to
prices it seems to me that one must of necessity consider these other
factors.
Acting Chairman King. You do not contend, do you, that every
business, whether it is steel or the grocery business, is going to have
profits entirely and in determining year in and year out the business
which he is conducting and the prices which he shall charge and all
factors, he has got to take into account losses as well as gains, good
years as well as bad years.
Dr. deChazeau. Yes.
Acting Chairman King. Dark days as well as sunshiny days.
Dr. deChazeau. I should agree that he must take those uiings into
consideration.
Sinc6 the location of the points in the scatter diagrams depends on
the number of weighted tons shipped in each year (i. e., the weight
accorded the actual tons of each product shipped) and the location
of these points determines the regression line (and, therefore, the
character of variable costs and the relative magnitude of such costs),
it must be assumed that the ratio of fixed to variable costs for the
average of all rolled and finished steel products is not only charac-
teristic for each of them but also characteristic for all other goods and
services supplied by the Corporation. But this is absurd. What are
variable and what are overhead costs for a particular product at a
particular time depend on the alternatives available to management.
An integrated plant Las capacity to roll a large variety of steel products
and a large proportion of the costs of any one is common to the rest in
the rolling and finishing operations and especially in prior processes
such as steel-making and pig-iron furnaces. \Vlien the rate of utili-
zation of capacity falls, it does not fall uniformly for all products and
all departments; and no matter how low it may fall for the plant as a
whole, there are technological obstacles to reduction in some depart-
CONCENTRATION OF ECONOMIC POWER 13627
ments below the minimum capacity of operating units (e. g., blast
furnaces). The additional costs of rolling or processing a particular
product or a particular specification of a product will vary widely at a
given plant and between plants depending on such factors as level of
output, combination of products and specifications, rate of change in
output, and so forth. In other words, at a given time, practically
all costs may be overhead so far as a given product is concerned.
Wliy doesn 't this condition, the real scope for managerial discretion
in the allocation of existing business to obtain lowest costs, reveal itself
. in the cost analysis under consideration? The use of an average, even
a weighted average, of all goods and services sold by the Corporation,
together with total expenses, is the obvious reason. A very wide
diversity in the cost characteristics for individual products may be
completely compensated in the over-all picture. In other words,
the behavior of the aggregate bears no necessary relation to that of
any one of its parts, and therefore its significance for the pricing of
any given product is indeterminate. Together with limitations on
the validity of conclusions previously outlined, this means that the
use of such a cost' analysis as a criterion for or a justification of pricing
policy cannot be accepted.
Mr. HiNRiCHS. Mr. Chairman, may I interrupt at this point?
I haven't the faintest idea what you have been talking about here.
Let me go back for just a minute and see if I can see the points that
you have covered. I thought I knew something about this too.
First of all, you say that where you have a wide variety of products,
a wide scattering of plants, the over-alL pictuje is necessarily confus-
ing, that you could make a detailed analysis for a given plant ^nd a
given product with far more certainty than you can make an overall
analysis for the Steel Corporation. Is that correct?
Dr. deChazeau. Yes.
Mr. HiNRicHS. Now, then, no one quarrels with that, least of all,
I presume, Mr. Yntema. Then you go ahead and you say that it
follows from this that you .can't draw any conclusions from average
price relationships or more particularly from aggregate cost relation-
ships to aggregate production relationships. It is necessary, is it not,
for a business enterprise that is making a sensible approach to a
question such as the question that is raised here as to general pricing
policy, to arrive at what you have described as a rough approxima-
tion of what those over-all relationships are. That would be true,
would it not?
Dr. deChazeau. Yes; I should say so.
Mr. HiNRiCHS. Do you know of any work which has been dqne in
the field of merging unlike products and costs in unlike plants which
is a more satisfactory method of arriving at that necessary approxima-
tion than the job that has been done here?
Dr. deChazeau. No
Mr. HiNRicHS (interposing). Are you criticizing this as a defective
utilization of methods which have been developed in recent years?
Dr. deChazeau. No. I am interested in whether the method as
applied to the entire Steel Corporation, deriving your relation of vari-
able cost to fixed cost from an historical price series, will in fact indicate
the extent of the fixed cost with relation to the variable cost, which is a
factor of importance, it seems to me, in' the pricing of the steel product.
124491— 41— pt. 26 4
13&28 CONCENTRATION OP ECONOMIC POWER
What I am critizicing is whether any such development will in fact
show you what that proper relationship is.
Mr. HiNRicHS. That is, your conclusion would be that the relation-
ships would be better arrived at through an intimate understanding
of price relationships in a given plant for a given product and through
a process of deductive economic reasoning rather than through a
statistical approach.
Dr. deChazeau. Or a statistical analysis of the operations of that
plant rather than a statistical analysis of the operations of the Steel
Corporation. Now I appreciate the limitations on Dr. Yntema's
work, which he himself pointed out, that he couldn't take an indi-
vidual plant and make such an analysis. I believe, however, that the
committee is interested in this study as it affects the question of
price flexibility or possible price flexibility in the industry as a whole.
It is from that point of view that I am criticizing.
Mr. HiNRiCHs. That is, as I understand not only what you are say-
ing in this paragraph, but in others, there are certain points at which
elements that were not taken into account by Dr. Yntema, in your
judgment, have given a less deep curve than the one that we had, as for
example in the general characteristic of carrying in the years of good
business a higher proportion of costs than are charged in years of bad
business?
Dr. deChazeau. Yes.
Mr. HiNRiCHs. I think there are other points where you similarly
have questioned, out of your experience, the precise steepness of the
curve. Now in this connection what you are questioning is the prob-
able error in the final result that comes out when one is using aggre-
gates. What you are saying essentially is a thing which Dr. Yntema
would agree with you in, I presume?
Dr. deChazeau. I hope.
Mr. HiNRiCHs. He would certainly be wiUing to give you the prob-
able error of his line of regression, to go technical, and what you are
saying is that that line or that $182,000,000 can in no sense of the word
be regarded as an absolutely fixed established sum, but that in the very
materials we are working with it is at best $182,000,000 more or less,
and the more or less may be very large.
The second thing that you are saying is that the evidence would be
more conclusive if it were substantiated out of a more particular
analysis of the cost operations in a particular plant engaged in making
a specific product and that the more one refines the product the less
statistical juggling one has to engage in, and the more accurate one's
final conclusions; is that correct?
Dr. deChazeau. Yes; but may I emphasize the importance of it?
Let me repeat. If this cost analysis is taken as not merely representa-
tive of the costs of the Steel Corporation, which incidentaUy is an im-
portant factor in the industry, but as representative of steel costs
generally (i. e., with relation to the size of variable costs contrasted
to fixed costs), the possibilities of price variation, of price reductions
in this industry, are practically insignificant and discussions of demand
elasticity are highly academic. It is true that Professor Yntema, for
his analysis, after deriving a figure of 0.3 to 0.4, assumes 1.0 as the
elasticity of demand for steel.
It wouldn't have made much difference if he had assumed 2.0 or
even 3.0; the result would be the same, although the magnitude of the
CX)NCENTtlATION OF ECONOMIC POWER 13629
results would be altered. Therefore, it seems to me that before one
can accept such a presentation of the costs as being the actual relation
of variable to fixed costs, one must know a little bit more than one does
at the moment.
Acting Chairman King. Dr. deChazeau, I am not a professor nor a
doctor nor a technologist, but I have some little practical appreciation
of the practical problems of life. I confess that I am not very clear
as to what you are trying to present for the consideration of the com-
mittee, but in order to bring it down to a concrete situation, let me
give you — so that I may understand whether I understand you, and
whether the matter which we are discussing is susceptible of concrete
presentation, and presentation such as the ordinary man would imder-
stand — a case of this character. How w®uld you, if you were running
the Corporation, fix your cost sheets and make your report in order to
determine these- variables and ponderables and imponderables, and
tangibles and intangibles which have been thrown at us here with
great ability?
I have in mind a corporation owning some mining properties. It
attempted to work them and failed; costs were too great; it found that
it had to "build a railroad; it had to ascertain the costs of that railroad
and its operation, and it had to build smelters. Those smelters didn't
always function properly; many of them had to be changed; new
processes, technological processes in the working out of orders were
developed, so that a proposition which started out with the expecta-
tion of costing only a few million, perhaps cost $75,000,000 before they
could make any profit at all.
Then in determining what their profits and their losses were", they
had to take into account the costs of operating the ming. That hdd
to be changed. Then, they had to take into account the cost of re-
moving the dirt and that cost a great deal. Wages changed. Then
came the various acts of Congress under which their taxes were in-
creased. Then came the question of depletion, difficulty in deter-
mining just what to allocate for depletion, what claims should be made
by reasonable depletion. There were variables there, variables tiiat
iranged many percent because some of the scientific men said you could
allow only so much for depletion, and others insisted that in view of
the fact that they were taking out the ore that capital was being
depleted.
Then there were the costs of the railroad and then the cost of the
smelters, and the wages differing; all those things; they had a nimiber
of integrated organizations all concentrating, though, m the final result
at. the end of the year when they had to write the balance shept, all
those costs and expenses and losses and profits were found there in
that balance sheet. How would you determine what to allocate to
the operation of the mine, what to determine for the smelter, the rail-
road, and so on? How are you going to do it? Is not this a practical
thing at the end of the year, the corporation says, "We have expended
$20,000,000; we have received $19,000,000. We have a deficit there
of $1,000,000." How are you going to allocate that? Would you say
they charged too much for the removal of the overhead, they charged
too much for depletion; they got too large a credit there and too large
a loss there? The company lost it or made it, and the balance sheet
showed everything they received -and everything they expended for
the year, would not that be an honest balance sheet?
13G30 CONCENTRATION OF ECONOMIC I'OWEK
Dr. deChazeau. Quite an honest balance sheet.
Acting Chairman King. Wouldn't that be the practical way that a
practical man would deal with his business affairs?
Dr. deChazeau. I know that it is bad form for one being asked a
question to reply with one, but I wonder if you considered what ele-
ments in that cost situation would be necessary elements for you if
you were fixing a price?
Acting Chairman King. Well, if I were fixing a price, as they were,
upon their cost and determining what their price should be, I would
take into account all of the things that I had expended in order to
make the copper, though part of it was the railroad, part of it the
smelter, and part of it was some other commodity; I would take all
those into account and put them into one balance to determine what
I had paid out and what my losses had been or my profits.
Then I would fix the price accordingly.
Mr. Feller. I wonder if I might attempt to clarify this? Am I
correct in understanding that Dr. Yntema has attempted to present
a formula under which it is possible to determine what the return of
the Corporation would be from changing the price of steel; in other
words by examining the elasticity of demand, the variation in demand
which results from the changes in price, and by examining the costs
of the Corporation as they appear on their books, he has attempted
to produce a formula which will tell the Steel Corporation or perhaps
the industry as a whole what sort of price changes to make in order
to make more money or in order to lose less?
Now that is a highly practical task and if the task succeeds Dr.
Yntema should receive the collective thanks and esteem of the steel
industry and a very substantial reward. Now as I understand it,
Dr. deChazeau is inquiring into the method of the constru'?tion of
that formula. Now again I understand one of the critical points in
the construction of that formula is in distinguishing between those
costs which are fixed, which do not vary with the rate of operation,
and those costs which are variable, those costl ..hich change as you
produce more or less steel.
Now may I go on and state my understanding?
Acting Chahman King. Aren't you assuming, that there is a datum
line, no change in certain activities?
Mr. Feller. Exactly, Senator.
Acting Chairman King. Every day.
Mr. Feller. What Dr. Yntema started out to do, was to try to
determine those costs which would be there if the Steel Corporation
produced only 1 ton of steel. In other words, the very rock bottom
of cost. Now the way he did that, as I understand it, subject to cor-
rection by Dr. Yntema, was to put down on a diagram the costs, to
put a diagram which indicates on the bottom the tons shipped by the
Steel Corpori\tion; on the side the total cost of the Steel Corporation;
to put down on that diagram a series of dots; each dot indicating the
particular rcj^ults for a given year. Then he took a pencil and drew
a line through those dots. Then he extended the line toward the
bottom and toward the top and where that line hit the chart he said,
"Those are the fixed costs," and he said thoy are $182,000,000, and
when Dr. Yntema comes back on the stand I should like to ask liim
whether anyone in the Steel Corporation had ever heard of that
$1 82,000,000 before he produced this study.
CONCENTRATION OF ECONOMIC POWER 13631
Now I understand that that is the situation, the issue before the
committee.
Acting Chairman King. I am afraid when professors disagree and
lawyers and so on, we will not reach any agreement. I am going to
interrupt the proceedings for a moment.
Dr. deChazeau. Since the points which have been raised pro and
con at this time will probably be discussed with Dr. Yntema, I think
it will be unnecessary for me to say any more at the moment.
If the contentions of the Corporation with regard to cost and demand
are admitted, one is forced to conclude that from any break-even
point a price reduction will bring losses and an upward price movement
will bring continuous and increasing profits.
Briefly, it is the contention of the Corporation that total costs are
linear and relatively steep, that variable costs are constant, and rela-
tively large, and that a demand elasticity as great as unity is in excess
of anytliing for which the industry may hope. The Corporation con-
cerns itself exclusively with results which might be expected with a
price reduction. But demand elasticity is equally applicable to price
increases with a corresponding decline in volume. On the assumption
that costs vary by a fixed amount per unit of putput (i. e., that the
cost function is a straight line) and that demand elasticity is unity
(i. e., that the total sum expended for steel is not affected by a change
in price), it is apparent that it would be increasingly profitable to
raise prices, disastrous to lower them; and that this situation is en-
hanced if demand is less elastic than unity (i. e., that a larger total
sum is expended on steel at a high than at a low price). The theo-
retical monopoly price would be at a point which allowed the sale of
a single ton.
That this monopoly price would ever be attained or approached Is
of course absurd. Elasticity of the curve in the vicinity of the break-
even point is no indication of its elasticity beyond the range of observed
price variations (a limitation of great importance, incidentally, on Dr.
Yntema's demand analysis). Price increases are- checked by the bar-
gaining power of large buyers (some of which are capable of producing
steel for their own requirements), by the potential substitution of
other products, by the competition of other steel companies suffering
from underutilization of capacity, by the force of public opinion in-
cluding the threat of Government intervention, and so forth.
The point is made merely to illustrate the inherent pressure toward
higher prices in the industry if the cost analysis of the corporation
be accepted either as characteristic for itself or for tlie industry. If,
as has been contended, a price cut on important business by any mem-
ber of the industry will be met immediately by his rivals whether the
price reduction is published or not published, the result is inevitable-
all sellers are worse off than before. On the other hand, any price
increase that can be made uniform throughout the industry, and
maintained, will result in gains for all. Accepting the cost data as
accurate, it is merely academic to consider whether price elasticity of
demand for steel is 0.5, 1.0, 2.0 or even 3.0. The question is inconse-
quential, for whatever the value of price elasticity within this range
the result would be altered only in degree. An elasticity far beyond
that contemplated by any student of the problem would be required
to make a policy of price reduction profitable.
13632 CONCENTRATION OF ECONOMIC POWER
UNITED STATKS STEEL CORPORATION'S ANALYSIS OF DEMAND FOR STEEL
Dr. deChazeau. I turn from a consideration of the cost study to
the study of demand. The object of Dr. Yntema's study of demand
is exceedingly narrow and its limitations must be borne in mind,
especially in any attempt to evaluate its significance, either as a guide
to industry pricmg policy or as an indication of the relative desirability
or possibility of price changes from a social point of view. The price
elasticity of demand for steel, which it purports to measure over the
period 1919-38, is defined as the percentage change in the quantity
of steel that would have been sold in a given year had the average level
of steel prices in that year been higher or lower than it actually was
by a certain percentage but everything else had been the same. It
is this last qualification which limits both the scope of the question
and the significance of the conclusion.
I wish to concern myself with three broad issues; namely, certain
technical criticisms of the measure of price elasticity of demand for
steel derived, the significance of such a concept of demand for indus-
trial pricing policies, the social problem raised by this analysis.
With regard to the first issue, I conclude that the method and data
employed have the net effect of reducing the apparent short-run
price elasticity of the demand for steel. I should point out here that
since the time is limited and since technical issues will be analyzed in
great detail by Dr. Bean, and especially because such suggestions as I
have would, in my judgment, make no substantial change in the result,
I shall pass over these points ve^^ briefly and not read my entire
manuscript.
In later analyses, a demand elasticity of unity (substantially greater
than that of 0.3 or 0.4 derived from this study) is used by Dr. Yntema
on general thigoretical grounds, there is reason to believe that the short-
run price elasticity of demand for steel is relatively low. A correction
therefore for the technical defects to be noted by me would probably
not make any important change in the ultimate conclusion. This
is the more certain if the cost analysis be accepted. The correlation
technique, however, is no more than a mathematical grist mill and
the significance of the results obtained cannot be greater than the
meaningfulness of the basic data which are subjected to this method
of treatment.
The measure of quantity sold is an ambiguous and changing aggre-
gate, the use of which tends to reduce elasticity to a minimum.
Dr. Yntema pointed this out himself, and after explaining that
individual steel products do not have the same economic importance
per unit of weight, they are not subject to the same demand conditions,
and their relative character and importance changes from year to
year, he proceeded to use an aggregate of such items.
Now the larger the aggregate employed and the more diverse the
products included in if, the smaller must be the apparent price elas-
ticity of demand on the assumption that other things are equal.
Although substitution of steel for other products and vice versa is
likely to be very small in the short run, even for individual products,
it is negligible for steel as a whole.
The measure of price change adopted tends to minimize the import-
ance of price in accounting for changes in volume.
CONCENTHIATION OF ECONOMIC POWER 13633
Dr. Ynoema used the Iron Age finished steel composite, which is
an arithmetic average of the published prices of eight steel products at
Pittsburgh. As he pointed out this morning, he also ran a later corre-
lation in which he used average mill nets for the Corporation and
found that no substantial change took place. It is important to note,
however, that the more inflexible the measure of prioe change adopted
(not so much in number of changes as in amount of change), the
greater the weight which will be given to factors other than price in
accounting for changes in the volume of sales. For example, had there
been no change whatever in the measure of price, the correlation
technique perforce would attribute all changes in volume of sales to
factors other than price.
With relation to -the use oi mill nets I merely add this point. I
accept Dt Yntema's statement that the use of the average mill nets
would not affect the results substantially. I criticize primarily the use
of an average which, unless it can be assumed that all prices move
together, will cover up the extent of the price change as it may affect
any given product, and therefore any given increase in demand. For
example, a break in sheet and strip prices around the middle of October
1939, first of $4 a ton, followed later by $2 a ton, which did not rep-
resent a general reduction in price of steel at that time, might well
have affected an increase in the demand for those steel products, but
mill net, the average mill net for all products, would show a reduction
which was much less than that for those particular products.
Acting Chairman King. Well, is there any relation between a
reduction in price and general consumption? I know of many in-
stances in which the reduction price of metal didn't make any material
increase in the consumptive demands. Does it follow as a rule that
when you reduce the price of a commodity there will be a great
increase in its consumption?
Dr. deChazeau. It does not always foUow as a rule; no.
Acting Chairman King. The exceptions are very numerous, aren't
they?
Dr. deChazeau. Very numerous.
The use of annual rather than monthly or quarterly data eliminates
the importance of the rate of price change as an independent variable
and, therefore, makes impossible a consideration of the timing of
price change. Steel prices are usually announced on a quarterly basis
although actual prices, through concessions, may be made from day to
day. Annual data eliminate seasonal factors in demand and reduce
the importance of speculative buying. But seasonal factors are more
properly eliminated statistically to leave, in the monthly or quarterly
figures, sales variations which may reflect price change. And specula-
tive buying is an integral part of the demand for steel of admitted
importance of the steel industry in the determination of price policy,
and of great social importance as well, whenever there is less than full
use of resources in the ecoiiomy. To nullify this speculative factor
in demand is to preclude an analysis of the timing of price change with
respect to consumer expectations and the demand for durable goods,
producers' expectations and the decision to invest, and changes in th6
prices of substitute and complementary goods.
By his selection of data, therefore. Dr. Yntema is precluded from
giving effect to the following factors, crucial in the concept of price
elasticity of demand :
13634 CONCENTRATION OF ECONOMIC POWER
(1) The ratio of steel product prices to the prices of substitute
materials or products in the manufacture of goods made from steel or
in consumption.
(2) The ratio of steel product prices to the prices of complementary-
goods used in the manufacture of goods made from steel.
(3) The rate of actual price change for steel products with regard
to the demand for those products.
(4) The timing of price change with regard primarily to the timing
of investment.
A fourth point which I note here I sliall pass over immediately.
Dr. Yntema worked out liis correlation for several relations, that is
one on the basis of shipments and one on the basis of bookings, and
so forth. I refer specifically to the latter. There is no convincing
r':^ason why Dr. Yntema adopts the figure of 0.3 to 0.4 as a maxi-
mum potential value of demand elasticity, rather than the figure 0.88
determined on the basis of estimated steel bookings. I shall leave
that for Dr. Yntema to discuss.
SIGNIFICANCE OF CONCEPT OF DEMAND FOR INDUSTRIAL PRICE
POLICY
Dr. deChazeau. My general conclusion is that, as a criterion of pric-
ing policy for the steel industry itself, the price elasticity of demand
measured by Dr. Yntema is inadequate.
Three important aspects of demand require separate analysis as a
basis for the pricing policy of any, seller: First, the cross elasticity of
the demand, by wliich is meant the percentage variation in quantity
of products sold by a given seller with a given percentage change in
his price on the assumption that tiiis price is not met immediately
by his rival. Second, the price elasticity of the market demand, and
third, shifts in the demand for any product at any given price. Dr.
Yntema neglects all but the second, the price elasticity of market
demand.
There is no error in neglecting cross-elasticity of demand, as 1 have
defined it, for the very obvious reason that, as has been noted many
times before this committee, no cut in price can take place among
important sellers for important business that is not immediately met
by rivals. Under those conditions, each seller must of necessity
consider, not the cross-elasticity of demand, but the elasticity of the
market demand in determining what price is desirable for him.
An average figure of price elasticity over the period 1919 through
1938 is almost certain to be erroneous as a criterion of price elasticity
of demand at any given time.
The demand for producers' goods, either raw materials or capital
goods, is a derived demand. In addition then to the degree of sub-
stitutabiUty and the cost of transfer from one material to another,
the elasticity of demahd for producers' goods at "any given time is
affected by two variables. First, if the prices of complementary
goods (i. e., labor and other materials) required along with the^ goods
in question are constant, the price elasticity of demand for that
producei-s' goods will be low — substantially less than that for the
finished product from which it was derived. Second, since the pro-
duction process is a time-consuming process (longer for capital goods
than fqr raw materials), the elasticity of demand for producers'
CONCENTRATION OF ECONOMIC POWER 13635
goods will vary with the expectations of buyers as to the potential
business situation and potential shifts in the demand for their finished
product over a future which varies in length with that of the pertinent
production process.
Since steel is primarily a producers' good, the price elasticity of
market demand will have two functional characteristics, neither of
which is given effect in Dr. Yntema's study.
(1) Price elasticity of demand for steel products is likely to vary
widelv with the amount of price change.
Small price changes, the only changefe measured by Dr. Yntema,
may have little or no effect on quantities pm-chased while large
changes may cause substantial variations. This is perhaps an un-
avoidable defect of the correlation method of estimating demand elas-
ticity from historical price series where price changes are small. It
was admitted as a possible qualification to his analysis by Dr. Yntema
(see "Exhibit No. 1411"). To my mind the error is graver than he
appreciates. He analyzes demand as though it were a continuous
function of price, that is, as though increases of demand occurred
with very small changes in the price. Even if so, it is hazardous to
project demand elasticity beyond the range of observed price changes.
But the considerations already noted suggest that, for producers'
goods, the demand is more likely to be a discontinuous function of
price. This means that with a small change in price, no change
takes place in quantity sold, but with a large change, you may get a
substantial increase in output.
Mr. HiNRicHs. What you mean there is that the increase in con-
sumption with the 20-percent reduction in price might be more than
twice as great as the increase in consumption with the 10-percent
reduction. You don't mean that Dr. Yntema has omitted any ob-
servations but that the changes in the price of steel in the past have
been so narrow as to restrict this study to the effect of comparatively
small changes in price?
Dr. deChazeau. That is perfectly right. Please bear in mind that
I am not criticizing Dr. Yntema as a statisticiaji. I have the greatest
admiration for Dr. Yntema's work. What I am calling attention to
are certain limitations as to the significance of the demand elasticity
as determined by the correlation method. That is my only objective.
(2) Price elasticity of demand for steel products is likely to vary
substantially from one stage of the business cycle to another, that is,
from one level of price for complementary goods and from one level
of business expectations to another.
The time of price change and the responsiveness of the price of
steel to other factors in the total market situation cannot be ignored
without invalidating the measure of demand elasticity derived. It
seems almost self-evident that no businessman could neglect with
impunity the importance of the timing of his price changes. By the
same token, any average measure of demand elasticity which abstracts
from it must prove an erroneous criterion of pricing policy. Such an
average demand elasticity over the period 1919 to 1938 has been
derived here by the correlation method. That is, actual changes in
sales were correlated with actual changes in the finished composite
price, and the demand curve was derived from it.
Mr. HiNRiCHS. May I interrupt with a second question on this
subiect? When you are talking about the relationship between the
J 3636 CONCENTRATION OF ECONOMIC POWER
changes in the price of steel and the use of steel and pointing out
that steel is a producer's good, very largely, and that the change in
the price of steel is only going to effect demand significantly if it is
passed on in the final product, you are distinguishing, are you not,
between two different points of view, with reference to which this
problem of price flexibility might be approached. From the business-
man's approach as to what he can reasonably expect to have happen
and how he ought to behave in ihe face of that, insofar as he is en-
gaged in making polic;^ the fact that a change in the price of steel is
not necessarily passed on, is a very important factor for the consider-
ation of the United States Steel Corporation as a business enterprise.
What you are opening up in that suggestion is the further suggestion
that there is a very real interest by the community at large in general
matters of pricing policy, that a change in the price of steel alone
would be likely to have a very insignificant effect, that changes in the
price of steel plus changes in the price of other products if reduced
costs of steel were fully passed on, might have a very much larger
effect that has been the case in the past, when very frequently those
changes in price were not passed on in a reduction in the price of
finished goods.
Is that a correct interpretation of the limitation that you drew
earlier in your discussion there?
Dr deChazeau. That is a point which I planned to make.
Dr. HiNRicHS. I'm sorry.
Dr. deChazeau. You have gone a little beyond, I thinly, any point
that I have made as yet. My main point here, if I may interpolate,
is merely this: That since the demand for producers' goods is likely
to vary with profit expectations, and since also insofar as steel is
a raw material which may be substituted for other materials, the cost
of transfer and substitutability may be involved, you might expect
two conclusions; first, that the demand elasticity would be very low
in the short run for a small change in price, but it might be much
greater with a larger change in price; and, second, that the demand
elasticity is likely to vary from one stage of the business cj^cle, and
therefore business expectations, to another stage.
That is the onlj^ point which I am making here, if you take a single
static — that is, "normal" — concept of demand elasticity throughout
the entire period of 1919 to 1938, it seems to me that theoretically
you must have an erroneous picture of demand with reference to
potential price changes for any given state of facts, even though, as
an average, it may have some meaning.
Acting Chairman King. When your commodity perhaps may be
subjected to competition from abroad, that is a factor, tangible or
intangible, affecting the question of elasticity, is that not true?
Dr. deChazeau. Senator, I should say it would not affect the
(juestion of elasticity but it will affect the question of what price you
charge.
Acting Cliairman Ivino. You relate elasticitj^ to prices, don't you
directly or indirectly?
Dr. deChazeau. Yes; but elasticity is a measure of the variation
in the proportion of steel which will be taken with a given propor-
tionate change in its price.
Noiv, imports from abroad or price competition from abroad may
have I ho elFert of foicing the Steel Corporation or forcing domestic
CONCENTRATION OF ECONOMIC POWER 13637
producers to reduce their prices irrespective of the elasticity of
demand.
The assumption that other things are equal, a necessary assump-
tion for the derivation of a statistical measure of price elasticity from
a time series by the correlation technique, is false and vitiates the
conclusion as a measure of the effect of a price change in a dynamic
situation.
When I say it is false, I mean as an actual market phenomenon,
one can change that assumption but it narrows the significance of
your conclusions.
As indicated a moment ago, the price elasticity of demand for a
steel product is likely to vary from one state of facts to another;
that is, either within the so-called business cycle or over longer periods
of time. More important for pricing pohcy than price elasticity at
any given time are shifts in the entire demand curve for the product.
That such shifts may take place in the absence of price change or
despite a price change for a particular product does not mean that
they may be neglected in the determination of pricing pohcy. Dr.
Yntema's analysis of price elasticity does neglect the impact of price
on shifts in the demand curve.
Dr. Kreps. Would you explain more clearly what you mean by
shifts in the. demand curve?
Dr. deChazeau. Yes; a shift in the demand curve represents an
increase in the quantity that will be taken at the same price; that is,
a change in the economic conditions, either by reason of a change in
profit expectations or by reason of a change in the general cost picture,
which will increase the demand for the product without a change in
its price; whereas demand elasticity is of necessity a measure of rela-
tive changes in quantity to be taken, with relative changes in price,
and therefore implicitly assumes that the conditions remain the same.
Dr. Kreps. To state what you have to say a little more clearly or
fully, Dr. Yntema's analysis assumes that at a given price or for an
average of prices for steel products, the same amount would be taken
throughout the period which he covers, namely, 1929, 1932, 1937, and
Dr. deChazeau. That is right. That is, it seems to me that in the
use of the demand elasticity figure which he has developed, which is
derived from the study of an historical series from 1919 to 1938, in
the use of that curvfe, he must assume that irrespective of the changes
in conditions, the Actual demand change with relation to the price
change would have been the same in any year.
Dr. Kreps. Do you regard that as probably true ,to fact, that at
any given price, the same amount of steel could have been sold both
m 1929, 1932, 1937, and 1938, or is that contrary to fkct?
Dr. deChazeau. Well, I clearly regard it as erroneous, as I have
just stated.
Mr. HiNRicHS. Pardon me, but did you state yourself correctly
there, that the same quantity of steel would be sold at the same price
in any one of these years, or did you mean to say that Dr. Yntema's
analysis indicates that a 10-percent reduction from the prices which
prevailed in 1932 would have tended to increase consumption in 1932
by 3 or 4 percent. That is, a reduction of 10 percent from whatever
price prevailed in 1932 when they were selling four and a half million
tons, would have yielded another 3 or 4 percent above the four and
13638 CONCENTRATION OF ECONOMIC POWER
a half million tons. Similarly, prices 10 percent lower than those
which prevailed in 1937 would have tended to produce 10 percent
greater consumption than the 13,000,000 tons that were sold in 1937?
It was that 10 percent related to 3 percent that you spoke of as being
constant?
Dr. deChazeau. It is the elasticity relationship.
Mr. HiNRiCHS. Would that be constant year after year as this is
presented? You didn't mean to say that at $75 a ton for a given
product, the same quantity would be sold in 1932 and 1937?
Dr. deChazeau. No; because obviously he points out the effect of
industrial conditions upon the total demands for steel as being a very
important factor.
Mr. HiNRicHS. Your point is merely then that there is logical reason
to believe that the effect of the 10 percent price reduction in one
phase of the business cycle may be very different?
Dr. deChazeau. Very different.
Mr. HiNRiCHS. Than the effect of a 10 percent price reduction in
another phase of the business cycle?
Dr. deChazeau. That is my point.
Mr. Feller. Can your point be restated in this way, not that Dr.
Yntema has overlooked the fact that there are changes in underlying
business conditions, but that he has derived the theoretical measure
of elasticity which can apply only if all other things are equal, and
all other things are never equal?
Dr. deChazeau. Yes; as I see it for the purpose of application of
the correlation technique to a historical series, with the object of
deriving a price elasticity of demand, you must assume that other
things are equal, and he has explicitly made that assumption. I have
no doubt whatsoever that Dr. Yntema would agree that conditions
are not equal from one time to another.
Now^ as he pointed out this morning, it was his own judgment
that although secondary repercussions of price change would inisrease
demand, such a possible shift in demand was far more than compen-
sated by his assumption of a unitary elasticity of demand, some three
times greater than the 0.3 to 0.4 elasticity which his study indicated.
Acting Chairman King. I don't quite understand. Professor
[laughter], how in this changing world, in this rather confused political
and economic and industrial situation, you can predicate any view
that prices will be the same tomorrow or the next day in any industry
or in relation to any product. I cannot conceive of a formula that
would compel or produce or result in a straight line, if I may use that
expression, of prices with respect to any particular product, and if Dr.
Yntema attempted to convey the idea that there was a formula under
the terms of which, in this changing economic situation, there would
be a constancy of prices, I did not understand it and I would not
agree with him, if that was his view.
Dr. deChazeau. Neither did I understand that to be his view.
What he was trying to indicate was the cost situation, on the one hand,
and the demand situation, on the other, which confronted the United
States Steel Corporation in making a price or in determining whether
a price reduction was desirable.
Acting Chairman King. Well, the cost situation can change as well
as the demand can change, can it not — if not there, at least frequently?
CONCENTRATION OF ECONOMIC POWER 13639
Dr. deChazeau. It changes in a stable way, namely, that the
variable cost is a constant and varies with the output.
Acting Chairman King. Well, I wouldn't agree with that.
Dr. deChazeau. I hope not. [Laughter.] That is what I have
been trying to bring out.
Acting Chairman King. Well, I am very glad that w- ehcited now
the proper interpretation of your observations.
Dr. deChazeau. If it could be assumed that the economy was
operating at, or that there was an effective tendency toward fiill use
of resources and that prices were governed by differential costs, this
atomistic approach to the problem of demand on the assumption that
other things are equal would be less fallible. But the larger, in terms
of employment, the industry under consideration, the more important
the impact of its pricing policies on the economy, the more character-
istic the adoption of prices only indirectly related to differential
costs, and the more widespread is unemployment of resources, the
more certain it becomes that other things are not and cannot be
expected to be equal. If other things cannot be assumed equal,
Dr. Yntema's analysis of price elasticity of demand cannot be con-
sidered a criterion of desirable pricing policy even for the United
States Steel Corporation.
Irrespective of short-run inelasticity in demand, prices of steel
products must be adapted to the long-run development of volume
business by consuming industries.
In the short period it is probable that the demand for most products
and especially that for producers' goods, either raw materials or
capital goods, is relatively inelastic. The short-run demand for
consumers' goods tends to be governed by habits of consumption.
Among producers' goods, the substitution of one raw material for
another is limited by technological conditions involving product
design, labor skills and installed capital equipment and the demand
for any given raw material is limited by the prices which must be
paid for complementary goods, including labor. Substitution and
the development of volume business in a steel product, therefore, is
probably a function of long-run pricing policies rather than short-run
and of adaptation of product to consumer needs more than either.
The automobile industry illustrates my meaning. Although at
any given time in the evolution of that industry, the price elasticity
of demand for steel products was probably low, the development of
steel products, such as sheet and strip, which could be stamped and
processed under mass production conditions, together with a down-
ward trend in prices of these products, has undoubtedly constituted
a major factor in the growth of the consuming industry and the
demand for steel. Neither the ratio of steel costs to total costs in
the consuming industry nor the existing price elasticity of the demand
for automobiles could measure this potential demand for steel. (By
way of parentheses, the ratio of steel costs to retail price of goods
made from steel, as used by the Corporation, understates the impor-
tance of steel cost in finished price since it neglects aU margins in
distribution commonly computed on a percentage basis. This is a
minor point, however, and the cori-eution perhaps would be small.)
It was the profit potentialities in a new method of production
wicth were altered by sticb changes. To measure the potential
13640 CONCENTRATION OF ECONOMIC POWER
demand for steel by the ratio of steel costs to total costs per unit of
finished product on the assumption that all other things are equal, is
to violate the dynamics of demand and to neglect the influence of
altered profit possibilities on the character of the productive process
and consequently on investment, employment and purchasing power.
What has probably been true of automobiles may still be true of
containers, housing and other consuming industries. The only point
here stressed is that the short-run price elasticity of demand for steel
products, on the assumption that other things are equal, falsifies the
character of the demand for steel.
Irrespective of short-run inelasticity of demand, price policies for
steel products must be integrated in time with investment opportuni-
ties if a desirable volume of business is to be maintained.
The demand for capital ^ goods, for other producers' goods, that is,
raw materials, and even for durable consumers' goods varies primarily
with price and income expectations of buyers. Thus the pricing of
any durable goods is of the utmost importance as it may affect the
timing of purchase. In an economy in which there is an effective
tendency toward full use of resources, this shift in purchases over
time, rather than any net increase in total demand in the long run,
might be expected to exhaust the possibilities in the timing of price
change. Even in this state of facts, the timing 'of price change would
be far from an incidental phase of price poHcy in its tendency to
correct business cycle changes and maintain volume of sales.
But in the presence of a large and possibly permanent volume of
unemployment, proper price changes may not merely shift an existing
demand in time, but, through their impact on investment, employment,
and income in the economy as a whole, they may bring a net increase
in total demand. Thus, a price policy dictated by a demonstrated
short-run inelasticity of demand for steel — i. e., a high price — may
defeat the interests of the industry..
Increased expenditures on steel at the higher price would limit
potential expenditures in other directions. These reduced expendi-
tures mean lower money incomes for producers of other goods,
decreased profit possibilities or increased losses, and consequently
increased unemployment. The effect is likely to be cumulative
because, characteristically in such a state of facts, actual steel prices
exceed differential costs while investment by the steel industry is
likely to be curtailed. In other words, hoarding by the steel industry
is more likely to increase than to decrease. The net result is a decrease
in the demand for steel.
To the extent that the steel industry is a price leader for other
industries, the restrictive effect on the national income is magnified
and the demand for steel is further restricted, That is, if other
industries follow steel prices, a rise in steel prices bringing a higher
f)rice in other industries, would have the same restrictive and cumu-
ative effect.
Finally, as to the social problem, if Dr. Yntema's analysis of demand
reflects faithfully the businessman's criterion of desirable price, he
has dramatized the conflict of private and social interest in pricing
policy which is the fundamental issue before the Temporary National
Economic Committee.
In an economy in which there is less than full use of resources, the
public interest in pricing pohcies centers on the impact of industrial
CONCENTRATION OF ECONOMIC POWER 13641
prices on total employmenc and income. After Dr. Laughlin Currie's
able exposition before this committee/ it is not necessary for me to
explain how national income is but another aspect of total expendi-
tures, primarily by businessmen, on production and investment. But
a price reduction for a given product which merely increases expendi-
tures on that product at the expense of substitutes may have no net
advantage for the economy as a whole. Indeed, unless it leads to
dishoarding in the industry so benefited, it may actually decrease the
total use of resources. Where there is extensive underutilization of
capacity, this is the most likely result. Thus, price reductions in
goods for which demand is elastic have a problematical effect on full
use.
On the other hand, price reductions in commodities for which
demand is relatively inelastic free purchasing power for expenditures
in other directions. Unless there is a net increase in hoarding, there-
fore, the effect on the economy is hkely to be stimulative. This is the
paradox of pricing policy in an economy characterized by less than
full use of "resources. In those industries in which demand is elastic,
private interest may dictate lower prices of problematical social value
while in those industries in which demand is relatively inelastic, private
interests may dictate high prices which bring upon industry and the
economy the consequences it is most anxious to avoid — reduced
demand, unemployment, lower income.
On the other hand, any individual firm (no matter how large) which
attempted alone to implement a contrary policy would probably en-
danger its fmancial solvency. To be effective in stimulating an
increase in demand (despite existing price inelasticity on the assump-
tion that other things are equal), price changes must cut across indus-
tries and complementary goods and must be integrated with invest-
ment schedules and a monetary policy conducive to such investment.
The attainment of such a coordinated program is beyond the scope
of any given firm or industry, no matter how well-disposed it may be.
Wherever (for any reason) the size of the individual firm is so large
as to force or induce price leadership, or the number of sellers of a
relatively homogeneous commodity in a given market is so small that
each determines his pricing policy with reference to the price elas-
ticity of the market demand for the industry product rather than the
cross-elasticity of the demand for liis own output, this confhct between
private and social interest in pricuig policy is of paramount importance.
To break up existing corporations by law into units sufficiently small
to resolve this confhct, although in many instances the obvious pro-
cedure, would in many others be inconsistent with the trend of legal
and judicial opinion over the past half century, and, in some, would
be inconsistent with the economics of production and distribution.
To subject such industries to direct Government control would project
the economy into a maze of bureaucratic regimentation inconsistent
with democratic institutions and processes of dubious merit unless a
positive and integrated program of desirable behavior had been de-
vised, and of doubtful necessity if such criteria of desirable price
behavior had been defined and Government was ready to implement
them with a consistent monetary and investment policy.
I urge upon this committee the necessity for a permanent Federal
agency empowered to coUect from basic industries necessary informa-
' Dr. Currie's testimony appears in Hearings, Part 9.
13642 CONCENTRATION OF ECONOMIC POWER
tion with regard to price, sales, costs, and investment which would
permit it to devise criteria of desirable and possible price changes and,
through other Government agencies, to coordinate such a program
with .public and public utility investment schedules and with central
banking policy.
Acting Chairman King. You are not intimating the structure of
our political and economic system, are you?
Dr. deChazeau. What do you mean by the structure of our political
and economic system?
Acting Chairman King. I think your observation might indicate
that if the Government had determined to take over the control of
private industry, then there would be no legal obstacle to that objec-
tive. You didn't mean to say that?
Dr. deChazeau. I didn't say that.
Acting Chairman King. You didn't mean to convey that idea?
Dr. deChazeau. I didn't mean to convey that idea. I am not a
lawyer and therefore I am not qualified to make any statement of
that sort. What I did intimate is that the pulverization of industry,
the breaking up of large corporations, might well be inconsistent with
the legal policy which has been followed as indicated by statutes and
as indicated also by interpretations of the court.
Dr. Kreps. In other words, if there were a large number of units
in an industry, cross-elasticity of demand would be the important
consideration?
Dr. deChazeau. Yes; that is, as a matter of fact, the only reason
why V anyon could assume that private business interest, which is
undoubtedly to maximize profits or, as is often unfortunately the
case, to minimize losses, is consistent with social interest or with the
interests of the economy as a whole. It is the impact of such self-
interest which alone would lead to a maximization of output governed
only by the producer's own cost situation.
This means that in fixing price or production policies, one thinks
primarily in terms of what I have called the cross-elasticity of demand,
the demand for the product of the individual firm on the assumption
that its price is not necessarily met at the moment by its rivals. So
soon as the number of sellers becomes suflSciently small that it is
impossible for anyone to operate on that basis, but each must consider
the impact of his price and production policies on his rivals, then he
tends to determine his pricing policy with relation to a measure of
the elasticity of the market demand.
Now, in a competitive industry, I have no doubt that you would
find the market demand had much the same inelasticity that Dr.
Yntema has found in steel. That does not affect pricing poHcy in
such competitive industries so long as the number of sellers is so large
that each one governs his price with relation to his cost.
Mr. Wooden. Dr. deChazeau, do you mean by that that this cross-
elasticity of demand is simply another name for competition, price
competition?
Dr. deChazeau. I think it is one of the essential conditions of
price competition. I have used it merely as a shorthand in order to
preclude using a long phrase each time.
Mr. Wooden. You would think it might be said to he the equivalent
of price competition cross-elasticity?
CONCENTRATION OF ECTONOMIC POWER 13643
Dr. deChazeau. I would say that it is the essential condition of it,
and I would say that cross-elasticity of demand ceases to be important
as soon as the number of sellers is sufficiently small that each one
must take into consideration the action of his rivals.
Mr. Wooden. Do you mean by that that price competition is
something that, so to speak, is 'impracticable among heavyweights
but it is all right among hghtweights and middleweights, or what?
Dr. deChazeau. Let me correct a possible misinterpretation.
You may have price competition. What I am talking about is th^
assumed socially desirable effects of price competition. In fact, when
you get price competition among large firms, you may well cut far
below your costs with unfortunate effects both in that industry and
for the economy.
I wouldn't want to discuss that at any great length at this time.
Dr. Keeps. In clearing up Mr. Wooden's question concerning full
competition, isn't the demand for the product of the individual firm
infinite?
Dr. deChazeau. That is true on the assumption that you have
complete homogeneity of product.
Dr. Kreps. In other words, business policy where the number of
units are large is always based on the practical experience that, by
reducing prices they can get any amount of increased demand?
Dr. deChazeau. For that reason it isn't necessary to reduce price
because they can sell their entire output at the market price.
Dr. Kreps. That's right. Now, business policy, and meaning by
business policy, policy which individual businessmen must follow in
an industry where the number of sellers is small, has to assum'e that
demand is inelastic. Is that correct?
Dr. deChazeau. Not to assume it; it has to measure market de-
mand rather than the demand for the product of the individual firm,
and it finds, as Dr. Yntema indicates, that that demand tends to be
relatively inelastic. I have no reason to doubt that that relative in-
elasticity (i. e., the price elasticity of the demand on the assumption
^that other things are equal) is probably correct. I do not consider it
to be of fundamental significance.
Dr. Kreps. Let us try to rise by gradual stages from considera-
tions of business policy to considerations that must be taken account
of by the T. N. E. C. Is it true that there are a number of practices
which are on the whole good business policy 'that may be deleterious
for the industry as a whole?
Dr. deChazeau. You are asking a very broad question. I have
no doubt but that there are, but I am not willing to specify ^t this
time. What I was indicating at this time was that, insofar as price
policy might be conditioned on this short-run price elasticity of the
demand on the assumption that other things are equal, it would be
deleterious from the point of view of social poUcy and that it roight
well be deleterious for the industry as a whole, even though no single
firm in the industry could operate on any other assumption withou t
endangering its solvency. That is, a'5 I see it, the dilemma, that is
the paradox, that is the problem. In order to make a price reduction
effective in bringing about a significant shift in demand, it is neces-
sary that it must cut across industry and probably cut across com-
plementary goods. If you assume that the prices of all other com-
124191— 41— pt. 26 5
13644 CONCENTRATION OF ECONOMIC POWER
plementary goods remain the same, then a price reduction in the
short-run can bring very httle shift, I should think. I should agree
with Dr. Yntema, on that assumption.
Dr. Kreps. Is industrial policy always consistent with economic
pohcy? That is, is it true that what is good for the industry as op-
posed to what is good for the individual business, is always good for
the economy? For instance, you indicated here that when industry
becomes organized into few units it has almost an inherent interest in
higher prices.
Dr. deChazeau. I should say that industrial policy may be op-
posed to the general economic interest insofar as each member of the
industry is interested in maintaining his position and precluding or
minimizing losses. From the point of view of the economy as a whole,
the bankruptcy of a firm is a matter of no importance. It may be the
step by which one gets greater efficiency and greater progress.
Dr. Kreps. In an economic pohcy which endeavors to maintain
competition, we have a profit and loss economy. Is that correct?
Dr. deChazeau. Yes; at all times.
Dr. Kreps. At all times. Therefore it is not necessary that each
firm in an industry, in fact it is almost the surest sign of lack of health
if each firm in an industry makes a profit. The normal situation is
that some firms make extraordinary profits, a large run of firms make
ordinary profits, and then there is a marginal firm, and then is it not
true (studies of the Tariff Commission have indicated such) there is
a group of "lunatic fringe" firms with bad luck or bad management
or what have you that are in process of elimination.
Dr. deChazeau. That is right, and that is likely to be characteristic
it all stages of the cycle.
Dr. i^REPS. So that the problem that the committee has to consider
and has to decide before it accepts whatever is valid for the business
of an individual firm as being valid for the economy as a whole is
whether or not this particular Corporation is, let us say, a marginal
firm or one even less efficient than the marginal firm? That is one of
the problems, at any rate, that the committee would have to consider.
Is that correct?
Dr. deChazeau. That would be correct, particularly if one were
talking about the level of costs rather than the' behavior of costs. I
have concerned myself with the behavior of costs rather than the level
of costs, and have not raised the problem as to whether the level of
costs as indicated in the Corporation's study is in any sense character-
istic of the industry and therefore whether the Steel Corporation may
not be a so-called extra-marginal firm whose efficiency is reduced by
reason of its size.
Dr. Kreps. One further question. Economic policy in turn has to
be integrated with what we may call public policy. Man doesn't live by
bread alone; economic aspects aren't the only aspects that are import-
ant. Sometimes public policy as expressed, for example, in the tariff,
obviouslj departs from sound economic policy in order to give the
favor and the cloak of government to special interests. Sometimes
we disregard economic policy in order to effect a policy of national
isolation, self-sufficiency, or of defense. Is that true?
Dr. deChazeau. I should agree with that.
Dr. Kreps. And would you agree that the province of the Tempo-
rary National Economic Committee is the province of economic policy
and of public policy as well as industrial policy and business policy?
CONCENTRATION OF ECONOMIC POWER 13645
Acting Chairman King. I was about to say that I think the statute
under which we are operating determines what our province is.
Dr. deChazeau. In answering the question directed to me, my
own feeling would be that obviously the committee is concerned with
what is desirable policy from the point of view of society as a whole,
as public policy and good economic policy, and whether the particular
industries investigated, in determining their own policies, determine
them in a way consistent with the attainment of the objectives of
social policy.
• Acting Chairman King. I presume we may proceed upon the theory
that in weighing these various questions that are brought before us we
weigh them in the light of the fact that we are a republic, we have a
constitutional form of Government, and this is a democracy and not a
Hitler foriri of government, or one approximating the totalitarian
attitude of goverrmients throughout the various parts of the world.
Isn't that right?
Dr. deChazeau. Yes; and that is why. Senator, in making the last
point I want to emphasize that it doesn't seem to me that the alterna-
tive for this committee is to accept business policy as it has historically
developed as inevitable, on the one hand, or, on the other hand, to
regulate that industry. It seems to me that what is necessary is
primarily some form of agency. Federal agency, empowered to collect
from basic industries necessary information with regard to actual
prices, sales, costs and investment which might permit the analysis
of what is desirable price policy from the point of view of society, and
that might also be in a position to operate through other Government
agencies to coordinate that program with public expenditures; with
public utility investment schedules primarily — I am merely indicating
places where this is possible ttironorli existing controls — and with a
central banking policy. Frankly, i consider it rather futile to criticise
private business corporations for thfeir pricing policies in terms of
results before any criterion has been developed as to what is desirable.
Mr. Ballinger. Could ,such a criterion be developed, in your
opinion? I mean, assuming we had all this information, could you
then sit down and teU the Government what would be a fair price in
steel?
Dr. deChazeau. I shouldn't presume to answer that in the aflEirroa-
tive. I would say this, that if such a criterion cannot be developed,
then we are at a most unfortunate impasse; because if you have
Government operation of industry, you must have such criteria of
desirable pricing policy, and if you have regulation of industry, you
must have such criteria, and, therefore, in dealing with private
industry, -before you can criticize it for its actual pricing policy, you
must have developed some criterion which is known.
Mr. Ballinger. Assuming that we concede your point, that com-
petition is impossible in the steel industry, that is, if it were possible
Dr. deChazeau (interposing). I didn't make that point, but, if you
wish to assume it, you may:
Mr. Ballinger. Well, for the sake of argument they have reached
a point where they couldn't afford to compete. Do you think reduc-
tion in the number of steel concerns in the United States has been
entirely in the public interest? WiU you agree with me that these
concerns have grown largely by the processes of merger and combina-
tion and acquisition of competitors' assets, I mean by artificial
13646 CONCENTRATION OF ECONOMIC POWER
Dr. deChazeau. You want my opinion and my judgment, I
assume. My judgment in the situation is that thej clearly have not
been in the public interest.
Mr. Ballinger. You think a smaller size would have been better?
Dr. deChazeau. Well, if you pin me down to what constitutes a
smaller size, I cannot answer for lack of knowledge of the magnitude of
economies of integration and mass production within the industry.
That requires, at least as a starting point, some sort of cost analysis
of the industry.
Mr. Ballinger. We couldn't say definitely, or you wouldn't be
willing to say definitely that we can produce steel cheaper under
these giant concerns which have been created by artificial processes
than might have been produced under concerns that were less inte-
grated and perhaps subjected to the competitive system. We don't
know about that, do we?
Dr. deChazeau. No.
Mr. Ballinger. We have always proceeded on the assumption
that whatever size it arrived at, by whatever method it got there, it is
good and in accordance have so protected it, without analyzing the
question of whether they are or not efficient, whether they result
in lower costs or not. I am pointing that out because that is the
trend as I have seen it.
Dr. deChazeau. That is right.
Acting Chairman King. You don't want us to deduce a conclu-
sion from your views that this committee is to consider the propriety
of destroying any branch of the capitalistic system and turning it
over to public ownership and public control?
Dr. deChazeau. Well, if I may repeat my statement. Senator,
since it seems to have been lost somewhere, I emphasize that to sub-
ject such industries, industries in which competition does not work
purely and perfectly
Acting Chairman King (interposing). I could assume that there
are industries in which competition does not exist or in which com-
petition in a reasonably short length of time would result in competi-
tion, even though temporarily there might be an arrested process of
competition.
Dr. deChazeau. There are ail sorts and varieties of competition.
What I am assuming is that there are industries in which the size
of the individual firms with relation to their market is so large that the
policies of the individu^il units within that industry are not condi-
tioned on their own cost situation with relation to the market price,
but are conditioned on the reaction of their rivals to a price or pro-
duction policy or investment policy determined by them. In the
determination of that policy, taking into consideration this impact
of rival policies,- you do not attain the same type of price competition
that you would have in other industries; and that price competition
which you do attain in such industries is not likely to give you prices
which are related to or determined by differential costs.
Dr. Kreps. Let me turn back to the analy^.s. Under a situation
in which there were a large number of firms, would any analysis of
market demand be legitimate which neglected cross-elasticity?
Dr. deChazeau. Analysis of market demand in such a situation
would have no significance so far as tho individual unit was concerned.
Dr. Kreps. In other words, if there were competition, full compe-
CONCENTRATION OF ECONOMIC POWER 13647
tition, the individual business in an industry could proceed on the
assumption that the elasticity of demand for its particular product-
was infinite.
Dr. deChazeau. That would be the assumption.
Dr. Keeps. The less the competition the more the elasticity of
demand diminishes from infinity down to 10, 5, and 0.3. Correct?
Dr. deChazeau. No; wait a moment. You are now talking about
the market demand elasticity.
Dr. Keeps. Yes.
Dr. deChazeau. It seems to me that the market demand price
elasticity might be quite inelastic, much less than unity, and still
the individual units within the industry might operate, would operate,
on the assumption that the demand for their product was of infinite
elasticity.
Dr. Keeps. That is the point I wanted to make. In other words,
it would be unimportant to the individual enterprise that the elas-
ticity of demand was for the product as a whole.
Dr. deChazeau. I should say that it is of no importance to the
individual farmer what the price elasticity in the demand for wheat is.
It becomes important only when you set up a triple A which is trying
to restrict output within the industry.
Dr. Keeps. Precisely.
Dr. deChazeau. It is beyond the power of any individual farmer
to affect that market price by whatever he might decide to do. Hence
he can operate on the assumption that the market demand for his
product is infinitely elastic even though the market demand for the
product may be very inelastic and in the short run is likely to be.
Acting Chairman King. Generally speaking, prices have gone
down during the past 25, 30, 40, or 50 years, in steel and automobiles,
in wagons?
Dr. deChazeau. Yes.
Mr. Ballingee. Not in steel.
Acting Chairman King. In automobiles and most of the major
portion of the commodities that enter into our personal, family, and
economic life. There has been a gradual reduction, has there not?
Dr. deChazeau. That is my impression, Senator.
Dr. Keeps. Although from studies of Dr. Frederick C. Mills and
others who have analyzed recent price history, is it not true that the
price of steel and certain other durable goods in the United States,
and by exception since 1929, have stayed relatively high? As Mills
points out, that is an experience quite unique in our history. It is
also an experience which is not known, for example, in Japan. The
relatively high price for building materials, for steel and for producers'
goods in general during the thirties tended to restrict the amount of
such products that can be bought and the amount of investment that
is made by the economy in general, did they not?
You are famihar, I take it, with Professor Mills' analysis comprising
three volumes, which tended to demonstrate that point?
Dr. deChazeau. In a general way, yes; but I wonder if v/e do not
get beyond the discussion of price elasticity when we talk about price
trends?
Dr. Keeps. Quite. It is only apropos of the problem that high or
low prices have no meaning unless set in terms of other prices?
Dr. deChazeau. That is right.
13648 CONCENTRATION OF l^XJONOiMIC POWIOR
Dr. Kreps. And relative to agricultural prices and food prices and
other items, the prices of steel and of steel products are still relatively-
high, even though there has been an absolute decrease since 1921?
Dr. deChazeau. Yes; I should agree that a downward trend in a
price is to be measured with relation to the prices of other products,
and I should agree likewise with the point which Dr. Yntema made
this morning, that price flexibiUty is most meaningful with relation
to cost flexibility. That is why it seems to me that the cost analysis
is so crucial to this entire discussion.
Acting Chairman King. Does that finish your statement?
Dr. deChazeau. Yes, sir.
Dr. Kreps. I had intended to call Dr. Yntema back to the stand,
but in view of the lateness of the hour, I should like to call Dr. Yntema
and the assistants who helped him, tomorrow. I suggest Dr. de-
Chazeau be recalled to the stand tomorrow.
IN MEMORIAM SENATOR WILLIAM E. BORAH
Acting Chairman King. This is deserving of consideration. When
we met yesterday we were advised by the chairman of the committee
of the passing of one of the members of this committee, and a
committee was named by the chairman to prepare appropriate
resolutions to offer, to be incorporated in our record. The committee
has acted and I am authorized as chairman of the committee to submit
the following for the committee, and ask that it be inserted in the
record:
In the death of Senator William E. Borah the Temporary National Economic
Committee has lost a quality of service of the highest order. His judgment on the
national economic problems with which the committee had to deal was broadened
by a lifetime of distinguished public service from which he had emerged as one of
the greatest and most consistent champions of free and independent business
enterprise.
His gift of public advocacy gave him great power in advancing the cause of
economic freedom. Public faith in his courage and integrity was so strong that his
presence on this committee became a guaranty of the honesty of its proceedings
and the good faith of its conclusions.
Yet in our sorrow over his death this committee consoles itself with the thought
that it has not entirely lost his judgment or his power of advocacy, or the influence
of his great prestige, Isecause during the year and a half that he served with us he
left the imprint of his views and his example on every member in a way which we
cannot forget: Therefore be it
Resolved by the Temporary National Economic Committee in meeting assem-
bled, That this committee spread upon its records the acknowledgment of its
indebtedness to Senator Borah; that we constantly keep in mind during our future
deliberations and hearings his counsel and example, and that in that way we guard
and preserve the great contribution which he has already made to our work.
We will adjourn until 10 :30 tomorrow morning.
(Thereupon, at 4 :50 p. m., the committee recessed until Wednesday,
January 24. 1930, at 10:30 a. m.)
INVESTIGATION OF CONCENTRATION OF ECONOMIC POWER
WEDNESDAY, JANUARY 24, 1940
United States Senate,
Temporary National Economic Committee,
Washington, D. C.
The committee met at 10:40 a. m., pursuant to adjournment on
Tuesday, January 23, 1940, in the Caucus Koom, Senate Office
Building, Joseph J. O'Connell, Jr., special assistant to the General
Counsel, Treasury Department, presiding.
Present: Mr. O'Connell (acting chairman), Senator King, Repre-
sentative Williams, Messrs. Lubin and Hinrichs.
Present also: Hugh White and Walter B. Wooden, representing the
Federal Trade Commission ; John V. W. Reynders and Walter White,
representing the Department of Commerce; William W. Werntz,
representing the Securities and Exchange Commission; A. H. Feller
representing the Department of Justice.
Acting Chairman O'Connell. The committee will be in order.
Dr. Kreps.
Dr. Kreps. In each case, the testimony of the group that has been
incorporated under the Temporary National Economic Committee
has been given to Dr. Yntema and other members of the staff of the
Steel Corporation so that they might make their comments.
I want to say that all of us are anxious to avoid undue and sterile
debate. All of us are equally anxious that the points of diflference in
the debatable area are clearly stated. I have asked Dr. Yntema to
make his comments upon Dr. deChazeau's remarks. I believe he
would like to have two of his assistants sworn in by the committee.
Is that correct, Dr. Yntema?
Dr. Yntema. Yes; I should appreciate that very much, Dr. Kreps.
Mr. Appert was primarily responsible for the study of the costs and
relation of cost to volume, and Mr. Lewis undertook the statistical
analysis of the demand for steel. I shall appreciate it if you will call
them at this tinie.
Dr. Kreps. I would like to call Mr. Appert and Mr. Lewis to the
stand.
Acting Chairman O'Connell. Will you each raise your right hand,
please? Do you and ea^h of you solenmly sHvear that the testimony
you are about to give in this proceeding will be the truth, the whole
truth and nothing but the truth, so help you God?
Mr. Appert. I do.
Mr. Lewis. I do.
13649
13650 CONCENTRATION OP ECONOMIC POWER
TESTIMONY OF HAROLD GREGG LEWIS, INSTRUCTOR IN
ECONOMICS, UNIVERSITY OF CHICAGO, AND RICHARD H.
APPERT, ATTORNEY AT LAW, RUTHERFORD, N. J.
Dr. Kreps. Mr. Lewis, for the purposes of the record, will you
state your full name?
Mr. Lewis. My name is Harold Gregg Lewis.
Dr. Kreps. Address?
Mr. Lewis. 1535 East Sixtieth Street, Chicago.
Dr. Kreps. You are instructor in economics at the University
of Chicago?
Mr. Lewis. That is right.
Dr. Kreps. You are also research associate of the Cowles Com-
mission for Economic Research at the University of Chicago?
Mr. Lewis. Yes.
Dr. Kreps. And you are the author of two of the exhibits which
have been- submitted here, "Exhibit No. 1411," ' entitled "A Statis-
tical Analysis of the Dfemand for Steel, 1919-38," and "Exhibit No
1412," 2 entitled "An Analysis of Changes in the Demand for Steel
and in Steel Prices, 1936-39"? Is that correct?
Mr. Lewis. That is correct.
Dr. Kreps. For the purposes of the record, Mr. Appert, will you
state your full name, please?
Mr. Appert. My name is Richard H. Appert.
Dr. Kreps. And your address at present?
Mr. Appert. My address is 62 'Ettrick Terrace, Rutherford, N. J.
Dr. Kreps. What position do you hold?
Mr. Appert. At the present time I am a lawyer with Mr. Olds'
firm.3
Dr. Kreps. I understand that you were formerly instructor in
accounting at Fordham University.
Mr. Appert. That is correct.
Dr. Kreps. And you have assisted particularly in the study of the
relation of volume to cost?
Mr. Appert. That is correct. I have done the work under Dr.
Yntema's direction and with the assistance of other members of our
group.
Dr. Kreps. Dr. Yntema.
TESTIMONY OF PROF. THEODORE OTTE YNTEMA, SCHOOL OF
BUSINESS, UNIVERSITY OF CHICAGO, CHICAGO, ILL.— Resumed
discussion of united states steel corporation studies
Dr. Yntema. We are glad to have the criticisms of our studies
offered by Dr. deChazeau in his testimony yesterday afternoon, and
we appreciate particularly the courtesy extended to us by Dr. Kreps
and by the committee in affording us the opportunity to comment on
the issues he has raised. While many points were mentioned in the
discussion yesterday afternoon, it is not possible without encroaching
unreasonably upon the time of the committee to deal with all of them.
Appendix, p. 13913.
Appendix, p. 13942.
Irving S. Olds, partner. White & Case; also a director of United States Steel Corporation.
CONCENTRATION OF ECONOMIC POWER 13651
I shall, therefoi'e, restrict my remarks to those which I regard as most
important.
First of all, I should like to clear up any misunderstanding which
may exist as to the purposes for which these studies were prepared.
They were not made with any idea of providing the United States
Steel Corporation or the steel industry with a formula which could be
used as a basis for price policy. As a matter of fact, steel men were
well aware of the characteristics of the demand for steel and the
behavior of costs long before we began this study. We have merely
applied the methods of statistical and economic analysis to the facts
and presented our findings to the committee in the simplest way we
could.
Our objectives in the analysis of demand and cost were these: First,
to ascertain approximately how the quantity of steel sold by the in-
dustry responded to changes in price, and second, to discover how costs
varied with output from the data which were available to us.
We have presented these findings to the committee in the hope that
they may throw some light on the possibilities, and on the limitations,
of increasing steel consumption by reducing price and on the extent
to which such price reductions could be borne by a company such as
the United States Steel Corporation.
Near the conclusion of his testimony, Dr. deChazeau said that if our
"analysis of demand reflects faithfully the businessman's criterion of
desirable price, he has dramatized the conflict of private and social
interest in pricing policy which is the fundamental issue before the
Temporary Economic Committee."
In the first place, there was never any suggestion on our part that
our analysis reflected or had anything to do with the businessman's
criterion of desirable price.
In the second place, and more important, the phrase, "conflict of
private and social interest in pricing poUcy" requires further clarifica-
tion.
In an economic system of private enterprise, each business seeks
and ought to seek to make the largest possible profit in the long run.
I suppose that most businessmen would like to get a higher price
for their products than they do, and I think it is correct to say that
it [would not be in the general social interest for them to obtain as
high a price as they woiild like to get.
If this is merely what is meant by the conflict of private and social
interest in pricing policy, it is an empty phrase. The real question is
whether the price level in the particular industry is such as to warrant
concern for the social interest.
There seemed to be some question yesterday as to why the steel
industry did not charge higher prices for their products if they could,
thereby, so obviously reduce their losses and increase their profits.
Certainly it is not because the steel companies do not want to
raise their profits from the levels which have prevailed over the past
10 years. The situation can only be explained by the fact that the
forces of competition are great enough to keep individual companies
from raising their prices.
Dr. Kreps. Would you care to elaborate that point a moment?
Isn't it generally true that monopoly defeats itself, that by stimulating,
as Mr. Chamberlin has pointed out in his well-known treatise on
13652 CONCENTRATION OF ECONOMIC POWER
monopolistic competition, by stimulating excessive capacity, by
causing other costs, particularly selling costs, to enter into the picture
the net results for the industry may none the less be disappointing
even though the factor of monopolistic competition operates virtually
unimpeded?
Dr. Yntema. I tliink you have made a correct statement of the
theory and the fact with respect to the sort of situation you describe.
Frankly, I don't think that is the explanation of why steel prices are
no higher than they are.
Dr. Kreps. Although that is the only possible assumption which
could have justified you in your analysis to neglect the cross-elasticity
of demand because otherwise your whole analysis of demand would
have no meaning for the Corporation. In a purely competitive situa-
tion the individual producer knows that he can sell his whole product
at the market price, in other words the demand for his individual
product in a purely competitive situation is infinite. Thus the farmer
sells all of his output at the going market price.
Dr. Yntema. Dr. Kreps, I think you impute to us objectives which
we did not have.
Dr. Kreps. Not objectives; I am asking you whether that is not
the assumption underlying your analysis. I must admit, in fact I
must pay tribute to the candor with which you have made that
assumption. It is one of the many things I admire in your study,
that you based it on the premises of monopoly and monopolistic com-
petition.
Dr. Yntema. The point I would like to make is this, that we were
not preparing a study to be used as a pricing policy by the steel
corporation or by the steel industry. We did, however, prepare the
study in the hope that this committee might be able to use this material
in appraising the level of prices in the industry over the past 10
years and to date with respect to the possibility of further reductions
and the results of such reductions in prices upon the industry.
We never undertook to deal with the problem of the demand for
the steel sold by an individual concern. If I may go on — I expected
to comment on that point.
There seemed to be some confusion yesterday between the elas-
ticity of demand for the industry and the elasticity of demand for
the individual concern. I should like to make it entirely clear that
our estimates referred only to the elasticity of demand for the indus-
try, that is, to the relation between the total quantity sold by the
industry and the price of steel, and that we never at any time at-
tempted to estimate the so-called cross-elasticity of demand, that is
the demand for the steel sold by the individual firm. That is not
relevant to the purposes which we had in mind.
Dr. Kreps. Mr. deChazeau, would you like to comment on that?
Dr. deChazeau. Merely at this time to make the point clear that
I indicated in my discussion, that Dr. Yntema had neglected that and
that in my judgment he had rightfully neglected it in view of the
pricing situation in the industry.
Dr. Yntema. At this time I should like to comment on a few of
the points in Dr. deChazeau 's ^''^ -.fn*'' '*
of the demand fo^* stool
CONCENTRATION OF ECONOMIC POWER 13653
Dr. deChazeau said:
The assumption that other things are equal, a necessary condition for the der-
ivation of a statistical measure of price elasticity from a time series by the
correlation technic, is false and vitiates the conclusions as a measure of the effect
of price change in a dynamic situation.
Frankly, I think that statement could only have been made on the
basis of a misunderstanding of our study. We did not assume other
things were equal. Our statistical analysis made allowance for the
effect of other factors and yielded an estimate of elasticity of demand
which would have been the elasticity upon the assumption that other
factors did not vary, but it is not a measure derived neglecting the
effect of other factors. We took specifically into account the effects
of these other factors and adjusted for them.
(Senator King assumed the chair.)
Dr. Keeps. Let me clarify this point. Your contention is not
that you neglected the other factors, but that you assumed them to
be constant?
Dr. Yntema. No; that is incorrect. We did not assume them to
be constant. We corrected automatically in the study, by the
methods we used. We corrected for the effects of the other causes
and the result, the measure of responsiveness of quantity of steel
sold to the price, represents an estimate of what would happen if
these other factors should stay constant.
Dr. Keeps. Yes; if they should stay constant.
Dr. Yntema. Yes.
Dr. Keeps. Which is exactly what we had in mind. ' I was going
to ask you otherwise whether you had, and if so, whether I had over-
looked finding the correction which you had made for the fact that
depression is primarily a heavy industry phenomenon. The steel
industry is the major industry. Therefore, the impact of unemploy-
ment or of restriction of production or of prices in steel upon general
business activity is highly substantial.
I was going to ask you whether or not you had made any correc-
tion. The answer is, I take it, that your analysis proceeds on the
assumption that those other factors were constant. As you just
phrased it a moment ago, you proceeded as if these other factors were
constant.
Dr. Yntema. No; I think we are now quibbling about language,
and I think we understand each other. I should not state our find-
ings in the words which you used, but I really don't think that we
would gain anything in further discussion as to the terms which we
employ.
Dr. deChazeau. May I make a statement?
Acting Chairman King. When doctors disagree, the patient suffers.
[Laughter.]
Dr. deChazeau. I doubt whether there is a real disagreement, and
I offer this merely to check with Dr. Yntema. As I understand it,
in the method of correlation employed or appUed, you take into ac-
count the effect through the concomitant variation process of factors
including price and including business profits and including changes
in the industrial situation, and then in order to^get at the price
elasticity of demand, you eliminate those changes which are due to
those other factors.
13654 CONCENTRATION OP ECONOMIC POWER
Dr. Yntema. That is correct.
Dr. deChazeau. Isn't that correct?
Dr. Yntema. Yes; that is correct.
Dr. deChazeau. That is what I mean by making the statement
that you assume other things are equal in your final price elasticity
of demand. You have eliminated the other factors from the variation
in sales. ^
Acting Chairman King. Doctor, are you a sufficient pragmatist
and realist to understand that at the end of the month or the end of
a period, a business organization, whether it has a number of activities
or is limited to one, makes its findings and discovers that it is in the
red or it made a profit. You admit that this is ordinarily the practical
way of conducting business?
Dr. Yntema. That is, of course, the fact.
Dr. Kreps. We all know that.
Acting Chairman King. Do these technical discussions which you
and Dr. deChazeau and Dr. Kreps have been indulging in throw any
light upon whether or not the steel industry or any industry for that
matter, at the end of the term, had a deficit or had a profit?
Dr. Yntema. I think they throw a great deal of light upon what
the deficit or profit would have been, if the price level of steel had been
different.
Acting Chairman King. Well, then, you can conceive of a hundred
different reasons which would have added to the profit or added to
the deficit?
Dr. Yntema. That is correct.
Acting Chairman King. Such as the level of wages, the prices of
raw materials, cost of transportation, and many, many other factors
that enter into the conduct of a business?
Dr. Yntema. That is absolutely correct. We assumed, however,
that the coihmittee was much iaterested in the relation of the price
of steel to the burden which reduction of price would have upon the
indus,try, and some of our remarks were addressed to thia,t point.
Acting Chairman King. Well, the proffessors may continue.
Dr. Yntema. Dr. deChazeau said:
As a criterion of pricing policy for the steel industry itself, the price elasticity
of demand measured by Dr. Yntema is inadequate.
I should merely like to point out that we never submitted our study
as a criterion of policy for the mdustry. I do not think it is possible
for the steel industry to have a pricing policy. There was discussion
with reference to whether the elasticity of demand for steel was the
same in depression periods and periods of prosperity. We simply
do not know from the examination of the data and we have not been
able to find out. We did not assume that it was necessarily constant.
Our result merely gives an average estimate of the elasticity, and we
should welcome any further light which can be thrown upon that.
Again, Dr. deChazeau said:
If other things cannot be assumed equal, Dr. Yntema's analysis of price elasticity
of demand cannot be considered a criterion of desirable pricing policy even for the
United States Steel Corporation.
With that I should agree, but I should point out that we never
thought that it should be regarded a criterion of desirable pricing
policy by the United States Steel Corporation.
CONCENTRATION OF ECONOMIC POWER 13655
Again, he said:
Irrespective of the short-run inelasticity in demand, the prices of steel products
must be adapted to the long-run development of volume of business by consum-
ing industries.
We addressed our remarks and our studies primarily to the possibility
of cyclical fluctuations in steel prices, and for that purpose, the study
or the consideration of short-run substitution over the period of this
cycle was relevant to our analysis.
Dr. Kreps. Shouldn't you rather have used quarterly or monthly
data in that event?
Dr. Yntema. There was a good reason for not employing quarterly
or monthly data. If such data were employed, it would be necessary
to bring into the study the effect not only of the level of steel prices
upon the consumption of steel, but also the effect of reductions. We
have found in our study that the effect of a reduction in steel price is
to scare off purchasers. As a matter of fact, if the price is reduced the
immediate result of that generally is to reduce and not to increase the
purchase of steel.
That is not invariably so, but generally that is the effect, and we
wanted to abstract that and leave it out of the picture. We preferred
to take not a month-to-month effect, but to take the effects over a
year-to-year period, which I think is more appropriate for this par-
ticular problem. |
Dr. Kreps. It may be appropriate for a trend problem, if the period
is sufficiently long but practically never utilized, is it, for measure-
ments embracing but one major cycle?
Dr. Yntema. I think it is in this case the better procedure to adopt,
but I do not wish to argue the point with you.
Dr. Kreps. At any rate it is contrary to accepted statistical pro-
cedure, isn't it?
RELATIONSHIP BETWEEN PRICES, DEMAND, AND COSTS
Mr. Woode;?;. Dr. Yntema, I understand you to take the view that
the demand for steel has only a minor effect upon the price, or rather,
the price has only a minor effect upon the demand?
Dr. Yntema. I think we ought to clarify that statement. Let
me put it this way: I should say that changes in the price level have
been less important, and, within any conceivable range, will be less
important in determining the quantity of steel that is sold^ than other
influences, such as profits, and the degree of activity in other parts of
our economic system.
Mr. Wooden. Well, do you think that the demand for steel would
be affected in any substantial degree by an increase in price?
Dr. Yntema. You mean the quantity of steel bought?
Mr. Wooden. Yes.
Dr. Yntema. I think it would be affected to some extent.
Mr. Wooden. Only to a minor extent?
Dr. , Yntema. It depends upon what you mean by those terms.
Let's put it this way, that it would be affected less proportionately
than the increase in price, and the effect upon the consumption would
probably be less over a period of years than the effects of many other
factors associated with the cyclical ups and downs in general busines^.
13656 CONCENTRATION OP ECONOMIC TOWER
Dr. Kreps. I must confess that I am not clear on that point.
Suppose the price of steel were raised 10 percent. Is it your assump-
tion that the amount of steel sold .would not diminish by any more
than 3 or 4 percent? Is that correct?
Dr. Yntema. But that is a rough estimate. We pointed out
Dr. Kreps (interposing). That is your estimate of the elasticity of
your demand curve?
Dr. Yntema. Let me say that we took that as the best guess we
could make. It may be lower than that. Many of us who studied
the figures think that the elasticity ts less, that by a 10 percent
decrease in price you wouldn't get even 3 or 4 percent increase in
volume. Some of those with wHom I talked think you wouldn't get
any increase.
Dr. Kreps. That is the best guess you could make?
Dr. Yntema. That is the best guess, with the evidence we have.
Dr. Kreps. Quite. Now, let's raise the price 20 percent. Do you
mean then that the amount of^teel purchased would decrease by only
6 or 8 percent? ""
Dr. Yntema. You are getting near the limit of the range of experi-
ence now.
Dr. Kreps. But not beyond the limit of the charts which you
submitted yesterday for the record.
Dr. Yntema. I think that our discussion^esterday had to do with
price changes of 10 percen generally. There was one case, however,
in the discussion of the j ice decrease which would be necessary to
bring the 1938 production up to the 1937 level where we talked about
a larger change than that. Frankly, the particular results there, the
particular quantities, are not very significant.
Dr. Kreps. I am referring to Chart B-5 of "Exhibit No. 1409," ^
in which you give in rather precise terms the total loss, as v^ell as
estimated additions to deficit if prices had been reduced asindicated.
Dr. Yntema. Yes; that ranges up to 18 percent, which is within
the range of experience on which our studies are based.
Dr. Kreps. Would you go on and say that if prices were increased
30 percent the amount of decrease in steel demand would be only 9 or
12 percent?
Dr. Yntema. If I may, I should like to be excused from answering
that question. I think it is in a realm not important for business
poHcy. I don't know and I don't think anyone knows.
Dr. Kreps. Isn't that the kind of price change that has occurred in
the industry according to your own figures, and therefore the kind
which may be vital for the purposes of the committee?
Dr. Yntema. I should say that if you are going to effect any great
increase in quantity of steel sold you would have to talk about changes
of that order of magnitude, and I think that if you are going to con-
sider those you must immediately consider the impact upon the
losses in the industry. And those losses would be so great that the
steel companies in the industry would almost immediately go into
bankruptcy unless they could pass on the reduction in prices to the
wage earners and to the others irom whom they buy materials.
Dr. Kreps. In other words, steel prices you feel are based on costs.
Is that correct?
Appendix, p. 13781.
CONCKNTRATION OF ECONOMIC POWER 13657
Dr. Yntema. I would never make the statement in quite that
form. I should say that cost is one factor which enters into price,
and it is a very important factor, and the relevance of cost to price
depends upon how long a period you have in mind. Prices might be
considerably less than cost in the short run. In the long run they will
tend to be approximately of the same order of magnitude.
Dr. Kreps. You will remember that yesterday you submitted a
chart of indexes of costs, actual costs, and of mill-net realizations.
As I understand it, mill-net realization reflects pretty well what the
consumer pays to the industry. At least that is in substance your
contention. Is that correct?
Dr. Yntema. I think that is a fair statement.
Mr. Reynders. Isn't it corrpct to say that the costs are a deterrent
against going indefinitely below in the price range?
Dr. Yntema. I'm sorry; I didn't heaTr the question.
Mr. Reynders. Is it not correct to say that the costs form a deter-
rent against undue reduction of price; that is, when you get below your
cost Ime you know you are in a danger zone and you begin to hesitate?
Dr. Yntema. That is correct; yes. The function of costs in the
processes which establish prices is to set a downward limit. Or-
dinarily a businessman will not undertake a venture which will bring
him in a smaller income than the additional costs resulting from that
venture. That is simply plain common sense and good economic
theory. The costs set a downward limit with respect to prices.
Mr. Reynders. And when you reach that cost line any disposition
to take business below that cost line will be actuated by yo-ur desire to
maintain your position in the industry or to meet some competition
from a competitor, which may be wise or unwise?
Dr. Yntema. Yes. You and I probably would use slightly different
words in saying the same thing, but I should accept that.
Dr. Kreps. Turning now to chart C-25 of "Exhibit No. 1409" *
which is on the easel, the lower line representing prices to the con-
sumer substantially — —
Dr. Yntema (interposing). May I interrupt? The lower line repre-
sents an index of mill-net yield. That does not represent the absolute
amount of prices.
Dr. Kreps. Yes; but it does mean that when the index reaches a
low of somewhere around 75 in the middle of 1933, the actual" price
to the consumer was probably somewhere in the vicinity of 25 percent
less than it was in 1926.
Dr. Yntema. Yes; that is correct.
Dr. Kreps. Now the actual costs, on the ot^e'r hand, also are
indices, are they not?
Dr. Yntema. Yes; the top is an index of the average cost per
weighted ton shipped.
Dr. Kreps. Therefore, irrespective of whether these are absolute
amounts or indexes, the relationship shown by your chart is the
relationship between prices to the consumer and costs to the Steel
Corporation.
Dr. Yntema. Well, if I were to say that I would qualify those
terms. I should say that the chart shows the relative movement of
' Appendix, p. 13835.
13658 CONCENTRATION OF ECONOMIC POWER
the prices to the consumers, and the average cost to the Steel
Corporation.
Dr. Kreps. The relationship, if you will observe, between costs
and prices, therefore, is, if anything, inverse; the higher the actual
costs go, the less theOopporation tends, historically speaking, to receive
from the consumer, as you brought out yesterday.
Dr. Yntema. That is correct in respect to these particular aver-
age costs.
Dr. Keeps. They sell a smaller volume and are doubly hurt be-
cause they also sell it at a lower price at the precise time when their
actual cost per unit is higher than it has been before. ,
Dr. Yntema. It is the actual average cost. If you look at the other
line you will see that the cost prices which they are paying for goods
and services do tend to vary in the same direction.
Dr. Keeps. But these deficits that you have estimated are on the
basis, are they not, of total costs at each level of output? If you take
the total costs and divide by total output, you get average cost, which
is represented by the upper line and therefore the line that is pertinent
to the problem of whether there were losses and pertinent to the prob-
lem of whether or not pricing pohcy was actually based on the behavior
of costs.
Dr. Yntema. No; I don't want to be quoted' as saying that in the
short run over the cycle that average costs determine what the price
will be. That we all know is not correct, and this chart demonstrates
beyond any possible doubt that that is not what did happen.
Dr. Keeps. I want to make clear for the committee one other
problem. I would like to have chart B-1 of "Exhibit No. 1409" ^
again put on the easel, please. When you speak of cost of steel you
don't mean steel, do you?
Dr. Yntema. That depends upon which particular statement of
mine you are referring to.
Dr. Keeps. This cost that you have here measured, "Relationship
between total costs of operations and volume of business," does not
represent a particular steel product?
Dr. Yntema. That does not represent a particular steel product.
We took some pains to point out that those costs are the total costs
of the Steel Corporation excluding certain miscellaneous, extraneous
operations. Those costs therefore extend beyond the production of
steel. Just very roughly, I should say that 90 percent of those
costs represent the costs of steel operations.
Dr. Keeps. Your figure was 89 percent.
Dr. Yntema. I say roughly 90 percent.
Dr. Keeps. Wliat is the rest of it?
Dr. Yntema*. The rest of it represents the cost of producing other
byproducts of the steel industry, the cost of producing cement, the
cost of furnishing various transportation services to outside
Dr. Keeps (interposing). Of the steel products that you have,
how many would you estimate'are roughly included?
Dr. Yntema. Well, I never undertook to count the number of
steel products.
Dr. Keeps. But it is many thousand?
Dr, Yntema. It is undoubtedly many thousand; it depends, on
how you define a product, however.
' Appendix, p. 13773.
CONCENTRATION OF ECONOMIC POWER 13659
Dr. Keeps. With the widest difference of quality and the widest
difference ia price per pound and price per ton?
Dr. Yntema. No, not of the widest difference in quality.
Dr. Kreps. Do you have the range with you?
Dr. Yntema. No. Let me point out now, if you are leading up to
the question of the aggregation of different items, that the relative
differences in quality and characteristics of these different products
are very much less than the differences and characteristics of the
products which statisticians and economists commonly combine in
. an index of physical volume.
Dr. Kreps. You mean to imply that such large differences exist
in, say, the quantity and quality of bushels of a standard grade of
wheat or a standard pound of sugar, with which, ordinary demand
and cost studies are concerned?
Dr. Yntema. No; that is not the point I was trying to make. I
was merely sayiug that many of the ind'^^'^s which are used, issued
by the Federal Reserve Board and by ot^or agencies of the Govern-
ment, are indexes of volume of production composed of such different
commodities as iron and steel, chemicals, textiles, and so on. What
I am saying is that the various steel products are more like each
other than the different components included in those indexes of
volume of production which have good standiag among economists
and statisticians.
Dr. Kreps. But you are measuring here the cost per ton of a
theoretical unit called steel in which the composition of steel varied,
did it not? For example, in 1932, did the figure of 11 percent for
items other than steel hold true, or was it considerably higher?
Dr. Yntema. I can't tell you, offhand, what the situation was.
Dr. Kreps. You did not stop to see i whether the composition of
the unit which you have asked us to accept as homogeneous was
actually homogeneous throughout the period?
Dr. Yntema. No; we never asked you to accept the composition
of the unit as homogeneous. That is not a correct description of
any index number of this type or any other sort. What we did do
was this. Instead of just adding tons of different products, we
assigned higher weights to the higher-cost products, lower weights
to the lower-cost products.
Dr. Kreps. Based on Values in some ye^r?
Dr. Yntema. Based on mill costs.
>Dr. Kreps. On the average for the period?
Dr. Yntema. Over a period of 5 years, from 1933 to 1937, inclusive,
I should say this — speaking strictly as a. professional statistician —
that I would stack this index up with any index of production that
you would care to name, and I should say that this would represent
the variations in volume for the products covered by the index with
at least as high a degree of accuracy as any index you can name.
Dr. deChazeau. I would just like to raise the point that the
validity of an index depends entirely upon the use to which it is put.
You may have production indexes which cover a much wider variety
of products than the . production index used here, but if it 's used
merely to indicate a trend in production and not used to determine
a cost, it seems to me that it would have inherently a greater validity.
I wonder whether Dr. Yntema would -comment on that.
124491— 41— pt. 26 6
13660 CONCENTRATION OF ECONOMIC POWER
Dr. Yntema. I would say this, that I regard this index as being
more than reasonably acceptable for the purposes to which we have
put it.
Dr. Kreps. You would regard this cost curve as being as acceptable
as the cost curves of special products which the United States Tariff
Commission publishes in its cost studies?
Dr. Yntema. The curves used by the United States Tariff Com-
mission, if I remember correctly, are of a very different type.
Dr. Kreps. They are also accounting costs, are they not?
Dr. Yntema. Yes; but if I remember, they represent a frequency
distribution of such costs. I don't like to get into such technicalities,
but I must point out that those so-called cost curves of average cost
of individual concerns, are entirely different from this type of study
we have presented.
Dr. Kreps. They are likewise cost curves for the industry, the
only difference being one of arrangement of the units. But let us
restrict the question to the nature of the unit. If I may be permitted
an analogy, what you have done is tantamount to taking a population
of 50,000 individuals of variegated races and tongues and nationalities
and asking us to accept a figure for average height and weight as
meaningful when the number of children or of Chinese varies from
1 percent in 1 year to 20 percent or more in another year.
Dr. Yntema. May I put the question
Dr. Kreps (interposing). The unit itself is what I am talking
about, not the arrangement.
Dr. Yntema. Would you say that an index of physical volume such
as prepared by the Federal Reserve Board is useful to show the
approximate fluctuations in the volume of business for the industries
included in that index?
Dr. Kreps. What you say you have given us is an actual cost
curve, not an index of production. They are not the same nor even
similar.
Dr. Yntema. No; but the question is fundamentally the same.
'The question is whether the index represents with reasonable satis-
faction the fluctuations "in the quantity of these products produced.
The question which I asked you I think is equivalent to the question
which you asked me.
Dr. Kreps. The two positions, I submit, are clearly stated, which
is all that is required.
Acting Chairman King. I wondered whether Dr. Kreps in his
analogy of 50,000 population with only 1 percent Chinese was meas-
uring their physical or their mental or their other qualities, their
capacity for work, or their capacity for idleness.
Dr. Kreps. It would make no difference what quality you selected,
when you have a unit which is as heterogeneous, which, if I may say
so, sir, is as nondescript a theoretical hash of varying composition as
is the unit called "ton of steel" in this analysis, the result obtained
is subject to a considerable amount of debatable evaluation. Now I
want to say immediately that if I had been in Dr. Yntema's shoes I
might have tried to do exactly what he did.
I also want to point out, however, that for purposes of evaluation
the point which I raise concerning the homogeneity of the unit will
be admitted by Dr. Yntema himself as being something which gave
him a great deal of worry.
CONCENTRATION OP EC?ONOMIC POWER 13661
Dr. Yntema. Yes; that is correct, but I should say that the range
of error due to the point that you make is not substantial for the
range of accuracy with which we are concerned.
Acting Chairman King. Trying to be a little practical and having
some little practical experience, having worked on the farm, I was
wondering when the farmer raises alfalfa and clover and wheat and
com and potatoes and sugar beets and all farm products, and at the
end of the year he balances up his accounts and he has sold so many
tons of potatoes, so many tons of corn, and so on, and he finds that
. he has in the bank $50. He has paid all his bills, all the cost, trans-
portation, sales cost, and so on, and he has only $50 in the bank. It
seems to me it would be pretty difficult to go back and say that "My
cost of lettuce was so much, my cost of potatoes was so much, my cost
of sugar-beets was so much," because the whole activity has been
consolidated and worked as a unit in every branch of that industry.
Dr. Yntema. That is correct.
Acting Chairman King. It seems to me that is the important
question, what is the final result of his labors during the year on the
farm. In nearly every industry with its various activities, its various
chains that lead out from perhaps a common center, you are going to
have cost problems, of course, various other problems, but after all
the question is. Did you make any money or did you lose?
Dr. Keeps. Would I understand you to say. Dr. Yntema, that the
knowledge of cost accounting by the farmer and by the Steel Cor-
poration and their knowledge of the costs of individual products was
probably not essentially different? They are both essentially equally
ignorant.
Dr. Yntema. I am in no position to answer that question. I donH
know what the farmers know about their costs, and I am not qualified
to say how much the' Steel Corporation knows about the costs, I
mean the costs of various products. I would say this, however.
Any allocation of overhead costs to different products is arbitrary;
and as soon as you start breaking up your costs, allocating overhead
to different products, you obtain a result which is not useful and
which is suspect for the kind of purposes with which we are here
concerned.
Acting Chairman King. Proceed.
Dr. deChazeau. If what you were really interested in was whether
the Corporation did make a profit or did not make a profit, would
you have gone to all this trouble
Dr. Yntema. No; that is not the reason for, nor the objective of,
the analysis. I have stated our objective several times and I. don't
think a restatement would clarify it, so I should like to proceed with
the discussion. ,
In dealing with our analysis of steel prices, volume and costs, Dr.
deChazeau raised a large number of detailed questions. We appre-
ciate these criticisms and the careful scrutiny which this document
received. Both Dr. deChazeau and some committee members
attached great significance to the extrapolation of the straight line .
representing the relation of costs to volume, to the zero point of
production, and to this $182,100,000 of fixed costs determined thereby.
Th^ point was made that this extension beyond the range of data
might be subject to considerable error and that, if in error, it would
13662 CONCENTRATION OF ECONOMIC POWER
seriously affect the validity of the results shown by the study. This
is not the case.
Dr. Kreps. There are statistical means of measuring that error,
are there not?
Dr. Yntema. Let me go on and explain. This extension merely
gives a convenient method of breaking the total costs into two groups:
Fixed costs and additional costs per ton.
Referring to chart B-1 in "Exhibit No. 1409'V entitled "Relation-
ship Between Total Costs of Operation and Volume of Business,
1938 Conditions," instead of using this $182,100,000 w^e could just
exactly as well have ^started from this point and said the total costs
of producing 4,000,000 weighted tons of tonnage products shipped was
$405,000,000, and that proceeding b.eyond that poiut but staying
within the range of the data, that the additional cost of operations
per additional ton of products shipped was reflected by the way in
which tnis line rose, the slope of the line, namely, $55.73 per ton.
The question of the extension of this line beyond the range of the
data is entirely unmaterial to the point we are trying to make. It
merely gives us a convenient way of stating our results.
Dr. deChazeau. I would like to point out that that point was
greatly overemphasized in the discussion yesterday merely because
the question was raised. I indicated, I beheve, that the extrapolation
of the figure of $182,000,000 is of httle importance for our purpose.
It is rather the character of the regression line and the rate at which
it rises, and the important thing in my mind was the fewness of obser-
vations and the possibihty that with changes in the allocation of costs
you might get a different type of distribution; that is the important
thing, not the extrapolation. I should admit what Dr. Yntema says.
Dr. Yntema. Yes, Dr. deChazeau is entirely correct in this latter
statement. I should be in complete agreement with it. The import-
ant question with reference to additional cost is the slope of this Une.
Acting Chairman King. Will you identify it so persons reading the
record may determine the points to which you are directing attention?
Dr. Yntema. The important question is the way in which the
dotted Une in the chart rises with increases in volume.
From inspection of the points, I submit to jou that, as far as this
particular evidence is concerned, a straight hrie represents as satis-
factory a description of the data as it is possible to obtain.
Dr. Kreps. That assumes, first, that each point such as the cost
in 1929 is accurately estabUshed, that there were no errors, say, in
allowance for obsolescence made by the management in 1929 which
they were not aware of until 1931 and '32.
Dr. Yntema. I shall come to that point shortly. I should not
accept your statement as it stands.
Dr. Kreps. Second, ^ou have absorbed two degrees of freedom,
have you not, in producing that line?
Dr. Yntema. Quite so. It is the minimum possible number of
degrees of freedom that you can absorb. I don't think we should
brmg this in.
Dr. Kreps. Wait a minute. For any average to be useful, and this
is a form of average, you need at least 30 observations.
Dr. Yntema. You do not.
I ApiMndix, p. 13773.
CONCENTRATION OF ECONOMIC POWER • l^QQ^
Dr. Kreps. For instance, if one person has an income of a million
dollars, and the other has no income, the two will have an average
income of half a miUion dollars a year, will they not? Such an average
is misleading, is it not?
Dr. Yntema. No; that is not correct. You do not need 30 obser-
vations.
Dr. Kreps. Then you need to correct by advanced statistical
methods for the fewness of the number of your observations, do you
not?
Dr. Yntema. No. Let me put this in plain English. You do not
need 30 observations to make a reasonable case with reference to the
relationship between two variables such as dollars of cost and tons of
production. If these points in the chart B-1, were scattered widely,
I should say that any line drawn through them would have relatively
little reliability. If these points were scattered in such a way that
every one of them lay precisely upon a straight line, I should say that
there was almost certainty that the relation in question was a straight
line relationship. These points do depart slightly, but very shghtly,
from such a straight line relationship, and I submit that this straight
line represents if not a relationship of absolutely perfect reliability,
at least a relationship of rather high rehability.
Dr. Kreps. This, of course, is a summary chart. The imderlying
data do show much of the wide scatter that Dr. Yntema is talking
about. This may be an instance such as Dr. Warren Persons, one
of the initiators of this technique, frequently warned against, in
which the errors or residuals due to errors of fit are perfectly correlated.
Dr. Yntema. I should like to attempt to get some perspective with
reference to the cricitism of the cost analyses which have been offered
to this committee. I had the feeling yesterday, as I listened to the
discussion, that the points suggested were, from a technical point of
view, most interesting, and I was very pleased to have them called to
our attention.
Acting Chairman King. You mean the testimony of Dr. de Chazeau.
Dr. Yntema. Yes, that is correct. I had the feeling, however,
that perhaps those who hstened to the testimony might not be able
to see the woods for the trees, and I should Uke, if I can, this morning,
to put these things in their proper perspective.
First of all, I should like to call attention to various concepts of
cost. In the long run, when a businessman is considering whether
or not he shall build -a plant and engage in a business, all the costs are
additional costs, there is no overhead, with reference to a problem of
that type. So if you take a long-run point of view, all costs must be
regarded as additional.
Dr. Kreps. Would you like to make your observations on costs
after our witness on costs has presented his testimony?
Dr. Yntema. I think it is very important at this point to do this.
Dr. Kreps. Then please do so.
Dr. Yntema. Because Dr. deChazeau has raised a fundamental
issue with respect to the interpretation of costs.
Acting Chairman King. Proceed.
Dr. Yntema. There is no possible quarrel among us, I think, with
reference to the additional costs. All costs are additional in the long
run.
13664 CONCENTRATION OF ECONOMIC POWER
In the second place, if I interpreted the statements yesterday cor-
rectly, we were criticized because our costs were static, not dynamic,
that they did not reflect other elements which changed in the business
situation as the volume of production by the steel industry changed.
Then, if I understood the criticism correctly and some of the sug-
gestions which have been made this morning, we were criticized because
our costs were not static enough. Now, I submit to the committee
that you must choose which concept you want to use in the particular
problem, and I suggest, therefore, that we keep clearly in mind two
ideas: First, what you might call dynamic costs, which represent the
actual history of costs, that is, the actual costs involved at these
various operating rates,- without attempting to eliminate in any way
whatsoever the effect of changes in efficiency or the effect of changes in
wage rates and other factors.
Now, if you want such a picture, we have it here
Acting Chairman King (interposing). Identifv it.
Dr. Yntema. In chart 1 of "Exhibit No. 1416", entitled, "Total
Costs (Unadjusted) and Volume of Business ; United States Steel Cor-
poration and Subsidiaries." ^ On the vertical scale is plotted millions
of dollars of actual costs incurred by the Corporation. Along the
horizontal scale is plotted milhons of weighted tons of all tonnage
products shipped.
The fluctuations in total cost and in tons of products shipped —
actual fluctuations without any adjustment whatsoever — are reflected
in this chart. Now, if that is what you are interested in, how actual
costs did fluctuate with volume otoperations, here is the picture.
I am not advocating this as an entirely satisfactory picture because
the other dynamic elements in the situation did not stay the same in
different business cycles; I am not advocating this as a statistically
useful concept, but only as a representation of what did happen. If
you extend this line, you will find that the fixed costs were lower than
"those represented in the other chart — B-2 of "Exhibit No. 1409"^ —
that they amounted to some 120 millions of dollars.
The average slope of this line, if we may take a straight line to repre-
sent the data — the fit is not so good in this case— the slope of the fine
in chart 1 in "Exhibit No. 1416" ^ is such that the increase in cost per
additional ton of product shipped is approximatelv $54.51 . Now, that
is something of a coincidence, that the additional cost in this case of
unadjusted total costs turned out to be approximately the same as the
additional cost in the case of the adjusted figures.
Let me go on and point out furthermore that this average line
reflects the changes in cost at wage levels prevailing when the wage
rates were lower than -they are today, except for the last few years.
It is however, also probably true that the wage rates today are more
inflexible than they were throughout this period, so that in -the future,
if you made up a chart of this sort, some factors would tend to make
the additional dynamic costs higher and some of them lower than they
are shown in this chart. I do not put a great deal of reliance m it,
but if you want a concept of how costs do change in the business cycle,
without attempting to adjust for other factors, I submit this is the
best evidence which we have available,
' Appendix, p. 14030.
> Appendix, p. iSTTrj.
» Appendix, p. U039.
CONCENTRATION OF ECONOMIC POWER 13665
Dr. Kreps. Wouldn't you rather say, how the allocation of cost
changes?
Dr. Yntema. No; I should not.
Dr. Kreps. Might not this— —
Dr. Yntema (interposing). This shows you how the total costs
change. There is no question of allocation here except insofar as
costs may not have been charged into the years in which you would
like to see them charged. There is some variation possible in some
of these costs, in charging them to one year or another. I shall
come to that point very shortly in dealing with the adjustments we
have made.
Dr. Kreps. Is it the accounting policy, do you know, to charge
costs to shipments as shipped? Is there a uniform managerial account-
ing policy, in that regard?
Dr. Yntema. That is a big (juestion. What do you mean by
uniform?
Dr. Kreps. There are various ways of allocating depreciation,
depletion, and all the other items that are subject to managerial
discretion ana managerial policy?
Dr. Yntema. That is correct.
Dr. Kreps. Now, you can allocate those in various ways?
Dr. Yntema. Well, within some range.
Dr. Kreps. That is correct.
Dr. Yntema. And the range can be very narrow in some instances,
and in others relatively large.
Dr. Kreps. But your evidence shows here that the allocation of
cost as historically made varies uniformly with shipments, does it not?
Dr. Yntema. No; I did not say that.
Dr. Kreps. Is it not rather uniform?
Dr. Yntema. No; I said the costs vary
Dr. Kreps (interposing). Costs as determined and allocated by
managerial decisions determine how much of depreciation and deple-
tion and other items you
Dr. Yntema (interposing). I should like to get the subject clearly
in mind in that sentence. It isn't the allocation necessarily that
varies; that is something to be investigated. The total costs do vary
with the shipments.
Dr. Kreps. I see; that is what I wanted to get straight.
Dr. Yntema. The second question is a reference to allocation and
I propose to take that up later.
Mr. HiNRiCHs. When you say that these are actual costs, the busi-
nessman understands immediately what you mean. From the lay-
man's point of view, these things that you refer to as actual costs
represent, in the first instance and for the largest part of the total,
out-of-pocket expenses with reference to which there is no question,
plus an accountant's allocation which necessarily involves questions
of judgment, and there is in that fact some policy question in terms
of how high those costs are?
Dr. Yntema. Yes; that is entirely correct and I am glad to have
the question because it is one of the points to which I wanted to
address my remarks in the next few minutes.
Mr. Werntz. May I ask there, wouldn't it be possible for the
management to allocate cash maintenance costs between the years?
13666 CONCENTRATION OP ECONOMIC POWER
Dr. Yntema. Would you let me come to that in just a moment?
I think we will get a more systematic treatment of the subject if
you could defer the question for the time being.
ANALYSIS OF OPERATING COSTS
Dr. Yntema. As I said, I think it is useful to get some perspective,
and I suggest that as a means of procedure, we refer to Table 8 in
"Exhibit No. 1416'V this table being entitled, "Analysis of Operating
Costs into Components, United States Steel Corporation and Sub-
sidiaries."
Mr. O'CoNNELL. What is the name of the pamphlet?
Dr. Yntema. It is "Exhibit No. 1416."
Mr. O'CoNNELL. Yes; but the name?
Dr. Yntema. It is entitled, "An Analysis of Steel Prices, Volume,
and Costs — Controlling Limitations on Price Reductions."
For the.period 1927 to 1938, the costs in the various classifications
are there aggregated. The total costs for that period excluding certain
miscellaneous items not connected with operations, amounted to
approximately $7,900,000,000. Of that total, two components ac-
counted for a very large proportion, namely, pay roll, $3,614,000,000,
which accounted for 45.8 percent of the total. (Percentages are not
in the table, I am supplying them from other computations.)
The item "Other expenses" on the extreme right of the table ac-
counted for 38.1 percent. Those two expenses, pay roll and other
expenses, accounted together for 83.9 percent of the total expenses.
Dr. Kreps. Could you give us some illumination upon what "Other
expenses" are? It is such a large item, almost as large as pay rolls.
Dr. Yntema. Yes; may I come to that in just a moment? I want
to take up in' detail the adjustment of these items.
Of the remainder, depreciation, and depletion, concerning which
you may hear much in the discussion, accounted for only 8 percent
of the total; taxes, other than Federal income and profits taxes and
social-security taxes, accounted for 5.4 percent of the total; the other
items are minor: interest, 1.5 percent; pensions, 0.9 percent; and
social-security taxes, 0.4 percent.
I suggest, therefore, that we focus our attention primarily on those
items which constitute the bulk of the costs, because it is only sub-
stantial errors in those items which could seriously affect the results
obtained in our adjusted costs.
From chart B-2 in "Exhibit No. 1409," also appearing in "Exhibit
No. 1416," entitled, "An Analysis of Steel Prices, Volume, and Costs,"
as chart 13 ^ in that document, the chart title being "Composition of
Total Costs of Operation in Relation to Volume of Business," it is
possible to get visually some impression as to the relative importance
of these components. It is easy to see that the pay roll is the biggest
item in the total; that 'the goods and services purcliased from others
constitute the next biggest item in the total; and that in comparison
•with these two, the other items are of relatively minor significance.
Let us consider the pay roU. The pay roll represents the out-of-
pocket expense of the Corporation for salaries and wages. Of that
total, the salaries account for a relatively small proportion. That is
> Appendix, p. 14040.
» Appendix, p. 14067.
CONCENTRATION OF ECONOMIC POWER 13667
represented in one of the charts which has been submitted to you,
chart E-4 of "Exhibit No. 1409," ^ and if you care to have it shown,
we can offer it on the easel.
Mr. HiNEiCHs. Pardon me, but are bonuses to officers included
there? I don't know the Corporation poHcy.
Dr. Yntema. Yes. I am informed that there have been no bonuses
in the Corporation since 1930. If I am incorrect in that, I should like
to have some official of the Corporation correct me. Before that time,
bonuses must have been a very small proportion indeed of the total
pay roll. I think that that would be an inconsequential item.
Acting Chairman King. Proceed.
Dr. Yntema. I do not think there is much question, therefore, about
the allocation of the pay roll as among the various years with reference
to whether or not the individuals rendered the services in those years.
There may later be some discussion about pay roll for maintenance
purposes and that we can take up when the point is raised.
Dr. deChazeau did call into question a point which is dealt with in
this next chart. It is E-17 in "Exhibit No. 1409," ^ entitled, "Earn-
ings Per Hour and Production, United States Steel Corporation and
Subsidiaries, April 1937, to November 1939." On the vertical scale is
plotted the earnings in cents per hour, and on the horizontal scale,
millions of tons.
This represents the relationship between average hourly earnings of
all employees and production, that is, the monthly production of
rolled and j&nished steel products.
What is apparent from this chart is that the average' earnings per
hour do not go up or down as production changes. During this
period the wage rate level did not change. Therefore, given a certain
wage rate level, the average earnings per hour do not fluctuate up and
down with the rate of operations. It is, therefore, appropriate in
adjusting the pay roll factor for variations in the wage rate, to divide
the total pay roU by the average hourly earnings to find out what the
pay roU would have been if the wage rate did not change. Perhaps
I should repeat that again. We have here in our original series total
pay roll per year. From the period 1927 to 1938, there were variations
in wage rates. Our adjustment consists in dividing the total pay roll
by the average hourly earnings, and as the chart which I have just
shown indicates, that does not involve dividing this total by something
which is related to volume of production. That gives us an approxi-
mately correct adjustment for variations in wage rates. I don't claim
it to be perfect, but it seems to me on the basis of the evidence that it
is a reasonable adjustment.
Mr. Appert. Dividing by average hourly earnings converts the
respective pay rolls into the man-hours for that year and then multiply-
ing by the average earnings for 1938 gives us the estimate of the pay
roll under 1938 wage rates.
Mr. HiNRiCHS. What you are saying in effect is, you are taking a
record of man-hours worked which you needed in order to arrive at
average hourly earnings and are multiplying by the prevailing average
hourly earnings in 1938.
Mr, Appert. Because the evidence indicates the average hourly
earnings are not dependent upon the rate of operation.
1 Appendix, p. 13871,
» Appendix, p. c-sgg.
13668 CONCENTRATION OF EtTONOMIC POWER
Mr. HiNRicHS. All you needed of this chart was to demonstrate
that within the period 1937-39 average hourly earnings had not been
significantly related to volume; you are inferring from that period and
the behavior during that period that that same thing was true during
the earlier periods which is a reasonable inference from the materials
that you have, but might be itself subjected to check.
Dr. Yntema. Yes, that is correct. If you would examine the
detailof the adjustment which appears in table 12, "Exhibit No. 1416,"
"An Analysis of Steel Prices, Volume, and Costs," ' you would find
some further light on your question. You will notice that for the
first 5 years, 1927 tlirough 1931, the average earnings stayed almost
constant, although there were fairly considerable changes in vol-
ume over that period, and you could by inspection of the adjusted
figures find out whether the relation of those observations to volume
was roughly the same as some of the others. That is merely an addi-
tional check upoT; the calculations.
We don't claim perfection for this. If wc could have made a direct
index of wage rates and salary rates, we would have preferred to do it
that way. We have used the best teclmic we can, and we leave it to
your judgment as to how satisfactory you think it is.
The second point to which I should like to call your attention is
the component of goods and services purchased from others. (Refer-
ring to chart B-2 of "Exhibit No. 1409," entitled "Composition of
Total Costs of Operation in Relation to Volume of Business.") ^ The
adjustment of that item is, I think, the crudest part of our statistical
analysis. Let us be quite frank about it. I should have been much
better satisfied myself if it had been possible to take the goods and
services purchased from others, to break them up by classes, and to
make an index number of the prices for each one of those classes and
then bring the aggregate together again after adjustment. It was
not possible to do that.
The goods and service's purchased from others are composed of such
items as raw materials, supplies, freight-in on materials and supplies,
public utility services, professional services, and so forth; it is an
extremely hf^terogeneous group.
Some of those items fluctuate in price rather widely in the business
cycle. Some of them fluctuate not at all. Now, this is what we have
done: We simply took the bull by the horns and decided, "We will
divide this item into two parts, half and half; one-half we will deflate
by dividing it by the Bureau of Labor Statistics Index number for all
commodities, excluding food and farm products, and the other we will
leave as it is, and then wc will add the two results together."
That is a very rough, crude statistical procedure. Let's be perfectly
frank about it. If we had been able to do it in a more satisfactory
way, we should have taken that procedure. Let me point out, how-
ever, that even though you had deflated the total of that item by an
index number, the Bureau of Labor Statistics index number of whole-
sale prices for all commodities, excluding only food and farm products,
the resulting difference in the additional cost per ton would only have
amounted to about $3. You get some idea, therefore, of the possible
error wliich might be induced because of the incorrectness of our
adjustment. It is not serious.
1 Appendix, p. 14(V42.
'Appendix, p. 13775.
CONCENTRATION OF ECONOMIC POWER 13669
Mr. Reynders. $3 per ton?
Dr. Yntema. The additional cost would have varied if we had over-
corrected by $3 per ton. If there is an error due to this, my estimate
is that it must be a relatively small item, even at that. This, I should
say, is one of the crudest parts of our analysis. If I were to criticize
this myself, that is one of the points to which I should call attention;
even at that I don't think it vitiates the general results, the general
order of magnitude which we show here for additional or variable costs
and fixed costs.
Dr. Kreps. In terms, however, of your average output, 10,000,000
tons, $3 a ton is $30,000,000 profit, is it not?
Dr. Yntema. That is not very large in relat'on to the total cost.
Dr. Kreps. And also not very large in relation to net profit?
Dr. Yntema. That is not a correot interpretation because if the
additional costs are reduced, the fixed costs are raised, so that it isn't
$30,000,000 difference in profit. That is not a correct interpretation.
Dr. deChazeau. There is a question still, of course, to be discussed
as to the possible allocation of goods and services purchased in one year
with relation to volume which would tend to bias your curve in the
direction of increasing your apparent variable cost.
Dr. Yntema. Both ways.
Dr. deChazeau. Possibly both ways.
Dr. Yntema. Yes, and let me point out why. What is the effect
of inventory adjustments in bad years?
Dr. deChazeau. Are you asking me the question?
Dr. Yntema. Yes.
Dr. deChazeau. You have the inventories and the data; I suggest
that you tell us.
Dr. Yntema. The effect cf inventory adjustments in bad years is
to increase the cost in the years of low operations.
Dr. Kreps. And decrease the cost in the years of high operation.
Dr. Yntema. Yes, and that makes the curve too flat and, therefore,
makes the additional cost too small. I mean that particular item gives
rise, you see, to a bias of exactly the opposite sort from this which
you have been discussing.
Dr. Kreps. Have you taken care to avoid that bias?
Dr. Yntema. No; we have not adjusted for that. We recognize
there are some elements of bias one way and some elements of bias
the other way.
Mr. Hinrichs. Where are inventory adjustments in other expenses?
Dr. Yntema. They would be included in all other expenses of ship-
ment. What we had to do was this: We had only the consolidated
statement available since we had separate figures on pay rolls, and
we took those out. There are some slight compensatory errors in the
distribution of these expenses between pay rolls and goods and services
purchased from others. One may be a little too large in some years
when the other is a little too small, but the effect of inventory charges
into cost of goods sold would appear here in goods 'and services pur-
chased from others.
Mr. Hinrichs. While I have interrupted, you said in answer to a
question by Mr. Kreps that in view of the magnitudes which were in-
volved that run up to a billion dollars a year, pretty nearly, at times,
that $30,000,000 was a relatively small item.
Dr. Yntema. In relation to the costs; not in relation to profits.
13670 CONCENTRATION OF ECONOMIC POWER
Mr. HiNRiCHS. No, no; that a question of estimating the location
of one of these points, $30,000,000 one way or another, was relatively
small against the total magnitudes that you have plotted there.
. Dr. Yntema. I'm sorry that I made that statement. I was in-
correct in so doing. That is not true, of course. The 30 million item
in the case of a biUion-dollar cost would be 3 percent and in the case
of $500,000,000 cost would be 6 percent, and that is a substantial
deviation.
Mr. HiNRiCHs. You are too good. I thought I was going to have
to stop you from letting 30 million go by because I don't want that
much.
Dr. deChazeau. May I clear up one thing? In connection with
the inventory adjustments, you are dealing, as I understand, not with
costs of goods manufactured but with cost of goods sold.
Dr. Yntema. That is correct.
Dr. deChazeau. When would the cost of goods taken into inven-
tory be charged, then, in terms of your total expense?
Dr. Yntema. Any inventory adjustments, writing down of inven-
tories at the end of the year, would increase the cost of goods shipped,
would increase the total cost charged into operations in that year.
We come finally to this index of shipments which has been the
subject of considerable discussion this morning as well as yesterday
afternoon. If I were asking questions about this index, I would ask
this question: There is a proportion of the products of the Corporation
not covered by this index. I should raise the question whether or not
the production of those products varied in proportion to the production
of the products we have included. We do not have satisfactory
measures of quantity of shipments for all those other products. We
do, however, have revenues, and this I will say, that based upon a study
of the fluctuation in revenues of these other commodities not included,
and upon a study of the revenues of those commodities which are
included, the inclusion 'of those additional commodities would not
make any substantial difference in the index nuraber. . And I would
say this further, that if you set 20 different statisticians or economists
to work on the construction of this particular index number, they would
come out with nearly; identical answers. If there is a possibility of
agreement on something, I should say it is here. I am quite willing
to defend this particular index, as a reasonably satisfactory measure of
production, to the last ditch.
There were many points that were raised which I have not discussed.
I don't like to leave those points without answer to them. On the
other hand, I am extremely reluctant to take the time of the committee
to discuss details which in my opinion are relatively insignificant. I
don't think they are really important in the total. I am perfectly
willing, however, to answer any questions, if any member of the
committee or any witness for the Government would like to raise such
question with reference to the significance of any item in the cost
analysis.
Acting Chairman King. Dr. Kreps, are you through with the
witness?
Dr. Kreps. Yes. The witness has made advanced comment
upon the presentations which follow, and I suggest, therefore, that the
presentations which follow be regarded as presentations of the other
side of the case. '
CONCENTRATION OF ECONOMIC POWER 13671
Dr. Yntema. May I say, Dr. Ejeps
Dr. Keeps (interposing). It was impossible to avoid it.
Dr. Yntema. Let me say this. Because of the fact that I had jfive
documents prepared by Government witnesses, I have not, as a matter
of fact, had an opportunity yet to read Mr. Taitel's statement. All I
have learned is through comments from Aljr. Appert, so in responding,
I could not have directed my remarks pointedly to the argument that
Mr. Taitel is about to make.
Mr. HiNRicHS. Will Mr. Yntema be available again?
Dr. Keeps. I should like to recall him after Mr. Ezekiel and Mr.
Taitel have presented their case.
Mr. Reyndees. Will that also be true in regard to Dr. deChazeau?
Dr. Keeps. No, Dr. deChazeau has an original presentation of
the Department of Justice which does not strictly belong to Dr.
Yntema's analysis. Dr. deChazeau may be dismissed.
Mr. Reyndees. I won't be able to be here this afternoon, and I
just wondered whether I would have an opportunity to clear up
some points in Dr. deChazeau's testimony.
Acting Chairman King. Proceed.
Mr. Reyndees. One statement you made was to the effect, as I
understand it, that while a 10 percent reduction in price might not have
a substantial effect upon demand, that a 20 percent reduction would
have that effect.
Dr. deChazeau. Yes, that statement was made not with relation
to any analysis, you understand, of demand, but merely derived from
the point that in the short run the substitutabihty of one raw material
for another, steel for other products, would be likely to be costly and
likely to be very small, and that particularly where the price of com-
plementary goods is not altered, is assumed constant, the increased
deniand for a given product may have quite an elasticity. The point
is made that with a very large decrease in the price of steel the demand
might have a different elasticity than with a small decrease because
of those obstacles to substitution and to increased demand involved
in unchanging costs of complementary goods.
Mr. Reyndees. On the other hand, your argument visualized a
reduction of 20 percent in steel prices. Isn't that true? I mean
leaving it to the discretion of the committee sitting here, it would
seem that a 20 percent reduction in steel prices would be something
that was reasonable in the order of possibility.
Dr. deChazeau. The percentage used was purely as a matter of
illustration, merely to draw distmction between a small price decline
and a large, price decline. There was no measure there of what I
should consider possible.
Mr. Reyndees. Would you wish to withdraw that suggestion of a
20 percent, or would you still have it in the picture?
Dr. deChazeau. I think I have withdrawn it as a statement of
what I consider to be a reasonable reduction in steel. My point was
merely that a large price reduction may have quite a different effect
from a small price reduction. -^
Dr. Keeps. I was going to ask Dr. deChazeau, if one takes a par-
ticular product, such as automobile sheet steel, hasn't there actuall.y
occurred more than the 20-percent reduction to which you referred?
Wasn't that the example which you had in mind, or one of the ex-
amples you had in mind?
13672 CONCENTRATION OF E(?ONOMIC POWER
Dr. deChazeau. There are examples of such price reductions, but
my point was a purely theoretical point, as to the significance of the
price elasticity of demand in the short run.
Mr. Reynders. The instance that Dr. Kreps suggested was, I
think, influenced very largely by improvement in the means of pro-
duction, and that gets into the continuous mill, which we know has a
very big reduction.
Dr. Kreps. Which is one of the benefits of low costs, particularly
when producers are forced to lower costs.
Acting Chairman King. Suppose we let the witness proceed. We
will not interrupt you with questions until we get through.
Mr. Reynders. Well, I think for the general understanding, it
should be clearly appreciated that the distribution of the sales dollar
of steel, taking the year 1938 ^ was divided 45 cents for labor, 38.9
for goods and services purchased, as shown there, and 8 percent for
taxes, making a total of 91.9 percent going out, which is out-of-pocket
expense for any manufacturer of steel.
Now, the committee here is engaged upon an investigation which
will give the Congress a correct picture of the possibilities of this
industry, and, for that reason, I am very anxious that no impression
should go forth that any reduction of that sort is in the range of possi-
bilities, taking it through the whole range of price.
Dr. deChazeau. May I answer that briefly, that the total expendi-
tures made by the Corporation for these various items do not represent
what are the additional costs, associated with a given product. Marry
of those costs are constant.
If your statement is that the Steel Corporation, in view of those
expenditures, could not take a 20-percent reduction in its revenue
without loss, I should admit it, but I should say that this would have
nothing to do with the possibility of reducing the price of a given steel
product 20 percent.
Mr. Reynders. Well, that is possible along the line indicated,
where the means of production might have undergone a very definite
change.
There is another point, Mr. deChazeau, that came up in the latter
part of your testimony, that had to do with your suggestion, or at
least you said you disapproved of the size of the units which had
grown up in the steel industry and regarded that — I am not in con-
troversy with you at all, but it is from your practical contact with the
steel industry — I understand you have visited many plants and you .
know the general situation, you are familiar with the location of raw
materials such as iron mines, coal, and so forth — in that connection,
what thought, may I ask, have you given to what is a suitable unit
of a steel plant of a steel corporation?
Dr. deChazeau. I have given it considerable thought, but in order
to give any statement with regard to it woidd involve a very complex
investigation wliich we were not able to make.
Mr. Reynders. It is really a very simple one. You can do it in
3 minutes.
Dr. deChazeau. The point-to which I made reference was merely
this, that when a plant expands through an increase in its production
facilities, then there is a self-corrective applied in that expansion
through increased cost, in a competitive situation.
> Referring to chart A-S of "ExliiMt No. 1409," appendix, p. 13757.
CONCENTRATION OF ECONOMIC POWER 13673
When a plant expands through merger, which in part reduces the
competition, the same corrective does not apply. Therefore, I think
that one can be suspicious of expansions by merger. Now, when one
presents evidence that the total costs of the Steel Corporation are so
large that they cannot make a profit, whereas other companies less
weU integrated or less thoroughly integrated, less extended in other
directions, are making a profit; then it does not seem to me that the
case has shown conclusively that the price of steel is too low. It
may be that the Corporation is too large for over-all efficiency.
Mr. Reynders. I am not taking the size of the Corporation par-
ticularly, because that is something in being, but for the perspective
of this committee, having raised the question at all, I think there should
be on the part of one who is so experienced as you are in the industry,
the ability to indicate what would be a reasonable size of a steel enter-
prise.
Dr. deChazeau. Well, I should say there that the reasonable size
of any integrated steel plant is so large in my estimation that even if
you broke up all of these corporations into those sizes, the number of
sellers would be so small in the market that each one would have to
take into consideration liis full effect upon his rivals. If that gives
you an answer without putting a dollar figure to it^in other words, I
feel that the fmidamental conditions in the industry require size.
Wlien a continuous mill alone requires an investment of from twelve
to twenty-five million dollars, when blast furnaces require an invest-
ment of around four and a half to five milUons, when your steel fur-
naces require an investment of upward of $600,000 — that is, any
integrated firm is hkely to have a very large investment and is likely
to be a very large plant. Now, whether the actual size of a plant is in
excess of efficiency, I am not qualified to make a statement.
Mr. Reynders. Well, there is one question: You would regard it as
necessary to have diversification of products, that is, you couldn't con-
template a steel plant that devotes itself to nothing but structural or
nothing but plates or nothing but tubes?
Dr. deChazeau. I could not contemplate an integrated plant which
confines itself to a single product. The investment is too large to
locate itself with relation to any given market or any given product.
The operating characteristics of blast furnaces and steel-making
furnaces require something close to capacity operation while they are
in operation, and therefore, in order to get a balance of utilization over
good years and bad and with shifts m demand, you need a multiplica-
tion of rolling facihties. For that reason, although you may establish
a nonintegrated firm close to a given market, or even possibly a semi-
integrated firm, I doubt whether it is possible to locate a fully inte-
grated firm with relation to a given product market.
Mr. Reynders. Well, getting down to figures, and what diversifi-
cation of products means, you are familiar with what is the annual
output of a continuous mill layout today?
Dr. deChazeau. Well, of course, it varies with the mill. A flat
rolled production of output, capacity output, up to 800,000 tons —
some of them are reported even up to a million tons.
Mr. Reynders. Now, such a plant, running to capacity, would re-
quire from 800,000 to a million tons of ingots, wouldn't it?
Dr. deChazeau. It would probably require more because of con-
version losses.
13674 CONCENTRATION OF ECONOMIC POWER
Mr. Reynders. Yes. Now, if you add to that, for instance, a
structural layout, structural and plates, those two items together
would be about the same capacity as the one of the continuous mill?
Dr. deChazeau. What you mean by structural, I take it, is a roll-
ing mill for shapes?
Mr, Reynders. I-beams, especially the broad flange beams
Dr. deChazeau (interposing). You don't mean a fabricating unit?
Mr. Reynders. Oh, no; entirely rolled products.
Dr. deChazeau. Well, now you are getting into a field in which
it seems to me there are men present here who are better qualified
to testify. I cannot. It is, after all, a fairly well-known fact as to
what the range hi cost iu a contuiuous mill is. When you get me
down to special mills, I am not prepared to quote you a dollar figure.
Mr. Reynders. I am only leading up to what the reasonable size
of the steel plant is, and that is, I think, something which this com-
mittee is very much interested in. I am merely bringing up the
various items because they are simple, extremely simple, and I think
you will find that if you" wish to diversify, iucluding structural ma-
terial and plates, you will have a gain of about 600,000 tons of an-
nual capacity, • and no plant of that kind would operate unless they
had a barn which would hold several hundred thousand, tons. Then
you have the category of pipe and wire, which* together might mean
a half nullion tons. If you idd those together you have something of
the order of 1,000,000, wel over a million tons of finished products,
perhaps a million and a qi: .rter, in fact much more than that — I beg
yoiu" pardon, it is really up to 3,000,000 tons when I add up these
various figures. That, then, would be necessary if you had a
diversified organization.
Dr. deChazeau. But those figures are probably very excessive, as
indicated by the much lower capacity of Lutegrated mills such as
Inland's mill in Chicago, or National Steel.
Mr. Reynders. Inland's is about the size I am talking about.
Dr. deChazeau. Yes. I haven't the figures before me.
Mr. Reynders. I can figure that; they are probably the size.
Dr. deChazeau. But it comes to a large figure. It is in the order
of a million.
Mr. -Reynders. That is the point I wanted to bring out. I had
3,000,000 tons here for the plant that I have outlined, and that is
about the size of Inland. I think there is Youngstown Sheet &' Tube.
Dr. deChazeau. Is that total finished roUed capacity or the ingot
capacity?
Mr. Reynders. That is the ingot capacity. That is the common
divisor. I think that Youngstown Sheet & Tube is about the
same size as Jones & Laughlin, so it would seem to me perfectly proper
to have on record here that a concern of about that size is not un-
reasonable; that is, when you are talking about units and diversified
units, you would reach some such size as we have here, and on the
basis of a reconstruction of cost today, if you know what that would
be for a ton, annual capacity.
Dr. deChazeau. No; there, are men better qualified to give you
that figure, but I would like to make this comment: In terms of the
actual development of mills you have reached that capacity. Whether
that capacity constitutes the minimum wliich is possible with efficient
operations docs not remain clear, or rather is not proved, and that is
why I hesitate to comment completely on those figures. I should
CONCENTRATION OF EXX>NOMIC POWER 13675
suggest, however, that the mere order of size in individual plants and
integrated groups, assuming that businessmen are interested in
eflficiency and increasing their profits through efficiency, would indicate
that the order of size is large, but I wouldn't comment on a 3,000,-
000-ton or a 2,000,000-ton or even a 1,000,000-ton size without much
more data than I have available.
Mr. Reynders. There is another element that enters into this,
Dr. deChazeau, and that is geographical distribution and geograph-
ical markets to be reached. A plant of the kind indicated here, if
located in one particular locality, would of course have restricted
markets and even going to this size you are still to an extent a local
concern, not covering the entire area of the United States.
Dr. deChazeau. May I say in connection with that that our study
of the distribution of products will throw considerable light on that
relation of markets to given producing areas ; although it will not give
a conclusive answer to the question, it will at least throw hght on the
actual distribution from a given producing area.
Mr. Reynders. My object here was to have in the record and for
the benefit of the committee an adequate idea as to what size really
constitutes and what it means in the steel industry. This type of
plant, if constructed on today's prices, would be of the order of about
a himdred dollars a ton, and that would then reach the size of about
$300,000,000.
Dr. deChazeau. That is right.
Mr. Reynders. That was what I had in mind, Mr. Chairman.
Mr. Wooden. I have a question I shQuld like to ask of Dr. Yntema.
Doctor, do you think any clear or accurate conclusion can be drawn
as to the feasibihty of a reduction in steel prices based upon a cost
study which includes the cost of such things as cement, the operation
of common carriers, and the operation of coal mines?
Dr. Yntema. After all, the operation of the coal mines when the
coal is used in the production of steel is just as much a part of steel
production as the processes, nearer the final products.
With reference to the operations of the railroads, insofar as they
carry the products used in the making of steel, they are part of the
steel-making processes.
Mr. Wooden. Insofar as they are common carriers?
Dr. Yntema. No; insofar as they carry the products used in the
aaking of the steel, they are part of the integrated steel process.
• Mr. Wooden. And what about cement?
Dr. Yntema. In the case of cement you have another industry.
I am not famihar with the technological development there. I i^der-
stand that the cement plants do to some extent use a byproduct of the
steel industry, but the poiut that is relevant, I thihk, is that the total
operations of the cement plants are scarcely a drop in the bucket in
comparison with the other items It wouldn't vitiate any findings
which I presented here.
Acting Chairman King. It is important to have iron ore too, is
it not?
Dr. Yntema. it is. necessary in integrated operations, of course.
Acting Chairman King. In the production of steel. Would there
be any objection to a steel company obtaining not only its coal supply
but its iron-ore supply as a part of its integrated activity? Would
that not make it perhaps to the advantage of the consumer ultimately?
124491— 41— pt. 26 7
13676 coNCEN;rRATiON of economic power
Dr. Yntema. Let me say with reference to questions of this type
that I am not a steel man and I am not competent to speak on ques-
tions of integration. I simply made a study of some phases of this
subject. I do not wish to commit myself on matters outside my
knowledge.
Acting Chairman King. The committee will take a recess until
2 o'clock.
(Whereupon, at 12:25 p. m., the committee recessed until 2 p. m.
of the same day.)
AFTERNOON SESSION
The hearing was resumed at 2 p. m. upon the expiration of the recess-
Acting Chairman King. Are you read)'^?
Dr. Kreps. Yes.
Acting Chairman King. The committee will be in order.
Call your first witness.
Dr. Kreps. Dr. Mordecai Ezekiel.
Acting Chairman King. Come forward, please. Doctor, will you
hold up your right hand? Do you solemnly swear that the evidence
you shall give in this hearing shall be the truth, the whole truth, and
nothing but the truth, so help you God?
Dr. Ezekiel. I do.
Acting Chairman King. State your name and residence.
Dr. Ezekiel. Mordecai Ezekiel, Washington, D. C.
TESTIMONY OF DR. MORDECAI EZEKIEL, ECONOMIC ADVISER
TO THE SECRETARY, DEPARTMENT OF AGRICULTURE, WASH-
INGTON, D. C.
Dr. Kreps. What position do you hold. Dr. Ezekiel?
Dr. Ezekiel. I am economic adviser to the Secretary of Agriculture.
Dr. Kreps. But as I un'derstand it, you are appearing in no respect
as a representative of the Department of Agriculture, but only as an
expert on statistical and economic matters, particularly on price
analysis?
Dr. Ezekiel. That is correct. I am not appiCaring for the Depart-
ment, but am appearing as an expert in statistical and economic
analysis.
Dr. Kreps. How long an experience have you had in the tech-
niques of statistical Research employed here by the Steel Corporation?
Dr. Ezekiel. I have been working approximately 20 years using
these methods, and have a textbook, Methods of Correlation Analysis,
which presents most of the methods Dr. Yntema has used.
'Dr. Kreps. As a matter of fact, it is a standard reference work in
this field. Tl^ese methods were first applied in what field. Dr.
Ezekiel?
Dr. Ezekiel, The methods have been used very widely in the field
of agriculture, particularly in the economic analysis of changes in
prices and production of farm products, in the Bureau of Agricultural
Economics jind agricultural colleges throughout the country.
Dr. Kreps. And have you utilized these methods in your economic
analyses concerning what might stimulate employment and the like?
Dr. Ezekiel. Yes, I have used them to some extent in industry.
They have been applied inuch less in industry, however, than they
CX)NCENTRATION OF ECX)NOivIIC POWER 13677
have in agriculture. I have given a good deal of attention also to
the general problem of unemployment and to the steps that industry
might take in dealing with the problem of unemployment and low
production.
Dr. Keeps. In that connection, it seems to me I remember two
books of which you are the author, one entitled "Twenty -five Hundred
Dollars A Year," and the other, "Jobs For All."
Dr. EzEKiEL. That is correct. I have two recent books on the
general subject of correcting industrial unemployment.
ANALYSIS OF DR. YNTEMA's STATEMENT CONCERNING PRICES, VOLUME,
COSTS, AND PROFITS
Dr. Kreps. Would you care to comment on the general significance
of the results which Dr. Yntema has secured in which he has utilized
your methods and their bearing, as you see it, on the general problem
of unemployment?
Dr. Ezekiel. Yes, that is what I am now prepared to do.
Dr. Kreps. You have a statement, have you?
Dr. Ea^KiEL. Yes, I have a statement summarizing my views on
these points.
The material that Dr. Yntema presented may be summarized jn
three broad statements:
First, that if the steel industry were to reduce its prices at any
time, the percentage gain in sales (due solely to the reduction in
price) would be at most no greater th^n the percentage reduction in
price, so that in consequence the gross income of the steel industry
would show no increase as a result of the price reduction.
Second, if the sales of steel were to increase at any time, the larger
output would lead to a reduction in production costs per ton, but
those costs per unit would not fall as rapidly as output rose, so that
total costs would increase as sales rose.
Third, that reduction in price would always reduce the profits or
increase the deficits of the Corporation. That result follows from
his argument, since total income would not increase with the increased
sales, whereas total costs would increase.
I have been over rather carefully the statistical technics used both
in the cost analysis and in the price analysis, and there are ^ good
many individual weaknesses in technic. Other witnesses will discuss
the detail weaknesses in the statistical analysis made. The points
that I am going to take up here are not differences of analysis, but
the interpretation made of the results obtained. What I piopose to
show is that even for the time accepting the results of the analysis in
the Steel Corporation documents, that the conclusions they draw from
those results do not necessarily follow.
Acting Chairman King. Do you mean to state that the factors
which they have taken into account do not exist or that they have
placed too much stress upon one factor and too little upon another?
Dr. Ezekiel. No, I am accepting for the time their measurement
of the effect of a change in price on quantity sold as measured by their
statistical analysis, but I shall draw attention to the fact that there
arei other concurrent elements in their analysis which they have
ignored in- their subsequent interpretation. Even though their
13678 CONCENTRATION OF ECONOMIC POWER
statistical analysis is taken as correct, the conclusions they reach are
not necessarily the right conclusions.
Acting Chairman King. You don't mean to state that those in
charge of the business, interested, of course, as they are in its develop-
ment and in the protection of the capital invested and of the labor that
is involved, you do not mean to state that they do not try to ascertain
the material condition, if I may use that expression, the utiUtarian
outlook, and then adjust their business to meet the situation, the rise
in prices or the decline in the market, in the consumptive demands
and so on?
Dr. EzEKiEL. I am not criticizing at all the technics of the industry
as it has operated. As I understand it, in fact the industry has oper-
ated to tins time without having these research results. These re-
search results have just been worked out and apparently, therefore,
are something different from the practical basis upon which the in-
dustry is operating. I am taking the research material which is in-
troduced as testimony to show that the industry had been justified in
not following a lower price policy and am reanalyzing the material to
show how quite different conclusions can be reached from the same
material.
The statement that they made, that it would never pa^ to reduce
prices, rests upon three assumptions that are made in their analysis,
two of which were not expHcitly stated in Dr. Yntema's presentation
or in the document submitted. Those assumptions are these: First,
that a reduction in the price of steel, while it might increase soiflewhat
the quantity of steel sold, would have absolutely no effect on the gen-
eral level of business activity.
Second, that reductions in the price of steel would have no effect
upon tlie level of prices involved in the cost of producing steel.
Third, that there is no possibility of bringing about concerted action
by the steel industry and other industries by which not only steel
prices but prices of many other products could all be reduced at the
same time, so as to get a much larger stimulus to activity than could
be produced from the effect of reducing steel prices alone.
Dr. Yntema referred 'briefly to this third point yesterday, but not
to the other two. .
Acting Chairman King. You wouldn't mean by tha,t that there
ought to be sort of a combination between the steel industry and
others with a view to raising or lowering prices?
Dr. EzEKiEL. That is the point I will discuss toward the end of
my statement.
Acting Chairman King. Running perhaps head-on with the Fed-
eral Trade Commission or with the Sherman antitrust law. You are
not advocating that, I assume.
Dr. F^ZEKiEL. Toward the end of my statement I discuss that point, .
so if you will permit me, I will put it off until I come to it.
Taking up each of these three points separately, the first is as to
the assumption made on the price of steel. The statistical method
.used in determining the over-all effect of a price reduction on the
q^uantity of steel sold is the method that is known as multiple correla-
tion analysis, a method by which you examine the simultaneous results
of several different factors all affecting another factor. By that
method it has beeil determined that the year-to-year changes in steel
output and sales in the period since the first World War have been
^ CONCENTRATION OF ECONOMIC POWER 13679
explained largely by three factors. The first of those factors is the
level of industrial activity, or the level of industrial profits and of'
national income. The second factor is the price of steel, and the third
factor is the long-time trend in the demand for steel.
Now, in measuring the effect of changes in each one of those factors
on the quantity of steel sold or shipped, the analysis assumes Ihat
only one factor is permitted to change at any one time while all the
others are held constant. That is, they have measured how much
change there would be in steel sales, if the price of steel was either
increas3d or decreased, while business ac'vity did not change and
while the trend factor was held constant, and then they have measured
how much a change in business activity would affect the sales of
steel, if meanwhile prices were held constant and trend were held
constant.
From that statistical analysis the conclusion has been reached that
if the steel prices were reduced but the other factors were held con-
stant, with no change in business activity and no change ia trend,
that at best sales would increase no more rapidly than the price fell
in percentage terms. That is what the elasticity of not more than
one means. But that conclusion in their analysis holds true only if
it were possible for steel prices to be reduced while industrial activity
meanwhile remained unchanged.
I want to consider for a moment \/hat does- actually happen when
steel prices are changed. Assume, for example, U. S. Steel Corpora-
tion would reduce its price, and as a result of its reduced price increase
its production and sales as they have shown. Not merely U. S. Steel
would reduce its price and output, but all the other steel companies
would reduce their price and increase their production and sales.
As a matter of fact, the analysis they present is not the analysis of
the tuect of price on U. S. Steel sales, but is an analysis of the effect
of price on sales of the entire industry. As a result of producing
more steel, they employ more men, they buy more materials, and they
pay more for the costs of the industry, so the disbursements for the
whole great steel industry for wages, material, power and freight,
would rise. Not only would disbursements in the steel industry
rise, but more steel being produced and purchased would mean that
more steel was being used in automobiles and construction, con-
tainers, machine tools, and all the other various uses. So that would
mean these other industries would also be employing more men, be
increasing their pay rolls, be shipping more products and increasing
their disbursements. As steel production increases, that carries with
it a change in the production of all other, industries. Their assump-
tion has been that all the time while the price of steel was reduced,
industrial activity has been unchanged, while only the price reduc-
tions affected steel sales.
I have brought with me a chart which shows how absurd it is to
assume that steel output could change without a corresponding change
in the level of industrial activities as a whole. This chart
~ Dr. Keeps (interposing). Dr. Ezekiel, for the purpose of identifi-
cation, would you like to have this chart introduced into the record?
Dr. Ezekiel. Yes; I would.
Dr. Keeps. Mr. Chairmaii, I should like to offer this chart, en-
titled, "Relation of Industrial Production, Excluding Iron 9,nd Steel,
to Steel Sales," for the record.
13680 CONCENTRATION OF t\?ONOMIC POWER
Acting Chairman King. It may be admitted, without objection.
(The chart referred to was marked "Exhibit No. 2183" and is
included in the appendix on p. 14119.)
Dr. EzEKiEL. What this fehart shows is the relation of changes
in steel shipments year by year to changes in the level of industrial
activity. The way the chart is prepared is this: Each year is repre-
sented by a single dot and that dot is placed both according to the
million tons of steel shipped that year, and according to the industrial
production excluding iron and steel; that is, according to the index
of industrial production in industries other than the steel industry.
Now, starting with a couple of low years after the War, thereafter
there is a very consistent relation between changes in steel shipments
and changes in the level of activity in other industries. Every single
time that there was an increase in steel output, there is an increase
in the activity of the other industries. For example, here from 1932
to 1933, steel shipments increased from around a little over 10,000,000
tons to about 16,000,000 tons, activity in the other industries in-
creased from about 73 to 82.
Dr. Kreps. Dr. Ezekiel, if you had limited yourself to the same
period as Dr. Yntema limited his analysis the dots for 1920 and 1921
would not appear; isn't that correct?
Dr. Ezekiel. Yes. Those years were, not used in his cost analysis.
Dr. Kreps. Then the conformance of the dots to the line would
be considerably closer; is that correct?
Dr. Ezekiel. Yes. You can see, except for those 3 years, how
closely all the other dots adhere to the line, showing the average
relation between change in steel sales, steel shipments and a change
in the general level of activity in industries other than steel.
Acting Chairman King. May we not postulate increase in steel
resulted from the increase in the other activities, instead of the
steel being the motivating, the moving cause, that the other industries
and activities were the ones, and they had their influence upon steel
and caused the increase or rise in steel?
Dr. Ezekiel. So far as the figures presented in the chart are con-
cerned, they do not tell us which is cause and which is effect. It
could be as you suggest, either way.
Acting Chairman King. You don't intend to say that steel is the
lever which determines the rise or the fall of the business activities^,
industrial activities, of our country?
Dr. Ezekiel. No, I am not saying that steel is the sole cause of
changes in business activity, but the point that I am making is that
changes in steel production and changes in activitv in industries
other than steel are so closely associated that regardless of which is
cause and which is effect, we have not in fact had changes in one with-
out changes in the other. So that to assume that you could make a
material increase in steel production without at the same time having
material increase in otlier industries, is to go against the record of the
past period.
Acting Chairman King. Wo have been led to believe by some of
the earnest and sincere and intelligent advocates of agriculture that
agriculture was the principal, the most important, industry and that
the condition of the farmers determines the condition of the rest of
our economic life; that if agriculture was prosperous, everything would
be prosperous, including the steel industry. You are not trying to
CONCENTRATION OF ECONOMIC POWER 13681
minimize the importanee of agriculture, you being in the Agricultural
Department?
Dr. EzEKiEL. It is true there as here, that large demand for
agricultural or farm products is associated with high buying power
in the cities, and high buying power on the farm in turn helps main-
tain city activity. But there, as here, there are two factors that run
together, and saying which one is cause and which one is effect, well,
all one can say is that they do move up and they do move down
together.
Dr. Keeps. Dr. Ezekiel, for purposes of removing all elements of,
shall I say, obfuscation, concerning this chart, is there anything
technically that can be said against this chart that is not equally
true of the other chart; that is Dr. Yntema's chart is taken to prove
that the prosperity of the steel industry is dependent on industrial
production, such an inference stands or falls on the same anah^tical
and statistical grounds as this chart; if one is admissible, the other is
admissible, and if one is not admissible, neither is admissible?
Dr. Ezekiel. Yes. The type of analysis is the same here as there,
that we are- reasoning from the fact that certain variables move or do
not move together.
Acting Chairman King. There may be some industries though that
are prosperous, profitable, whereas other industries are not pros-
perous, o^-er the same period?
Dr. Ezekiel. Well, this chart, of course, does not attempt to
measure profits. It simply takes the observed facts as to volume of
output and says that the record shows that when there is a large
volume of steel output, there is also a large volume of output in other
industries; when there is a low volume of steel output, there has
always been a low volume of output in other industries. It does not
go beyond that point.
Acting Chairman King. Well, does not the increased output in
other industries lead to an increased output in steel?
Dr. Ezekiel. Opinions can differ on that. I believe, however,
that the record will show that the changes in steel production have
been more responsible for changes in the total of industrial activity
than have the changes in most other industries; that steel production
is more variable than the total output of other industries; and that
the heavy capital goods, such as steel, may play a predominant part
in the business cycle. For example, the output of agricultural pro-
ducts varies only 10 or 15 percent from high to low, whereas the output
of steel during the same period has varied from as low as 10,000,000
tons to as high as 35,000,000 tons.
Acting Chairman King. Well, I can't help but believe that the
prosperit;^ of agriculture determines in part at least the prosperity of
the steel industry. I recall when a boy, working on the farm, when
our products, farm pioducts, brought a good price, we would buy
more wagons and more reapers, more mowers, and more steel com-
modities which we used in connection with agricultui-al development,
and the purchase of those steel commodities, of course, furnished an
additional market for the steel industry, so that the steel industry
prospered when we prospered; that is to say, their output increased
when our output of agricultural commodities had a reasonable price
13682 CONCENTRATION OF ECONOMIC POWER
I don't know which was the chicken and which was the egg there,
but I know that the prosperity of the farmers brought prosperity to
other branches, including the steel industry, and I cannot quite con-
ceive of the theory that the s'teel industry is the lever that moves our
entire economic and industrial fabric.
Dr. EzEKiEL. Well, if you will pardon me, sir, I am not trying to
say that it is the sole lever, but I am trying to say that it is one of the
levers. It is quite true, as you state, that when farmers have income
and can buy more tractors and more automobiles, that helps the
demand for steel, but it is also true that only a relatively small pro-
portion of steel goes into such uses, as compared to the total amount
of stegl going into railroad cars and the construction industry and all
the many other industrial uses of steel, as well as farm machinery and
the farm portion of automobiles.
Acting Chf».irman King. That would mean, of course, that the
railroad industry, which has been one of the largest consumers of steel,
when it is prosperous, it adds to the prosperity of the steel industry?
Dr. EzEKiEL. Quite So.
Acting Chairman King. And if the railroad industry is in the
doldrums, as it has been for a number of years, in part due to the
development of trucks, waterways, and other means of transporta-
tion, it affects quite materially the profits and business of the steel
industry.
Dr. EzEKiEL. The sc<e point I am making at this time is that if a
reduction in steel prices does increase the quantity of steel sold, that
means more products are being hauled over the railroad, more steel
is being used, and there is an associated increase in other industries.
That is as far as I am going at the moment.
Mr. HiNRipHS. Your materials here are essentially the same as the
materials that Dr. Yntema presented when he mentioned the relation-
ship between industrial activity and steel output, and you are merely
calling attention to tl ». fact that by virtue of the fact that a change in
steel price stimulates to some extent volume of business, there is a
secondary reaction, so that if his figure of three-tenths of a percent
were to be taken as correct it would inevitably be somewhat larger
than that, you are not saying how much larger at this stage of the
game, but that there must be some addition beyond the thing that
comes from the processes that Mr. Yntema used by virtue of this
repercussion from the stimulated industrial activity. Is that correct?
Dr. EzEKiEL. That is the point to which I am now coming. For
these statistical purposes it is very convenient to measure the separate
effects of business activity and of steel prices on the amount ot steel
production and shipments, but after that has been done it is quite
erroneous to assume that you can reduce steel prices and increase
steel sales without at the same time increasing business activity.
The analysis that w^-s presented in the documents in Dr. Yntema's
statement yesterday indicated that cbanges in business activity wei^
about 10 times as important as were changes in prices, in accounting
for changes in steel production. The tables in his statistical analysis
show that the percent of the variation in steel consumption explained
by business activity was of a magnitude about 10 times as great as
the percent of the variation explained by steel prices. But then when
he came to estimate how much mcrease in production you get for
rcd\!ctiou m price he took into account only the eJect of price on sales
OONCEKTRATION OF ECONOMIC POWER 13683
and made no allowance whatever for the effect of the associated
increase m business activity on steel, so that in considering price
alone and in failing to consider the associated changes in business
activity, the analysis presented has left out of account the most
important single factor. In consequence, it has gravely understated
the increase in sales which might be expected to follow from a given
reduction in price. For that reason it is quite possible that a reduc-
tion in steel prices would cause a very much larger increase in sales
than the computations presented assumed, and it is therefore likewise
quite possible that sales would increase so greatly with a lower level
of price that the gross income of steel producers would rise as the
price of steel fell or was reduced.
The second weakness in the analysis lies in the assumption that
the price level of the factors entering into steel production would
remain the same regardless of the price of steel. It is well known, for
example, that steel scrap is an important material in steel production,
and it is also customary that scrap prices and pig iron prices move
fairly closely together, so that cheaper iron and steel would mean
probably less expensive scrap. Furthermore, the analysis of costs
presented by the Steel Corporation took into account the wholesale
price level as one of the factors with which to adjust the "other costs"
item of costs. They show that the material costs move with the
general price level. Steel and iron are major industrial products,
so that it is not unreasonable to assume that changes in the price of
steel and iron may exercise a price leadership not only on steel scrap
but throughout many industrial prices. As a matter' of fact, the
record of the past 20 years does show that steel prices and the prices
of other industrial products have moved together quite closely through
most of the past 20 years, have had very similar general price move-
ments. So to the extent that both scrap and other industrial prices
were influenced by reductions in iron and steel prices, the costs of
producing steel would be reduced, at least relative to that which they
have shown. So even assuming the accuracy of their cdst analysis, it
is still quite possible that a reduction in steel prices and an increase
in steel output would cause materially less increase in Corporation
expenditures than their calculations have shown. You see, I am not
criticizing their calculations; I am cricitizing their failure, in estimating
the increase in total cost of the Corporation, resulting from an increase
in output, to take into account that the lower iron and steel price at
which they output was produced would also mean lower prices for some
of the products that they were buying, and therefore not as much
increase in total cost as they have shown, based upon the 1938 level of
prices.
Acting Chairman King. That is to say, you would be criticizing
the steel management for not producing more steel when they got down
to 30 or 40 percent of their capacity, or 25 percent, even though at that
time they couldn't find a market for the amount of steel which they
were producing at that small percentage of capacity. You think they
ought to have continued their output, even tahave lowered the prices,"
for the purpose of stimulating activities in other enterprises?
Dr. EzEKiFL. What I am considering is, had they lowered their
price, how much of an increase in sales that would have made and how
much of an increase in total costs they would have faced. They have
presented materials which show that had they lowered their price they
13684 CONCENTRATION OF ECONOMIC POWER
would not have increased their sales any more than they reduced the
price, so they would not have increased their gross income, and that
tliey would have increased their total costs materially so they would
have lost more by doing it. What I am attempting to show is that
their analysis of their own data does not necessarily lead to that con-
clusion, that they have left out of account certain very important
factors, first on the demand side and then on the cost side, which makes
the result that might have been obtained materially different from
that which they have presented to you.
Mr. Walter White. On the demand side, Dr. Ezekiel, doesn't
it make a difference whether you make your price reduction at a time
when industrial activity is increasing or whether you make it at a
time when general industrial activity may be static or declining?
Dr. Ezekiel. I am not an expert in the steel industry, I don't
know the answer, but I might say that no material has been presented
by the Corporation to show that there is such a difference. The fact
that they have never tried reducing tbeir prices during periods of
depression to see if they could move more material, that is, reducing
prices in the magnitude of the type discussed here, means we don't
have any experience to go on. We do know in the field of agriculture
that when farmers reduce tbeir prices very greatly they can as a result
of those low prices move into consumption just as much as they moved
into consumption before. We do know that throughout the depres-
sion except when the drought came along, farmers continued to pro-
duce almost as much as they had been producing before and moved
into consumption that continued output.
Actmg Chairman King. Is that true of cotton?
Dr. Ezekiel. Yes; by and large it is true of cotton. Even for
cotton the very low prices resulted in a sustained domestic consump-
tion of cotton from 1936 to date which was not so much below the
predepression levels.
Acting Chairman King. You are not taking into account the mil-
lions of dollars with which we have had to subsidize the cotton
industry by reason of the decline in the consumptive demands, closing
of markets, the lack of demand for cotton.
Dr. Ezekiel. The greater part of that loss of consumption was in
the foreign market rather than the domestic, and that is outside of
the thing I was trying to present now.
Acting Chairman King. Except by way of analogy. If I had time
I would call your attention to the dechne in the production of zinc and
lead and copper and all of the metals during the period when the
prices went down until copper was sold at 4 and 5 cents a pound, per-
haps at a loss of 4 or 5 cents a pound, and yet with those low prices
it didn't stimulate the consumptive demand of copper.
Dr. Ezekiel. It is quite true, sir, that most industries have oper-
ated the same as steel has operated, and that is the precise point why
this question of industrial price poUcy is so important, because it is
true that most industries follow the practice of maintaining their
prices with only at most moderate reductions. You mentioned some
that reduced more, but most of the heavy industries did not. Auto-
mobiles sold in '33 at much the same price as they sold in '29, and they
took out the reduction in demand in a greatly reduced output, turning
their workers out into the street to take care of themselves, or let the
Government take care of them. Farmers take the depression in low
CONCENTRATION OF ECONOMIC POWER 13685
prices; industry takes the depression in low employment and their
workers stay idle.
Dr. Keeps. I am not sure that I imderstand you correctly. You
don't mean to say that there has not been a reduction in absolute steel
prices during the period from 1929 to 1932, do you?
Dr. EzEKiEL. No, the point I was making is that the reduction in
price has been very small contrasted to the reduction in output, and
very small contrasted to the reduction in the prices of farm products
and many other industrial products.
Dr. Kreps. When we speak of steel prices being relatively high, we
mean relative to what?
Dr. Ezekiel. Well, it depends on the base we use. We may say
steel prices are high relative to 1929, or we may say they are high
relative to other products.
Dr. Keeps. Wlien you said that the steel prices did not dechne,
that we had no experience of a decline in steel prices, you meant no
experience of a decline in steel prices which was greater, relatively,
than the decline in prices in general?
Dr. Ezekiel. That is correct. I might say also we have very little
experience with a decUne in steel prices even as great as in the prices
of other products.
Dr. Keeps. In general it is true that steel belongs to an order of
producer's goods which as Dr. Mills has pointed out in a three volume
analysis rose relatively in price during the depression, became rela-
tively more expensive. Is that correct?
Dr. Ezekiel. Yes, you had to use more of other maferials to buy
steel; that is, a ton of steel in 1932 and '33 cost a great many more
bushels of wheat, or bushels of corn, or bales of cotton, or even suits
of clothes, than it did in 1929.
Dr. Keeps. And that is the important point for the problem of
investment, and the Uke, is it not?
Dr. EZEKiEii. It is one of the important points; I wouldn't say it
was the exclusive one.
I have considered the price analysis presented and the cost analysis
presented, and I have shown that even accepting the results secured
by the accounting and statistical technics used in the studies presented
by the Corporation, that there still remains the probabiUty that a re-
duction in price would produce more increase in income and less in-
crease in cost than the Corporation has assumed. In consequence, a
reduction in price would cause less increase in deficit than the Corpo-
ration assumes and might even produce an increase in profit.
This reinterpretation of the results of the analyses presented may
prove of interest to many of the people in the steel business who may
have been puzzled "by the dilemma to which the original analyses led.
As these analyses stood, they sum up in this conclusion, which was
stated on page 38 of "Exhibit No. 1416," entitled "An Analysis of
Steel Prices, Volume and Costs," and I quote:
That unless the elasticity of demand for the product exceeds 1 by a substantial
margin, the theory that price reduction in and by -itself would produce profits
through increased volume is utterly fallacious, not only for the United States
Steel Corporation, but for any business or any industry. * * * Application
of the theory of increased profits through price reduction could thus only produce
loss to the enterprise which adopted it.
13686 CONCENTRATION OF ECONOMIC POWER
If that statement were turned the other way around it would equally
have proved that any industry situated as the steel industry is situ-
ated, would increase its profit by increasing its price and reducing its
output, since to increase the price would bring an increase in total
income, while the reduced volume would produce a aet reduction in
costs.
So that many other industries in the same position at the same
time could all make more money by producing less product and by
selling it at a higher price. That is the conclusion which the steel
analysis leaves.
They don't carry it to that point, but presumably the largest profit
would be reached when 1 ton of steel was produced by the industry
and sold at a price- of a billion dollars a ton, since that would involve
practically no cost except the fixed cost and you would have all profit.
It is obvious if all corporations attempted to follow this theory at the
same time that not only would most workers be unemployed, but
nobody would be able to pay the high prices for the product.
FAILURE OF HIGH PRICES TO PROMOTE BUSINESS
Dr. EzEKiEL. So the theory that all businesses can make more
money by all producing less and all charging higher prices at the same
time obviously cannot be true. It may be comforting to those who
have been troubled by this fundamental dilemma to which the results
lead that these conclusions rest upon a very narrow view of the in-
dustrial process and only hold true on the assumption that the steel
industry operates without effect upon the rest of the economy.
The reinterpretation I have just presented may be summarized
briefly in the statement that the conclusions of the Steel Corporation
have made no allowance for the fact that with changes in the prices of
iron and steel products and the volume of steel operations there would
also be associated changes in the prices of other products and in the
volume of operation of the economy as a whole.
So when allowances are made for that broader effect of changes in
steel prices, the conclusions reached may be quite different from those
reached by the Corporation. In fact, they would have to be very
greatly different unless the further conclusion is to follow that private
enterprise organized in the corporate system can only lead, if they
follow the profit motive, to .smaller and smaller output and more
and more unemployment, and that the only way that increased pro-
duction and fuller employment can be reached would be by substituting
some other method of economic organization in place of the present
right of corporate managers of industry to decide what prices to es-
tabhsh. It is not beheved that either the stcd companies or thif
committee would wish to reach that further conclusion.
Mr. Walter White. Isn't your conclusion in that respect depen-
dent upon the nonexistence of competition to affect prices.
Dr. EzEKiEL. I am taking the statements made by the Steel
Corporation for the industry in their presentation as representing the
way that the industry operates. They discuss in their presentation
the possibility of increasing prices or decreasing prices as if it lay
within their power to make the choice quite without the effect of
competition.
CONCENTRATION OF ECONOMIC POWER 13687
Mr. White. That is not my recollection of the testimony that
Dr. Yntema gave. I don't think you were here, but that point came
up. He was asked whether 50 percent in reduction in production
and 50 percent increase in price would make more money for the
Corporation and he said "yes." When asked why they didn't do it,
he said competition in the industry, so I feel that is apphcable to
industry as a whole on that particular argument.
Dr. EzEKiEL. I am sorry, I didn't hear Dr. Yntema's verbal
discussion. I have read the various documents, particularly the
document which compares the cost andlysis with the price analysis,
and as I read that document, it read in terms of "if this price policy
were followed, such-and-such results would follow; if this price policy
were followed, such and such results would follow," and included the
statement which I quoted to the effect that it would never pay any
industry like steel to reduce its price. As I pointed out, the same'
data just as readily lead to the opposite conclusion that it would
always pay any industry like steel to raise its price. I -haven't con-
sidered how many other industries are situated like steel, but have
pointed out that if there were many, and all of them followed the
principle that they could make the biggest profit by always raising
price, it would be pretty hard on everybody else.
There is a third range of possibiUty which lies rather outside the
scope of the material considered in the Steel Corporation's statement.
Dr. Yntema, I believe, had just one passing reference to it; it was
indicated in the third assumption that I stated in the beginning.
Throughout their analysis the Corporation claimed thafr changes in
steel prices have Uttle or no effect on the demand of the final consumer
for the products finally made from steel. That result foUows, they
claim, since the price of steel makes up such a relatively small fraction
of the cost of finished automobiles, houses, tractors, locomotives,
watches, and other products. It is rather interesting to note exactly
that same argument, that the price of their product makes only a very
small percent of the finished product, is made by lumber men in
explaining why reduced lumber prices would not increase the sale
of houses, by members of building unions in explaining why reduced
per hour wage rates, for bricklayers and carpenters, would not increase
the sale of houses, and by farmers iu explaining why reduced wheat
prices do not increase the sale of bread.
Dr. Keeps. Dr. Ezekiel, do you mean to say that we could use this
same multiple correlation technic which Dr. Yntema has exploited to
prove an inelastic demand for steel and establish by identical methods
an iuelastic. demand for, let's say, a particular group of craftsmen of
union labor in the steel industry?
Dr. Ezekiel. No; that is not my argument at this point. I am not
referring to the technic of the statistical analysis here, but rather to the
computation that since steel, say, makes only one-tenth of the cost of
a finished automobile, if you reduce the price of steel 10 percent, you
would reduce the automobile price only 1 percent, and that doesn't
make any difference.
'Dr. Kreps. Isn't that the identical argument of the plasterer
when he says, "Why shouldn't I get $26 a day? It makes a difference
of only a fraction of a percent in the cost of a house. There will be
no stimulation of the demand for houses. People are not going to
buUd any more houses if I take only $18."
13088 CONCENTKATION OF ECONOMIC POWER
Dr. EzEKiEL. Exactly.
Dr. Kreps. Therefore, if labor argues that way, what do we call
that type of reasoning? Is that the lump of labor fallacy, so-called?
Dr. EzEKiEL. I am not trymg to attach terms to these; I am just
trying to point out the argument involved.
Dr. Keeps. Isn't it true that is called the lump of labor fallacy?
Dr. EzEKiEL. Yes; it is in the case of labor.
Dr. Kreps. Therefore, when it is argued that general social policy
should be based upon computations of this sort, actually you have a
lump of business fallacy on advanced mathematical stilts.
Dr. EzEKiEL. Yes; I guess you can caU it that; I realize that as
the steel industry is now organized, and' as the activity legally per-
mitted corporations of this country is now circumscribed, there is
probably no wa}"^ by which changes in steel prices can be coordinated
with changes in other prices, although it might be interesting to point
out in this connection that as far as I am aware, the anti-trust laws
are solely directed against combinations in restraint of trade, and
that so far as I know, there has never been a case to test whether
combinations for the expansion of trade would be similarly illegal.
QUESTION OF CONCERTED ACTION TO EXPAND PROJ)UCTION
Dr. EzEKiEL. But as bus'ness corporations do operate, it is no doubt
true that there is no existir ; means by which reductions in steel prices
could be brought about co currently with reductions in cement prices,
lumber prices, freight rates, automobile prices, furniture prices, houses,
and perhaps even wage rates per hour, though not necessarily incomes
per year of workers engaged in some of the more highly paid trades
such as steam fitters, carpentiers, and bricklayers. It is perfectly ob-
vious that if some means could be found by which concerted reductions
could oe made in the prices of many products at the same time, the
additions of these savings all down the line would add up to a very
much greater reduction in price of the final finished product than would
be possible if only a single industry made the change in price. So in
periods of great economic contraction such as that which occurred in
1930 to '32, or again in late 1937 and '38, it should be possible for
such concerted reductions jn price to be accompanied by concerted
expansions in output. The Steel Corporation itself has shown that
such an increase from a low to a high output would greatly reduce the
cost of output per ton produced.
There have been many discussions of the possibihty of concerted
action by industrial units which have seemed to assume that the only
way that such concerted action could be brought about would be for
the Government itself to take over the ownership of the industries,
either through pubHc ownership, to use the American term, or social-
ism, to use other terms, or else to assume that the only way that can
be brought about would be for the Government to take* over complete
control of the production policy of a corporation through some sucb
form as fascism. The argument made on this point has always
seemed to lead to that conclusion — that either you would have
socialism or fascism, either complete ownership or complete govern-
ment regimentation in order to bring about any such considered action
by industry. But it seems to me that there are other possible technics
by which a democratic goveniment can find ways to cooperate with
industrial producers in assisting them to develop concerted programs
CONCENTRATION OP ECONOMIC POWER 13689
of production and price change which would not involve either of
those extreme forms of economic organization. We do know that in
agriculture the farms of this country are still o-wTied by individual
farmers, and we know that the. programs of faTm production are
workeci out democratically with ue participation and approval of
individual farmers, with a very great deal of planning from the
bottom up through township committees, (*ounty committees. State
committees, and regional committees • so that the farm programs
that are in effect are not programs imposed from Washington, but are
programs worked out from grass roots up, representing the decision
of farmers as to what they feel needs to be done after they go over the
fact.
Yet at the same time through these programs the producers pf the
major export crops working through the A. A. A. and associated
programs have found a means of taking concerted action with respect
to the acreage and price of their major crops without involving either
socialism or fascism. So the fact that it hag been possible to work out
democratic procedures and carry through concerted action in the field
of agriculture may suggest that parallel democratic procedures could
be -developed in the field of industry, .and that through those demo-
cratic procedures, production might jbe increased, prices might be
reduced, and employment raised on a larger scale than individual
industries have been able to estabhsh and maintain during recent
years.
Mr. HiNRicHS. Pardon me, but have I misunderstood what has
happened imder A. A.- A? Was the procedure one whpch reduced
prices and increased prodiiction, or was it merely the fact that dis-
cussions had taken place democratically, in which you are suggesting
the parallel?
Dr. EzEKiEL. No; it is obvious that A. A. A. has not been aimed
at the further expansion of agricultural production in the face of the
very low market for farm products which have resulted from a low
industrial production. I am bringing the A. A. A. experience into
evidence merely to show that i.t is possible to bring producers together
on concerted programs of action, and that those concerted programs of
action can be devised through democratic procedures.
Mr. Wooden. Dr. Ezekiel, do you think that method of procedure
is applicable and practical in the case of an industry where you have
9 or 10 concerns controlling 80 percent of the output?
Dr. Ezekiel. Yes. I beheve it would be infinitely easier to sit
down around a table with 8 or 10 men and work out what each indus-
try should do than to go out into the field to develop an agricultural
program by discussing it with two or three million cotton producers
or one milhon wheat producers, and get those several million producers
to take concerted action.
Mr. Wooden. It would be decidedly easier to get concerted action?
Dr. Ezekiel. Yes. It should be easier in industry. If you did get
concerted action among each of the great industries, each of them
expanding its production in the proper proportion, then you would
have larger markets and you would have a material reduction in the
final price which would make it possible to move the products into
consumption.
Mr. Wooden. Was not cooperation among the farmers for the pur-
poses of decreasing production and increasing price?
13690 CONCENTRATION OF ECONOMIC POWER
Dr. EzEKiEL. Yes, sir; that is what Mr. Hinrichs has just pointed
out.
Mr. Wooden. You would favor that in the case of all industries?
Dr. EzEKiEL. No. As matters now stand, if a single industry takes
concerted action to decide what it should do, whether that single
industry be agriculture or whether it be the petroleum industry or the
anthracite coal industry or the bituminous coal industry, operating
as a single industry, the thing it is most likely to do is to say, "The
market for our product is only so much; national activity is so much,
and this limits the market for our product. Therefore, we as a single
industry, must be careful not to produce too much for our market."
That, of course, is what the steel industry has done in the past, when
it has so greatly cut down its production.
What I am proposing, however, is that a iiumber of the great
industries, each at the same time, expand their production so that
their iotsd market will not be determined by the present relatively
low production and national income of the whole country, but will
be determined by the much larger production and national income
which they are all proceeding to create. One industry by itself
cannot expand national production as effectively as can many indus-
tries working together. Only concerted action taken among many
great industries at the same time can, by that action, make a bigger
market for the product.
Mr. Hinrichs. Pardon me, but you said one industry cannot
expand national production, and a little earlier we had one industry
that was exj)anding national production. You mean that one
industry, acting by itself, cannot expand national activity to a
sufficiently large extent to benefit or to be certain of benefiting
financially from that expansion in national activity? You don't
mean to go back on your earlier statement that that 3 percent expan-
sion on a 10 percent reduction price is going to have in and of. itself
some beneficial effect on the rest of the economy, though it might
be bad for the stockholders of the Steel Corporation?
Dr. EzEKiEL. Yes. My statement should be that one industry
by itself cannot produce anything like as much effect on national
income as can concerted action by many industries. You are quite
correct:
And now, the testimony that the Steel Corporation has put in,
that it would never pay them to reduce prices, suggests that private
corporations, if they continue to operate in the next few years with the
same philosophy as that which has controlled their operations in
recent years, will never fmd ways to solve the large and continuing
unemployment. And yet, if private enterprise is to survive, business
must find a way under private enterprise to solve the problem of
unemployment and to provide a continuing rise in the standard of
living — more goods to consume for each day's work.
Now, in concluding this statement, I would like to indicate that I am
quite aware of the fact that no way has yet been developed and put
into action by which the officers of the Steel Corporation or any of
the great corporations similarly situated could take such concerted
action to reduce prices and increase production in many industries
concurrently as that which has just been suggested above, and I
would also like to indicate that the problem is a much larger one
than the problem of prices and production alone, because expansion
CONCENTRATION OF ECONOMIC POWER 13691
in production and in consumption and employment can be continued
and maintained only if the buying power that is made available to the
workers of the country is increased rapidly enough so that the con-
sumer demand for the various products produced rises in proper
proportion to what is being produced. In other words, if you increase
production and don't give your consumers, the people of the country,
enough income to buy the output, you cannot maintain increased
output.
On the other hand, if you do increase buying power in proportion
to the increased output, then that will call into action still further
increases in production and further expansions in plant capacity.
The problem of devising a program of concerted industrial expansion
so as to secure a proper proportioning of the changes in prices, wages,
production, employment, investment, and expansion in plant and
equipment, is much more intricate and extensive than can appro-
priately be discussed at this point. However, as I understand the
job which the Temporary National Economic Committee is attacking,
it is to suggest ways by which the productive facilities of this country
can be used to the fullest extent so as to secure a steadily rising
standard' of Hving and so as to insure that the steady increase in
technological efficiency and in industrial arts and sciences shall produce
hereafter a steadily rising level of consumption for all our citizens,
instead of producing, as they have so often up to this time, a steadily
increasing proportion of our citizens who are barred from normal
participation in society.
While the Steel Corporation alone is not in a position to bring about
such modifications in laws and methods of business organization as
might lead toward this result, the T. N. E. C. is in a position to
consider such changes. . It is for that reason that I have presented this
final discussion of ways in which concerted price reductions might
bring about higher profits and higher employment even if individual
corporations such as the Steel Corporation, are not now in a position
to undertake such concerted operations.
Acting Chairman O'Connell. Are there any questions?
Mr. HiNEicHS. I have two further questions. You speak of
concerted action to achieve price reductions. I suppose you are dis-
tinguishing there between concerted action and simultaneous action.
It is conceivable that under conditions of competition between small
producing units, it would not require willful action of any significant
extent to reduce prices. Industrial producers would find themselves
in much the same position that farmers find themselves in. Prices
of farm products, as I understand it, have gone down more pr less
simultaneously, not because of any concerned action by farm pro-
ducers, but because of the effect of competition against large producing
capacity in periods when national income goes down. Is that not
correct?
Dr. EzEKiEL. Yes; that is quite correct; but if all of our industries
were organized on as competitive a basis as most farm production is
organized, or was organized prior to A., A. A., and for that matter, as
competitive as most cotton textile proaucers are organized and many
dress goods producers, then when people are unemployed, when
demand is low, competition would force prices to fall together, and
you would get as a result of competition, an expansion in employment
and expansion m production. But as our economy is organized today,
124491 — 41— pt. 26 8
13692 CONCENTRATION OF ECONOMIC POWER
not only steel but many other great industries, seem to be so organized
that prices do not fall when demand goes down, or at least, prices do
not fall anything like as rapidly as they do in other industries. In-
stead, output is reduced as demand falls.
Mr. HiNRicHS. That is, your prices under competitive conditions,
willy-nilly would normally fall to something approximating the vari-
able costs of the marginal producer; that is, if we were to assume
that these variable costs oi $55 a ton represented marginal costs,
irrespective of whether or not it is a good thing for the owners of a
particular business, you would expect large amounts of unused
capacity to drive prices down to something approximately close to
that $55 level. Is that correct?
Dr. EzEKiEL. Yes; we had it in agriculture. Prices fell far below
the level that covered the computed cost of all the costs involved in
the business, but farmers produced and sold in spite of that because
there wasn't anything else they could do under competition.
Mr. HiNRiCHS. Now, the corollary of what you have been saying
with reference to concerted action to reduce prices and expand pro-
duction, if I understand you correctly, is that any concerted action
to maintain prices, while it may be beneficial to the particular industry
involved, is by just that extent deleterious to all of the rest of the
economy. All the other units that are also dependent upon the total
volume of national income and industrial activity suffer from the
curtailment that is involved in the maintenance of high prices. Is
that correct?
Dr. EzEKiEL. Yes; that is correct and is borne out by the experience
of the last depression. During the last depression, beginning in 1929,
each of these industries that have only limited competition sought to
restrict their output so as to. maintain their price, but each of them
suffered from the fact that many other industries were doing the same
thing. As a result, demand as a whole was collapsing, and even with
restricted output, these industries still had difficulty maintaining
price.
I would like to go back to a couple of the other points you made
earlier, just to clear up the issues raised. You were quite correct
that I used the term "concerted action" as opposed to "simultaneous
action." Many of these industries do have such a concentration of
corporate control, have production in the hands of such a small
number of producers, tli^t anything like free competition just cannot
take place. The only way they can take action of the sort I have
described is if they definitely plan it, if they definitely get together
and arrange, both in each controlled industry and between the different
industries, to move together on a program to reduce price and increase
production.
May I make one more point on this question that Dr. Hinrichs
raised? You asked if the fall in price in competitive industries would
not tend to produce that result — increased production — automatically.
In that connection it is interesting to contrast what happened to the
steel industry itself after the very high prices of the immediate post-
war period, in 1919 and 1920, and after the prices of 1929.
After the war-inflation boom, when demand fell off in 1921 and 1922,
there was apparently much more competition in steel then than there
is now, because steel prices did come down very, very rapidly. Other
prices came down even more rapidly. The decline in nominal steel
CONCENTRATION OF ECONOMIC POWER 13693
prices from the high in 1919 to the low in 1921 was as great as in all
commodities, but the decline in actual mill net yield was not so great,
according to the data filed by the United States Steel Corporation.
The depression of 1921-22 was over very promptly, at least com-
pared to this past depression. After the war boom, the post-war
depression involved a rapid d^op in price of almost all industrial
products, and brought about a prompt increase in production and
increase in employment, so that by 1923, new high levels were being
made in production. But during the depression of 1929, industrial
prices did not fall as greatly, and instead of having a short depression,
even though a hard one, with a prompt recovery, the recovery didn't
come and didn't come, and stUl has not come in full measure.
Mr. Wooden. If, as you say, it is impractical to expect. competition
in an indu'^^try such as steel, and if no one of the 9 or 10 producers who
control collectively some 80 percent of the output wiU reduce prices as a
matter of competition, what reason is there to expect that they will do
it if they are permitted to take concerted action?
Dr. EzEKiEL. Well, I haven't suggested that merely by permitting
them to take concerted action, it would necessarily result in their
doing so. ' In fact, I was discussing not merely concerted action in the
steel industry, but concerted action in a number of great industries at
the same time, in the steel industry and the automobile industry and
the cement industry and the glass industry and in a number of other
industries.
Mr. Wooden. Under some form of Government control, you mean?
Dr. EzEKiEt. Under some form of Government participation, suffi-
ciently to insure that the action taken by industry would be- in the
general welfare. I have discussed elsewhere various forms of organi-
zation that might be used to bring about that result.
Dr. Keeps. Dr. Ezeldel, are you familiar with the proposals of
Prof. O. M. W. Sprague of Harvard University in this regard?
Dr. EzEKiEL. I have a general acquaintance with them. I am not
sure that I can expound them in full.
Dr. Kreps. As you imderstand his proposals, does he recommend
that in the building field there be some such, as you have said, con-
certed action or simultaneous action, in order to reduce costs of
housing?
Dr. EzEKiEL. Yes; I believe that he pointed out that if you really
wanted to get housing costs down, you had to get reductions aU along
the line, and that if you get reductions by all the persons who partici-
pated in the housing, you could expect to get such an increase in
houses sold as to bring about a benefit for all involved.
Dr. Keeps. And as I understand your point it is the pertinent
consideration, if we want to increase the volume of housing. If we
took each particular commodity, we could show that in each particular
case that a reduction in price would have a negligible effect on the
total volume of housing; isn't that correct?
Dr. Ezekiel. Yes; I believe that is correct for housing.
Dr. Keeps. We could prove it for each individual, particular case,
and yet when we added it up, our results, instead of being a sum total
of the individual effects, would be according to Dr. Sprague and accord-
ing to your analysis, diametrically opposite. The social effect, in
other words, is. entirely different from the sum total of individual
effects. The results of particular studies, of each of the materials that
13694 CONCENTRATION OF ECONOMIC POWER
goes into a house and each of the skilled groups of ^ftbor, do not give
us a guide for public policy with respect to housing; is that correct?
Dr. EzEKiEL. Well
Dr. Kreps (interposing). It is the total picture that you regard as
important?
Dr. EzEKiEL. Yes; studying any one pait of it does not necessarily
tell what the possibilities are if the problem is dealt with as a whole.
Acting Chairman O'Connell. Are there any other questions?
Have you any other questions you would like to ask, Dr. KJreps?
Dr. Keeps. No, you may dismiss the witness.
Acting Chau-man O'Connell. Thank you very much, Doctor. I
think that is all.
Dr. Kreps. I would now like to sunamoit Mr. Martin Taitel to the
stand.
Acting Chairman O'Connell. Have you been sworn yet?
Mr. Taitel. No; I have not.
Acting Chairman O'Connell. Do you solemnly swear that the
testimony which you are about to give in this proceeding shall be
the truth, the whole truth and nothing but the truth, so help you God?
Mr. Taitel. I do.
TESTIMONY OF MARTIN TAITEL, SENIOR CONSULTING ECONO-
MIST, WORK PROJECTS ADMINISTRATION, WASHINGTON, D. C.
Dr. Kreps. For the purpose of the record, Mr. Taitel, will you
state your full name and address?
Mr. Taitel. Martin Taitel.
Dr. Kreps. And your address?
Mr. 'Taitel. Chevy Chase, Md.
Dr. Kreps. You have been formerly with the N. R. A. as economic
adviser on codes and statistician?
Mr. Taitel. Yes.
Dr. Kreps. Where did you receive the bulk of your training in
statistics?
Mr. Taitel. Under Professor Yntema at the University of Chicago.
I should like to say that I am very proud to have received my training
from Professor Yntema; I hope he will be as proud of the product of
his training.
Dr. Kreps. You have prepared a statement for us. Is it the one
to which we heard something of a rebuttal this morning?
Mr. Taitel. I should say a rebuttal in part.
Dr. Kreps. You may proceed.
Mr» Taitel. The steel industry has rather generally been regarded
as an industry with high "fixed" costs, that is, one o£ these industries
in which unit costs of production decline as output increases.
So far as prices in such an industry are based upon costs, the
pricing policy would tend to be one that provides for dechning prices
as the volume of output increases. Prices in the steel industry,
however, have not followed this pattern. They have tended to
remain relatively fixed. The typical practice has been to increase
prices with increased volume rather than to decrease prices as sales
expand. Such price behavior is much more consistent with a situation
m which increasmg output is associated with constant or rising costs.
CONCENTRATION OF ECONOMIC POWER 13695
UNITED STATES STEEL CORPORATION ANALYSIS OF COSTS IN RELATION
TO PRICE DECISION-MAKING
Mr. Taitel. The statistical analysis of costs presented to this
committee by the United States Steel Corporation is designed to
dciend the pricing system practiced by the Steel Corporation. It is
designed to show that the price policy actually pursued by the corpora-
tion has been in considerable^ measure dictated by its costs. The
illusion is created in the analysis prepared by the Corporation that the
great bulk of the Corporation's costs vary directly with the number
of tons of steel it produces. This illusion has been attained by dis-
solving the distinction between fixed and variable costs. All costs,
except bond interest and pensions, are treated as if they were made
up of fixed and variable elements which can be segregated by means
of highly refined statistical techniques, but cannot be segregated on
tRe basis of direct observation. The net result of such manipulation
is the one obtained by the Corporation, namely, costs are in the
main shown to be "variable'-' costs. But, unless one accepts the highly
sophistical theories upon which the numerical calculations are based,
one cannot accept the Corporation's analysis as a true reflection of the
cost situation in the steel industry as a basis for price policy.
Any cost accountant or statistician working with cost figures can
attain a variety of cost-volume relations by varying the methods of
computing or stating costs. This is strikingly illustrated by the
sharply contrasting results obtained by the "Iron Age" and the United
States Steel Corporation. Mr. T. W. Lippert, metallurgical editor
of "Iron Age", presents, in this year's January 4 issue of that journal,
a production-profit curve based upon his examination of "production-
profit data of two large steel companies — both integrated producers
and both makers of practically all types of steel, from fine wire to
structural shapes and including low alloy steels." The results pre-
sented differ from the comparable results of the Corporation's analysis.
I should like, Mr. Chairman, at this point, to offer for the record
a chart entitled, "Contrast in Production-Profit Computations."
Dr. Kreps. Will you display the chart, please?
Acting Chairman O'Connell. It may be admitted.
(The chart referred to was marked "Exhibit No. 2184" and is
included in the appendix on p. 14120.)
Mr. Taitel. The Corporation's analysis purports to show that
profits vary directly with output; namely, the addition to profit is
the same for each additional ton of steel sold.
Mr. Lippert's analysis, on the pther hand, purports to show that
the relation between profits and production is decidedly not of this
direct character but that changes in output produce profits of varying
magnitudes, depending upon the rate of capacity at which plants are
operated. According to his analysis, the rate of profit per additional
ton of output increases rapidly between the break-even point —
roughly 45 percent of capacity — and about 80 percent of capacity.
Above the 80-percent level there is only a very small profit per
additional ton of output until a rate of capacity somewhere around
the 90-percent level is reached, after which a loss is associated with
each additional ton to full capacity. Furthermore, according to Mr.
Lippert's study, below the break-even point losses increase very
13696 CONCENTRATION OF ECONOMIC POWER
slowly as operations are reduced to about 20 percent of capacity and
tbea increase sharply as the operating rate approaches zero.
Dr. Kreps. Mr. Taitel, will you turn to the chart and indicate the
significance of what you haVe just testified?
Mr. Taitel. This is Mr. Lippert's curve and he shows that from
the break-even point
Dr. Kreps (interposing). Just a second. That is Mr. Lippert's
curve reproduced from the January 4 issue of "Iron Age" of this
Mr. Taitel. Yes.
Dr. Kreps. Do you know who Mr. Lippert is?
Mr. Taitel. He is the metallurgical editor of "Iron Age."
Profits from the break-even point — these are total profits — increase
much more sharply than output. The increase in slope of this curve
as it goes above the break-even point indicates that the additional
profit per ton increases with an increase in the percent of capacity
operatmg..
Dr. Kreps. How do6s that contrast with the lower chart, which is
based, I take it, upon the materials that have been adduced by the
Steel Corporation?
Mr. Taitel. The lower chart represents a situation in which the
profit per additional ton is constant. , As I have plotted it here,
I believe for each additional ton, the addition to profit is about $18.
Under Mr. Lippert's computations, losses do not increase very rap-
idly as operations decline below the break-even point.
Dr. Kreps. Of what significanae is that fact?
Mr. Taitel. According to Mr. Lippert's computations, steel plants
are able to operate between 20 and 45 percent of capacity without
much change in the total loss; within that range of output there is
very little change in the loss, assuming, of course, no change in prices.
Dr. Kreps. You mean if the top curve is true to fact, I take* it.
Is that right?
Mr. Taitel. Yes. I should perhaps note that in both cases the
price structure is assumed to be the same, so that receipts would in-
crease directly with output.
Dr. Kreps. Supposing you had a 10-percent reduction in price,
does it make any difference whether the condition of the industry is
that as represented in the top chart as compared with that in the
bottom chart?
Mr. Taitel. It makes a good deal of difference. Speaking in ap-
proximate general terms, a 10-percent reduction in price would bear
less heavily on profits assuming Mr. Lippert's curve is correct when
operations are above the break-even point but below about 80 percent
of capacity. The reverse is true below the break-even point but above
about 20 percent of capacity.^
Dr. Kreps. If you were to take those rather startling loss figures
of Dr. Yntema — those* estimated loss figures guessed at but precisely
stated in dollar figures, thus giving them a specious and spurious
plausibility and reminding one of the cynic's definition of statistics
as the science that states an uncertainty with precision — how would
' In precise terms, the adverse effect upon profits of a price reduction when output is at a given level is
greater for the production-profit curve with tne lesser slope between that output level and the new level of
©utput. Since, however, the scales for the two curves are not the same— Mr. Lippert providing no indica-
tions of absolute amounts of profits oriosses— it is not possible to stat« exactly the ranges of output for which
lb* Adverse effect is greater or less for one curve than for the other.
CONCENTRATION OF ECONOMIC POWER 13697
those estimated losses be changed if the cost curve in the industry
corresponds to the pattern of profits shown in the upper part of your
chart ("Exhibit No. 2184") as contrasted with that shown in the lower
portion of your chart?
Mr. Taitel. I am not sure that I understand the question.
Dr. Keeps. I will repeat the question in a different form. Would
the estimated loss from a reduction in the price of steel be less or
greater if the condition that Mr. Lippert describes is true, than it was
represented to be on the charts that Mr. Yn tenia showed?
Mr. Taitel. The adverse effect upon profits resulting from reduc-
tions in price would be less in the case of Mr. Lippert's curve when
operations are approximately between the break-even point and 80
percent.
Mr. Walter White. Do you know what statistics and figures Mr.
Lippert had available from which he derived his curve?
Mr. Taitel. He says he used production and profit data for two
large steel companies. I have not seen the detailed figures. I have
simply reproduced the chart as it appears in "Iron Age."
Mr. Walter White. Does it show that a decline in profits occurs
after a certain volume has been passed?
Mr. Taitel. That is correct. Beyond about 90 percent, total
profits decline, that is, for each additional ton to expand from about
90 to 100 percent there is a loss.
Mr. Walter White. Is that because it is in an ineflScient plant,
do you know?
Mr. Taitel. He states that it is due to general inefficiencies — trying
to meet particular orders of particular customers at particular times,
bringing in of obsolete capacity, and so on.
Mr. Wooden. Is it to be understood that the United States Steel
Corporation is not one of the two that are included in Mr. Lippert's
study?
Mr. Taitel. There is no statement to that effect in Mr. Lippert's
article. He does not state the names of the two companies. I think,
though, that this chart expresses more eloquently than any words at
my command the different results which can be obtained from studies
of cost records in the steel industry.
My analysis of the Corporation's cost analysis is directed toward
this point: The arbitrary nature of the allocation of costs as between
years or over portions of the output makes it impossible for particular
cost-volume computations such as the Corporation has presented to be
the all-important basis of decisions as to prices. The conclusion is not
that the Corporation does not have to reckon with its money costs.
Rather, the conglusion is that the kind of cost-volume relation which
the Corporation derives is not the one relevant to price decision-making
under actual operating circumstances.
To establish my main conclusion, it is necessary to establish two
others. First, that the Corporation in particular, and almost anj
business firm in general, has a choice as to when and in what amount it
will charge a considerable number of items of expenditure to costs.
And, second, that the cost-volume relation obtained by the Corpora-
tion's analysts is in large part the consequence of the particular times
at which it has chosen to charge certain expenditures to costs, and of
the particular accounting procedures by which it has chosen to be
governed in aUocaling costs.
13698 CONCENTRATION OF ECONOMIC POWER
I want to make very clear the fact that I am questioning neither the
validity nor the usefulness of the accounting procedures or records of
the Corporation, nor the statistical methods or procedures used by the
Corporation's analysts when 'they are directed toward purposes other
than the one now under discussion. In fact, I, as one who pretends to
be qualified, want to pay tribute publicly to -the skillful and ingenious
use which has been made of highly refined and advanced statistical
techniques. I must, however, point out that even the most ingenious
and advanced methods of accounting and statistical analysis may not
provide the correct answer to the particular issue to which they are
addressed.
The essential requirement of a costing system is that it shall be use-
ful in terms of particular purposes. Thus, in a book written by three
eminent authorities — Professors Sanders, Hatfield and Moore —
published by the American Institute of Accountants in 1938, and
sponsored by the Haskins and Sells Foundation, there appears the
following statement:
Since the income statement is prepared for the information of owners, managers,
creditors, and taxing authorities, and for regulatory and other purposes, those
accounting practices are best which serve these purposes in the most reliable and
helpful manner.
It sometimes becomes necessary to prepare separate statements to serve the
several purposes.
An eminent economist — Prof. J. M. Clark — in his classic study,
"The Economics of Overhead Costs", published in 1923, points out —
* * * the cost-accounting conceptions of cost do not agree with cost as used
by the general accountant, and they disagree because they are wanted for different
purposes.
Typi(Sally — and I believe this is true of the Corporation — the costing
system is designed to be useful for operating, tax or public-statement
purposes. That such a system should, without any recasting- of
accounts, provide data showing the "actual" or "true" division be-
tween fixed and variable costs or the "actual" or "true" shape of the
marginal cost curve is not to be expected. Consequently, since
costs as entered on the books are for general purposes, a cost analysis
based upon book costing contains no inherent validity.
Allocation of costs as between years or as between segments of out-
put is and must be in part arbitrary no matter what accounting prin-
ciples and practices are followed. Many items of cost have no observ-
able economic or physical connection to the output with which they
are associated. While practical considerations require their alloca-
tion, the guides themselves are not sufficient. Within wide hmits
set by custom, allocations are molded to show particular results for
the particular purposes of the allocater.
In making the allocations the .overpowering tendency- is to use
accounting procedures which wiU place costs on the books when there
is output and receipts' against which to charge them. A variety of
reasons on the part of management may explain this — reluctance to
a"dopt accounting methods which might show large losses in poor
year?, efforts tp minimize tax liabilities, desire to allocate building and
equipment expenses as equslly as possible over aU units of output,
and so forth. Taken together, these underlying motives operate (a)
to minimize the fixed costs, (6) to raise the variable costs, and, (c)
CONCENTRATION OF ECONOMIC POWER 13699
to show constant marginal costs, when comparisons between book
cost and output are made. Thus, the true picture is distorted.
There is another tendency underlying ordinary costing which leads
in the same direction. Some choice as to timing exists for certain
types of expenditures. The tendency is to vary such expenditures with
the volume of receipts. At the same time, there is the tendency to
charge them to current operations, to consider them sunk costs, the
sooner off the books the better.
The effect of the particular accounting procedures used upon the
results obtained from a study of over-all cost-volume such as the
Corporation's is so great that it cannot be neglected. In fact, it may
be said that the accounting procedures themselves are major determi-
nants of the statistico-arithmetic results. Particular consideration
must be given to the allocations of charges as between years since it
is the shape of the cost curve which expresses whether unit costs
increase, decrease, or remain constant with increases in output. While
(a) the items included in cost and (6) the total amounts of those items
charged to costs over the life of the business are also factors, yet they
do not loom large in comparison with the allocation of the items
included as between years, particularly in an analysis such as the
Corporation's in which practically all expenditures are included. I
shall not discuss these two elements; partly because they do not
appear impc rtant, but also because the published sources provide
no adequate material for determining their effect upon the cost-
volume analysis.
The chief items subject to allocation over accounting periods are
depreciation, depletion, amortization, maintenance, repairs, in-
tangibles such as patents, and similar items. Clearly when such
expenditures are actually charged to costs is just as important as how
much is charged to costs. That the Corporation has in the past
made serious errors in the timing of the charges is indicated by the
establishment in 1935 of a "Reserve for amortization of investments
in subsidiaries" of 181 milhon dollars. This reserve was established,
presumably, because of the undercharging of depreciation in the
period prior to 1935. In 1935 and later years the reserve is drawn
upon to increase current profit figures— about $7,000,000 in 1936 and
$8,000,000 in 1937. Thus, when "costs" are shifted as between years
the results of a cost-volume analysis are different from those which
might have been obtained had such shifting not been indulged in.
The periods to which expenditures are charged as costs are some-
times the result of advanced planning, such planning being based upon
the estimated Hfe of assets, the estimated output or both. Errors in
such estimates are corrected by adjustment on the books as they are
recognized. Final adjustment always occurs at the time of disposal
of the assets. For it is only upon final liquidation that actual costs
are known. Business, however, must make interim estimates; hence
the errors.
The method of handling such adjustments is, however, extremely
important if the figures are to be used for an analysis of costs. Typi-
cally, a revision of cost figures for prior years is not made by revising
the figures for the earlier years but by adjusting the figures for the
current year. This may not be a serious matter if the adjustment is
not included as an item of current costs but as an adjustment to
surplus. But, if the adjustment is charged to current costs (or spread
13700 CONCENTRATION OF ECONOMIC POWER
over current and future costs), any cost- volume analysis based on
them will be seriously warped.
The Corporation's method of handling one type of adjustment is
indicated in the annual report for 1928, which states that —
The large increase in the provisional allowances by subsidiary companies in
1928, compared with 1927, is attributable to a considerable extent to the rather
substantial amounts charged off for obsolescence of property investment cost in
connection with abandonment of old plants not theretofore fully depreciated.
Thus, what might be construed as an adjustment for inadequate
depreciation prior to 1928 became an operating charge in 1928.
An interesting discussion of the extent to which the United States
Steel Corporation erred in computing profits and, therefore, costs during
the twenties is contained in a paper by W. A. Hosmer, in "Business and
Modern Society", published by the Harvard University Press in 1938.
I refer those who are interested in further study of the matter to
Professor Hosmer's very excellent paper.
In the ultimate analysis, because of the discretionary elements of
all cost allocation, everything that may be shown by the cost-volume
relation is explicitly or implicitly assumed by the accounting proced-
ures as appropriate. It is impossible to demonstrate that any par-
ticular allocation is most vali'^ except for a particular purpose. All
that we can show is that a particular method of allocation gives a
particular :Lelation between charges to costs and volume. This is
what the Corporation's analysis shows at most, for example, for
depreciation and depletion.
The sources upon which my analysis is based are (1) the Corpora-
tion's
annual reports, and (3) the Corporation's registration statement
filed with the Securities and Exchange Commission. I am convinced,
however, that the essential nature of my conclusions would not be
changed had I had acces's to the sources at the disposal of the Corpora-
tion's analysts.
In order to bring out most clearly the full effect of the arbitrary
nature of cost allocation and classification, I have cast my analysis
within the same general statistical framework as that used by the
Corporation. In so doing, I do not imply that that framework is
above criticism. In this connection, the following observations are,
I think, pertinent.
1, I have used the Corporation's own measurements of output,
i. e., the Corporation's own figures on weighted tons of products
shipped, though I do not admit the validity of the methods used to
devise them. The principal defects of these- measurements for pur-
poses of cost analysis revolve around the weights used to convert
quantities of a wide variety of different products into homogeneous
units of output and the lack of consideration given to changes in
capacity during the period covered.
2. Though I do not admit the validity of the particular applica-
tion, I have used the least squares or correlation technique, though
perhaps not with the same degree of excellence as the Corporation's
analysts, for deriving the sum.mary relation between a cost category
and volume, and have labeled the statistical results (technically esti-
' AppendU.p. 14032.
tion's study entitled, "An Analysis of Steel Prices, Volume and Cc
I believe identified as "Exhibit No. 1416," ' (2) the Corpora ti
CONCENTRATION OF ECONOMIC POWER 13701
mates of parameters) as "fixed" and "variable" costs in the same
manner as has the Corporation. The principal defect of the technique
is that it takes no account of the interdependence of the cost measure-
ments for the various years. This defect is extremely critical. Be-
cause it exists in the technique, no account is taken of such facts as
this: If depreciation is charged to costs in one year, it cannot be
charged in another year so that relatively high charges in one year
tend to involve relatively low charges in another year.
3. I have not attempted to measure the effects of such inadequacies
as may exist in the Corporation's adjustments of pay-roll and "other
expenses" to "1938 conditions." The pay-roll adjustment as made
takes no direct account of the possible effects upon average hourly
wage rates of differences in the proportions of employees in the vari-
ous occupations at different outputs. Furthermore, the facts upon
which the adjustment for increasing labor efficiency was made seem
more appropriately to indicate (a) declining unit labor cost with the
expansion m output during 1927-29 and 1934-37, and (6) inadequate
allowance for changing compositions of the working force during
1930-33. With regard to "other expenses" it seems pertinent at
least to raise the question as to whether a somewhat modified general
index of prices is appropriate for deflating the amounts paid by the
Corporation for what must be a rather specific composition of goods
and services.
That the purpose for which accounting statements are made de-
termines, in part, the way in which expenditures and charges are
classified can be illustrated by contrasting the segregation of accounts
for public statement purposes with the segregation of accounts for
the cost analysis.
I have six tables. Shall I insert them one by one?
Dr. Kreps. Would you prefer to insert them as a group at this
time?
Mr. Taitel. I believe that would be most convenient.
Dr. Kreps. Mr. Chairman, I should like to insert into the record a
series of six tables titled as follows: Table I, "Reconcihation of Total
Costs Before Bond Interest and Inter-Company Items in .'Analysis'
and Registration Statement, 1935-37""; Table II, "Comparison of
Break-Down of Lumped Costs in the 'Analysis' and in Registration
Statement, 1935-37"; Table II-A, "Additions to Reserves Charged to
Cost of Goods Sold, Etc., 1935-37"; Table III, ' Taxes Other Than
Federal Income and Social Security Taxes, 1927-38"; Table IV,
"Taxes Other Than Federal Income and Social Security Taxes, 1927-
38 — Recom.puted 'Fixed' and 'Variable' Costs"; Table V, "Mainte-
nance and Repairs, 1927-38"; Table VI, Stripping and Development
Experises, 1927-38".
Acting Chairman O'Connell. Who prepared these charts?
Mr. Taitel. I prepared these tables.
Acting Chairman O'Connell. And the source of the material?
Mr. Taitel. The sources are indicated on the table. They have
been taken from the three general sources I indicated, the Cost
Volume Analysis, the annual reports, and the registration statement.
Acting Chairman O'Connell. They will be admitted.
(The tables referred to were marked "Exliibit No. 2185" and are
included in the appendix on p. 14121.)
13702 OONOENTRATION OF ECONOMIC POWER
Mr. Taitel. a reconciliation of the figures in the Corporation's
"Analysis" with those in the S. E. C. registration statement for the
three years 1935-37 is shown in Table I of the group of tables iden-
tified as "Exliibib No. 2185." To obtain the same total costs, (1)
"Expenses for dismantling, moving, and rearranging of existing facil-
ities, less the value of salvage recovered in connection therewith" have
to be omitted although classified as operating expenses in the registra-
tion statement; (2) "Plant and organization survey expenses" have
to be included although classified as an income deduction in the regis-
tration statement; and (3) "Discoimt on purchases" has to be in-
cluded although classified as other income in the registration statement
and annual reports.
In the annual reports and registration statement there is a functional
classification of accoimts. Cost of goodi-sold, and so forth (including
intercompany items), apparently includes all items which are con-
strued to be allocable to specific items of output. Thus, certain
amortization, rents and royalties, and maintenance and repairs, are
charged directly to cost of goods sold. At times, also, certain taxes
(in minor amounts) have been so charged. Included also, are gross
operating expenditures for transportation and miscellaneous operations
(both shown separately at times).
Another general functional category is "Other operating expenses"
which apparently covers items which are not deemed to be specifically
alloc9.ble to items of output. Presumably, only steel operations are
covered. The major items are (1) general administrative and selling
expense?, (2) depreciation and depletion, and (3) taxes.
Finally, there is a third functional classification — "other income"
and "income deductions." This includes items apparently considered
to be nonoperating in character, such as dividends, rents and royalties,
capital losses, and so forth.
In the Corporation's cost analysis the classification of accounts
used for public statement purposes is retained only in part. The
bulk of the costs are redistributed into two classifications: (1) Pay
roU, and (2) other expenses. It was not possible for me to recast
them along functional lines since the Corporation's public statements
do not contain the necessary data. It was possible, however, to indi-
cate the character of some of the items included in the two categories.
This is shown in Tables II and II-A^ where the two types of break-
downs are compared.
The two bases of classification are not contradictory; they are
just different. And the reason they are different lies largely in the
different purposes for which the Corporation has prepared them.
That the Corporation has made an inadequate division between
fixed and variable costs, even in teims of its own analysis, is clearly
illustrated by the treatment accorded in the cost anatysis to taxes
other than Federal income and social security taxes. This tax item
includes mainly State and local property taxes but also the Federal
capital stock and excise and miscellaneous taxes. The break-do\\Ti
is shown in Table III of "Exhibit No. 2185." For some years the tax
figures apparently represent the accrual of tax liabilities and the
difference column indicates the extent of allocation as between years.
Capital stock taxes should not have been lumped with the other
taxes. The Corporation was not subject to such taxes prior to 1982..
• Of "Exhibit No. 2186," appendix, pp. 14121 and 14122.
CONCENTRATION OF ECONOMIC POWER 13703
SO that they should at least have been segregated. But most impor-
tant is the fact that capital stock taxes under "1938 conditions"
depend, not upon output, but upon decisions by management based
upon «5«timates not only.of future costs but also of future output and
prices. Stated otherwise, declared values for tax purposes are deter-
mined within limits by management forecasts of net incomes, i. e.,
the estimated relation between costs and receipts; they are not
determinants of net incomes in the sense in which a property tax is.
Thus, reduction to "1938 conditions" for purposes of a cost analysis
imphes that the 1938 figure should best be used for all years, i. e.,
that the capital stock tax is probably best considered as a fixed cost.
The contention might be made (and to some extent is implicit in the
Corporation's analysis) that declared values would s^ary with output.
But this assumes that prices will vary with output in such a way
as to make it profitable for the Corporation to vary the declared values
with output. Realistically, such an assumption, not to mention the
assumption of accurate forecasting, has no place in a cost analysis,
particularly one for the Corporation in view of the fact that in 1938
a greater tax was allocated to costs than in 1937.
Appropriate treatment of the capital stock tax would destroy what
Uttle reliability there is in the analysis of taxes on pages 13-14 of the
Corporation's analysis.
Dr. Keeps. Whaf exhibit?
Mr. Taitel. "Exhibit No. 1416." They account for a good share
of the rise in taxes between 1932 and 1938 as the figures in Table IV
of "Exhibit No. 2185"- show. Exclusive of capital stock* taxes there
is no significant difference between the relation of taxes and weighted
tons for 1932-38 and the relation for 1927-31. The $43,200,000 item
for 1937 is the most extreme observation, being about 15 percent
above the next largest one. This suggest? not a shift in the tax
burden between the two periods, but rather some peculiarity in the
1937 tax charges. _
The effect of using tax figures appropriate for general-statement pur-
poses, but not appropriate for cost-volume relations, upon the results
of the Corporation's analysis is substantial. Results of a recomputa-
tion of the tax regression both including and excluding 1937 data are
shown in Table IV of "Exhibit No. 2185." Capital-stock taxes have
been considered as a fixed cost at the 1938 level.
Both of the recomputations show a much higher "fixed" and a much
lower "variable" tax cost than is shown by the Corporation's analysis.
Even with the 1937 observation included, "fixed" costs are raised by
almost 25 percent and "variable" costs lowered by over 50 percent
with reference to the results of the Corporation's analysis. The recom-
putation excluding 1937 shows "fixed" costs to be raised almost 30
percent and "variable" costs lowered almost 65 percent.
A study of the maintenance and repair expenditures of the Corpo-
ration shown in Table V of "Exhibit No. 2185" illustrates the tendency
of the Corporation to charge some expenditures to costs when made.
It also provides another illustration of the effects of an inadequate
segregation of costs upon the results of a statistical cost analysis. In
this latter case it is the Corporation's treatment of maintenance on
railroad properties which may be suitable for some purposes of the
Corporation but which is definitely misleading for purposes of coat
analysis.
13704 OONOENTRATION OF ECJONOMIC POWER
The bulk of the maintenance and repair expenditures (all in the case
of railroad properties) are charged to costs as made, although small
portions are passed through reserves each year. By and large, how-
ever, maintenance is charged as the work is done — not when the par-
ticular outputs making maintenance necessary occurred. Further-
more, no segregation appears in the published sources (except to a
very limited extent in the registration statement) between maintenance
required regardless of output and the additional amounts of mainte-
nance required for each level of output. And in the cost analysis,
maintenance is presumably buried in pay-roll and other expenses —
although the figures are available — even though the amounts charged
to costs in some years have been almost twice as great as depreciation
and depletion.
Included in the maintenance and repair accounts of the Corporation
are the expenditures on lis railroad properties which are always
charged to costs when made. But all of the operating and maiate-
nance expenses of its railroads should not be included in those costs
which are presumably comparable with steel shipments. (For a
wider range of factors this point is discussed and carefully minimized
in the Corporation's analysis, pp. 39-42, "Exhibit No. 1416.") Part
of the other transportation and miscellaneous operajtions should
also be excluded. But it was not possible to do'so since the necessary
accounts are not shown separately in the annual reports. However,
other than railroad maintenance charges are relatively minor items —
they only accounted for ibout 3 percent of the total maintenance
expenditures in 1929, wh reas railroad transportation accounted for
about 20 percent.
The allocation of practically all maintenance and repair expendi-
tures to costs in the year in which they occur is improper. There
is, of course, a considerable amount of leeway as to when such expendi-
tures are made. That the Corporation's practice reflects the element
of flexibility is indicated in the 1932 annual report which says that
"maintenance expenditures * * * include a substantial amount
expended m order to keep inactive departments prepared for resurnp-
tion of operations when business improves." Thus, the Corporation
charged to 1932 costs, maintenance expenditures which it admits
were necessary either on account of past or future operations but
not to current operations. And it should be noted that only a very
small part ($469,000 out of $28,000,000) of the current expenditures
were not charged to current costs, while $1,300,000 was charged to
current costs to build up the reserve account.
Application of the statistical technic used in the Corporation's
cost analysis to the maintenance and repair data gives the following
results:
"FUed"
costs per
year
"Variable-
costs
per ton
Coefficient
of cor-
relation
Total (1927-38)
$5,320,000
1,600,000
3,720,000
$8,356
6.320
2.036
0.08
Excluding railroads (1927-36)
.97
Railroads (by subtraction) ' —
.
GONOENTRATION OF ECONOMIC POWER 13705
No adjustments have been made for wage and price changes similar
to those in the analysis. Such adjustments would tend somewhat
to raise the "fixed" and lower the "variable" costs.
If my analysis of maintenance and repair expenditures presented
thus far is correct the Corporation's analysis is biased to the following
extent:
1. "Variable" costs are overstated between $1 and $2 per ton
because of the inclusion of railroad maintenance exp^iditures in
excess of those "attributable" to shipments.
2. "Fixed" costs are understated by the inclusion of the total
railroad maintenance expenditures.
The results indicate that the Corporation has practically adopted
the "cost when spent" principle for maintenance. They also indicate
that the Corporation has come close in its maintenance accounting
to the principle of spreading such costs equally over all units of output.
Had "equal spreading" been fiUly accomplished, the "variable" cost
computations for mamtenance excluding railroads would have been
lower by about 10 cents per ton than the $6.32 figure obtained. The
two figures are so close as to warrant the suspicion that they are the
results of a conscious design.
The Corporation's policies with regard to depreciation and depletion,
as stated in its S. E. C. registration statement, are —
1. Depreciation is charged on the straight-line method. When
the actual operating rate is less than the predetermined average
rate charges are reduced but less than proportionately ahd in no case
by more than 50 percent. (Railroad equipment is, of 'course, depre-
ciated at rates approved by the Interstate Commerce Commission.)
2. Depletion is charged by prorating the investment costs over the
estimated recoverable quantitj^.
The Corporation's policy with regard to depreciation is only one
of many that might have been followed, and is apparently a com-
promise between charging equal amounts per unit and equal amounts
per annum. On an equal charge basis, the statistical computations
would show an annual "fixed" cost of about 53 million dollars but
no "variable" cost; on an equal per unit charge basis, the computa-
tions would show no "fixed" costs but a "variable" cost of about
$5.30 per weighted ton. It is easily seen that the actual results of
29.5 milUon dollars for "fLxed" and 2.37 for "variable" costs obtained
by the Corporation are about halfway between these two extremes.
Our analysis assumes, of course, that the depreciation and deple-
tion figures which the Corporation uses in its cost analysis are the
appropriate^ ones in the sense that they are computed in accordance
with a general pohcy that does not' change from year to year. But
the records at my disposal indicate that this may not be so.
(Senator King assumed the Chair.)
Acting Chairman King. Depreciation and depletion, especially in
mines, that is largely fixed by statute.
Mr. Taitel. It is fixed by statute in terms of statements for tax
purposes. It is not fixed by statute in terms of what corporations
may charge on their own books for their own purposes.
Both the 1927 and 1928 depreciation figures in the Corporation's
cost analysis include about 11.5 million dollars, of charges to a bond
13706 OONOENTRATION OF ECONOMIC POWER
sinking fund reserve used "to cover amortization of appreciated cost
to it (the holding company) of investment in stocks of subsidiary-
companies in excess of their own investment in tangible property."
Such charges were made in years prior to 1929 but not in later years.
It is not clear whether these charges simply make up for under-depre-
ciation on the books of the subsidiaries, i. e., represent the basis
for transfers in excess of earnings from the subsidiaries to the holding
company in order to fulfill the conditions of the bond indenture, or
whether they are bona fide amortizations of intangibles by the hold-
ing company. Just how these amounts should have been handled in
the cost analysis depends, of course, upon the continuity of depre-
ciation policy during this period.
Also included in depreciation were charges of about $1,000,000 in
1932, $400,000 in 1933, $450,000 in 1934, a^d $400,000 in 1935
"normally included in the value of the season's production of ore
carried in 'inventories'." These charges belong, of course, in other
accounts. However, their inclusion in the depreciation account has
only a minor effect on the statistical results.
Conversely, about 7 million dollars in 1936, 8 million in 1937, and
probably a somewhat smaller amount in 1938 were excluded from the
depreciation item included in the cost analysis. These amounts rep-
resent charges against a special reserve set up 'in 1935 on the books
of the holding company presumably to make up under-depreciation
in prior years. The extent to which changes in depreciation rates
occurred is not clear.
The depreciation and depletion figures used in the cost analysis do
not include a comparable item for amortizing "investment in strip-
ping and development of mines and structural erection equipment."
Acting Chairman King. Do not take into account the credit to
which they would be entitled in their balance sheet?
Mr. Taitel. What I am saying is that the depreciation and deple-
tion figures used in the Corporation's analysis as depreciation and
depletion figures do not include the item on the books of the Corpora'-
tion for investment in stripping and development of mines and in
structural erection equipment. Those expenses are not included in
the Corporation's analysis under the item of depreciation and deple-
tion.
Acting Chairman King. Would that be to their advantage m
obtaining offsets?
Mr. Taitel. Well, they are charged or entered as costs on their
books, but they are in separate accounts. The cost analysis, having
taken the depreciation figures, as I understand it, from the annual
reports, does not include the stripping and development ejqpenses,
and the structural erection equipment expenditures. The item is
charged directly to property accounts and not through reserves.
The Corporation's policy with regard to this item is, according to
the registration statement, to charge the strij-ping and development
part to cost in the same manner as depletion, and the structural erec-
tion equipment part in the same manner as depreciation. Another
indication of the depreciation and depletion character of the item are
the wide differences from year to year between expenditures and
charges to costs shown in table VI of "Exhibit No. 2185."
OONOENTRATION OP ECONOMIC POWER 13707
No doubt the Corporation has reasons which are valid in terms of
operating and related purposes for not handling the stripping and
development expenses as depreciation and depletion.
Acting Chairman King. VMiere would you place upon the books
that you were keeping the cost for stripping and for depreciation?
^Vhat column would you put those costs in?
Mr. Taitel. The iuetlvod of accounting which the Corporation uses
in handhng its stripping and development expenses is a perfectly good
one. The only difference between that method and the method used
for the regular depreciation and depletion accoimt is this: When an
expenditure is made for stripping and developing, it is entered directly
as an investment; when a charge is made to costs, the amount is
entered in the same investment account, instead of in a separate
depreciation or depletion account.
I do not intend to imply that the Corporation is using bad accounting
practices. It is a perfectly legitmiate procedure and the procedure
is stated very clearly in the annual reports.
Mr. Walter White, You mean by that that they charge stripping
expense, for instance, to ore that is being mined from somewhere else,
in the current year, instead of to the ore which will be mined under-
neath that stripping in a subsequent year?
Acting Chairman King. Perhaps the question is not clear. Will
you repeat it?
Mr. Walter White. ^Vhether the stripping is charged, or the
stripping over a certain ore body is charged to operating expense, in
connection with oie receipts from somewhere else — that is, that year's
operation is not held in a suspense account to be charged against it,
or wliich underlies the particular body o^ ore that is being stripped? •
Mr. Taitel. As I understand the accounting procedure, as the
annual report states it,' when they make an expenditure for stripping,
they charge it to the property account, that is, it is an addition to the
property asset, it is an investment. In essence, it is -a suspense
account, as you put it. Later, when the ore is mined, it is charged
to the ore mined.
Mr, Walter White. I should think that would be the proper or
usual practice.
Mr. Taitel. I don't know whether it is the usual practice. I
simply state that that is what the Corporation does.
But whatever the reasons for so doing, those reasons cannot justify
the treatment of stripping and development expenses in the Corpora-
tion's cost analysis. Particularly should the item have been segregated
in the cost analysis since it is included in "Other Exisenses," a cost
category treated in the cost analysis as if it represented current pur-
chases of goods and services.
The "variable" cost for stripping, and so forth, is higher by about
14 cents per weighted ton than the over all average of 40 cents for the
12 years. This does not indicate a close correspondence with a poUcy
of allocation directed toward obtaining equal per unit costs, but rather
a tendency on the part of the Corporation to charge larger amoimts
T^er unit the larger the output and the income realized. In view of
such a tendency, it may be said that the particular treatment accorded
to stripping and related expenses biases the "variable" cost derived
in the Corporation's analysis slightly, upward and the "fixed" cost
slightly downward.
124491— 41— pt. 26 9
13708 OONOENTRATION OF ECONOMIC POWER
Criticisms of the nature presented do not, of course, negate the
hard facts of total costs. In the long run, cumulated total costs as
they appear in the income account as charges approximate actual
cumulative money costs. And over a long period of time, money
costs are determinative. They must be met by receipts if a profit
is to be made and if the Corporation is to, remain in business. But
this is true only in the so-called long run. In the short-run periods,
and it is in the short run that prices are made, total costs as figured in
the Corporation's analysis are not the facts upon which pricing is
based.
Nor is the implication that a cost-volume relation pertinent to
pricing would have been obtained had more appropriate treatment
been accorded to the various cost items mentioned. Rather, with
regard to the technical features of the criticism our examination of
the figures upon which the Corporation's analysis rests has shown that,
in terms of the Corporation's own framework of analysis — the
applicability of which to the problem I have grave doubts — the
Corporation's estimate of "fixed" cost is biased downward and the
estimate of "variable" cost is biased upward.
My conclusion with regard to the substantive features of my analysis
is that a division between fixed and variable costs obtained from a
statistical analysis of historical data, such as the Corporation has made,
bears but a nebulous relation to the actual division of fixed and variable
costs which bears upon a particular act of pricing. The question
may fairly be raised whether the Corporation has ever before had
prepared for the guidance of its executives cost analyses of the type
presented to this committee for the purpose of helping those executives
solve tjieir pricing problems. And I am led to believe that the Cor-
poration's cost analysis is not. a description of what in fact has guided
its pricing policy but is being used as a rationalization of the actual
pricing practices pursued by the Corporation in the past.
It is, of course, possible to modify the statistical analysis used by the
Corporation's analysts so that it would present cost schedules relevant
to the particular pricing problems the Corporation faces. Such
modification, however, would in my opinion stop little short of destroy-
ing the whole theoretical basis on which the Corporation's analysis
rests. But it would provide the best descriptive measurements
available from the bag of tricks of modem methodology.
The modification would be based upon an insistence that the division
between fixed and variable costs or a cost-volume relation meaningful
for an actual price decision is not a unique relation applicable to all
possible circumstances and therefore to none. For a general state-
ment, the most appropriate presentation is in terms of upper and lower
limits. As the conditions under which the cost figures are to be used
converge more and more to a concrete pricing situation, the limits
would be narrowed. For a particular pricing situation, the range
between the lower and upper cost estimates can probably be small
enough for practical purposes. But the estimates of this type would
bear no consistent relation to the Corporation's cost-volume curve.
In general, the lower limit for a cost estimate used for pricing in
connection with a particular prospective volume — the real hard cost
that must be covered if busniess is to be accepted— would be much
below the one indicated by the Corporation's analysis. It might be
OONOBNTRATION OF ECONOMIC POWER 13709
above. But where it would fall depends upon a wide range of practical
circumstances, such as the condition of the plant, the position of
material suppliers, the degree to which the working force needs to be
rearranged for production at the contemplated new level, and so on.
The upper limit, in general, would be above the cost-volume curve
of the Corporation's analysis. This must be so since the Corporation
in particular, and almost any business in general, is not immune from
the drive to cover sunk costs as quickly as possible. For a general
statement applicable to a whole range of ordinary practical circum-
stances, it might be taken as the minimum amount which a business
must cover in order to carry on over a very long period. This is
roughly equivalent or in rough conformance with accounting princi-
ples as usually stated and applied. But in a particular pricing situa-
tion the pertinent upper limit of costs bears no necessary or consistent
relation to the generalized upper limit.
In brief summary, the theoretical calculations submitted by the
United States [Steely Corporation analysts]! would regard as highly
interesting applications of refined econometrics, but of little use to the
committee as a description of the actual considerations upop which
steel price decisions are based.
Acting Chairman King. Any questions? Thank you very much,
Professor.
Call your next witness, Dr. Kreps.
Dr. Kreps. I would like to keep Mr. Taitel on the stand. While
we have greatly taxed the patience of Dr. Yntema, I am sure he
would like to Inake some comments on Mr. Taitel's analysis, reserving
such specific questions as he may have on Dr. Ezekiel's analysis,"which
Dr. Ezekiel ought himseK to answer, until Friday morning, when Dr.
Ezekiel can be back in town. Unfortunately, he had to leave at 4
o'clock this afternoon.'
Acting Chairman King. Would you prefer to resume the stand
now, Doctor, or perhaps. you would rather wait until the morning?
Dr. Yntema. I should much prefer to comment on this tomorrow
morning. I have had five documents to read and this is the first
time I have had an opportunity to hear this discussion. I could deal
with the general aspects of it now, and in a sense, I did so this morning.
But I think that my comments would be more valuable to the com-
mittee if I could make them to morrow morning.
Acting Chairman King. I thirk perhaps that would be better.
> Dr. Kreps. I should hke to point out to the chairman that the
hearings might thereby be considerably delayed. We have a witness
for tomorrow morning and tomorrow afternoon. All of us wou^d like
to' make these hearings as brief as possible.
Dr. Yntema has had this paper ' since yesterday. As he has him-
self acknowledged, he made some comment this morning. If possible,
and if it does not transgress on his patience and his energy too much,
he would really help us to keep the hearings brief if he could make his
comnients this evening. The hour is still early.
Acting Chairman King. I think if the witness has to go over his
paper and present it^ then if you ask the man offhand to reply to it,
I think you ought to give him a Uttle more time.
' Refers to a prepared statement from which Mr. Taitel read.
13710 CONCENTRATION OF ECONOMIC POWER
Dr. Yntema. Senator, this is merely the point
Acting Chairman King (interposing). And there would be no
delay as far as the committee is concerned. If it would not delay you ,
if Dr. Yntema can come back
Dr. Yntema (interposing). I can make comments on the general
question. I cannot obviously, on such short notice, comment on
highly technical points. I am prepared to comment both on Dr.
Ezekiel's paper and on the paper which Dr. Bean is scheduled to
present.* I am quite willing to proceed if the committee so desire.
Acting Chairman King. Well, if you are wilb'ng to proceed, if you
think that would be better, that you would make a better presentation
by waiting until tomorrow, as far as the chairman is concerned, we
won't force a ^vitness to testify when he isn't quite ready to testify.
Dr. Yntema. The general point to which I should address myself
would be this, that in evaluating criticism which has been offered this
afternoon — and I think in fairness, we should say this is a scholarly
statement that we have just heard — in evaluating the criticisms which
have been offered, I think it is important to keep in mind the relative
magnitude of the various parts of the costs and the portions to which
these comments apply.
As I pointed out in the discussion this morning relative to table 8 of
"Exhibit" No. 1416," ^ there are two major elements in the total costs:
The pay roll and the "other expenses." With reference to the matter
of depreciation and depletion, the allocation among various years is,
of course, arbitrary, and I think that Mr. Taitel would be reluctant
himself to specify exactly what allocation was the proper one.
With reference to the matter of maintenance and the allocation of
these other costs among years, the propriety of the allocation depends
eventually upon what you meaji by a variable cost and by a fixed cost.
From one point of view, you might insist that those maintenance and
repair charges should be allocated equally over every one of the years.
From that point of view, they would all become fixed costs. Or you
might say that they should all be allocated in proportion to the volume
of output. Or you might say that they should be allocated in such a
way as businessmen do allocate them, confronted as they are with the
ups and downs of business.
Now, the last is substantially the definition of variable costs which
we have used. We have in our separation of costs between fixed and
variable, attempted to adjust as best we could, for the effects of
changes in the wage rates and the prices and the tax rates, which the
Corporation must pay at different points in the cycle. We have at-
tempted to adjust also for the change in efficiency as reflected by the
downward trend in the costs. We have not eliminated such in-
equahty in the distribution of maintenance items as results from ad-
justments to the ups and downs in volume of business, and I submit
to the committee that that is the appropriate treatment of the item.
I should take issue with the criticism which has just been made,
that maintenance cost is really a cost which ought to be allocated
equally year by year so that it would thereby become fixed rather
than variable. But I have no quarrel with that; it is merely a matter
of definition of terms. I suggest that the definition I used was the
appropriate one.
' Dr. Bean's testimony begins infre, p. 13719.
> Apoendix, p. 14040.
CONCENTRATION OF ECONOMIC POWER 13711
Acting Chairman King. How could you make that a fixed charge
when one plant with the same uses may become obsolete in, say, 5
3^ears, and another in 20 years? How could you say that that should
be a fixed charge and should be the same during that entire period?
Dr. Yntema. My opinion is that we have handled the situation
correctly for the purpose with which we were concerned, namely, a
study of how costs fluctuated in the business cycle, and the possibili-
ties of reduction in costs in the cycle. It seems to me that the best
guide we could take in the situation is what businessmen did do when
they were confronted with the kind of phenomenon which we are
now considering. That was their reaction to the pressure of changes
in volume of output.
Acting Chairman King. I know of cases where valuable machinery
by reason of some technological development has become obsolete
within 2 or 3 years, whereas other machines treating ores lasted for
several 3^ears. In the first instance, they became obsolete in 1 or 2
years, and should have been charged off during thai, year instead of
being continued over several years.
Dr. Yntema. May I read for the committee one of the introduc-
tory paragraphs in this statement by Mr. Taitel :
So far as prices in such an industry are based upon costs, the pricing poHcy would
tend to be one that provides for declining prices as volume of output increases.
Prices in the steel industry , however, have not followed this pattern. They have
tended to remain relatively fixed. The typical practice has been to increase
prices with increased volume rather than to decrease prices as sales expand. Such
price behavior is much more consistent with the situation in which increasing
output is associated with constant or rising costs.
That paragraph, it seems to me, abstracts to a considerable extent
from the factors which really do account for the cyclical changes in
costs and prices. The outstanding characteristics, which are im-
portant in determining the price in the market, are the tremendous
shifts in demand in the business cycle. Those are outside the control
of the Steel Corporation or the steel industry.
The second consideration which cannot be neglected is the changes
in the prices which the Corporation and the industry must pay for the
materials and for the labor services which they must have for the
production of their product. It seems to me that those have been
neglected i'l the particular suggestion.
I should like to call attention to the chart which was submitted by
Mr. Taitel. I do not know, and I do not think the witness who
presented it knew, the basic material from which it was constructed.
Judging from what I know about the total costs in the Corporation,
and from what I know about the cost behavior in many of the sub-
divisions, I am inclined to be extremely skeptical as to the validity
of the chart, at least mitil we know the basic data upon which the
chart is constructed. I think that we ought to reserve judgment
with respect to its validity.
Acting Chairman King. I didn't hear all the testimony.
Dr. Kreps. The chart is on the easel, "Exhibit No. 2184."
Acting Chairman King. Does the chart indicate a fixity of price
for continued periods of time, say for coal or for ore or for freight
rates, notwithstanding what we know to be the fact that has varied
very much in price during the past few years, particularly since the
13712 OONOENTRATION OF ECONOMIC POWER
Bituminous Coal Act, and the price of ore varies, too, by reason of
many conditions
Dr. Yntema (interposing). Senator, I wish I knew the material
from which this chart was prepared. I don't.
Dr. Kreps. You are on the same ground that we are when we try
to evaluate your charts. Isn't that correct?
Dr. Yntema. No; that is not correct.
Dr. Kreps. We have in no case seen either the basic material or
the work sheets. We know fully as much about Mr. Lippert's chart as
we know about any of yours.
Dr. Yntema. But we don't even know in this case in the chart
presented by Mr. Taitd whether there has been any adjustment for
the prices paid for the materials, any adjustment for wage rates, any
adjustment for changes in "efficiency. There is no description from
the material presented as to what the cost means from which this
chart is derived, and I simply don't understand it, and I don't see
how the conmiittee can possibly understand it.
Acting Chairman Ki5ig. Does it assume ai sort of continuous line,
using the charts that we have, for costs and for those conditions?
Dr. Yntema. Senator, I simply don't know; I just don't know
what it means.
Acting Chairman King. We know that there are constant variables
in so many of the activities connected with the steel industry, and for
that matter, all industries.
Dr. LuBiN. May I ask the witness: Dr. Yntema, are you acquainted
with Iron Age?
Dr. Yntema. Yes.
Dr. LuBiN. Do you accept it as a reputable journal?
Dr. Yntehja. What do you mean by "reputable" journal?
Dr. LuBiN. In the sense that the members of the industry quote
it as representing the points of view of conditions in the industry and
as a good reporter of conditions in the industry.
Dr. Yntema. For some statistical purposes, I would say it is a
good reporter. As far as individual articles are concerned, especially
in a case as technical as this, I should say I would want to reserve
judgment in appraising any particular article.
Dr. Lubin. Do you know Mr. Lippert?
Dr. Yntema. I do not.
Dr. Lubin. Do you know anything about his standing in the
profession?
Dr. Yntema. I do not.
Mr. Lubin. Do you know whether members of the Corporation or
members of the industry look upon him as a responsible person?
Dr. Yntema. I do not.
Dr. Lubin. In other words, then, when you question his data, you
are not questioning hi^ competency.
Dr. Yntema. Oh, no, not at all. I am merely saying that I dont
know what this means, and until we have a description of what it is,
I suggest that we can't interpret it properly. He may be completely
competent. I don't mean to imply he is otherwise.
There recurs in the statements by Mr. Taitel as weU as others the
implication that we here have presented a guide for pricing policy
by the Steel Corporation and by the steel industry. May I once again
repeat that such is not the case, that we have attempted to present
CfONOENTRATION OF ECONOMIC POWER 13713
a description of how costs vary with output upon certain assump-
tions. We have attempted to describe how the volume of steel sold
by the industry varies with the price of steel to give some basis for
judging what would be the effects of a change in the price of steel
upon the quantity sold and upon the profits and losses of a corpora-
tion such as the Steel Corporation.
Dr. Kreps. The point you are making is very important. I want
to underscore it. You emphasize that your analysis is "not a guide
for pricing policy by the Steel Corporation and by the steel industry."
It certainly is not a guide for the T. N. E. C. For example, let us
examine a little more closely the costs which you are talkmg about.
They do not apply to any particular steel product? Correct?
Dr. Yntema. The costs apply to all the steel products and the
other operations of the Steel Corporation.
Dr. Keeps. To what you call the "product mix"
Dr. Yntema (interposing). The composite total of aU products.
Dr. Keeps. They do not apply to costs within any particular
plant.
Dr. Yntema. They apply to costs within all plants.
Dr. Keeps. They only apply to costs within aU plants of the Steel
Corporation. They do not apply to costs within other plants and
outside the Corporation, nor to the totality of such other plants in
other corporations, nor to the steel industry as a whole.
Dr. Yntema. That is quite correct; our analysis of costs was
necessarily restricted- to the material we had in the Steel Corporation^
Dr. Keeps. In other words, your cost curve is what you regard a
convenient summary of cost experience as you have seen it in the
Steel Corporation throughout the period which you covered.
Dr. Yntema. If I may speak frankly, I don't like, the word "con-
venient." I would say that io is an inappropriate description.
Dr. Keeps. I withdraw the word "convenient." The summary is
one that you feel enabled you to get a good glimpse and to give us a
good glimpse of what seemed to you to be the cost relationship for
those 50,000 products in that varying product mix, composed of steel,
cement, and the like.
Dr. Yntema. Yes, I think that is a correct statement.
Dr. Keeps. These are theoretical costs, not actual costs.
Dr. Yntema, They are actual costs adjusted to eliminate the effect
of variables which we did not want to leave in because they woiild
becloud the picture.
Dr. Keeps. But they are not actual costs of any actual product.
Dr. Yntema. But of the actual group of all products.
Dr. Keeps. Nor are they costs actually incurred in any actual plant.
Dr. Yntema. I think we covered that.
Dr. Keeps. The distinction is extremely important and is one upon
which there i&no disagreement. I merely want to help Dr. Yntema
clear up the confusion.
Dr. Yntema. May I say in concluding these remarks that I appre-
ciate very much indeed the scientific character of the discussion which
has just preceded, but that, in appraising it, I don't think it alters
substantially my views as to the applicability of our findings to the
purposes for which they were designed.
Mr. Fellee. May I just ask this question? Throughout this dis-
cussion I have kept recurring in mv mind this nnestion: Hero is n vfurv
13714 OONOENTRATION OF BCION JMIC POWER
important problem to be examined. Dr. Yntema has examined it in
one-way. Is it possible to examine it in another way? Dr. Yntema,
I should like to ask you from your experience with the materials at
hand in the Corporation, would it have been possible to have dealt
with this problem of the variations in costs at different rates of output
by taking the costs of a particular plant of the Corporation, and instead
of using these historical aggregate costs, to have considered the costs of
the various operations that go into making steel?
Dr. Yntema. Mr. Feller, we considered that with great care before
we undertook our analysis. Our decision as to procedure was dictated
by these considerations : First, we were fearful if we came to the com-
mittee with costs which involved necessarily arbitrary allocations of
overhead that we should be criticized for any allocation that we made
because any such allocation is to some extent arbitrary. In the second
place, we thought the committee would be more interested in the total
picture than in the picture presented in the individual plant. We did,
as a matter of fact, make numerous studies of the variation in costs
with output for short periods of time in individual plafl.ts, all of which
confirmed the findings which we have here presented with respect to
linearity, the straight line behavior of total costs with volume. Of
course, the absolute level of those costs was not material for the.
general purpose which we had in mind.
I should like to point out also that even if we were to take an
individual plant we would encounter the same type of difficulties thatQ
we encountered here. Furthermore, the work would mount to un-
reasonable proportions because each plant produces not one product
but many products, and for each plant, therefore, it would be neces-
sary to construct an index of production to relate to costs, and we
should have to make all the types of adjustments we have here. The
job would be so complicated if attacked in that way and the results
would be subject to such great question that we did not think' it
desirable to approach the problem in that manner.
Dr. LuBiN. I should like to ask Mr. Yntema just a question or
two in regard to methodology. Mr. Yntema, if United States Steel
Corporation would ask you to appear and do a job for them on cost-
ing, there are six or seven products on which there is doubt in the
minds of certain officials as to whether the price they are charging
is the right price, j"ight, being the economic price, would you proceed
to (Jo the job the way you did this one here?
, Dr. Yntema. The question you are asking is a very different ques-
tion from the one we posed, and I should say that consequently I
should approach it in a different way. I should go on further, if I
may. Dr. Lubin, and point out that your question is still an ambig-
uous one.
Dr. Lubin. Let's be specific. A certain vice president "of a very
large corporation fcomipented to me recently, he doesn't know how
his prices are fixed. He was interested in throe products. If he
asked you to come in and check up and find out whether the price
they are charging is a proper economic price, particularly in view of
its relationship to costs, and give thorn a cost picture
Dr. Yntema (interposmg) . The proper economic price, according
to my view of things, is the price that you can get in the market.
It is not determined in the short run by costs; it is determined
primarily by what your competitors will offer the product for and
CONCENTRATION OF ECONOMIC POWER 13715,
what the public will pay for it. It doesn't seem to me that in a
competitive situation the function of costs in the short run is to
serve as a basis for the establishment of price. That is what I re-
ferred to in raising the issue with respect to the interpretation of
your question.
Dr. LuBiN. Let me change my question and be more specific. If
you were asked by this official to tell him whether or not the price
that' they are receiving is such that it makes the Corporation a good
profit on the basis of its cost, would you proceed the way you have
in this?
Dr. Yntema. No, because the question is different from the ques-
tion W which we addressed ourselves. We were not concerned in
this particular problem with the individual product prices, the
reasons for that I have tried to point out. If it had been possible
to present unambiguous results, and the burden of expense and time
had not been toe great, we should have been only too glad to attack
that problem, but we worked (a) under limitations of time and ex-
pense, and (6) also under limitations that are theoretically inherent
in the problem, that is, with reference to the allocation of overhead
or fixed costs.
Dr. LuBiN. In other words, frankly, my purpose in asking this, I
am trying to formulate in my own mind what these figures are good
for.
Dr. Yntema. These figures are good, if I may say it again, for
these purposes; we wanted to find out on the demand side how the
total quantity of steel sold would vary with the price. You might
approach that product by product. I think you would get into
difficulties which would make your problem insoluble if you did so,
because of the substitutability among products. In the second place,
we wanted to find out how the costs in a concern such as the Steel
Corporation varied with output to discover what would be the effects
of price reductions or price increases upon the costs of such a corpora-
tion. We merely want to present such information to the committee.
Dr. LuBiN. In doing that latter, you had to use certain arbitrary
assumptions.
Dr. Yntema. It depends upon what you mean by "arbitrary
assumptions."
Dr. LuBiN. You had to make a decision. Shall we allocate this way,
or that way?
Dr. Yntema. That is just what we didn't do for the most part.
We attempted to select a method which would minimize the necessity
of arbitrary allocation. That is why we did what we did; we didn't
want to engage in any more arbitrary allocation than was necessary.
Dr. LuBiN. Wasn't selection of the method itself an arbitrary
thing? You might have selected another method.
Dr. Yntema. That is quite right, but the selection of the method
is arbitrary only insofar as there is a set of alternatives for the pur-
pose you have in mind.
Dr. LuBiN. What I am trying to get to. is whether or not Mr.
Feller hasn't really struck the kernel of things, that the same job
might have been done by somebody equally well with a different set
of assumptions and different set of methods and gotten entirely
different results, and although you might not have agreed, there
might have been people wiio did, in fact you might have found two
13716 CONCENTRATION OF ECONOMIC POWER
corporations doing this thing differently, getting different results, and
both having authority for their methods.
Dr. Yntema. I see what you are driving at, I think. Let me say
this in answer to it: I think it is possible, given sufficient time and
money and patience, to study the relationship of the costs which are
associated directly with individual plants in relation to their output.
The problem is an overwhelmingly great one if you are going to cover
any considerable territory, and it still leaves untouched a certain
important realm, that is, namely, the allocation of the overhead not
associated with those individual plants. To some extent you can get
part way in the problem, but there still is a margin of indeterminancy
in any solution that you obtain by it.
Acting Chairman King. That becomes more of a problem if you
have several hundred or several thousand commodities growing out
of this same general activity.
Dr. Yntema. You would have the same problem, plant by plant,
as you have in deahng with the total corporation.
Dr. LuBiN. I had an old professor on'^e you probably know him,
Henry Carter Adams, a man who institute . the first series of accounts
for the Interstate Commerce Commission. The first sentence he
always told us in his classes was that accounting was purposive ; your
purpose would determine the method you used. I wonder whether
the same wouldn't be appUcable to the testimony you have presented.
Dr. Yntema. I think that is a very important question which you
have raised. We selected this method on the basis of two considera-
tions, its applicability of method and the exigencies of time and
expense. And I should like to point this out, if I may: That neither
ordinary accounting records nor cost-accounting records reveal
immediately the kind of data which we needed, and that is why we
resorted to this type of analysis which we think is applicable to the
problem for which we designed it.
Mr. Feller. I should like to ask you another question along this
line. As I understand the method used, the fundamental, the basic
datum in the whole tiling is a chart, a scatter diagram, as statisticians
call it, on which you place 12 dots. It all began from that, didn't it,
the cost- volume relationship in each of 12 years?
Dr. Yntema. I think there is some misapprehension on the part of
those who havf^ participated in the discussion as to the relative import-
ance of the demand side and the cost side. I should say that the
inelasticity of the demand for steel is important as well as the cost
behavior. I should deny that it all started in this one httle scatter
diagram.
Mr. Feller. I agree with you and I would, like to amend that. I
was addressing my remarks entirely to the cost analysis. I may say
that my question might be quite different if I were addressing myself
to the demand. Just to narrow dowTi my question, have you tried,
or is it possible to try, to put more dots on by taking more years?
In other words, supposing you went back to 1906 and plotted that,
have you tried checking your results by taking more years?
Dr. Yntema. The reason for the selection of the years which we
took was that we did not have satisfactory records for the t.djustment
of the pay rolls for prior years. I don't want the impression to remain
that the evidence in the chart is the only evidence on \Nnich we based
our final conclusions. We confirmed this study, as I said, by numerous
CONCENTRATION OF ECONOMIC POWER 13717
other studies, and speaking now as a statistician and economist, I
was amazed to find how closel;^ in so many cases the total costs were
almost precisely a linear function of output. It was one of the real
surprises of my life as an economist.
Mr. Feller. Was it possible to test the result, that is, the linear
result, by analyzing the profit-and-loss record of the Corporation?
Dr. Yntema. I don't think that would give nearly as satisfactory
results as this. If you take the profit or loss record, the profits or
losses are a function both of the prices charged and the costs which
are involved, and it does not seem to me that tnat is an appropriate
type of analysis.
Mr. Feller. I would like to put it to you concretely. Supposing
you took the records of operation of the Corpojation from July 1938
to January 1940 — now, that is a period that covers six quarters — in
which there has been no substantial price change, wouldn't that be a
good test period, not necessarily to derive a fine like this, but to test
whether or not the line that you have derived is an accurate one?
Dr. Yntema. Well, there are difficulties in that. For one thing,
you will run into some problems in connection with seasonal varia-
tions of cost. There is one difficulty iji connection with t^at that is
rather serious.
We did test out our material by studying the variation of pay roUs
and of hours worked with output, and got confirmation of our findings.
I should be rather skeptical of the sort of procedure you suggest, and
again I want to make clear that we did not have all the time in the
world to undertake the study. We were limited.
We were engaged in a large number of other studies besides this one,
and we made the best of what facilities we had. I doubt very much
that the profit-and-loss analysis you suggest would throw great light,
however, on the problem
Mr. Wooden. Might I ask a question? Dr. Yntema, can you say
whether the average price level in 1939 was less than in 1938?
Dr. Yntema. May I answer that with a chart? I think I can
show that very easily.
Mr. Wooden. I mean for steel.
Dr. Yntema. May I show you that by chart?
Mr. Wooden. Any way you like.
Dr. Yntema. Let me present this chart, if I may. This chart is
numbered C-9, "Exhibit No. 1409." It is entitled, "Reported Com-
posite Price and Composite MUl Net Yield." ^ The heavy line in
the chart represents an index of the composite mill net yield to the
United States Steel Corporation subsidiaries. The dotted line is the
reported composite price by "Iron Age." This line extends in 1939
through November, and it is apparently from inspection of the chart
that the average price reported by "Iron Age" is substantially lower in
1939 than in 1938, and also that the composite mill net yield to the
Corporation is similarly substantially lower in 1939.
Mr. Wooden. Is it as much as 10 percent lower?
Dr. Yntema. The reduction from the high point in 1938 to the
low point m 1939 is from 105.9 for the mill net yield to 91.4 That
would be a drop of 14.5 points, a drop of approximately 14 percent
from the high point in 1938.
« Appendix, p. 13815.
13718 CONOENTRATION OF ECONOMIC POWER
Mr. Wooden. On page 16 of your prepared statement which you
read, you estimated that a 10 percent decrease from the 1938 average
level would produce only 11 percent increase in volume.
Dr. Yntema. No, we didn't estimate that. What we said was that
was more than the most optimistic response which could be expected.
We didn't mean to imply that we thought that was a reasonable
expectation.
Mr. Wooden. Was that more than could be expected?
Dr. YNTEkA. Oh, yes. Our best guess is that the elasticity of de-
mand was probably not more than 0.3 or 0.4, that a 10-percent reduc-
tion in price probably wouldn't of itself bring about more than perhaps
a 3 or 4 percent increase in volume. That was a very crude estimate ;
it may be more than that, or less than that.
Mr. Wooden. Is that consistent with the fact that during 1939 the
total volume for the industry iiicreased as much as 65 percent?
Dr. Yntema. Our statement is not at all inconsistent with the facts.
We went to considerable length to point out that the primary factors
determining the quantity of steel sold, are industrial profits, national
income, or, if you want to put it in another form, other industrial
activity, and that the price is a relatively minor factor. I should say
that the evidence for the years which you cite bears out exactly the
findings which we have submitted to the committee, that there were
other factors far more important than price which determined the
course of events in those years.
Acting Chairman King. Any other questions?
Dr. Lubin. Is it possible to find out how much these other factors
weigh as opposed to price changes?
Dr. Yntema. That is a far-reaching question, Dr. Lubin, and a
very good one. One of the devices open to us is to attempt a multiple
correlation analysis. The results of such analysis have been taken
somewhat more seriously by the committee than I thought they would
be taken. I should regard the results of our correlation analysis as
merely confirmatory of our other findings, and I should not attach
great precision to the results which are obtained thereby. I think
they do, however, give us some evidence on a very complicated, diffi-
cult problem.
Acting Chairman King. Thank you very much. Doctor.
Dr. Kreps. I suggest that Mr. Taitel be di-: /-^ed as a witness.
Dr. Yntema will probably wish to make comments later on.
Dr. Yntema. I should like to respond to several of the points which
Dr. Ezekiel raised in his testimony
Dr. Kreps. That may be done tomorrow.
Acting Chairman King. The committee will stand adjourned until
10:30.
Dr. Kreps. I should like to add a word of thanks to Dr. Yntema,
who consented while the material is still fresh in our minds, to stay
with us and give us the benefit of his additional comments.
Acting Chairman King. I think we are indebted to him.
(WTiereupon, at 4:45 p. m., the committee recessed until 10:30 a. m.,
Thursday, January 25, 1940.)
INVESTIGATION OF CONCENTKATION OF ECONOMIC POWER
THURSDAY, JANUARY 25, 1940
United States Senate,
Temporary National Economic Committee,
Washington, D. C.
The committee met at 10:45 a. m., pursuant to adjournment on
Wednesday, January 24, 1940, in the Caucus Room, Senate Office
Building, Joseph J. O'Connell, Jr., special assistant to the General
Counsel, Trfe ary Department, presiding.
Present: Mr. O'Connell (acting chairman), Senator King, Repre-
sentative Wilhams.
Present also: John V. W. Reynders, representing the Department
of Commerce; Walter B. Wooden, representing the Federal Trade
Commission; Frank P. Smith, representing the Securities and Ex-
change Commission; Martin Taitel, senior consulting economist.
Work Projects Administration; A. H. Feller, special assistant to the
Attorney General, and Melvin G. deChazeau, consulting economist,
Department of Justice.
Acting Chairman O'Connell. The committee will please be in
order.
Dr. Kreps being ill today, Mr. Feller is going to take his place.
Will you introduce the first witness, Mr. Feller?
Mr. Feller. Mr. Chairman, I imagine that my ignorance of the
science of econometrics is almost unparalleled, and I may be in con-
siderable difficulty in conducting examination of this very abstruse
science.
Yesterday afternoon, the committee may remember, Dr. Yntema
pointed out that his analysis had two parts of great significance.
One part was the analysis of cost and the other was the analysis of
demand. The testimony yesterday was Concerned mainly with the
analysis of the cost situation by Dr. Yntema. I should like to call to
the stand Dr. Louis H. Bean, who, I understand, will discuss the other
problem, the problem of demand.
Acting Chairman O'Connell. -Will you hold up your right hand,
Dr. Bean, please. Do you solemnly swear that the testimony you
are about to give in this proceeding will be the truth, the. whole truth,
and nothing but the truth, so help you God?
Dr. Bean. I do.
TESTIMONY OF DR. LOUIS BEAN, ECONOMIC ADVISER.
DEPARTMENT OF AGRICULTURE, WASHINGTON, D. C.
Mr. Feller. Dr. Bean, will you state your name and position for
the reporter?
Dr. Bean. Louis H. Bean, economist in the Bureau of Agricultural
Economics of the Department of Agriculture.
13719
13720 OONOENTIIATION OF ECONOMIC POWER
Mr. Feller. How long have you been with the Department of
Agriculture?
Dr. Bean. I have been with the Department of Agriculture 16
years.
Mr. Feller. And you are primarily a statistician?
Dr. Bean. I am primarily an economist, doing a great deal of
statistical work.
Mr. Feller. Where did you receive your training as statistical
economist?
Dr. Bean. I am a graduate of the University of Rochester, Roch-
ester, N. Y., and Harvard School of Business; sinee graduating from
the Harvard School of Business, I have been with the Bureau of
Agricultural Economics.
Mr. Feller. Do you have a statement, Dr. B;ean?
Dr. Bean. I have a statenaent.
Mr. Feller. Will you proceed?
EXAMINATION OF UNITED STATES STEEL CORPORATION ANALYSES
Dr. Bean. An examination of the four U. S. Steel Corporation
statements on demand analyses for steel in the container, automobile
and railroad industries and for all industries combined, reveals grave
statistical defects. These are defects in methods of analysis as well
as in assumptions and in data. In the first three of these studies, the
important objective was to reveal the effect of price on consumption,
but the methods and data used were inadequate, with the result that
the quantitative conclusions arrived at are unreliable, and so gen-
erally recognized by the authors. In some cases adequate data were
not fully utilized and important price-volume relationships remained
undetected.
In the study dealing with total steel consumption by ail industries,
several of the quantitative analyses presented are statistically im-
reliable because of the wide range within which the "true" relation-
ships between price and volume may lie. No account was taken of
the extent to which one or two extreme observations influenced the
results obtained. In certain cases where the analyses show little
influence of price on volume, a close examination of the data used
reveals substantial price influence ; and in cases where low prices were
found to be associated with low volume, the underlying relationships
can be shown to be just the opposite.
The conclusion of the analysis of the demand for steel in the auto-
mobile industry is that the elasticity of the demand for steel used as a
raw material in the automobile industry is very low, and that a 10-
percent reduction in the price of automotive steel would increase car
sales by only 1% percent and the consumption of automotive steel
by not more than 2 or 3 percent.
In support of this conclusion, there is presented on page 19 a scatter
diagram purporting to show the relation of automotive steel prices to
total automobile steel consumption for the years 1924-38. It is
claimed that this scatter diagimn of percentage changes in the annual
consumption and annual average price "fails to indicate that lower
steel prices aro associated with greater quantities of steel purchased
and vice versa." Actually, the basic relationship does indicate that
lower steel prices are associated with increased consumption of steel
CONOENTRATION OF ECONOMIC POWER 13721
in automobile production. The failure of the authors to note this is
due to the fact that there are other factors besides the price of steel
which cause changes in the total consumption of steel in automobiles
and that statistically it is not possible to determine the effect of one
factor on another when the other important elements in the problem
are not taken into account.
(Senator King assumed the Chair.)
Acting Chairman King. General improvement in conditions would
necessarily increase the production and sale of automobiles, without
much regard for the price of steel.
Dr. Bean. If that factor is not taken into account, Senator, then
the underlying effect of price on volume cannot be clearly seen.
Acting Chairman King. There are so many factors that must be
taken into account in determining the question of costs, cost prices.
Dr. Bean. That I take it was gone into quite thoroughly yesterday.
In a moment, Senator, I think I shall be able to indicate to you how
it is possible to take this factor of demand into account in this partic-
ular illustration.
Had the authors of this analysis taken into account changes in the
level of demand, they would have found an inverse relation between
the price of automotive steel and automobile steel consumption.
Even in the scatter diagram on page 19 it is clear that the relation of
price to consumption is negative, as is indicated by the observations
for the years 1925 to 1929, inclusive. Similarly, the observations for
the years 1930-32 and 1933-37 show the inverse relationships but at
different levels of demand.
May I take a moment to illustrate what these words mean. The
chart on page 19-
Mr. Feller (interposing). Page 19 of "Exhibit No. 1413," ^ entitled
"Analysis of the Demand for Steel in the Automobile Industry."
Dr. Bean. This is an enlargement of that chart.
Now, the reason I call your attention to it is that it is used as a sort
of clincher to the aigument that price is relatively unimportant, and
that as far as this illustration is concerned, it does not show that lower
prices are associated with increased, consumption.
I have already pointed out that in certain years, there is an inverse
relationship. If you examine the data for. the years 1925, 1926, 1927,
1928, and 1929 only, you find that the relation of price change to
volume change is negative; that is, reductions in volume being asso-
ciated with price advances. If you take another group of years in
sequence, 1933, 1934, 1935, 1936, and 1937, these observations also
lie along a, negatively sloped regression. Similarly, if you take the
price decline in years of business depression, namely, 1930, 1931, 1932,
and 1938, you also have a suggestion that the relationship is negative.
I want to come back to this illustration in another moment.
Acting Chairman King. Doctor, have you taken into account in
your study the, what shall I say, psychological condition of the people
resulting from apprehensions of bad or good legislation, bad or good
times, foreign affairs, possibility of protracted peace or. of a violent
war; are not all of those things to be taken into account when you are
trying to formulate a rule, if you can call it a rule, for the determina-
tion of the cause of the rise or decline in prices?
' Appendix p . 13993.
13722 oonoeNtration of economic power
Dr. Bean. Your question, Senator, leads me to think that I ought to
make clear the position in which I appear here as a witness. I have
not undertaken to analyze the supply and demand factors for the
steel industry. I have spent a great deal of time over these reports
that have been submitted to you.' They are the kinds of studies for
which many of us statisticians and economists have been waiting for
years. They contain a great deal of very useful and interesting
information. Having studied factors which affect agricultural and
industrial prices, I have looked at these reports quite minutely to
glean from them fundamental knowledge about the iron and steel
industry. I am here merely to pass on to you, who. undoubtedly have
not examined these things in as close detail as I have, some of the
over-all impressions that, as a statistician, I obtained from studying
these reports. I have no original, contribution to make except,
perhaps, to indicate to you' the method by which I examined these
statistical reports.
Acting Chairman King. Of course, you woidd concede that the
situation in Asia, where Japan has obtained for a number of years
large quantities of scrap and steel, would affect the prices more or less,
or affect the market or the demand for steel?
JDr. Bean. I would, yes.
Acting Chan-man King. And similarly, conditions in Europe, if we
have peace or war?
Dr. Bean. I think so le of these unusual conditions show up
suggestively in some of t! e material that I want to go over with you.
By introducing the additional factor of changes in industrial pro-
duction (as a measure of changes in the level of demand) into this
analysis, it can be shown that for, a given reduction in automobile
prices, the associated increases in automobile-steel consumption have
varied directly with the rate of change in business.
When changes in industrial production are inircduced in this
analysis, we find that the price changes in the years 1933, 1935 and
1936 are associated with increases in industrial production of 14 to
19 percent.
Price changes in the years 1925, 1926, 1928, and 1929 with increases
in industrial production of 4 to 10 percent; in the years 1927 and 1924
with decreases in industrial production of 2 and 6 percent respectively;
and in the years 1930, 1931, 1932, and 1938 with decreases of 16 to
22 percent in industrial activity. The only 2 years for which the data
seems to be out of line are 1934 and 1937 when factors other than the
relatively small changes in industrial production apparently supported
automobile-steel consumption at a relatively higher level. If bookings
instead of the indirect measure of consumption used here were avail-
able, it is quite likely that these 2 years (1934 and 1937) would also
show as consistent a relation of price to volume as we find for all the
other years in this analysis.
,In other words, in this chart ^ that I have just described there are
15 observations, 13 of those 15 lie in such a position as to indicate an
inverse relation of price to volume, and the 2 exceptional years,
were undoubtedly years of greater anticipations, speculation in indus-
try, that distort the price volume relationship.
' Exhibits introduced by U. S. Steel Corporation.
' Chart 6 of '.'Exhibit No. 1413"; appendix, p. 13993.
CONCENTRATION OF ECONOMIC POWER 13723
Mr. O'CoNNELL. You mean a decrease in prices associated with
substantial increases in demand in those years.
Dr. Bean. May I illustrate by referrmg to chart 6 of "Exhibit
No. 1413." »
I would like to take years when business was rising rapidly, some-
where between 15 and 19 percent per year. There are three cases
of that sort in this analysis. They are these years right here, 1933, 35,
and '36, and I think any of you dealing with just those 3 years would
observe that the relation of these three points is such as to indicate a
. negative relationship at this level of demand. These other years
when business was rising rapidly are here. Let's now take years
when business was rising moderately, somewhere between 4 and 10
percent a year, instead of 15 to 19. The years of that type are 1925,
'28, '26, and 2 additional years, '34 ajid '37.
There is another type of situation, namely, when business is declin-
ing rapidly, as in the years 1930, '31, '32, and '38, and the underlying
relationship here too is a negative one. These data show that the
underlying relationship of price changes to volume changes is negative,
if the different business situations are taken into account. This
illustration is not conclusive evidence that price changes and volume
changes do not reveal a negative relationship.
Acting Chairman King. You appreciate, I suppose, Doctor, that
the introduction into our stream of economic business life, a very
large Federal contribution, such as 2 or 3 billion of dollars duriag the
bonus period, and the large appropriation of more than $3,600,000,000
just for relief and other matters in 193^3, would necessarily have some
effect upon business and upon prices, it would revive in some instances
and possibly result in a decline in other avenues of business activity.
Dr. Bean. But none of these things. Senator, alter the basic fact,
if it is a fact, as revealed here that the relation between price and
volume is basically inverse. That relationship may be hidden com-
pletely by the things that you have mentioned, but the underlying
fact of an inverse relation between price and volume isn't altered.
The things you mention may swamp that basic relationship, if we
may put it that way, or cancel it; but price nevertheless has its basic
influence and my point is merely that there is an inverse relationship
between price and volume that can be revealed by the very materials
which in this report it is claimed do not reveal it.
Acting Chairman King. Well, prices are affected by the volume of
currency and the credit extended with the banks. Isn't that true?
Dr. Bean. Yes.
Acting Chairman King. And, of course, business then reacts to
those changes which result from the condition just alluded to.
Dr. Bean. That is quite true. We are, however, talking about a
report, Senator, which purports to portray the relation of price to
volume with all of these other things held constant or held unchanged.
Acting Chairman King. I am not talking about that report because
some of these reports and some of the addresses of learned doctors
and statisticians may not always afford a true basis for conclusions
which may be reached with respect to causes and effects in the rise and
fall of prices and in expansion and recession of business.
1 Appendix, p. 13393 .
124491 — 41— pt. 26 10
13724 CONCENTRATION OF ECONOMIC POWER
Mr. Feller. Dr. Bean, would it be correct to say that your testi-
mony is directed to precisely the point that the Senator has just
raised? In other words, does analysis and demand presented on behalf
of the Corporation reveal a true picture of the relationship between
price and demand?
Dr. Bean. As far as this particular illustration is concerned, it does
not to me. The statement is made in the report ^ that this analysis
does not reveal an inverse relation of volume and price. By making a
simple refinement in that analysis, I do find an inverse relationship.
The mere introduction of this additional factor representing changes
in the level of demand not only gives clear indication of a negative
relationship of price and volume but greatly clarifies the interacting
and at times the counteracting influences of price changes and demand
changes on the changes in the volume of automobile steel consump-
tion. Thus for a 10 percent reduction in automobile steel prices, the
associated increase in automobile steel consumption was about 80
percent when business activity was increasing at a rate of about
15 percent per year. It was about 50 percent when business activity
was increasing at a rate of 5 to 10 percent per year, and about 20 to 30
percent with business conditions remaining unchanged. On the
other hand, with business activity falhng around 20 percent per year,
a 10 percent reduction in automobile steel prices was accompanied
by a reduction in automobile steel consumption of about 30 percent.
These relationships are not to be taken as final but merely as an
indication that the general conclusion contained in the analysis of
demand for automobile steel of a 0.2 to 0.3 elasticity of demand under
normal conditions appears to be a substantial understatement.
The effect of price on the amount of steel used per car appears to be
understated in this analysis. This may be judged by comparing the
price of automotive steel (given in table 10 of "Exhibit No. 1413"^ with
the amount of steel consumed per motor vehicle. ^ There is here an un-
mistakable inverse relationship between the price of steel and the
amount of steel per car.
For example, in 1924 the price of automotive steel is given as 3.45
cents, the amount of steel consumed per motor vehicle is given as ap-
proximately 0.8 of a ton. By 1928 the price ot automotive steel had
been reduced to 2.67 cents and the amount of steel per motor vehicle
had increased to 1.4 tons. Similarly, for the period 1929 to 1933 we
find the price in 1929 of 2.69 cents associated with steel consumed per
vehicle of 1.16 tons; and the lower price of steel in 1933 of 1.89 cents
associated with steel consumed per vehicle of 1.8 tons. Conversely,
the higher prices for steel in 1934-38 are associated with smaller
steel consumption per car, averaging about 1.4 tons.
The exact relationship between the price of automotive steel and the
consumption of steel per vehicle is difficult to determine. For the
period of the 1920's the gross relation between price and consumption
of steel per vehicle suggests an elasticity of 2 or more. For the period
1933 to 1936 the gross relation between price and volume of steel per
vehicle suggests an elasticity of 1. For the entire period 1924 to 1938
the over-all relation suggests an average elasticity of somewhat less
than 1. These figures are greatly at variance with the conclusion of
the U. S. Steel Corporation analysis that "price considerations have a
1 "Exhibit No. 1413".
> Aiipendix, p. 13997.
OONOENTRATION OF ECONOMIC POWER 13725
minor influence in determining the consumption of automobile steel"
and that the elasticity of demand fdr steel as a raw material in the
automobile industry is probably no higher than 0.2 or 0.3.
Mr. Feller. May I jilst ask you a question there? The -relation-
ship which you have pointed out between the,' price 6f steel and the
amount. of steel used per car is a relationship, as I^ understand it,
which is derived from observation of the historical data, and it does
not necessarily suggest, does it, that the changes in the amount of steel
per car were due entirely to the price factor; i^^ ther words, they
may have been due to technical factors ^nd to ne .styling and to !new
technological equipment which made it possible to produce lighter
sheets or something of that sort?
Dr. Bean. That is right. I merely call attention to these factors
for this reason, that on the face of it, there is a significant inverse
relationship between consumption and price. The report did not
analyze that relationship so as to give an answer to the question
you raise.
With respect to the demand for steel in the container industry, it
is the conclusion of this analysis that the price of tin plate is of minor
significance in determining tin plate consumption, and /that in view
of the inelastic demand for product^ packed in * tin cans and the
relatively small proportion of the price represented by the cost .of
tin plate, a reduction in the price of tiu plate would be ineffectual in
increasing the consumption of steel.
This study fails to develop any statistical proof of the major assump-
tions involved in these conclusions. In fact, there is a good deal of
evidence in the data contained in this report which the authors- failed
to utilize and which shows price of both tin plate and of canned goods
as much more important than was assumed here. In this study,
there is no analysis of the relation between retail prices and con-
sumption of canned products to indicate the nature of the elasticity
for canned food products.
The conclusion that the elasticity here is small is merely inferred
from an analysis of the National Resources Committee which corre-
lated national income with consimiption of canned fruits. The fact
that that study cannot be used as a basis for inferring the nature of
the relation between price and volume consumed and that the relation
may actually be one of substantial elasticity is suggested by an exami-
nation of the relation between the price of canned tomatoes, the
size of the pack of canned tomatoes, and the price of tin plate, the data
being given in tables 12 and 13 of the report. If we assume that the
price of canned tomatoes is determined by supply as measured by the
size of the pack and by the price of tin plate, one of the important
cost items, it is possible to observe both the nature of the relation
of supply to price and of the relation of the price of tin plate to the
price of a can of tomatoes. An analysis set up along these lines indi-
cates, for example, that a pack of 14,000,000 cases iD:the 1920's was
associated with a price of 13 cents per can, when the price of tin plate
was kept at $5.50 per base box; and with a price of something under
12 cents when the nrice of tin nlate was reduced to $5.20 or $5.25.
The elasticity of demand for canned tomatoes has changed a great
deal during the period from 1923 to 1938, if the relation between the
size of the pack and price may be taken as indicative. During the
1920's it appears that a 10-percent increase in the size of the pack
13726 CONCENTRATION OJT ECONOMIC POWER
was associated with approximately a 2-percent decline in price, the
price of tin plate remaining michanged. In the more recent years,
particularly 1935 to 1938, a 10-percent increase in the size of the
pack has been associated with a 5-percent reduction in price.
These facts suggest that the elasticity of demand for some of the
canned food products may be quite different from that assumed by
the authors of the analysis of the demand for steel in the container
industry.
Furthermore, the price of tin plate as a factor in determining the
price of canned goods may be more important than the authors of
these studies indicate. This is suggested by a comparison that may
be readily made between the price of tin plate and the composite
price of canned goods as shown in one of the charts on the board,
and I shall point to that in a moment. It appears here that there
is a high degree of correspondence and therefore correlation between
the price of tin plate and the combined price of canned tomatoes,
peas, and corn. Compared with the relation between the price of
tin plate and canned goods prices in the 1920's, the prices of canned
goods in the last 5 years have been relatively low. This reflects
the effect of the large volume of production of canned goods, but
except for that fact there is apparently an underlying relation between
the price of tin plate and the price of canned goods which the authors
failed to investigate and reveal.
- And graphically, these words mean this
Mr. Feller (interposing). Dr. Bean, do you want to have that
in^'-oduced for thg record?
Dr. Bean. Yes, please.
Mr. Feller. Mr. Chairman, I should like to offer for the record,
chart entitled — I can't see that from here.
Dr. Bean. "Price of tin plata and of canned goods for the years
1923 to 1938, inclusive."
Acting Chairman King. It will be received.
(The chart referred to was marked "Exhibit No. 2186" and is
included in the appendix on p. 14124.)
Mr. Feller. Would you also state for the record the sources
from which the data were taken?
Dr. Bean. The price of tin plate is taken from the report on the
demand for steel in the container industry, the one I am just now
discussing.
Mr. Feller. That is "Exhibit No. 1415." '
Dr. Bean, And the composite price of canned tomatoes, peas, and
com is also taken from this report, except that we have combined
the three columns in one of the tables of this report as a composite
average.
The over-all relationship between the price of tin plate and the price
of these canned goods is of this sort, that for the years 1923 to 1928 or
1929 you have both prices of tin plate and prices of canned product?
at one level. Then both decline to approximately the same level in
1933, and both rise sharply in 1934. Since 1934 the price of canned
goods has been relatively stable, but at a somewhat lower level than
would be indicated by the previous relation between tin-plate prices
and canned-goods prices, and I am suggesting that the very large
volume of production of canned goods is partly responsible, if not
> Appendix, p. 14010.
OONOENTRATION OP ECONOMIC POWER 13727
entirely responsible, for keeping the price of canned goods relatively
lower than the price of tin plate, taking the basic relation between
them as that which existed prior to 1925.
The real significance of this illustration is that something more
probably needs to be done in all of these analyses in the way of relating
the price of raw materials, such as steel, to the price of the specific
things in which steel is used. Here we have a very definite suggestion
that steel is a predominant factor in the price of this consumer product,
for which the general assumption was made that steel prices have
very little influence.
The general conclusion that the price of tin plate is of minor signifi-
cance in determining its consumption is not borne out by the aggregate
volume and price data given in this report. If we relate the volume of
total production of tin plate as given in table 9 to the price of tin plate
given in table 12 (and adjust the price for changes in the general level
of prices of goods other than farm food and iron and steel products
referred to later), we find that the volume of tin-plate consumption
has been subject to a marked expansion in demand ever since 1921,
but that price has also been a factor in the volume consumed. On
the average, price has been close to 60 percent as important as all other
factors combined. Similarly, if volume of tin-plate output and indus-
trial production are examined as factors in determining the price of
tin plate, we find that tin-plate production is fully as important, if not
more so, than industrial activity as a price factor.
Before commenting on the study of the demand for steel by all
industries combined, it may be well to point to a general criticism of
the price-analysis technique used in the analyses of demand for steel
in the container, railroad, and automobile industries.
In the studies on the demand for steel in these industries, the relative
unimportance of price is deduced from a set of scatter diagrams show-
ing the relation between the year-to-year percentage changes in price
of the appropriate kind of steel and percentage changes in the factor
which represents the use of steel by the particular industry. In the
case of containers this latter factor is the pack of tomatoes, corn, and
peas separately ; in the case of the railroad industry it is the deflated
value of investment by railroads and railroad steel consumption ; and
in the case of the automobile industry it is automobile steel consump-
tion. None of these scatter charts displayed any significant correla-
tion and it was concluded that the price of steel was of very little
importance, if any, in determining steel consumption in these indus-
tries.
One of the difficulties in this procedure is the use of only year-to-
year percentage changes in one factor to show the existence or absence
of relationship without taking into accoimt other factors as Avell. It
is elementary in statistical analyses that no reliable conclusions re-
garding the effect of one factor on another can be drawn unless the
complicating influences of other factors known to be present have
been isolated. It is therefore surprising that there is no indication
of any attempt to include in this set of analyses any of the factors,
other than price, which might affect the demand for steel in the par-
ticular industry. Under these circumstances, there is a good chance
that no effect of price on steel consumption wouH be indicated even
if price variations had really caused, say, half the variation in con-
sumption. This point we have ^^Iready illasLafeed l . dealing with
13728 CONCENTRATION OF ECONOMIC POWER
chart 6 in the study on the Demand for Steel in the Automobile In-
dustry/ and showed that where no negative relationship seemed to
exist, the addition of an essential factor revealed unmistakably the
existence of a negative relation between volume of consumption and
price.
Turning now to the analysis of the demand for steel by all indus-
tries combined, in contrast with the foregoing studies, this one does
make an attempt to establish statistically the nature of price elas-
ticity for steel. Multiple correlations were made between various
sets of factors and the elasticity of demand was measured for each
price quantity regression obtamed. The equations for nine such
analyses are given in the appendix, but only four of these are discussed
in the text.
All of these four analyses are inadequate in isolating any reliable
estimate of the elasticity of demand. This becomes apparent when
we look beyond the equation, which we can do by plotting the price-
volume regressions and the individual observations adjusted for the
influence of other factors than price for close scrutiny, and if the
committee will bear for a moment with some of this technical hngo I
wUl elucidate with some graphic material.
We have plotted the price-volume relations and the mdividual sets
of data that they represent in one of the- charts. In the first of these
price-volume relations purporting to show the net efiFect of price on
steel ingot production, practically no influence of price was discovered
by the U. S. Steel Corporation analysts. The same data can, how-
ever, be made to reveal a negative relationship between the price of
steel and steel ingot production instead of the slightly positive in-
fluence found by the authors of this analysis, indicating that the
results here are, to my mind, indeterminate. The second price
quantity relation is derived from tli- same data including two addi-
tional years, 1920-21, and the rate l hange in industrial production
instead of a trend factor, as the additional elements. The mere addi-
tion of these two years gives a fictitious positive relation between
price and steel ingot production. This positive relationship is
determined entirely by one point far beyond the range of the other
observations. But for this one point the method and data used here
would give a statistically insignificant and equally questionable price-
volume relationship.
The third price-volume relationship is obtained in an analysis using
industrial profits, consumer income, and a composite price of steel in
relation to shipments. The relation of price to shipments as indicated
in this study is a slightly negative one, indicating very httle influence
of price on volume and is also statistically insignificant. This result
also is determined by two observations far remo v'ed from the range of
the other observations. Excluding th^^se two observations," the nega-
tive relation of price to. volume is much more pronounced and appears
to be fuUy three times as important as a factor-determining volume
than tUe authors found it to oe.
The fourth price-volume relation is obtained by relating steel book-
ings (instead of shipments) to industrial profits, consumer income, and
the composite price tt steel. This is the only study that indicates a
substantial negative relation between the price of steel and steel book-
ings. The relationship might be even more pronounced if a more
1 "Exhibit No. 1413," appendix, p. I39S1 at 13993.
CONCENTRATION OF ECONOMIC POWER 13729
adequate representation of steel prices were used or if a lag of several
months between price and volume were taken into account.
May I clarify these words with a chart containing these four relation-
ships?
Mr. Feller. Mr. Chairman, I offer the series of chartsentitled "The
Net Regression of Volume on Price."
(The chart referred to was marked "Exhibit No. 2187" and is
included in the appendix on p. 14125.)
Mr. Feller. As I understand it, these charts indicate analyses of
data which was contained in the reports submitted by the United
States Steel Corporation.
Dr. Bean. Yes. They are the relationships which are given in one
of the tables in the general report on demand for steel, with these
stated in four different equations and summarized in a tabulation.
Here we have plotted the equations and also the individual observa-
tions that are given in the study, so that we could observe whether
or not the relationships as given by the equations are satisfactory.
With respect to the first equation, I find that it is necessary to make
a slight alteration in the relation of industrial production to production
of steel, getting a little more pronounced influence of industrial pro-
duction on steel than is represented by the published equation, and
when that slight alteration is made I find that the relation between
price of steel and production is somewhat different from the one given
by the equation. Instead of finding a positive relationship between
price and production I find somewhat of a negative relationsliip merely
as a result of improving in this analysis the relation of industrial pro-
duction to steel production. We have here a negative relation of price
to volume instead of the positive one found by the United States Steel
Corporation analysts.
In the second equation there is given a positive relation of price to
volume; the higher the price the larger the volume of steel production.
Note that there is just one point way outside of the range of the
rest of the data which really controls the slope of that line, and by
the mathematical method used that is correct. One point far removed
from the body of the data controls the slope of this relationship.
Mr. Feller. Pardon me, Dr. Bean, what is that point? I can't
read it.
Dr. Bean. 1920. The year 1920 in this analysis shows price of
steel to have been veiy high as compared with the prices in all the
other years subsequent to 1920. In the first equation that year was
not used, therefore we found only a moderate slope of relation between
price and volume, but the inclusion of that one year of very high prices
necessarily shows a much greater influence of price on volume than
was obtained in the other case, and you can readily see that if that
one year were left out of the analysis all of the observations would be
contained within a very limited area, and that it would be difficult
to envision any kind of a relationsJiip of price to volume.
If any importance is to be placed on this particular part of the
analysis, you need to be aware of the fact that it is controlled almost
entirely by one observation, that it that one observation were left
out the result would be quite indefinite.
With respect to the thud equation, I again find that the relation of
price to volume is influenced to a large extent by two observations,
in this case the years 1923 and 1937 that are considerably removed
13730 CrONCENTRATION OF ECONOMIC POWER
from the area where the other price and volume data He, and that if
these 2 years were left out of the analysis, a much different result
would be obtained within this elipse.
(Mr. O'Connell assumed the Chair.)
Mr. Reynders. May I ask what is the year of that low point?
Dr. Bean. One is 1938, the other is 1934
Mr. Reynders. Does that practically balance the high?
Dr. Bean. It does in this particular analysis; yes. By the mathe-
matical method used here one point offsets the other and therefore
you get a relatively insignificant influence of price on volume, but
note that the result would be quite different if you leave out these
2 extraneous years.
Mr, Reynders. Can't you leave out the two in the lower ring?
Dr. Bean. You may do that, but if you leave out these two, then
the relationship takes on a steeper slope, which is more significant
than the one contained in the report. If, as you suggest, you leave
out also 1938 as an extraneous year or as one that doesn't belong in
this analysis, then you still find that the relation of price to volume
is greater than the one indicated here. For the group of years 1924,
'25, '26, '27, '28, and '29, the relationship is not the one presented
in the United States Steel Corporation report, but something of a greater
slope, and if you deal similarly with the other years, 1930, '31, '32, '33, and
'34, you again get a slope which is steeper than the one given by the
published equation, so that accepting your suggestion that we leave
out also 1938 as well as '23 and '37, I still find that the basic relation-
ship as indicated by these data is something more significant than that
given by the equation.
Finally, there is the fourth study where bookings are used instead
of shipments or production, and here we do find a substantial relation
of price to volume. The report indicates that the elasticity is 0.88,
in other words a 10-percent increase in price associated with close 'to
a 9-percent decrease in volume, or vice versa, and it is to me significant
that the effect of price becomes more significant as these analyses
come nearer to using more adequate data representing demand (in
other words bookings is a more adequate measure of demand in a
deniand study than is the volume of production).
In general, it may be said that all of the foregoing analyses give
such unreliable results that the authors themselves discard their show-
ings as to the nattire of elasticity of demand for steel and resort to the
assumption oi vr 'o elasticity as a basis for the further analyses of
costs in relation to volume and of losses in relation to price reductions.
In their conclusions as to the effect of a given price reduction on the
volume of steel, the authors of these studies fail to take into account
the effect that such a price reduction would have on the general
average of price of goods directly and indirectly affected by steel
prices. They also fail to take into account the additional effect of
the increased volume of steel due to a price reduction on business in
general and therefore on steel, and that point Dr. Ezekiel elaborated
yesterday. That there is a positive relation between steel activity is
well known and demonstrated in these demand studies. They have
not demonstrated the close relation that exists between the prices of
iron and steel and the general level of prices of other goods. In the
chart labeled "Indexes of prices of iron and steel and other commodity
prices," this relationship is shown in a general way. While it is not
CONCENTRATION OF ECONOMIC POWER 13731
possible to indicate the effect of prices of iron and steel on other prices
quantitatively — or of other prices on the prices of iron and steel —
that effect nevertheless needs to be taken into account in any analyses
of the relation of lower steel prices to volume and profits or losses.
And the chart that I just referred to, I would like to introduce
into the record also.
Mr, Feller. I offer for the record the chart entitled, "Index of
Wholesale Prices of Iron and Steel and of Other Goods, for the years
1919 to 1938, inclusive."
Acting Chairman O'Connell. That will be admitted.
(The chart referred to was marked "Exhibit No. 2188" and is
included in the appendix on p. 14126.)
Dr. Bean. The source of these data, in the case of iron and steel,
is the Bureau of Labor Index of Iron and Steel Prices, the other price
index is derived from the Bureau of Labor series, the all-commodity
index, by removing from it three groups — the group called farm
products, the group called foods, and the third group, iron and steel;
so that we have here a comparison annually between the fluctuations
in the composite price of iron and steel and the composite of all other
prices, exclusive of iron and steel and exclusive of the highly variable
prices of foods and farm products.
By and large, there is a very close correspondence between these
two price indexes, with two exceptions. One occurs in 1923, when
iron and steel prices rose quite sharply, as contrasted with a fairly
stable average for all other prices. I believe that the series introduced
in the record by the Corporation the other day of mill net yields does
not show quite this rise in steel prices for 1923. The other departure
occurs after 1929, when the price of iron and steel remains at a
relatively higher level than the prices of all other products, excluding
iron and steel and the agricultural products.
But throughout these years, except for the fact of one index being
at a higher level than the other, there is a very close correspondence
in the year-to-year behavior, and I introduce this material for no
other purpose than to suggest that the question of how much the
price of iron and steel affects other prices has not been answered, and
that it is something that ought to be fully looked into.
Mr. Reynders. There is a further divergence from last year, of
course?
Dr. Bean. Correct. In 1938, steel prices held at about the same
level while prices of the other things, other than farm and food,
which went down slightly.
Acting Chairman O'Connell. Any members of the committee have
any questions?
Mr. Wooden. Dr. Bean, you have used the expression "an inverse
relationship between price and volume." I think we all understand
what that means. In other words, as price decreases, the volume
tends to increase. You have also used the expression of a negative
relationship between price and volume. You don't mean the same
thing by those two expressions, do you?
Dr. Bean. Yes, I do. A line that rises to the right is, as a rule,
called a positively inclined line, whereas one that declines to the right
is called a negatively inclined line, and a negatiyely inclined line in
this case is the same thing as showing an inve^e relation "between
price and volume.
13732 CONCENTRATION OP ECONOMIC POWER
Mr. Wooden. Js it your conclusion that there is this inverse rela-
tionship between the price of steel and the volume of steel sold, but
sometimes that relationship is offset or overcome, or you might say
reversed, by the strength of other factors?
Dr. Bean. Yes, and that is true of practically all other prices, even
agricultural prices.
Acting Chairman O'Connell. I believe there are no further ques-
tions at this time. Thank you very much, Dr. Bean.
Mr. Feller. The next witness will be Dr. Yntema, recalled, who,
as I understand, will comment on Dr. Bean's testimony.
TESTIMONY OF PROF. THEODORE OTTE YNTEMA, SCHOOL OF
BUSINESS, UNIVERSITY OF CHICAGO, GHICAGO, ILL.— Resumed
Dr. Yntema. Mr. Chairman, I have just been listening with great
interest, as I always do listen with great interest to any statement by
Dr. Bean. He has been critical of some of the statistical work which
we have done, and I shall be deeply critical of some of the suggestions
which he has made. It should be made clear, however, that the area
of our agreement is undoubtedly far greater than the area of our dis-
agreement. I think that is a statement in which Dr. Bean himself
will probably concur. Tliis is just a matter of keeping a sense of
proportion with respect to the criticisms that are offered.
In evaluating the analyses of demand which we submitted to the
committee, Dr. Bean neglected almost entirely the main line of our
argument, and concentrated his attention on the secondary evidence.
In our studies we approached the problem of determining elasticity of
demand by two general types of analyses. The first of these involves
making a rough estimate of the elasticity of demand for some of the
principal products made from steel, then the calculation of the pro-
portion of steel cost to the price of these products, and finally the
derivation from this evidence of the relative response in the quantity
of steel sold to the price of steel in each of the respective industries
producing these products.
Tnis type of analysis is well known to economists, and as Dr.
deChazeau indicated at one point in his testimony, leads unmistakably
to the conclusion that the elasticity of demand for steel must be very
low.
Dr. Bean has offered very little, if any, criticism of this type of
analysis, which constitutes, I may say, the mainstay of our conclusions.
The second Hne of approach which we employed consisted in study-
ing the relation of the quantity of steel purchased by the industry to
the price of steel and other important variables determining its de-
mand. As we shall indicate, some of his criticisms of this analysis
are not of substantial validity, and at some points we think he has
not quite fairly presented the statements which we have made.
I should like to emphasize that a considerable part of Dr. Bean's
discussion has had to do with the gross or unadjusted relations (as he
has taken care to point out) between changes in the price of steel and
changes in its consumption in the automobile, container and railroad
industries. This material which we presented was relatively imim-
portant and could have been dropped out entirely from the studies
without impairing in any way their validity.
CONCENTRATION OF ECONOMIC POWER 13733
In criticizing this part of our studies, Dr. Bean has failed to under-
stand the argument we were making, and I must say in fairness to him
that we in turn ought to be criticized for faihng to make our point clear.
I tliink that the criticisms in this case are with reference to the lack of
unambiguity on our part and not with reference to the content of the
argument that we were trying to present.
Now, to get down to cases. In determining the demand for steel
consiimed by the automobile industry, we based our conclusions
primarily upon the study of the demand for automobiles made by Roos
and von Szeliski for General Motors Corporation and upon the
proportion of the cost of steel to the retail price of an automobile.
To this major part of our argument, Dr. Bean has offered no objec-
tion whatsoever. Concentrating his attention on the relatively un-
important part of the analysis, he first quoted a portion of a sentence
out of context, thereby distorting its meaning. Referring to chart 6
on page 19 of "Exhibit No. 1413"^ entitled "An Analysis of the
Demand for Steel in the Automobile Industry," he said that we claim
that this scatter diagram showing percentage change in annual con-
sumption and average annual price (which he has just exhibited to
the conmiittee) "fails to indicate that lower steel prices are associated
with greater quantities of steel purchased and vice versa."
He then criticized us for failing to attempt to find from these data
what the relationship between steel consumption and steel prices
would have been if allowance had been made for the other factors.
Let me read the sentence in its entirety:
Since the scatter diagram fails to indicate that lower steel prices are associated
with greater quantities of steel purchased and vice versa, it justifies the view that
price considerations have a minor influence in determining the consumption of
automobile steel.
That language is by no means clear. The idea we meant to convey
was that, in the past, changes in the price of steel have not been the
major determining influence, that they were offset or swamped, I
think, was the phrase that Dr. Bean iiimself used, by the effect of
other influences.
We should be the first to suggest that in the data there probably is
some evidence of a negative relationship, that is, an inverse relation-
ship, between changes in prices and changes in quantity, although
the effect of the change in price upon the quantity bought is very
slight indeed.
Taking these data in the scatter diagram, Dr. Bean has attempted
to derive such a net relationship between price and quantity pur-
chased by the automobile industry through the technique of graphic
multiple correlation analysis, for the invention of which all statisticians
are greatly indebted to him. It is generally recognized that this
technique is one of the most powerful and also one of the most danger-
ous devices for analyzing data. Its application in this case illustrates
both these characteristics.
At one point in his discussion, Dr. Bean uttered a profound truth.
I quote:
It is elementary in the statistical analyses that no reliL,ble conclusions regarding
the effect of one factor or another can be drawn unless the complicating influences
of other factors known to be present have been isolated.
1 Appendix, p. 13993.
13734 OONOENTRATION OF ECONOMIC POWEU
This is a fundamental principle, absolutely essential to the applica-
tion of the teclmique wliich he has used.
I suggest to you that Dr. Bean's analysis fails to reveal the true,
underlying relationship between the price of steel and the consumption
of steel in the automobile industry, because he has himself neglected
important complicating factors which would have been apparent to
him if he had been in position, and had had the opportunity, to give
the problem more extended and intimate study.
If I may revert to the chart, I should like to suggest to jou that
Dr. Bean has either undertaken too little or too much. The mference
that we intended to be drawn from this chart, entitled, "The Relation
of Automobile Steel Price to Automobile Steel Consumption,"
appearing on page 19 of exhibit No. 1413, was simply that from the
scatter of all of these observations showing changes in automobile
steel consumption and changes in automobile steel price, it was not
apparent that steel price was the controlling factor. Changes in
steel prices were not the controlling factor in determining changes in
automobile steel consumption. I think Dr. Bean would be the first
to agree with that.
We merely meant to convey the idea — which we did not convey
satisfactorily, as witness the fact that Dr. Bean misinterpreted us — we
merely meant to convey this idea, that steel price is not the controlling
factor, that there were other very important factors which he has
himself recognized.
The procedure which Dr. Bean suggested is nothing new to us.
We have spent many weary days trying to do what he has tried to
do this morning. The diiference between the procedure which he
employed, however, and the procedure which we employed, is this,
that we think he still failed to take into account many variables which
are extremely important in this problem, and that the inferences which
he draws from this chart consequently are not reliable.
For example. Dr. Bean said that it is possible to consider the rela-
tionship between changes in automobile steel consumption and auto-
mobile prices in the years 1933, 1935, and 1936, and the years preceding
and from them, form some sort of a reasonable inference with respect
to the net effect of changes in automobile steel price, upon automobile
steel consumption. If I overstate the case, I hope that you will
correct me, Dr. Bean.
He says that such a relationship may be inferred because in those
years, the change in industrial production from the year preceding
was of about the same level. Now, let me suggest to vou that the
change in industrial production is by no means the only important
factor in this situation. For example, the change in consumer income
from the year preceding is a far more important factor in changing
the demand for automobile steel than changes in general industrial
production.
A series of deflated consumer incomes, which I confess is none too
satisfactory, exhibited the following characteristics: In 1933, there
was a decrease in deflated consumer income of 3.6 percent; in 1935,
an increase of 3.5 percent; and in 1936, an increase of 13.1 percent.
In the second place, the changes in the purchase of steel by the
automobile industry reflect the mcreases or the decreases of their
inventories. If you want to find the underlying relationship, neglect-
ing changes in mventory, betv^een the percentage change in price
CX)NC?ENTRAT1UN OF ECONOMIC I'OWI^R 13735
and the percentage change in consumption, that factor is an ex-
tremely important one, particularly, I should say, in the year 1933,
when we moved from a period of declining busmess to a period of
advancing business.
In the third place, I should, for your consideration, point out the
fact that Senator Kling so aptly cited a few minutes ago, and that
is whether or not the change in psychological attitude on the part of
those who buy automobiles and those who produce the automobiles,
whether the changes in psychological attitude from the year pre-
ceding, in 1933 as compared with 1932, in 1935 as compared with
1934, and in 1936 as comi)ared with 1935, were in fact the same.
That happens to be a very important factor in the demand for auto-
mobiles and the demand for steel in the production of automobiles,
and I submit to you that those changes were not in fact the same.
In the fourth place, the demand for automobiles and the demand
for the steel in them depend upon the proportion of automobiles in
existence which are about to be scrapped, and the changes in that
situation from the year preceding. I should raise the question as to
whether or not the change in that factor actually remained the same.
I say there is some difference in that factor which ought not to be
neglected.
There is still a further question that ought to be considered, whether
or not the changes in the prices of other related products from 1932
to 1933, from 1934 to 1935, and from 1935 to 1936, were substantially
the same, because those changes in the other prices are related to this
net relationship which you are inclined to find between, automobile
steel consumption and automobile steel prices.
The reason we did not pursue this particular line of attack is that
we did not have confidence in it. As 1 said before, I suggested to you
that Dr. Bean has either tried to do too much or has done too little,
because the only conclusion I would be willing to derive from this
information presented to you this morning is that the price of steel is
not the controlling force in determining the quantity of steel con-
sumption; that it is swamped by other forces. I do not think that
there would be substantial disagreement about that.
If you get into the study of net relationships, you are getting into
exceedingly dangerous territory. We recognized that in the multiple
correlation analysis, and we wrote page after page, pointing out the
qualifications to which any such analysis must be subjected, and we
used extreme care in making plain those qualifications.
In continuing with the discussion of the relation between the price
of steel and its consumption in the automobile industry, Dr. Bean
dealt with the problem of the variation in the steel consumed per
automobile in relation to the price of steel. In attacking that par-
ticular problem, he employed data which we discarded because they
were not applicable to the purpose. He took the reported figures on
steel consumption in the industry and divided it by the number of
cars, to obtain the average steel consumed per car. That neglects
the effect of inventory changes in the hands of the automobile com-
panies. We considered using the material and discarded it for that
particular purpose because the data were not as satisfactory an indica-
tion of the steel per car as is obtained by the use of average weights
per car. Those data appear on page 23 of "Exhibit No. 1413," and
with reference to that particular set of materials, Dr. Bean has offered
13736 CONCENTRATION OF ECONOMIC POWER
no criticism whatever, I suggest to the committee that that is
the appropriate study and the materials which he used arc far loss satis-
factory for that purpose.
In dealing with the study of the demand for steel in the container
industry, Dr. Bean did not go into such great detail and it is therefore
not possible for me to offer such specific criticism of his comments.
There are one or two points, however, that I do not think should be
passed by without some comment.
He said:
If we assume that the price of canned tomatoes is determined by supply, as
measured by the size of the pack and by the price of tin plate, one of the important
cost items, it is possible to observe both the nature of the relation of supply to
price and of the relation of the price of tin plate to the price of a can of tomatoes.
Now, if you are considering the demand side of the situation, and
you have the size of the pack, that is, the quantity of tomatoes packed
to be sold, and that pack is sold, I cannot see, and I do not see, how
anybody else could find that the price of tin plate is a relevant factor
in determining the price of tomatoes. It is the quantity of tomatoes
sold in the packed form that determines the price. The price of tin
plate might be higher or lower but it would have no effect whatever
upon the price of tomatoes. That is simply elementary economics.
Mr. Reynders. In that connection, could you refresh your memory
as to what the fluctuation in the price of tin plate, what effect that
would have on the price of a can of tomatoes?
Dr. Yntema. The evidence on that
Mr. Reynders (interposing). I am speaking now of the fluctuation
in price, not the price itself.
Dr. Yntema. The fluctuations in the price of tin plate were rela-
tively small over this period, ranging from $5.50 to $4.43 per base box,
a change of something in the neighborhood of 20 to 25 percent. The
proportion of the cost of tin plate to the price of the goods «old at retail
is about 10 percent. That would mean that the effect upon the final
price of canned goods would be approximately 10 percent of the 25
percent, or in the neighborhood of 2 or 3 percent. That would be the
effect which you would expect if it is passed on entirely to the consumer
of the product.
May I revert again to one of Dr. Bean's charts? Dr. Bean pre-
sented this chart showing the price of tin plate and of canned goods,
numbered "Exhibit No. 2186." This represents the price of ^tin plate
from 1923 through 1938 and also the price of canned tomatoes, peas,
and corn. I cannot quote Dr. Bean precisely in his comments on the
chart, but if I understood him correctly, he conveyed the idea that
thi; chart suggests that these fluctuations in the price of tin plate
have been of some substantial importance in affecting the price of
canned tomatoes, peas, and corn.
Now, Dr. Bean has pointed out at some length the importance of
keeping constant the effect of other important variables, and it seems
to me that it is not possible at all to draw any reliable conclusion from
these two series unless we do exactly what ne has told us we should
do namely, allow for the effecj: of changes in the general price level
and the national income, which also showed somewhat similar fluctua-
tions during this period.
Acting Chairman O'Connell. If they did show somewhat similar
variations, would it change the character of the relationship?
CONCENTRATION OF ECONOMIC POWER 13737
Dr. Yntem\. It would make the answer indeterminable. I don't
say that there is nothing whatsoever to Dr. Bean's suggestion; I
simply say that it is incorrect statistical procedure, to infer, from
the evidence which has been presented, that fact. If you did have the
other evidence, the three series would move so much alike that the
precision of any inference you could make would be far less than might
appear upon a supei-ficial examination of these data.
Dr. Bean said:
In the studies on the demand for steel in the container, railroad and automobile
industries, the relative unimportance of price is deduced from a set of scatter
diagrams showing the relation between the year-to-year percentage changes in
price of the appropriate kind of steel, and percentage changes in the factor which
represents the use of steel b^- the particular industry.
Merely again to obtain the proper perspective, that was decidedly
secondary evidence, and w^e did not deduce the relative importance
from that evidence; we merely used that material to point out that
other factors as well as price were of controlling importance in deter-
mining the quantity of steel bought by the various industries. Our
analyses would not be one whit less convincing if that material were
eliminated entirely from them. In fact, some criticis have told me
that they tliink that our presentation would be improved by that
omission. I think that is a fairly debatable point.
In general, then, summing up Dr. Bean's criticism of the use of
these scatter diagrams, I suggest, first, that we have failed to make
ourselves clear on them, that the point wliich we intended to convey
is a perfectly proper point, but that we may perhaps have used not
quite entirely unambiguous language, and if so, that fault is ours and
we appreciate his calling it to our attention.
In the second place, we should deny that it is so easy to ascertain
from inspection of these charts the sort of negative relationship which
he -has pointed out. The problem is far more complicated than he
has made it. I suggest once more that either we must do more than
he has indicated or otller^vise, simply make the rough inference which
we did make.
Finally, Dr. Bean offered some suggestions with reference to the
analysis of the demand for steel by multiple correlation methods,
and I should like to respond to them.
This net chart of Dr. Bean's is entitled, "The Nqt Regression of
Volume on Price," and is numbered "Exhibit No. 218f."
The question might be raised, "Why did we incorporate in
"Exhibit No. 1411" ^ a study showing a net relationship that as price
goes up the average production and consumption of steel by all
consuming industries increases."
The same question might be raised with respect to Relationship
II. Now, we did not intend to convey to the committee or to anyone
reading this report, that is "Exhibit No. 1411" (and I think that a
simple reading of our text will indicate as much) that we thought
that these relations, i. e., Relationships I and II in this exhibit,
represented actually how consumption and production of steel would
vary with price. The fact that we did present them ought to be
taken as evidence of good faith, to show how difficult it is to find the
true relationship by this method.
' Appendix, p. 13913.
13738 OONOKNTIIATION OF ECONOMIC POWER
Frankly, I should say that in Relationship I, as we have pointed
out in the document entitled "A Statistical Analysis of the Demand
for Steel, 1919-38," "Exhibit No. 1411", we think that we have not
taken entirely satisfactory variables into account.
In that particular study we investigated the relationship between
the fluctuation of production of steel ingots and castings and the price
of steel and industrial production and a time trend. The inference
to be drawn from this study is that price of steel, industrial produc-
tion, and time trend do not account with entire satisfaction for the
production of steel ingots and castings. It is a cardinal principle in
statistics that when your results don't make sense you discard them.
In this particular case I should say that the results do not conform
entirely with expectations and that the analysis must therefore be
carried further,
Mr. Lewis.' In both of these relationships we very carefully
pointed out in "Exhibit No. 1411" that the basic data on which they
were computed were inadequate for drawing any conclusion from
them, and we ourselves did not draw any conclusions from them. In
several pages we pointed out that steel ingot production was not a
satisfactory measure of steel sales, nor was industrial production, on
the other hand, a satisfactory measure of the shifts in demand
assuming price constant.
Dr. Yntema. From our own i-eport, "Exhibit No. 1411" therefore,
you can read without difficulty that we did not attach great significance
to this finding. It would have made a far better appearing report if
we had simply omitted tliis entirely (referring to Relationships I and II)
and present ad just one result. We elected, however, to show to the
committee what was involved in this sort of study and to give some
idea of the qualifications which must be kept in mind in interpreting it.
The point that Dr. Bean has raised with respect to Relationship II
in this "Exhibit No. 2187" ^ is certainly correct, that the slope of this
line representing how production fluctuates in relation to the composite
price of finished steel is determined largely by this one observation for
1920, and no statistician would therefore attach great weight to it.
We did not attacji any great significance either to this particular
finding.
In Relationship III we find that the average behavior of shipments
of steel in relation to price is such that as price goes up shipments go
down. Dr. Bean has suggested that if we eliminate two observations
1937 and 1923, from this Relationship III, that the average line of
relationship would be steeper. Well, that of course is no basis for
statistical procedure. Merely because two points fail to conform to
the others is no basis for eliminating those points. If there were a sound
logical reason for eliminating those two particular points, then their
elimination ought to be considered, but it is one of the worst types of
statistical manipulation to omit ]^oints merely because they happen
to fail to conform to the rest of the observations.
May I suggest this, please. I am not accusing Dr. Bean of poor
statistical technique. Dr. Bean is one of the best statisticians whose
acquaintance it is my pleasure to have made. I am merely suggesting
that Dr. Bean's point is a good one that you ought to investigate
whether there is some characteristic which those points have, and I
' Harold Gregg Lewis, economist, University of Chicago. See supra, p. 13650.
» Appendix, p. 14125.
CONC?ENTRATION OF ECONOMIC TOWER 13739
think that that should be taken mto account in going further with the
analysis.
Mr. Lewis. May I interrupt? In the first place, Dr. Bean has said
that if we actually eliminated these two observations we should get a
net relationship between price and volume, which would be approxi-
mately three times as large as the one we actually obtained. Of course
even if we did that, our result would still be significantly less than unit
elasticity. Even if we eliminated these points, the final conclusion we
would get would be that changes in steel price do not lead to larger
proportional changes in steel volume. Actually, since this report '
was written — unfortunately Dr. Bean has not had an opportunity to
study the data in the same way that we have — I have attempted to
account for the fact that these observations, 1937, 1923, 1924, 1938,
1933, are extreme observations. I have found that if we include addi-
tional factors such as the rate of change in the price of steel, that is,
essentially, if we study speculative buying of steel, that these points
lie very close to this line. Not only is that true, but our results lead
us to the conclusion that actually the elasticity we should so obtain
would be lower than the elasticities which were obtained in this report.
Dr. Bean's major comment on this last relationship in which steel
bookings are related to composite price of steel, industrial profits, and
consumer income, is that first we have probably used an inadequate
measure of price. I do not quite understand what criticism he has of
that price. However, may I point out that since this computation
was made, we have actually used the index of mill yet yield, which is a
very carefully constructed mill net index, and we have found if we sub-
stituted that index for the price we used in the report, entitled "A
Statistical. Analysis of the Demand for Steel, 1919-38", "Exhibit
No. 1411," that our results would not be changed significantly.
Secondly, he has suggested in connection with this last relationship
that if we lag price or consumption, we should probably get a some-
what different result. I should agree with that, but in order to dem-
onstrate the reasonableness of such a lag, he should have to give some
underlying logic whereby "that lag could be given some definite
economic meaning.
Dr. Yntema, May I offer in conclusion just a few remarks? The
area of agreement is far greater than that of disagreement, and Dr.
Bean and I, I am sure, do not have substantially different views in
these matters, and I want to emphasize this point, that part of the
difficulty has arisen due to our own fault in not making perfectly
clear what we intended to show.
There is one point to which Dr. Bean reverted in concluding his
testirnony with respect to our alleged failure to take into accouht the
additional effect of the inpreased volume of steely due to a price re-
duction, on business in general, and, therefore, on steel. This is the
same point that Dr. Ezekiel made yesterday, and smce we are to
discuss Dr. Ezekiel's^ statement tomorrow, I should like to defer
comment on this particular point until that time.
Acting Chairman O'Connell. I would like to recess now. Wha,t
is your pleasu^^e?
Mr; Feller. Dr. Bean wanted to make one observation.
Dr. Bean. Just two comments. I do not believe I was unfair in
the' quotation from "E.xhibit No. 14.13" that I cited in respect to
■ "Exhibit No. Hir'.'appendix, p. 1.391J?.
124491— 41— pt. 26 11
13740 OONOENTRATION GF ECONOMIC POWER
chart 6 of that exhibit.^ If any of you will read it, leave off the word
"since," and the last phrase in that sentence, and you will get the
gist of what the authors apparently had in mjjid. I tried to find
their thouglit. The only place where I could find it was in that par-
ticular sentence to which they have added the word "since," and I
don't think the word "since" alters the meaning of that sentence.
The second point, as to whether or not this material has been pre-
sented to the committee in such a way as to leave with the committee a
full understanding of the complications, the inadequacies and all the
rest. It is my impression that the reports do not quite do that job, and
I turn for the moment to page 28 ^ of the "Analysis of the Demand for
Steel" ("Exhibit No. 1411") reading next to the last paragraph on that
page: "In the graphical analyses that were made of the various de-
mand relations, there were clear indications that if the lags of ship-
ments and industrial profits behind bookings were removed, Relations
III and IV would both give about the same results for the elasticity of
demand, yielding a figure of 0.3 to 0.4." It is the next sentence I
really have in mind. "The evidence and argument adduced in the
preceding pages of this paper support the conclusion that such a
value — or one even lower — for the elasticity of demand for steel is
not a statistical happenstance, but a reality."
I would underscore the word "reality," It is my impression that
this particular paragraph which will be generally read — if anything
is read in these reports it will be paragraphs of that sort and not the
equations and not the qualifications — gives much greater importance
to the analyses contained in these reports than I think the analyses
are entitled to. Finally, if these analyses are not entitled to be taken
with a^reat deal of reliability, then I would say to Dr. Yntema that
they should not have been included with the implication that they
corroborate his basic conclusions.-
Dr. Yntema. May I comment? I think we aU owe a debt of
gratitude to Dr. Bean for bringing out more clearly in the discussion
considerations which ought to be kept in mind in evaluating the
evidence. I should like to call his attention to a word in the sen-
tence which he read. This sentence begins, "The evidence and argu-
ment"— If I had used as the subject of that sentence "statistical
analyses", without any logical argument accompanying it, I should
never have stated that conclusion. The reasort why we concluded
that the low elasticity demand for steel is not a statistical happen-
stance but a reality is that it "made sense."
Mr. Feller. Mr. Chairman, before we adjourn, and I suggest that
we adjourn until 10:30 tomorrow morning, I should like to express on
behalf of Dr. Kreps his profound gratitude to the Steel Corporation,
to Dr. Yntema, to his staff, and to Drs. deChazeau, Taitel, Ezekiel,
and Bean for their work in bringing out this important material. On
behalf of myself I should like to make just one observation. Ever
since I heard of the science of econometrics, I have been very skeptical
of it. After Ustenmg to the most eminent practitioners of tliat science
or art, my skepticism has not been substantially lessened. Mr.
Justice Holmes once said that a page of history is worth a volume of
logic, and after listemng to the discussion here, I wondered whether
• Appendixp. 13993.
> Of the original document.
CONCENTRATION OF ECONOMIC POWER 13741
it cpuldn't be paraphrased by saying a page of business facts is worth
a vokime of econometrics.
Acting Chairman O'Connell. I think that the committee has the
same view that you do as to paying our tribute to Dr. Yntema and
his staff for the work they have done, and on behalf of the committee,
I also want to thank Dr' deChezeau, Dr. Ezekiel, Dr. Bean, and Mr.
Taitol for the work they did in helping to present this rather com-
plicated picture to the committee.
We will recess now until 10:30 tomorrow mormng.
(Whereupon at 12:25 p. m., the committee recessed until 10:30
a. m., Friday, January 26, 1940.)
(Testimony on the Iron and Steel Industry is resumed and con-
cluded in Hearings, Part 27.)
APPENDIX
The following exhibits, Nos. 1409 to 1418, were introduced and
ordered to be placed on file with the committee during Hearings,
Part 20. They were subsequent!}' ordered to be printed and are
reproduced herewith, with the exception of "Exhibit No. 1418," which
is included in Hearings, Part 27.
Exhibit No. 1409
Table of Contents
section a financial
Title Of Chart
Assets, From 1901
Assets, 1938 Compared with 1901
Capital, Surplus and Liabilities
Earnings and Cash Dividends
Earnings and Cash Dividends per Share of Common Stock
Ratio of Earnings to Net Assets:
From 1920
From 1902
Distribution of the Sales Dollar, 1929-1938, inclusive
Distribution of the Sales Dollar, Year 1938
Payroll and Earnings per Dollar of Sales
Payments to Employees and to Investors per Dollar of Sales
Taxes per Dollar of Sales
Total Taxes and Earnings Available for Dividends
Assets, Earnings and Taxes, 1937-1938 Compared with 1902
State and Local Taxes
Taxes (State and Local, Federal, and Social Security)
Taxes Paid in 1937 and 1938
Average Assets, Annual Sales and Earnings, 1929-1938, inclusive
Ratio of Sales to Total Assets (U. S. Steel and Other Companies), year 1938
Average Monthly Prices of Common and Preferred Stocks
SECTION B COSTS
Relationship between Total Costs of Operation and Volume of Business, 1938
Conditions
Composition of Total Costs of Operation in Relation to Volume of Business
Relationship between Sales and Costs — Effect of Reduction from Average 1938
Prices'
Increases in Volume Needed to Compensate for Various Decreases in 1938
Prices Compared to Probable Resulting Increases in Volume
Estimated Additions to 1938 Deficit — How Deficit Would have Increased If
Prices Had Been Reduced
Relationship between Sales and Costs — Effect of Reduction from 2nd Half 1938
Pirices
Increases n Volume Needed to Compensate for Various Decreases in 2nd Half
1938 Prices Compared to Probable Resulting Increases in Volume
Unadjusted Costs and Volume of Business Compared with Estimated Costs for,
Corresponding Volumes under 1938 Conditions
How Costs Have Increased, 1938 Compared with 1931
How Costs Have Increased, 1937 Compared with 1929-1930 Average
13743
13744 CONCENTRATION OF ECONOMIC POWER
SECTION C PRICES
Title of Chart
Composite Mill Net Yield and CosJ, per Weighted Ton Shipped (1926= 100)
Average Yearly Base Prices of Principal Steel Products (1924 = 100)
Reported Base Price and Mill Net Yield:
Heavy Structural Shapes at Pittsburgh
Heavy Structural Shapes at Chicago
Plates at Pittsburgh
Bars at Pittsburgh
Bars at Chicago
Standard Pipe at Pittsburgh
Cold Rolled Sheets
Reported Composite Price and Composite Mill Net Yield (1926= 100) :
From 1926
From 1912
Proportion of Steel Cost in Price of Finished Product, 1938
Automobile Steel Consumption and Steel Prices in United States
Automobile Steel Consumption and Automobile Production in United States
The Basing Point Method of Quoting Delivered Prices:
Diagram 1
Diagram 2
Diagram 3
Diagram 4
Diagram 5
Diagram 6
Diagram 7
Diagram 8
Diagram 9
Diagram 10
Diagram 11
The Uniform F. O. B. Mill Price Systeip:
Market Territories of Major Mills Producing Steel Sheets
Detailed Map of Counties in E. Ohio and W. Pennsylvania
Explanation of Unadjusted and Adjusted Freight Absorption
Average Delivered Price and Freight Absorption (U. S. Steel 'rporation Sub
sidiaries), February, 1939
Average Delivered Price and Freight Absorption (Selected Producing Companies),
February, 1939
Breakdown of Average Delivered Price (U. S. Steel Corporation Subsidiaries),
February, 1939
Breakdown of Average Delivered Price (Selected Producing Companies), Feb-
ruary, 1939
Relation of Mill Net Yield and Reported Base Price, February, 1939
SECTION D CAPACITY AND PRODUCTION
Total Ingot Capacity (U. S. Steel Corporation Subsidiaries and Other Steel Pro-
ducing Companies)
Steel Ingot Capacity Compared with Population (U. S. Steel Corporation Sub-
sidiaries and Total United States)
Total Ingot Production (U. S. Steel Corporation Subsidiaries and Other Steel
Producing Companies)
Steel Ingot Production Compared with Population (Total United States)
Ingot Capacity and Production (U. S. Steel Corporation Subsidiaries)
Ingot Capacity and Production (Total United States)
Per Cent of Ingot Capacity Operated (U. S. Steel Corporation Subsidiaries)
Steel Production and Manufacturing Production (Federal Reserve Indexes,
1923-1925=100)
World Ingot Production (By Principal Steel Producing Countries), 1929, 1937
and 1938
CONOENTKATION OF ECONOMIC POWER
13745
SECTION E LABOR
Title of Chart
Number of Employees and Ingot Production (1929=100)
Actual Number of Employees and Number That Would Have Been Required on
Basis of 1929 Hours per Week
Ingot Production and Number of Employees (U. S. Steel Corporation and Sub-
sidiaries, 1929=100), 1937 Compared with 1929
Ingot Production and Numl^er of Employees (Total Steel Industry, 1929=100),
1937 Compared with 1929
Employment and Payroll by Classes of Employees, 1937-1938 Average
Employees by Age Groups, May 1, 1938
Skilled* Semi-skilled and Common Labor Employees, Year 1938
Payroll and Component Factors (1929=100)
Wages and Hours (U. S. Steel Corporation Manufacturing Subsidiaries)
Wages and Hours (U. S. Steel Corporation Manufacturing Subsidiaries and All
Manufacturing Industries)
Average Weekly Earnings Compared with Cost of Living (1929=100)
Wage Rates and Steel Prices (1914=100)
Average Earnings per Hour and Common Labor Rate
"Earnings per Hour and Steel Prices (1926=100)
Earnings per Hour and Production, April, 193V -November ,1939, inclusive
SECTION F — MISCELLANEOUS
Flow Chart of Steelmakinj
Section A — Financial
Assets — U. S. Steel Corporation and subsidiaries
Fixed Assets
(Land, Buildings,
Equipment, Etc.)
Current Assets
(Cash, Invento-
ries, Etc.)
1901
1902.
1903.
1904
1905
1906
1907.
1908
1909.
1910.
1911.
1912.
1913.
1914.
1915.
1916.
1917.
1918.
1919.
1920.
1921.
1922.
1923.
1924.
1925.
1926.
1927-
1928.
1929.
1930.
1931.
1932.
1933.
1934.
1935.
1936.
1937.
1938.
1,291,883,
., 300, 521,
,328,132,
, 336, 882,
,330,097,
, 320, 077,
, 372, 972,
, 393, 522,
, 401, 507,
, 430, 212,
, 460, 303,
, 448, 175,
, 465, 498,
, 457, 853,
, 443, 300,
,472,623,
, 521, 836,
, 563, 937,
,573,661,
, 606, 758,
, 644, 795,
,631.579,
,639,158,
, 078, 208,
, 692, 197,
, 667, 391,
, 709, 779,
,661,123,
, 541, 492,
, 677, 327,
, 683, 982,
, 650. 816,
, 653, 923,
, 626, 143,
, 338, 522,
, 350, 037,
,410.432,
, 166, 519,
$190, 644, 992
211,276,652
216, 079, 027
207. 929, 556
242, 355. 569
279, 237, 763
284,366,811
243. 539, 140
290, 920, 288
281, 806. 984
278, 984, 551
327, 324. 854
335. 087, 691
334. 379, 563
405, 241, 095
610, 404, 307
927,713.414
007, 680, 052
792, 220, 835
823,788,417
694, 310, 235
709, 074, 010
781.724,062
735. 986, 097
753, 445, 627
786, 747, 687
723, 803, 437
780, 906, 264
744,691.068
717,217.277
595, 820. 720
507,915,913
448, 973, 131
457, 968, 505
483. 878, 883
513, 939. 237
508. 296, 375
544, 759, 493
$1, 482, 528, 825
1,511,798,510
1,544,211,164
1,544,812,404
1,572,453,335
1, 599, 315, 431
1,657,339,629
1, 637, 061, 469
1,692,428,137
1,712,019.845
1, 739, 288, 534
1.775.500.109
1,800,586,323
1.792,233.493
1. 848. 541, 861
2. 083. 027, 974
2, 449, 650. 206
2.571.617.175
2. 365, 882. 382
2. 430. 646, 963
2. 339. 105. 310
2. 340. 653. 216
2. 420. 882, 704
2,414, 194, €66
2, 446, 643, 331
2, 454, 139, 185
2. 433, 583. 169
2, 442. 030. 233
2. 286, 183, 656
2.394,644,611
2. 279. 802, 813
2. 168. 732, 222
2, 102, 896. 880
2. 084, 112, 287
1,822.401,742
1.863,976,619
1. 918, 729, 289
1,711,279,006
Data are as shown on books at the end of each year. Data for 1901 are .partially estimated.
Fixed assets include good will and other intangible items, as well as tangible property.
All property values are "net," after the deduction cf rererves for Jepletien, depreciati'yi.,etc.
Current assetsiQcJade a relatively^mall amount of other a-s.'^te, e. g , mining royaltiis, deferred charges,
to. Intel -company profit in inventories-has been eliminated froW ctnrent assets.
13746 CONCENTRATION OF ECONOMIC POWER
Assets — 1938 compared with 1901 — U. S. Steel Corporation and subsidiaries
1901 (April 1st)
1938 (December
31st)
Land, Buildings, Equipment, etc.
Intangibles
Current Assets
$545,500,0C0 $1,166,519,512
749, 207, 806 1
.'j4, 694, 676 544,759,493
$1,489,402,482 $1,711,279,006
Allocation of fixed assets in 1901, as between land, buildings, equipment, etc., and intangibles, is that of
U. S. Bureau of Corporations made in 1911.
ASSETS
U. S. STEEL CORPORATION AND SUBSIDIARIES
While fixed assets of U. S. Steel Corporation ce now carried on the books
at less than the value of fixed assets at the time of the organization in 1 90 1 ,
the Corporation's physical plant is much greater today. This largely results
from two causes:
(1) The elimination from time to time of all intangible values. When the
Corporation was formed, various going businesses were acquired at prices
in excess of the value of their tangible property, resulting in intangibly
assets of about $750,CX)0,000 (as later determined by the U. S. Bureau of
Corpo.'atlons), representing the good-will or earning power of these busi-
nesses. While originally of real value, it has been deemed prudent to write-
down from time to time the value of all such intangible items, gogd-wlll now
being valued at $1.00.
(2) The reserve for obsolescence and depreciation was Increased to the
extent of $270,000,000 in 1935, principally because of improvements in
manufacturing methods v/hich made existing facilities of older design less
valuable.
CONCENTRATION OF ECONOMIC POWER
13747
7\SSETS - 1938 COMPARED WITH 1901
U. S. STEEL CORPORATION AND SUBSIDIARIES
MILLIONS OF DOLLARS
CURRENT ASSETS
1,600
1,200
1
WND, BUILDINGS,
EQUIPMENT, ETC
1901
(Apr. 1st)
ASSETS OF THE CORPORATION ARE CONSERVATIVELY VALUED
1. Values of property are based on reports of
Departments of U. S. Government
2. Property was certified to be conservatively
valued by independent engineers as of Dec. 31, 1937
3. Intangible values have been written down to $ 1
The amount of intangibles originally included in the property account of
U. S. Steel Corporation at its organization on April 1 , 1 901 represented the
good will or earning power of the various going businesses, which were con-
solidated to form the Corporation, over and above the value of their
tangible assets. This amount has been written off from time to time and
was finally reduced to $1.00 in 1938.
Of the total of nearly $750,000,000 of intangibles, $182,000,000 was
written off against current earnings, $285,000,000 against capital surplus,
and the remainder against earned surplus.
13748 CONCENTRATION OF ECONOMIC POWER
Capital, surplus and liabilities — U. S. Steel Corporation and subsidictries
Year
Preferred
Stock
Common
Stock
Surplus and
Reserves
Funded Debt
Current Li-
abilities
Total
$510, 205, 743
510,281,100
360, 281,' 100
360, 281, 100
360, 281, 100
360, 281, 100
360, 281, 100
360,281,100
360.281,100
360,281,100
360, 281, 100
360,281,100
360, 281, ICO
360, 281, 100
360,281,100
360,281,100
360. 281, 100
360,281,100
360,281,100
360, 281, 100
360, 281, 100
360,281,100
360, 281, 100
360, 281, 100
360, 281, 100
360, 281, 100
360, 281, 100
360, 281, 100
360, 281, 108
360, 281, 100
360. 281, 100
360,281,100
360,281,100
360,281,100
360, 281, 100
360, 281, 100
360, 281, 100
360, 281, 100
$508, 227, 394
.508, 302, 500
508, 302, 500
508, 302, ,500
508, 302, 500
,508, 302, 500
,508. 302, 500
,508, 302, 500
508, 302, 500
508, 302, 500
,508, 302, 500
.508, 302, 500
508, 302, 500
508, 3t .2, 500
508, 302, 500
508, 302, 500
508, 302, 500
508, 302, 500
508, 302, 500
508, 302, 500
508, 302, 500
508,302.500
508, 302, 500
508, 302, 500
508, 302, 500
.508, 30?, 500
711, 623, 500
711,623,500
813,284,000
868, 743, 500
870, 325, 200
870, 325, 200
870, 325, 200
870.325,200
870, 325, 200
870, 325, 200
870, 325, 200
652, 743, 900
$29, 550, 140
■ 72,827,866
61,956,410
60, 048, 347
89, 595, 628
120,648,046
137,351,182
127,412,522
152,932,904
190,531.446
196, 817, 873
201,966,918
235, 872, 934
220, 679, 690
271,071,558
491,698.022
617,577,414
687, 607, 2,53
738,017,793
818,406,123
813,758.392
801,660,015
8.56,910,754
882,163,4.30
922, 177, 402
944, 859, 235
752, 226. 629
777,180,077
848,811,007
925, 487, 442
864, 810, 665
764, 707, 681
706, 870, 895
679, 719, 257
408,023,691
411,969,078
445, 083, 957
370, 143, 288
$380, 365, 8.30
370. 560. 792
574, 130, 514
576, 342, 203
573, 506, 549
566,411,776
606,341,022
598, 033, 493
609, 766, 907
600,070,012
622,251,002
644, .5.38, 723
637, 552, 728
661,102,374
644, 0,52, 373
629,803,916
623, Oo7, 609
617,644,840
602, 209, 725
586,812,045
572,514,762
571,756,017
557, 985, 323
540,488,518
537, 964, 166
519, 574, 424
500, 529, 307
480, 429, 556
134,788,423
123, 054, 594
119,063,247
114,921,210
110,398,829
117,496,363
114,240,603
117,843,431
125,707,961
248, 849, 388
$.54, 179, 718
49,826,2.52
39, 540, 640
39, 838, 2.54
40, 767, 5.58
43, 672, 009
45, 063, 825
43.031,854
61,144,726
52,834,787
51,636,059
60,410,868
58,577,061
41,867,829
64, 8,34, 330
92, 942, 436
340,351,583
397,781,482
157,071,264
156, 745, 195
84, 248, 556
98, 753, 584
137,403,027
122,9.59,118
116,918,163
121,121,926
108,922,633
112,516,000
129,019,125
116,977,975
6.5, 322, 601
48, 497, 031
55,020,856
56, 290, 367
69,531,148
103, 557, 710
117,331,071
79, 261, 330
$1, 482, 528. 825
1902.
1903
1,. 511, 798, 510
1,544,211.164
1904
1 544 812 404
1905
1.572,453,335
1906
1,599,315,431
1907
1,657,339,629
1908
1909
1,6.37,061,469
1,692.428,137
1,739,288,534
1,775,500,109
1,800,586,323
1,792,233,493
1915
1,848,541,861
1916
2,083,027,974
1917
2, 449, 550, 206
1918
2,571,617,175
1919
2, 365, 882, 382
1920
2, 430, 546, 963
1921
2, 339, 105, 310
2, 340, 653, 216
2, 420, 882, 704
2,414,194,666
2, 445, 643, 331
1926 — .
1927
1928
2, 454, 139, 185
2, 433, 583, 169
2, 442, 030, 233
1929
2, 286, 183, 655
1930
1931
2,394,544,611
2,279,802,813
1932
2, 158, 732. 222
2,102,896,880
2,084.112,287
1, 822, 401, 742
1936..
1, 863, 976, 519
1937
1,918,729,289
1,711,279,006
1938
Data are as shown on books at the end of each year. Data for 1901 are partially estimated.
Premiums on common stock sold are included with surplus.
Surplus is exclusive of inter-Company profit in inventories.
Reserves do not include depreciation, depletion and amortization reserves, which are applied to the credit
of gross property investment.
Purchase money obligations and minority interest are included with funded debt.
CONCENTRATION OF ECONOMIC POWER
13749
CAPITAL, SURPLUS AND LIABILITIES
U. S. STEEL CORPORATION AND SUBSIDIARIES
THE CORPORATION HAS A RELATIVELY SMALL INDEBTEDNESS
U. S. Steel Corporation has a sound financial structure, with a relatively
small amount of liabilities and a comparatively large amount of surplus
and reserves, and capital stock. Present capitalization is represented
entirely by tangible assets.
A large part of the funded debt was retired In 1929, resulting in substantial
reduction In fixed charges further fortifying the Corporation against
business depressions.
The decrease in surplus and reserves since 1930 reflects depression period
losses, the increase in reserve for obsolescence and depreciation in 1935,
and the write-down to $ 1 .00 of the intangibles remaining In 1938.
The Increase In common stock In 1927 was the result of a 40% stock divi-
dend, that In 1929 was due to the sale of additional common stock for
cash, and that in 1930 was in connection with property acquisitions, viz..
Atlas Portland Cement Company, Oil Well Supply Company and Colum-
bia Steel Corporation; the decrease in 1938 was due to the change in the
common stock from $ 1 00 par value to no par value, with a stated capital of
$75 per share.
13750 CONCENTRATION OF ECONOMIC POWER
Earnings and cash dividends — U. S. Steel Corporation and subsidiaries
Year
Earnings
(After In-
terest)
Earnings
(After Com-
mon and
Preferred
Dividends
Preferred
Dividends
Common
Dividends
Total Cash
Dividends
$61,807,993
90,306,525
5.5,416,653
30, 267, 529
68, 585, 492
98. 128, 587
114,565,564
45, 728, 714
79, 073, 695
87, 407, 185
55, 300, 297
.54, 240, 049
81, 216, 986
23, 496, 768
75,833,832
271,531,731
224, 219, 564
125,317,377
76. 794, 582
109, 694, 228
36, 617, 017
39, 653, 455
1(8,707,065
85, 067, 192
90, 602, 653
116,667,405
57. 896, 836
114,173,775
197, ,592, 060
;04.421,571
13,038,142
' 71, 175, 705
' 36, 501, 123
■ 21, 667, 780
1, 146, 708
50, 583, 356
94. 944, 358
' 7, 717, 454
$19,828,824
34,253,657
12,304,917
5,047,852
43,365,815
62, 742, 860
69,179.837
10, 342, 987
33,521,918
36, 772, 383
4, 665, 495
3, 605, 247
30, 582, 184
1 16, 971, 984
44, 260, 374
201,835,585
107,505,437
28, 935, 350
26, 159, 780
59,059,426
' 14,017,785
' 10, 981, 347
54, 259, 994
24, 266, 340
29,801,801
55, 866, 553
12,893,514
39, 140, 453
108, 523, 343
18,836,097
' 49, 165, 485
'91,891,868
' 43, 706, 745
■ 28, 873. 402
1 6, 058, 914
144,002
27, 695, 427
1 32, 937, 131
$26, 752, 539
35, 720, 178
30, 404, 173
25, 219, 677
25, 219, 677
25, 219, 677
25,219,677
25, 219, 677
25, 219, 677
2,5, 219, 677
25, 219, 677
25, 219, 677
25, 219, 677
25, 219. 677
25, 219, 677
25, 219, 677
25, 219, 677
25, 219, 677
25, 219, 677
25, 219, 677
25, 219. 677
25, 219, 677
25, 219, 677
25, 219, 677
25, 219, 677
25, 219, 677
25, 219, 677
26, 219, 677
25, 219, 677
25,2il9,677
25, 219, 677
20, 716, 163
7, 205, 622
7, 205, 622
7, 205, 622
50. 439. 354
58, 545, 679
25, 219, 677
$15, 226, 630
20, 332, 690
12, 707, 663
$41, 979, 169
56,052,868
43,111,736
1902 -
1903
1904. .
1905 -- -
1906
10, 166, 050
10, 166, 050
10,166,050
20, 332, 100
25,415,125
25, 415, 125
25, 415, 125
25, 415, 125
15, 249, 075
6, 353, 781
44, 476, 469
91, 494, 450
71, 162, 350
25, 415, 125
25, 415, 125
25,415,125
25,415,125
29, 227, 334
35, 581, 175
35,681,175
35, 581, 175
49, 813, 645
49, 813, 645
63, 849, 040
60, 365, 797
36, 983, 950
35, 385, 727
35 385 727
1907
1908
35 385 727
1909
45 651 777
1910
50 634 802
1912 -
50, 634, 802
1917...
1918
1919
1920
50 634 802
1921
50 634 802
1922
50 634 802
1923
54 447 071
1924
60,800 852
1925
60, 800 852
1926
60,800 852
1927
1928
1929
1930 -.-
1931
1932
89, 068, 717
85, 585, 474
62. 203, 627
20, 716, 163
1933
7 205 622
1934
7 205 622
1935
7,205 622
1936
50,439,354
1937.-.-
8, 703, 252
67, 248, 931
Earnings are after all charges, Including interest, bond premium and discount, all taxes, and additions to
bond sinking funds which were later applied to amortization of intangibles.
Earnings would be slightly lower if the special addition to depreciation reserve of $270,000,000 in 1935
could be accurately apportioned over prior years.
CONCENTRATION OF ECONOMIC POWER
13751
EARNINGS AND CASH DIVIDENDS
UNITED STATES STEEL [1 CORPORATION AND SUBSIDIARIES
Cn 0> CT> Ci
1 \ COMMON
1 Idivdenos
t t ft +-'l
A
PREFERRED \ .. / \
>VDVIDENDS / i /••■•
\
' ^ ,>i__i«i^._::
\
-'^ / ^ *
:l7t::::i:;:;:::::::::::
r-^A
a^ CT> CTi CT^
•5j-u3ooo<Nj«e-tDooo
U. S. Steel Corporation earnings and cash dividends on the common stock
have fluctuated greatly since 1901.
Since 1930, earnings have been insufficient to cover preferied dividend
requirements in all but two years. Less than the full amount of the preferred
dividend was paid in each of the years 1932-1935. inclusive, and the
accumulated arrearages were paid off from the earnings of 1936 and 1937
The common stock has received no dividend since 1931, with the exception
of $1.00 per share paid in 1937.
13752 CONCENTRATION OF ECONOMIC POWER
Earnings and cash dividends per share of common slock — United Slates Steel
Corporation
Year
Earnings
per Share
of Common
Stock
Cash Divi-
dends per
Share of
Common
Stock
Year
Earnings
per Share
of Common
Stock
Cash Divi-
dends per
Share of
Common
Stock
1901
$6.90
10.74
4.92
.99
8.53
14.34
15.61
4.03
10.59
12.23
5.92
5.71
11.02
1.34
9.96
48.46
39.15
19.69
10.15
$3.00
4.00
2.50
1920 ... -
$16. 62
2.24
2.84
16.42
11.77
12,86
17.99
10.28
12.50
22.61
9.42
1 1.40
1 11.08
1 7.09
" 5. 39
" 2.77
2.91
8.01
13.78
$5.00
1902
1921
5.00
lOOSf''
1922
5.00
1904
1923...
1924 -.
5.75
1905
7.(X)
1906
2.00
2.09
2.00
4.00
5.00
5.00
5.00
5.00
3.iX)
1.25
8.75
18.00
14.00
5.00
1925 .-
7.0O
1907
1926
7.00
1908
1927 •
7.00
1909
1928
7.00
8.00
7.00
4.25
1932 ..--
1933 --
1934
19.35
1936
1937
1.00
1938
I Indicates loss.
Earnings data used are the consolidated earnings of U. S. Steel Corporation and subsidiaries after all
charges, including interest, bond premium and discount, all taxes, and additions to bond sinking funds
which were later applied to amortization of intangibles.
Earnings would be slightly lower if the special addition, to depreciation reserve of 5270,000,000 in 1935 could
be accurately apportioned over prior years.
Calculation of earnings per share is based upon average of common shares outstanding at beginning and
end of year; earnings are after preferred dividend requirement, regardless of amount paid.
Ratio of earnings to net assets — U. S. Steel Corporation and subsidiaries
Year
Earnings
(Before
Interest)
Net Assets (As-
sets less Cur-
rent Liabilities)
Ratio of
Earnings to
Net Assets
(Percent)
1920
$139, 043, 581
65. 109, 283
68, 020, 445
136,718,703
112,377,701
117,711,771
143,425,343
113,960,340
139,919,784
212,536,930
110,061,667
18, 507, 766
> 65, 862. 244
> 31, 336, 670
I 16,616,728
6, 106, 488
55, 501, 787
100,085,446
644, 874
$2,273,801,768
2, 254, 8.56, 754
2,241,899,632
2,283,479,677
2, 291, 235, 548
2,328,72.M68
2.333,017,257
2, 324, 660, 536
2. 3'29, 514. 233
2, 157. 164, 530
2. 277, 566. 636
2. 214. 480. 212
2,110.235,191
2. 047. 876, 024
2.027,821,920
1.752,870,594
1,760,418,809
1.801,398.218
1,632,017,676
6.12
1921 -
2.89
1922 - -
3.03
1923
5.99
1924
4.90
1925
5.05
1926
6.15
1927
4.90
1928
6.01
1929
9.85
1930
4.83
1931
0.84
1932
' 3.12
' 1.53
■0.82
0.35
3.15
1937
6.56
1938 - -
0.03
I Indicates loss.
Earnings are before interest but after all other
including all taxes.
CONCENTRATION OF ECONOMIC POWER
13753
RATIO OF EARNINGS TO NET ASSETS
(EARNINGS BEFORE INTEREST - TOTAL ASSETS LESS CURRENT LIABILITIES)
U. S. STEEL CORPORATION AND SUBSIDIARIES
12
12
10
10
>
8
/
\
8
/
\
6
f
\
6
t-
V
/
s
__
s
/
t-
z
4
\
/
1
/
4 5
lU
>
/
\
...
. .
/
. ,
'-'
\
1920-1938
1
a:
?
3 4";
2 0^
'^
0
^
/
a.
0
\
^
-2
\
^
-2
\
/
-4
-4
Since 1920, the roMo of earnings of U. S. Steel Corporation to the com,
bined investment of stockholders and bondholders has averaged approxi-
mately 3.4%. For the past ten years the ratio has been slightly less
than 2%.
13754 CONCENTRATION OF ECONOMIC POWER
Ratio of earnings to net assets — U. S. Steel Corporation and subsidiaries
Earnines Net Assets (As-
(Before sets less Cur-
Interest) ! rent Liabilities)
Ratio of
Earnings to
Net Assets
(Percent)
1004
1905
1906
1907
1908
1909
1910
1911.
1912
1913
1914
1915
1916
1917.
1918.
1919
1920.
1921.
1922.
1923
1924
1925.
1926.
1927.
1928.
1929.
1930.
1931.
1932.
1933.
1934.
1:^35.
19b6.
1937
$1,461,972,
1,504,670,
1, 504, 974,
1, 531, 685,
1, 565, 643,
1,612,275,
1,594,029,
1,631,283,
1, 659, 185.
1,687,652,
1,715,089,
1, 742, 0O9.
1, 750, 365.
1, 783, 707.
1, 990, 085,
2, 100, 198.
2, 173. 835,
2,208,811,
2.273,801,
2. 254, 856,
2,241,899,
2, 283, 479,
2,291,23,'),
2. 328. 725.
2,333,017,
2. 324. 660.
2, 329, 614,
2, 157, 164,
2, 277. 566,
2.214. iSO,
2,110,235,
2. 047. 876.
2,027,821,
1, 752, 870,
1,760,418,
1, 801, 398,
1,632,017,
7.63
5,39
4.01
6.42
8,20
8,31
4,83
6 78
7,11
5.12
5,06
6,57
3.24
6.09
15.25
12.10
7.74
4.84
6.12
3! 03
5.99
4.90
5.05
6,15
4.90
6.01
9,85
4,83
0,84
'3.12
1 1.53
'0.82
0.35
3,15
5,56
0,03
Indicates loss.
Earnings are before interest but after all other charges, including all taxes.
CONCENTRATION OF ECONOMIC FOWEU
13755
RATIO OF EARNINGS TO NET ASSETS
(EARNINGS BEFORE INTEREST - TOTAL ASSETS LESS CURRENT LIABILITIES)
U. S. STEEL CORPORATION AND SUBSIDIARIES
u
I \ -
1 , \ '^'L-v ,, ^
^ .. . ^2 -.ll.2^f.\.j Yd
L „ js .2 s<? - :. -. "-AyESi? l; ^.. . ^
S 3 ^^ V
GX\_ _ ir.^'/ 3 s
V / ^
t---:::& ::.::!
-fi
-6
900
902
904
906
908
910
912
914
916
918
Sines organlza+ron, the ratio of earnings of U. S. Steel Corporation to
the combined investment of stockholders and bondholders has averaged
approximately 5.1%; since 1920, the ratio has been about 3.4%; and for
the past ten years the ratio has been slightly less than 2%,.
Distribution of the sales dollar
1929-1938 inclusive-
subsidiaries
-U. S. Steel Corporation and
Classification
Dollar
Amount
Per Cent
of Sales
Payroll (Wages and Salaries)
$2, 804, 198, 490
2,379,954,228
512, 132, 759
464,685,657
64,885,182
252, 196, 770
72, 467, 364
42.8
Goods and Services Purchased from Others
36.4
Depreciation and Depletion
7.8
Taxes (Federal, State and Local)
7. 1
Bond Interest (Including Premium and DLscount)
1.0
Dividends on Preferred Stock
3 8
Available for Dividends on Common Stock
1. 1
Sales and other Revenues
$6, 550, 500. 450
100 0
Payroll represents, wages and salaries paid to all employees of all companies. The relatively small con-
struction payroll has been excluded as constituting capital expenditures subsequently recoverable through
depreciation charges.
The amount available for dividends on common stock does not represent the total amount paid but only
the portion provided by sales and revenues during the period covered.
Sales and other revenues represent the total amount available for the payment of all expenses and other
obligations. In eliminating inter-company business, amounts applicable to transportation companies
were partially estimated.
124491— 41— pt. 26- —12
13756 CONCENTRATION OF ECONOMIC POWER
Distribution of sales dollar — year 19S8 — U. S. Steel Corporation and subsidiaries
Payroll (Waees and Salaries) c
Goods and Services Purchased from Others.
Depreciation and Depletion.
Taxes (Federal, State and Local).
Available to Apply on Bond Interest...
$275. 364, 898
237,454,811
49,193.448
48, 842, 131
544, 874
45.0
38.9.
8.0
8.0
0.1
Sales and Other Revenues.
$611,400,162
Payroll represents wages and salaries paid to all employees of all companies. The relatively small con-
struction payroll has been excluded as constituting capital expenditures subsequently recoverable through
depreciation charges.
The amount available to apply on bond interest does not represent the total amount paid but only the
portion provided by sales and revenues during the period covered.
Sales and other revenues represent the total amount available for the payment of all expenses and other
obligations. In elimfnating inter-company business, amounts applicable to transporta. ion companies
were partially estimated.
DISTRIBUTION OF THE SALES DOLLAR
U. S. STEEL CORPORATION AND SUBSIDIARIES
1929-1938 INCLUSIVE
AVAIUBLE FOR COMMON DIVIDENDS 1
PREFERRED DIVIDENDS 3.8«
BOND INTEREST 1 0«
TAXES 7 1«
DEPRECIATION AND DEPLETION
GOODS AND SERVICES
PURCHASED FROM OTHERS
36.4 C
During the past ten years. 42.8^ of every dollar of sales and other
revenues of U. S. Steel Corporation and subsidiaries were paid to enn-
ployees In wages and salaries. Despite a relatively high degree of
integration, the Corporation spent 36.4{* for goods and services pur-
chased from others, e. g., scrap, non-ferrous metals. Inward freight,
electric power, tools, lubrication, etc. Depreciation of plant and equip-
ment amounted to LQif. Taxes absorbed 7.1^.
There remained for the bondholders and stockholders only 5.9^, of which
1.0^ went for bond interest, 3.8^ went for preferred dividends and l.l^
were available for dividends on the common stock.
CONCENTRATION OF ECONOMIC POWER
13757
DISTRIBUTION OF THE SALES DOLLAR
U. S. STEEL CORPORATION AND SUBSIDIARIES
YEAR 1938
AVAIWBLE TO APPLY ON BOND INTEREST 0 1«
TAXES
GOODS AND SERVICES
PURCHASED FROM OTHERS
38.9(
INCOME WAS INSUFFICIENT TO COVER BONO INTEREST BV $7,717,454, THE LOSS FOR THE YEAR
In the year 1938, 45.0^ out of every dollar of sales and other revenues of
U. S. Steel Corporation and subsidiaries were paid to employees in wages
and salaries, 38.9^ were absorbed by goods and services purchased from
others, 8.0^ by depreciation of plant and equipment, and 8.0^ by federal,
state and local ta»es.
The amount remaining was insufficient to cover bond interest by
$7,717,454, the loss for the year. After payment of preferred dividends,
the loss for the year was $32,937, 1 3 L
13758 CONCENTRATION OP ECONOMIC POWER
Payroll and earnings per dollar of sales — U. S. Steel Corporation and subsidiaries
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915.
1916.
1917.
1918
1919
1920
1921.
1922.
1923.
1924.
1925.
1926
1927
1928.
1929.
1930.
1931.
1932.
1933.
1934.
1935.
1936.
1937.
Thousands of Dollars
Payroll
Sales and
Other
Revenues
422, 1S7
39.^ 275
324, 682
400, 382
484, 535
504, 749
331, S07
442,506
492, 574
433, 036
535, 490
561, 745
413.166
524, 922
903, 033
1,276,358
1.328,248
l! 295,' 849
725, 945
809, 310
' 920,' 742
1,023,812
1,087,165
961,980
1.010,952
377. 179
423. 201
.544, 173
791, 697
1 028,751
611,400
Payroll per
Dollar
of Sales
$0. 285
.306
.307
.313
.305
.319
.363
.343
.355
.373
.354
.369
.393
.337
.292
.272
.341
.432
.449
.458
Earnings
per Dollar
of Sales
,107
.113
.181
.124
.024
1.247
1.097
1.051
.002
.064
.092
1.013
> Indicates loss
Payroll represents total wages and salaries paid to all employees of all companies and includes a relaiivily
small amount of construction payroll, which it was not pos.<!ible to exclude in early years.
Earnines are after all charges, including interest, bond premium and discount, all taxes, and addition? to
bond sinking funds which were later applied to amortization of intangibles. Earnings after interest but
before additions to bond sinking funds were about one cent more per dollar of sales than those shown, for
the years lOOl-lCTS, inclusive.
Sales and other revenues represent the total amount available for the payment of all expenses and other
oblieations. In eliminating inter-company business, amounts applicablejzi transportation companies wen-
partially estimated.
CONCENTRATION OF ECONOMIC POWER
13759
PAYROLL AND EARNINGS PER DOLLAR OF SALES
U. S. STEEL CORPORATION AND SUBSIDIARIES
^
/VkIa.'' /^^/^\ / ^°
__^/-'^^A J
- ' 1 m
c/o
EARNINGS A ^
J»nER INTEREST) J^ \l "
^^tT"l"'"/l'"'°
---- ~ "\~~y/~
X-A- -10
-20
1// -20
8gSg§2^2:22S
922
924
926
928
330
2.^
934
936
938
940
. ^ ^ ^
Ever since the organization of U. S. Steel Corporation in 1 90 1 , the propor-
tion of the sales dollar going to employees in the form of wages and sal-
aries has had an upward trend, increasing from about 30# in 1901 to about
45^ in 1938. The portion remaining as earnings available for dividends to
stoclcholders. however, has declined, even more than the portion going to
employees has increased.
13760
CONCENTRATION OF ECONOMIC POWER
Payments lo employees and to investors per dollar of sales — U. S. Steel Corporation
and subsidiaries
Thousands of Dollars
Payroll
Interest
and Cash
Dividends
Sales and
Other
Revenues
Payroll per
Dollar of
Interest
and
Dividends
per Dollar
of Sales
1906..
1006.
1907.
1908.
1909.
1910.
1911.
1912.
1913.
1914.
1915.
1916.
1917.
1918.
1919.
1920.
1921.
1924.
1925.
1926.
1927.
1928.
120,528
120,764
99,778
128, 053
147, 766
160,826
120, 511
151,663
174,955
,161,419
189, 352
207,206
162,380
176, 801
263, 386
347, 370
452,664
479, 548
581, 557
332,888
322, 678
469, 503
442, 459
456, 740
467, 409
430, 727
413, 700
420,073
391, 271
266,871
133, 913
163, 150
210, 504
251, 577
77,354
68,748
55,311
54,986
64,787
64,738
66,650
77,056
81, 265
81, 779
83,204
83, 936
73,700
64,327
101, 739
147, 704
127, 041
80, 779
79, 984
79, 127
79,002
88,111
S7, 910
87, 559
101,097
100, 779
104, 014
91,226
67, 673
26,030
12, 370
12, 257
12, 165
55,358
72,390
33, 482
422, 187
395, 275
324,682
409,382
484, 535
504, 749
331, 807
442,506
492, 574
433, 036
535,490
561, 745
413, 166
524, 922
903, 033
1, 276, 358
1, 328, 248
1, 109, 898
1, 295, 849
725, 945
809, 310
1, 093, 552
920, 742
1,023,812
1, 087, 165
961, 980
1, 010, 952
1, 094, 074
840, 226
551, 126
288, 664
377, 179
423,201
544, 173
791, 697
1, 028, 761
611,400
$0,285
.306
.307
.313
.305
.319
.363
.343
.355
.373
.354
$0,183
.174
.170
.134
.134
.128
.201
.174
.165
Payroll represents total wages and salaries paid to all employees of all companies and includes a relatively
small amount of construction payroll, which it was not possible to exclude in early years.
Interest includes bond premium and discount.
Sales and other revenues represent the total amount available for the payment of all expenses and other
obligations. In eliminating inter-company business, amounts applicable to transportation companios
were partially estimated.
CONCENTRATION OF ECONOMIC POWEIt
13761
PAYMENTS TO EMPLOYEES AND INVESTORS PER DOLU\R OF SALES
U. S. STEEL CORPORATION AND SUBSIDIARIES
en
^'° -.'zz"''\A^'yt'^\' ..
^ :;/-''\ /
</5
H 30 ?-■ \' "
O
'° wMrffc^
S
0
J?^ .^ M^ in
,0 _ .^
WTW :
900
902
904
906
908
910
912
914
916
918
920
922
924
926
928
930
932
934
1936
938
1940
From 1902 to 1938, payroll payments to employees have absorbed an
increasing proportion of the sales dollar of U. S. Steel Corporation and
subsidiaries, while interest and cash dividend payments to investors have
absorbed a decreasing proportion.
Taxes per dollar of sales — U. S. Steel Corporation and subsidiaries
Total
Taxes
Sales and
Taxes per
lotal
Taxes
.Sales and
Taxes per
Year
Other
Dollar lof
Year
Other
Dollar of
Revenues
Sales
Revenues
Sales
1920.
$68,820,598
$1, 296, 849, 375
$0. 053
1930...
$49,523,594
$840,226,222
$0. 069
1921
37, 683, 727
725, 944, 864
0.052
1931
34,247,632
651, 126, 423
0.062
1922 _
35, 798, 450
809, 309, 543
0.044
1932
31,737,202
288, 663, 837
0.110
1923 _
55,082,523
1,093,551,939
0.050
1933
31,709,993
377,179,040
0.084
1924
45, 276, 855
920, 742, 443
0.049
1934
35, 780, 385
423,201,194
0.085
1925
60, 923. 191
1,023,811,883
0.050
1935.
38,575,010
544, 172, 546
0.071
1926
52, 542, 237
1,087,164,574
0.048
1936
52,150,945
791,696,719
0.066
1927
46,755,461
961,979,849
0.049
1937
88, 048, 237
1,028,760,629
0.086
1928.
61,233,103
1,010,952,092
0.051
1938
48, 842, 131
611,400,162
0.080
1929
55,386,167
1,094,073,678
0.051
Taxes include all federal, state and local taxes of all companies.
Federal tax adjustments made retroactive to years in which applicable; distribution of adjustments to
years 1917-1920, inclusive, partly estimated.
Sales and other revenues represent the total amount available for the payment of all expenses and other
obligations. In eliminating inter-company business, amounts applicable to transportation companies were
partially estimated.
13762
CONCENTRATION OF ECONOMIC POWER
TAXES PER DOLLAR OF SALES
12
U.J
>. SI
EEL
gORPORATION AND SUBSIDIARIES
"^ 1 1 M 1 1 1
12
10
8
-
—
—
—
-^
-
\
-
y.
J
^
10
8
/
\
/
trt
/
•v
1
V-
z 6
6
y
'
m
4
\
^
__
_
4
o
■
2
0
2
0
?
3j^f^RSSlS?2Sg3S?5S^SS!SiS!^S^S
a>CT>CT>oicn<j>ai<ncT>cncri(naia>cn CTia^CT>aiaS<j>
From 1920 through 1930, U. S. Steel Corporation and subsidiaries paid out
each year in taxes approximately 5(' out of every dollar of revenue from
sales and other sources. In 1938 the tax collectors took about 8^ out of
every dollar of revenue.
The decrease In taxes per dollar of sales subsequent to 1932 was not the
result of a decline in the total amount of taxes paid, which increased from
$31,710,000 in 1933 to $88,048,000 in 1937 and $48,842,000 in 1938, but
was due to such taxes being apportioned over an increased amount of
sales.
Toial taxes and earnings available for dividends
subsidiaries
U. S. Steel Corporation and
Year
Earnings
Available for
Dividends
Total Ta.xes
Year
Earnings
Available for
Dividends
Total Taxes
1920
$109,694,228
36,617,017
39, 653, 455
108, 707, 065
85, 067, 192
iiO, 602, 053
116.667.405
.S7, 896, 836
114,17.3,775
197, 592, 060
$68,820,598
37, 683, 727
35, 798, 450
55, 082, 523
45, 276, 855
60,923,191
52, 642, 237
46. 755, 461
51,233,103
55,386,167
1930
1931
1932
1933. -.
$104,421,571
i:i, 0.38, 142
171,175,705
1 36, 501, 123
1 21, 067, 780
1,146,708
51. .583. 356
91.944,358
'7,717,454
$49, 523, 594
1921
1922
1923
34, 247, 632
31,737,202
31,709,993
35, 780, 385
1925
1926. .
1927
1935
1936
1937
38, 575, 010
52,150,945
88, 048, 237
1928
1938
48,842,131
1929
' .' urillcs loss.
1 aniinps arc after nil chiirpes, including interest, bond' premium and discount, all taxes, and additions
t" boii'l •■■inking funds wliicli were later applied to amortisation of intangibles.
T*te iiicluile all federal, state and local taxes of all companies.
pedejal tax udjusin:outs made retroaetivc to years in which npplieat'!e; distribution of adjustments to
years l9i;-iU20, inclusive, partly estimated.
CONCENTRATION OF ECONOMIC I'OWER
13763
T
200
160
1 120
o
° 80
u.
O
CO 40
z
o
^ 0
z
-40
-80
OTAL TAXES AND EARNINGS AVAILABLE FOR DIVIDENDS
U. S. STEEL CORPORATION AND SUBSIDIARIES
/
\
160
120 <
o
80 ^
o
40 w
z
o
0 .;j
z
-40
-80
EARNINGS
AVAILABLE FO
DIVIDENDS
/
\
y
/
\
/
N
/
\
1
I
^
L
'•«..
.^
_TA)
ES
^^
^
y1
/
\
.
"~
,^'
i/
\
y
/
\
/
1
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
From 1 930 through 1938, U.S. Steel Corporation's total tax bill amounted
to about $410,615,000, whereas during this period earnings available for
dividends to stockholders were about $ 1 27,072,000. Thus, the Corporation
during the last nine years has paid in taxes over three times as much as the
earnings available for dividends to stockholders.
Assets, earnings and taxes — U. S. Steel Corporation and subsidiaries
Item
1902
1937
1938
$1,511,798,510
90. 306, 525
2,391,466
$1,918,729,289
94,944,358
42,882,565
31,749,768
13,415,904
EarninRs Available for Dividends
Taxes:
State and Local (Excl. Social Security)..
17,717,4.')4
32,044,825
5, 488, 091
Social Security
11 309 215
2,391,466
88,048,237
48,842,131
1 Indicates loss.
Assets are as shown on books at the end of each year, including intangibles.
Earnings are after all charges, including interest, bond premium and discount,
to bond sinking funds which were later applied to amortization of intangibles.
Taxes include all federal, state and local ta.xes of all companies.
taxes, and additions
13764
CONCENTRATION OF ECONOMIC I'OWElt
ASSETS, EARNINGS AND TAXES
U. S. STEEL CORPORATION AND SUBSIDIARIES
MILLIONS OF DOLLARS
ASSETS
•EARNINGS A
FOR DIVIC
1
/AILABLE
ENDS
90
75
TAXES
1
SOCIAL
SECURITY
FEDERAL
.EXCLSOCStO
-45
-30
-
1
STATE 1 1
AND LOCAL ■ ■
_ II
1
I §1 "
1902
1937
1938 ■
c
1902
1937
1938
From 1902 to 1937-1938 (these two years being averaged), assets of U. S.
Steel Corporation increased 20%, earnings available for dividends
declined 51% and taxes rose 2750%.
In this comparison the year 1902 was selected because it was the first full
year of the Corporation's operation; an average of the years 1937 and
1938 was used because operations were high in one year and low in the
other year, the average being considered representative of present day
conditions.
State and local taxes — U. S. Steel Corporation and subsidiaries
Year
Amount
Year
Amount
1902
$2, 391, 466
2, 972, 600
3, 052, 967
3, 646, 490
4, 356, 126
5, 383, 924
5, 361, 160
7,597,871
8,078,585
8,846,422
9,117,678
11,296,095
11,433,763
11,804.650
14,390,155
13.577,204
17,501,453
21.968,387
30,581,138
1921
$29, 227, 488
1903
1922
31,251,245
1904 .
37, 005, 965
1905
31,513,311
1906
35, 298, 993
1907
1926
35, 266, 010
1908
1927
34, 469, 585
1909
1928
35. 8.54, 669
1910
1929
37,617,085
1911
1930
35, 954, 861
1912
1931
34, 145, 185
1913
31,0f>5,300
30, 335. 893
1916
1934 -.
31,255,688
1916
1935
32, 433. ?67
1917
1936
35. 397, 155
1918
1937 ....
42, 882, 565
1919 .
1938
32, 044, 825
1920
Data exclude social security taxes.
CONCENTRATION OF ECONOMIC POWER
13765
STATE AND LOCAL TAXES
U. S. STEEL CORPORATION AND SUBSIDIARIES
ml
anmi
~ 40
— 35
30
- 25
- 20
-- 15
— 10
— 5
— 0
State and local taxes of U. S. Steel Corporation increased steadily from
^bout $2,400,000 in 1902 to approximately $ I 3,500,000 In 1917. Between
1917 and 1920, these taxes more than doubled, although the Corpora-
tion's Ingot dapaclty increased less than 1 .57o and tts investment In prop-
erty account only aisout 5% during this period. From 192 I through 1938,
state and local taxes varied between $29,000,000 and $43,000,000, the
variation being largely due to differences in volume of operations. How-
ever, even In 1932. the year of lowest operations, these taxes amounted
to over $30,000,000.
The great Increase In state and local taxes after 1917 was not the result of
a corresponding increase in taxable property, but was largely caused by
increased assessments and tax rates, state and local.
Taxes — JJ. S. Steel Corporation and subsidiaries
State and Local
(Excl. Social
Security)
Federal (Excl.
Social Security)
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
456, 239
547, 205
076, 558
763, 544
624, 198
276, 227
28.5,876
378, 434
769, 082
568, 733
102, 447
671, 902
374, 100
524, 097
141,643
416, 670
749, 768
488, 091
$3, 337, 120
13,415,904
11, 309, 215
$37, 683, 727
35, 798, 450
55, 082, 523
45, 276, 855
50, 923, 191
52, 542, 237
46, 755, 461
51, 233, 103
55, 380, 167
49, 523, 694
34, 247, 632
31, 737, 202
31,709,993
35, 780, 385
38,57.5,010
52, 150, 945
88, 048, 237
48, 842, 131
Taxes include all federal, state and local ta.\es of all companies.
Federal tax adjustments made retroactive to years in which applicable; distribution of adiustments to
years 1917-1920, inclusive, partly
13766
CONCENTRATION OF ECONOMIC POWER
TAXES
U. S. STEEL CORPORATION AND SUBSIDIARIES
SOCIAL
SEClffilTY
~(f£OtRAL-
AND ST«TEl
i
n
^
m
. m
40
30
20
10
The heavy burden to U. S. Steel Corporation of state and local taxes is
ever present, whether the Corporation is operating at a profit or a loss.
While federal taxes, excluding social security taxes, are less In years of tow
operations because of lower income taxes, the amount of federal taxes now
paid in a prosperous year is much greater than formerly. Federal taxes in
1937 were about $31,750,000 as contrasted with about $17,769,000 in
1929, an increase of'nearjy 80%, despite the fact that production in 193 7
did not reach the level of 1929.
Social security taxes now add an additional $10,000,000 to $15,000,000
annually to the Corporation's tax bill.
The Corporation paid nearly $50,000,000 in taxes in 1938, despite the fact
that a deficit, after interest, of nearly $8,000,000 was incurred.
, Taxes paid in 1937 and 19S8—U
S. Steel Corporation and subsidiaries
Item
J937
1938
1937-38 Average
$88,048,2.37
261, 293
13, 579, 086
$1,028,760,629
8, 703, 2.52
$336.97
$0.40
$8.56
$10. 12
$48, 842, 131
202,108
7, 159, 543
$611,400,162
8, 703, 252
$241. 66
$6.70
$7.99
$5.61
$68, 445, 184
231,700
Tons of Iron and Steel Shipped
10, 369, 664
Sales and Other Revenues
$820,080,390
Shares of Common Stock
8, 703, 252
$295. 40
$6.50
$8.35
Taxes Per Share of Common Stock
$7.86
Taxes include all federal, state and local taxes of all companies.
Iron and steel shipped includes rolled and finished steel products, pig iron, ferro-manganese and ing .is.
Sales and other revenues represent the total amount available for the payment of all expenses and other
obligations. In eliminating inter-company business, amounts applicable to transportation companies
were partially estimated.
In calculating amount of taxes per ton of steel shipped, taxes of cement manufacturing subsidiaries,
amounting to $1,142,634 and $901,744 in 1937 and 1938, respectively, were eliminated: taxes of certain com-
panies, the operations of which are not entirely ■ lated to the production, sale and distribution of steeJ,
were included in full because satisfactory alloca was not possible.
CONCENTRATION OF ECONOMIC POWER
13767
U. S. Steel Corporation and subsidiaries paid $136,890,368 in taxes during
1937 and 1938
This is equivalent to an average cost per year of:
$295.40 per employee,
$6.50 per ton of iron and steel shipped
$8.35 per $100 of sales,
$7.86 per share of common stock.
An average of the years 1937 and 1938 was used because operations were
high in one year and low in the other year, the average being considered
representative of present day conditions.
Average assets, annual sales and earnings — U. S. Steel Corporation and subsidi-
aries— 1929-1938 inclusive
Year
Total Assets
Sales and Other
Revenues
Enmines
1929 .
1930—
1931
It::;:::;::;:::.:. :;.;;:; :::;:::::::;:::::;::;:;::
1934..
$2,286,183,655
2,394,544,611
2, 279, 802, 813
2, 158, 732, 222
2,102,896,880
2,084,112,287
1,822,401,742
1,863,976,619
1,918,729,289
1,711,279,008
2,062,265,902
$1,094,073,678
840, 226, 222
551, 126, 423
288, 663, 837
377, 179, 040
423, 201, 194
644, 172, 646
791, 696, 719
1,028,760,629
611, 400, 162
655,050,045
$212,536,930
■110,061,667
18, ')07, 766
1 65, 862, 244
1 31, 336, 670
1 16, 616, 728
IMS": -
1936 _.
1937
65, 601, 787
W38....
< Indicates loss.
Total assets are as shown on books at the end of each year, including intangibles.
Sales and other revenues represent the total amount available for the payment of all expenses and other
obligations. In eliminating inter-company business, amounts applicable to transportation companies were
partially estimated.
Earnings are before interest but after all other charges, including all taxes.
13768
(CONCENTRATION OF ECONOMIC rOWKK
AVERAGE ASSETS, ANNUAL SALES AND EARNINGS
U. S. STEEL CORPORATION AND SUBSIDIARIES
1929-1938 INCLUSIVE
MILLIONS OF DOLLARS
500 1,000 1,500
ASSETS
SALES
EARNINGS
A HUGE INVESTMENT IN J^ELATION TO SALES IS CHARACTERISTIC OF THE STEEL
BUSINESS. EARNINGS WERE 5.9% OF SALES BUT ONLY 1.9% OF ASSETS.
The integrated production of steel requires a heavy investnnent in iron
ore and coal nnines. transportation facilities, coke ovens, blast furnaces and
steel nnills. The ratio of sales to assets, therefore, is low.
Although earnings of U. S. Steel Corporation and subsidiaries during the
period 1929-1938, inclusive, were 5.9% of sales and other revenues, they
were only \.9% of assets.
Ratio of sales to U-.tal assets — Year 1938
Millions cf Dollars
Ratio
of Sales
to
Assets
Company
MUlions of Dollars
Ratio
of Sales
to
Assets
Company
Sales
Total
Assets
Sales
Total
Assets
Kroger Qroc. & Baking
231.3
793.8
501.7
135.2
154.9
253.1
282.4
42.0
1,067.0
102.2
57.9
306.5
286.1
89.0
170.1
276.7
406.6
62.5
1,598.0
174.4
399
259
175
175
91
91
67
67
59
General Electric
International Paper
U.S. Steel...
Du PonfCExcL O. M.
Interest)
Kennecott Copper
Glen Alden Coal
Pennsylvania R. R
Consolidated Edison...
Equitable Office Build-
ing
209.5
97.5
611.4
235.4
89.1
35.1
385.0
139.4
3.1
374.5
220.7
1,711.3
720.0
342.9
139.0
2, 322. 4
1,061.1
36.8
5fl
Sears, Roebuck
General Foods
36
U.S. Rubber
American Tobacco
International Harvest-
er
American Woolen
General Motors
Warner Brothers
33
26
26
17
13
g
Source: Moody's Manual of Investments (except for U. S. Steel data).
Sales include aU revenues resulting from the sale of goods or services or from other activities in whicli the
company is enk-aged. In most instances the figures are net, after deduction of the amount of returiis and
allowances. Slight differences in account classification exist but they are not sufllcient materially to impair
comparability.
Total assets are as of end of fiscal year, usually December 31st.
CONCENTRATION OF ECONOMIC POWER
13769
RATIO OF SALES TO TOTAL ASSETS
YEAR 1938
COMPANY
KROGER CROC. & BAKING
399»
SWIFT
259
SEARS. ROEBUCK
175
GENERAL FOODS
175
U. S. RUBBER
91
AMERICAN TOBACCO
91
INTERNATIONAL HARVESTER
69
AMERICAN WOOLEN
67
GENERAL MOTORS
67
WARNER BROTHERS
59
GENERAL ELECTRIC
56
INTERNATIONAL PAPER
44
'U. S STEEL
36
DU PONT (EXCLG.M. INTEREST)
33
KENNECOn COPPER
26
GLEN ALOEN COAL
25
PENNSYLVANIA R. R.
17
CONSOLIDATED EDISON
13
EQUITABLE OFFICE BIDG
8
100%
200%
I
Source. Mooiy'i Manuai o/ Inv^itmenti
In the year 1938 sales of U. S. Steel Corporation and subsidiaries were
equal to 36%, of total assets.
Companies which perform only a small part of the entire process of
production, fobrication, and distribution characteristically have a high
ratio of soles to total assets. On the other hand, highly integrqted
companies, such as U. S. Steel Corporation, which perform a large port
of the entire process from the production of raw materials to the fabri-
cation and distribution of the finished product, as well as companies such
as railroad and utility companies requiring a heavy investment "for the
services rendered, have a low ratio of sales to total assets.
When turnover is high, profit margins con be low. When turnover is
low, profit margins must be higher in order to produce an adequate
return on investment.
Average monthly prices of common, and preferred stocks — U. S. Steel Corporation
Jan..
Feb.
Mar.
Apr.
May
June
July.
Aug.
Sep..
Oct..
Nov.
Dee.
Common Preferred
$174.94
180.25
182.69
184.06
174.50
178. 44
200.13
234. 63
241. 44
200.25
170. 38
172. 81
$142. 51
142.50
142.63
142. 76
142. 50
140.50
140.13
141.44
143. 13
141. 19
140. 38
141.07
Jan .
Feb.
Mar.
Apr.
May
June
July.
Aug.
Sep..
Oct..
Nov.
Dec.
Common Preferred
$175.31
$141.88
183. 06
141.94
186. 00
1 14, 25
189. 50
145. 19
174.75
145. 13
162. 69
145.50
161.63
145. 69
163.88
145. 94
164. 25
148. 69
151.81
148.88
143.88
146. 38
140.94
143. 13
13770 c^N^'^
Average mon
iONTRATION OF l^CONOMTC POWER
thly prices of common ar^d preferred stocks-U. S. Sleel Corporation-
^ Continued
Common I Prcfprred
$38.56
43.69
45.63
44.63
48.25
46.56
$94.13
.75
109.75
108.19
112.88
113.25
$48.38
$118.25
57.31
124.50
64.06
129.75
63.69
127.00
57.88
122.63
61.63
126.82
62.50
126.63
G7.88
135.50
71.25
138.13
74.69
145.00
75.69
150.32
76.75
143.32
^Source: Data based on quotations on New York Stock Exchange fr
'%^KeIa?e''aveSes of monthly high ar-^ lo-.v auotation.s.
oinniprclal and Financial Chron-
CONCENTRATION OF ECONOMIC POWER
13771
3yvHS a3d savnoQ
«Oin
AVERAGE PRICES OF COMMON AND PREFERRED STOCKS
AVERAGES OF MONTHLY HIGH AND ^OW PRICES ON NEW YORK STOCK EXCHANGE
UNI 1 ED STATES STEEL CORPORATION
i
^
3
-Ii
/^: -
X
-J
g
1 3 •
§s
« : .
.. s
■ ^ —
-
§
^^- *- "
^
i
-» >
"^
i
TS £
*JL\«-g
i
\ !3^Q
\"^' I"'
Si
It' a
I ' ^
§
)P ]i
a*
1
rL
K,
1
*^
( ">
1
o\
E (
r^a
- 1 1
S >
' lu
^ i "^
V :^^
s 1
.
t c
} ^ :
h"
1 t
S "~
-- .^
V
- 1 i
• ''
: i 1
I
^
- I 1
: t --
"p •
- i ^
Y'
-^ r-
— —
- ~ -!
- s 1
S-
?
1
\
|8 8 8SSS5^ o ooovD^ 3
3avHS y3d savnoa ^
1 J
g 1
JEir
m
m
J 1
124491— 41— pt. 26 15
13772
CONCENTRATION OF ECONOMIC POWER
EARNINGS AND CASH DIVIDENDS PER SHARE OF COMMON STOCK
UNITED STATES STEEL CORPORATION
50
I EARNING
... An _ _ JALLLL
S
RE
HAR
40 ^
<
^ 9n 1- JL
- Vr 20 a!
,.iSAJ.\ ,0 ;
t '; 4/MP n
ffi^l....^L.„ i
V-4- 10
-20
-20
§i§§i8§sgi
From 1901 through 1930, consolidated earnings of U. S. Steel Corporation
and subsidiaries available for dividends on the common stock of the Cor-
poration averaged roughly $ 1 0 per share. About one half of this was paid
out in dividends and most of the remainder retained in the business was
investc i«,T plant and equiprrient.
From 1931 through 1938, earnings per share were exceeded by losses.
The common stock has received no dividend since 1 93 1 , with the excep-
tion of $1.00 per share paid in 1937.
Section B — Costs
Relationship between total costs of operation and volume of business — 19S8 condi-
tions— United States Steel Corporation andr subsidiaries
Year
Millions'of
Weighted
Tons of
Products
Shipped
Costs-1938
Conditions
(Millions of
Dollars)
Year
Millions of
Weighted
Tons of
Products
Shipped
Costs— 1938
Conditions
(Millions of
Dollars)
1927
13.0
14.0
15.1
11.9
8.1
4.4
954.5
966.2
979.0
838.8
628.9
436.0
1933
6.2
6.1
7.6
11.0
13.2
7.8
512.0
1928
I934 ..
510.0
1929
1935
610.3
1930
818.2
916.2
1932
1938
614.3
Average relationship: Costs =$182, 100,000 plus $55.73 per weighted ton of products shipped.
Total costs are adjusted to 1938 interest, pension, wage, and tax rates, to 1938 price level, and to 1938
eflBciency.
Weighted tonnages are actual tonnages, adjusted for change in proportions of high and low cost products
and for the equivalent tonnage of average cost rolled and finished steel products represented by products
other than steel.
CONCENTRATION OF ECONOMIC POWER
13773
RELATIONSHIP BETWEEN TOTAL COSTS OF OPERATION
AND VOLUME OF BUSINESS - 1938 CONDITIONS
U. S. STEEL CORPORATION AND SUBSIDIARIES
1200
1100
1000
to
900 ^
1100
1000
^ 900
g 800
'=' 700
o 600
^ 500
''
■
]-'
■'
1927
^':
1928
*1929
1936
.^
i^^
^
./'
800 ^
700 °
600 o
500 2
»^
^
1934
^.<
-.
•'^
-1 ^°°
i 300
200
100
0
^^
^
^
300 2
100
n
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
MILLIONS OF WEIGHTED TONS OF ALL TONNAGE PRODUCTS SHIPPED
NOTt: TOTW. COSTS «OJU$TtO TO 1938 INTWEST, !«. POBIOH, AND WAK RATES; TO 1938 «1CE LEVELS; AND TO 1938 EFFICIENCY
The average relatidnshlp between volume and cost in the operations of U^ S. Steel Corporation and
subsidiaries, as Indicated by the costs for each of the years 1927 to 1938 adjusted to 1938 conditions,
is such that the total costs as of 1938 for any volume of business may be estimated by multiplying the
weighted tons of products shipped by $55.73 and adding $182.(00.000. These costs are exclusive of all
non-operating income and expense and of all intercompany transactions, but they cover all operations
of the Corporation's subsidiaries and, hence, do not represent merely the cost of producing steel.
In using weighted tonnages as the measure of volume, the weighted tonnage figure for each year rep-
resents the number of gross tons of average cost rolled and finished steel which would be the cost
equivalent of the actual tons of various types of steel and other products sold during the year.
The costs embraced In the $182,100,000 are those which are incurred regardless of differences in
volume ranging from 17.7% to 90.4% of annual ingot capacity and, hence, may be termed "fixed
costs". The $55.73 per ton represents the additional costs Incurred with the expansion In operations
represented by each weighted ton of product shipped. Since the average cost amounts to $55.73 per
ton plus the pro rata portion of fxed costs and since the $ 1 82, 1 00,000 of fixed costs can be distributed
over more units as production is increased, the average cost per ton will obviously decrease as volume
rises, but the decrease is not as great as is sometimes popularly supposed.
J 3774 CONCENTKATKJN UF ECONOMIC TOWER
Composition of total costs of operation in relation to volume of business — United
States Steel Corporation and subsidiaries
Item
Costs That
Must Be Met
Regardless
of Operat-
ing Rate
Additional
Costs for Each
Additional
Weighted Ton
of Product
Shipped
$8,300,000
7,700,000
29,500,000
24,200,000
62,100,000
2,500,000
47,800,000
$0.00
0 00
Pensions
Depreciation and Depletion
2 37
Taxes other than Social Security and Federal Income
I 43
Payrolls
29 10
Social Security Taxes
1 1*1
Goods and Services Purchased, etc
21 67
$182,100,000
Data are based on cost-volume relationship indicated by 1927 to 1938 costs, adjusted to 1938 conditions.
Weighted tonnages are actual tonnages, adjusted for change in proportions of high and low cost products,
and for the equivalent tonnage of average cost rolled and finished steel products represented by products
other than steel.
CONCENTRATION OP ECONOMIC POWER
13775
COMPOSITION OF TOTAL COSTS OF OPERATION
IN RELATION TO VOLUME OF BUSINESS
U. S. STEEL CORPORATION AND SUBSIDIARIES
1200
3 4. 5 6 7 8 9 10 U 12 13 14 I
MILLIONS OF WEIGHTED TONS OF ALL
TONNAGE PRODUCTS SHIPPED
NOTE: 19271938 EXPERIENCE ADJUSTED TO 1938 CONDITIONS
17 IS^lNltfiEsl
The total costs of operation of U. S. Steel Corporation and subsidiaries embrace some items of cost
which, at 1938 interest, wage and tax rates and 1938 price levels, would remain constant throughout the
entire range of volume within which the Corporation has operated during the period 1927 to 1938, while
other items of cost vary directly with the volume of business as indicated by the volume of shipments.
Of the $182,100,000 of fixed costs, over one-third represents payroll while less than one-sixth consists
of depreciation and depletion. Of the items which vary with Increases in volume, payrolls and-goods and
services purchased represent by far the most important items, representing over 90% of the additional
costs per ton.
The above costs cover all operations of U. S. Steel Corporation and subsidiaries and do not reflect
merely the cost of producing steel. The weighted tons shipped, which are the equivalent gross tons of
average cost rolled and finished steel products represented by the actual tons of products shipped,
are used as an indicator of the volume of all operations. 18,000.000 weighted tons represent capacity
operations.
13776
CONCENTRATION OF ECONOMIC POWER
Relationship between sales and costs — effect of reduction from average 1938 prices —
United States Steel Corporation and subsidiaries
Item
Fixed
Variable
(Per Weighted
Ton Shipped)
Total Costa - -
$182, 100, 000
$65.73
$71.86
Revenues from Transportation and Miscellaneous Operations
5.80
Total Sales and Revenues, Average 1938 Prices
77.66
$64.67
5.80
Tntftl Salp« ftnd 'RnvfiTiiifi.q Avprafffl 1938 PrlfiPS Less 10%
70.47
Cost-volume relationship is that indicated by 1927-1938 costs, adjusted to 1938 conditions. Variation in
costs with changes in volume suppose no changes in wage, interest or tax rates, in pension payments, or in
materiaJ prices. , . , , ,
Weighted tonnages are actual tonnages, adjusted for changes in the proportions of high and low cost
products, and for the equivalent tonnage of average cost rolled and finished steel products represented by
products other than steel.
CONCENTRATION OF ECONOMIC POWER
13777
RELATIONSHIP BETWEEN SALES AND COSTS
EFFECT OF REDUCTION FROM AVERAGE 1938 PRICES
U. S. STEEL CORPORATION AND SUBSIDIARIES
1100 -
1000-
900-
1 800-
1 1 1 1 1 1 1
/
A
4
1100
1000
900
800 °^
700 o
600 ^
500 1
400 1
300
200
100
0
<=> SAUS MO REVENUES
lacB SALES AND REVENUES
/
^
m
1
/f
i
y
S
-
^
L
EVE
.^
-
\
i
Q 700 -
S 600-
i 500-
i ^-
300-
200 ^
100 -
oi
y
\
7
4
f
-
/
i
f
r
—
TO
AL
CO
TS
/
(
f
f
-
-
/
f
i
V
-
-
4
/
1
0
NOTE: COSTS ARE
1 2 3 4 5 6 7 8 9 10 U 12 13 14 15 16 17 1
MILLIONS OF WEIGHTED TONS OF ALL
TONNAGE Pf?ODUCTS SHIPPED
3ASE0 ON 1927-1938 EXPEftlENCC ADJUSTED TO 1938 CONDrTIONS
3
While an increase in the volunr.e of steel sold results In a considerable reduction in costs per ton, the
reduction is not so great as to permit of any sizeable reduction in price without a much greater relative
increase in volume. At the average amount of sales and revenues per weighted ton prevailing in 1938,
total sales and revenues would be sufficient to cover total costs if shipments amounted to about 8,300,-
000 weighted tons or more, which is equivalent to an operating rate of 40% to 45% of capacity, de-
pending upon the type of products shipped. A reduction of 10% from the average 1938 prices
would so reduce the total sales and revenues that the break-even point would not be reached until ship-
ments had reached about 12,400,000 weighted ton:, as indicated by the intersection of the dashed line
with the total cost line. Hence, a 10% reduction in price could be offset only by a 48.8% increase In
volume. This relationship Is not confined to the break-even point, for to net any particular amount of
profit or loss at prices 10% below the average 1938 level would require a volume 48.8% above that
required at average 1938 prices.
13778 CONCENTRATION OF ECONOMIC TOWER
Increases in volume needed to compensate for various decreases in 1038 prices com-
pared to probable resulting increases in volume-^U. S. Steel Corporation and
subsidiaries
Percentage Reduction
in Price
Percentage
Increase in
Volume
Needed
Probable
Percentage
Increase,
Assuming
Elasticity
of 1
Percentage Reduction
in Price
Percentage
Increase in
Volume
Needed
Probable
Percentage
Increase,
Assuming
Elasticity
ofl
1
3.4
7.0
10.9
15.1
19.6.
24.5
29.8
35.5
41.8
48.8
56.4
1.0
2.0
3.1
4.2
5.3
6.4
7.5
8.7
9.9
n. 1
12.4
12
64.8
74.2
84.8
96.7
110.3
125.8
143.9
165.0
190.3
220.8
13.6
2 -
13
14.9
3..
16.3
4
15
17 7
5
16
19 1
6
17
20 5
7
18
22 0
8
19
23 5
9
20
25 0
10
21
26 6
Estimates of increase in volume needed based on cost-volume relationship indicated by 1927-1938 costs
a<ljusted to 1938 conditions, and suppose no change in wage, interest, or tax rates, in pension payments, or
in material prices.
CONCENTRATION OF ECONOMIC POWER
13779
INCREASES IN VOLUME NEEDED TO COMPENSATE FOR
VARIOUS DECREASES IN 1938 PRICES
COMPARED TO PROBABLE RESULTING INCREASES IN VOLUME
U. S. STEEL CORPORATION AND SUBSIDIARIES
240
220
200
UJ
180
S
—>
^
160
z
(O
140
s
120
z
Ui
100
s
^
80
'^
60
40
20
n
r
(1
r
r
r
]
,
-
NCR
N VO
NF
^SES
LUM
OED
n
'
, ,
,i:
^.4*
1
,
■■
PROBABLE
RESULTING
'INCREASES
IN VOLUME
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
PERCENTAGE DECREASES FROM AVERAGE 1938 PRICES
NOTE: PROBABLE RESULTING INCREASES IN VOLUME BASEDON ASSUMPTION THAT ELASTICITY OF DEMAND EQUALS 1.
The Increases in the volume of s+eel soH which would have been likely to have resulted
fronn decreases in the average 1938 prices would have been but a small fraction of the
percentage Increases necessary to compensate for the respective reductions.
The elasticity of demand for a product is measured by the ratio of the relative resulting
increase in the volume to the relative decrease in price. Both actual business experience
and statistical analyses Indicate that the elasticity of demand for steel is less than I . Thus
cutting prices in half could no more than double volume. It is on the basis of an elasticity
of I that the probable Increase in volume has been computed.
The estimate of the Increase In volume needed takes into consideration the effect of the
increased volume In reducing costs per ton.
13780 CONCENTRATION OP ECONOMIC POWER
Estimated additions to 19S8 deficit — How deficit would have increased if -prices had
been reduced and volume had increased to same relative extent — U. S. Steel Cor-
poration and subsidiaries
Percentage Reduction in Price
Estimated Ad-
ditional Loss,
Assuming
Elasticity of
Demand for
Steel of 1
Percentage Reduction in Price
Estimated Ad-
ditional Loss,
Assuming
Elasticity of
Demand for
Steel of 1
1
$3,900,000
7,900,000
12, 000, 000
16,200,000
20, 500, 000
24,900,000
29,300,000
33,900,000
38,500,000
43,300,000
jj
$48 100,000
2
12 .-.
13
14
53, 100, 000
3
58, 200, 000
68, 700, 000
74, 200, 000
17 ---.
79, 800, 000
18
85, 500, 000
19 _
91,400,000
20 - — --
97, 400, 000
Estimated additional loss based on cost-volume relationship indicated by 1527-1938 costs adjusted to 1938
conditions, and supposes no change in wage, interest, or tax rates, in pension payments, or in material prices .
CONCENTRATION OF ECONOMIC POWER
13781
ESTIMATED ADDITIONS TO 1938 DEFICIT
HOW DEFICIT WOULD HAVE INCREASED IF PRICES HAD BEEN REDUCED
AND VOLUME HAD INCREASED TO SAME RELATIVE EXTENT
U. S. STEEL CORPORATION AND SUBSIDIARIES
96 872
ESTIMATED ADDITIONS TO
DEFICIT IF PRICES HAD BEEN
REDUCED AS 'NDCATED
792
712
63.2
55.2
H-47.2
392
3L2
H- 232
15.2
PERCENTAGE REBUCTION IN 1938 AVERAGE PRICE
The elasticity of demand for a product is measured by the i
increase in volume to the relative decrease in price.
itio of the resulting relative
Analyses of the demand for steel indicate that steel has an elasticity of demand no
greater than I. Thus, cutting steel prices in half could no more than double the volume.
If an attempt had been made to stimulate the volume of steel sold during the recession
of 1938 by decreasing prices further than was actually required by competition, the in-
crease in volume which would have resulted would not have been sufficient to compen-
sate for the price reduction. On the contrary, any further decrease in prices would have
served but to increase the 1938 deficit, and the greater the reduction, the more the
deficit would have increased.
13782 CONCENTRATION OF ECONOMIC POWER
Relationship between sales and costs — Effect of reduction from 2nd half of 1938
prices — U. S. Steel Corporation and subsidiaries
Item
FUcd
Variable (Per
Weighted
Ton Shipped)
Total Costs
$182, 100, 000
$55.73
,
5.80
73.13
Sales Average 2nd Half 1938 Prices Less 10%
$60 60
Revenues from Transportation and Miscellaneous Operations
5 80
Total Sales and Revenues, A-verage 2nd Half 1938 Prices Less 10%
66 40
Cost-volume relationship is that indicated by 1927-1938 costs, adjusted to 1938 conditions. Variation
in costs with changes in volume suppose no changes in wage, interest or tax rates, in pension payments, or
in material prices.
Sales per weighted ton prevailing in 2nd half of 1938 represent the average sales per weighted ton for 1938
reduced proportionately to the extent to which the selling value per weighted ton or rolled and finished
steel products shipped during the 2nd half of 1938 was less than the average selling value of rolled and
finished steel products for the entire year.
Weighted tonnages are actual tonnag&s, adjusted for changes in the proportions of high and low cost prod-
ucts, and for the equivalent tonnage of average cost rolled and finished steel products represented by
products other than steel.
CONCENTRATION OF ECONOMIC POWER
13783
RELATIONSHIP BETWEEN SALES AND COSTS
EFFECT OF REDUCTION FROM 2nd HALF 1938 PRICES
U. S. STEEL CORPORATION AND SUBSIDIARIES
1100 -
1
1100
1000
900
800 1
700 o
600 o
500 1
400 -
300
200
100
0
c i SALES AND REVENUE
y
-
/
Y
/
/
\
//
fe
^
\
i
f
(f
/f
\
/
a 7°° ~
4
/
(O
/:
V
n
r
o 5°° "
/
'I
Y
TO
AL
CO
TS
s ^°°"
/
/
//*
r
200^
/
J
t
^
'
~
-
~
0^
y
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1
MILLIONS OF WEIGHTED TONS OF ALL
TONNAGE PRODUCTS SHIPPED
NOTE COSTS ARE BASED ON 1927.1338 EXPERIENCE. ADJUSTED TO 1938 CONDITIONS
3
The reduction in the cost per ton of steel resulting from an increased volume of production Is not so
great as to permit of any sizeable reduction in price without a much greater relative Increase In volume.
At the average amount of sales and revenues per weighted ton prevailing during the second half of
1938, after the June 24, 1938 price reduction, total sales and revenues would be sufficient to cover
total costs if annual shipments amounted to about IO'/2 million weighted tons or more, which is equiva-
lent to an operating rate of about 50% to 55% of capacity, depending upon the type of product
shipped. A reduction of 10% from this price level would so reduce the total sales and revenues that the
break-even point would not be reached until over 17 million weighted tons were shipped, as Indicated
by the Intersection of the dashed line with the total cost line. This would mean that the Corporation and
its subsidiaries would not break even until operations had reached 907o of capacity. In which case opera-
tions would have to be carried on at the impossible rate of 1 30% of capacity to earn a return as
modest as SYo o" t^ie investment in tangible assets.
13784
CONCENTRATION OF ECONOMIC P0W1<:R
Increases in volume needed to compensate for various decreases in 2nd half 1938 -prices
compared to probable resulting increases in volume — U. S. Steel Corporation and
subsidiaries
Percentage Reduction
in Price
Percentage
Increase in
Volume
Needed
Probable
Percentage
Increase,
Assuming
Elasticity
ofl
Percentage Reduction
in Price
Percentage
Increase in
Volume
Needed
Probable
Percentage
Increase,
Assuming
Elasticity
ofl
^
4.0
8.4
13.1
18 3
24.0
30.2
37.2
44.8
53.5
63.1
74.1
1.0
2.0
3.1
4.2
5.3
6.4
7.5
8.7
9.9
11.1
12.4
12
86.7
K)1.3
118.3
138.4
162.7
192.4
229.7
277.9
342.6
434.1
13 6
2
13
14 9
3
14
16 3
4
15
17 7
5
16
19 1
6
17
20 5
7
18
22 0
8 - --
19
23 5
10
21
26 6
Estimates of increase in volume needed based on cost-volume relationship indicated by 1927-1938 costs
adjusted to 1938 conditions, and suppose no change in wage, interest, or tax rates, in pension payments, or
in material prices.
CONCENTRATION OF ECONOMIC POWER
13785
INCREASES IN VOLUME NEEDED TO COMPENSATE FOR
VARIOUS DECREASES IN 2nd HALF 1938 PRICES
COMPARED TO PROBABLE RESULTING INCREASES IN VOLUME
U. S. STEEL CORPORATION AND SUBSIDIARIES
440
7 8 9 10 U 12 13 14 15 16 17 18 19 20 21
PERCENTAGE DECREASES IN PRICES
NOTE: PROBABLE RESULTING
IN VOLUME BASED ON ASSUMPTION 1
r ELASTICITY Of DEMAND EQUALS I
The probable increase In the volume of sales which would result from a decrease in steel prices from
the level prevailing subsequent to the June 24, 1938 price reduction, is but a small fraction of the per-
centage increase which would be necessary to compensate for the price decrease. The divergence be-
tween the increase in volume needed and the increase in volume which would probably result from the
price reduction is even greater than the divergence with respect to reductions from tne average 1938
prices.
The elasticity of demand for a product is measured by the ratio of the relative resulting increase in the
volume to the relative decrease In price. Both actual business experience and statistical analyses indicate
that the elasticity of demand for steel Is less than I. Thus cutting prices in half coukJ no more than double
volume. It is on the basis of an elasticity of I that the probable Increase In volume has been computed.
The estimate of the increase in volume needed takes into consideration the effect of the increased vol-
ume in reducing costs per' ton.
13786 CONCENTRATION OF ECONOMIC POWER
Earnings per hour and steel prices
[Earnings per hour=earninBs per hour of all employees of United States Steel Corporation and subsidiaries;
steel prices=Iron Age composite price of finished steel]
1929
1830
1931....:..
1932
1933
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct _
Nov
Dec
1934
Jan
Feb..
Mar
Apr
May
Jun
Jul _.
Aug.......
Sep
uct
Nov
Dec
1935
Jan
Feb
Mar
Apr
May
Jun
Jul..
Aug
Sep
Oct
Nov
Dec
1936
Jan
Feb
Hourly earnings
Cents
per
hour
68.7
69.1
61.4
56.9
57.7
57.3
56.1
54.3
53.3
56.3
59.6
62.- 7
66.0
66.5
65.5
72.0
71.0
70.6
71.7
72.3
73.0
72.3
73.0
73.1
72.5
73.4
72.7
73.0
72.7
73.3
72.5
72.5
73.0
73.3
74.2
73.9
1926=
100
102.8
103.0
103.6
92.1
85.3
86.5
85.9
84.1
81.4
79.9
84.4
89.4
94.0
96.9
99.0
99.7
98.2
107.9
106.4
105.8
107.5
108.4
109.4
108.7
110.0
109.0
109.4
109.0
109.9
108.7
108.7
109,4
109.9
111.2
110.8
Steel prices
Cents
per
pound
2.209
2.048
1.957
1.901
1.885
1.873
1.867
1.817
1.802
1.890
1.9,50
1.933
1.945
1.945
1.945
1.945
1.988
2. llf
2.11
2.056
2 056
2 056
2.056
2. 056
2 056
2.056
2.056
2 056
2 056
2.056
2.05
2.05
2.05
2.05
2.062
2.062
2.062
95.4
88.5
84.5
82.1
81.4
80.9
80.6
78.5
77.8
78.6
81.1
81.3
81.6
84.2
83.5
84.0
88.8
88.8
1936
Mar
Apr
May
Jun
Jul :..
Aug
Sep
Oct
Nov
Dec
1937
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug -
Sep
Oct
Nov
Dec
1938
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep. -
Oct
Nov
Dec
1939
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Hourly earnings Steel prices
Cents
per
hour
72.8
73.0
72.4
73.7
74.2
73.9
72.5
71.9
75.0
77.8
78.0
78.5
83.1
89.8
S9.5
89.4
90.1
89.4
87.7
89.6
89.2
90.8
91.0
91.9
93.1
90.6
90.1
89.0
89.2
88.9
89.5
90.2
89.0
89.9
91.2
89.6
109.1
109.4
108.5
110.5
111.2
110.8
108.7
107.8
112.4
116.6
116.9
117.7
124.6
134.6
134.2
134.0
135.1
134.0
131. 5
129.4
129.8
131.5
133.0
134.3
133.7
136.1
136.4
137.8
139.6
135. 8
135.1
133.4
133.7
133.3
134.0
134.8
134.2
135. 2
133.4
134.8
136.7
134.3
Cents
per
pound
2.021
2.028
2.028
2.033
2.091
2.091
2.096
2.116
2.116
2.199
2.249
2.249
2.459
2.512
2.512
2.512
2.512
2.512
2.512
2.512
2.512
2.512
2.512
2.512
2.512
2.512
2.606
2.459
2.300
2.300
2.293
2. 255
2.286
2.286
2.286
2.286
2.286
2.286
2.2.16
2.236
2.236
2.236
87.3
87.6
87.6
87.8
90.3
90.3
90 5
91.4
91.4
95.0
97.1
97.1
106.2
108.5
108.5
108.5
108.5
108.5
108.5
108.5
108.5
108.5
108.5
108.5
108.5
108.5
108.3
106.2
99.4
99.4
99.0
97.4
98.7
98.7
98.7
98.7
98.7
98.7
97.5
96.6
96.6
96.6
Steel prices are monthly averages of weekly figures.
The 1926 base for earnings per hour of all employees of U. 8. Steel Corporation and subsidiaries was esti-
mated from data on the total steel industry compiled by National Industrial Conference Board, as Cor-
poration data are not available prior to 1929.
CONCENTRATION OF ECONOMIC POWER
13787
EARNINGS PER HOUR AND STEEL PRICES
1926 = 100
r\
A
w^
EARNINGS
f
(US,
SUBS
S.C. »N0
;
r
P
n
%
\,
/
J
f
=^
■^
s
W-
J STEEL PRICE
-^ (IRON *GEI
\j
^
•
1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940
So^'cc Srte/ Pf-ces - '^oo Age ctyrpos'ti^ p'<ce o
Payroll is such an Important element of costs in steel production that
increases in earnings per hour usually cause higher steel prices. However,
earnings per hour are now about 35% above the level of 1926, wherisas
steel prices are slightly lower than they were in that year. Since wages
an<J salaries average about 45% of the costs of U. S. Steel Corporation
and subsidiaries, a 35% Increase In hourly earnings causes approximately
a l57o increase in total costs.
124491— 41— 'pt 26 14
13788 CONCENTRATION OF ECONOMIC POWER
Unadjusted costs and volume of business compared with estimated costs for correspond-
ing volumes under 1938 conditions — U. S. Steel Corporation and subsidiaries
Year
Millions of
Weighted
Tons of
Products
Shipped
Unadjusted
Costs
(Millions
of Dollars)
Year
MiUions of
Weighted
Tons of
Products
Shipped
Unadjusted
Costs
(Millions
of Dollars)
1926
14.9
13.0
14.0
15.1
11.9
8.1
4.4
956.7
867.0
884.5
724^9
539.4
361.2
1933
6.2
6.1
7.6
11.0
13.2
7.8
414 4
1927
1934
442.9
1928
1935
539 2
1929
1936
731 8
1930
1937
900 5
1931
1938
614 5
1932
Estimated relationship of cost to Volume under 1938 conditions: Costs =$182, 100,000 plus $55.73 per
weighted ton of products shipped.
Unadjusted costs are as per profit and loss statements submitted to Federal Trade Commission, February
7. 1939, exclusive of Federal income taxes, miscellaneous non-operating income and expense, and of inter-
company items.
Estimated relationship of cost to volume is that indicated by 1927-1938 costs adjusted to 1938 interest,
tax, pension and wage rates; to 1938 price levels; and to 1938 eflBciency.
Weighted tonnages are actual tonnages, adjusted for change in proportions of high and low cost products
and for the equivalent tonnage of average cost rolled and finished steel products represented by products
other than steel.
CONCENTRATION OF ECONOMIC POWER
13789
UNADJUSTED COSTS AND VOLUME OF BUSINESS COMPARED WITH
EST. COSTS FOR CORRESP. VOLUMES UNDER 1938 CONDITIONS
U. S. STEEL CORPORATION AND SUBSIDIARIES
1100
1000
1 900
=i 800
o
° 700
o 600
- 500
2 400
i 3°o
200
100
0
/-
1100
1000
CO
900 ^
800 g
700 °
600 S
500 -
400 2
300 1
200
100
0
,'-'
'
'Y'-'
''
,^-
1928
'"■
.^ 'r936
1930
1933^
-i
.,''
IW
j931
^.'
''■
[.934
_^-
■-''
1932
1933
'-'
UNADJUSTED COSTS
EST. COSTS UNDER '38 CONDITIONS
'
L
^r
"
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
MILLIONS OF WEIGHTED TONS OF ALL TONNAGE PRODUCTS SHIPPED
Total costs of United States Steel Corporation and subsidiaries, exclusive
of Federal income toxes, the net amount of miscellaneous non-operating
income end expense, ond of inter-company items, hove been greater for
a given volume of sfiipmenfs in recent years tfion formerly. Tfie year 1938,
for instance, may be compared with 1935 and 1931, while 1937 may te
contrasted with 1927 and 1928.
The dotted line represents the estimated total costs ot the various volumes
if 1938 interest, tox, ond wage rates, pension payments, material prices
and technological conditions had prevoiled throughout the period. The
combined effect of the various factors increasing and decreasing costs,
such OS increased woges, taxes, and material prices, and improved tech-
nology and efficiency has been to raise the cost level obove the levels of
eoch of the preceding yeors.
How costs have increased-
U. S. Steel Corporation and subsidiari
with 1931
-1938 compared
Item
1931
1938
Per Cent
Increase
(•38 over
'31
Payroll (Wages and Salaries)
$253, 178, 649
197, 874, 481
47, 317, 895
34, 247, 632
$551, 126, 423
18, 507, 766
7, 676, 744
$275, 364, 898
237,454,811
49, 193. 448
48,842,131
$611, 400, 162
544,874
6, 659, 253
9
Goods and Services Purchased from Others
20
Depreciation and Depletion
4
Taxes (Federal, State and Local)
43
Sales and Other Revenues
Earnings (Befnre Interest)
-97
Rolled and Finished Steel Products ShTp]^d (Tons)
-13
Payroll represents wages and b 'aries paid to all employees of all companies. The relatively small con-
struction payroll has been excluded, as constituting capital expenditures subsequently recoverable through
depr-^ciation charges.
Sales and other revenues represent tie total amount available for the payment of all expenses and ether
obligations. In eliminating inter-compary business, amounts applicable to transportation companies were
partially estimated.
13790
CONfENTTlATION OF ErONOAriC T»OWRR
HOW COSTS HAVE INCREASED
U. S. STEEL CORPORATION AND SUBSIDIARIES
1938 COMPARED WITH 1931
.— — ""depranddepl ^-
(WAGES AND SALARIES)
In 1938, shipments of rolled and finished steel products by U. S. Steel Corporation subsidiaries were about
1,000,000 tons less then in 1931. However, largely because of higher prices and differences in the types
of products sold, sales and other revenues were opproximotely $60,000,000 higher than in 1931.
But in 1938
Payroll was up
Goods and Services Purchased were up
Depreciation end Depletion were up
Taxes were up
making a total increase in cesfs of
with the result that
Earnings before Interest were down
$22,000,000
40,000.000
2,000,000
14,000,000
$78,000,000
CONCKNTl{A'ri(»N Of' l':c:(»N()MlC TOWEIl 1371)1
How costs have increased — If. S. Steel Corporation and subsidiaries— 1037 compared
with 1929-'S0 average
Item
I929-'30 Average
1937
Per Cent
Increase
'37 over
'29--30
Payroll (Wages and Salaries)
$387, 416, 114
306,010,460
60, 912, 142
51,511,936
$967, 149, 950
161,299,298
13,429,325
$426, 330, 944
353, 434, 790
60,861,212
88, 048, 237
$1,028,760,629
100,085,446
12, 789, 841
10
Goods aud Services Purchased from Others
16
Depreciation and Depletion
Rolled and Finished Steel Products Shipped (Tons)
-5
Payroll represents wages and salaries paid to all employees of all companies. The relatively small con-
struction payroll has been excluded as constituting capital e.xpenditures subsequently recoverab'e through
depreciation charges.
Sales and other revenues represent the total amount available for the payment of all expenses and other
obligations. In eliminating inter-company business, amounts applicable to transportation companies were
partially estimated.
13792
CONCENTRATION OF ECONOMIC POWER
HOW COSTS HAVE INCREASED
U. S. STEEL CORPORATION AND SUBSIDIARIES
1937 COMPARED WITH 1929-'30 AVERAGE
,^-' TAXES ^^
.^— -D'EPR. AND DEPj ■^-^
In 1937, shipments of ro'led and finished steel products by U. S. Steel Corporation subsidiaries were about
640,000 tons less tbon the overoge of 1929-'30. However, lorgely because of hgher prices and differ-
tnces in the tyoes of products sold, soles and other revenues were opproximotoly $62,000,000 higher
'fhj„ ,n 1929 -JO.
But in 1937
Payroll was up $39,000,0iD0
Goods and Services Purchased were up 47,000,000
Taxes were l p 37,000,000
.naming a total increase m costs of $123,000,000
with the result that
Earnings before Interest were down $61,000,000
CONCENTRATION OF ECONOMIC POWER 13793
Composite mill net yield and cost per weighted ton shipped — U. S. Steel Corporation
and subsidiaries
[1926 = 100]
Composite Mill Net Yield
Jan Feb Mar Apr May Jun Jul Aug
Oct Nov Dec
1926-
1927..
1928..
1929..
1930..
1931..
1932..
1933..
1934-.
1935..
1936..
1937..
1938..
94.2
92.4
7816
77.0
87.1
92.1
89.0
91.4
105.4
93.2
94.2
91.6
79! 1
76.0
88.1
92.0
97.0
93^9
91.2
82.3
75! 6
87.4
91.9
87.6
93.3
105.9
95.8
78.7
75.0
87.1
95.8
104.3
95.1
100.1
95.9
94.3
94.2
8l!4
77.7
74.5
88.5
92.0
87.1
98.0
104.4
94.8
96.4
93.8
94.3
80^4
79.2
74.6
87.4
91.2
96.3
92.9
95.0
86.6
79.9
79.5
73.5
91.8
90.5
87.3
101.6
97.9
91.4
96.3
92.4
95.4
86.0
79.8
79.3
75.0
92.9
101.9
96.2
91.4
99.9
95.9
92.7
94.5
85.0
81.9
79.0
77.2
91.9
90.0
88.8
103.4
95.9
91.4
99.6
94.9
92.9
94.3
79.4
9<.3
89.6
105! 7
93.7
92.2
99.9
95.2
93.9
94.3
78.2
82.6
92.5
90! 0
104.8
91.6
93.0
99.8
93.5
93.7
94.0
82.0
80.2
77.9
83.5
89.9
89.6
90.6
105.3
92.2
The composite mill net yield index represents the amount, relative to that for 1926, received per ton by
U. S. Steel Corporation subsidiaries (after freight) from sales of a representative constant assortment of all
principal products.
Cost per Weighted
Ton Shipped
Year
Cost per Weighted
Ton Shipped
Year
Actual
Cost
Estimated
Cost at
1926 Vol-
ume
Actual
Cost
Estimatea
Cost at
1926 Vol-
ume
1926
100.0
103.6
98.3
91.8
105! 6
129.6
100.0
101.1
97.2
92.0
92.8
92.2
91.7
1933. . .
105.8
115.0
112.7
105.9
108.7
124.5
1927 - -
1934.
1928
1935
96 6
1929
1936
1930
1937
106 3
1931
1938
107 6
1932...
Actual cost per weighted ton shipped is total cost, exclusive of bond interest. Federal income taxes, mis-
cellaneous non-operating income and expense, and of inter-company items, for all subsidiaries of U. S.
Steel Corporation, divided by the number of weighted tons shipped. Weighted tonnages are actual ton-
nages, adjusted for change in proportions of high and low cost products and for the equivalent tonnage of
average cost rolled and finished steel products represented by products other than steel. The cost of opera-
tions not related to the production of steel is included in total cost, but since such cost is a small percentage
of the total and since the other operations tend to expand and contract with the volume of steel production,
the relative change in the total cost per weighted ton may be considered fairly indicative of the change in
the cost of producing steel.
Estimated cost if 1926 volume maintained is the actual cost per weighted ton shipped adjusted to 1926
volume on the assumption that the percentage change in the average cost per ton as the result of a given
change in volume would have been the same in each of the respective years as it is estimated to have been
under 1938 conditions.
13794 C'TONCENTRATION OF ECONOMIC POWER
Section C — Prices
Average yearly base prices of principal sleel products — reported by Iron Age
Rails
Structural
Shapes
Plates
Standard Pipe
Bars
Year
Dollars
per
Gross
Ton
1924
= 100
Cents
per
Pound
1924
= 100
Cents
per
Pound
1924
= 100
Dollars
^%
Ton
1924
= 100
Cents
per
Pound
1924
= 100
1924
1025 -....-
43.00
43.00
43.00
43.00
43.00-
43.00
43.00
43.00
42. 44^
39.26
36.37
36.37
36.59
41.86
41.77
40.00
100.0
100.0
100.0-
100.0
100.0
100.0
100.0
100.0
98.7
91.3
84.6
84.6
85.1
97.3
97.1
93.0
2.19
i;9.5
1.83
1.87
1192
1.69
1.62
1.57
1.68
1.78
1.80
1.85
2.21
2.17
2.10
100.0
90.9
83! 6
85.4
87.7
77.2
74.0
71.7
76.7
81.3
82.2
84.5
100.9
99.1
95.9
2.12
1.91
1.88
1.82
1.87
1.93
1.69
1.62
1.57
1.61
1.78
1.80
1.85
2.21
2.17
2.10
100.0
90.1
85! 8
88.2
91.0
79.7
76.4
74.1
75.9
84.0
84.9
87.3
104.2
102.4
99.1
70.30
70.30
70.30
69.57
69.84
70.30
67.45
6,5.29
64.89
61.63
66.32
68.40
62.01
69.17
67.00
63.00
100.0
100.0
100.0
99.0
99.3
100.0
95.9
92.9
92.3
87.7
94.3
97.3
98! 4
95.3
89.6
2.20
2.02
2.00
1.84
1.87
1.92
1.71
1.63
1.57
1.64
1.81
1.81
1.93
2.40
2.35
2.21
100.0
1926
1927
1928
85 0
1929
87 3
1930
77 7
1931
74 1
1932
71 4
1933
74.5
1934
82.3
1935
1936
87.7
1937
109.1
1938
106.8
1939 >
100 5
AVire Nails
Hot Rolled Sheets
Cold Rolled Sheets
Tin Plato
Year
Dollars
per Keg
1924 = 100
Cents
per
Pound
1924 = 100
Cents
per
Pound
1924=100
DoUars
per Base
Box
1924=100
1924
2.89
2.72
2.65
2.54
2^57
2.10
1.88
1.95
1.99
2.52
2.53
2.13
2.67
2.60
100.0
94.1
91.7
87.9
89.3
T2.7
65.1
67.5
87.2
87.5
73.7
92.4
90.0
84.4
. 2.79
2.45
2.37
2.20
2.04
2.12
1.99
1.86
1.71
1.62
1.85
1.85
1.92
2.35
2.25
2.08
100.0
87.8
84.9
78.9
73.1
76.0
71.3
66.7
61.3
58.1
66.3
66.3
68.8
84.2
80.6
74.6
5.00
4.39
4.30
4.17
4.03
4.06
3.64
3.13
2.80
2.48
2.96
2.95
3.02
3.49
3.31
3.13
100.0
87.8
86.0
83.4
80.6
81.2
72.8
62.6
56.0
5.50
5.50
5.50
5.48
5.25
5.35
5.19
4.94
4.69
4.43
5.25
5.25
5.25
5.22
5.31
5.00
100.0
1925
100.0
1926
100.0
1927
99 6
1928
95.5
1929
97.3
1930
94.4
1931
.89.8
1932
85.3
1933
49.6
80.5
1934
59.2
95.5
1935
59.0
60.4
69.8
66.2
62.6
95. 5
1936
95.5
1937
94.9
1938
96.5
90.9
Data for 1939 are on basis of first 8 months.
CONCENTKATIUN OF ECUNlLMJC I'UWEU
18795
AVERAGE YEARLY BASE PRICES OF PRINCIPAL STEEL PRODUCTS
REPORTED BY IRON AGE 1924 = 100
II
r'"j: T"Ti-
TTT': ""' JV
i /il» \. .u«
rw« ^ S
RA
t- I 'i'm-
;-l I -..- --|;;f-
'{' " " Y
/ 80 --- ^s,^j/
_L-.50 ~ J_
iiiil
iiii iiiii
im illiiiili
Considerable flexibility exists in steel prices. Not only do steel prices
fluctuate v/idely but, also, prices of different steel products fluctuate in
varying degree and direction.
As compared with 1924, prices of steel today are generally lower, whereas
wage rates are roughly 30% higher.
13796
CONCENTRATION OF ECONOMIC POWER
Reported base price and mill net yield — heavy structural shapes at Pittsburgh
[Cents per pound]
Base
Price
&
Base
Price
Mill
Yield
Price
Mill
Yield
1926
May ..
1930
Sep
1934
Jan
1.90
1.90
1.90
1.90
1.90
1.94
2.00
2.00
2,00
2.00
2.00
2.00
1.81
1.79
1.81
1.80
1.78
1.81
1.82
1.84
1.85
i.Si
1.83
1.73
1.69
1.65
1.61
1.60
1.60
1.60
1.60
1.68
1.63
1.61
1.57
1.52
1.51
1.51
1.47
1,80
1.80
1.80
1.80
Feb
Oct.
Mar
Jul
Nov .-
Dec
May
Sep
Jan
Feb
Oct
Jul
Nov
1935
Aug
Dec
Sep
Jan.. _
1.80
1.80
1.80
1.80
1.80
1.80
1.80
1.80
1.80
1.80
1.80
1.80
Oct .
1931
1.74
Nov
Dec
Mar
1.78
1.64
1.65
1.65
1.65
1.65
1.65
1.63
1.60
1.60
1.60
1.60
1.50
1.47
1.47
1.5C
1.50
1.48
1.49
1.48
1.46
1.46
1.44
1.47
1.46
May..
1927
Feb
Jan
1 75
Mar
Jul
1 76
1.98
1.90
1.90
1.88
1.80
1.80
1.80
1.80
1.78
1.75
1.77
1.80
1.82
1.83
1.80
1.81
1.78
1.76
1.78
1.73
1.72
1.71
1.71
1.71
Apr
Aug
1 76
Jan
May ""'
Sep
1 74
Feb
Oct
Mar
Jul .
Nov
Dec
May'::"."::"
Sep .
Jan .
Oct
Jul
Nov
1936
Dec
Sep..
1.80
1.80
1.80
1.80
1.80
1.80
1.90
1.90
1.90
1.90
1.90
1.90
1932
1,74
Feb .
Mar...
1.72
1.60
1.50
1.52
1.60
1.60
1.60
1.60
1.60
1.60
1.60
1.60
1.60
1.39
1.37
1.33
1.35
1.34
1.36
1.42
1.37
1.40
1.44
1.42
1.47
1.72
May
1928
Feb
Jun
1 72
Mar
Jul
1.71
1.81
1.85
1.85
1.85
1.85
1.85
1.85
1.90
1.90
1.90
1.90
1.90
1.74
1.74
1.76
1.77
1.75
1.77
1.77
1.77
1.77
1.78
1.80
1.79
Apr
Aug
1.73
Jan
May
Sep
1.77
Feb
Oct " "
1.79
Mar
Jul -
Nov
1.79
Dec
i^:::::::::::
Sep
Jan _
.Oct
Jul
Nov
1937
Aug
Sep
Dec
Jan
2.05
2.05
2.21
2.25
2.25
2.25
2.25
2.25
2.25
2.25
2.25
2.25
Oct
1933
1.84
Nov
Feb.
1.85
Dec
Mar... .-
1.88
1.60
1.60
1.60
1.60
1.60
1.60
1.60
1.60
1.60
1.70
1.70
1.70
1.45
1.42
1.46
1.49
1.46
1.46
1.46
1.48
1.55
1.59
1.63
1.63
Apr
1.93
May
1929
2.00
Mar
Jul
2.10
1.90
1.90
1.90
1.95
1.95
1.95
1.95
1.95
1.95
1.90
1.90
1.90
1.80
1.80
1.79
1.80
1.80
1.80
1.80
1.79
1.78
1.78
1.78
1.76
2.17
Jan .
May
Sep...
2.19
Feb
Jun
Oct .
2.18
Mar
Jul
Nov
2.20
Apr
Aug
Dec
2.20
May
Sep
Jan
Jun
Oct
Jul -
Nov
1938
Dec
Sep :::::::::::
Jan
2.25
2.25
2.25
2.25
2.25
2.22
2.10
2.10
2.10
2.10
2.10
2.10
1934
Feb
2.14
Dec
Mar
Apr
2.18
1.70
1.70
1.70
1.74
1.85
1.85
1.81
1.80
1.70
1.67
1.66
1.68
1.69
1.70
1.G9
1.67
2.16
May
1930
2.12
Mar
Jul
2.05
1.83
1.80
1.80
1.80
1.77
1.70
1.71
1.70
2.02
May
Sep
2.01
Feb":":"":":
Oct
2.01
Mar
Jul
Nov
2.00
Apr
Aug
Dec . -
1.98
Base prices are as rer)ortcd by Iron Age and are monthly averages of weekly figures.
Mill net yield is an average of yields at Clairton and Homestead plants of U, S. Steel Corporation subsidi-
ary which represent net sales of heavy structural shapes to domestic market (after freight) divided by num-
ber of tons shipped, converted to cents per pound.
CONCENTRATION OF ECONOMIC TOWER
13797
5
4
Q
z 3
o
Q.
Ul
0.
z
o
1
REPORTED BASE PRICE AND MILL NET YIELD
HEAVY STRUCTURAL SHAPES AT PITTSBURGH
5
4
Q
3 z
■=>
o
Q-
CC
Q-
2
z
(
EPORT
:d
^7
^
^
\=
^
-^
\
^".
■^
r. "'O'**^^' n
y
If^
r^
/
/
■^
^
s
NET YIELD
jesioiARy)
1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1935 1937 1938 1939 1940
Base prices of heavy structural shapes at Pittsburgh, as reported by
Iron Age, have shov^n considerable flexibility since 1 926.
There has been even more fluctuation in the mill net yield, that is, the
amount per pound actually received by the U. S. Steel Corporation subsidi-
ary after deduction of cost of delivery. Such mill net yield declined 25%
from the high of 1929 to the low of 1932. The Increases In prices In 1937
were the resolt of increased wages and other costs.
Factors tending to lower mill net yield with respect to reported base
price are principally (a) reductions from base price, and (b) excess of actual
cost of delivery over freight added to base price in computing the
delivered price. Factors tending to raise mill net yield with respect to
reported base price are principally (a) extras for special finish, quality, size,
heat treatment, etc., and (b) extras for small quantity.
13798 CONCKNTRATTON OF EOONOMTC POWER
Reported base price and 7nill net yield — Heavy structural shapes at Chicago
[Cents per pound]
Base
Price
Mill
Yield
Base
Price
Mill
Yield
Base
Price
Mill
Yield
1926
May -.
1930
Sep
1934
2.10
2.10
2.10
2.10
2.10
2.10
2.10
2.10
2.10
2.10
2.10
2.10
1.90
1.87
1.91
1.90
1.91
1.91
1.93
1.93
1.96
1.95
1.96
1.97
1.83
1.79
1.75
1.75
1.71
1.70
1.70
1.70
1.75
1.71
1.70
1.72
1.65
1.64
1.64
1.60
1.85
1.85
1.85
1.85
1.79
Feb
Oct
1.83
Mm
Jul
Nov
1.80
Dec .
Jan.
1.85
M^::::;::::::
Sep
Jun.
Oct
Jul
Nov...
1935
Aug
Dec
Sep
Jan
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
Oct
1931
1.86
Feb
1.85
Mar.--
1.86
1.71
1.72
1.70
1.75
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.60
1.67
1.69
1.70
1.66
1.69
1.64
1.68
1.64
1.63
1.64
1.62
Apr
1.85
May
1927
Feb
Jun
1.86
Mar
Jul
1.89
2.10
2.03
2.00
2.00
2.00
2.00
2.00
1.94
1.90
1.85
1.87
1.90
1.92
1.90
1.93
1.90
1.90
1.89
1.87
1.85
1.85
1.82
1.79
1.79
Jan
May
Sep
1.87
Oct
Mar
Jul
Nov
1.86
Dec
1.83
May
Sep
Jan
Oct
Jul
Nov
1936
Aug
Dec
Sep _.
Jan
1.85
1.85
1.85
1.85
1.85
1.85
1.95
1.95
1.95
1.95
1.95
1.95
1932
Feb.
1.85
Dee
Mar
Apr
1.84
1.68
1.65
1.68
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.59
1.58
1.60
1.63
1.66
1.70
1.65
1.68
1.65
1.65
1.84
May
1928
Feb
Jun
1.83
Mar
Jul . .
1.84
1.91
1.95
1.98
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
1.89
1.88
1.88
1.86
1.85
1.86
1.84
1.86
1.84
1.86
1.87
1.87
1.87
Jan
May::::::::":
Sep
1.84
Feb
Oct
1.90
Mar .
Jul
Nov
1.91
Anr
Aug
Dec
1.93
Xy:::::::::::
Sep
Jan
Jun
Oct
Jul
Nov
1937
Aug
Sep
Dec
Jan.. .
2.10
2.10
2.25
2.20
2.30
2.30
2.30
2.30
2.30
2.30
2.30
2.30
1933
Nov
Feb
1.94
Dec
Mar
1.96
1.70
1.70
1.70
1.70
1.77
1.70
1.70
1.69
1.65
1.75
1.75
1.76
1.63
1.62
1.63
L59
1.54
1.53
1.60
1.63
1.69
1.71
1.75
2.04
May
1929
Feb
2.05
Mar
Jul
2.12
2.00
2.01
2.05
2.05
2.05
2.05
2.05
2.05
2.05
2.03
2.00
2.00
L94
l!91
1.90
1.88
1.89
1.91
1.90
1.88
1.87
1.83
May
Aug
2. 16
Jan
Sep
2.21
Feb
Jun
Oct .
2.28
Mar
Jul
Nov
2.29
Dec
2.32
May
Sep
Jan
Jun!
Oct....
Jul
Nov..
1938
Aug
Dec
Sep
Jan
2.30
2.30
2.30
2.30
2.30
2.26
?. 10
2.10
2.10
2.10
2.10
2.10
Oct
1934
2.32
Nov
Feb
2.29
Dec
Mar
2.31
1.76
1.75
1.75
1.75
1.90
1.90
1.86
1.85
1.80
1.78
1.80
1.80
1.81
1.81
1.84
1.81
2.25
May
1930
Feb
Jun
2.24
Mar
Apr
Jul
2.14
1.99
1.95
1.94
1.90
1.84
1.82
1.79
1.79
Aug . ...
2.11
Jan
May
Sep
2.12
Feb
Jun
Jul
Oct
2.09
Mar
Nov
2.12
Aug
Dec
2.11
Base prices are f»s reporteil by Iron Age and are monlbly averages of weekly figures.
Mill net yield Is that of South Cliicago pluut of V. S. Stt't-i Corixiratiou subshiiHry, and represents net
SHies of heuvy structural shapes to donie.sllc iimrkel (after fril(dit) divided l>y iiuinbcr o( loiis slapped,
•-■ouverteil tn twuts iicr pound.
C()NCt:NTliATlUN OF ECONOMIC I'OWER
13799
5
4
^ 3
o
Q.
a:
a.
z
LJ
O
REPORTED BASE PRICE MID MILL NET YIELD
HEAVY STRUCTURAL SHAPES AT CHICAGO
5
4
Q
a:
0.
z
U)
O
1
\
# .
REF
noTrn
■7
\.
^
^BASE PRIC
■NAdRON AGE)
a
^
rf
i
ty^
V.
ILL NE
T YIEL
D
SUBS
1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940
Base prices of heavy structural shapes at Chicago, as reported by Iron Age,
have shown considerable flexibility since 1926.
There has been even more fluctuation In the mill net yield, 1. e,, the amount
per pound actually received by the U. S. Steel Corporation subsidiary after
deduction of cost of delivery. Such mill net yield declined 24% from the
high of 1929 to the low of 1933. The Increases in prices In 1937 were the
result of Increased wages and other costs.
Factors tending to lower mill net yield with respect to reported baie price
are principally (a) deductions from base price and (b) excess of actual cost
of delivery over freight added to base price In computing delivered price.
Factors tending to raise mill net yield with respect to reported base price
are principally (a) extras for special finish, quality, sI-t, heat treatment,
etc, and (b) extras for small quantity.
13800 CONCENTRATION OF ECONOMIC POWER
Reported base price and mill net yield — Plates at Pittsburgh
[Cents per pound]
MiU
Net
Yield
1.86
1.85
1.85
1.84
1.80
1.80
1.80
1.78
1.75
1.77
1.80
1.81
1.85
1,85
1.85
1.85
1.85
1.85
1.90
1.90
1.90
1.90
1.90
1.90
1.90
1.90
1.95
1.95
1.95
1.95
1.95
1.95
1.94
1.90
1.90
1.75
1.76
1.75
1.75
1.77
1.77
1.84
1.85
1.85
1.86
1.84
1.82
I; 79
1.76
1.75
1.73
1.74
1.70
1.71
1.72
1.70
1.73
1.73
1.75
1.74
1.73
1.70
1.73
1.74
1.73
1.75
1.75
1.76
1.75
1.77
1.77
1.79
1.70
1.70
1.69
1.70
1.68
May
Jun.
Jul-
Aug-
Sep.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
Jun.
Jul-.
Aug.
Sep.
Oct.
> .V.
Jan.
Feb.
Mar.
Apr-
May
Jun.
Jul..
Aug-
Sep.
Oct.
Nov-
Dec.
Jan.,
Feb.
Mar.
Apr.
May
Jun.
Jul-
Aug.
Sep.
Oct.
Nov-
Dec
Jan. .
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Mill
Net
Yield
1.73
1.69
1.65
1.61
1.60
1.64
1.65
1.65
1.65
1.65
1.65
1.52
1.60
1,60
1.60
1.60
1.60
1.60
1.60
1.60
1.60
1.60
1.60
1.60
1.55
1.50
1.53
1.60
1.60
1.60
1.70
1.70
1.70
1.52
1.53
1.48
1.47
1.47
1.50
1.51
1.49
1.43
1.43
1.42
1.40
1.39
1.40
1.43
1 40
1.38
1.38
1.34
1.33
1.36
1.32
1.40
1.34
1.36
1.34
1.37
1.41
1.47
1.36
1.31
1.32
1.33
1.33
1.42
1.52
1.55
1.59
1.70
1.73
1.70
1.72
1.70
1.72
V74
1,77
1,85
1.75
1.85
1.77
1.81
1.80
1.80
1.84
Oct.
Nov
Dec.
Jan..
Feb.
Mer.
Apr.
May
Jun.
Jul.-
Aug-
Sep-
Oct.
Nov.
Dec-
Jan.
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep.
Oct.
Nov.
Dec.
Jan.,
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep.
Oct.
Nov.
Dec.
Jan.,
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep.
Oct.,
Nov.
Dec.
Mill
Net
Yield
1.80
1.80
1.80
1.80
1.80
1.80
1.80
1.80
1.90
1.90
1.90
1.90
1.90
1.90
2,05
2.05
2.21
2.25
2.25
2.25
2.25
2.25
2.25
2.25
2.25
2.25
2.25
2.25
2.25
2.25
2.25
2.22
2.10
2.10
2.10
2.10
2.10
2.10
Base prices are as reported by Iron Age and are monthly averages of weekly figures.
Mill net yield is that of Homestead plant of U. S. Steef Corporation subsiiiiarj and represents net sales
of sheared plates to domestic market (after freight) divided by number of tons shipped, converted to cents
per pound.
CONCENTRATION OF ECONOMIC POWER
13801
5
4
? 3
2
a.
^ 2
z
UJ
O
1
REPORTED BASE PRICE AND MILL NET YIELD
PLATES AT PinSBURGH
5
4
3 i
o
OL
a:
Q.
2 CO
z
LU
o
1
/7
^
^
^
*'\/-
-^
\
REPOR
BASEP
(IRON
n
TED
RICE,
f
^
^
y
^
^
IM''
ILLNE
ITOU
SUBS
7YIEL
ssc.
)
1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940
Base prices of plates, as reported by Iron Age. have shown considerable
flexibility since 1926.
There has been even more fluctuation In the mill net yield, that is, the
amount per pound actually received by the U. S. Steel Corporation subsidi-
ary after deduction of cost of delivery.. Such mill net yield declined 32%.
from the high of 1929 to the low of 1933. The increases in prices in 1937
v/ere the result of increased wages and other costs.
.Factors tending to lower mill net yield with respect to reported base
price are principally (a) reductions from base price, and (b) excess of actual
cost of delivery over freight added to base price In computing the
delivered price. Factors tending to raise mill net yield with respect to
reported base price are principally (a) extras for special finish, quality, size,
heat treatment, etc.. and (b) extras for small quantity.
13802 CONCENTRATION OF ECONOMIC POWER
Reported base price and mill net yield — Bars at Pittsburgh
[Cents per pound]
2.00
2.00
2.00
2.00
1.95
2.00
2.00
2.00
2.00
2.00
2.00
2.00
1.89
1.85
1.85
1.79
1.92
1.92
1.93
1.96
1.96
1.95
1.93
l!90
1.89
1.84
1.77
1.77
1.76
1.80
1.80
1.81
1.80
1.82
1.81
1.83
1.84
1.86
1.85
May
Jun.
JuL-
Aug
Sep.
Oct.
Nov.
Dec.
Jan..
Feb
Mar.
Apr.
May
Jun.
Jul.-
Aug.
Sep.
Oct.
Nov.
Dec.
Jan.-
Feb-.
Mar.
Apr.
May.
Jun..
Jul...
Aug..
Sep..
Oct..
No\\^
Dec.
Jan..
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug
Mill
Price Yield
1.75
1.73
1.65
1.64
1.61
1.60
1.60
1.60
1.64
1.62
1.60
1.56
1.55
1.64
1.55
1.65
1.57
1.65
1.57
1.65
1.55
1.65
1.54
1.65
1.50
1.63
1.49
1.60
1.50
1.60
1.46
1.60
1.49
1.60
1.46
1.58
1.47
1.50
1.50
1.52
1.60
1.60
1.60
1.60
1.60
1.60
1.60
1.60
1.60
1.46
1.40
1.45
1.47
1.45
1.45
1.48
1.45
1.51
1.50
1.52
1.45
1.48
1.50
1.47
1.51
1.38
1.43
1.42
1.43
1.49
1.51
1.52
1.75
1.75
1.75
1.79
1.90
1.90
1.82
1.73
1.75
1.71
1.80
1.81
1.81
1.81
1.85
Sep.
Oct.
Nov
Dec
Jan.
Feb
Mar.
Apr.
May
Jun.
Jul .
Aug.
Sep.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep.
Oct.
Nov.
Dec.
Jan..
Feb-
Mar.
Apr-
May
Jun.
Jul..
Aug-
Sep.
Oct.,
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep.
Oct.
Nov.
Dec-
Base Mill
Price Yield
1.80
1.81
1.80
1.84
1 80
1.84
1.80
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.95
1.95
1.97
2.05
2.05
2.05
2.20
2.20
2.40
2.45
2.45
2.45
2.45
2.45
2.45
2.45
2.45
2.45
2.45
2.45
2.45
2.45
2.45
2.41
2.25
2.25
2.25
2.25
2.25
2.25
Base prices are as reported by Iron Age and are monthly averages of weekly figures.
Mill net yield is that of Duquesne plant of U. S. Steel Corporation subsidiary and represents net sales
ofbars, rounds, etc., O. H., to domestic market (after freight) divided by number of tons shipped, converted
to cents per pound.
CONCENTRATION OF ECONOMIC POWER
13803
REPORTED BASE PRICE AND MILL NET YIELD
BARS AT PITTSBURGH
r
7
\
Ji>
K
^
'^
1
=^
REPO
BAS£
WON
tTEO
J^
-*
4
/ ,
>
V*
laN
if
)
1926 1927 1928 1929 1930 1931 1932 1933 1934 193S 1^6 1937 1938 1939 1940
Base prices of bars at Pittsburgh, as reported by Iron Age. have shown con-
siderable flexibility since 1926.
■ There has been even more fluctuation in the mill net yield, that is, the
amount per pound actually received by the U. S. Steel Corporation sub-
sidiary after deduction of cost of delivery. Such mill net yield declined
n°la from the high of 1929 to the low of 1933. The increases in prices in
1937 were the result of increased wages and other costs.
Factors tending to lower mill net yield with respect to reported base
price are principally (a) reductions frorp base price, (b) excess of actual
cost of delivery over freight added to base price in computing the
delivered price, and (c) quantity discounts. Factors tending to raise mill
net yield with respect to reported I- ase price are principally (a) extras for
special finish, quality, size, heat treatment, etc., and (b) extras for small
quantity.
124491— 41-^pt. 26 IE
13804 CONCENTRATION OF ECONOMIC POWER
Reported base price and mill net yield — bars at Chicago
[Cents per pound]
Mill
Net
Yield
2.10
2.10
2.10
2.10
2.10
2.10
2.10
2.10
2.10
2.10
2.10
2.10
2.10
2.03
2.00
2.00
2.00
2.00
2.00
1.91
1.95
1.98
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.01
2.05
2.05
2.06
2.06
2.05,
2.05
2.05
2.03
2.00
2.00
1.99
1.95
1.96
1.91
2.08
2.09
2.08
2.08
2.06
2.09
2.09
2.11
2.12
2.10
2.13
2.12
2.13
2.06
2.06
2.04
2.03
2.05
2.00
l!96
1.93
1.96
1.97
2.00
2.01
2.00
2.03
2.02
2.03
2.03
2.04
2.05
2.06
2.04
2.04
2.03
2.03
2.04
2.05
2.03
2.03
1.99
1.99
1.96
1.93
1.94
1.93
May
Jun.
Jul..
Aug.
Sep.
Oct.
Nov.
Dec-
Jan..
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep.,
Oct.
Nov.
Dec-
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep.
Oct.
Nov.
Dec.
Jan..
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep..
Oct.
Nov.
Dec.
Jan...
Feb...
Mar.,
Apr...
Mny..
Jun...
Jul-...
Aug...
Mill
Net
Yield
1.85
1.83
1.76
1.76
1.71
1.70
1.70
1.70
1.71
1.72
1.70
1.75
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.60
1.68
1.65
1.68
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.70
1.69
1.65
L80
1.80
1.80
1.80
1.95
1.80
2.00
1,80
1.97
1.84
2.00
1.»ft
1.97
1.95
1.95
1.87
1.97
1.85
2.00
Oct.
Nov
Dec-
Jan..
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep..
Oct..
Nov.
Dec.
Jan..
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep..
Oct..
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep.
Oct.
Nov.
Dec.
Jan..
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep.,
Oct.
Nov.
Dec.
Base
Price
MUl
Net
Yield
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.85
1.90
1.90
1.90
1.90
1.90
1.90
1.90
1.90
1.90
2.00
2.00
2.(K)
2
2.10
2
2.10
2.
2.10
2.
2.25
2.25
2.45
2.50
2.50
2.50
2.50
2.50
2.50
2.50
2.50
2.50
2.60
2.50
2.50
2.50
2.50
2.45
2.25
2.25
2.25
2.25
2.26
2.25
Base prices are as reported by Iron Age and are monthly averages of weekly figures.
Mill net yield is that of Gary plant of U. S. Steel Corporation subsidiary and represents net sales of bars,
rounds, etc., O. H., to domestic market (after freight) divided by number of tons shipped, converted to cents
per pound; classification broadened January, 1934.
CONCENTRATION OF ECONOMIC POWER
13805
REPORTED BASE PRICE AND MILL NET YIELD
BARS AT CHICAGO
-^
»^
V-,
'"'■
M,^
NET YIELD
us.at
y
//'
\
\:
^■■-
\
3«^
, y
■^ REPORTE
)
6<«/w
v/
<n
»N*Ot
1926 1927 1928 1929 1930 1931 1932
1934 1935 1936 1937 1933 )S39 1940
Base prices of bars at Oittago, as reported by 'ron Age, have shown con-
siderable flexibility since 1926.
There has been even more fluctuation in the mill net" yield, that is, the
annount per pound actually received by the U. S. Steel Corporation after
deduction of cost of delivery. Such mill net yield declined 22 7o f""©"! ^^^
high of 1929 to the low of 1932. The increases in prices in 1937 were the
result of increased wages and other costs.
Factors tending to lower mill net yield with respect to reported bise
price ate principally (a) reductions from base price, (b) excess of actual
cost of delivery over freight added to base price in computing the
delivered price, and (c) quantity discounts. Factors tending to raise mill net
yield with respect to reported base price are principally (a) extras for spe-
cial finish, quality, size, heat treatment, etc., and (b) extras for small
quantity.
13806
CONCENTRATION OF ECONOMIC POWER
Reported base price and mill net yield— standard black welded pipe at Pittsburgh
[Dollars per gross ton]
Base
Price
Mill
Net
Yield
Base
Price
Mill
Net
Yield
Base
Price
MUl
Net
Yield
1926
May
1930
Sep
1934
78.74
78.74
78.74
78.74
78.74
78. 74
78.74
78.74
78.74
78.74
78.74
78.74
83.59
82.41
81.71
87.70
82.46
83.46
81.51
81.73
82.05
82.43
81.90
79.43
74.48
74.48
74.48
74.48
74.48
74.48
74.48
74.48
74.99
73.22
74.96
74.38
75.00
74.01
73.83
70.89
76.61
76.61
76.61
76.61
73.34
Oct.
72.09
Jul- -
Nov
72.65
Dec
72.64
May
Sep.
Jan
Oct
Ftil
Nov
1935
Dec
Bep
Jan
76.61
76.61
76.61
76.61
76.61
76.61
76.61
76.61
76.61
76.61
76.61
76.61
Oct
1931
73.17
Feb
74.18
Mar.
73.12
74.48
74.48
74.48
74.48
71.22
72.62
72.62
72.62
72.62
72.62
72.62
72.62
73.54
74.26
74.18
73.39
69.66
69.90
70.48
71.66
71.78
70.16
70.74
71.42
Apr
72.91
May
1927
73.34
Mar --
Jul..
73.02
78.74
78.74
78.74
78.74
78.74
78.74
78.74
78.74
78.74
76.76
74.80
74.80
83.36
82.11
83.02
82.95
83.62
82.40
81.12
81.19
80.93
78.29
75.75
77.10
72.92
May. ..-.
Sep.
72.62
^eb ""
Jun
Oct
71.43
Mar
Jul
Nov
74.14
tKy :
Aug
Dec
73.36
Sep
Jan ,.
Feb -
Jun
Oct
Jul
Nov
1936
Aug
Dec
Sep ..
Jan
76.61
72.78
69.22
68.32
68.32
68.32
68.32
68.32
68.32
68.32
68.32
68.32
Oct
1932
73.19
71.90
Mar
67.45
72.62
72.62
72.62
72.62
72.62
72.62
72.62
72.62
72.80
72.80
72.80
72.80
72.21
68.64
70'. 16
66.88
70.49
68.43
68.83
68.82
68.22
68.64
65.25
Apr--- ---.
62.30
1928
May
65.17
65.96
Mar .
Jul
64.70
76.83
76.83
76.83
78.27
78.74
78.74
78.74
78.74
78.74
78.74
78.74
78.74
78.60
78.11
78.47
78.00
78.12
80.47
80.21
80.31
80.63
79.48
80.42
80.11
65.19
Jan
May :
Sep
65.86
Feb
Jun ^
Jul ^.
Aug
Oct
64.57
Mar
Nov
64.39
Apr
Dec
65.11
May
Sep
Jan
Jim
Oct
jSl
Nov
1937
Dec
Sep
Jan
68.32
68.32
77.82
79. 52
79.52
79.52
79.52
79.52
79.52
79.52
79.52
79.52
Oct
1933
66.18
Mnv
Feb
65.99
DM
Mar
65.27
72.80
72.80
72.80
64.96
64.96
69.96
69.16
69.16
69.16
69.16
69.16
69.16
68.36
69.39
68.14
60.07
50.63
63.46
61.12
60.43
64.49
65.97
65.38
64.73
Apr --
72.38
1929
;May
75.96
Feb
70.41
Mar....-
Jul-- -.
73.41
78.74
78.74
78.74
78.74
78.74
78.74
78.74
78.74
78.74
78.74
78.74
78.74
79.47
79.61
78.33
80.37
79.34
78.11
79.29
79.26
77.52
79.02
80.75
79.87
73:13
May -
Sep
74.45
Jun
Jul
Oct
74.70
Mar .
Nov
72.52
Dec
71.74
uiy :"
Sep
Jan
Oct
Jul.
Nov-
1938
Dec
Sep
Jan
79.52
79.52
79.52
79.52
79.52
79.52
70.56
70.56
7a 56
70.56
70.56
70.56
Oct
1934
76.96
Feb
73.58
Mar
75.56
69.16
69.16
69.16
71.02
76.61
76.61
76.61
76.61
66.80
67.78
66.34
66.33
67.95
66.42
73.93
74.44
Apr
75.59
1930
Feb
jSn
75.11
Mar
jSi .:::::::::::
66.76
78.74
78.74
78.74
74.48
78.56
79.63
79.14
70.89
66.00
Jan
May
Sep
65.58
Feb
Jun
Oct :::
67.32
Jul.
Nov
66.60
Dec
66.70
1
Base prices are as reported by Iron Age (converted to a gross ton basis) and are monthly averages o( weekly
figures.
Mill net yield is that of National plant of U. S. Steel Corporation subsidiary and represents net sales
of standard black welded pipe to domestic market (after freight) divided by number of tons shipped. This
classification of pipe includes some pipe sold on a base diflerent from the base price to which the mill net
yield is here compared, but the two base prices show the same general trends and fluctuations.
CONCENTRATION OP ECONOMIC POWER
13807
REPORTED BASE PRICE AND MILL NET YIELD
STANDARD BUCK WELDED PIPE AT PIHSBURGH
1
^
rs
REPORl
ED
-^
"h
fV
BsT"'
IRON AGE) rr;
r^-~>A
\__
If^
i
V-i
1^
\z
a
u
UMILL Na YIELD
1 (TO u.s.sc.
I SUBSIDIARY)
^0 A
1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940
Base prices of standard black welded pipe, as reported by Iron Age, have
shown considerable flexibility since 1926.
There has been even more fluctuation in the mill net yield, i. e.. the amount
per ton actually received by the U. S. Steel Corporation subsidiary after
deduction of cost of delivery. After decreasing gradually from 1926 to
1929, the mill net yield declined 37% from the high of 1929 to the low of
1933. The Increases in prices in 1937 were the result of increased wages
and other costs.
Factors tending to lower mill net yield with respect to reported base price
are principally (a) reductions from base price, (b) excess of actual cost of
delivery over freight added to base price in computing the delivered price,
(c) quantity discounts and (d) deductions for quality, size, etc. Factors tend-
ing to raise mill net yield with respect to reported base price are principally
(a) extras for special finish, quality, size, steel treatment, etc., and (b) extras
for small quantity.
13808
CONCENTRATION OF ECONOMIC POWER
Reported base price and mill net yield — cold rolled sheets
[Cents per pound]
Base
Price
MiU
Yield
Base
Price
MiU
Yield
Base
Price
Mill
Yield
1926
May
1930
Sep
1934
Jan
4.45
4.38
4.35
4.28
4.24
4.15
4.15
4.15
4.25
4.25
4.25
4.25
■4.61
3.75
3.66
3.60
3.60
3.50
3.45
3.38
3.30
'3.68
2.95
2.95
2.95
2.95
3 28
Feb
June
Oct
3.38
Mar "
Jul
Nov
Dec .
3.34
3.39
MaV:::::::::::
Sep
Jan
Feb
Jim.. . .
Oct
JuL
Nov
1935
Dec
Sep...
Jan
2.95
2.95
2.95
2.95
2.95-
2.95
2.95
2.95
2.95
2.95
2.95
2.95
Ort
1931
3.26
3.10
Dec
Mar
3.18
3.30
3.90
3.24
3.10
3.03
3.02
3.10
3.10
3.10
3.10
3.10
3.02
'3.12
Apr
3.17
May
1927
Feb
Jun
3.36
Mar
Jul
3.24
4.18
4.15
4.15
4.15
4.15
4.25
4.26
4.25
4.25
4.16
4.12
4.00
'4.41
Apr
Aug
Sep . . .
3.27
Ian
May """■
3.27
Feb
Oct
3.27
Mar
Jul
Nov
3.10
Aug..
Dec
Jan
3.06
t/ay.::::::::::
Sep...
Jun
Oct
1936
Jul
Nov
Aug
Dec
2.95
2.95
2.95
2.95
2.95
2.95
3.05
3.05
3.05
3.05
3.05
3.25
sep^.v.v::::::::
Jan ,
Feb
3.0O
Oct
1932
Feb
2.93
Nov
Mar
2.85
Dec
2.90
2.80
2.86
2.90
2.89
2.85
2.85
2.81
2.75
2.66
2.63
2.65
'3.07
Apr
2.85
May
3.05
1928
Mar
JuL.
3.07
4.00
4.08
4.15
4.04
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.08
•4.09
3.15
Jan
May
Sep....
3.15
Feb
Oct :
3.13
Mar
Jul
Nov
3.17
Apr
Aug
Dec
3.08
Npay-"::::::::::
Sep..
Jan
un
Oct
1937
Jul
Nov
Dec...
3.25
3.25
3.49
3.55
3.55
3.55
3.55
3.55
3.55
3.55
3.55
3.55
Sep
Jan
3.00
Oct
Feb
3.12
1933
Nov
Mar
3.27
Dec
2.35
2.25
2.30
2.30
2.34
2.29
2.40
2.47
2.76
2.75
2.75
2.76
■2.66
Apr
3.35
May
3.48
1929
Feb
Jun
3.67
Mar
Jul -
3.63
4.10
4.10
4.10
4.10
4.10
4.10
4.10
4.08
4.00
4.00
4.00
3.98
'4.23
Apr
Aug
3.54
Jan
May
Sep
3.52
Feb
Oct
3.58
Mar
Jul
Nov
3.54
Dec.
3.61
i^y:::::::::::
Sep
Jan
Jun
Oct
1938
jSi
Nov
Dec
3.55
3.60
3.45
3.46
• 3.50
3.35
3.35
3.35
3.35
3.23
3.36
3.35
Sep
Jan
3.54
Oct
1934
Feb
3.36
Nov
Mar
3.35
Dec
2.75
2.75
2.75
2.85
3.16
3.16
2.99
2.95
2.88
3.17
3.06
3.08
3.23
3.16
3.27
3.25
Apr
3.24
May
3.38
1930
Feb
Jun
3.32
M^v.v.v.".".::::
jS^v.".:::::::::
3.12
3.90
3.90
3.88
3,80
'3.68
Apr
Aug
3.08
Jan
Sfay
Sep
2.97
Feb
Oct
2.78
Mar
Jul
Nov
2.67
Apr
Aug
Dae. .
2.69
' Yearly average.
Base prices are as reported by Iron Age and are monthly averages o( weekly figures. Iron Age data are
for 20-gauge cold rolled sheets at Pittsburgh from September 1926 to April 1938; data prior and subsequent
to that period have been adjusted to that gauge.
Mill net yield is a weighted average of yields of plants of U. S. Steel Corporation subsidiaries, which repre-
sents net sales of cold rolled and automobile sheets to domestic market (after freight) divided by number
of tons shipped, converted .to cent? poi pound. Dataarefor plants of American Sheet and Tin Plate Com-
pany prior to 1937, and thereafter for Vandergrift and Oary plants of Carnegie- Illinois Steel Corporation.
CONCENTRATION OF ECONOMIC POWER
13809
REPORTED BASE PI^ICE AND MILL NET YIELD
COLD ROLLED SHEETS
'\y-
>
>—
:^
NETY
3 US S.
BSIDWR
K
u
-^^
MILL
<T
S
ELD
r
\
REPORTED
BASE PRICE
(IRON AG£I
■s
y
•J-
\F
\
1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940
Base prices of cold-rolled sheets, as reported by Iron Age, have declined considerably
since 1926 as well as having shown considerable fluctuation during the period.
The monthly mill net yield, i. e., the amount per pound received by the U. S. Steel
Corporation subsidiary after deduction of cost of delivery, has shown even more
fluctuation than the published base price. The mill net yield curve prior to 1 934 appears
to fluctuate less than the reported base price because it is based on annual averages,
monthly data not being available.
Over the period the mill net yield has declined from an average of 4.6 K per poijnd in
1926 to a l938 1owof2.67(; per pound, a decrease of 42%.
Factors tending to lower mill net yield with respect to reported base price are princi-
pally (a) reductions from base price, (b) excess of actual ccst of delivery over freight
added to base price in computing the delivered price, (c) quantity discounts and (d)
deductions for quality, size, etc. Factors 'ending to raise mill net yield with respect to
reported base price are principally (a) extras for special finish, quality, size, heat treat-
ment, etc., and (b) extras for small quantity.
13810 CONCENTRATION OF ECONOMIC POWER
Reported composite price and composite mill net yield
[1926=1001
Com-
posite
Price
Mill
Net
Yield
100.3
99
100.0
100
100.3
99
100.1
100
99.6
100
99,7
fi9
100.0
99
1(X).0
99
1(10.0
99
100.0
99
100.0
99
100.0
99
98.5
95.9
96.5
96.2
95.7
95.5
95.4
96.4
96.0
93.0
92.1
92.1
98.7
b
98.0
97.0
r
96.6
ay
95.9
n
96.4
96.3
96.3
96.9
94.9
95.2
93.6
1928
92.3
93.7
94.1
94.0
93.0
93.0
92.6
93.3
93.3
93.7
94.2
94.5
93.4
b
93.4
93.3
r
93.4
94.3
93.8
92.9
92.4
p
92.7
92.9
93.9
93.7
1929
94.7
94.7
94.7
96.0
96.2
96! 2
95.6
95.4
94.9
94.7
95.2
94.2
b
94.2
ar
93.9
94.2
94.3
95 0
ig
95.4
p
94.5
94.3
94.3
3C
94 0
1930
93.4
92.8
92.7
90.6
88:2
b
91.6
ar
91 2
pr
89.9
n
88.0
Jul..
Aug.
Sep.
Oct.
Nov
Oec.
Jan..
Feb.
Mar.
Apr-
May
Jun.
Jul..
Aug.
Sep.
Oct.
Nov.
Dec.
Jan..
Feb.
Mar.
Apr.
May
Jun.
Jul.,
Aug.
Sep..
Oct.
Nov.
Dec.
Jan..
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep.
Oct.
Nov.
Dec
Jan. .
Feb..
Mar.
Apr..
May.
Jun..
Jul...
Aug..
Sep..
Oct..
Nov.
Dec.
Com-
posite
Price
Mill
Net
Yield
87.2
86.3
85.9
85.6
85.4
84.8
86.6
86.0
85.0
83.7
83.3
82.0
85.4
85.6
85.6
85.3
85.3
84.8
84.4
84.0
84.0
84.0
83.8
82.5
82.2
83.2
82.3
81.8
81.4
80.0
81.3
80.2
81.2
81.1
81.2
82.4
82.4
82.4
82.7
82.7
82.7
82.5
82.0
82.0
81.4
80.9
80.6
78.5
77.8
78.6
81.1
81.3
81.6
84.2
83.5
84.0
78.6
79.1
79.3
78.7
77.7
79.2
79.5
79.3
79.0
78.8
77.0
76.0
76.6
75.0
74.5
74.6
73.5
75.0
77.2
79.4
82.6
83.5
87.4
91.8
92.9
91.9
93.3
92.5
89.9
Jan.
Feb.
Mar.
Apr.
May
Jun.
Jul-.
Aug.
Sep.
Oct.
Nov.
Dec.
Jan..
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep.
Oct.
Nov.
Dec.
Jan..
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep..
Oct.
Nov.
Dec.
Jan..
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep.
Oct.
Nov.
Dec.
Jan..
Feb..
Mar..
Apr.-
May.
Com- Mill
posite Net
Price Yield
88.8
88.8
88.8
87.8
90.3
90.3
90.5
91.4
91.4
95.0
108.5
105.4
108.5
105.1
108. 5
105.9
1(»S. 5
104.3
108.3
104.4
106.2
102.7
99.4
97.9
99.4
96.2
99.0
95.9
97.4
93.7
98.7
91.6
98.7
92.2
98.7
98.7
98.7
CONCENTRATION OP ECONOMIC POWER 13811
Reported composite price and composite mill net yield — Continued
11926=100]
Com- Mill
posite Net
Price Yield
Com- Mill
posite Net
Price Yield
Com- Mill
posite Net
Price Yield
1939
Jul
96 6 1 91 4
Aug
Sep
Oct
Nov
The reported composite price index Ls based upon the Iron Age composite price, which is an arithmetic
average of the reported base prices of eight representative finished steel products.
The composite mill net yield index represents the amount, relative to that for 1926, received per ton by
U. 8. Steel Corporation subsidiaries (after freight) from sales of a representative constant assortment of all
principal products.
REPORTED COMPOSITE PRICE AND COMPOSITE MILL NET YIELD
1926 = 100
COMP
OSITE PRICE
"X
^
*s»«.
\
NTI
^
- r^f
rn
'"'•!»<
A
V
\
^
cT
"^
^,
(TOU
SUBSI
OSITE
TYIEL
S.S.C.
lARIES)
D
100 s
1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940
Base prices of steel, as Indicated by the composite price reported by
Iron Age, have shown considerable flexibility since 1926.
There has been even more fluctuation in the mill net yield, i. e., the amount
received per ton by the U. S. Steel Corporation subsidiaries on the various
products after deduction of cost of delivery. The mill net yield index
declined from an average of 100 in 1926 to a low of 73.5 in 1933. The in-
creases in prices in 1 937 were the result of Increased wag js and other costs.
Factors tending to lower mill net yields with respect to reported base
prices are principally (a) reductions from base price, (b) excess of actual
cost of delivery over freight added to base price in computing the deliv-
ered price, (c) quantity discounts and (d) deductions for quality, size, etc.
Factors tending to raise mill net yield with respect to reported base prices
are principally (a) extras for special finish, quality, size, heat treatment,
etc., and (b) extras for small quantity.
13812
CONCENTRATION OF ECONOMIC POWER
Reported composite price and composite mill net yield
[1926=1001
Com-
MUl
Net
Yield
Com-
posite
Price
Mill
Net
Yield
Com- Mill
posite Net
Price Yield
1912
Jan
1916
Jan
1920
Jan
61.6
60.5
60.2
61.9
62.9
63.5
65.0
67.1
69.5
71.7
73.4
74.5
57.9
57.3
57.5
57.6
58.1
68: 9
59.6
60.8
60.9
61.9
89.0
95.2
105.7
112.8
118.8
116.2
114.0
115.9
119.4
123.4
130.5
141.6
67.3
68.7
7! 0
74.2
77.3
79.2
82.3
84.0
86.0
88.2
90.8
152.7
152. 7
164.2
170.9
168.0
166.3
169.1
172.7
170.8
162.5
154.2
133.1
Feb
Feb
Feb
122 2
Mar
Mar
Mar
122 3
Apr
Apr .
Apr
123 5
^y :~
May : I":'
May .
124 0
Jun - -
Jun .
Jun ...
123.8
Jul
Jul ... .
Sep
Oct
cfft::::::::::::
Oct
Nov
Nov
Nov
125 8
Dec
Dec
Dec.
126 5
Jan
Jan
1913
1917
1921
Jan
76.6
76.3
77.1
77.3
74.6
72.9
72.0
70.2
67! 3
65.0
63.2
63.3
64.3
65.2
65.7
66.0
66.4
66.2
66.1
6.5.9
65.3
64.2
62.9
146.2
151.2
161.5
177.5
197.1
216.2
230.4
226.7
218.1
149.8
148.8
148.6
98.8
103.5
107.3
111.9
115.4
118.7
123.3
127.5
131.5
134.4
134.2
132.8
129.1
123.1
115.9
114.0
114.5
109.0
102.2
96.0
93.0
91.3
88.8
86.4
126 1
F^-
Feb
Feb "
126 5
Mm-" :
Mar
Mm-' ""
126 6
May'::::::::
May
May":':":"":"
Jul-
Jul
Aug
Jul
106.1
Aug..
Sep
Aug
98 7
Sep
Sep
94 6
Oct "
Oct
Oct
90.4
Nov
Nov . .
Nov
Dec
Dec
Dec
Jan
Jan
Feb.
Mar
1914
1918
1922
Jan..
62.7
63.8
63.6
62.6
61.5
60.4
60.5
62.5
63.5
62.5
60.3
59.0
61.2
60.8
61.2
61.1
60.5
59.9
59.0
59.2
59.4
59.6
59.5
153.3
153.3
153.3
153.3
153.3
153.3
153.3
153.3
153.3
153.3
153.3
149.5
139.3
137. 6
136.1
135.8
137.9
138.2
139.4
139.5
140.6
142.1
142.6
140.2
84.6
82.2
82.2
85.1
87.7
90.1
91.5
87.0
101.6
103.6
103.3
102.4
K.I
Feb..
Feb...
Mar
Mar
82 8
Apr
Anr
Apr
82 7
May"
May
May
82 9
Jun
Jun
Jun
Jul
Jill
Jul .
84.2
AUB
Sep
86.2
Oct
Oct .
Oct
87.3
Nov
Nov
Nov
88.6
Dec
Dec
Dec
90.0
Jan
Jan
1916
1919
1923
Jan
59.7
60.3
61.0
62.1
61.9
62.4
63.5
65.3
67.3
70.6
76.4
83.8
58.2
58.0
58.1
58.8
59.1
59.6
60.1
60.4
61.0
62.1
63.1
64.5
145.6
145.6
141.8
130.9
130.5
130.5
130.5
130.5
129.8
131.8
133.2
134.3
136.1
134.7
132.1
125.4
123.9
122.0
121.9
121. 5
121.2
121.5
120.8
121.6
103.7
109.5
115.5
119.9
119.7
118.6
118.6
us: 6
118.6
lis. 6
118.0
91.9
Feb
Feb
Feb .
Mar
Mar
Mar
96.0
Apr
triy
Apr
98 4
mi
Alay
Jun
100.4
Jun
Jun
101.9
Jul...
Jul
Jul
Aug
Sep
Sep
106.2
Oct
Oct
Oct
107 7
Nov
Nov
Nov
110.4
Deo
Dec ..
Dec
CONCENTRATION OF ECONOMIC POWER 13813
Reported composite price and composite mill net yield — Continued
[1926=100]
Com-
posite
Price
Mill
Net
Yield
118.6
117.5
115.0
112.0
100.2
108.0
105.8
103.6
101.8
101.0
101.6
103.6
112.6
112.7
113.0
112.6
111.0
108.3
107.3
105.3
104.2
102.0
101.4
101.2
104.1
104.6
104.7
102.0
100.6
99.6
99.5
98.7
97.7
98.3
99.8
100.6
101.9
102.3
102.9
103.0
102.3
100.7
100.2
100.3
99.7
99.7
99.5
99.6
100.3
99
100.0
100
100.3
99
10(1. 1
10(1
99.5
100
99.7
99
100.0
99
100.0
99
100.0
99
100.0
99
100.0
99
100.0
99
98.5
95.9
96.5
96.2
95.7
95.5
95.4
95.4
95.0
93.0
92.1
92.1
98.7
98.0
97.0
96.6
95.9
96.4
96.3
96.3
95.9
94.9
95.2
93.5
Jan..
Feb.
Mar.
Apr-
May
Jun.
Jul..
Aug.
Sep.
Oct.
Nov.
Dec.
Jan..
Feb.
Mar.
Apr-
May
Jun.
Jul-.
Aug-
Sep.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr-
May
Jun."
Jul..
Aug.
Sep.
Oct.
Nov.
Dec-
Feb.
Mar-
Apr.
May
Jun.
Jul..
Aug.
Sep.,
Oct.
Nov.
Dec.
Com-
posite
Price
MUl
Net
Yield
92.3
93.7
94.1
94.0
92! 6
94^2
94.5
92.9
92.4
92.7
92.9
93.9
94.7
94.7
94.7
96.0
96.2
96.6
96.2
95.6
95.4
94.9
94.7
95.2
94.2
94.2
93.9
94.3
94.2
94.3
95.0
95.4
94.5
94.3
94.3
94.0
93.4
92.8
92.7
90.6
88.8
87^2
86.3
85.9
85.6
85.4
84.8
92.4
91.6
91.2
89.9
86.6
86.0
85.0
85.4
82.2
85.6
83.2
85.6
82.3
85.3
81.8
85.3
81.4
84.8
80.4
84.4
79.9
84.0
79.8
84.0
81.9
84.0
80.0
83.8
81.3
82.5
80.2
Jan..
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep..
Oct.,
Nov.
Dec.
Jan..
Feb.
Mar.
Apr-
May
Jun.
Jul..
Aug-
Sep.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep.
Oct.
Nov.
Dec.
Com- Mill
poslte Net
Price Yield
81.4
80.9
80.6
78.5
77.8
78.fi
81.1
81.3
81.6
84.2
83.5
84.0
13814
CONCENTRATION OF ECONOMIC POWER
Reported composite price and composite mill net yield — Continued
[1926=100]
Com-
posite
Price
Mill
Net
Yield
Com-
Mill
Net
Yield
Com-
posite
Price
Mill
Net
Yield
1936
Mar. - --
1937
May
Jun..
Jul
Aug
Sep
1938
Jan
89.1
88.1
87.3
87.6
87.6
87.8
90.3
90.3
90.5
91.4
91.4
95.0
89.0
89.1
87.6
86.4
87.1
88.2
87.3
88.1
9o!o
90.6
106.2
108.5
108.5
108.5
108.5
108.5
108.5
108.5
108.5
108.5
93.3
95.8
98.0
99.8
101.6
101.9
103.4
105.7
104.8
105. 3
108.3
106.2
99.4
99.4
99.0
97.4
98.7
98.7
104.4
Feb
Apr
102.7
Mar
May
97.9
Jul
Oct
93.7
Sep
Nov
Dec
Jan,
Feb --.
Mar
M^ay:;::::;:;::
Jun
Jul -.
91.6
An?
Oct
92 2
Sep
Nov _
Oct
Dec
1939
Nov
Jan
Dec
1938
98.7
98.7
98.7
98.7
97.5
96.6
96.6
93.2
1937
108. 5 105. 4
95.8
95.1
97.1
91.4
92.3
108.5
108.5
108.5
105.1
105.9
104.3
94.8
Jan
Mar
92.1
91.4
The reported compositp price index is based upon the Iron Age composite price which is an arithmetic
average of the reported base prices of eight representative finished steel products.
The composite mill net yield index represents the amount per ton, relative to that for 1926, received by
U. S. Steel Corporation subsidiaries (after freight) on a representative constant assortment of all principal
products.
CONCENTRATION OF ECONOMIC POWER
13815
's ZD " "^ — ■- "^ s "" i
-^'-'o-0 0>-C •C-n?f->-
-Oco^aO oogg.
. . «J o S o ?--2 S o
-D=oJ3-§ :£8£-t®
I E fe s S.-^ »-^1 8.S
^^ s:^ll 2^ S i"^
® .1 .If ^ -o E « I ^ §
it:? o -"S S3-_£ E
-c"e-S| g-S. i~-» §
1^ :-o ii o.|S2f
6
lr|i .1
§S I
sag R
SU^SWnN XBQNI
13816
CONCENTRATION OF ECONOMIC POWER
Proportion of steel cost in price of finished product, 1938 — typical products of
portant steel consuming industries
J'roduot
Estimated
Price
EstlEnated
Cost of
Steel
Steel Cost
Remarks
Reinforced Concrete
Road.
Dairy Barn
$55,000
$2,600
$218
$4,000
$938,000
11.8«S
$730
$112
$33,400
$2,260
$400
$83
$7
$172
$55,077
1.22*
$77
$25
pi 2, 300
$898
0.7
3.2
3.4
4.3
6.9
10. 4
10.5
22.3
36.8
39.9
One mUe of twenty foot concrete highway
in Indiana. Steel cost is cost of 16,000
pounds of reinforcing bars.
Roof sizes 30' x 60'. Steel cost Is cost of
Electric Refrigerator. . .
galvanized sheets for roof only, and ex-
cludes other items such as nails, staples,
Six ou. ft. capacity refrigerator. Steel cost
is cost of 200 pounds of steel, principally
sheets, and excludes steel in motor and
refrigerating unit.
Apartment BuUding...
Can of Food Products-
cost is cost of 8,000 pounds of radiators,
boiler, water tahk, pipe, metal lath, nails,
gutters, conduit, etc.
SU story flre-proof, walk-up apartment
house in N. Y. Steel cost Is cost of fabri-
cated but unerected structural steel.
Average price of cans of tomatoes, peas and
com In N. Y. Steel cost Is cost of tin
plate.
Typical low-priced four door sedan de-
livered In N. Y. Steel cost is cost of
2,800 pounds of sheets, strip, pig iron,
bars, etc.
Average ef fifteen low-priced farm imple-
ments. Steel cost estimate is based on
average weight, and steel cost of $60 per
ton.
One mUe of new track, including ballast,
ties, rails and fastenings. Steel cost is
cost of raUs and fastenings.
Typical 60 ton capacity hopper car. Steel
Farm Implement-
Railroad Track-.
Freight Car
cost Is cost of 38,000 pounds of plates,
shapes, wheels, axles, etc.
Above data are necessarUy estimates but are believed to reflect, with reasonable accuracy, the propor-
tion of steel cost In the price of the finished product.
CONCENTRATION OF ECONOMIC POWER
13817
PROPORTION OF STEEL COST IN PRICE OF FINISHED PRODUCT
OTHER COSTS INCL PROFIT I 1 STEEL COST
REINF. CONCR. ROAD
DAIRY BARN
ELEC. REFRIGERATOR
FRAME HOUSE
APARTMENT BID'G
CAN OF FOOD PROD'S
AUTOMOBILE
FARM IMPLEMENT
RAILROAD TRACK
FREIGHT CAR
PRICE OF FINISHED PRODUCT = 100%
20% 40% 60%
A 10% REDUCTION IN THE PRICE OF STEEL CAN HAVE BUT RELATIVELY LIHLE
EFFECT ON THE PRICE OF MOST FINISHED PRODUCTS
The cost of steel is a relatively smjjl part of the price of most finished
products. For example, th"fe.csst of the steel in a low-priced automobile
averages about 10% of the delivered price to the ultimate consumer
Consequently, a 10% reduction in the cost of steel, if entirely passed
on to the consumer, would reduce the price only I %.
13818 CONCENTRATION OF ECONOMIC POWER
Automobile steel consumption and steel prices in United States
Year
.Automo-
bile Steel
Con-
sumption
(Tlious.
Or. Tons)
Automo-
bile Steel
Price
(Cents
per
Pound)
Compos-
ite Steel
Price
(Cents
per
Pound)
Year
Automo-
bile Steel
Con-
sumption
(Thous.
Or. Tons)
Automo-
bile Steel
Price
(Cents
per
Pound)
Compos-
ite Steel
Price
(Cents
per
Pound)
1923 .
4,182
2,981
4.886
5,486
4,895
6,963
6,545
4,406
3.73
3.45
3.04
2.99
2:67
2.69
2.44
2.697
2.505
2.334
2.315
2.202
2.165
2.209
2.048
1931
1932
1933 -
3,149
1,864
3,530
4,101
6,016
6,712
6,077
3,619
2.18
1.98
1.89
2.22
2.22
2.^
2.73
2.60
1 957
1924
1925
1.901
1.879
1926
1934
1927..
1928
1936
1929
1937
1930
1938
2 394
Source: Automobile steel consumption data are based on figures of Iron Age. Steel price figures are from
Iron Age, automobile steel price data representing average of prices of hot rolled strip and hot and cold rolled
sheets.
AUTOMOBILE STEEL CONSUMPTION AND STEEL PRICES
IN UNITED STATES
9
8
7
6
CO
1 '
4
u.
O
<o 3
z
o
=! 2
s
1
9
8
7
6
5 g
4 2
K
3 ^
<A
2 £
o
1
^TOMQB
LESl
EEL
^
^
^
v.
\
UNbU
/
\
/
N
s
/
\
\
^^
AU
TOMO
3ILE
\
i
y
^
=^
*==:!
===
EtLHKICL
=5=
^
^
...
:^
STEE
.PRI
E
r
^illilllsllilllll^
Automobile steel consumption and steel prices have little, If any, relation
to each other. Hence, factors other than steel prices must govern auto-
mobile steel consumption.
CONCENTRATION OF ECONOMIC POWER 13819
Automobile steel consumption and automobile production in United States
Year
Automobile
Steel Con-
sumption
(Thous.
Or. Tons)
Automobile
Production
(Thous.
Cars)
Year
Automobile
Steel Con-
sumption
(Thous.
Or. Tons)
Automobile
Production
(Thous.
Cars)
1923. .. .
4,182
2,981
51486
4,895
6,963
6,545
4,406
4,180
3,738
4,428
4,506
3,580
4,601
5,622
3,510
1931
3,149
1,864
3,530
4,101
6,016
6,712
6,977
3,619
2.4?2
1,431
1 986
1924
1932
1925
1933
1926
1934
2,870
4 120
1927
1935
1928
1936
4,616
5,016
2,655
1929 _...
1930
1938
Source: Automobile steel consumption data are based on figures of Iron Age. Automobile production
data are from Statistical Abstract of the United States, and Automobile Manufacturing Association.
AUTOMOBILE STEEL CONSUMPTION AND AUTOMOBILE PRODUCTION
IN UNITED STATES
, z
5 o
-1
i
7
6
CO
§ 5
:i 4
3
2
1
0
JTOMC
>OII t
ETC CI
/
<,
CONSUMPTION
(GMSS TONS)
/
"^
y
s
/
/
/
J
\
/'
^
V.
/
x
y
\
1
\
/
N
J
y']
/
\
K
/
/.
N
'^
PW
»ua
(CARS)
OH
.
s 1 1 i i i
§|gS||||g||2
Production of automobile steel is closely associated with and dependent
upon production of automobiles. Accordingly, the dsmand for automobile
steel is determined by factors which govern the demand for automobiles,
rather than by factors, such as price of automobile steel, within the control
~f *Ua <teel industry.
124401— 41-rpt. 2(j-
13820 CONCENTRATION OF ECONOMIC POWER
The Basing Point Method of Quoting Delivered Prices in the Steel
Industry
The basing point practice in the steerindustry is a simple method of quoting
delivered prices, which results in the competition of many geographically sepa-
rated steel producers at the markets for each of the diversified products of modern
steel mills. It is not a price-fixing medium nor does it result in high prices. It
does not stifle price competition but rather extends the benefits of such competition
to all consumers.
This basing point practice has evolved over a period of more than half a century
to meet fundamental economic conditions in the steel industry. Delivered prices
result from the buyer's need to know the cost to him of steel delivered at his plant,
since transportation charges from mill to consumer are often a substantial part of
the value at the place of consumption.
The producer of steel must take into consideration all of the elements of cost
involved, from the transportation of raw materials, through the processes of con-
verting such raw materials into steel products, to the final delivery of such prod-
ucts to the consumer. It requires more than four tons of raw materials to produce
one ton of finished steel. The location of facilities for producing pig iron and steel
ingots must be determined largely by the factor of raw material assembly costs.
This Hmits the location of blast furnaces and open hearth furnaces to a few areas
where the raw materials are readily available. In turn, the economies of integra-
tion cause the location of rolling mills near the steel producing units. Large well-
integrated mills, designed to supply the scattered markets of the entire country,
have been constructed in such areas. These mills produce many diversified prod-
ucts in order to utilize ingot capacity to the fullest extent and achieve low produc-
tion cost per unit. A modern integrated mill must serve more than its immediate
area; it must reac'h many of the important markets for its diversified products in
order to obtain an even flow of orders. Thus, concentration of production facili-
ties in a few areas and wide distribution of products is a rule in the steel industry
enforced by economic considerations. The result is competition at all consuming
points between several geographically separated producers.
.The demand for steel is subject to enormous fluctuations in the business cycle.
The capacity of the industry, including reserve capacity, is not more than sufl5cient
to supply the needs of the country during periods of high demand, such as 1929,
1937 and the present time. Less capacity would result in scarcity and high prices
during such periods. The problem of adjustment to the fluctuations of the busi-
ness cycle is solved in the most economical way. While the industry is constantly
constructing new facilities to incorporate technological advances, the older mills
which, although outmoded, have not served their full useful life, are retained in
reserve to meet the demand at high levels of consumption.
Most criticisms of the basing point method disregard entirely these fundamental
economic facts. The steel industry is often judged by criteria derived from ab-
stract theory, based upon imaginary conditions which cannot exist. Natural de-
viations from these criteria are arbitrarily assumed to be evils and are, without
demonstration, ascribed to the basing point method. Critics sometimes rest their
case solely upon bland assertions and rhetorical exaggeration. In many instances,
mere name-calling is resorted to. Thus, in the language of some critics, the prac-
tice of meeting competitive prices at a distance becomes "freight absorption' ; the
resulting difference in mill net returns becomes "price discrimination"; the result-
ing shipments from other than the mill nearest the destination becomes "cross-
hauling"; and .the realization of a competitive advantage due to superior geo-
graphical locat'uin becomes "phantom freight."
Competitive forces determine the pirices quoted at all destinations. To obtain
Dusiness in a market at a distance froih his mill, a producer must meet competitive
prices quoted by other producers nearer to such markets; he must pay the freight
necessary to transport the steel product to the consumer; and he will therefore
realize a lower mill net return than on sales to consumers nearer his mill. This
enables him to operate his mill at a lower unit cost and thus to sell to the nearby
consumer for less than he otherwise could.
There will always be some shipments of similar products past each other in
opposite directions unless competition between geographically separated pro-
ducert, is arbitrarily limited to the marginal territory between their mills. Even
under the uniform f. o., b. mill price system proposed by the Federal Trade Com-
mission, shipments would not always be made from the nearest mill. The alleged
economic waste resulting from cross-shipments must be balanced against the
countervailing advantages to the public of a competitive system, and also against
CONCENTRATION OF ECONOMIC POWER 13821
the economic losses which would follow from artificial limitation of marketing
territories.
If an isolated producer is located nearer than other producers to an important
market, he wiU be able to realize a higher miU net return. In so doing, he may
be merely taking proper advantage of his superior geographical location, or he
may need such higher return to compensate for his additional costs in assembling
and processing raw materials. He can obtain higher mill net returns than some
of his competitors either by announcing a higher price at his mill, or by merely
meeting the competitive delivered prices of other producers. . Characterizing tht
latter practice as the collection of freight charges which are not paid is a distortion
of the facts.
Transportation of steel products by water vehicles and trucks has received
attention unwarranted by its true importance, and significant factors in the
situation have been overlooked. The practical availability of each of these
mediums of transportktion is circumscribed by many inherent limitations.
The producer located so as to be able to transport some products by water has
an advantage over other producers not so located, which he is properly entitled
to realize by a higher mill net return. His advantage often lies merely in the
ability to reach markets from which rail freight rates would bar him. Where all
the circumstances warrant it, the advantage is passed on to consumers by lower
delivered prices. The producer's advantage, however, is one which may easily
turn into a disadvantage. If be gives one consumer the benefit of the saving
resulting from water transportation, he may soon have to make the same price to
aE consumers in the area and ship by rail, with freight disadvantages which will
lower his rpill net returns. Shipment by truck seldom involves an appreciable
freight saving, and often involves additional freight cost. The added expense
and inconvenience to the producer in truck shipments justify any additional
charges' made.
The proposed alternative to the basing point method is a uniform f. o. b. rnill
price system. The effects of this system would be extremely complex, and are
therefore largely unforeseeable. Its exponents propose it in the name of abstract
theory, and have outlined its characteristics and effects only with respect to the
elimination of supposed evils of the basing .point method. They have never
described the operation of the system nor analyzed its effects in relation- to the
economic facts of the steel industry.
The uniform f. o. b. mill price system is expected by its exponents to eliminate
high cost, inefficient and supposedly uneconomically located miUs and to break
up concentration of production facilities, by forcing the erection of small mills
in all parts of the country. Such results, even if they wo\iId be accomplished by
the system, would conflict with basic economic factors, and necessarily increase
present production and transportation costs.
The system is also expected..to increase existing competitton. This is to be
accomplished by the extraordinary means of arbitrarily limiting the competition
between miUs not adjacent to each other to marginal territory. Each mill, or
group of mills, would be restricted in distribution to a circumscribed area subject
to only slight possible variations in size. Each c'ustomer would be confined to a
single or a very few sources of supply. The capacities of mills would be limited
to the consumption in the prescribed territories, and any existing additional
capacity would have to be scrapped. Serious dislocations in the steel industry
and in industries dependent upon it would be inevitable.
'Under a uniform f. o. b. mill price system, local monopolies and high assembly
and production costs would displace the present widespread competition and low
costs. I
The following diagrams, prepared by United States Steel Corporation, are,
designed to clarify the operation of the basing point method of quoting delivered
prices in the steel industry by means of a few simple illustrations of typical
basic situations. While a good deal of mystery seems to have been made of
the basing point practice, it is actually simple and easily understandable. These
diagrams do not purport to give a complete picture of the price competition which
exists in the sale of steel products. In particular, it must not be concluded that
the prices of steel products at any consuming point are inflexibly determined by
the application of the basing point practice — competition between different
steel producers is keen .and often results in delivered prices at the consuming
point considerably lower than the delivered prices which would reuslt from the
application of the basing point method as illustrated by these diagrams.
October 30, 1939.
13822
CONCENTRATION OF ECONOMIC POWER
THE BASING POINT METHOD
Most steel products are sold on a Delivered Price basis.
Diagram 1 : How the Delivered Price is computed.
&
■0
DELIVERED PRICE
Base price is used herein In the sense of the announced price at the basing
point, wltnout freight or extras, and delivered price is used herein in the
sense of the price at the consuming point, not including extras.
If the base price at a basing point is $40 per ton and freight therefrom
to a consuming point Is $4 per ton. the delivered price at such consuming
point is $44 per ton.
Note: The bqse price o( $40, used In this and subsequent diagrams, is purely arbitrary
and is not to be talen as an actual price. Prices vary for different steel products.
CONCENTRATION OF ECONOMIC POWER
13823
THE BASING POINT METHOD
Diagram 2: Explanation of Freight Disadvantage and Freight Absorption.
Mill at (A) has lowest Base Price plus Freight to jx].
Mills at (B) and © are at a Freight Disadvantage;
to sell at [x] they must absorb Freight.
BASING /-TN BASE PRICE $40
POINT \^ FRT OISADV 2
T MILL NET 38
G>
BASE PRICE $40
FR-T OISADV 0
MILL NET 40
'con;
4f-
•<5)
BASE PRICE $40
FITT OISADV. 3
MILL NET 37
A, B and C are basing points, at each of which a base price of $40 per ton
is announced by the mills located there. X Is a consuming point. If the
freight rate from each basing point to X were added to the base price
at each basing point three delivered prices would result, $43. $45 and
$46 per ton, the delivered price from A being the lowest. However, a
consumer at X naturally will not pay more than $43 per ton, the lowest
delivered price quoted. Consequently, competition forces the mills at
B and C also to quote a delivered price of $43 per ton at X, which re-
sults In their mill net returns being reduced to $37 and $38 per ton, re-
spectively, whic^ amounts are below their base prices.
Since such reduction below the base price is necessitated by the freight
disadvantage of the mills at B and C. it Is often called "freight absorp-
tion" by the critics of the basing point practice. The inference Is that
such mills are paying a higher freight rate than they charge to the
customer. In reality, they are not charging any amount for freight, but
are quoting such a delivered price as Is necessary to meet the comfjetltlon
of the more favorably located mill at A, and paying the freight necessary
to make dellverv at the consuming point
13824
CONCENTRATION OF ECONOMIC POWER
THE BASING POINT METHOD
Diagram 3: Explanation of first type of Freight Advantage and so-called "Phantom Freight"
Mill at (A) has lowest Base Price plus Freight to [x].
Mill at © charges the same Delivered Price. Having a Freight
Advantage of $ 1 ^)ver (A), © realizes a Mill Net
$ 1 higher than (A). This $ 1 is so-called "Phantom Freight".
NON-BASING
POINT
NO BASE PRICE ©
fRT ADV. $1
MILL NET 41
&
CONSUMINr
I 1 POINT
►[x>
A and B are basing points, at each of which a base price of $40 per ton is
announced by the mills located there, c is not a basing point because
the mill at c sees fit not to announce bf>se prices at c, but merely meets
the competitive delivered prices of othJrjriHs. The mill at c at any time
In its discretion may decide to make c a basing point. At consuming
point X, the lowest delivered price from a basing point mil! is $44 per
ton, computed upon $40 base price plus $4 freight from A. If the mill at
B sells at X, it is at a freight disadvantage of $ I per ton and, accordingly,
its mill net return will be $1 per ton less than its base price.
The non-basing point mill at c, however, has a freight advantage of $ I per
ton over the mill at A in selling at X. By meeting the competitive delivered
price of A at X, the mill at c receives a mill net return of $41, or $1 more
than the mill at A which may explain why it has not decided to become
a basing point. This amount has been characterized by critics of the
basing point practice as "phantom freight", the inference being that the
mill at c charges the customer a higher freight rate than it pays. In reality,
the mill at c merely names a delivered price at X which permits it to
realize the benefit of its freight advantage due to superior geographical
location. It would realize this same advantage if it announced a base
price at c of $41 per ton.
Smce there are today very few non-basing point mills, very little freight
advantage of this type is now realized in the steel industry.
CONCENTRATION OF ECONOMIC POWER 13825
THE BASING POINT METHOD
Diagram 4: Explanation of second type of Freight Advantage and so-called "Phantom Freight"
Mill at (D has lowest Base Price plus Rail Freight to [x]
Mill at (a) charges the same Delivered Price.
When mill at (A) ships by water it has a Freight Advantage
of $1 and realizes a Mill Net $1 above its Base Price.
This $1 is so-called "Phantom Freight".
BASE PRICE $40'^^============================^DEL1VEREO BASE PRICE
FRT ADV. (WATER) 1 FREIGHT (WATER) $2 price $43 MILL NET
MILL NET $41
Note: When mill at (a) ships by rail it is at a Freight Disadvantage
of $1 and realizes a Mill Net $1 below its Base Price.
A and B are basing points, at each of which a base price of $40 per ton
is announced by the mills located there. X is a consuming point, at vvhich
the lowest delivered price, using rail freight rates, Is $43 per ton. If the
mill at A sells at X, and ships by rail, It Is at a freight disadvantage of $1
per ton. If it meets the delivered price of B at X and ships by rail, it will
realize $ I per ton less than Its base price.
If the mill at A ships by water to X, It has a freight advantage of $1 per
ton. If it meets the delivered price of B at X and ships by water, the mill
at A realizes the benefrt,of Its water freight advantage and obtains a
mill net return $1 per tor higher than Its base price. This advantage, when
thus realized, Is also characterized by critics of the basing point method
as "phantom freight".
In reality, steel mllfs realize very little freight advantage or "phantom
freight" of this type.
13826
CONCENTRATION OF ECONOMIC POWER
THE BASING POINT METHOD
Diagrams: Determination of. Boundary between Natural Market Territories
The Boundary dividing the Natural Market Territories of mills at
Basing Points (t) and (B) is the line 0-0 connecting the
points at which the Delivered Prices from (A) and (§) are equal.
j^*^' DEUVEREO
7
CONSUMING X-a^
/C-^
\y>
BASING / <' DEUVEREO
POINT /^^^ PRICE J45
(X^-l iRilGHJJS _^ -
BASE PRICE
$40
0
BOUNDARY
E»45|
""^T-'CON
FREIGHT $5
.---®
A and B are basing points, at each of which a base price of $40 per ton is
announced by the mills located there. X. Y and Z are consuming points,
each of which is equi-distant freightwise from A and B. The line 0-0 con-
necting these points is the boundary between the natural market territories
of the mills at A and B. As thus defined, natural market territory refers to
the area in which a mill can sell at the competitive delivered price without
realizing less than its base price. In sales at X, Y and Z and at points on its
side of the line 0-0, either the mill at A or the mill at B realizes a mill net
return equal to its base price.
CONCENTRATION OF lOCONOMIC FOWEll 13827
THE BASING POINT METHOD
Diagram 6: How shipping beyond Boundary of Natural Market Territory reduces Mill Net.
When mill at (§) sells to [x], its Mill Net is $40.
When mill at (§) sells to [y], its Mill Net is only $37 because:
1. Freight is $2 higher.
2. Delivered Price is $1 lower.
I-
»•*
/I
pSNT(J).___JiE!GHIiL__J2'!lV[x}*- i''l'£.fi2.~~rir^
DELIVERED
PRICE $41 _
/I
BASING -
BASING
•~... POINT
BASE PRICE $40
MILL NET (XorY)
$40
BASE PRICE $40
MILL NET (X) 40
MILL NET lY) 37
0
BOUNDARY
A and B are basing points, at each of which a base price of $40 per ton Is
announced by the mills located there. At consuming point X, located on
the boundary 0-0 between the natural market territories of the mills at A
and B. the delivered price Is $42 per ton, calculated with reference to the
freight rate from either basing point. At consuming point Y, the delivered
price is $41 per ton. equivalent to the base price at A of $40 per ton, plus
$ I freight from A to Y.
When the mill at B sells at X, Its mill net return equals its base price, but
when It sells at Y, its mill net return is reduced by the effect of two factors.
First, the delivered price Is $ I per ton less than at X. Second, the freight
rate from B to Y is $4 per ton, or $2 more than from B to X. Thus, the mill
net return of ttie mill at B Is reduced $3 per ton below its base price.
It may be advisable for the mill at B to sell at Y and realize a mill net
return $3 per ton below its base price In order to obtain a higher operat-
ing rate and thus secure a lower average unit production cost.
13828
CONCENTRATION OF ECONOMIC POWER
THE BASING POINT METHOD
Diagram?: Non-basing Point Mill.
Mills at Basing Points (a) and (f) realize full Base Pnces on sales in their
respective Natural Market Territories.
Non-basing Point mill at © has no Base Price and meets the Delivered Prices
of (A) and (§) when it sells in their respective Natural Market Territories.
(
NON-
BASING
POINT
MILL
©
)
BASING
POINT
©
BASE PRICE
$40
BASING
POINT
®
BASE PRICE
$40
0
BOUNDARY
A and B are basing points, at each of which a base price of $40 per ton is
announced by the nnilis located there, c is not a basing point, because
the mill at c sees fit not to announce base prices at c, but merely meets
the competitive delivered prices of other mills. The line 0-0 is the boundary
of the natural market territories of the mills at A and B. Since the mill at
c announces no base price at c, with which its mill net retums may be
compared, it has no natural market territory in the sense in which that term
Is used in these diagrams. Either the mill at A or the mill at B can sell at
points nearer c than A or B, respectively, without reducing its mill net
return below its base price, as long as the mill at c follows its practice of
meeting competitive delivered prices of other mills.
CONCENTRATION OF ECONOMIC POWER 13829
THE BASING POINT METHOD
Diagram 8: Effect of naming new Basing Point
After (C) becomes a Basing Point the Boundary 00 between (A) and (f)
ceases to be significant
Mill at (C) then has a Natural Market Territory, bounded by NN and N'N', in
which it establishes lower Delivered Prices than (A) or (|)
To sell in this territory, mills at Basing Points (A) and (B) must now absorb freight
©
BASE PRICE
»40
BOUNDARY BOUNDARY BOUNDARY
When the mill at C decides to announce a base price at C and C thus
beconnes a basing point, the line N-N beconnes the boundary between
the natural market territories of the mills at A and C and the line N'-N'
becomes the boundary between jhe natural market territories of the mills
at B and C. The line 0-0, which marked the former boundary between the
natural market territories of the mills at A and B, now ceases to be signifi-
cant, because the mill net returns of the mills at A and B are reduced on
sales to any point between N-N and N'-N', respectively, regardless of
whether or not such point is beyond the line 0-0. The natural market terri-
tory of the mill at C has been carved out of the former natural market
territories of the mills 'at A and B.
13830
CONCENTRATION OF ECONOMIC POWER
THE BASING POINT METHOD
Diagram 9: Illustration of Cross-hauling.
Products shipped from (A) to [Y] go past products shipped from (§) to [Y|
This involves Cross-hauling only if:
1. The products shipped are identical.
2. Shipments occur at substantially the same time
CONSUMING
POINT
FREIGHT $6
BASING
POINT
/CN BASE PRICE
BASING rK^
,/'
POINT \^
BASE PRICE
$40
FREIGHT J6
CONSUMING
POINT
A and B are basing points, at each of which a base price of $40 per ton is
announced by the mills located there. X and Y are consuming points, at
each of which the lowest delivered price calculated with reference to the
nearest basing point, is $42 per ton. When the mill at A sells at Y or when
the mill at B sells at X, In either case meeting the competitive delivered
price at the consuming point, the mill net return Is reduced $4 per ton
below the base price. When the mill at A ships to Y and the mill at B ships
to X. the shipments cross each other, in a broad sense, and this is said
by critics of the basing point practice to constitute "cross-hauling". This
is a very controversial term. Under a proper interpretation of the word,
there is no cross-hauling unless the products shipped fro/n A to Y and from
B to X are Identical and unless the shipments occur at substantially the
sanoe time. "Cross hauling" Is the necessary result of competition In the
steel industry.
CONCENTRATION OF ECONOMIC POWER
13831
THE BASING POINT METHOD
Diagram 10: Effect of Basing Point price differential (Supplementary to Diagram 3).
In selling at [x] or any point up to the Boundary 0-0 of its Market Territory, mill at (§)
realizes its higher Base Price and freight added equals freight paid.
In selling at |Y|, mill at (C) realizes less than its Base Price, although:
1. It realizes more than mill at (A).
2. Freight added exceeds freight paid. o
BASE PRICE $40
BASE PRICE $45
0
BOUNDARY
A Is a basir>g point, at which a base price of $40 per ton is announced by the mill located there. C is a
new basing point at which a base price of $45 per ton has recently been announced by the nnill located
there. X and Y are consuming points at which oelivered prices of i46.50 and $46 per ton, respectively,
are quoted.
Before C became a basing point, the mill at C merely met the delivered price of $46.50 quoted at X
by the basing point mill at A, and realized a mill net return of $45 per ton, which was $5 more than that
of the mill at A. This amount is characterized by critics of the basing point method as "phantom freight."
When C becomes a basing point, with the announcement of a base price of $45 per ton by the mill
at C, the price at X is unchanged, since the combination of base price at C plus freight from C to X
equals the delivered price computed with reference to A. The mill at C still realizes a mill net return of
$45 per ton, which is $5 higher than that of the mill at A. This" amount can no longer be characterized
as "phantom freight."
Lllcewlse, before C became a basing point, the mill at C merely met the delivered price of $46 Quoted
at Y by the mill at A, and received a mill net return of $44 per ton, which was $4 more than that of
the mill at A. When C becomes a basing point, with a base price of $45 per ton, the lowest delivered
price at Y is still the base price at A plus freight from A to Y. The mill at C still receives a mill net return
of $44 per ton. or $4 more than that of the mill at A. Although the mill net return of the mill at C is now
$ I lower than Its base price. It may still be charged with collecting "phantom freight," since the freight
paid (C to Y) is less than the freight used (A to Y) in computing the delivered price at Y.
Actually, the higher mill net returns realized by the mill at C, both before and after C is named a bas-
ing point, represent the proper realization by the mill at C of Its superior geographical location with
respect to sales at X and Y. The mill at C may need such higher returns to meet higher raw material
assembly or production costs.
13832
CONCENTRATION OF ECONOMIC POWER
THE BASING POINT METHOD
Diagram 11: Freight Disadvantage reduced by Water Shipment (Supplementary to Diagram 4).
Mill at (a) has $2.50 Freight Disadvantage when shipping to [x]
by rail, and Mill Net, js reduced $2.50
Mill at (A) has only $.75 Freight Disadvantage when shipping to [x]
by water, and Mill Net is reduced only $.75
DELIVERED PRICE $42
BASE PRICE $40
A and B are basing points, at each of which a base pn.e of $40 per ton
is announced by the mills located there. X is a consuming point at which
the lowest delivered price is $42 per ton, calculated witti reference to the
freight rate from B. If the mill at A sells at X, and ships by rail, it is at a
freight disadvantage of $2.50 per ton, and realizes $2.50 per ton less than
its base price. If the mill at A ships by water to X, it is still at a freight
disadvantage of $.75 per ton and realizes $.75 per ton less than its base
price.
The advantage of the water shipment merely reduces the freight disadvan-
tage of the mill at A from $2.50 per ton on rail shipments to $.75 per ton,
when shipped by water.
-^
MARKET TERRITORIES OF MAJOR MILLS
PRODUCING STEEL SHEETS
Assuming Each Such Mill Adopts A Uniform P.O. B. Mill
'^rice Equal To Prevailing Base Prices
LACKAWANNA
PITTSBURGH
6- VANOERGRIFT
WEIRTON, BEECH 80TT0«
6. STEUBENVILLE
YOUNSSTOWN, CAMPBELL
6. M' DONALD
WARREN
CLEVELAND
CANTON. MASSILLON.
PORTSMOUTH
ASHLAND
MIDDLETOWN
GRANITE CITY
CHICAGO
GARY, INDIANA HARBOR
MONROE
DETROIT
124491 — »0— pt. 28 (Face p.
LEGEND
A APOLLO
BB BEECHBOTTOM
Br BRACKENRID6E
Bu BUTLER
Ci CANTON
CI CLEVELAND
0 ORAVOSBURG
L LACKAWANNA
M MASSILLON
N MILES
P PITTSBURGH
SP SPARROWS POINT
S STEUBENVILLE
V VANDERSRIFT
W. WARREN
We WEIRTON
Y YOUNGSTOWN
o- McDonald
DETAILED A\AP OF COUNTIES
IN WESTERN PENNSYLVANIA AND EASTERN OHIO
Showing towns controlled by mills producing steel sheets
each such mill adopts a uniform F.O.6. mill price
equal to prevailing base price.
VHVi\ — 40— pt. 2B (Face p. 13833) No. 2
CONCENTRATION OF ECONOMIC POWER
13833
Some Effects of Proiposed Uniform- F. O. B. Mill Price System
A uniform f. o. b. mill price system would require each mill to name a price at
the mill applicable to aU sales. The combination of mill price and transporta-
tion charges from mill to destination would determine the delivered cost of steel
to the consumer. Generally, the delivered cost at any destination would vary
as between different producing mills, and the consumer naturally would buy
from the mill offering the lowest delivered cost, i. e., the mill having the lowest
combination of mill price and freight from mill to the particular destination. A
mill could sell only in the area in which the combination of its mill price and
freight from mill to destination would be the lowest.
MAP NO. 1 MARKET TERRITORIES OF MAJOR MILLS PRODUCING STEEL SHEETS
Map No. 1 illustrates some immediate consequences of the introduction of the
proposed system. Seventeen major sheet producing points are shown, with sell-
ing territories to which mills at such points would be restricted in the sale of
sheets under the proposed system, assuming present freight rates and mill prices
equal to present base prices at nearest basing points. (This results in equal mill
prices at all mills, except at Detroit and Granite City, where the mill price is $2
higher, and at Monroe, where the mill price is $3 higher.) On this map, terri-
tories are determined by counties according to freight rates to one or two key
towns, and mills adjacent to each other are considered as a single producing
point.
Local monopolies are the rule; competitive areas the exception. AU important
sheet markets are in the monopoly area of a single producing point, or a produc-
ing point group such as Chicago-Gary. Sheet consumption is slight in areas of
competition.
Size of monopoly areas depends generally upon the proximity of other m^ills.
Sparrows Point and Lackawanna together have monopolies of all North Atlantic
seaboard markets. Chicago-Gary has a monopoly of Wisconsin, Northern Il-
linois, Iiforthern Indiana, and Western Michigan. In contrast. Western Penn-
sylvania and Eastern Ohio mills, having large capacities, are restricted to small
areas.
Territorial allocation is extremely arbitrary. Warren and Youngstown mills,
for example, each have monopolies in their home counties; they share territories
to the north of Warren, but Warren, passing Youngstown, reaches territory
south of Youngstown, which Youngstown cannot reach. Chicago-Gary mills
share most territory which either can reach, including the home county of each ;
but Chicago, passing Gary, has a monopoly in certain Indiana counties; and
Gary, after passing exclusive Chicago territory, has a monopoly in many other
Indiana counties.
Note.— This map necessarily pictures the situation at a single relative price level. Differences in rela-
tive price levels would change market territories of mills, but would not eliminate the pattern of local
monopoly and fixed territories here shown.
A mill in close proximity to other mills would tend to swallow up their entire market areas in seeking to
widen its own market area by lowering its mill price.
At any relative level of their prices, two mills could generally compete in only one important market,
since automatically the price of the first would exclude the second, or the price of the second would exclude
the first, from every other such market.
MAP NO. 2 DETAILED MAP OF COUNTIES IN WESTERN PENNSYLVANIA AND EASTERN
OHIO
Map No. 2_shows miscroscopically a section of the area sliown on Map No. 1,
and further illustrates in detail some immediate consequences of the introduction
of a uniform f. o. b. mill price systern. On Map No. 1, selling territories are
determined by county. On Map No. '2, each town is separately marked with
the symbol of the mill or mills which could sell in such towns under the proposed
system, assuming mill prices equal to present base prices at nearest basing points,
and present freight rates. The closer scrutiny thus afforded indicates that the
problem is more complex than Map No. 1 shows.
Local monopolies are even more striking on Map No. 2. In many areas shown
on Map No. 1 as enjoying competition from more than one producing point, there
are actually only a few towns where competition occurs. Many towns are ac-
cessible to only one mill. For example, Columbiana County, Ohio, is shown on
Map No. 1 as an area of competition between Warren mills (including Niles)
and Weirton-Beechbottom-SteubenviUe mills. Actually, as shown on Map No.
2, three towns are accessible to Niles only, one town to Steubenville only^ and
13834 CONCENTRATION OF ECONOMIC POWER
one town to Canton only. Similarly, Westmoreland County, Pennsylvania, is
shown on Map No. 1 as an area of competition between the Pittsburgh group of
mills and the Butler-Vandergrift-Apollo mills. Actually, as Map No. 2 indicates,
the Dravosburg mill has a monopoly in six towns, and the Vandergrift mill in
another, while Dravosburg and Apollo compete in one town, and Pittsburgh,
Brackenridge, and Vandergrift compete in another.
Mills located close to each other are shown as groups on Map No. 1, and thus
competition between them throughout the group territory is indicated on such
map. The separate detailed consideration of each mill and each town on Map
No. 2 shows local monopolies and arbitrary restrictions of selling points even in
such territories. For example, Cambria County, Pennsylvania, is in the joint
territory of Butler-Vandergrift-Apollo on Map No. 1. Actually, as shown on
Map No. 2, Apollo has a monopoly in five towns, while Apollo and Vandergrift
compete in three others. In Weirton-Beechbottom-Steubenville territory:
Beechbottom has a monopoly of all towns in Belmont County, Ohio; in Jeffer-
son County, Ohio, Steubenville a monopoly in two towns, and Canton in another,
while Steubenville and Weirton compete in one town.
Note.— While this map is necessarily based on a single relative level of prices at different mills, it is not
believed that the pattern shown on the map would greatly change at any pdssible relative level. Prices
at adjacent mills would be identical, since each would be oblieed to meet another's price reductions im-
mediately, or lose its entire selling territory.
Composite mill net yield and cost per weighted ton shipped, U. S. Steel Corporation
and subsidiaries
[1926=100]
Year
Composite Mill Net Yield
Year
Jan.
Feb.
Mar.
Apr.
^ay
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
1926....
99.8
100.0
99.6
100.2
100.1
99.8
99.8
99.5
99.9
99.6
99.9
99.8
1926
1927....
98 7
98.0
97.0
96.6
95.9
96.4
96.3
96.3
95.9
94.9
95.2
93. 5
1927
1?28._.-
93.4
93.4
93.3
93.4
94.3
93.8
92.9
92.4
92.7
92.9
93.9
93.7
1928
1929....
94.2
94.2
93.9
94.3
94.2
94.3
95.0
95.4
94.5
94.3
94.3
94.0
1929
1930....
92.4
91.6
91.2
89.9
88.9
88.0
86.6
86.0
85.0
83.7
83.3
82.0
1930
1931.-..
82.2
83.2
82.3
81.8
81.4
80.4
79.9
79.8
81.9
80.0
81.3
80.2
1931
1932 ...
78.6
79.1
79.3
78.7
77.7
79.2
79.5
79.3
79.0
78.8
78.2
77.9
1932
1933...
77.0
76.0
76.6
75.0
74.5
74.6
73. 5 •
75.0
77.2
79.4
82.6
83.5
1933
1934 ...
87.1
88 1
87.4
87.1
88 5
87.4
91.8
92.9
91.9
93.3
92.5
89.9
1934
1935 ...
92.1
92.0
91.9
91.9
92.0
91.2
90.5
90.8
90.0
89.6
88.8
89.6
1935
1936....
89.0
89.1
87.6
86.4
87.1
88.2
87.3
88.1
88.8
89.6
90.0
90.6
1936
1937 ...
91.4
92.3
95.8
98.0
99.8
101.6
101.9
103.4
105.7
104.8
105.8
m7
1938 ...
105.4
105.1
105.9
104.3
104.4
102.7
97.9
96.2
95.9
93.7
91.6
92.2
i9;w
1939....
93.2
94.1
95.8
95.1
94.8
92.1
91.4
91.4
91.4
92.2
93.0
The composite mill net yield index represents the amount, relative to that for 1926, received per ton by
II. S. Steel Corporation subsidiaries (after freight) from sales of a representative constant assortment of all
principal products.
Cost per Weighted
Ton Shipped
Year
Cost per Weighted
Ton Shipped
Year
Actual
Cost
Estimated
Cost at
1926
Volume
Actual
Cost
Estimated
Cost at
1926
Volume
1926
100.0
103.6
98.3
91.8
96.9
105.6
129.6
100.0
101.1
97.2
92,0
92.8
92.2
91.7
1933
1934
105.8
115.0
112.7
105.9
108.7
124. 5
85.1
92.0
1935
96.6
1936
99.6
1937..
106.3
1938
107.6
1932
Actual cost per weighted ton shipped is total cost, exclusive of bond interest. Federal income taxes, mis-
cellaneous non-operating income and expense, and of inter-company items, for all subsidiaries of U. S. Steel
Coriioration, divided by the number of weighted tons shipped. Weighted tonnages are actual tonnages,
adjusted for change in proportions of high and low cost products and for the equivalent tonnage of average
cost rolled and finished steel products represented by products other than steel. The cost of operations
not related to the production of steel is included in total cost, but since such cost is a small percentage of
the total and since the other operations tend to expand and contract with the volume of steel production,
the relative change in the total cost per weighted ton many be considered fairly indicative of the change in
tlie cost of producing steel.
Estimated cost if 1920 volume maintained is the actual cost per weighted ton shipped adjusted to 1920
volume on the assumption the.t the percentage change in the average cost per ton as the rseult of a given
change in volume would have been the .same in each of the re^spective years as it is estiraateii to have been
under 1938 conditions.
CONCENTRATION OF ECONOMIC POWER
13835
COMPOSITE MILL NET YIELD AND COST PER WEIGHTED TON SHIPPED
U. S. STEEL CORPORATION AND SUBSIDIARIES
1926 = 100
A
/
/
, AC
UALC
OST
/
V
^-v-
Y?
K
Q
■<^
x^
,__
ESTI^
AT -t
ATEO COST
evOLUM^^
^
I
\
^
r
^ft»
Yp
"OSITE
TYIEL
V^
1
^v^
^y
IILLB
D
1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940
150
140
130
120 ^
110 S
s
100 i
90 2
a
z
80 -
70
60
50
The composite milt net yield index, which is indicofive of the level of steel
prices, hos generally been lower thon the index of octuol costs per ton
since 1926. This is true even if the effect on costs of chonges in the Operoting
rate is eliminated, as shown by the index of the estimoted costs per ton if
1926 volume hod obtained throughout the period. In the base year 1926,
with the various subsidiaries operoting ot on overoge rote of 89%, U. S.
Stjel Corporation realized 6.2% on its investment.
124491— 41— pt. 26
13836
CONCENTRATION OF ECONOMIC POWER
Relation of mill net yield to reported base price, U. S. Steel . Corporation Subsidies,
February 1939
Cents per Pound
Item
H'y Struc-
tural
Shapes
(at Pitts-
burgh)
H'y Struc-
tural
Shapes
(at
Chicago)
Plates (at
Pitts-
burgh)
Cold
Rolled
Sheets
2.100
.145
2.100
.099
2.100
.154
3,200
Extras
.041
Total - - - -- -
2.245
2.199
2.254
3.241
Mill Net Yield (Incl Extras) - ...
2.062
.122
2.106
.035
2.061
.119
2.908
Freight Absorption ..-- - --
.145
Total - - - . -
2.184
. .061
2.141
.058
2.180
.074
3.053
Price Reductions
.188
Reported Base Price -
2.100
2.062
2.100
2.106
2.100
2.061
3.200
Mill Net Yield (Incl Extras)
2.908
.038
•.006
.039
.292
•Mill net yield is in excess of reported base price.
Reported base prices are as reported by Iron Age.
Other data are frf^m U. S. Steel Corporation subsidiaries' reports to T. N. E. C. in answer to Questionnaire
Form B, entitled "Distribution and Pricing of Selected Steel Products."
Data on heavy structural shapes and plates at Pittsburgh and Chicago are for shipments from Homestead
and South Chicago plants, respectively. Data on cold rolled sheets are for shipments from Irvin and Gary
plants.
Freight absorption is adjusted for basing point price differentials.
CONCENTRATION OF ECONOMIC POWER
13837
QNnOd a3d SiN30
oi
o in o in o 1
oc s
. c. ^ ^ in o 1
Q. 9
P
%%
LlJ §
ili
a
£
IJ
!■
pm
%
2a
1
5
ii
1
rdco
Q '
2=i
QNnOd a3d SiN30
QNnOd aSd SiN30
1; CJ O
aNODd a3d SiN30
11
If
8 J
If
d °
8 §
5.1
2 c
II
II
II
«5 -£
1 i 2|
S 9
i
J
s
.9 S.
3i
I -
B =
1%
§ i
§1
s -si
9- ~
Is
111
- 5 ^
« ^
It
5i § i
S ° -i:
sis
13838 CONCENTRATION OF ECONOMIC POWER
Average delivered price and freight absorption, February 1939 shipments of selected
steel products — U. S. Steel Corporation subsidiaries
Dollars per
Net Ton
Per Cent of
Delivered
Price
Db.ivered Price (Incl. Extras)....
Freight Absorption (Unadjusted)
Freight Absorption (Adjusted)...
65.00
1.99
1.33
100.0
3.6
2.4
Data are from U. 8. Steel Corporation subsidiaries' reports to T. N. E. C, In answer to Questionnaire
Form B, entitled "Distribution and Pricing of Selected Steel Products."
Delivered price is "Total Invoiced Delivered Value," as shown on Form B.
Unadjusted freight absorption is difference between "Actual Freight Paid or Allowed on Shipments
from Mill to Destination" and "Freight Charges Added to Base Prices to Arrive at Invoiced Value," as
shown on Form B.
Adjusted freight absorption (as defined in Department of Justice "Supplement to Form B Tables,"
discussion of Table 9) is such unadjusted freight absorption decreased by the amounts by which the basing
point prices applicable on certain sales were greate- than the base prices at other basing points.
CONCENTRATION OF ECONOMIC POWER 13839
AVERAGE DELIVERED PRICE & FREIGHT ABSORPTION
U. S. STEEL CORPORATION SUBSIDIARIES
FEBRUARY 1939 SHIPMENTS OF SELECTED STEEL PRODUCTS
DOLLARS PER
NET TON
60
100.0%
DELIVERED FREIGHT FREIGHT
PRICE ABSORPTION ABSORPTION
(UNADJUSTED) (ADJUSTED)
Source: Antwei to T. N. E. C Queaionnait'
The unadjusted freight absorption (i.e., the difference between "freight
added" and "freight paid") per net ton on the domestic shipments of
selected steel products from selected mills of U. S. Steel Corporation sub-
sidiaries made during February 1939 averaged $1.99 a ton or 3.6% of
the average delivered price of $55 per ton, including extras. The adjusted
freight absorption (i.e., the unadjusted figure minus basing point price
differentials) averaged $1.33 a ton or 2.4% of the average delivered price.
13840 CONCENTRATION OF ECONOMIC POWER
Average delivered price and freight absorption, February 1939 shipments of selected
steel products — selected producing companies
Item
Dollars per
Net Ton
Per Cent of
DeUvered
Price
Delivered Price (Incl Extras)
56.06
1.77
1.16
100.0
Freight Absorption (Unadjusted)
3 2
Freight Absorption (Adjusted)
2.1
Data are from Department of Justice Summary compiled from T. N. E. C. Questionnaire Form B,
entitled "Distribution and Pricing of Selected Steel Products."
Delivered price is "Total Invoice Delivered Value" as shown on Form B.
Unadjusted freight absorption is difference between "Actual Freight Paid or Allowed on Shipments
from Mill to Destination" and "Freight Charges Added to Base Prices to Arrive at Invoiced Value," as
shown on Form B.
Adjusted freight absorption (as defined in Department of Justice "Supplement to Form B Tables,"
discussion of Table 9) is such unadjusted freight absorption decreased by the amounts by which the basing
point prices applicable on certain sales were greater than the base prices at other basing points.
CONCENTRATION OF ECONOMIC POWER
13841
AVERAGE DELIVERED PRICE & FREIGHT ABSORPTION
SELECTED PRODUCING COMPANIES
FEBRUARY 1939 SHIPMENTS OF SELECTED STEEL PRODUCTS
DOLLARS PER
NET TON
60
50
40
30
20
10
3.2%
2.156
DELIVERED FREIGHT FREIGHT
PRICE ABSORPTION ABSORPTION
(UNADJUSTED) (ADJUSTED)
Source: Department of Justice Summary
Compiled from T. N. E. C Questionnaire Form B.
The unadjusted freight absorption (i.e., the difference between "freight
added" and "freight paid") per net ton on domestic shipments of selected
steel products from selected mills of selected producing companies mode
during February 1939 averaged $1.77 a ton or 3.2% bf the average
delivered price of $55.06 per ton, including/ extras. The adjusted freight
absorption (i.e., the unadjusted figure minus basing point price difPerentiais)
averaged $1.16 or 2.1% of the average delivered price.
13842
CONCENTRATION OF ECONOMIC POWER
Breakdown of average delivered price, February 19S9 shipments of selected steel
products — U. S. Steel Corporation subsidiaries
Item
Dollars per
Net Ton
Per Cent of
Delivered
Price
66.00
6.71
49.29
3.72
1.99
0.66
1.33
100.0
Freight Paid
10 4
Mill Net Yield (Includtng Ertras)
80 6
Freight Added
6.8
Freight Absorption (Unadjusted)
3.6
Basing Point Price Dlflerentlals
1.2
Freight Absorption (Adjusted)
2.4
Data are from U. 8. Steel Corporation subsidiaries' report to T. N. E. C. In answer to Questionnaire
Form B, entitled "Distribution and Pricing of Selected Steel Products."
Delivered price Is "Total Invoiced Delivered Value" as shown on Form B.
Freight paid is the "Actual Freight Paid or Allowed on Shipments from Mill to Destination," as shown
on Form B.
Mill net yield Is delivered price less freight paid.
Freight added is "Freight Charges Added to Base Prices to Arrive at Invoiced Value," as shown on
Form B.
Unadjusted freight absorption Is difference between such freight ijaid and such freight added.
Basing point price dlflerentlals (ta defined In Department of Justice "Supplement to Form B Tables,"
discussion of Table 9) are amounts by which basing point prices applicable on certain sales were greater
than the base prices at other basing points.
Adjusted freight absorption (as defined In Department of Justice "Supplement to Form B Tables,"
discussion of Table 9) is diSerence between freight absorption (unadjusted) and such basing point price
differentials.
CONCENTRATION OF ECONOMIC POWER
13843
BREAKDOWN OF AVERAGE DELIVERED PRICE
U. S. STEEL CORPORATION SUBSIDIARIES
FEBRUARY 1939 SHIPMENTS OF SELECTED STEEL PRODUCTS
DOLURS PER
NET TON
60
50
40-
30
"DELIVERED"
PRICE
$55.00
20-
10-
FRT
PAID
$5.71
MILL
NET
YIELD
$4929
msi
PRICE
/DIFFERENTIALS $0.66 I FRT ABSORPTION
/ FRT ABSORPTION f (UNADJUSTED) $ 1.99
►^(ADJUSTED) $1.33
FRT ADDED $3.72
Source: Answer to T.N.E.G. Questionnaire Form
The "Freight Charges Added to Base Prices to Arrive at invoiced Value",
as shov/n on Form B, on domestic shipments of selected steel products from
selected mills of United States Steel Corporation subsidiaries made during
February 1939, amounted to 6.8% of the delivered price, while the "Actual
Freight Paid or Allowed on Shipments From Mill to Destination", as shown
on Form B, amounted to 10.4%.
The unadjusted freight absorption amounted to 3.6% of the delivered
price, and the adjusted freight absorption, after deduction of the amount
of basing point price differentials, amounted to 2.4% of the delivered price.
13844
CONCENTRATION OF ECONOMIC POWER
Breakdown of average delivered price, February 1939 shipments of selected steel
products — selected producing companies
Item
Dollars per
Net Ton
Per Cent of
Delivered
Price
65.06
4.77
50.29
3.00
1.77
.61
1.16
100.0
Freight Paid
8 6
Mill Net Yield (Including Extras)
91.4
Freight Added
5.4
Freight Absorption (Unadjusted)
3.2
Basing Point Price Differentials . . .
1. 1
Freight Absorption (Adjusted)
2. 1
Data are from Department of Justice Summary compiled (rom T. N. E. C. Questionnaire Form B,
entitled "Distribution and Pricing of Selected Steel Products."
Delivered price is "Total Invoiced Delivered Value" as shown on Form B.
Freight paid Is the "Actual Freight Paid or Allowed on Shipments from Mill to Destination," as shown
on F6rm B.
Mill net yield is delivered price less freight paid.
Freight added is "Freight Charges Added to Base Prices to Arrive at Invoiced Value," as shown on
Form B.
Unadjusted freight absorption is difference between such freight paid and such freight added.
Basing point price differentials (as defined in Department of Justice "Supplement to Form B Tables,"
discussion of Table 9) are amounts by which basing point prices applicable on certain sales were greater
than the base prices at other basing points.
Adjusted freight absorption (as defined in Department of Justice "Supplement to Form B Tables,"
discussion of Table 9) is difference between freight absorption (unadjusted) and such basing point price
differentials.
CONCENTRATION OF ECONOMIC POWER
13845
BREAKDOWN OF AVERAGE DELIVERED PRICE
SELECTED PRODUCING COMPANIES
FEBRUARY 1939 SHIPMENTS OF SELECTED STEEL PRODUCTS
DOLLARS PER
NET TON
60
50
40
30 DELIVERED"
PRICE
$55.06
20
10-
FRT ABSORPTION
(UNADJUSTED) $1.77
FRT ADDED $3.00
MILL
NET
YIELD
$50.29
Source: Department c^ Justici Summary
Complied from T.N.E.C '
The "Freight Charges Added to Base Prices to Arrive at Invoiced Value",
OS shown on Form B, on domestic shipments of selected steel products from
selected mills of selected producing companies made during February 1939,
amounted to 5.4% of the delivered price, while the "Actual Freight Paid or
Allowed on Shipments from Mill to Destination", as shown on Form B,
amounted to 8.6%.
The unadjusted freight absorption amounted to 3.2% of the delivered
price, and the adjusted freight absorption, after deduction of the amount of
basing point price differentials, amounted to 2.1% of the delivered price.
13846 CONCENTRATION OF ECONOMIC POWER
Explanation of unadjusted and adjusted freight absorption
Item
Dollars per
Net Ton
Item
DoUarsper
Net Ton
Freight Paid
17.60
8.40
9.20
49.00
42.00
Basing Point Price Diflerential
Freight Absorption (Adjusted)
Base Price at Producing Point
Freight Absorption (Adjusted)
Mill Net Yield
7.00
Freight Absorption (Unadjusted)
Basing Point Price (Houston)
2.20
42.00
Basing Point Price (Chicago)
CONCENTRATION OF ECONOMIC POWER
13847
EXPUNATION OF UNADJUSTED & ADJUSTED
FREIGHT ABSORPTION
Chicago
Dallas
DELIVERED
PRICE -^
$57.40
FREIGHT
PAID
$17.60
BASING POINT-
PRICE
DIFFERENTIALS
$7
FRT ABSORP.
M
FREIGHT
ADDED
$8.40
FREIGHT
ABSORPTION
(UNADJUSTED)
$9.20
Houston 0
The unadjusted freight absorption is the difference between freight paid,
$17.60, and freight added, $8.40, or $9.20. However, since the base price
applicable on the sale includes a difPerential over the base price at the
producing mill, the unadjusted freight absorption is partially ofFset by the
amount of differential. The adjusted freight absorption is the difference
between the unadjusted freight absorption of $9.20 and the differential of
$7.00, or $2.20.
The mill net yield is reduced below the base price only by the amount of
the adjusted freight absorption.
13848
CONCENTRATION OF ECONOMIC POWER
Section D — Capacity and Production
Total ingot capacity— U. S. Steel Corporation subsidiaries and other steel producing
companies
Capacity in Thousands of
Gross Tons
Capacit
y In Thousands of
U.S.
Orcrss Tons
U.S.
Steel in
%of
Year
Steel In
Year
%of
U.S.
Steel
Other
Compa-
nies
Total
U.S.
Total
U.S.
U.S.
Steel
Other
Compa-
nies
Total
U.S.
Total
U.S.
1901 _..
"9,431
1 12,032
21,463
43.9
1920
22,353
33,284
65, 637
40.2
1902
' 10,033
1 12, 667
1 22,700
44.2
1921
22.694
34,683
67, 377
39.6
1903
1 11,211
1 12, 689
'23,900
46.9
1922 ,..
22.694
35,723
68,417
36.8
1904
' 11,548
I 13, 642
26,190
45.8
1923
22,802
35,843
58,645
38.9
1905
' 12,882
' 13,418
'26,300
49.0
1924
22, 816
36,616
59, 432
38.4
1906 .-
I 13,445
' 13, 955
'27,400
49.1
1926
23,125
38,012
61, 137
37.8
1907
14,777
1 13, 723
'28,500
51.8
1926
22,749
35,064
67, 813
39.3
1908
15,590
' 14,710
'30,300
61.6
1927
23,17:^
36, 855
60,032
38.6
1909
17, 157
I 16,843
'34,000
50.6
1928
23,762
37, 703
61. 465
1910
17,845
" 17, 355
135,200
60.7
1929
24,202
39,582
63,784
37.9
1911
18,083
1 17,917
'36,000
60.2
1930
26, 163
40,003
68,166
38.6
1912
18, 822
' 19, 178
'38,000
49.6
1931
26, 075
42,905
68,980
37.8
1913
18,496
' 20,504
'39,000
47.4
1932
27,841
42,499
70, 340
39.6
1914
18, 998
20,691
39,689
47.9
1933
27,342
42,849
70, 191
39.0
1915.. _
19,228
22,066
41.294
46.6
1934
27,342
> 42, 413
' 69, 755
39.2
1916
20,841
24,947
45,788
45.6
1935
27,342
42,704
70,046
39.0
1917..
22,046
27,568
49, 614
44.4
1936
26,657
43, 133
69,790
38. 2
1918
22,207
30, 334
52,641
42.3
1937
25, 772
44,003
69, 775
36.9
1919
22, 340
32, 143
H483
41.0
1938
25,790
46,804.
• 71,694
36.0
Source: Corporation records and American Iron and Steel Institute. Data as of January 1st each year.
' Partly estimated.
» Figures for 1934 and subsequent years i jlude only that portion of capacity of steel for castings used hy
foundries operated by companies producii steel ingots.
Tennessee Coal, Iron and Railroad Coupany data included in Corporation figures beginning with Jan-
uary 1, 1908.
CONCENTRATION OF ECONOMIC POWER
13849
TOTAL INGOT CAPACITY
U. S. STEEL CORPORATION SUBSIDIARIES AND OTHER STEEL PRODUCING COMPANIES
eg Csi CM
50
"
^
1
*"
-
rr
""
-
-
25
n
>. STEEL CAPAC
% OF TOTAl
i
i
i
\
I
I
c
0
I
c
5
i
a
t
0
5
0
0
0
s
I
I
I
T
I
s
3
I
s
]
^
\
s
\
s
I
o
2
Sexjrze: CofVOratxjn tKorfh ami Amn lioo (f S*eW /ns(.
Ingot capacity of the steel industry increased steadily until 1932 (the
decrease ^n the total curve in 1 926 was due to a readjustment of capacity
data by the American Iron and Steel Institute, rather than to an'abandon-
ment of facilities to produce steel). Since 1932 the capacity of the country
has remained practically unchanged, as a result of the reduced demand for
steel, particularly from the railroad and construction industries.
U. S. Steel Corporation's portion of the total capacity of the country has
decreased from a high of 52% in 1 908 to 36% in 1 938.
13850
CONCENTRATION OF ECONOMIC POWER
Steel ingot capacity compared with population — U. S. Steel Corporation subsidiaries
and total United States
Year
Capacity— Thous.
Gross Tons
Population
Total U. 8.
(Thou-
sands)
Capacity Per Cap-
ita—Pounds
U. 8. Steel
Total U. S.
U.8.8.
U.S.
1884
2,821
13,675
4,529
5,270
16.866
6,461
16,803
1 7, 145
7,487
18,607
9,726
110,864
12,002
113.086
H169
1 16. 635
118,900
21,463
122,700
123.900
25.190
126.300
127.400
128.600
130.300
134.000
135.200
136,000
138,000
139,000
39,689
41.294
45,788
49,614
52,541
54,483
65,637
67,377
68,417
68.645
59. 432
61, 137
57,813
60,032
61, 465
63, 784
65,166
68,980
70, 340
70. 191
' 69. 755
70. 046
69.790
69. 776
71,594
66, 379
56,658
59) 217
60,496
61, 775
63.056
64,361
65.666
66,970
68.275
69,580
72! 189
73,494
74,799
76,129
77, 747
79, 365
80,983
82,601
84,219
85,837
87,455
89,073
90.691
92, 267
93, 682
95, 097
96, 512
97,928
99,343
100,758
102. 173
103.688
105.003
106.543
108,208
109,873
111, 537
113, 202
114, 867
116.632
118. 197
119.862
121. 626
123,091
124, 113
124, 974
125. 770
126,626
127. 521
128. 429
129. 2.S7
130,085
114
1885
145
1886
175
1887
190
1888
217
1889
234
1890
242
1891 - -
249
1892
256
1893
288
1894
310
1895
350
1896
379
1897
406
1898
432
1899
495
1900 ---
556
1901
I 9, 431
• 10, 033
« 11.211
« 11, 548
112,882
•13,445
14.777
15,590
17, 157
17,845
18,083
18,822
18.496
18,998
19,228
20,841
22,046
22,207
22,340
22,353
22,694
22,694
22,802
22.816
23.125
22,749
23.177
23,762
24,202
25,163
26,075
27.841
27.342
27,342
27.342
26.657
25. 772
25.790
272
283
310
313
343
351
378
392
424
433
432
443
429
435
434
463
483
480
476
470
470
463
458
451
450
437
439
444
446
458
471
499
487
483
480
465
447
444.
618
1902
641
1903 .-
661
1904
683
1905
700
1906
715
1907
730
1908
762
1909
840
1910
855
1911
861
1912
895
1913
905
1914 .
908
931
1916
1,018
1917
1,088
1918
1,136
1919.. _
1920
1,162
1. 170
1921
1,188
1922
1,191
1923 .
1,178
1924
1,176
1925
1,192
1926
1,111
1927
1.138
1928
1.149
1929
1.176
1930 - .
1.186
1931
1.245
1932
1933
1.261
1.250
1934
1.234
1935
1.230
1936
1.217
1937
1,209
1938
1.233
Source: American Iron and Steel Institute, U. 8. Census Bureau and Corporation records.
I Partly estimated
• Flp:ures for 1934 and subsequent years include only that portion of capacity of steel for castings used by
foundries operated by companies producing steel ingots.
Capacity data are as of January 1 of each year; population data are as of July 1 of each year.
CONCENTRATION OF ECONOMIC POWER
13851
STEEL INGOT CAPACITY COMPARED WITH POPULATION
U. S. STEEL CORPOKATION SUBSIDIARIES AND TOTAL UNITED STATES
.500- ^-^-
^ — rTTrrr
M J .tOTALi
i
M
P
L 1
1200 1 1 i
000 Mi
T 1 '^ - CAPACITY
1 j |PER CAPITA^,
^
¥
1
i
"t4®i
1!
iil
us. STt
CAPACir
i-Jj
1
1
i 1
"T 1 Mlt
-^Tr iL
.iiiiiii
PER CAP!
44'
ti
300 +i ilXk
ili-i'-'-tti
TT
M
TtttTTIj
T^i
^ ff"m I
iM HI
l'!
liH
i!
ill
!' 1
1500
'^%^
a><j>cio^ocn-3io^ci
A-»f, //OT <■• S»^ (--Jt W U S Ceniu! (
The increase In ingo> cspftcity of U. S. St?el Cofpora+ion since 1901 hss
roughly l-ept pace wiih fna growth in population of the United States.
Ingot capaci+>/ of the roal steel incusfrs', hcwaver. has Increased mor«
rapidly +han population.
124491 — 41— pt. 26-
13852
CONCENTRATION OF ECONO.MIC POWER
Total ingot ■production — U. S. Steel Corporation subsidiaries and other steel producing
companies
Production in Thousands
of Gross Tons
U.S.
Steel
'"Tlf
U.S.
Year
i
Production in Thousand.-
of Gross Tons
U.S.
Steel
Year
U.S.
Steel
Other
Com-
panies
Total
U.S.
U.S.
Steei
Other
Com-
panies
Total
U.S.
ln%of
Total
U. S
IflOl
1902
8,855
9,760
9,174
8,413
12,006
13,529
13,100
7,839
13, 355
14, 179
12,763
16.001
18. 6.56
11, 826
16,376
20,911
20,285
19,583
17,200
4,618
8,197
5, 361
4*4;
8,018
9, 8W
10, 263
6.184
10,600
11,916
10,923
14,350
14, 646
11,687
15, 775
21, 863
24, 776
24,879
17, 471
13, 473
14, 947
14,536
13,360
20,024
23, 398
23,363
14,023
23,905
26.096
23,676
31, 251
31,301
23, 613
32, 151
42. 774
45.061
44, 462
34, 671
es.7
66.2
63.1
60.7
60.0
57.8
66.1
56.9
53.8
64.3
53.9
St.!
53.2
50.3
50.9
48.9
45.0
44.0
49.6
1
1 1920
1 192!
19, 278
10,966
16,082
20, 33C
16, 479
18,899
20,307
18.486
20,106
21,869
16, 726
10, 0K2
4,929
8.04'
8, 660
11.131
16, 9f/S
18,532
9, 307
22,856
8;818
19, 521
24,614
21,453
26, 495
27,987
26,4^3
31, 438
34. .'VU
2S. 'm
15, 803
8. 7:.2
15,185
1 17,375
2-i, 962
30, 860
32,036
18.953
42,113 ' 45.8
1903
1904 ...
e::::::::
1907
1908
1905
1910...
1911..
1912
1913 ...
1914
1916 .
1916
1917
ma
19i9
1922
1923.
1924
1925
1926
1927.
1928.
1929
1930
' 1531 .
1932
1333
1934.
1935. -
1936
1937
ffl
35, ««
44 944
37 932
4.S. 594
48. 294
44, 93--
51,544
56, '.33
40,699
25,046
13,681
23, 232
'26,055
34,093
47,768
.50,569
28,350
45.2
45.2
43.4
41.6
42.0
41.2
39.0
38.8
41.1
38.9
36.0
34.6
33.2
32.6
35.4
36.6
33.1
fiourco: Corporation records and American Iron and Stfiel iDotitate. Datii iuclude production of castings.
' .Figures tor 19j4 and subsequent years include only that portiin of production of steel for castings used
by fouQdrie.s operated by companies producing steel ingots.
Tennessee Coal, Iron and Kaiiroad Company data included in Cjrporation figures tegiDnins will: Jan-
uary 1, 1908. .
CONCENTRATION OF ECONOMIC POWER
13853
TOTAL INGOT PRODUCTION
U. S. STEEL CORPORATION SUBSIDIARIES AND OTHER STEEL PRODUCING COMPANIES
-
-
s
1
U.S.
>
IMII
STEEL PROOUC
IS OF TOTAL
Tl
5N
S.
S
\
"
J
'
-
-
^
-
-
-
..
«
■
-
(
1
I
^
I
I
I
I
I
I
3
c
^
i
3
0
I
3
«
?
«
I
I
I
1
3
I
1
I
^
I
1
1
.•JCOrti an</ After. Iron C^ Steci ti
ingot production has shown great variation from year to year, because the
demand for steel products fluctuates so widely with changes in general
business conditions.
The portion of the country's ingots produced by U. S. Stee| Corporation
has been declining with few interruptions since 1901. Whereas the
Corporation produce ' 66% of the total in J 90 1, it produced but 33%
in 1938.
13854 CONCENTRATION OF ECONOMIC POWER
Steel ingot production compared with populaiion-total United States
1884..
1885..
1886..
1887..
1900.
1901.
1902.
1903.
1904.
1905.
1906.
1907.
1908.
1909.
1910.
Produc-
tion
(Thou-
sands of
Gross
Tons)
Popula-
tion
(Ttou-
sands)
1,551
1,712
2,563
3,339
2,899
3,386
4,277
3,904
4,928
4,020
4,412
6,115
6,282
7,157
8,933
10,640
10,188
13,474
14,947
14,635
13,860
20,024
23, 398
23,363
14,023
23,955
26,095
mi::::::::::::.— 1 23,676
Produc-
tion
per
Capita
(Pounds)
55, S
56,658
57,938
69,217
60,496
61, 775
63,056
64,361
65,666
66.970
68,275
69,580
70,885
72, 189
73,494
74,799
76,129
77, 747
79,3(65
80,983
82,601
84,219
85,837
87,455
89,073
90,691
92,267
1925..
1926..
1927.
1928.
1929.
1930.
1931.
1932.
1933.
1934.
1935.
Ingot
Produc-
tion
(Thou-
sands of
Gross
Tons)
Popula-
tion
(Thou-
sands)
Produc-
tion
per
Capita
(Pounds)
31, 251
31,301
23,513
32, 151
42, 774
45,061
44,462
34,671
42, 133
19,784
35,603
44,944
37, 932
45,394
48,294
44, 935
51,544
56,433
40,699
25,946
13,681
23,232
1 26, 055
34,093
47,768
1937'::::::::: 50, 669
95,097
96,512
97,928
99,343
100,758
102, 173
109, 873
111,537
113,202
114,867
116,532
118, 197
119,862
121, 626
123, 091
124, 113
124,974
125, 770
126, 628
127, 521
128,429
129,257
130,085
927
851
963
,039
741
468
244
414
461
599
833
876
mZs for 1934 and subseauent years include only that portion of production of steel for c^stmgs used
by fortes operated by companies producing steel ingots.
CONCENTRATION OF ECONOMIC POWER
13856
STEEL INGOT PRODUCTION COMPARED WITH POPULATION
rOTAL UNITED STATES
1 ^i
INGO ,
PRODUCTION /W
(
an _ -
Ml
!\
yllt-M
ji^f'- T POPULATION
y
iiiiiiiiiii
iiiiiiiiiiiiiiiiii
200 §
150 ^
1200
1000
800
iiiisslsigmg^sgSgSggfgaSISI
: Amtr. Inn O-Su^liA vdU.S. C ,« (
From 1884 thrpugh 1929, steel ingot production in this country expanded
more rapidly than population and production per capita increased from 63
to 1,039 pounds. Since 1929, the situation has been reversed; the peak in
1937 was lower than that of 1929 and in 1938 production was only 488
pounds per capita.
13856 CONCENTRATION OF ECONOMIC POWER
Ingot capacity and production — U. S. Steel Corporation subsidiaries
[Monthly Production and Average Monthly Capacity in Thousands of Grass Tons]
1920
1923
1926
1929
1932
1936
1938
Jan
1,676
1,630
1,858
1,468
1,548
1,522
1,415
1,500
1,569
1,658
1,685
1,769
1.780
1,560
1,809
1,755
1,865
1,694
1,660
1,738
1,599
1,778
1,616
1,476
1,701
1,634
1,996
1,812
1,721
1,637
1,631
1,711
1,651
1,668
1.553
1,593
1,753
1,733
1,975
2.019
2,142
1,970
1,982
2,009
1,772
1,754
1,495
1,266
661
549
569
457
432
342
263
291
361
392
376
348
909
957
879
899
753
7.S8
923
951
1,039
1,070
1,059
685
Feb
548
Mar -
Apr -
711
642
Sfay - --
647
587
Jul .
612
Aug
764
838
Oct -.
1,047
Nov -
1,224
Dec
1,092
Aver. Cap'y . . .
1,863
1,900
1,899
2,017
2,320
2,278
2,143
1921
1924
1927
1930
1933
1936
1939
Jan
1,803
1,349
1,103
760
743
612
501
622
648
939
978
919
1,736
1,738
1,934
1.529
1,206
1,015
876
1,120
1,227
1,316
Mn
1,466
1,695
1.649
1.993
1.780
1,781
1 492
1,328
1,446
1,311
1,335
1,339
1,339
1,480
1,652
1,792
1,690
1.719
1.489
1.283
1.382
1,340
1,140
905
854
350
349
305
466
655
926
1,204
1,000
790
785
569
648
997
1,037
1,215
1,472
1,515
1,473
1,409
1,534
1,490
1,590
1.578
1.598
1,076
Feb
1,037
Mar " ::
1,177
1,018
mi" :::::::::::::;::::::::
963
988
Jul - -..:_-
1.047
Aug -
Oct
Nov..... ...
Dec
Aver. Cap'y
1,891
1.901
1,931
2,073
2,278
2,221
2,143
1922
1925
1928
1931
1334
1937
Jan
962
1,048
1, 3.59
1,346
1,446
1,456
1.359
1,279
1,246
1,493
1,553
1,634
1,832
1,674
1,859
1,578
1,440
1,401
1,348
1,461
1,472
1,699
1,613
1,623
1,657
1,715
1.882
1,792
1,731
1.498
1.497
1,606
1.633
1.821
1.669
1,616
1,042
1,013
1,241
1,134
997
837
743
661
597
608
642
667
602
693
917
976
1,194
1,162
536
616
487
461
608
618
1.738
T 639
1,964
1,865
1,695
1,790
1,603
1,131
800
552
Feb... -.
Mar
Apr
Jun
m
Aug --
Oct - -
Nov ---
Dec
1,891
1,927
1.080
2,240
2,278
2,148
Data include Ingots and castings.
CONCENTRATION OF ECONOMIC POWER
13857
INGOT CAPACITY AND PRODUCTION
U S. STEEL CORPORATION SUBSIDIARIES
MILLIONS OF GROSS TONS
1920 1922 1924
1930 1932 1934 1936
Ai four di'tmcf peiiods m the inrervai 1920 1929, denands upon Ingot
production facilities exceeded U. S. Steel Corporation s capacity. About
1929 the Corporation inaugurated a prorjiarn of plant ^nodern;za^ion
which was accompanied by an increase in ingof capacity. Having been
commenced, this program wa; carried on during the years 1930-1932 in
spite of the business depression, the magnitude and duration of which
could not be foreseen.
Since 1932, some obsolete capacity has been retired. At preosnt. the ingot
capacity of the Corporation is no mere than sufficient tc provide the
steel which it required at periods of peak demand.
13858 CONCENTRATION OF ECONOMIC POWER
Ingot capacity and production — total United States
[Monthly Production and Average Monthly Capacity in Thousands of Gross Tons]
1920
1923
1926
1929
1932
1935
1938
Jan
3.624
3,402
3,917
3,132
3; 423
3,539
3,328
3,562
3,561
3,581
3,133
2,779
3,841
3,472
4,067
3,964
4,216
3,767
3,631
3,696
3,357
3,5<7
3,134
2,863
4,132
3,785
4,469
4,196
3,928
3,734
3,635
3,987
3,913
4,074
3,076
3,467
4,545
4,372
5,118
4,999
5,339
4,951
4,898
4,988
4,673
4,579
3,556
2,932
1,600
1,496
1,448
1,273
1,137
923
815
856
1,003
1,099
1,043
871
2,915
2,817
2,910
2,682
2,675
2.294
2,303
2,962
2,869
3,192
3,200
3,121
1,794
Feb
1,726
Mar """
2,038
Apr
1,951
May : :
1,831
jun .
1,660
Jul .
2,008
2,580
se^v::::::: ::::::::::::::: ::::::::::
2,692
Oct
3,158
Nov
3,618
Dec
3,185
Aver, cap'y
4,636
4,887
4,818
5,315
5,862
5,837
5,966
1921
1924
1927
1930
1933
1936
1939
Jan. ,
2,517
1,999
1,795
1,387
1,446
1,146
918
1,300
1,342
1,847
1,897
1,630
3,650
3,826
4,207
3,348
2,640
2 066
1,878
2,553
2,828
3,125
3,121
3,569
3,823
3,S45
4.575
4,163
4.083
3,626
3,232
3,529
3,298
3,345
3,155
3,203
3,808
4,067
t;!l
4,014
3,445
2,945
3,085
2,863
2,714
2,2.30
1,995
1,030
1,087
910
1,361
2,005
2,599
3,210
2,905
2,313
2,112
1,540
1,822
3,086
3,002
3,384
3,991
4,097
4,035
3,975
4.247
4,214
4,601
4,389
4,491
3,225
Feb... -
3,037
Mar — ..:
3,459
Ap'
3,021
May
2,970
Jun
3,175
Jul . .
3,214
Aug
3,823
4,299
Oct
6,480
Nov
5:551
Dec -
5,246
Aver. Cap'y
4,781
1922
4,953
5,003
5,431
5,849
6,816
6,088
1925
1928
1931
1934
1937
Jan
2' 071
2,814
2,902
3,219
3,128
2,953
2,629
2,818
3,410
3,430
3,301
4,193
3,752
4,194
3,584
3.455
3,205
3,084
3,421
3,490
4,028
4,081
4,549
4,345
4.246
3,778
3,841
4,217
4,186
4,693
4,306
4,055
2,534
2,570
3,083
2,794
2,574
2,149
1,907
1,733
1,560
1,605
1,607
1,313
2, 025
2,243
2.836
2,976
3,447
3,102
1.509
1,399
i;502
1,633
1,991
4,786
4.498
5,303
5,155
5,237
4,254
4,631
4,958
4,362
3,449
2,189
1,496
Feb " "" "" '
Mar -
uii: :::":::::::::::::::::::::
Jul
Aug
Sep
Oct "
1^:::::::::::::::::::::::: :::::::::;:::
Dec
4,868
5,095
6,122
6,748
6,813
6,815
Data include production of open-hearth, Bessemer, crucible and electric ingots, but exclude production of
castings. Portion of the 1939 monthly data represented by production of crucible and electric ingots was
estimated.
CONCENTRATION OF ECONOMIC POWER 13859
Percent of ingot capacity operated — U. S. Steel Corporation subsidiaries
Year
Average
Rate for
Year
Peak Month
Average
Rate-
Peak
Month
1920
86.2
48.3
70.9
89.2
72.2
81.7
89.3
79.8
84.7
90.4
66.5
n'.7
29.4
31.6
40.8
63.4
72.0
36.4
51.9
March
96.0
1921
January
95.7
82.0
May
94.8
March _
102.2
192S
March _
96.8
1926 -
March
97.5
1927
March
95.6
1928
March .
1929
May
98 3
1930
March.. .
82.9
1931
March
64.9
1932
March
22.7
1933
July
55.1
May .■
47.0
72.0
1937
May
89.7
1938
57.7
19391
September
67 6
Per cent of ingot capacity operated Is based on ingots and castings.
Peak month rate is calculated from actual monthly capacity.
' First 9 months.
i386C
CONCENTRATION OF ECONOMIC POWER
PER CENT OF INGOT CAPACITY OPERATED
ANNUAL AVERAGE CONTRASTED WITH PEAK MONTH
U. S. STEEL CORPORATION SUBSlDlARltS
i rOS 1953 ARE c< e\sis '.
In the steel business it is imperat've that there be sufficient capacity to
meet peak demands. Customers' orders vary as to s-ze f\n6 quali+y to such
an extent that steel must be rolled to order and cannot ordinarily be talcen
from stock. Some steel products, such as sheets A'hich a'e to be subjected
to deep-drawing as in ihe manufacture of autoniobile fenders, m^jst bs
used in a comparatively short time after production in order to obtain the
best results.
The extent to which the curve representing peak monthly operations is
above the curve representing average operations is an indication of the
extent to which the capacity needed to meet peak demands even in years
of low production is in excess of the average capacity operated. It is
apparent that the entire amount of the capacity of the United States Steel
Corporation subsidiaries is needed to meet peak demanoi in prosperous
yeers.
CONCENTRATION OF ECONOMIC POWER
13861
steel production and manufacturing production- Federal Reserve indexes adjusted
for seasonal variation
Jan..
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep..
Oct..
Nov.
Dec-
Jan..
Feb..
Mar..
Apr..
May.
Jun..
Jul...
Aug..
Sep..
Oct..
Nov..
Dec.
Jan..
Feb..
Mar-
Apr..
May.
Jun..
Jul...
Aug..
Sep..
Oct..
Nov..
Dec.
Mfrg
Jan..
Feb.
Mar.
Apr.
May.
Jun..
Jul...
Aug.
Sep..
Oct..
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep..
Oct..
Nov.
Dec.
Jan..
Feb..
Mar..
Apr..
May.
Jun..
Jul...
Aug..
Sep..
Oct..
Nov-
Dec.
Jan..
Feb..
Mar..
May.
Jun..
Jul...
Aug..
Sep..
Oct..
Nov..
Dec.
[1923-1925 = 100]
Mfrg
Jan.,
Feb-
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep..
Oct.-
Nov.
Dec.
Jan.-
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep..
Oct..
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep..
Oct..
Nov.
Dec
Jan..
Feb..
Mar..
Apr..
May.
Jun..
Jul...
Aug..
Sep..
Oct..
Nov..
Dec.
Feb..
Mar..
Apr..
May.
Jun..
Jul...
Aug..
Sep..
Oct..
Nov..
Dec.
Jan..
Feb..
Mar.
May.
Jun..
Jul...
Aug..
Sep..
Oct..
Nov..
Dec.
Steel Mfrg
Jan.
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep..
Oct..
Nov.
Dec
29
63
31
61
22
M
36
66
48
77
71
03
99
102
«0
01
«fi
83
60
76
47
70
60
73
Jan.
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep..
Oct..
Nov.
Dec.
13862 CONCENTRATION OF ECONOMIC POWER
Steel production and manufacturing production — Federal Reserve indexes adjusted
for seasonal variation — Continued
[1923-1925-100]
Steel
Mfrg
Steel
Mfrg
Steel
Mfrg
1938
Jul
1938
Jan
1939
Jan
62
50
49
50
47
46
76
75
76
73
73
74
62
70
76
90
108
101
82
87
89
95
103
104
94
87
79
73
89
100
100
Feb
Aug
Feb
97
Sep..
93
91
Apr...
Oct
^/y
May
Nov
97
Jun
Dec .
Jul..
> 101
Source: Federal Reserve Board.
> Preliminary.
STEEL PRODUCTION AND MANUFACTURING PRODUCTION
FEDERAL RESERVE INDEXES ADJUSTED FOR SEASONAL VARIATION
160 p^
1
1 —
19
23-
1925
= 1
DO
1
r"
. —
160
140
120
to
Z 100
_
—
—
—
/
^
—
—
—
—
—
140
120
100 £
y
u
k
^
^
MANUFACTURING
PRODUCTION
\
\
/
,
\i
5
i 80
\
^
^
//
\J
80 i
\
fv\
h
\f
\
2 60
^ 40
20
0
\
SJ"
/
60 2
Q
40 ?
20
I
\
V
./ IRON AND STEEL
K PRODUCTION
1924 1926 1928 1930 1932 1934 1936 1938
Soor,t f«*.-y /?«»-»» Soi'rf
The steel business in the United States is good when general business is
good and vice versa. There is marked correlation in the fluctuations in
steel production and those of all manufacturing production, although the
peaks and valleys of steel production are somewhat more pronounced
than those for all manufacturing.
CONCENTRATION OF ECONOMIC POWER 13863
World ingot production by principal steel producing countries
Production in Thou-
sands of Gross Tons
Country
Production in Thou-
sands of Gross Tons
1929
1937
1938
1929
1937
1938
United States
56, 433
18,786
4,645
9,636
60,569
20, 176
17, 544
12,964
28,350
22,876
18,156
10, 394
,Tnpftn
2,258
9,546
2,089
4,015
5,719
7,795
2,054
3.808
6 860
Germany •
Fraqce
6 077
U. S.S. R
Italy
2,285
United Kingdom.
2,249
« Includes Austria but not Czechoslovakia.
Source: American Iron and Steel Institute, except production of Japan, 1938, estimated by Iron Age
(foreign figures based on information received from abroad).
Data Include production of steel ingots and castings.
UNITED
STATES
GERMANY
U.S.S.R.
FRANCE
ITALY
BELGIUM
WORLD INGOT PRODUCTION
BY PRINCIPAL STEEL PRODUCING COUNTRIES
MILLIONS OF GROSS TONS
20 30 40
'. 'ion & Slul //at tnd Im Agr
The United States produces more steel than any other country In the world.
In the last ten years, however, its steel production has declined, while that
of most of the other countries has increased, notably Germany and Russia.
13864 CONCENTRATION OF ECONOMIC POWER
Ingot capacity and production
[Total United States]
Month
1920
ly Production and Average Monthly Capacity in
Thousands of Gross Tons
1923
1926
1929
1932
1935
1938
Jan
3,624
3,402
3,917
3,132
3,423
3,539
3,328
3,562
3.661
3,681
3,133
2,779
3,841
3,472
4,067
3.964
4,216
3,767
3,531
3,696
3,357
3,577
3.134
2,863
4.132
3,785
4,469
4,106
3,928
3,734
3,635
3.987
3,913
4,074
3,076
3,467
4, ,545
4,372
5,118
4,999
5,339
4,961
4,898
4,988
4,573
4,579
3,656
2,932
1,500
1,496
1,448
1,273
1,137
923
815
856
1,003
1,099
1,043
871
2,915
2,817
2,910
2 682
2.675
2.294
2.303
2,962
2.869"
3.192
3,200
3,121
1 764
Feb
1 726
Mar
2 038
Apr - -
1 951
Xy" "
jun .
1,660
Jul .
2 008
Auk . -
2,580
Sep : ...
2,692
Oct - -.
3, 158
Nov
3,618
Dec .
3,185
4,636
4,887
4,818
6,315
5,862
5.837
1921
2,517
1,999
1,795
1,387
1,448
1,146
918
1,300
1,342
1,847
1,897
1,630
1924
3,650
3,826
4,207
3,348
2,640
2,066
1,878
2,653
31125
3,121
3,569
1927
1930
1933
1,030
1,087
910
1.361
2.006
2,599
3,210
2,905
2,313
2,112
1,640
1,822
1936
1939
Jan — ---
Feb... -
3.823
3,846
4.675
4,163
4,083
3] 232
3,529
3.298
3,345
3,155
3,203
3.808
4.067
41142
4,014
3,445
2,945
3,085
2,863
2,714
2.230
1,995
3,086
3.002
3.384
3,991
4,097
4,035
3,975
4,241
4,214
4,601
4! 491
3.225
3,037
Mar... ---
3,459
Apr— - -
3,021
May..... —
2,970
Jun -- -
3,175
Jul
30, 214
Aug -.--
3,823
Sep
4.299
Oct --.-
5,480
Nov.... - -
Pec
5,551
6,246
Aver. Cap'y
4,781
4,953
5.003
5,431
6,849
5,816
6,088
1922
1925
1928
1931
1934
1937
Jan - ----
1,893
2,071
2.814
2.902
3,219
3,128
2,953
2,629
4,193
3,752
4,194
3,584
3.455
3,205
3,084
3,421
3,490
3,889
3,903
3,971
4,028
4,081
4, 549
4,345
4,246
3,778
3,841
4,217
4,186
4.693
4,306
4,055
2,534
2,570
3,083
2,794
2,574
2,149
1.907
1,733
1.660
1,605
1,607
1,313
2,025
2,243
2,836
2,976
3,447
3,102
1,509
1.399
1,286
1,602
1,633
1,991
4,786
4,498"
5.303
5.155
5,237
4,254
4,631
4! 362
3,449
2,189
1,496
Feb - -
Mar ---
Apr
Sfay.
Jan.:
Jul
Aug
Se^.
2,818
3,410
3,430
3,301
c^:::...,
Nov - ---.
Dec -
Aver. Cap'y
4,868
5,095
5,122
5,748
5,813
5.815
CONCENTRATION OF ECONOMIC POWER
13865
INGOT CAPACITY AND PRODUCTION
TOTAL UNITED STATES
MILLIONS OF GROSS TONS
6
5
4
3
6
5
4
3
X'
._j
"a
M
VERA
ONTH
*PAC
t
LY
TY
H
--
1
'"
ik
1
L
k
Ai*
^
A
/
\
Vj
n
i^
^
1
V
'^
%
L
Ipi oduc
TION
i
i
r/
L
i
1
2
—
\
I-
4
—
—
r
\
V
f
I
1
u
-"
J-
'
2
1
0
0
-
V.
1920 1922 1924 1926 1928 1930 1932 1934 1936 1938
SLrre; American /ran & Sl^l hstitiM
Capacity of the steel Industry is -not excessive. Unused or Idle capacity
should not be confused with "excess" capacity. Ingot copacity, an accepted
basis for determining rotes of operations, reflects roughly operations of
finishing capacities. Even in periods of peak demand; orders are not dis-
tributed among products in such a way as to make possible full utilization
of all finishing facilities. In practice, therefore, operations probably would
never be maintained ot 100 per cent of either ingot or finishing capacity
because of lack of coordination between demand and capacity for various
products. Production might, therefore, be expected to run below capacity
even at the peak of the cycle.
In times of emergency, or under the pressure of extraordinary demands on'
the industry, it might occasionally be possible to attain an operating rate
slightly in excess of 100 per cent of rated capacity for short periods by
bringing into operotion obsolete facilities, lengthening the work week,
eliminating holidays, or by other means.
13866 CONCENTRATION OF ECONOMIC POWER
Section E — Labor
Number of employees and ingot production — U. S. Steel Corporation and subsidiaries
(000 Omitted)
Index No.'s
1929=100
(000 Omitted)
Index No.'s
1929=100
No. of
Empl'8
Ingot
Prod'n
(Or.
Tons)
Empl's
Prod'n
No. of
Empl's
Ingot
Prod'n
(Or.
Tons)
Empl's
Prod'n
1929
Jan
1933
Jon
243
244
24";
255
259
261
263
263
260
265
248
241
1,753
1,733
1,975
2,019
2,142
1,970
1,982
2,009
i:772
1,754
1,495
1,266
96
96
97
101
102
103
104
104
103
101
98
95
96
95
108
111
118
108
109
110
97
96
82
69
151
147
140
144
151
171
190
201
201
190
192
189
350
349
305
466
655
926
1,204
1.000
790
785
569
648
60
68
65
67
60
67
76
79
79
76
76
76
19
Feb
Feb
19
Mar
Apr
Mar.
Apr
17
26
36
51
Jun
May
Jun
Jill
Jul
6fl
66
Aug
Aug
Sep
Sep
43
oc?: ::::::::
Oct
g?J-:::::::
Nov
Dec...
31
Jan
19
30
1934
Jan
242
248
252
256
261
262
259
256
253
248
244
1,480
1,652
1,792
1,690
1,719
1,489
1,283
1,382
1,340
1,140
905
854
96
98
)0f
10
10b
103
102
101
100
98
96
94
81
91
93
94
81
70
76
74
62
49
47
185
187
190
195
■604
208
203
192
186
178
176
174
602
693
917
976
1,194
1,152
536
516
487
461
508
618
73
74
75
77
81
82
80
76
73
'0
70
69
33
Feb""
Feb
38
Mar
Apr
Mar_
Apr
50
/ay
Jun
May
Jun
66
63
j"l ■■
Jul
29
Aug
Aug
28
Sep.
Sep.
27
Oct
Oct....
25
^tlv"-:::
Nov
Dec
28
34
Jan
1931
1935
Jan
238
237
237
237
233
221
207
203
199
194
192
189
1,042
1,013
1,241
1,134
997
837
743
661
597
608
642
567
94
94
94
94
92
87
82
80
79
77
76
75
67
56
68
62
54
46
41
36
33
33
35
31
182
191
195
196
197
197
196
197
197
190
198
196
909
933
957
879
899
753
758
923
951
1,039
1,070
1,059
72
75
77
78
78
78
77
78
78
77
78
77
60
Feb
Feb
51
Mm"::::::::
Apr
Mar
Apr
63
48
/ay
Jun
May
Jun
49
41
Jul.....
Jul.
43
Aug ■.
61
Sep :.
Sep
62
Oct ::
Oct....:...
67
g^c^v.::::::
g^I".::-:-::
59
58
Jan
Feb
19
32
19
36
Jan
186
182
182
173
167
169
165
150
152
154
158
164
551
549
669
457
432
342
263
291
361
392
376
348
73
72
72
68
66
63
61
69
80
61
62
61
30
30
31
25
14
16
20
22
20
10
196
197
201
211
220
223
229
235
238
239
240
239
997
1,037
1,215
1,472
1.515
1,473
1,409
1,534
\:Z
77
78
79
83
87
88
91
93
94
94
95
94
55
Feb
57
Mar
Apr
Mar
Apr
67
81
May
Jun
May
83
81
Jul
Jul
77
Aug
Aug
84
Sep .
Sep
82
Oct .
Oct
87
Nov
Dec
To".::::::
87
88
CONCENTRATION OF ECONOMIC POWER 13867
Number of employees and ingot jtrodudion — U. S. Steel Corporation end eubsid-
iaries — Continued
(000 Omitted)
Index No.'8
1929-100
(000 Omitted)
Index No.'s
1929=100
No. of
Empl'8
Ingot
Prod'n
(Or.
Tons)
Empl's
Prod'n
No. of
Empl's
Ingot
Tons)
Empl's
Prod'n
1937
Jul..
1938
Jan
240
24fi
2(4
261
267
273
277
278
276
271
266
236
1;1
1,964
1,866
1,696
1,790
1,603
■•IS
662
96
97
^
106
108
109
110
109
107
101
93
06
90
104
102
108
102
93
98
88
62
44
30
192
191
192
107.
206
207
612
764
838
1,047 .
1,224
1,002
76
76
76
78
81
82
34
Frt"
AoK..
42
M«
Sep.:: —
^: . —
67
^
Nov
Dec
67
60
Jul
Jan
1939
^^
Nov
207
209
211
196
212
214
214
214
236
1,076
i;037
1,177
'Sg
968
1,047
1,269
1,431
82
83
83
77
84
84
86
86
93
60
Feb
67
.1938
Mar
Apr
66
66
Afty
Jun
63
220
211
206
204
202
197
685
711
642
847
887
87
83
81
8i
80
78
38
30
39
83
SS
33
64
Jan
Jul
87
Feb
Aug.
Sep
.70
Mar
78
Apr
Oct
i^y
Jun
g"J.-.:::::::
Number of employees repreaents number on rolls during each month.
Ingot production data Include production of iqgots and castings.
124491.^41- pt. 2«
13868
CONCENTRATION OP ECONOMIC POWER
NUMBER OF EMPLOYEES AND INGOT PRODUCTION
U. S. STEEL CORPORATION AND SUBSIDIARIES
1929 = 100
140 1 1 1 1 1 1 1 1 1 1 F 1 il40
k
^
yn
i^
\
\
MBERO
PLOYEE
/?
^
^.j
V
\
\
K
\.
Si
Xr-
1
i
1 1
\
\
N
f^/
J
!
/
/
\'
V
\r
r
VlNGOT
-PR DUCTIO
fN
y
1
1929 1930 1931 1932 1933 193!* 1935 1936 1937 1938 1939 1940
During depression periods, the number of employees of'U. S. Steel Cor-
poration has not declined os much as ingot production. To a large
extent, this has been due to the Corporotion's policy of sharing the
available work so far as practicable among the maximum number of
employees.
In 1937, there were more employees than i
ingot production was less.
1929. despite the fact that
CONCENTRATION OF ECONOMIC POWER
13869
Actual number of employees and number that would have been required on basis of
1929 hours per week — U. S. Steel Corporation and subsidiaries
Actual
No.of
Empi's
(B^l'l,
HrsJ
Addtl.
Number
Empl'd
Actual
No.of
Empi's
Empi's
(Basis '20
Addtl.
Number
Empl'd
253,138
251,782
215,223
164,390
263,138
335,978
160^020
00,340
Mar
1036
1930
16,804
'55'iSi
73,990
200,700
211,008
210,664
222,079
239,452
234,972
237, 570
238,044
340,014
238,781
162,044
184,975
180,234
197,882
195, 183
190. 370
'209,802
216,670
209,363
209,838
38,656
Apr
M;033
1933
Jun
25 097
jSf.".:::::::::::::::
34,369
151,010
147,360
130,686
143,022
160,661
170,767
100,170
200,740
200,633
189,696
102,438
180)220
77,703
81,659
78,116
80,061
97,173
130,478
163,947
160,773
138,967
129,338
119, 128
120,418
73,217
66,710
66,460
63,861
63,478
. 40289
36^223
30^076
61,666
60 357
73,310
68,811
Aug
35,602
JiuiuBrv
Sep
27,768
Feb
Oct
23,374
Mar
Nov
30,661
Anr
Dec :
38.943
^fi^:::::::::::::
Jan
1937
Jul
Aog
S^D
240,360
244,602
264,011
260,665
267,052
272,656
276,897
278,178
276,202
270, 616
255,788
235,565
207,683
222,365
220,810
230,673
228,902
239,606
230,748
231,815
228,378
194,982
167,767
142,367
oSt ' "
32.776
Feb
22,337
Dm ::
Mar,.
24,102
Apr
29,892
1034
May
38,160
Jun.
33,060
Jul..-.
4o;i4e
185,433
186,686
190,163
JHeeo
204,033
207,731
203,416
192,038
186,003
178,426
175,737
174,350
118,688
126,074
186,824
188,207
165,805
162,318
119,760
116,803
103, ««
106,692
104,986
104, 157
71,846
60,612
54.320
56,462
«,138
46,413
83,656
76,236
81,904
71,834
70,761
70,108
Aug
46,363
Jan
Sep........
47.827
Feb
0(?t
75,633
Mar -..
Nov...
88,031
Apr -
Dec
93,308
»fty-...-
Jan
Jtm.
1938
Jul.... -
Aug
Sep. ..:„.
Oct
220,270
210,680
204,810
203,816
201,623
106,808
102,021
101,811
103,321
107,271
206.700
307,370
120,683
126,257
130,830
126.730
120,014
120,611
111,389
glS
138, 773
163,614
149,468
90,687
Nov
Feb
84 323
Dec
Mm-'
74 4%
Apr
78,085
1935
Jun
76,287
jt:::::::::::::
80.633
181,825
ii
lOsiftlO
106,634
106, 624
106,685
197,084
106,075
Its
1«^02S
%%
144,210
142,601
151, 134
140, 650
1481800
63.131
45^482
61,060
5% 676
H232
60^206
60 876
52:324
54,023
44,461
48^426
4?; 086
Aug
67, 911
Jan
Sep :
62,443
Feb
Oct
ss,m
Mar
Nov
62,095
Apr
Deo : ::
6/, 902
*^::::::::::::::
Jan
Jun
1030
JuL
Aug
Sep
206,715
208,005
211,011
105,764
212,381
213,728
214,205
148,648
141.050
146,644
150. 139
149.758
Ort
68,167
Feb
50,213
Mar
40,784
Apr
63,814
193P
64,689
Jul
64.447
Jan. . ..
196,858
197, 4«3.
152,617
163,608
43,241
33,766
Feb . .
Actual number of employees represents number on rolls during each month.
Equivalent number of employees was obtained by multiplying the actual n _ .
year or month by number of hours worked per week during the period and dividing by the number of
; the actual number of employees in esch
hours worked per week daring 1929.
13&70
CONCENTRATION OF ECONOMIC POWEB
ACTUAL NUMBER OF EMPLOYEES
& NUMBER THAT WOULD HAVE BEEN REQUIRED ON BASIS OF 1929 HOURS PER WEEK
U. S. STEEL CORPORATION AND SUBSIDIARIES
250,000
200,000
A
250,000
200,000
150,000
100,000
50,000
-^
\»
TUAL
M8ER
A
■^
h
\y
f
\
V
Jl
A
r'-
/
\/
(\f^
\
\
t
A
rv/l 1
EQUW. NUMBER
ON BASIS OF 1929
HOURS PER WEEK
^y
V
V.
J
0
ADDITIO
100,000
1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 "
^AL NUMBER OF EMPLOYEES DUE TO REDUCTION OF HOURS PER WEEK
^
^
\/
^r
./v
/
K
>A
50,000
0
0
ANNUAL DATA ir
/
/
\j
^
y
1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940
>9-l932, MONTHIY DATA IHEREAntR
U. S. Steel Corporfttion's policy during recent depression years of sharing
the available work so far as practicable among the maximum number of
employees has made it possible to give work to a larger number of
employees than would otherwise be required. The additional number of
employees has varied from a minimum of about 25,000 during periods of
high operating activity to a maximum of roughly 90,000 during periods of
business depression.
Ingot production and number of employees — U. S. Sted Corporation and subsidiaries
Item
(0 months)
1837 In %
0^1929
Monthly Ingot Production (Tons).
Number of Employees
Hours per Week.
Earnings per Hour..
Ettnlngs per Week.
1,822,401
2^3,138
47.2
10.643
$3a33
1,783,291
263,391
39.6
lasie
133.16
Ingot production represents average monthly production of Ingots and castings.
Number of employees represents the average number on rolls.
Hours per week, earnings per hour and earnings per week data are for all wage earners of U. 8. Steel Cor-
poration and subsidiaries.
CONCENTRATION OF ECONOMIC POWER
13871
INGOT PRODUCTION AND NUMBER OF EMPLOYEES
U. S. STEEL CORPORATION AND SUBSIDIARIES
1929 = 100
INGOT NUMBER OF
PRODUCTION EMPLOYEES
•100
loo-
se-
60-
40-
20-
-60
-40
20
1929 1937
1929 1937
TECHNOLOGICAL ADVANCES HAVE NOT RESULTED IN
FEWER EMPLOYEES
Technological advances in the art of steel making have no+ resulted in fev^er em-
ployees being required by U. S. Steel Corporation. The year 1929 and the first nine
months of 1937 are two periods in which production of steel was comparable. Be-
tween these two periods the Corporation expended huge amounts for mor§ modern
and continuous types of equipment to keep pace with technological advances.
Average monthly ingot production in the first nine months of 1937 was 2% less
than in the year 1929, whereas the average number of employees in the first nine
months of 1937 was 4% more. The employees in 1937 worked about 8 hours less per
week than in 1929 but. due to increases of about 27% in hourly wage rates, actu-
ally earned more per week in 1937 than In 1929.
13872 CONCENTRATION OF ECONOMIC POWER
Ingot production and number of employees — Total steel industry
Monthly Ingot Production (Tons).
Number of Employees
4.670,869
428,319
1937 (9
months)
1937 in %
or 1929
Ingot production represents average monthly production of all ingots, and excludes production of castings.
Number of employees represents number of wage earners of iron and steel manufacturing companies
reporting to American Iron and Steel Institute (1929 figure partially estimated, based on data of U. S. Census
of Manufactures on wage earners in Blast Furnaces, Steel Works and Rolling Mills).
Employment and payroll by classes of employees — U. S. Steel Corporation and
subsidiaries
Class o( Employees
Number of
Employees
Total PayroU
Wage Earners:^
228,281
170,241
$366, 610. 690
1938
204,643,987
Average -
199,261
280, 627, 289
Operating Salaried:
1937 . .:..
22,669
21, 791
$65, 938, 276
1938. -
49, 311, 604
22,230
52, 624, 940
Qen'l Administrative and Sales:
1937 - .
10, 343
10,076
$30,478,817
1938
28,353,741
Average -. -
10, 210
29, 416, 279
Number of employees represents the average number on rolls during the year.
Payroll figures include construction payroll.
CONCENTRATION OP ECONOMIC POWER
13873
EMPLOYMENT AND PAYROLL BY CLASSES OF EMPLOYEES
U. S. STEEL CORPORATION AND SUBSIDIARIES
1937-1938 AVERAGE
THOUSANDS OF EMPLOYEES
50 100 150
WAGE EARNERS 199
OPERATING SALARIED 22
GEN'L ADMIN. & SALES 10
WAGE EARNERS 281
OPERATING SALARIED 53
GEN'L ADMIN. & SALES 29
MILLIONS OF DOLWRS OF PAYROLL PER YEAR
100 200 300
General administrative and sales employees receive only a small part
of the total payroll of U. S. Steel Corporation and subsidiaries. During
1937 and 1938 wage earners and operating salaried employees, repre-
senting 957o of the working force, received 927o of the total payroll.
13874 CONCENTRATION OF ECONOMIC POWER
Employees by age groups — U. S. Steel Corporation and subsidiaries
Age Group
Number of
Employees
Per Cent
of Total
Ago Group
Number of
Employees
Per Cenl
of Total
3/229
21,409
24,100
25.771
24,560
26 207
25,312
1.7
10.9
12.3
13.2
1Z6
13.4
12.9
61-56
20,664
14,106
7,977
2,333
66-60
7.2
28-30
61-66
4.1
Over 65
1.2
Total
196,667
100.0
46-60 .
Data are as of May 1, 1938 and cover employees carrying group life insurance; on date indicated, insured
employees represented about 97% of ^;he total of all employees.
EMPLOYEES BY AGE GROUPS
U.S. STEEL CORPORATION AND SUBSIDIARIES
MAY 1, 1938
I
mom
5 !5 S S S S S
•S cScMMrr}?*!??*}?
YEARS OF AGE
HALF OF EMPLOYEES ARE OVER 40 YEARS OF AGE
About half the employees of U. S. Steel Corporation are over 40 years
of ago. There are more employees between the ages of 41 and 45 than in
any other group.
CONCENTRATION OF ECONOMIC POWER 13875
Skilled, semi-skilled and common labor employees — U. S. Steel Corporation and
subsidiaries — year 1938
Class of Employees
Number of
Employees
Per Cent
of Total
Skilled and Semi-Skllled
179, 672
22,436
88 0
Common Labor . ....
11 1
Total
202,108
100 0
Number of employees represents the average number on rolls during the year.
Administrative and sales employees, representing 6% of the total in 1938, are included with skilled and
semi-skilled employees.
SKILLED, SEMI-SKILLED AND COMIVION LABOR EMPLOYEES
U. S. STEEL CORPORATION AND SUBSIDIARIES
YEAR 1938
COMMON LABOR
EMPLOYEES
11?5
SKILLED AND
SEMISKILLED
EMPLOYEES
89«
The steel industry Is one that requires skilled labor. In 1938 about ^'i%
of all employees of U. S. Steel Corporation and subsidiaries were stilled
or semi-sltilled.
13876 CONCENTRATION OF ECONOMIC POWER
Payroll and component factors — U. S. Steel Corporation and subsidiaries
Payroll
(000
Omitted)
Number
of Em-
ployees
Earnings
Hour
Hours
Index Nos. W29=100
Pay- No. of
roU Em pi's
E. per H. per
Hour Week
1920
1930
1931
1932
Jan..
Feb.
Mar.
Apr.
May
Jun..
Jul-
Aug.
Sep..
Oct..
Nov.
Doc.
Jan..
Feb.
Mar.
Apr.
May
Jun..
Jul..
Aug.
Sep..
Oct..
Nov.
Dec.
Jan..
Feb.
Mar.
Apr.
May
Jun..
Jul..
Aug.
Sep..
Oct..
Nov.
Dec.
Jan..
Feb.
Mar.
May
Jun..
Jul..
Aug.
Sep..
Oct..
Nov.
Dec.
253, 138
251, 782
215, 649
164,319
(Above represent average
monthly payroll)
$35,006
32,606
22,239
11,159
$0,686
0.6S7
0.691
0.614
46.2
43.3
34.3
25.4
$9,070
151,010
$0,569
23.8
26
60
83
8,705
147, 369
0.677
25.6
26
68
84
8,557
139,585
0.573
24.2
24
55
84
8,895
143,922
0.561
25.7
25
57
82
10,807
150, 651
0.543
29.8
31
60
79
13, 769
170,767
0.533
sa.3
39
67
■78
17,663
190, 170
0.663
37.4
50
75
82
19, 591
200,749
0.596
37.0
66
79
87
17:238
200,633
0.827
32.0
49
79
91
17,095
189 696
0.646
31.5
49
75
94
15,561
192,438
0.659
2S. 6
44
76
96
16,198
189,229
0.658
29.4
46
76
96
$15,322
185,433
$0,660
28.3
44
73
96
16,497
186, 686
0.666
31.2
44
74
97
18,231
190, 153
0.666
33.0
52
75
95
19, 705
194,669
0.720
32.8
56
77
105
22,648
204,033
0.710
35.3
65
81
103
22, 741
207, 731
0.706
36.1
66
82
103
17,627
203,416
0.717
27.2
50
80
106
17,282
192,038
0.723
28.1
49
76
105
14. 972
186,693
0.730
25.8
43
73
106
16, 779
178,426
0.723
27.6
45
70
105
15, 226
175, 737
0.730
27.6
43
69
106
15, 572
174, 350
0.731
27.6
44
69
107
$19, 080
181, 825
$0. 725
32.7
65
72
106
19,733
191,026
0.734
36.2
56
76
107
21,411
194, 992
0.727
34.1
61
77
106
20,752
196, 260
0.730
33.8
59
78
106
21 2.S8
197,284
0.727
33.5
61
78
106
19,930
197.269
0.733
32.1
67
78
107
20,212
195,940
0. 725
32.2
58
77
106
21, 427
196,534
0.725
33.9
61
78
106
20,589
196, 524
0.730
33.5
59
78
106
22, 684
195, 585
0.733
35.7
65
77
107
22, 014
197, 984
0.742
34.9
63
78
108
22,455
195, 975
0.739
35.1
64
77
108
$22, 896
195, 858
$0. 733
36.0
65
77
107
22, 201
197,463
0.733
38.3
63
78
107
24, 167
200, 709
0.728
37.3
69
79
106
26,662
211,008
0.730
40.5
76
83
106
28,009
219,664
0.724
39.8
80
87
106
28,912
222, 979
0.737
41.0
83
88
107
29, 567
229, 452
0.742
39.3
84
91
108
30,128
234, 972
0.739
39.2
86
93
108
30, 097
237, 570
0.725
40.8
86
94
106
31, 774
238,944
0.719
41.7
91
94
105
31,119
240, 014
0.750
40.3
89
95
106
33,336
238. 781
0.778
40.6
96
94
113
CONCENTRATION OF ECONOMIC POWER 13877
Payroll and component factors — U. S. Steel Corporation and subsidiaries — Contd.
Pajrroll
(000
Omitted)
Number
of Em-
ployees
Earnings
Hour
Hours
wiek
Index Nos
. 1929 = 100
'„T
No. of
Em pi's
E.per
Hour
H. per
Week
1937
Jan
$33,177
32,293
39,065
40,857
41,906
42,517
42,383
42,394
39,625
34,484
28,749
26,477
240,359
244,602
254,911
260,565
267,052
272, 655
276, 897
278, 178
276,202
270, 515
256,788
236,665
$0,780
0.785
0.831
0.895
0.894
0.901
0.894
0.877
0.863
0.866
0.877
39.9
42.0
41.8
40.9
39.6
40.6
38.6
38.5
3313
30.3
27.9
95
92
112
117
120
121
121
121
113
99
82
73
95
97
100
103
105
108
109
110
109
107
101
93
114
114
121
131
130
130
131
130
126
126
128
Feb . ..
Mar
90
Apr
89
Xy "
86
Jun
88
Jul
83
Sep".: : :
Oct
Nov
Dec —
1938
Jan
$23,560
20,923
23,842
22,655
22,321
21,961
21, 194
22,897
23,146
25,358
27,186
27,177
220,270
210, 580
204,819
203,816
201,623
196, 898
192,021
191,311
192, 321
197,271
206,709
207,370
$0,887
0.896
0.892
0.908
0.910
0.919
0.931
0.906
0.901
0.890
0.892
27.2
27.7
29.4
28.6
27.5
28.3
26.8
29.8
31.2
34 5
33.3
67
60
68
65
64
63
61
65
66
73
78
78
87
83,
81
81
80
78
76
76
76
78
•81
82
129
131
130
132
133
134
136
132
131
130
130
130
69
Feb
60
Mar
Apr
Uny
62
60
Jun
Jul
68
Aug
65
Sep.' ::.: :"
68
Oct.
Nov
Dec
72
1939
Jan
$27,223
26,344
29,499
26,404
26,684
28,308
27.884
30,980
206,716
208,995
211,011
195, 764
212,381
213,728
214,205
214,108
$0,894
0.899
0.895
0.902
0.890
0.899
0.912
0.896
33.2
35.1
35.3
33.5
31.9
34.4
32.3
36.4
78
75
84
73
76
81
80
88
82
83
83
77
84
84
85
85
130
131
130
131
130
131
133
131
72
Feb
76
Mar
76
Apr
73
May-- "-:-:-:■'-
Jun - .
Jul
Aug
79
Payroll figures include construction payroll.
Number of employees represents number on rolls duiing month.
13878
CONCENTRATION OF ECONOMIC POWER
PAYROLL AND COMPONENT FACTORS
U. S. STEEL CORPORATION AND SUBSIDIARIES
1929 = 100
PAYROL
f\
"*
~N
/
' \
kJ
\
A
/Aj/"
/
\
tJ'"
\
V
h
V.
J
1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940
EARNINGS .^
PER HOURr^-
-^
-«A
K^
, .
y ^ NUMBER OF
y\. run i-ivrcc
■^
>N^
-^
/^
^
^M-
^
S^
vA
A
fT
^ HC
PER
URS \
week\
7
AA
^-11
>f
v
1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940
. D*T» 192»I932. MOHTHIT DAT* WOXKXUi
Total payroll is affected by throe factors, (a) number of employees, (b)
number of hours worked, and (c) hourly wage rates.
Since 1929. hourly earnings of employees of U. S. Steel Corporotion
have been well maintained and since 1936 have been approximately 30%
above the 1929 level. Decreases in total payroll in the depression periods
have largely been due to reductions in number of employees and hours
worked per week, made necessary by the absence of orders for steel
products.
CONCENTRATION OF ECONOMIC POWER 13879
Wages and hours — U. S. Steel manufacturing subsidiaries
Ave
Year or Month ^^|?
Ho
rage
tags
sr
ur
Average
Hours
Per
Week
Average
Earntags
Per
Week
Year or Month
Average
Earnings
Per
Hour
Average
Hours
W^le'k
Average
Earntags
Per
Week
1929 $0
641
634
615
618
625
632
660
676
825
843
580
586
586
650
661
646
639
648
647
660
661
663
663
661
669
662
660
656
660
659
660
664
667
664
663
658
669
665
661
680
687
47.5
43.2
31.3
22.0
29.4
28.6
33.4
40.2
37.4
27.5
26.3
29.7
32.1
32.4
35.6
35.9
24.6
25.3
23.1
25.2
27.3
28.2
32.1
34.3
33.5
33.5
32.8
30.5
31.2
33.6
33.2
35.6
36.1
35.2
36.5
37! 9
41.4
40.6
41.6
39.4
$30.47
27.38
19.28
11.42
15.43
18.09
22.02
27.20
30.84
23.16
15.27
17.39
18.79
21.03
23.14
23.18
15.63
16.37
14.94
16.39
17.77
17.10
20.97
22.66
22.06
22.19
21.65
20.01
20.25
22.12
21.95
23.66
23.41
23.37
24.17
25.25
24.99
27.53
26.77
28.20
27.07
Aug
$0,683
.664
.660
.691
.726
.726
.730
.794
.865
.872
.872
.881
.871
.840
.814
.805
.806
.808
.820
.834
.853
.856
.868
.877
.863
:iro
.839
.832
.832
.836
.841
.846
.848
:iJ
.862
39.6
41.2
42.6
41.2
41.6
41.4
43.1
43.0
42.1
40.0
40.8
38.0
38.1
37.4
31.7
28.1
25.0
24.1
24.7
27.0
26.1
24.9
26.4
23.8
27.4
29.2
31.3
33.7
32.2
31.9
33.3
33.9
31.6
31.7
32.1
30.0
35.4
$27.04
Oct
Nov
Dec
30.07
Jan. 'W
30.06
Feb
31.46
1936
Mar
34.11
1937
Apr
36.43
1938
May.......
34.88
Jan '34
Jun"
35 66
Feb
Jul
33 61
Mar
Aug
33.19
Anr
Sep
31 38
^y
0(*" ""■
25 80
jSn^:::::::::::::::
Nov
22.61
Jul
Dec.
20.13
Jan. '38
19.44
Sep;.::::::::::::::
Feb
20.21
o5
Mar
22.63
Nov
Apr
22.27
Dec
^ay ' '"■
21.31
Jan. '35
Jun
22.03
Feb
Jul
20.89
M„
Aug
23.37
Apr
Sep
24.66
^y :
Oct :
26.26
jSn
Nov
m:3o
jSi
Dec
26.82
Aug
Jan. '39
26.54
Feb . ..
27.84
qK
Mar
28.64
Nov
26.62
Dec
May
28.90
Jan. '36
Jun .. .. .
27.83
Frt.
Jul : ::::
26.42
Ma
Aug .. .:
30.49
Apr
Sep . . ..
wii : :
Oct :: ::
-Tnn
Nov . . . .
Jul.-
Dec
Data are for wage earners only, exclusive of all salaried employees.
Figures prior to 1933 are partially estimated, based on samples.
13880
CONCENTRATION OF ECONOMIC POWER
WAGES AND HOURS
U. S. STEEL CORPORATION MANUFACTURING SUBSIDIARIES
AVERAGE EARNINGS PER HOUR
1.00
« so
i:
.20
0
60
50
« 40
a:
g 30
^ 20
10
0
40
30
3 20
o
10
0
ANNUAL DATA 19M
1.00
so «
.60 5
.40 g
20
0
60
50
40 „
30 i
20 ^
10
0
40
30
20 1
10
r\
y^
^^
_^
_^
J ^
^
1929 '30 31 '32 '33 '34 '35 'SS '37 '38 '39 1940
AVERAGE HOURS PER WEEK
^
^ A
'\,
\
A
hi/*
^
^
^
'I
J
1929 -30 '31 '32 '33 '34 '35 '36 '37 '38 '39 1940
AVERAGE EARNINGS PER WEEK
A
N
s.
A
0''
/^
\
J
jj
\
V
^
/I
f
1929 '30 '31 '32 '33 '34 "35 '36 '37 '38 '39 1940
•1933. MONTHLY DATA THEREAHER
Average earnings per hour of wage earners of U. S. Steel Corporation manufacturing subsidiaries are
about 30% higher today than in 1 929.
Due to the Corporation's policy of sharing the available work so far as practicable among the maximum
number of employees during depression periods, average hours per week reached a level in 1938 almost
as low as that of 19^2. With the improving demand for steel whi..,i began in the Fall of 1938, more work
became available and by August 1939, average hours per week had increased to about 35'/2-
Average earnings per week at the low point of 1938 were substantially above the low point of 1932, on
account of the higher hourly rate. In August 1939, despite a twelve hour shorter work week, weekly
earnings were slightly higher than in 1929.
CONCENTRATION OF ECONOMIC POWER
13881
(Vages and hours — U. S. Steel Corporation manufacturing subsidiaries and all
manufacturing industries
Year or Month
1929.
1930.
1931.
1932.
1933.
1934.
1935.
1936.
1938
Jan., '34.
Feb
Mar
Apr
May.....
Jun
Jul
Aug
Sep
Oct
Nor
Dec
Jan., '35.
Feb
Mar
Apr
May.. .
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan., '36.
Feb
Mar
Apr
May-.
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan., '37.
Feb
Mar
Apr -
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan., '38. .
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan., '39. .
Feb
Mar
Apr
May.....
Jun
Jul
Aug
Average Earnings per
Hour
U. S. Steel All Mfrg,
$0,641
61.-)
518
625
$0,460
632
.548
660
.568
676
.676
m
.643
843
.646
680
.533
685
.631
586
.631
650
.541
651
.561
645
.550
639
.556
648
.555
647
.559
av)
.553
651
.564
6,53
.560
653
.564
47.5
43.2
31.3
22.0
29.4
33^4
40.2
37.4
27.5
29! 7
32.1
32.4
36.6
35.9
24.5
25.3
23.1
25.2
27.3
26.2
32.1
34.3
33.5
33.5
32.8
30.5
31.2
33.6
33.2
35.6
36.1
35.2
36.5
38.4
37.
41.
40.6
41.5
41.2
42.6
41.2
41.5
41.4
43.1
43.0
42.
40.0
40.
38.0
38.1
37.4
31.7
25^0
24 1
24.7
27.0
26.1
24.9
25.4
23.8
27.4
29.2
31.3
33.7
32.2
31.9
33.3
33.9
31.5
31.7
32.1
All Mfrg.
37.9
34.7
36.6
38! 5
35.3
3518
36.3
36.2
35.4
34.9
33.4
34.0
34^3
34.1
35.2
35.2
36.4
36.6
36.4
35.8
35.4
35.2
36.6
37.4
38.2
37.8
38.7
37.3
37.4
38.6
38.7
39.2
39.2
38.5
39.4
38.7
40.5
40.6
41.1
40.4
41.0
40.4
39.8
39.2
37.9
37^4
37.6
35.4
34.4
33.2
34.3
34.5
34.2
34.4
34.4
34.7
36.3
36.9
37.4
36.5
37.1
36! 9
37.1
36.4
36.7
37.2
$30.47
27.38
19.28
11.42
15.43
18.09
22.02
27.20
30.84
23.15
15.27
17.39
18.79
21.03
23.14
23.18
16.63
16.37
14.94
16.39
17.77
17.10
20.97
22.66
22.06
22.19
21.65
20.01
20.25
22.12
21.95
23.66
23.41
23.37
24.17
25.25
24.99
27.53
26.77
28.20
27.07
27.04
27.35
30.07
30.06
31.45
34.11
36.43
34.88
35. 55
33.51
33. 19
31.38
25.80
22.61
20.13
19.44
20.21
22. .53
22.27
21.31
22.03
20.89
23.37
24.65
26.26
28.30
26.82
26.54
27.84
28.54
26.62
26.90
27.83
26.42
30.4'J
13882
CONCENTRATION OF ECONOMIC POWER
Wages and hours — U. S. Steel Corporation manufacturing subsidiaries and all
manufacturing industries — Continued
Year or Month
Average Earnings per "
Hour
AveraM Hours per
Average Earnings per
Week
U. S. steel
All Mfrg.
U. 8. Steel
All Mfrg.
U. 8. Steel
All Mfrg.
Oct
Nov
Deo
Source: Corporation recordsland U. S. Bmeau of Labor]Statistlcs.
U. 8. Steel data are for wage earners only, exclusive of salaried employees; data prior to 1933 are partially
estimated, based on samples.
Data for all manufacturing industries are those of U. 8. B. L. 8. and cover wage earners In 89 manufactur-
ing Industries; data are not available prior to 1932.
WAGES AND HOURS
U. S. STEEL M'FR'G SUBSIDIARIES AND ALL MTR'G INDUSTRIES
AVERAGE EARNINGS PER HOUR
1
I. S. STEEL /^
^^
=:=_
—
^
-^MANUFACTURING
1
-7
•32 '33 '3* -35 '36 '37 "38 39 1940
60p
A
VER/
^GE
HOU
RS
>ER WEE
K
r-i
^^
ALL
\,
^\
f^
^x
^
Zj
30
\y
y^
'\f
U. S.' STEEL
^•
.
1929 -30 '31 -32 '33 '34 '35 '36 '37 '38 '39 1940
AVERAGE EARNINGS PER WEEK
...
Ja
N
V
A
MAr
UFAC
h
^
fJ
s
-^
/b
TURI
IG
....
20 :;
1929 '30 '31 '32 '33 '34 '35 '36 '37 '38 '39 1940
ANNUAl AND MONTHLV DATA SHOWN Soufrt. CofiaMoo itmik and USB LS.
Average earnings per hour of wage earners in U. S. Steel Corporation manufacturing subsidiaries a'e
considerably higher than earnings per hour in manufacturing industries generally.
Average hours per week in Corporation subsidiaries tend to fluctuate with hours per week 'n ail manu-
facturing industries but in greater degree. In periods of low operation, weekly hours in Corporation,
subsidiaries are below those in all manufatturlng industries, and In periods of high operation they are
above.
Average earnings per week in Corporation subsidiaries tend to be above those in all manufacturing
industries even though hours per week in the former are the lower, as in 1939. This is because of the
higher hourly rate In Corporation subsidiaries which, in 1939, was more than 20 cents an hour above that
In manufacturing industries generally.
CONCENTRATION OF ECONOMIC POWER 13883
Average weekly earnings compared with cost of living — wage earners of U. S. Steel
Corporation subsidiaries
Cost of Living
Casli Earnings
Real Earn.
Year
1923-25
=100
1929=100
Dollars per
Week
1929=100
ings
1929-100
1929
99.3
97.2
88.9
80.3
75.7
78.3
80.6
81.9
84.0
83.6
100. D
97.9
89.5
80.9
7«.2
78.9
81.1
82.5
84.6
84.1
30.32
27.19
19.44
11.48
15.00
17.93
21.63
26.79
29.93
23.06
100.0
89.7
64.1
37.9
49.5
59.1
71.3
88.4
98.7
76.1
100.0
1930
91.6
1931
71.6
1932
46.8
1933
65.0
1934
74.9
1936
87.9
1936
107.2
1937
116.7
1938
90.5
Source: Corporation records and U. 8. B'lreau of Labor Statistics.
Yearly cost of living data computed from U. 8. B. L. S. first, middle and Iflit Of year data by weiigblng flrgt
and last of year figures by one and the middle of the yeer figure by.two.
Real earnings equal cash earnings dlTided by cost if living.
AVERAGE WEEKLY EARNINGS COMPARED WITH COST OF LIVING
WAGE EARNERS OF U. S. STEEL CORPORATION SUBSIDIARIES
1929=100
120
120
100
80
60
40
20
0
RE
ALEAfiN
NGS^
Ul
i 80
a
2 40
20
0
I
b
1ST OF U
VING
y
^
/
-5^
y
V
UJ
i
3
Z
'x
o
2
^
s.
r-
\
<i
^<.
KEARNII
SS
"-
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
MOTC: KM. EAKNINeSEQWl CASH EARNINGS WVIOEB BY COST Of UVINO
Sout..- Cop<»l«n«a«4«/USffXS
In 1938, average weekly earnings of all wage earners of U. S. Steel
Corporation subsidiaries were more than 20% below those in 1929.
However, since the cost of living in 1938 was also below that in 1929.
real weekly earnings (i. e., cash earnings divided by cost, of living) of
Corporation wage earners were only 10% below the 1929 level.
With improved business conditions in 1939. weekly earnings of Corpora-
tion wage earners had recovered by August to approximately the 1929
level. At that time, the cost of living was still about 1 5% below the 1929
average, so that real eamings of Corporation wage earners were nearly
20% above the 1 929 point.
-124491— 41— pt. 26-
13884
CONCENTRATION OF ECONOMIC POWEP
V/age rates and steel prices
(Wage rates=U. S. Steel Corporation Mfrg. subidlaries' basic rate per hour for common labor (Pitts-
burgh district); steel prices=Iron Age composite price of finished steel]
STEEL PRICES (IRON AOE)
Year
Price Per
Net Ton
1914 = 100
Year
Price Per
Net Ton
1914 > 100
$44.14
40.80
41.14
37.36
34.14
35.20
35.40
38. M
37.30
32.64
33.52
30.84
30.54
33.22
28.66
30.66
53.34
83.82
70.84
62.30
154. 01
142.36
143. 55
130.36
119. 12
122.82
123. 52
134. 19
130. 13
113.89
116.96
107.61
a06.56
115.91
100.00
106.97
186.11
292.46
247.17
217.38
1920
$74.74
48.74
42.48
53.94
50.10
46.68
46.30
44.04
43.30
44.18
40.96
39.14
38.02
37.68
40.66
41.16
41.54
49.28
47.88
46.20
260.78
1901
1921 -
170.06
1902
1922 ..-
148.22
1903
1923
188.21
1904
1924
1925
174.81
1906
162.88
1906 - ----
1926 .- -
161. 66
1907
1927
153 66
1906
1928
151 08
1909
1929....
1930
154.16
1910
142 92
1911
1931
136 57
1912
1932
1933
1934
132. 66
1913
131 12
1914
141.87
1915
1936
143. 61
1916
1936
144 94
1917
1937
171. 95
1918
1938
167.06
1919
1939 >
157. 71
' steel price data for 1939 are basis of first 9 months.
WAGE RATES (U. S. S. C. Subs.)
DateEflective ^HoiJ*'^ 1914-100
Jan. 1,1901..
Jun. 1,1902..
Jan. 1, 1904..
Apr. 1, 1905..
Jan. 1,1907..
May 1, 1910..
Feb. 1, 1913..
Feb. 1, 1916..
May 1, 1916..
Dec. 16, 1916.
May 1, 1917..
Oct. 1, 1917..
Apr. 16, 1918.
Aug. 1, 1918..
Oct. 1, 1918- .
75.0
80.0
72.5
77.5
82.6
87.6
100.0
110.0
126.0
137. S
150.0
165.0
190.0
210.0
234.1
Date Effective
Feb. 1, 1920..
May 16, 1921.
Jul. 16, 1921..
Aug. 29, 1921.
Sep. 1, 1922. -
Apr. 16, 1923.
Aug. 16, 1923.
Oct. 1,1931--
May 16, 1932.
Jul. 16, 1933.-
Sep. 16, 1933.
Apr. 1, 1934--
Nov. 16, 1936
Mar. 16, 19.37
Rate Per
Hoi^
$0. 5128
.4125
.37
:30
256.4
206.2
185.0
150.0
180.0
200.0
220.0
195.0
166.0
200.0
212.5
235.0
262.6
312.5
From August 16, 1923 to September 16, 1933, common labor is that of 10 hour men; subsequent to Septem-
ber 16, 1933, rat" is that of 8 hour, non-continuous labor.
CONCENTRATION OF ECONOMIC POWER
13885
WAGE RATES AND STEEL PRICES
1914 = 100
300
^ 250
er
i 200
Z
300
.
f
I
250 ^
X
LU
00
200 S
Z
150 X
UJ
100 ^ .
50
0
y
i4
re
WAG
tRATi
i-i
;!
.--
i
TE
EL
PI
m
E£
•
f
>
1
1
-.^
UJ
O
^ 100-
50 -
^>
S ^
-
>
^
^
"
j
I
»v
-'
-'
■■-
r-
-
•■
r
'
. U S S<~/ G.pc«,«y, ™.,y««^ »iW..«,, Rmto,9* <t*rt, !«< ,« p« /«.,
^ 1 s II
.feroynmonbtor
1
Wage rates today, as represented by the common labor rate per houc
of manufacturing subsidiaries of U. S. Steel Corporation in the Pittsburgh
District, are more than ff ur times as high as at the beginning of the cen-
tury. Current steel pric js, as represented by the Iron Age composite
price of finished steel, jre practically at the same level as in 1900, with-
out attempting any adjustment for the great improvement in quality
which has occurred during the inter/al.
13886
CONCENTRATION OF ECONOMIC POWER
Average earnings per hour and common labor rate- — U. S. Steel Corporation
manufacturing subsidiaries
AVERAGE EARNINGS PER HOUR-ALL WAGE EARNERS
1929
1930
1931
1932
1933
1934:
Jan
Feb
Mar
Apr -_.
May --.
Jun
Jul
Aug
Sep
Oct
Nov
Dec
1935:
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
1936:
Jan -
Feb
Mar
Apr
May
Jun
Jul
Aug
DaU prior to 1933 partially estimated,
$0. 641
.634
.615
.518
.526
.580
.585
.586
.650
.651
.645
.639
.648
-.647
.650
.651
.653
.653
.661
.659
.662
.660
.656
.650
.659
.660
.664
.667
.658
.659
.665
.661
.680
.687
.683
1936:
1937:
Sep..
Oct..
Nov-
Dec-
$0.
Jan.-
Feb..
Mar.
Apr..
May.
Jun..
Jul..
Aug.
Oct.-
Nov.
Dec.
1938:
Jan..
Feb..
Mar.
Apr..
May.
Jun._
JuL-
Aug-
Sep--
Oct..
Nov.
Dec-.
1939:
Jan.-
Feb..
Mar.
Apr..
May.
Jun..
Jul._
Aug.
664
660
691
725
726
730
794
865
872
872
881
871
840
814
805
805
820
834
853
856
868
877
853
844
840
832
836
841
846
848
862
based on samples.
COMMON LABOR RATE-COMMON LABOR EMPLOYEES IN PITTSBURGH DISTRICT
Date Effective:
Jan. 1, 1901.-
Jun. 1, 1902- -
Jan. 1, 1904- -
Apr. 1, 1905- -
Jan. 1, 1907..
May 1, 1910--
Feb. 1, 1913.-
Feb. 1, 1916- .
May 1, 1916-.
Dec. 16, 1916.
May 1, 1917-.
Oct. 1, 1917..
Apr. 16, 1918.
Aug. 1, 1918-.
Oct. 1, 1918..
From August
September 16, Ifl
Rale per Hour
.. $0. 15
-- . 16
-- . 145
.. .155
.. .165
.. .175
-. .20
-- .22
.- .25
.. .275
-. .30
.. .33
-- .38
.. .42
-- .4683
Date Effective— Con. R<^' ver Hour
Feb. 1, 1920 $0. 5128
May 16, 1921.
Jul. 16, 1921-.
Aug. 29, 1921.
^p. 1, 1922..
Apr. 16, 1923.
Aug. 16, 1923-
Oct. 1, 1931 -.
May 16, 1932.
Jul. 16, 1933-
Sep. 16, 1933.
Apr. 1, 1934..
Nov. 16, 1936.
Mar. 16, 1937.
4125
37
40
425
47
525
625
6, 1923, to September 16, 1933, common labor rate i
3, rate is that of 8-hour, non-continuous labor.
that of 10-hour men; subsequent to
CONCENTRATION OF ECONOMIC POWER
13887
AVERAGE EARNINGS PER HOUR AND COMMON LABOR RATE
U. S. STEEL CORPORATION MANUFACTURING SUBSIDIARIES
1
A
A-
^
1
1
}
—
—
— V
EARNINGS __^
PER HOURp^
^
r~"
_^_
>
\__
_^
r
i
—
—
' 'l
J"
J-
COMMO^
BOR RA
1 J
"L.
J
1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940
The common labor rate at Pittsburgh Is a goo?i measure of the trend of
earnings per hour of employees of U. S. Steel Corporation manufacturing
subsidiaries.
Vacation wages paid in the summers of 1937, 1938 and 1939 caused the
Increases in earnings per hour during those periods. Vacation wages paid
to wage earners of manufacturing subsidiaries in 1937 and 1938 amounted
to $3,744,346 and $3,1 14,134, respectively.
13888
CONCENTRATION OF ECONOMIC POWER
Earnings per hour and steel prices — Earnings per hour — Earnings per hour of all
employees of U. S. Steel Corporation and subsidiaries; steel prices — Iron Age
composite price of finished steel
Hourly Earnings
cents
perHr.
1926=
100
102.8
103.0
103,6
92.1
cents 1926=
per lb. 100
2.209
2.048
1.957
1.901
Afi.9
85.3
1.885
S7.7
86.5
1.873
fi7.?.
85.9
1.867
fifi.l
84.1
1.817
64.8
81.4
1.802
53. 3
79.9
1.820
f)«.3
84.4
1.878
59.6
89 4
62.7
94.0
1 890
«4.6
96.9
1.950
6S.9
98.8
1.933
65.8
98.7
1.945
109.9 2.062
109. 9 2. 040
95.4
88.5
84.5
82.1
81.4
80.9
80.6
78.5
77.8
78.6
81.1
81.3
81.6
84.2
83.5
84.0
66.0
99.0
1.946
84
66.5
99.7
1.946
84
«.■;, 5
98.2
1.94S
84
72.0
107.9
1.988
85
71.0
106.4
2.118
91
70.6
105.8
2.118
91
71.7
107.5
2.056
88
72.3
108.4
2.056
8«
73.0
109.4
2.056
88
72. 3
108.4
2.056
KM
73. 0
109.4
2.056
73.1
109.6
2.056
88
72.5
108.7
2.056
88
n 4
110.0
2.056
72,7
109.0
2.056
88
73,0
109.4
2.056
88
72.7
109.0
2.056
88
7;<.3
109.9
2.056
88
72.5
108.7
2.056
8H
72.5
108.7
2.066
88
73.0
109.4
2.056
88
n 3
109.9
2.062
89
74.2
111.2
2.062
89
73.0
110.8
2.062
89
Mar.
Apr.
May
Jun.
Jul.
Aug.
Sep.,
Oct.,
Nov.
Dec.
Jan..
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep.
Oct..
Nov.
Dec.
Jan.,
Feb.
Mar.
Apr.
May
Jun.
Jul..
Aug.
Sep.
Oct.
Nov.
Dec.
Jan..
Feb..
Mar..
Apr..
May.
Jun..
Jul...
Aug..
Hourly Earnings Steel Prices
cents 1926=
perHr. 100
cents 1926-
per lb. 100
72.8
109.1
2.021
73.0
109.4
2.028
72.4
108.6
2.028
73,7
110,6
2.033
74.2
111,2
2.091
73.9
110,8
2.091
72.5
108.7
2.096
71.9
107.8
2.116
75 0
112.4
2.116
77.8
116,6
2.199
88.7
133.0
2.512
89.6
134.3
2.512
89.2
133.7
2.512
90.8
136.1
2.512
91.0
136.4
2.606
91.9
137.8
2.459
93.1
139.6
2,300
90.6
135.8
2.300
90.1
135.1
2.293
89.0
133.4
2.255
89.2
133.7
2.286
88.9
133.3
2.286
89,4
134,0
2.286
89,9
134.8
2.286
89.5
134.2
2.286
90.2
136.2
2,286
89.0
133.4
2.266
89.9
134.8
2.236
91 2
136.7
2.236
89.6
134.3
2.236
87.3
87.6
87.6
87.8
90.3
90.3
90.5
91.4
91.4
95.0
78.0
116.9
2.249
78,5
117.7
2.249
83,1
124.6
2.459
1
89,8
134.6
2.612
1(
89,5
134.2
2.512
V
89.4
134.0
2.512
90.1
136.1
2.612
\{
89,4
134.0
2.512
V
87.7
131.6
2.512
V
86.3
129.4
2:512
V
86.6
129.8
2.512
I
87.7
131.5
2.512
u
97.1
97.1
06.2
08.5
.08.6
08.6
6
.08.6
08.6
08.6
08.5
08.5
08.6
08.6
08.5
08.6
08.3
06.2
99.4
99.4
99.0
97.4
98.7
98.7
98.7
98.7
98.7
98,7
97.6
96.6
99.6
96.6
Steel prices are monthly averages of weekly figures.
The 1926 base for earnings per hour of all employees of U. S. Steel Corporation and subsidiaries was
estimated from data on the total steel industry compiled by National Industrial Conference Board, as Cor-
poration data are not available prior to 1929.
CONCENTRATION OF ECONOMIC POWER
13889
Earnings per hour and production — U. S. Steel Corporation and subsidiaries —
April 19S7-November 1939
1937
1938
1939
Month
Earnings
Per Hour
Thousands
or Net Tons
Produced
Earnings
Per Hour
Thousands
0/ Net Tons
Produced
Earnings
Per Hour
Thousands
of Net Tons
Produced
Jan
$0 887
537
467
687
532
506
522
466
623
625
734
852
777
$0,894
.899
.895
.902
.890
.899
.912
.896
.893
.890
.893
9S1
Ft.:.:
896
892
908
910
919
931
906
901
890
892
Mar.
993
1,483
1,452
1,438
1,278
1,247
1,154
869
648
S^y::::::::::::::::: ::::::::
895
894
901
894
877
863
877
Jul
879
Aug
1 033
Sep —
1 198
Oct :: :" "
1 537
Nov
1,629
Dec
Average relationship: Earnings per hour=$0.901 minus $0.0007 for each hundred thousand net tons
produced.
Earnings per hour are average hourly earnings of all employees.
Production data represent monthly production of rolled and finished steel products.
100
90
80
§ 70
I ^
§ 50
{2 40
g 30
20
10
0
EARNINGS PER HOUR AND PRODUCTION
U. S. STEEL CORPORATION AND SUBSIDIARIES
APRIL 1937 - NOVEMBER 1939
1 1
AVERAGE
.
.^
100
90
80
70 -
60 S
50 ^
40 g
30 1
20
10
0
REL
ATION
SHIP
♦
♦
*~
• 1939 MONTHS
■ 1938 MONTHS
♦ 1937 MONTHS
-J
c
.1 .2 .3 .4 .5 .6 .7 .8 .9 1.0 1.1 12 1.3 1.4 1.5 1.6 1.
MILLIONS OF NET TONS
i/e A^tago H>j'lf £a,r,r,g, cf All Empl^ms
7
While average hourly earnings of all employees of United States Steel
Corporation and subsidiaries vary somewhat from month to month, the
voriations are small and bear very little relation to the rate of production.
Averoge hourly earnings tend to be less than one cent per hour lower
when operations ore at 90% of capacity than when operations are af 25%
of capacity.
13890 CONCENTRATION OF ECONOMIC POWER
Ingot production and number of employees, total steel industry
Item
1929
1937 1937 In
(9 months) % of 1929
Monthly Ingot Production (Tons)
4, 670, 869
428,319
4, 798, 000 105
NllPlN^r of nimplnypAa
521 303 122
Ingot production represents average monthly production of all ingots, and excludes production of castings.
Number of employees represents number of wage earners of iron and steel manufacturing companies
reporting to American Iron and Steel Institute (1929 figure partially estimated, based on data of U. 8.
Census of Manufactures on wage earners in Blast Furnaces, Steel Works and Rolling Mills).
CONCENTRATION OF ECONOMIC POWER
13891
INGOT PRODUCTION AND NUMBER OF EMPLOYEES
TOTAL STEEL INDUSTRY
1929 = 100
INGOT NUMBER (
PRODUCTION EMPLOYEE
120 [ ■
n I
)F
S
-120
NUMBERS
1 1 1
III
-100
-80.
-60
UJ
m
2
z>
z
■SSi
ggg
X
LU
Q
■z.
- 40-
20-
1
-40
-20
X
UJ
Q
z
1
i
0-
]
is
92<
3 1
(9(
93-
/lONT
7 1925
HS)
) 193:
(9 MONT
-0
^S),
TECHNOLOGICAL ADVANCES HAVE NOT RESULTED IN
FEWER EMPLOYEES
From 1929 to 1937, a period morked by important installations of continu-
ous types of equipment to keep pace with tecfinological advances, employ-
ment increased relatively more in the steel industry generally thap it did
in the U. S. Steel Corporation. Average monthly ingot production for the
industry in the first nine months of 1937 was 5% more than in 1929, whereas
the average number of employees in the first nine months of 1937 was
22% more.
13892
CONCENTRATION OF ECONOMIC POWER
ill
I
til tb ^ ^
rr^
CONCENTRATION OF ECONOMIC POWER 13893
Exhibit No. 1410
SOME FACTORS IN THE PRICING OF STEEL
This is an analysis made in connection with studies by the United States Steel
Corporation in preparation for hearings on the steel industry before the Temporary
National Economic Com r.ittee.
October 30, 1939.
table of contents
Introduction
The Demand for Steel
Immediate Source of Demand
Geographic Distribution
Characteristics of Demand
Derived Nature of Demand
Durability and Demand
Postponability of Purchase of Durable Goods
Total Demand for Steel is Inelastic
Effect of the Substitution Factor
Potential Elasticity of Demand from a Particular Producer
The Supply of Steel
Geographic Concentration
Technological Aspects
Capital Investment Requirements
Factors in Expenditures for New Plants and Equipment
Source of Funds
Incentives for Investment— Profit Motive
Incentives for Investment— Obsolescence
Siie and Number of Producers
Channels of Distribution
Jobbers and Warehouses
Importance of Outlets
Summary
Characteristics of Cost in the Industry
"Overhead" or "Fixed" Costs
•Additional" Costs
Average Costs
The Dynamics of the Market for Steel
Costs and Demand
Psychological Factors
Chiaraeterlstic Patterns of Action by Sellers in the Market for Steel
The Basing Point Method of Quoting Delivered Prices
Economic Roots of the Basing Point Method
Relation of Competition to Profits, Capacity and Costs of Distribution
Profits
Capacity
Distribution Costs
Selling Expense
Freight Absorption
Conclusion
The Function of the Steel Industry in the National Economy
As a Source of Raw Material
As a Factor in Employment
As a Factor in the Growth of the Nation
Introduction
How much does the price of steel influence the quantity sold? What is the
relationship of cost to the price of steel? What degree of price competition
is desirable, and possible, in the steel industry? Why does the steel industry
quote delivered prices and why does it use the basing point method of quoting
delivered prices? Does the steel industry perform its proper function in the
national economy? Before these questions can be answered a careful analjTsis
must be made of the fundamental factors underlying the demand-supply situation
in the industry.
Subject to some exceptions with respect to particular products, the salient
characteristics of demand and supply in the steel industry may be summarized
as follows:
(1) The demand for steel is marked by tremendous cyclical fluctuations.
(2) The total demand for steel is inelastic, i. e., the total quantity of steel
bought from the industry would not be greatly different at any
particular time if the price were higher or lower.
(3) In contrast, the demand for steel from a particular producer usually
possesses great potential elasticity. In other wor.ds, buyers will
readily shift from one producer to another in response to a difference
in price. This is due to the informed character of the buying of steel.
Buyers have excellent technical knowledge of the product to be
purchased; and since nearly all steel is purchased on specification,
the identical grade and type of steel may be obtained for the most
13894 CONCENTRATION OF ECONOMIC POWER
part from any one of a number of producers. Furthermore, the large
size of individual purchases makes it worth-while for buyers to shop
for the lowest possible price.
(4) The cost structure in the industry is marked by substantial fixed costs
which must be met regardless of the amount of steel produced. * Even
more significant is the fact indicated by the operating experience
of the United States Steel Corporation and its subsidiaries over
the past ten years that the additional cost per unit of output re-
mains approximately the same regardless of the rate of operations
provided labor rates, prices of raw materials, etc. remain constant.
As a result of these two characteristics the average cost of each
unit of the entire output is higher than the additional cost per addi-
tional unit of output for practically the whole range of operations
up to the limits of practical capacity. Finally, the cost of labor and
of other goods and services purchased from others (which together
constitute about 80 percent of the total cost in the case of the sub-
sidiaries of United States Steel Corporation), are largely outside
the control of the management of the steel producer.
(5) Producers of the great bulk of the tonnage of steel products sold in the
respective consuming areas are relatively few in number.
These characteristics of the steel industry, of course, do not coincide with
the conditions necessary for the "perfect" price competition of classical economic
theory. The theory of "perfect" price competition, for example, assumes each
buyer and seller to be too small to influence the market price; any seller is supposed
to be able to reduce his price and expand his production without fear of reactions
on the part of competitors. This is not true of the market for steel. As a con-
sequence of potential shiftability of buyers in response to price concessions, there
is an incentive to obtain business bv price reduction even below average cost
as long as the price of the additional units so sold is above the additional cost
thereof, but in actual competition in the steel industry such a tenda,ricy is modified
to some extent by the difl3culty of continuing to offer lower prices than competitors
since competitors meet price concessions almost immediately. Furthermore,
"perfect" price competition does not take into account the consequences of the
presence in the market of relatively few, but large, buyers, nor the size of their
individual orders. It overlooks the relative difficulty of new producers entering
the market and many other factors of importance in the competitive situation in
the steel industry. In appraising this situation it should be recognized that the
conditions requisite for theoretically "perfect" price condition have rarely, if
ever, been approached in any industry, and could never be generally achieved
in a manufacturing industry such as steel. Accordingly, it is hardly reasonable
to judge competitive practives in the steel industry by imaginary standards based
on abstract conditions which cannot possibly be fulfilled, and which probably
never have been fulfilled in any industry.
Waiving the reasonableness of the application of the criteria, it is pertinent
to inquire what the consequence of "perfect" price competition would be in the
steel industry. If such a theoretical state of competition prevailed, each producer
would take aU the business he could get so long as the price yielded more than
the additional cost of producing the additional ton of steel so sold. If the demand
exceeded the capacity of existing producers, the price of steel would sky-rocket,
being limited only by the magnitude of the demand. If, however, the demand
declined to less than the existing capacity, the price would drop abruptly to the
level of the additional cost per additional unit of the least efficient producer
remaining in the market. In such a situation producers would cover little, if
any, of their overhead. Producers, therefore, would be operating at heavy
losses whenever existing capacity was not being fully utilized, and would recoup
these losses by high prices and large profits during the peak of prosperity. In
major depressions the efficient as well as the marginal concern would fail to sur-
vive unless it had accumulated an extraordinarily large cash balance.' Under
such conditions existing capacity would be reduced with the result that the steel
industry would become a bottle-neck in the succeeding rise in the business cycle
by limiting the possibility of increased production and creating a premature
boom in prices before the rest of the economy could achieve full employment.
' In the case of the subsidiaries of United States Steel Corporation these costs are approximately 30%
of total cost at 40% of capacity operations, 20% at 70% capacity and 16% at 100% capacity.
> If the subsidiaries of United States Steel Corporation sold steel at a price only equ-1 to the additional
cost of additional units of production, It Is estimated that the loss to the Corporation would be approximately
$182,100,000 a year. Under these conditions the Corporation could not survive for more than a few years.
CONCENTRATION OF ECONOMIC POWER
13895
Actually, of course, these characteristics of "perfect" price competition would
not be tolerated. The cut-throat struggle in depression and the sharp increases
in prices and profits in prosperity, as well as the bottle-neck in capacity, would
be the object of attacks by legislators, economists and others.
This paper is an attempt to outline the numberous factors involved in the pricing
of steel with the hope that a re-statement of fundamentals will contribute to a
clearer understanding of prices and price structure in the steel industry.
The Demand fob Steel
immediate source op demand
Orders for steel come mostly from companies using the products of the steel
ndustry as raw materials in making goods or as equipment in producing services.
Companies purchasing steel have been classified, and estimates of the percentage
of the total steel production of the United States purchased by each class have
been made as follows:
Percentage Distribution of Hot Rolled Iron and Steel Production Among Major
Consuming Industries •
Industry
1938
1932-38
Average
1926-31
Average
Automotive
17.3
18.8
6.1
9.1
4.7
7.4
7.5
3.5
3.6
1.6
0.3
20.1
20.8
16.0
10.1
8.4
6.0
6.0
5.5
4.2
3.6
0.9
0.6
18.0
16 3
Construction
19.9
Railroads
17.9
Container . .. ....
4.7
Agriculture .. . .
6 0
Oil, Oas and Water - .
8.3
Exports . .... .... .
6.9
Machinery ..
3.8
ShipbuUding . „
0.9
Mining :: : : :
0.7
Miscellaneous ....
'15.6
< M. W. Worthing, Distribution of Steel Products to Major Consuming Industrie', United States Steel
Corporation, October 30, 1939. Computations made by apportioning Individual hot-rolled product
totals on the basis of Iron Age distribution reports and by allocating jobber snipments to ultimate con-
sumers.
» "Miscellaneous" for the period 1926-31 includes "Furniture and Furnishings."
la connection with the above classification interesting observations may be
made. First, the purchasers of steel are principally companies engaged in the
production of producers' and consumers' durable goods. An exception is the
container industry which manufactures tin cans, an article classified as a perishable
good since it is generally used but once and discarded. Second, in recent years
there has been a marked increase in the percentage of steel purchased by con-
sumers' durable goods industries, such as the automotive and household appliance
industries, and a decrease in the percentage of steel purchased by producers,
durable goods industries, such as the railroad industry. In this connection "Mis-
cellaneous," which has shown such rapid growth", includes many industries pro-
ducing consumers' durable goods such as refrigerators, air conditioning units,
stoves, etc. Third, "Exports" in some years account for an appreciable amount
of total steel sold. Since the economics of export trade involves conditions not
present in the domestic market, the subject of prices and pricing methods in the
steel export trade have not been included in this study.
Most industries purchasing steel are characterized by large companies; in the
automotive, container, agricultural implements, household durable goods and
shipbuilding industries, a relatively few large companies comprise a substantial
percentage of the total production of their respective industries.' In purchasing
their steel requirements these large companies usually come into the market with
orders of considerable magnitude. The demand for steel therefore consists, to a
great degree, in large-sized orders placed by relatively few companies.*
• Big Business: Its Growth and Its Place, Chart 3, p. 42 (Twentieth Century Fund.) Exhibit No. 8S8
submitted to the T. N. E. C, July 11, 1939, (based on Census of Manufactures).
« Sales statistics of the subsidiaries of United States Steel Corporation show that In 1937, 941 customers bad
billings over $100,000 each and accounted for 73% of gross sales; In 1938, 663 customers had billings over $100,000
each and accounted for 68% of gross sales.
13896 CONCENTRATION OF ECONOMIC POWER
GBOGBAPHIC DISTRIBUTION
Orders for steel arise for the most part in concentrated geographical areas. The
bulk of tonnage business originates in a belf extending east of the Mississippi,
and north of the Ohio rivers, tapering ofif toward Philadelphia and New York;
but important markets exist outside this zone, particularly for products required
by the oil and canning industries. Although major markets for particular steel
products vary both as to location and degree of importance, the principal centers
of the composite demand for steel in their general order of precedence are: *
1. Detroit 6. Youngstown 11. Cincinnati
2. Chicago-Gary 7. Milwaukee 12. Houston
3. Pittsburgh 8. San Francisco 13. Buffalo
4. Cleveland 9. Newark 14. St. Louis
5. Los Angeles 10. New York 15. Toledo
Characteristics of Demand
The demand for steel is subject to tremendous cyclical fluctuations. This is due
primarily to the great cyclical fluctuations in the demand for producers' and
consumers' durable goods in the manufacture of which steel is consumed.
DEBIVteD nature OF DEMAND
The demand for new durable goods is highly sensitive to changes in the demand
for services which the durable goods perform. This may be demonstrated by a
simple theoretical illustration. A railroad needs five hundred cars filled to ca-
pacity to carry 10,000,000 passengers a year. Each year fifty cars normally wear
out and are replaced. More people decide to travel by railroad and passenger
traffic increases 10 percent, so that 11,000,000 passengers a year must be accom-
modated. T lis requires fifty more cars which must be acquired .immediately to
meet the inc? 3ased demand for passenger service. Therefore, in the year that this
increase occurs the railroad has to buy one hundred cars instead of the fifty
usually purchased for the normal replacement program. Thus a 10 percent in-
crease in the demand for passenger service results in a 100 percent increase in the
demand for railroad passenger cars. This is sometimes called by economists the
"acceleration principle." It works in reverse too. If passenger traffic decreased
10 percent there would not be any demand at all for new passenger railroad cars;
5ince only four hundred and fifty cars would be required to carry the 9,000,000
passengers left, no additional cars would be needed to replace the fifty worn out.
[n other words, a 10 percent decrease in demand for passenger service would cause
a 100 percent decrease in the demand for new durable goods to perform such
service.
DURABILITY AND DEMAND
The longer the life of durable goods the more sensitive is the demand for the
new durable goods to changes in the demand for services. For example, in the
simple theoi-etical illustration given above the average life of the railroad car was
presumed to be ten years. Fifty cars normally had to be replaced annually.
However, if the average life had been five years, 100 cars per annum would have
to be replaced. In that event a 10 percent increase in the demand for passenger
service would have resulted in only a 50 percent increase in the demand for new
railroad cars, and a 10 percent decrease in the demand would have resulted in a
50 percent decrease in the demand for new equipment. On the other hand, if the
average life of a car had been twenty years, only twenty-five cars would have to
be replaced annually. Therefore a 10 percent increase in the demand for service
would have caused 200 percent increase in the demand for new railroad cars.
In the event of a 10 percent decrease in the d'^mand for service, the replacement
demand for new equipment would not only disappear entirely, but twenty-five
additional cars theoretically would be removed from service and be available to
meet the normal replacement demand in the following year.
Thus, while the demand for new durable goods is highly sensitive to change in
the demand for services which the durable goods perform, the degree of such
sensitivit.v and the magnitude of the resultant fluctuation in demand depends on
> Based on estimates made in 1937 for the subsidiaries of the United States Steel Corporation of a "normal"
Industry-wide market for the following products: heavy rails: heavy structural shapes; plates, sheared and
oaiversal: fabricated otructural work; merchant bars, including reinforced concrete bars and Ught structural
Ihares: black sheets; galvanised sheets; hot rolled strip; rods, wire and wire products; tin mil' products;
pipe and tubing.
CONCENTRATION OF ECONOMIC POWER 13897
the life span of the durable goods; fluctuations in the d. mand for new du able
goods will be progressively greater as durability increases. In actual practice
many qualifications to this principle exist,* nevertheless it is fundamental in the
demand for durable goods.
POSTPONABILITY OF PURCHASE OF DURABLE GOODS
The purchase of durable goods usually can be easily postponed, and is postponed
when income is scant or prospects for the profitable use of additional durable goods
are discouraging. As a result of postponability of purchase, producers' durable
goods industries feel an immediate effect on demand resulting from the con-
traction of producers' income as expenditures for capital goods are deferred and
the income of the purchaser is directed primarily to meeting necessary out-of-
pocket expenses. In addition, even though the immediate business outlook is
favorable, expenditures for capital equipment may be postponed if the long term
business outlook is unfavorable; the business man must anticipate a reasonable
return over the life of the investment before tying up his capital in durable equip-
ment. After a prolonged depression, with purchases of durable goods almost
completely eliminated, increased profits and returning confidence as to the future
may stimulate a great upward surge in the demand for replacements previously
postponed and also for new equipment for expansion.
In like manner, consumers' durable goods industries feel the impact of declining
consumer income, as funds available are used to buy the necessities of life and
existing consumers' durable goods, such as automobiles, are made to last longer
than anticipated, or are discarded without replacement under stringent con-
ditions. Increased consumer income, actual and anticipated, will create a strong
revival in demand for consumers' durable goods as replacements are made and
new equipment purchased.
As previously indicated, the "acceleration principle" becomes more potent as
durability of a product increases. As a result the magnitude of expansion and
contraction in demand for products of the durable goods industries will be greater
than for non-durable goods industries. These fluctuations of demand for ne\'/
durable goods will be further magnified by the postponability of purchase of these
goods; a producer wUl buy coal, oil or electrical energy long after he has decided
he must postpone purchase of capital equipment, and a consumer must buy food,
clothing and other necessities even though he cannot afford a new car or a
refrigerator.
TOTAL DEMAND FOR STEEL IS INELASTIC
The magnitude of these cyclical fluctuations in demand cannot be materially
affected by adjustments in the price of steel because the total demand for steel
is itfelastic. This is due, first, to the derived nature of the demand for steel, and,
second, to the limited number of substitutes for basic steel products, and con-
versely the limited number of products for which steel may be substituted.
As previously indicated, the demand for steel is derived from the demand for
the services which products made of steel perform. If a change in the price of
steel is to influence the demand for the finished product in which the steel is used,
two conditions must exist: the cost of steel must represent a substantial per-
centage of the selling price of the finished article, and the demand for the finished
article itself must be such that it responds to changes in its price. This is not
generally the case; steel as a raw material usually represents a small percentage of
the total cost of the finished iproduct, and the major industries purchasing steel
have a rather inelastic demand for their products.'
The automotive industry, which during recent years has been the largest single
customer of the steel industry, is a typical example of the derived nature of the
demand for steel and the resultant inelasticity of such demand. The cost of steel
in a low-priced automobile retailing between $700.00 and $800.00 is about $85.00,
or roughly 10 percent of the retail price. Roos and von Szeliski in a recent study
' (a) The actual age distribution of the stock of durable goods in use might change in dlflerent years
(b) The effect of obsolescence is to increase replacement rates and therefore limit the magnitude of fluctua-
tions, if it is a constant factor from year to year. If an erratic factor, it would increase the fluctuations if It
occurred in normal or above normal years, or if it occurred in sub-normal years it would limit the fluctuation.
(c) For producers' durable goods, obsolescence can be brought about by shifts in demand, the development
of new products, the introduction of new techniques of production, discovery of new resources or new meth-
ods of using resources, migration of industry from one afea to another, and similar changes. In the field of
consumers' durable goods, style changes and shifts in consumers' demands are among the causes which may
result in shortening the otherwise useful life of durable goods.
"> The approximate proportion of steel cost in price of finished product for various items is as follows:
mile of railroad, 36.7.%; apartment building, 10%; automobile, 10%; can of food, 8%; frame house, 6.2%;
electric refrigerator, 3.4%; dairy barn, 3.2%; mile of reinforced highway, 0.7%.
13898 CONCENTRATION OF ECONOMIC POWER
contained in "The Dynamics of Automobile Demand" » estimated 1.5 to be a
representative average of elasticity of demand for new automobiles; i. e., for
every 1 percent decrease in the price, the automobiles sold would increase 1.5
percent. Since steel costs represent 10 percent Qf retail price, a 5 percent decrease
in steel prices would permit a 0.5 percent reduction in the price of automobiles,
and according to such elasticity of demand. would increase automobile sales to the
extent of 0.75 percent. The resultant increase in the demand for steel by the
automobile industry would be neghgible.
EFFECT OF THE SUBSTITUTION FACTOB
Substitution of steel for other materials, or a reverse substitution, is not an
important factor in the cyclical fluctuations in the demand for steel. If, through
lower prices, steel could invade a major market served by other products, or if
high relative steel prices meant invasion of major steel markets by substitute
products, there would be imparted to the total demand for steel a degree of
elasticity not now present. Steel possesses more physical strength per dollar
of investment than any other existing product; wood and concrete have a restricted
field in which they may be substituted for heavy steel. Glass, plastics, rubber,
alxmiinum and certain alloys may serve as substitutes in specialized fields; but
even in these cases price may be only one of many competitive factors involved.
Therefore, price reduction would result in very little additional steel being sold
as substitutes for other products, and a price advance, unless abnormal, probably
would not result in additional competition from substitute products.
POTENTIAL ELASTICITY OF DEMAND FROM A PABTICULAR PBODUCEB
Although the over-all demand for steel is inelastic and the total quantity bought
would not be substantially different if the price within reasonable limits were
lower or higher, the demand for steel from a particular producer possesses great
potential elasticity. This readiness of a buyer to sh\ft from one producer to
another because of a lower price is due to the informed character of the buying
of steel. Technical knowledge of the product to be purchased is available through
laboratories of individual purchasers, trade associations and independent research
agencies; exactly the same steel may, for the most part, be obtained from any one
of a number of producers. Furthermore, the large size of individual purchases
makes it worth-while for buyers to seek the lowest possible price. This propensity
to shop is enhanced by knowledge of latest price quotations, by familiarity with
psychological and other factors resulting in a "buyers" or a "sellers" market for
aU or particular prodiicts, and by a general understanding of approximate costs
of steel production ; indeed, a few purchasers » of steel operate completely inte-
grated steel works to supply a portion of their requirements, and others i" have
semi-integrated and non-integrated capacity.
Thus, potentially, the demand for steel from an individual producer is'elastic
and buyers are often in a position to exert bargaining pressure to obtain the lowest
possible prices, especially when the steel industry is not operating near capacity.
The Supply of Steel
geographic concentration
The most economical source of steel is that location at which the raw materials
can be assembled, the steel produced and delivery to the market effected at the
lowest possible total cost. In determining plant location '^ assembly costs are
most important; more than four tons of raw materials must be assembled for every
ton of steel produced, i^though production costs are subject to variations due
primarily to geographical wage rate differentials, these variations are supple-
mentary to and, in a measure, compensatory for otherwise uneconomical assembly
or delivery costs.
The approximate amounts of principal raw materials required per ton of pig
iron are: 4075 pounds of iron ore (assuming ore of a reasonably high metallic
content), 2700 pounds of coking coal and 900 pounds of limestone. Another 1500
pounds of coal may be consumed for power and heating before a ton of finished
' Publication of the General Motors Corporation based upon papers presented at a Joint meeting of the
American Statistical Association and the Econometric Society in Detroit, Michigan, on December 27, 1938.
' Ford Motor Company and International Harvester Company.
'• American Car and Foundry Co.; American Locomotive Co.; Atchison, Topeka and Santa Fe Railroad
Co.; Continental Can Co.; Simonds Saw and Steel Co.; Tlmken Roller Bearing Co., Inc.
" Availability of a large water supply Is important In steel mill location.
CONCENTRATION OF ECONOMIC POWER 13899
aieel product has left the mills. The greater proportion of the raw materials is
used in the blast furnace, but integrated steel works have developed from blast
furnace plants because (a) as steel approaches the finished stage the cost of ship-
ment becomes a smaller percentage of the cost of the product to the buyer; (b)
integration assures more constant and reasonably full utilization of blast furnaces
and open hearths; (c) economies of converting molten iron into steel and other
heat conservation factors are important in the economical production of steel'.
Limitations imposed by the necessity for the most favorable combination of
assembly, production and delivery costs have confined steel production to a few
geographical areas.
The most favorable combination of the three variables is probably to be found
at Lake Erie and Lake Michigan ports and in the Pittsburgh district (including
.the Mahoning and Ohio Valleys) . These locations '^ were primarily determined
by the assembly costs of Lake Superior ores which are the backbone of the steel
industry in the United States and supply about 82 percent of the ore consumed in
the country, and of the finest metallurgical coking coals which are found in Western
Pennsylvania, West Virginia and Kentucky. The assembly cost. of lirnestone,
which is well distributed and the least important of the major raw materials, is
usually an incidental factor.
Comparative assembly costs at principal production centers in this area have
been estimated as follows:
Estimated Assembly Costs in the Production of Pig Iron, Summer of 19S7 '
[In dollars per gross ton of pig iron]
Iron Ore
Coal
Flux
Total
Annual Blast-Furnace
Capacity
Producing Center
Thousands
of Gross
Tons
Percentage
.ofU.8
Total
Weirton-Steubenville _ .
$5,508
5.804
3.497
3.497
3.497
6.193
3.487
$0,468
0.284
2.714
2.909
3.249
1.979
3.867
$0,337
0.337
0.241
0.241
0.086
0.170
0.241
$6,313
6.425
6.452
6.647
6.832
7.342
7.595
2,093
11, 521
2,685
3,267
1,423
6,592
10,266
Pittsburgh _._ __..
Cleveland
Buffalo - ._..
Detroit ._
Youngstown
23.0
6.4
6.5
2.8
13.2
20.5
Chicago
Total
1 Worthing, Marion, "Coiaparative Assembly Costs in the Manufacture of Pie Iron", Pittsburgh Btisinest
Review, v. VIII., No. 1, January 31, 1938, Pp. 21-25, Table 1.
Assembly costs at these locations vary; the importance of each component of
the costs is emphasized by the difference of $1,17 in favor of Pittsburgh over Chicago
due entirely to Pittsburgh's fortunate position in the center of the finest metallur-
gical coking coal fields in the country.
Although primarily based upon assembly costs, the growth of these great steel
production centers to their present size would not have been possible if outlets
fop at least a considerable part of their products did not exist fairly close at hand;
aU the production centers, coincide with, or are adjacent to, major centers of steel
demand. However, the location of these production centers depends only in part
on relative assembly costs and the magnitude of local demand for a particular
product; it depends, among other thngs, on the conformation of the market for
each product and for the group of products that may be economically produced
together."
For example, hot rolled sheets, cold rolled sheets and tin plate, which are
produced at the Gary sheet and tin mills of a subsidiary of United States Steel
Corporation with a'll the attendant economies of large scale production, are products
of virtually the same integrated process. Major outlets for hot rolled sheets are
Chicago, Detroit, and Indiana with important sources of demand in Iowa, Minne-
sota, and Ohio; Detroit is the principal market for cold rolled sheets, and Chicago
" Although Pittsburgh historically was established as a steel producing center before Lake Superior ores
and coking coal came into general use, its growth and the maintenance of its dominant position has been
based on its economical accessibility to these resources.
'3 Committee on Iron and Steel Price Research, National Bureau of Economic Research Conference on
Price Research, Proposa.'i /or Research on Prices and Pricing Policies in the Iron and Steel Industry (1939).
124491 — 41— pt. 26 21
13900 CONCENTRATION OP ECONOMIC POWER
is an important market for tin palate. A similar situation exists at the Irvin
Works of this same subsidiary in the Pittsburgh district which rolls the same three
products. Ohio and adjacent West Virginia counties, Pittsburgh and Philadelphia
are maior markets for its hot rolled sheets. Cold rolled sheets are principally
shipped to Cleveland, other Ohio centers and Philadelphia. The Irvin Works
may also supplement Gary in the Detroit market with hot and cold rolled sheets
in periods of peak demand, while Metropolitan New York is the major market
for its large ouput of tin plate.
This market structure of groups of products that may economically be produced
together accounts in part for production patterns with such apparent inconsisten-
cies as limited capacity at Detroit and excess capacity, as compared to local de-
mand, at Pittsburgh. The effects of historical development and the immobility
of steel making equipment will be discussed later.
Birmingham, Alabama, and vicinity is another location with a favorable
combination of assembly and production costs. Assembly costs at Birmingham
are undoubtedly the lowest in the country — iron ore, coal and flux being in close
proximity. In this case low assembly costs compensate in part for the com-
paratively poor quality of the raw materials; iron content of the ores is low and
phosphorous content high, making conditioning and sintering desirable; the coal
requires washing before coking. With wage rates lower than other districts,
production costs are also economical, although basic wage rates have been rising
in the South. These advantages of assembly and production costs are offset by re-
moteness from major markets; a substantial part of the tin plate produced at the
large plant recently er^ted by Tennessee Coal, Iron & Railroad Company,
another subsidiary of United States Steel Corporation, at Fairfield, Alabama, is
shipped to the West Coast and Hawaii.
Sparrows Point, Maryland, is strategigally located. Based on the use of
high grade imported ores, iron ore costs have been estimated " to be less at
Sparrows Point than at Lake Erie and Pittsburgh area plants, which advantage
is offset, in part at least, by higher assembly costs for coal and limestone. Its
accessibility to the large markets of the eastern seaboard, and its ability to com-
pete on the West Coast via all-water transportation due to tidewater facilities,
make economical distribution costs a major factor in the favorable location of
Sparrows Point.
Combined assembly, production and delivery costs make possibldi integrated
steel production on a commercial basis in only one other geographical area at the
present time; '^ Colorado and Utah both possess iron ore, fair coking coal and
limestone in sufficient quantities and within reasonable assembly distance of
each other. Due to prohibitive distribution costs, however, this district must
depend, in the main, on local demand for special products. At Pueblo, Colorado,
the Colorado Fuel and Iron Corporation, cognizant of this situation, produces
principally rails and track accessories for Western roads, and wire products for
farm and ranch consumption. At Ironton, Utah, the Columbia Steel Company,
a subsidia>y of United States Steel Corporation, operates a blast furnace whose
pig iron output is taken in part by its West Coast steel mills near Los Angeles
and San Francisco and in part by local buyers. California steel mills also use a
considerable amount of scrap obtained locally.
Although it is an important steel consuming area, the West Coast cannot sup-
port more than limited steel making capacity due to high assembly costs, par-
ticularly in the face of competition from Birmingham and Sparrows Point, both
of which can serve this area on a more economical basis.
The principal steel producing centers of the nation, therefore, are confined to
particular geographical areas where the raw materials for steel making can be
economically assembled. Differences in the development and activity of these
producing areas have been determined to a considerable extent by the relative
costs of transporting steel to consuming areas. Many small non-integrated
mills, howeyer, are located outside the major producing areas where they may
use local scrap, merchant pig iron or semi-finished steel to produce steel for con-
sumption in the local area or may specialize in particular products to distribute in
more widespread markets'.
t" Maryland State Planning Commission, The Iron and Steel Industry— Float Furnaces, Steel Works and
Rolling Mills, November 1938, p. 14.
'» With the exception of certain areas with small local ore deposits, capable of supporting limited operations,
i. e., ore deposits of New Jersey, Eastern Pennsylvania, and the Adirondacks, economically accessible to
Pennsylvania coal fields.
CONCENTRATION OF ECONOMIC POWER 13901
TECHNOLOGICAL ASPECTS
Steel making equipment installed at the producing centers is both costly and
immobile; the economies of size inherent in steel manufacture have been important
factors in determining the design of modern mills. The result is that the small
plants of fifty years ago have been succeeded by complex and gigantic operating
units.
Twenty years ago the coke used m blast furnaces was principally made in banks
of simple beehive ovens, usually located at the mine. Today, it is made at or
near the steel plant in long batteries of by-product coke ovens with alternating
coking and heating chambers topped by coal larries, off-takes and collecting mains.
In close proximitj' stand the tall cooling towers and scrubbers, the ammonia house
and benzol plant used to obtain dumerous by-products from the tars and gases
emanating from the coking ovens, which are today recovered and put to use.
In 1880 the capacity of the most efficient blast furnace, a comparatively simple
unit, was one hundred tons per day; at present the newest and most eflicient
furnaces are rated at 1100 to 1200 tons per day. This increased output has been
accomplished not only by increase in size and better blast furnace practice, but
by mechanical improvements and the development of auxiliary equipment. A
blast furnace plant today is enormous and complicated. The furnace is a tall
circular structure 90 to 100 feet high, built of firebrick and reinforced externally
by a close-fitting steel shell. It is provided with apparatus for hoisting iron ore,
coke and limestone to the top where they are charged into the furnace. Large
pipes carry the gas generated in the furnace to the stoves where it is used for
heating purposes. Beside each furnace stand four cylindrical stoves nearly
as high as the furnace itself. These stoves heat air to high temperatures before
it is blown into the furnace at the rate of five tons of air for every ton of iron
produced. The impurities in the raw material are either burned out or accumu-
lated in the slag which gathers on top of the molten metal. This slag is removed
through the higher of two tapping holes. Through another tapping hole the
molten iron is drawn at periodic intervals either into ladles to be carried to huge
containers known as mixers subsequently to be taken to the open hearth and
Bessemer converters, or into runners leading to the pig iron casting beds. A
boiler house, power plant, pumping station, tflrbo-blower, stockyard, ore bridge,
car dumper and raw material bins, all constitute important parts of blast f'urnace
equipment. .
The steel making equipment is equally complex and has increased in size as it
has become more efficient. In 1899 the average open hearth furnace had a
capacity of 22 tons per heat; in 1938 the average furnace capacity was 95 tons
per heat and the largest 400 tons per heat. Even more spectacular has been the
radical improvement in design and the increase in size of continuous rolling mills
for flat rolled products in recent years. This acceleration of growth has been so
dramatic that in 1936 a continuous rolling mill with a capacity of as much as
600,000 tons of finished flat rolled steel per year was unprecedented; yet in March
1938 a continuous strip mill was opened with an annual capacity of approximately
one million tons."
Equipment used in each stage of modern steel making is usually so combined
as to perform a series of vertically integrated operations; conservation of heat
and power requires continuous processes. Assurance of adequate sources of raw
material and the elimination of purchasing expenses at each stage of operations
are important factors in promoting further integration.
Vertical integration is a dual development in the industry. Non-integrated
and semi-integrated producers desiring independence from producers of semi-
finished steel and the owners of raw material reserves, and influenced by the
possibility of additional savings, integrate toward the sources of their raw materials.
Partly as a result of such movement and partly due to the dechne in demand for
steel used in producers' durable goods industries, producers of semi-finished and
heavy steel have obtained outlets for their productive capacity by integration
towards more highly finished products.
CAPITAL INVESTMENT REQUIREMENTS
This combination of huge uriits vertixially integrated requires large capital
investment. A modern blast furnace of about 1,000 tons capacity with the
auxihary equipment above m utioned costs four to five million dollars. The
i« Republic Steel Company continuous strip mill, Cuyahoga Valley, Ohio.
139D2 CONCENTRATION OF ECONOMIC POWER
average investment required for a modern steel works of efficient size is approxi-
mately $100,000,000. Such a mill would be capable of producing about 1,000,000
tons of ingots per annum and would have diversified finishing equipment of suffi-
cient capacity to con vert about half the output into billets and other semi-finished
steel and the other half into sheets and strip. Such an investment would not
include operations prior to the assembly of raw materials at the plant site, i. e.,
the plant would be integrated only from coke plant to continuous rolling mills.
Operating units may be and sometimes are much larger; a single continuous hot
and cold rolling finishing plant alone may require an investment of $60,000,000.
Such large and complex equipment cannot be moved in response to geographical
shifts in demand, and only extraordinarily great differential advantages of a new
location j ustify scrapping existing facilities embodying large unamortized invest-
ment and long remaining service life. New areas of demand usually develop only
for particular products or groups of products and it may be more economical for
the established producer to install sufficient capacity at the existing location to
compete in the new markets than to build integrated steel works at the source of
the new demand. This decision may depend first, on the combination of products
that can be economically produced together, and second, on whether the steel
demanded can be produced by integrating new facilities with unused capacity at
the existing location. Modernization and expansion at the established location
may be rational; and> the development of an individual company at a particular
location may thus be perpetuated.
FACTORS IN EXPENDITURES FOR NEW PLANTS AND EQUIPMENT
The number of producers of any particular steel product bears a rather direct
relationship to the minimum investment required to become such a producer.
It is pertinent to inquire first, the source of the funds for such capital expenditures
and second the inducements necessary for the investment of these funds.
Source of Funds. — Funds for investment in npw plants and equipment may be
obtained from any one or a combination of the following sources: (1) Outside
capital; both existing companies and promoters of new companies may borrbw
through the medium of notes and bonds or sell stock to obtain funds from this
source. (2) Accumulated earnings; the availability of this source of funds over
the years enabled existing companies to promote and keep pace with the upward
trend in- national steel consumption, and in addition helped small non-integrated
and semi-integrated steel companies grow into large integrated units. (3) De-
preciation and other reserves; this has been the primary source of funds for replace-
ment and modernization programs.
Incentives for Investment — Profit Motive. — The normal incentive for investment
is prospective profits. This may cause the expansion of existing companies; the
^development of non-integrated and semi-integrated companies into integrated
companies being a case in point; or it may induce new companies to enter the
field usually as non-integrated or semi-integrated specialists. The formation of
a new integrated steel company, except by merger, would not be likely today
since: (1) A large capital investment is necessary. (2) The technological and
organizational difficulties in forming such a company are great. (3) The difficulty
of obtaining an immediate market for the output of such a new company would
be tremendous; great losses in early years would therefore seem inevitable.
Incentives for Investment — Obsolescence. — Obsolescence has been an important
motive for capital expenditures by the steel industry in recent years. This has
been due to: (1) New production techniques; the introduction of continuous hot
strip mills and continuous cold reduction proce.sses has brought about a major
technological revolution in the industry. (2) The development of new products;
cold reduced sheets and cold reduced tin plate have practically displaced the hot
rolled products in major markets. In order to remain in markets demanding the
new and better products, companies have had to purchase new equipment and
construct new plants. (3) Shifts in demand; e. g., the marked increase in the
demand for sheets, strip, tin plate and other steel required by consumer goods
industries, and the decline until very recently in the demand for rails, plates and
structural shapes. This shift has caused expansion of existing companies both
to meet the new demand and to obtain outlets for otherwise unutilized ingot
capacities.
CONCENTRATION OF ECONOMIC POWER 13903
SIZE AND NUMBER OF PRODUCERS
The producers of the bulk of tonnage steel are large in size and relatively few
in number, which is a natural development in an industry requiring great capital
investment as the result of large scale equipment, vertical integration and, in
certain cases, horizontal integration. Principal producers (including subsidiaries)
and their respective percentages of total ingot capacity for the year 1938 are
indicated in the following table: '^
Percentage Distribution of Capacity Among Producers of Steel Ingots and Steel for
Castings— 1938
Percentage of
Total I Annual
Name of Corporation Capacity
United States Steel Corporation 35. 3-
Bethlehem Steel Corporation 13. 7
Republic Steel Corporation 8. 9
Jones & Laughlin Steel Corporation 5. 0
National Steel Corporation 4. 7
Youngstown Sheet & Tube Company 4. 3
Inland Steel Company 3. 9
American Rolhng Mill Company 3. 6
Wheeling Steel Corporation 2. 4
Other smaller companies .- 18. 2
Total 100. 0
1 This total does not include those companies that produce steel only for castings.
However, ingot capacities should not be the sole criteria of the size and number
of producers, especially in the consideration of markets for particular products,
since the number of companies and the percentage of the total that each has
capacity to produce varies with individual steel products. The number is deter-
mined by: (1) The minimum investment required in equipment to produce the
product, the prospective return thereon, and the relative simplicity of the opera-
tion. The investment formerly required for a steel mill and past profit margins
must be considered in a study of any particular company, since most of the
present producers entered the market under conditions different from those which
today would face a newcomer. (2) The technological history of the product and
the equipment used to produce it. (3) The nature of the demand for the product;
its diversity and geographical distribution. (4) The historical development of
the producers.
The percentage of the total represented by the capacity of any individual pro-
ducer is principally a reflection of: ''1) The historical development of that producer,
particularly with reference to product specialization; (2) The technological history
of the product and the equipment used to produce it; (3) The producer's location
with respect to demand.
Under the influence of these factors there are distinct variations in the character
and total number of producers of each steel product, and in the percentage of
total capacity possessed by each producer for each product, and in the geographical
distribution of their plants.
CHANNELS OF DISTRIBUTION
Approximately 80 percent of the steel produced by the steel industry is sold
directly to consuming industries through the sales organizations of the producing
companies so that in the majority of cases "sellers" and "producers" are inter-
changeable terms in the market for steel.
Jobbers and Warehouses. — The balance of the steel sold passes through the hands
of jobbers, warehouses and other distributors which are essential in the sale of
standardized products in small lots to widely scattered consumers, or where
geographical conditions such as exist on the West Coast make this form, of dis-
tribution particularly ecanomical.
" American Iron and Steel Institute, Iron and Steel Works Directory of the United States and Canada,
1938, pp. 401-402.
13904 CONCENTRATION OF ECONOMIC POWER
Although the jobber market is an important factor in the distribution of steel,
the "influence of this form of distribution on the pricing and marketing of the
majority of steel products is negligible. Jobber outlets are, however, important
elements in the marketing of galvanized sheets', concrete reinforcing bars, standard
pipe, tubes, and merchant wire products.
Importance of Outlets. — Maintenance of outlets for semi-finished and finished
steel is important for most members of the steel industry. The acquisitions of
such outlets, by integrated producers, which occur from time to time, involve a
change in the distribution pattern of the industry.
The supply side of the steel market from a long term viewpoint is marked by
these characteristics: (1) The areas of production are geographically concentrated
in a few districts because of location of raw materials and transportation costs.
(2) Large size equipment and vertical integration are typical of the industry;
some companies are also horizontally integrated, while a number of semi-integrated
or non-integrated companies are specialists in particular products. (3) Large
capital investment is necessary; however, for certain products the investment
necessary to become a producer is relatively much smaller than for others, and
this seems to be an important controlling factor in determining the number of
producers of a given product. (4) Generally speaking, producers are large in
size and few in number, although in particular cases major producers of specialty
products may be smaller non-integrated or semi-integrated units. (5) Investment
in new plants and equipment arises both in response to prospective profits and
as a result of obsolescence.
In contrast with many types of markets the steel market is one not easily entered
by producers, or withdrawn from, once enlry has been accomplished. The
large investment required, technological and organizational difficulties, and the
problem of obtaining an immediate market are obstacles to entry. The non-
recoverable costs that must be sunk in a steel company are not conducive to with-
drawal if there is an opportunity for any, return in excess of out-of-pocket expenses.
In much the same manner, che supply side of the steel market differs from other
markets in that productive capacity cannot be easily adjusted to meet changing
market conditions. Once capacity is installed, it is inelastic . and cannot be re-
moved except by scrapping, which ordinarily does not appear desirable due to
the large investment involved; nor can capacity be easily expanded except by
heavy capital expenditures requiring a considerable time interval.
Characteristics of Cost in the Industry
"overhead" or "fixed" costs
There are certain costs in the steel industry which are approximately the
same regardless of the amount of steel produced, i* These costs are sometimes
known as "overhead" or "fixed" costs. In the case of the United States Steel
Corporation and its subsidiaries such "fixed" costs are composed of the following
elements in the approximate percentages indicated:
" United Staters Steel Corporation & Subsidiaries Components of "Fixed" Costs,
under 1938 Conditions
Item
Approximate
Percentage
Interest
4.1S6
Pensions
k^
vio
Payroll . "::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::;:
84.10
Social Security Taxes
1.37
Goods and Services Purchased from Others
26.26
Total.- . . ...
100.00
" This presumes a company in operation. Complete shutdown naturally would decrease these costs
sharply.
CONCENTRATION OF ECONOMIC POWER 13905
Other steel producers may have dififerent percentages for the components of
their "fixed" costs depending on the degree of integration and their capital struc-
ture. However, regardless of their composition, such costs are relatively large
for major producers in the steel industry, and with low operating rates are a
substantial percentage of total costs."
"additional" costs
The costs over and above "fixed" costs represent the "additional" cost inci-
dental to the production of each additional ton of steel, assuming the steel mill
is already in operation. Recent studies of the experience of the United States
Steel Corporation and its subsidiaries over the past ten years indicate that the
addition to the total costs arising from the production of each additional ton '"
of steel is the same regardless of the operating rate at which the additional output
is obtained as long as the other factors affecting costs remain constant. This
phenomenon of constant "additional" costs covers an observable range of output
which extends from around twenty percent of capacity to slightly beyond ninety
percent of the physical limit of output. It is not certain that this relationship
would hold true as the physical limit of capacity is reached since at that point the
equipment may become overtaxed and for various reasons operate less efficiently
and at greater cost. Less efficient reserve units may also be placed in service to
meet the peak levels. In those circumstances the additional costs incidental to
the production of an additional unit of output would cease to be constant and would
probably rise sharply. The percentage composition of these costs in the case of
the United States Steel Corporation and its subsidiaries is indicated in the following
table.
United States Steel Corporation & Subsidiaries Components of "Additional"
Costs, under 1938 Conditions
Approzlmat
Percentage
Taxes (other than social security and Federa,! income) 2. 57
Depreciation and Depletion 4. 25
Payroll 52.22
Social Security ^axes 2. 08
Goods and Services Purchased from Others 38. 88
Totali. 100.00
For companies less integrated than the United States Steel Corporation and
its subsidiaries the percentage attributable to "Goods and Services Purchased
from Others" would increase and the percentage of other components decrease.
AVERAGE COSTS
Since the average cost of producing a given ton of steal is the sum of the "ad-
ditional" costs plus an amount equal to the "fixed" costs divided by the number
of units produced, this "average" cost must necessarily be higher than the
" See the following table:
United States Steel- Corporation & Suhsidiaries Percentage of "Fixed" to Total Costs at Various Hates
of Operation, under 19Sl Conditions
Operating Rate
Percentage of
"Fixed" to
Total Costs
Operating Rate
Percentage of
"Fixed" to
Total Costs
10
57.2
43.9
35.8
30.1
26.0
60
22.9
30
:::::::::::::::::::::::::;:..
80
18.5
50 ::::::::::: ::
100
" The tons mentioned in describing the cost pattern are weighted tons. This means that each ton of
Rolled and Finished steel product or of other tonnage product which is of a type whose average cost is less
than the average cost of all Rolled and Finished steel products, is made to count as less than a full ton
while tons of products of a class which is on the average more costly than the average cost of Rolled and
Finished .steel products, are made to count more than a full ton. In this way the number of tons of all
tonnage products shipped has been converted into equivalent tons of average-cost Rolled and Finished
steel products. The result is that total costs of various tonnages shipped are made comparable where they
would not be if unweighted tonnages had been used.
13906 CONCENTRATION OF ECONOMIC POWER
"additional" cost for nearly the whole range of operations almost to the limits
of capacity.^'
The components of the average cost of producing a ton of steel are, to a certain
degree, largely outside the control of steel producers; wage rates tend to be
inflexible and lag in adjustment, prices paid for goods and services are often
fixed by outside agencies as they are in the case of railroad rates, interest is
determined by factors in the money market, taxes are established by law, and
depreciation and depletion charges cannot long be disregarded.
The Dynamics of the Market for Steel
COSTS AND demand
The inelasticity of the total demand for steel and the aforementioned charac-
teristics of cost in the steel industry place definite limitations on the financial
ability of the industry to increase production by decreasing prices. Assuming
that each 1 percent decrease jn price would increase consumption of steel 1
percent, a 10 percent decrease in the average level of steel prices prevailing during
1938 even though offset by a 10 percent increase in the quantity of steel sold,
would have increased the deficit ^^ of the United States Steel Corporation from
$8,758,572.P0 to $52,058,672.00. This estimate is most conservative, since there
is every indication that the elasticity of the demand for steel is not as great
as assumed above.
Despite this overall price-volume-cost relationship in the industry, the potential
elasticity of demand for the product of an individual steel company and the
internal problems arising within individual companies from this characteristic
cost pattern further affect the market for steel.
Except in periods of high operations, and more particularly in times of slack
demand, there is a tendency to cut prices below average costs so long as the
price for the additional unit sold is above the "additional" cost necessary to
produfS- tich additional ton of steel. The large size-of individual orders and the
potential shiftability of buyers of steel in response to price considerations accen-
tuate such a tendency, particularly wlren, due to the inelastic nature of the total
denrtand for steel, the problem fOr the individual producer is to obtain a share
of the going business. Thus it is that in periods of restricted demand, knowing
that anything above his "additional" costs contributes something toward "over-
head" or "fixed" costs which must be met in any event, the producer will cut
prices below his average costs if he feels he can obtain additional business for his
mills thereby. This inherent tendency to cut prices, however, is offset to Some
extent by the knowledge that competitors will meet price concessions as soon as
they become known.
PSYCHOLOGICAL FACTORS
Buyers and sellers of steel react differently at various stages of the lousiness
cycle; this is natural in an industry marked by large cyclical fluctuations in the
demand for its products. In depression the tendency toward price cutting grows
as buyers bargain more sharply and sellers scramble for what business there is
in an effort to reduce deficits mounting under the burden of "overhead" or "fixed"
costs. In better times buyers are less averse to paying higher prices, and sellers
no longer under the goad of operating losses are reluctant to make price conces-
sions. Therefore, in part at least, cyclical fluctuations in steel prices are attrib-
utable to changes in the psychology of buyers and sellers.
characteristic patterns of ACTION BY SELLERS IN THE MARKET FOR STEEL
The factors mentioned above have resulted in phenomena that reappear each
time the steel industry passes through a full cycle in demand.^' In'a jising cycle
" In the case of United States 'Steel Corporation and its subsidiaries, the average cost of all operations
per ton of steel shipped, under 1938 conditions, would be $55.73, plus an amount equal to $182,100,000 Cthe
total "fixed" costs) divided by the number of tons produced.
» Deficit after deduction of bond interest, but before Federal income and profit taxes and exclusive of
non-operatinR income and expense.
" The pattern outlined has perhaps been oversimplified since (1) all products do not pass through each
phase of the cycle simultaneously, making the pattern more confused than it appears in this outline; (2) the
existence of jobbers and distributors complicates the situation with respect to certain products; (3) in addi-
tion, the homan factor is unpredictable, making it difficult for businessmen always to rationalize their
actions as they participate in a highly competitive market.
CONCENTRATION OF ECONOMIC POWER 13907
as demand increases, average costs in the industry decrease as additional unitfi
are produced, but these decreases are usually soon offset by higher raw material
prices, and increased labor and other costs. In addition, as already indicated,
the psychology of the buyers^ and sellers changes and the industry may feel that
the time is propitious for an increase in prices, not only to cover increased costs,
but also to compensate for past losses and to accumulate resources for possible
future periods of depression. Quite naturally, however, producers of steel do
not care to take the risk Oi losiiig their share of business by an increase in prices
which may not be followed by their competitors.^^ The natural result is that
the industry is inclined to wait for some large producer to. announce higher
prices. This natural phenomenon in the rising cycle is sometimes called "price
leadership". So long as the term is used to describe a natural phenomenon re-
sulting from factors inherent in the industry and involving no collusion or other
violation of the anti-trust laws, there is little objection to the term.
In the falling cycle, average costs increase as demand and production decrease,
accentuated in part by the continuance of high wages which have a tendency to
become inflexible, or in any event to lag in their adjustment to the lower level of
production. In the early stages of the decline in demand, the industry, aware of
the inflexibility of the total demand for steel and faced by rising average costs per
unit of output, naturally is averse to cutting prices when the prices they are
getting on the going business barely cover their costs. From past experience
the industry is aware that any weakening of prices leads buyers to hold off pur-
chasing in the expectation that prices will go still lower. Then too, the steel
producer may be optimistic about an improvement in general business conditions
in the near future. However, sporadic price cutting soon breaks out spurred by
the individual producer's hope of obtaining an additional share of the going
business. Concessions soon become general knowledge in the trade; and while,
for a period, some producers may not care to compete on the basis of these con-
cessions, eventually all producers must meet competition at the going prices.
The Basing Point Method of Quoting Delivered Prices
The basing point method of quoting delivered prices in the steel industry has
developed over a long period of years in response to the fundamental economic
factors of that industry. Two authorities on the economics of the steel industry
succintly point to the basic fallacy in the reasoning of most critics of this pricing
method when they state that "Intelligent appreciation of the pricing problem in
the steel industry has suffered from a failure of most commentators to distinguish
between the basing point system as a medium or mere mechanism for the trans-
lation of policy into action and the economic roots of that primary policy itself."^'
ECONOMIC ROOTS OF THE BASING POINT METHOD
In quoting prices manufacturers of steel must take certain basic factors into
consideration: (1) The cost of transportation from steel mill to destination may
be substantial in relation to the value of steel shipped. Consumers of steel are
interested in the cost of steel at the place where they use it. Therefore, most
consumers want to know the lowest delivered price at which they may purchase
the steel they require.' (2) Consumers of steel are located in different parts of
the country and although more steel may be sold in some sections than in others,
even major markets for the same steel product may be geographically widespread.
(3) Producers of steel must locate their plants at points where raw materials
may be economically assembled. This confines major steel producing centers to
a few geographical areas. Modern steel making equipment is large and complex;
it requires great capital investment and is extremely immobile once installed.
(4) To insure economical and reasonably stable operations, steel producers must
sell large quantities of steel and since consumers of the group of steel products
that may economically be produced together may be located in different areas, the
producer must be able to quote prices at diversified locations. The extent
'* Does not apply where all capacity of a pwrticular product is booked substantially ahead.
» de Chazeau and Stratton, Economics of the Iron and Steel Industry, by Daugherty, de Chazeau and
Stratton, p. 578 (McGraw-Hill Book Company, 1937).
13908 CONCENTRATION OP ECONOMIC POWER
to which he may economically serve diflFerent consuming areas will be determined
by the most economical combination of assembly costs of raw materials, produc-
tion costs and the cost of delivering finished steel to important markets. (5)
Producers of steel have large "fixed" costs, which must be met regardless of the
number of tons produced so long as operations are continued. Although these
producers realize that the total quantity of steel consumed cannot be greatly
influenced by reductions in steel prices, they do know that the quotation of a
delivered price only slightly below other quoted delivered prices may influence
the placement of substantial orders with a particular producer. Since competition
for available business is keen, and particularly so when low rates of operation
make the "fixed" costs burdensome, a knowledge of the level at which competition
must be met in quoting prices at a definite location is valuable in preventing
completely disorganized markets that might prove disastrous to the industry.
The multiple basing point method of quoting delivered steel prices is a simple
pricing medium which has evolved over a long period of time to meet the peculiar
characteristics of the steel industry. It is an open price method of quoting
delivered prices at diversified locations. Such open prices are similar to list
prices which may be and are reduced to meet competition. As a pricing medium
it permits the consumer to bargain with a number of producers for both steel
and service at the lowest possible price and at the point where he needs it. It
serves producers by permitting them to compete in diversified markets to obtain
the volume and even flow of orders necessary to economical operations. In
essence, it provides an orderly medium by means of which consumers and pro-
ducers of steel may trade to their mutual benefit.
Relation of Competition to Profits, Capacity and Costs of Distribution
Price competition is necessary in any industry operating in a capitalistic
system. Is the steel industry competitive? Efforts at such determination too
easily lead into the realms of economic sophistry. Criticism and defense of
competition in the industry should not be based on abstract criteria which fail
to take into account the fundamental phenomena involved; it should be based on
tangible evidence.
Edward Chamberlin in his notable work, "The Theory of Monopolistic Com-
petition", demonstrates that evidence of imperfect functioning of competition
may be found in any one, or a combination of three, undesirable elements.^*
The first is excessive profits resulting from high monopoly prices. The second is
excessive productive capacity induced by high prices which encourage the entrance
of producers into the market, until the reduced volume of each lowers profits to
the minimum level, althoQgh the original high prices remain. The third is
excessive selling costs which contribute to higher prices if selling costs per unit are
greater than the decrease in production costs resulting from the increased volume
of production. Selling costs are simply one element of distribution costs, and
Mr. Chamberlin, although he does not do so, could apply his thesis to all distri-
bution costs with equal force. Assuming that excessive profits, excessive capacity
and/or excessive costs of distribution are criteria of the lack of competition, what
is the position of the steel industry with respect to these standards?
Profits in the steel industry are not excessive. From 1919-1928 inclusive,
the average return on investment was 5.1 percent; from 1929-1938 the average
rate of return was 2.4 percent."
A study based on a composite of financial statements of leading companies in
their respective industries illustrates the comparative earnings of other industries
and the steel industry for the period from 1929 to 1937 inclusive.
" Chamberlin, Edward, The Theory of Monopolistic Competition, Chapters V and VI, Harvard University
Press, 1938.
" Steel Facts, August 1939, No. 35. p. 3. Since the years, components and sources are different this flRure
naturally does not agree with that for "Iron and Steel" in the table which follows.
CONCENTRATION OF ECONOMIC POWER 13909
Ratio of Earnings to Net Assets — 1939-37 Inclusive (Earnings Before Interest in
Percent of Total Assets Less Current Liabilities) Steel Industry Compared With
Other Industries
7*!aTninn EdTTling
Industry — Continued. Ratio
Motion Pictures 5. 6
Building and Real Estate. _^ 5. 2
Telephone & Telegraph . 5. 0
Paper and Products 4.9
Oil Producing and Refining _. 4.8
Metals (Non- Ferrous)- 3. 8
Rubber & Automobile Tires. 3. 7
Railroads (Class I) 3. 6
Railroad Equipment i . _ 3. 1
Steel and Iron 2. 0
Textiles & Apparel 1.5
Coal 1.1
Industry: Ratio
Tobacco and Products 12. 3
■Vutomobiles and Trucks 11.7
Household Products 10. 6
Office Equipment 10.3
Automobile Accessories 10. 2
Chemicals and Fertilizers 10. 1
Leather and Shoes 9.3
Retail Trade 9.0
Electrical Elquipment &
Radio 7. 6
Food Products 7.6
Public Utilities 6.0
Machinery (Industrial & Ag-
ricultural).. 5.7
Sources: Standard Trade and Securities, Standard Statistics Company, Vol. 31 #20 Section 3 for 1927—
1935. Vol. 89 #15 Section 6 for 1936 and 1937.
On the basis of these figures the steel industry can hardly be accused of excessive
profits. Are these low profits caused by excessive capacity?
Capacity of the steel industry is not excessive. Unused or idle capacity should
not be confused with "excess" capacity. Past experience indicates that even in
periods of peak demand orders are not distributed among products in such a way
as to make possible fuU utilization of all facilities. In practice, therefore, opera-
tions probably would never be maintained at 100 percent of finished steel capacity
because of lack of coordination between demand and capacity for various products.
Production might, therefore, be expected to run five or ten "percent, or even more,
below capacity at the peak of the cycle.
In times of real emergency, or under the tremendous pressure of excessive de-
mands on the industry, it might be possible, by bringing into operation obsolete
facilities, lengthening the work week, eliminating holidays, and by other means,
to attain an operating rate in excess of 100 percent. This last happened in
May 1929.
True, the steel industry had a large amount of unused capacity during recent
depression years, but this is reasonable and to be expected in an industry with
capacities that are rigid and immobile and whose rate of operations is so controlled
by the tremendous cyclical flu&tuations in the demand for steel. If the industry
is to have facilities to supply the peak or near-peak demand, it must have idle
capacity during the periods of lower demand. An industry which, in the partial
recovery of 1937, produced steel ingots for three- successive months in an amount
roughly equivalent to the average monthly capacity for the industry in the high
production year of 192'9, cannot have "excessive" capacity if it is to take care of
the demands of a norndal recovery which would only have to be about 10 percent
greater than the peak months of 1937 to utilize the present full capacity of the
industry. The vital importance of existing capacity is emphasized by current
conditions which make it imperative for the steel industry to produce steel in
quantities never before equaled in its history. Quite conceivably, with any
capacity less than it presently possesses, the steel industry would become a bottle-
neck and prevent full normal recovery.
13910
CONCENTRATION OF ECONOMIC POWER
DISTRIBUTION COSTS
The steel industry does not have excessive distribution costs. In a study of
distribution costs of 312 manufacturers in 1931 ^^ "Iron and Steel and Their
. Products," a very broad classification, ranked among those having the lowest
distribution costs. The steel industry proper undoubtedly had even lower dis-
tribution costs than those companies included in the classification "Iron and
Steel and Their Products," if the records of the United States Steel Corporation
and its subsidiaries are in any way indicative of the average distribution costs for
the steel industry.
Selling Expense. — The major elements in the distribution cost study referred
to are "direct selling costs" and "advertising and promotion costs." These two
items combined represented 11 percent of net sales of those companies reported as
component manufacturers of "Iron and Steel and Their Products"; in 1931, the
same year used in the aforementioned study, direct seUing costs and advertising
and promotion costs were 3.1 percent of net sales for the United States Steel
Corporation and its subsidiaries.'^*
Freight Absorption. — An element more or less peculiar to the steel industry is
the amount paid by a steel producer for the transportation of steel from the steel
mill to the customer over and above the amount of the freight charge included in
his computation of the delivered price under the basijig point method of quoting
delivered prices. This results from competition in the steel industry, as a pro-
ducer in order to share in the business must meet the delivered price of a com-
petitor whose steel mill is nearer freight-wise to the customer. This is sometimes
called "freight absorption" by critics of the basing point practice.
A broad sampling 3" of shipments for the month of February 1939 by the
American Steel & Wire Company, Carnegie-Illinois Steel Corporation and
Tennessee Coal, Iron and Railroad Company, three subsidiaries of the United
States Steel Corporation, showed average "freight absorption" of $1.99 per ton
equivalent to 3.75 percent of the net sales return to the companies on these ship-
ments, and 3.6 percent of their delivered value to the customer.^' In view of
the fact that "freight absorption" plus selling expenses and advertising and pro-
motion costs for the steel industry are less than just the selling expenses and
'» See the following table:
Distribution Costs of 31$ Manufacturer', 19SI
[In Per Cent of Net Sales]
Product
Percent
Product
Percent
ConsuiULT Products:
Drugs and Toilet articles
48.8
sail
32.9
32.2
31.6
31.0
28.7
27.1
26.5
24.7
22.6
21.7
21.2
18.9
Consumer Products— Continued.
Agricultural Supplies
18 4
Paints & Varnishes
Tobacco Products
18 3
Furniture
Sporting Goods
18.2
Heating Equipment
Radio Equipment
Industrial Products:
16.5
Office Equipment & Supplies...
25.8
23.7
21.7
20.4
Chemicals and Allied Products..
Electrical Equipment- . —
Iron and Steel & Their Products.
Nonferrous Metals
Transportation Equipment
19.9
Household Appliances
19 7
Automotive
19.0
Clothing
Home Furnishings
18.5
15.5
An Analysis of the Distribution Costs ofSli Manufacturers, As.sociation of National Advertisers and
the National Association of Cost Accountants, New York. 1933, pp. 64, 106.
" Percentage of selling expenses and advertising and promotion costs to net sales for the United States
Steel Corporation for 1926 is 1.34%; for 1027, 1.65%; for 1928, 1.61%; for 1929, 1.53%; for 1930, 2.29%; for 1931,
3.07%,; for 1932, 4.32%; for 1933, 3.22%,; for 1934, 3.32%; for 1935, 2.79%; for 1936, 2.27%,: and for 1937, 1.98%i
"Temporary National Economic Committee, Form B. Distribution and Pricing of Selected Stee
Products for month of February 1939.
" "Adjusted" freight absorption, i. e., the above mentioned unadjusted freight absorption less basing
point price differentials, averaged $1.33 per ton, equivalent to 2.4% of the delivered value for the above
named subsidiary companies. Data based on Form B returns for the 55 steel companies reporting show
that "unadjusted" freight absorption for those companies averaged $1.77 per ton, or 3.2% of delivered
value, and "adjusted" freight absorption averaged $1.16 per ton, or 2.1% of delivered value. (See T. N.
E. C. Exhibit No. 1409, Charts C27, C28 and C31.)
CONCENTRATION OF ECONOMIC POWER 13911
advertising and promotion costs of nearly every other industry ,^2 it canftot be
charged that distribution costs in the steel industry are excessive.
Since excessive profits, capacity and distribution costs are not present in the
steel industry, it may reasonably be concluded that, although the economic
factors in the steel industry are such that it cannot survive for long under con-
ditions of cut-throat competition, it is sufficiently competitive to be free of the
alleged evils of lack of competition.
Conclusion
THE FUNCTION OF THE STEEL INDUSTRY IN THE NATIONAL ECONOMY
There remains one question of vital interest. Does the steel industry perform
its proper function in the national economy?
As a Source of Raw Material.— The steel industry primarily supplies a basic
raw material for the production of other goods and services. Properly to per-
form its function it must continuously provide material meeting the exacting
and changing demands of a great variety of industries each of which has diversified
requirements. The steel industry has consistently dbne so, as is clearly evidenced
by the industrial growth of the United States. The steel industry has developed
new products and improved the old ones, both on its own initiative and in close
cooperation with the steel consuming industries. In fact, if it were not for the
steel industry, many of the major improvements in products of other industries
would not have been possible. For example, the streamlined all-steel automobile
would have been impossible to construct fifteen years ago since it depends upon
the deep drawing quahties and strength of the modern cold rolled sheets. Due
primarily to the recently introduced cold reduced tin plate certain fruits and
vegetables are now available throughout the year as canned products. Beer
could not be sold in cans so readily if the steel industry had not developed a
special type of tin plate which can withstand internal pressure. New stream-
lined trains use high tensile, low alloy steels and stainless steels which have been
developed by the steel industry. Special heat treatments have been discovered
which, when applied to rails, insure better and longer service.
32 See the following table:
Selling Expenses and Advertising and Promotion Costs of SIB Manufacturers in 1931
[In Per Cent of Net^Sales]
Product
Direct
Selling
Costs
Advertis-
ing & Pro-
motion
Total
Consumer Products:
11.3
17.1
14.8
15.8
21.3
11.5
10.9
11.5
12.8
12.9
15'l
8.7
9.1
8.2
1:5
5.4
14.6
11.8
10.0
9.4
10.6
12.0
9.0
10.2
8.8
5.1
18.4
7.5
6.1
7.9
3.2
6.7
6.0
0.3
6.2
6.8
4.0
3.7
2.9
3.7
2.2
1.6
8.2
3.6
5.3
4.4
3.0
3.1
2.5
3;o
2.0
1.1
1.7
1.3
29.7
Paints and Varnishes
2* 6
Furniture
20 9
Heating Equipment
23 7
Office Equipment and Supplies
24 5
Confections and Bottled Beverages
18 2
Petroleum Products
16.9
Hbusohold Appliances
19.6
Clothing
15.3
Shoes....
Hardware
11 3
9.8
Tobacco Products
Sporting Goods ..
11.4
12.0
Radio Equipment
10 7
Industrial Products:
11.9
11.8
Electrical Equipment
15 0
Iron and Steel and Their Products
11 0
Nonferrous Metals
11 3
Transportation Equipment
10 5
Textiles
6 4
An Analysis of the Distribution Costs of SIS Monvfactmers, Associations of National .\dvertiser
and the National Association of Cost Accountants, New York, 1933, pp. 64. 160.
13912
CONCENTRATION OF ECONOMIC POWER
To produce these better products and still keep costs down, the steel industry
over the years has constantly improved its equipment and has developed entirely
new equipment such as the continuous sheet and strip mills which so recently
revolutionized the industry. It cannot be said that the steel industry has been
remiss in providing better materials to be used by other industries to make
products and provide services. This functioning of the steel industry to supply
new and better steels is particularly germane to the pricing problem since quality
improvements are usually not reflected in price series. In addition, many types
of steel which are in actuality new products may be known by the names orginally
applied to the products they replaced and as a result the new products and the
old may be included in single price series although they may have little or no
homogeneity.
As a Factor in Employment.' — Steel prices would be even more important to
the national economy if they influenced the amount of goods that could be sold
by companies for which the steel industry is a source of supply, and so affected
the rate of employment in those industries. This study has indicated that the
price of steel is of negligible importance as a factor in the demand ♦'or goods made
of steel because of the small perce age of the cost of the steel as related to the
cost of the finished product. Steel prices have little effect on national production
or employment. This is not to imply that the steel industry may charge any
price its whim or fancy may dictate. Competition among producers, and bargain-
driving purchasers with large orders to place, keeps prices at levels which some-
times do not even cover costs.
It has been charged by some that steel prices have remained firm in the face
of falling demand, and as a direct result production and pay rolls have declined
drastically. If the implications of this charge could be sustained it would be a
serious indictment. But they cannot be sustained. This study has shown that
the demand for steel is derived from the demand for goods made of steel. This
demand depends in turn on such factors as the level of national income and con-
fidence that in the future there will be opportunity for the profitable use of addi-
tional durable goods. The total demand for steel is inelastic; that is, the total
quantity of steel bought from t' e industry would not be substantially different
at any particular time if the pric : were higher or lower. The steel industry must
have orders on hand before it ca ^ produce; steel is made to exacting specifications
for particular uses; the very buikiness of such steel items as might be made in
anticipation of future demand prevents their heavy production for inventory.
If there is lack of confidence in the future and declining national income, produc-
tion and consequently hours of employment, will decrease despite all efforts of
steel producers. Only confidence in the future and actual or anticipated increase
in national income can create production and. resultant employment in the steel
industry.
Despite" the negligible influence of price on demand for steel, and waiving the
fact that the composite published price of steel is more flexible than critics often
suppose, and the further fact that net yields received by the industry are more
flexible than indicated by published figures,^^ what adjustments would have to
be made if steel prices were cut appreciably? Since substantial "fixed" costs
must be" met regardless of the amount of steel produced, prices cannot be out of
line with total costs over any considerable period.
What costs could be adjusted if prices were substantially reduced when the in-
dustry was operating at 50 percent of capacity? Based on cost data of the
United States Steel Corporation and its subsidiaries previously discussed, pay-
rolls would be approximately 50 percent of total costs at that rate of operation;
goods and services purchased from others, 34 percent; taxes and depreciation and
the following table:
Indexes of Prices
[1926=100]
Year
Composite
Iron Age
Prices
U. S. R. C.
Mill Net
Yields
Year
Composite
Iron Age
Prices
V. S. S. C.
Mill Net
Yields
100.0
95.1
93.5
95.4
88.5
84.5
82.1
100.0
, _ 96.5
9? 3
94.5
87.9
81.3
78.8
1933 -.
81.2
87.8
88.9
89.7
106.4
103.4
76.7
1927
1934....
89.1
1928
1935
90.9
1928 .
1936
88.6
1930 .
1937
99.6
1931.
1938
99.8
1932 .
CONCENTRATION OF ECONOMIC POWER 13913
depletion about 7 percent each; and the remaining 2 percent of total costs would
represent interest to bondholders and pensions to retired workers. There is no
getting away from taxes; they must be paid. Depreciation and depletion charges
could be overlooked for short periods, but not for long. If interest were not
paid, the Company would be forced into bankruptcy. The remaining 84 percent
of total costs represents payrolls and goods and services purchased from others.
Goods and services purchased from others perhaps could be obtained at lower
prices by sharp bargaining where the prices are not fixed by law as they are in the
case of railroad rates. Payrolls remain. They are 50 percent of total costs.
There is very little doubt that any appreciable cut in steel prices over the long
run would have to be met by reducing wage rates.
As a Factor in the Growth of the Nation. — This study has discussed the productive
capacity of the, steel industry and indicated the reasons why unused capacity may
be present in certain periods, but excess capacity, in the sense that it is not
necessary to the economic well-being of the industry and of the nation, is absent.
It has been shown that assembly costs of raw materials, the geographical location
of markets for products that may be economically produced together, the im-
mobility of steel-making equipment, the huge investment required therein, and
the historical development of individual companies are more important than the
pricing method in accounting for the existence of more capacity in certain dis-
tricts than local consumption might seem to dictate. It has been pointed out
that steel-making capacity has developed in every area where raw material as-
sembly costs, costs of production and nearness to consuming markets have been
conducive to such development. On these bases it cannot be contended that the
price structure of the steel industry has been instrumental in the preservation of
uneconomic capacity nor in the prevention of the expansion of economic capacity.
In brief, the steel industry has efficiently performed its function in the national
economy, has materially assisted in the development of this country, and has
ever been prepared to meet the needs of the nation in each forward surge of
prosperity as well as in times of national emergency.
Exhibit No. 1411
A STATISTICAL ANALYSIS OF THE DEMAND FOR STEEL, 1919-1938
This is an analysis prepared by the Special Economic Research Section of
United States Steel Corporation, composed of Messrs. Edward T. Dickinson,
Jr., Ernest M. Doblin, H. Gregg Lewis, Jacob L. Mosak, Mandal R. Segal,
Dwight B. Yntema and Miss Marion W. Worthing. The work of this group
was under the supervision of Theodore O. Yntema, Professor of Statistics,
University of Chicago. This analysis was written by H. Gregg Lewis who had
the benefit of suggestions from other members of the staff. It is issued by United
States Steel Corporation.
November 1, 1939.
CONTENTS
I. statement of the Problem
II. Summary of Conclusions
III. Some General Considerations on the Demand for Steel
IV. Method of Analysis
V. Factors Which Might Be Expected to Influence the Quantity of Steel Sold
VI. The Period Studied
VII. The Demand Relation Hypothesis
VIII. The Statistical Findings
I. Statement of the Problem
This analysis,! undertakes to measure the importance of the level of steel
prices in determining the quantity of steel ^ sold. More specifically the question
to be considered in this study is:
If the average level of steel prices in any year had been higher .
or lower than it actually was by a certain percentage, but every-
.' No attempt will be made In this paper to summarize or criticize previous statistical studies of the demand
for steel, except as this study does so by implication. However, the following reports should be consulted
in connection with this paper: Henry L. Moore, Economic Cycles: Their Law and Cause (New York, 1914);
Roswell H. Whitman, "Statistical Investigations in the Demand for Ironand Steel", Ph. D. dissertation.
University of Chicago, 1933; National Resources Committee, Industrial Committee, Patterns of Resource
Vne (Preliminary Edition for Technical Criticism); (Washington, 1939), pp. 63, 65, 128-129, 131-132.
' Throughout this paper the term steel should be understood to include only those products sold by the
steel production industry— i. e., what is generally understood as the steel-works and rolling mills industry—
to consumers outside that industry. The term pTodndf-madefrom-steel includes all products into which
steel so defined enters as a raw material of production.
13914 CONCENTRATION OF ECONOMIC POWER
thing else had been the same,' by what percentage and in what
direction would the quantity of steel sold in that year have
changed? In other words, what is the price elasticity of demand
for steelf
II. Summary ofConclusions
The analysis of the following pages indicates that in the period 1919 to 1938,
year to year fluctuations in the quantity of steel sold are accounted for in major
part by changes in economic factors other than the price of steel. Only a very
small part of tlie changes in steel sales can be attributed to steel price changes.*
The statistical analysis indicates, although not entirely conclusively, that the
demand for steel is very inelastic, i.e., that changes in the level of steel prices
(other conditions of steel demand remaining the same) cause much smaller per-
centage changes in the opposite direction in the quantity of steel sold.' The
best estimate of the elasticity of demand for steel indicated by this analysis is
approximately .3 or .4.
This means that very large reductions in price would be necessary to effect
significant increases in the volume of sales. Such price reductions would decrease
the gross income of the steel producers, while at the same time increasing their
total costs of production.
The major factors affecting the demand for steel, such as consumers' income,
industrial profits and business anticipations, seldom remain constant. In the
period 1919-1938 fluctuations in these and other factors were of such great mag-
nitude and importance that it would have been impractical to attempt to maintain
the level of steel production by compensatory changes in steel prices.
III. Some General Considerations on the Demand for. Steel
It may seem that the economic and statistical problems involved in an econo-
metric analysis of the demand for steel are simple. The demand for no other
product, however, is more complex, or presents greater analytical problems.
A. steel is not a homogeneous commodity
The steel industry is generally pictured as a mass-production industry, selling
only a few types of steel products, a pound of which is like every other pound of
the same type in physico-chemical composition, degree of processing or fabrication,
general shape and dimensions.
Actually, the steel industry produces thousands of steel products, most of which
are practically made-to-order to the chemical, physical, shape, and dimension
specifications of each buyer.' And each of the many steel products has its own
price.
It ii obvious that a demand analysis cannot reasonably be made for each of
these innumerable steel products. Thus, one is confronted at the outset with
the problem of combining all steel products into a composite who^e quantity and
price can be measured.'
B. steel is a raw material, a producers', not a consumers', good
Steel as it is sold by the steel producers usually is not a finished product ready
for use (consumption) by the ultimate consuming public. It is a raw material
used by its buyers, along with labor, machines, and other raw materials in the
production of products-made-from-steel.*
Thus the demand for steel does not depend solely and directly upon t^e con-
ditions determining consumers' purchases — but is indirectly derived from the
conditions affecting the output of products-made-from-steel.
3 Except to the extent that changes in other factors averting the demand for steel are caused by the change
in the level of steel prices.
< See Section VIII, pp. 25-28. In the years 1919-1938 chanpe,s in the level of steel prices were generally
of smaller relative magnitude than changes in other factors afTecting the demand for steel. It is obvious
that large price changes have greater effects than small price changes. But the effects of the changes in
steel prices were so smalt in the period studied as to afford no basis for the inference that considerably
greater price changes would have been more than a minor influence in determining the volume of steel
produced and sold.
» See Section VIII, pp. 27-28.
» See, for example, the list of steel products in the Census of Manufacturers, 1929 (United States Depart-
raont of Commerce, Bureau of the Census, 1933T; pp. 953-958. Each type listed is composed of many differ-
pnt steel pro^iucts sharing only the common characteristic s of the type. See also the list of steel products
for which prices are published weelvly in the steel trade journal. The Iron Age.
7 See pp. 4 1 -44 for a further discussion of this problem.
» See p. 1, footnote 2 above for a definition of steel and prcducts-made-from-steel.
CONCENTRATION OF ECONOMIC POWER 13915
This complicates the analysis because the amount of steel sold to a orodnopr
of products-made-from-steel depends largely on: producer
(1) His current and expected output of products-made-from-steel This is
m turn dependent upon an interrelation of'^numerous factors^uch as-
fhi''i^^'%?>!''"'''l* Tk"^ expected costs of production, including not only
the cost of the steel he uses, but many other costs as well
prc^dU'ttm^tSslee™"'"'"*' °' *^^ ""^'^^* ^" "^^^^ ^« ««^ ^-
wii?depend upon * ''^ '*^'' ^^ T' ^^^ ""^* °^ Product-made-from.steel, which
of 5LT ^ni'^'^'^K °.-?^i''^' characteristics of his product-made-from-steel:
(b)Th\ price oTsfee^" "^''"'''' ^^^ ""^ ^^' production methods. '
(c) The cost of using substitutes for steel.'"
C. STEEL is USED IN THE PRODUCTION OF MANY WIDELY DIFFERING KINDS OF
PRODUCTS-MADE-FROM-STEEL
That products-made-from-steel are almost innumerable and widelv diverse in
kind IS a point that need not be labored. One has only to observe 'the number
of products-made-from-steel which enter into everyday activity
nf ill tl^r.? r"' ^'"^,^^^ relations among the factors, determining the outputs
of all types of products-made-from-steel were more or less identical: the diversHv
of products-made-from-steel would present no great analytical difficStiesTndP
from'si^eV nterTT' '^^ '''''■ I^^ ^^vious. ho'wever, tha't since proSute-m\l^^
,v™;- enter into so many diflfenng aspects of economic activity, the deter-
minations of their outputs must also differ greatly. The way in which the outmft
«ndl3?'''' f determined is.certainly much different ?rSiat for Lto^^^^^^^
and that for automobiles different from that for battleships auiomoDiies,
It IS clearly an impossible task to make an analysis of the output of everv tvne
n P^«duct-made-from-steel.H Thus again we. have the index number probleS
of combining the many products-made-from-steel, the factors which dete?miS
nu^mVe?KroiL'^corpSs:^i ^^^^ ^'^' ^^ ^^^ ^ -^P"* int^ot rfalrCS
D. STEEL IS LARGELY USED IN THE PRODUCTION OF DURABLE GOODS '^
onl^rZ.f!^'"^''^^'^' e^o^o^icf to observe that goods are valuable-that is, can
command a pnce on the market-only for the services thev provide." Thus the
demand for goods essentially is derived from the demand 'fo? the services of thl
The pecuhar characteristic of durable goods is that they can provide services
over a long period of time. Once a stock of durable goods has been built u^fj
minishi^iflnr?'^ ''°"?°^^' societies-it is possible to obtain an atoostS
yZuZl r t^^'^^'T' f'-o'" them for u long period of time without the production
JJJI^ , "^l' '^''"'^i- r?"«^«P<'««. that is, may go on without a correspond-
wlfenTf "!« . °^ new durable goods. New durable goods wiU be produced only
when It IS economically desirable to replace "worn out" durable goods Tnd to
124491 — 41— pt. 26 22
13916 CONCENTRATION OF ECONOMIC POWER
3nlarge the stock of durable goods. Thus it is obvious that the production of
new durable goods tends to be largely dissociated from the consumption of the
services of the stock of durable goods. For the same reason, the amplitude of
cyclical fluctuations in the production of new durable goods will tend to be greater
than the variations in ^he consumption of durable goods and in the production
and consumption of perishable goods. Since the demand for steel is derived
largely from the production of new durable goods, it follows that there will be
great cyclical fluctuations in the quantity of steel sold.
Upon what factors does the demand for new durable goods depend? Inasmuch
as the conditions of demand for prodvcers' durable goods differ in some respects
from those for consumers' durable goods, each of these types will be discussed
separately.
(1) Factors Affecting the Demand for New Producers' Durable Goods. — Broadly
speaking, a producer will not. purchase a new durable producers' good unless he
can reasonably expect that the return attributable to the new good over its "life
span" will be sufficient to cover all the costs (including a reasonable profit) attrib-
utable to the purchase and use of the good. That is, the purchase must be
expected to be a profitable one.
Among the most important factors determining the profitability of such pur-
chases are:
(a) The current demand and the future demand expected by the producer
for his output of goods and services.
(b) His present stock (number of units, age and eflSciency of the units, and
expected life span of the stock) of durable goods.
(c) The purchase price of the new durable good, including financing
charges.
(d) The expected life span and efficiency of the new durable good. That
is, the expected life "capacity" of the new durable good.
(e) The "costs of using" the good — i. e., the labor, material, managerial
costs, etc., involved in the use of the good.
(f) The expected sale price per unit of the output of the good.
(g) The current and anticipated costs of (including the "costs of using")
substitutes, such as labor, for the new durable goods.
(2) Factors Affecting the Demand for Consumers' Durable Qoods. — The most
important factors affecting the demand for consumers' durable goods are: "
(a) The current and anticipated amount of consumers' disposable cash
income.
(b) The distribution of such income among economic classes.
(c) The size (number of units, age distribution, efficiency, and expected
life span) of the stock of consumers' durable goods.
Cd) The present and anticipated price of the new durable good.
(e) The costs of operating (including maintenance) the new durable good
(f) The cost of obtaining competing consumer services, including the
"costs of living."
(g) Consumers' tastes.
It is apparent that the demand for new durable goods is determined by a complex
composite of factors. Moreover, not all of the factors are directly measurable.
Since the complete "fund of services stored" in durable goods can be used up
(consumed) only over a more or less long pe'rod of time, anticipations are of para-
mount importance in determining the output of new durable goods. Thus there
arises the problem of "measuring" changes in producers' and consumers' states of
mind."
There is a further and very important analytical problem. Since the amount
of any commodity bought and sold depends not only on its price but also on a
complex set of other factors, an analysis which attempts to isolate the influence of
price is more difficult when the other factors are numerous, important, and subject
to great or rapid changes. If, for example, as is the case for certain staple agri-
cultural commodities, only a few factors other than price tend to be important in
determining the quantity sold, and tend also to follow a slow and regular routine
of change, the problem of isolating the effect of price is simplified. In the case of
durable products-made-from-steel, however, factors other than price are numerous,
exert very important effects, and tend to have large and irregular variations.
'• See Roos and von SzeliskI, op. cit.. for an analysis of the demanil for automohllcs.
" Inasmuch as current anticipations depend for the most part on the recent and current behavior of factors
which in many cases can be measured, an approximate measure of anticipations can often be obtained from
study of the measurable factors. See, for example, pp. 22-23.
CONCENTRATION OF ECONOMIC POWER 13917
Thus the problem of isolating the effect of the price of steel on the quantity of
steel sold is exceedingly difficult. '^
E. STEEL IS DURABLE AND CAN BE STORED
Since steel itself is durable, it may be kept in stock for fairly long periods without
serious physical deterioration." Thus, purchasers of steel may currently buy
more steel than they require for current (or anticipated near future) consumption,
building up a stock of steel for future production requirements. Conversely,
the building up of such a stock in the past enables a steel purchaser currently to
buy less steel than he consumes, the balance of such consumption coming from
depletion of his steel inventories. If changes in the size of steel inventories in the
hands of consumers (buyers) tend' to be large, then it is obvious that the size of
such inventories is an important factor influencing the sales of steel producers.
The size of steel inventories in the hands of consumers ^^ will depend for the
most part on:
(1) Buyers' anticipations as to future prices of steel.
(2) Their expected production requirements, which will depend largely on
their expected sales of products-made-from-steel.
(3) The expected length of time it will take to get delivery from steel
producers on future orders of steel.
(4) The cost of carrying such inventories.
If the steel buyer expects that prices of steel shortly will be higher, or that
near-capacity operations of steel producers may delay delivery on his orders at a
time when his steel requirements will be high, he may currently buy more than he
needs for current consumption, stocking steel as protection against future higher
prices or delivery delay. On the other hand, if his steel requirements turn out
to be smaller than expected, he may find himself with unnecessarily large inven-
tories of steel on hand. Thus he may consume from stock, curtailing his buying
below his current production requirements.
However, such changes in inventories, which are largely speculative, for the most
part exert only a short run effect on steel buying. The effect usually is a short
run shift in the time of the actual purchases, Tvithout changing the total amount
of steel bought over a one or two year period from what it otherwise would have
been.
The reasons for this are:
(a) Inventories of steel cannot be reduced below a certain minimum
(which depends largely on the level of the producers' operations) without
serious inconvenience. This is especially true when there is danger of delay
in delivery of orders of steel.
(b) On the other hand, the cost of carrying inventories and the risks in-
volved tend to set an upper limit to their size. The larger the inventories,
the higher is the carrying expense, and the further into the future must the
user anticipate prices of steel and his own production requirements. Such
anticipations become more risky as they extend longer into the future. The
situation seldom arises when the coets of carrying are low enough, and the
future certain enough to justify changing inventories by more than a few
months' production requirements.
'This is not to say, however, that year-to-year fluctuations in steel inventories
are unimportant in explaining year-to-year changes in steel buying. In periods
of rapid change in business activity and business outlook — such as the period from
the middle of 1936 to the middle of 1938 — changes in the size of inventories may be
very important.
Thus in analyzing the demand for steel it is necessary to include as a factor net
changes in steel inventories in the hands of steel buyers, or in the absence of such
data, the factors upon which the size of steel inventories depends.
" An excellant discussion of the problem of isolating the effect of price in the derivation of quantity-price
demand relations is contained in Henry Schultz, Theory and Measurement of Demand (Chicago, 1938),
pp. 61-104.
" There are, of course, exceptions to this statement; for example, cold reduced auto sheets should be used
promptly.
so This section deals only with changes in inventories of steel in the hands of consumers. However, steel
producers themselves may keep stocks of steel. Inasmuch as the largest part of the steel produced is made
to order to the buyer's specification, changes in inventories of finished steel in the hands of producers are
ordinarily small. There is some evidence, however, that changes in inventories of steel ingots, semifinished
steel/ and standard types of finished steel in the hands of producers may at times be quite large. Such
changes of inventories in the hands of producers are relevant to the discussion of this paper only if it is neces-
sary to derive estimates ;3f steel sales from figures on steel production. For a discussion of the latter problem
see Appendix VIII, Section 1.
13918 CONCENTRATION OF ECONOMIC POWER
F. STEEL IS NOT SOLD IN A SINGLE ONE-PBICB MARKET 2'
Largely because steel producers and steel buyers are located over a wide area,
and also because it is impossible at all times for all buyers and sellers of steel to
have "perfect knowledge of the market," there tend to exist at any time certain
differentials between the prices paid for the same type of steel by different buyers."
These differentials are of two main types:
(1) First there are the more or less permanent price differentials between
buyers in different geographic areas. These differentials have arisen partly
from varying costs of assembling raw materials and converting. them into
finished products at different locations, partly from varying costs of trans-
portation of the finished product into different areas, partly from the forces
of competition, and partly from certain long established institutional arrange-
ments in the pricing of steel. ^^ These same forces, however, tend to keep the
differentials more or less constant, so that year to year changes in the price
of steel are about the same in all areas.^*
(2) The second type of price differential is the concession from the pre-
vailing price. Because of competition among steel producer.'^, it is obviously
advantageous at certain times for certain steel producers to offer steel at lower
prices than their competitors." By so doing they can often take a substantial
share of the steel market away from competing steel companies. ^^ However,
the same forces of competition require that such price concessions be kept
from the knowledge of competitors; otherwise the concessions will be met and
become general." When concessions do become general, data on the price
cuts ordinarily become market knowledge available to the steial, trade journals
who report "going" market prices. ^^
The combination of these two types of price differentials means that at any time
there tends to be more than one price for the same type of steel. Thus there
arises the problem of combining these prices into single composite prices for the
various types of steel. 2'
IV. Method of Analysis
In order to make the discussion of the following pages clear it is necessary to
define the terms "quantity of steel sold," "products-made-from-steel," "inven-
tories of steel in the hands of producers of products-made-from-steel," and
"jobbers' stocks."
(i) By quantity of steel sold is meant the physical quantity, i. e., tonnage of
steel sold by steel producers.
(ii) A product-made-from-steel is any finished producers' or consumers'
good {not service) into which steel enters as an actual raw material of produc-
tion. Automobiles, steel bridges, rails-laid are examples. Products-made-
from-steel are of two broad categories:
(a) Products-made-from-steel produced for sale. Automobiles, house-
hold appliances, agricultural implements are in general examples of
this type.
(b) Products-made-from-steel produced for the producer's own use.
Rails-laid, bridges and highways, pipe-lines-laid are examples of this
category.
»i This section is not to be interpreted either as an attempt to describe fully or to appraise the pricing and
sellin? arrangements in the market for steel.
" Obviously this problem is not confined to the marketing of steel. Such price differentials will almost
always arise when there is more than one seller and one buyer.
" Discussions of the baslng-point method of pricing are especially relevant here. See the description in
Daugherty, de Chazeau and Stratton, Economics of the Iron and Steel Industry (McOraw-Hill Book Com-
pany, New York. 1937), Vol. I, pp. 533-544.
" Mill net indexes for different basing point areas support this conclusion.
" Price concessions are especially advantageous (from the short-run point of view of the individual seller)
for producers (I) having relatively small steel producing capacity (2) operating at a low percentage of capacity.
By making concessions such producers may gain enough business to raise their operation to a rate at which
they can make substantial profit gains (or loss reductions).
M Sec the recent discussion in Paul M. Sweczy, "Demand under Conditions of Oligopoly," Journal of
Political Economy, v. XLVII, No. 4 (Aug., 1939), pp. 568 et seq.
>' See Sweezy, op. cit.
" Undoubtedly there are times when price concessions arc important, and when it is difficult for the trade
Journals to verify or measure the extent of the concessions. Ordinarily the trade journals can measure the
extent of -the concessions only when the market for some type of steel "breaks wide open."
'• This, too, is a difllcult index number problem that has never been satisfactorily solved. Ideally the
solution requires separate demand analyses for each group of buyers subject to the same price differentials.
CONCENTRATION OF ECONOMIC POWER 139] 9
(iii) By inventories of steel in the hands of producers of -products-ynade-from-
steel — or simply steel inventories — is meant the total of:
(a) The physical quantity of steel on order and held in raw material
inventories by producers of products-made-from-steel; and
(b) The equivalent measure of the amount of steel held by producers
in work-in-process inventories (partiallv fabricated products-made-
from-steel). ^o
(iv) The term jobbers' stocks means the physical amount of steel held by
jobbers — i. e., steel middlemen.
The quantity o-f steel sold in any year then is obviously equal to the number of
units of products-made-from-steel produced in that year multiplied by the average
amount of steel used in the production of each, plus the net change in steel inven-
tories from the beginning to the end of that year plus the similar net change in
jobbers' stocks.
For example, suppose there is only one type of product-made-from steel, say
automobiles. During the year automobile producers manufacture 5 million cars
using 1.5 gross tons of steel in the production of each car. Then 7.5 million gross
tons of steel would be required to produce the 5 million cars. At the beginning
of the year automobile producers held 0.5 million gross tons of steel in steel inven-
tories (in unfabricated form or in equivalent measure in work-in-process), and 1.0
million gross tons at the end of the year. It is obvious then that automobile pro-
ducers (producers of products-made-from-steel) must have bought 8.0 million gross
tons of steel during the year, 7.5 million of which was used in prodiiction and 0.5
miUion to increase their .steel inventories. Of this 8.0 million tons, 7.0 million
were bought directly from steel producers, and 1.0 million from jobbers. Jobbers
meanwhile increased their stocks from 1.0 million to 2.0 million tons, so that they
must have bought 2.0 million tons from steel producers, 1.0 million to increase
their stocks, and 1.0 to sell to producers of products-made-from-steel.
Steel producers then must have sold 9 milhon gross tons of steel during the year:
7.5 million of which went into the production of products-made-from-steel;
0.5 million to increase the steel inventories of producers of products-made-
from-steel; and
1.0 million to increase jobbers stocks.
It is therefore clear that in determining the effect of a change in the price of
steel on the quantity of steel sold. in any year, it is necessary to find out how such
a price change would affect:
(i) The physical volume of production of products-made-from-steel.
(ii) The average quantity of steel used in the' production of each unit of
product-made-from-steel.
(iii) The net change in steel inventories,
(iv) The net change in jobbers stocks.
It will be helpful to consider how each of these is determined.
(1) Volume of Production of Products- Made-from-Steel. The output (number
of units) of products-made-from-steel in any year may be looked upon as made
of three parts:
(a) The quantity of such products sold.
(b) The net change in producers' inventories of products-made-from-steel
produced for sale.
(c) The output of products-made-from-steel manufactured for the pro-
ducer's own use.
The sum of the first two is obviously equal to the output ci products-made-
from-steel produced for sale.
(a) The main factors which determine the volume of sales of products-
made-from-steel have been discussed in Sections III-B and III-D. Since
the price of steel will in general affect the volume of sales only indirectly
through Its effects on the prices of products-made-from-steel. the problern
becomes the two-fold one of first isolating the influence of the prices of prod-
ucts-made-from-steel on their sales, and then determining how the price of
steel affects prices of products-made-from-steel.
This second problem is itself an extremely diflScult one, for the prices of
products-made-from-steel depend on many factors beside the price of steel
>" That is, the amount of steel used In the production of inventories of partially finished products-
from-steel.
13920 CONCENTRATION OF ECONOMIC POWER
The price of steel will exert its eflfect on the price of products-made-from-steel
through its effect on their costs of production, and the importance of the
effect will depend in large measure on the importance of steel costs in the
total unit costs of productioni
In general it seems likely that an increase in the price of steel, other things
remaining the same, will increase unit costs of production and thereby the
prices of products-made-from-steel. The increase in the prices of products-
made-from-steel would probably lead to a reduced number of units sold.
For a reduction in the price of steel, the converse is probably true.
Moreover, it is reasonable to assume that ordinarily:
(i) Sales of products-made-from-steel will not be very responsive to
changes in their prices. That is, changes in the prices of products-
made from-steel do not lead to much larger percentage changes in sales.''
(ii) Changes in the unit costs of production of products-made-from-
steel lead to approximately equivalent changes in their prices.
(iii) Changes in the price of steel generally lead to changes in the costs
of production of prodncts-made-from-steel in approximately the pro-
portion of unit steel costs to total unit costs of production.
Since steel costs are generally only a small fraction of the total costs of pro-
duction of products-made-from-steel, it follows that one may reasonably
expect changes in the price of steel to lead to much smaller percentage changes
(in the opposite direction) in the sales of products-made-from-steel.
(b) The size of inventories (in the hands of producers) of products-made-
from-steel produced for sale depends in major part on :
(i) The current and expected volume of sales of products-made-from-
steel.
(ii) Current and expected labor costs of production of products-made-
from-steel.
(iii) The current and expected length of time required to make
delivery on orders of products made-from-steel.
Of these three factors it seems reasonable to expect that only the first would
be significantly affected by changes in steel prices. Since net changes of
inventories of products-made-from-steel tend to be small relative to sales ana
since sales are only slightly responsive to changes in the price of steel, it thus
seems likely that steel price changes are of negligible importance in affecting
steel sales through their effects on inventories of products-made-from-steel.
(c) The major factors determining the output of products-made-from-
steel produced for the maker's own use have been discussed in section III'-D.
The price of steel will affect the output of such products mainly through its
effects on tb-^ir costs of production, maintenance, and operation. Since the
output of durable producers' goods is probably not very responsive to changes
in their costs of production, operation, and maintenance, and since steel costs
are generally only a very small part of such costs, changes in the price of steel
lead to much smaller percentage changes (in the opposite direction) in the
output of products-made-from-steel manufactured for the producers' own
use.
(2) The Average Amount of Steel Used in the Production of Each Unit of
Products-Made-From-Steel. In section III-B it was pointed out that the
amount of steel used per unit of output of a product-made-from-steel depends
largely on:
(a) The technological characteristics of the product-made-from-steel; of
steel and substitute raw materials; and of the production methods.
(b) The price of steel.
(c) The cost of using substitutes for steel.
"The first set of factors (a) however, tend to change very slowly, so that
they may be classed as "long-run" factors which are relatively unimportant
in determining year to year fluctuations in the average amount of steel used
per unit of product-made-from-steel.
A rise (or fall) in the price of steel relative to the cost of using substitutes
will ordinarily lead to a decrease (or increase) in the amount of steel used per
unit of product-made-from-steel. However, it is extremely doubtful if a
change in steel prices in any year has more than a negligible effect in induc-
ing substitution in that year (or even in the next two or three). The reasons
for this are obvious:
(i) The tastes of buyers of products-made-from-steel tend to change
very slowly, and thus retard the rate of substitution.
" The demand for durable goods is generally not. very elastic.
CONCENTRATION OF ECONOMIC POWER 13921
(ii) There are a limited number of ^dbstitutes that are economically
and technically suitable. Moreover technical conditions of production
rigorouslj^ limit the amount of substitution that can take place.
(iii) The use of substitutes for steel, or the substitution of steel for
other factors of production generally requires great changes in the type
of plant, equipment, and labor required. A change in the price of steel
relative to the cost of substitutes must normally be substantial and
persist for several years before the investment and labor training costs
required by substitution will be undertaken.
(3) Net Changes in Inventories of Steel in the Hands of Producers of Products-
Made-From-Steel, and Net Changes in Jobbers' Stocks.
The main factors determining net changes in inventories of steel in the hands of
producers of products-made-from steel and net changes in jobbers' stocks were
discussed in section III-E, and will not be elaborated further here.
Once it is known how (1) the output of products-made-from-steel, (2) the
average quantitj' of steel per unit of this output, (3) inventories of steel, and (4)
jobbers' stocks are determined— including the influence of the price of steel in
such determination, then it will be evident how the quantity of steel sold is deter-
mined, ^nd how steel price changes would affect the quantity of steel sold.
Although the above approach is ideal in that it enables the analyst to get a clear
picture of the separate ways the price of steel acts in determining the quantity of
steel sold, it is impossible to follow it here. Information is lacking at critical
points:
(1) Data on the output of important products-made-from-steel are lacking.
Moreover, even if output data were complete, the great diversity of products-
made-from-steel as well as the fact that they change in nature from year to
year makes the problem of combining them into an economic composite an
almost impossible one.^^
(2) Data on the quantity of steel used in the production of different types
of products-made-from-steel, as well as data on steel inventories and jobbers'
stocks are also almost completely unavailable.
Thus it might seem that the whole problem would have to be stranded without
■dv, answer. However, it is obvious from the discussion of the previous pages
that the quantity of steel sold depends upon the factors which determine (R) the
output of products-made-from-steel, (S) the average quantity 'of steel per unit
of the output, (T) steel inventories, and (L'^) jobbers' stocks. Thus if the most
important of these basic determining factors can be measured, if such measures
are available, and if reasonable hypotheses — determined from economic logic and
empirical observation — can be set up as to the relation between these factors, then
the problem may not be insoluble. However, it should be obvious that the results
obtained by such procedure will not be as conclusive as those obtained by the
approach outlined on preceding pages, since there will not be the intermediate
checks on hypotheses as to the demand relationships that the first approach offers.
The problem remaining then is a threefold one:
(1) The clear definition and measurement of the most important factors
determining the above economic variables (R), (S), (T), and (U).
(2) The setting up — on a priori and empirical grounds — of an hypothesis
as to the way these factors act together in determining (R), (S), (T), and (U),
and thus the quantity of steel sold.
(3) The statistical testing of the hypothesis.
V. Factors Which Might Be Expected to Influence the Quantity of Steel
Sold ^3
From the discussion of the previous pages it is clear that the following factors
might reasonably be expected to influence the quantity of steel sold:
(1) The price of steel — including both the level and the direction of change.
(2) Consumers' disposable cash income.
(3) The distribution of the income among income classes.
(4) The stock (number of units and efficiency) of durable goods — both
consumers' and producers'.
(5) The cost of living.
(6) The prices of goods and services which compete with products-made-
from-steel for the outlays of producers and consumers.
(7) The costs of maintaining and operating products-made-from-steel.
-32 See section III-C.
" See Appendix VIII below for further liscussion of the problem of defining and measuring the important
factors m the demiind for steel.
13922 CONCENTRATION OF ECONOMIC POWER
(8) Industrial profits.
(9) The psychological atmosphere — i. e., producers' and consumers' anti-
cipation as to future economic conditions.
(10) Industrial production-.
Since some of these variables are very highly related to others, however, and
since others tend to change slowly and smoothly from year to year, certain of
them were omitted in the actual analysis. The factors which were used in the
final statistical analysis were:
(1) The price of steel — both its level and direction of change.
(2) Industrial production — both its level and direction of change.
(3) Consumers' income — both its level and direction of change.
(4) Industrial profits— both its level and direction of change.
(5) The cost of living.
(6) A time-trend variable.
As will be pointed out below ^* these six factors can be taken as approximately
representing all of the preceding ten.
Vri. The Period Studi'=:d
The period 1919 to 1938 was chosen for analysis for the following reasons:
(1) It was a long enough period to provide observations on the nature of
the demand for steel under practically all types of conditions so that somewhat
general inferences could be drawn from the data. The period covered includes
both years of boom and years of dechne.^'
(2) This period is of more current interest than earlier periods, because
the inferences drawn are of more accurate current application.
(3) Data for years prior to 1919 are very often not available.^'
However, after the analysis was begun it -^as found desirable to exclude the
years 1919-1921 from some of the demand relations. The analysis indicated
that the situation in these three years was abnormal because of the World War.
The magnitude and the direction of the fluctuations in economic activity were not
typical of the rest of the period, and tl;ie inclusion of these years, it was thought,
obscured the ordinary steel demand relations. The statistical analysis, however,
was in most cases carried through for both the complete and the abbreviated
periods.
Annual data^ rather than monthly data or data for periods longer than a year,
were selected for analysis for three reasons:
(1) Monthly data were not available for some of the series.
(2) The use of monthly data unnecessarily complicates the analysis for
the purposes of this paper because it introduces short-term factors — such as
seasonal variations and short-run speculative activity — which are practically
excluded by using annual data."
(3) The use of longer-period data was considered undesirable because
(a) A much longer period of years would have to be studied in order
to get a sufficient number of observations.
(b) The effect of year to year changes in demand conditions on steel
sales was desired.
(e) It is extremely difficult to isolate the causative effect of price when
longer-period data are used. The use of longer-period data introduces
many new factors into the analysis which can be considered as unim-
portant in studying year to year changes.
VII. The Demand Relation Hypotheses '*
The final problem remaining prior to the actual statistical determination of
the demand for steel is that of setting up an economically logical hypothesis as
to the way the factors considered in section V act together in determining the
quantity of steel sold. This is by far the most important part of the whole analysis
X See-section VII-A.
« When annual data 8i« used, the statistical technique here employed requires a period as long ss fifteen
of twenty years in order to get a suS'cient number of observations. The reasons for this are tecbrlcsl and
will not be discussed here.
3« For example, rebaMe ^lata on consumers' Income, industrial profits, the ccst of living, and industrial
production are lot avsilabVin good form before 1919.
" Ideally, of course, it wo id be desirable to use monthly, or even shoiter p^iod data, since intra-year
variaiions tend to alTect anrtual measures. However, th« extra analysis was considered to be too great to
compensate (or the small loss of Information Involved in using annual data.
» See Roos and 7c n Szeliskl, op. cit., section III.
CONCENTRATION OF ECONOMIC TOWER 13923
of the demand for steel. It is obvious that the final inferences drawn — concerning
the influence of the price of steel on the quantity of steel sold — will depend on the
demand relation hypothesis set up.
The problem of setting up a demand relation hypothesis for steel is a perplexing
one. Products-made-from-steel are so numerous ana so diverse that it is almost
impossible to analyze the way economic factors act together in determining the
output of even the most important. Moreover, data which would be helpful are
lacking at critical points. Then, too, information as to the economic-technical
problem of the amount of steel used per unit of p»'oducts-made-f rom steel is almost
Chart 1
DEMAND FOR STEEL AND INFLUENCING FACTORS
IN UNITED STATES
S2«
si
11-:
STEEL INGOT
/\
J
/
s
1^1^
\
1
,
\
/
1
/
BOOKINGS^
\
\
f
I.
\
I
>
N
WteeH
\
^
V
^
^,
f
\
1
1
^
^
<^
>
1
^^
-•
k^
,Jci..lcoL
,'
/
zi
">,
J
\
*^^
\
INCOME 1 ,
'"'
"•s
•<
^
r-'
^
V
\
N
f'
»-
/
PROFITS
\
/
>
/
\
/
\
/
/
N
f
/
-L-i41-
^,
/
V
.^
/-
PRODUCTION
(I921. 25-1001
-•,
/
\
<^
1
>
V
Fl^
^
W
_
-
y
U
H
ISHED
ST
;el
completely unavailable. A similar situation exists for the problem of setting up
an hypothesis as to the determination of steel inventories.
The lack of information at critical points, and the absence of a completely
suitable body of economic theory have forced recourse to what is largely an
empirical determination of the demand hypothesis.
A. ACTUAL VARIABLES INCLUDED IN THE DEMAND RELATIOI^ HYPOTHESIS
Five general hypotheses as to the actual variables to be incl-uded in the demand
relation hj'pothesis were set up. The basic variables included in these various
hypotheses are shown in Chart 1 and Appendix I.
13924 CONCENTRATION OF ECONOMIC POWER
The five general hypotheses were:
The quantity of steel sold '» depends upon:
(h-1) The price of steel,*" and the volume of industrial production."
(h-2) A time-trend variable in addition to those of (h-1). ^
(h-3) The same variables as (h-2) and in addition two variables
measuring respectively the rate of change in the price of steel
and the rate of change in the volume of industrial production. <*
(h-4) The price of steel/" a time-trend, consumers' supernumerary
income," and industrial profits."
(h-5) The same variables as (h-4), and in addition three variables
measuring respectively the rates of change in the price of
steel, supernumerary income, and industrial profits.**
In (h-1) it was assumed that industrial production measured accurately the
composite influence of all factors affecting the demand for steel except the price of
steel. It was assumed that industrial production reflected the composite effect
of the most important demand factors, viz., industrial profits, consumers' income,
the replacement pressure on the stock of durable goods, and, also indirectly
the psychological outlook. - Since all of these factors have actually been more or
less highly correlated with industrial production, such an assumption is not
unreasonable.
In (h-2) an additional time-trend factor was, included. The time-trend was
included explicitly as a variable to act as a proxy measure for all factors influencing
the demand for steel which tend to change slowly and smoothly over a long period
of time. Thus it serves as a composite measure for such factors as population,
the size of the stock of durable goods, and long time changes in various price and
cost levels (including the level of the prices of steel), industrial technology, and
people's tastes. The inclusion of such a variable makes it possible partly to
isolate the effects of these long-run factors.
It is commonly recognized that a very important factor determining the current
level of activity in durable goods production — and thus in steel sales — is the busi-
ness outlook of producers and consumers, their anticipations as to future prices,
- profits, income, etc. Such anticipations are very largely determined by the ra-
pidity and direction of change in recent and current business activity. If present
levels of activity are higher than they have been in the recent past, it is easier to
believe that conditions will continue to improve than if activity is currently on
the decline. For this reason the rate of change of industrial production was
included in (h-3) as a factor measuring changes in anticipations. Similarly the
rate of change in the price of steel has been included in (h-3) as a measure of steel
buyers' anticipations as to the near future price of steel.
In (h-4) and (h-5) industrial profits and consumers' supernumerary income and
their respective rates of change have been substituted for industrial production
and its rate of change to measure the composite of factors other than the price of
steel influencing the demand for steel.
Of the five general hypotheses it would seem that |(h-5) is probably the most
complete and the most reasonable. The final answer, of course, cannot be given
until the form of the five general hypotheses is set up and tested.
'• Measured by steel ingot production, and estimated shipments and bookings. See Appendices II
and Viri.
*o The Iron Age '■omposite price of finished steel. See Appendix VIII.
*> The Federal Reserve Board index of manufacturing production excluding iron and steel. See
Appendices VI and VIII.
« The rates of change are for any year in each case measured by the link relative for that year. (The link
relative is equal to that year's figure divided by the figure for the previous year.)
" See Appendix IV.
" See Appendix V.
« The measure of the rates of change is the link relative, except for profits where the rate of change is
measured by first differences. (The first difference for any year is equal to the figure for that year less the
figure for the previous year.)
CONCENTRATION OF ECONOMIC POWER 13925
B. THE FORM OF THE DEMAND RELATION HYPOTHESIS
The next step in the analysis is the formulation of an hypothesis as to the way
the economic variables act together in determining the demand for steel.
Each of the five general hypotheses outlined above was studied by familiar
graphical multi-factor correlation techniques," in order to find out
(i) What mathematical relation seems to be the most reasonable expression
of the relation between the factors.
(ii) Whether any of the five general hypotheses should be discarded or
modified.
The graphical analysis indicated that for all of the hypotheses a simple additive
relation would probably give as satisfactory results as any other (such as the
multiplicative or combinations of the additive and multiplicative). ■•'
It was also decided from the graphical analysis to use only (h-2), (h-4), and a
modification of (h-3) which excluded the rate of change in the price of steel **
and the time-trend.
Thus four mathematical relations were formulated for further examination by
mathematical statistical techniques. Translated verbally these relations were:"
Relation I. — Production of steel ingots and castings is equal to:
Price of steel multiplied by a constant value
plus Industrial production multiplied by (another) constant value
plus Time (in years) multiplied by (another) constant value
plus a constant balancing value.
Relation II. — The same as relation I plus
The rate of change of industrial production multiplied by a con-
stant value and excluding the time-trend.
Relation III. — Estimated steel shipments are equal to:
Price of steel multiplied by some constant value
plus Industrial profits multiplied by some constant valuef
plus Supernumerary income multiplied by some constant value
plus Time (in years) multiplied by some constant value
plus a constant balancing factor.
Relation IV. — The same as relation III except that estimated steel bookings
were substituted for estimated steel shipments. ^o
The only problem remaining was to find the numerical values of the various
constant multiplying and balancing factors in the relations. Once this was done
it was easy to find out how much of a change in the quantity of steel sold (as
represented by bookings, shipments, or ingot production) has been associated
with a given change in the price of steel, industrial production or any other of
the independent variables.
« See Henry Schultz, op. cit., pp. 184-186, including the sources cited in footnote 7 on p. 185.
*' The additive relation has in its favor the simplicity with which the statistical analysis may be carried
out. More complicated forms of mathematical relations have, of course, certain logical advantage.^ arising
from their greater generality. It is well known, however, that if it is desired to study a demand relation near
the average values of its variables, the linear arithmetic form gives practically the same results as more com-
plex forms. Since it was considered feasible to study the relation only near its average values, and since there
was no clear indication from the graphical analysis that a more complicated form was a more likely one, the
additive relation was selected. However, the statistical analysis was also carried through for (h-2) using a
simple multiplicative (linear logarithmic) relation. See Appendix VII.
<« The graphical anaylsis indicated that no significant information would be added by the rate of change
terms in (h-5) and the rate of change of the price of steel in (h-3) , and that the inclusion of these terms might
break down the statistical analysis.
<» These relations are stated in mathematical form in Appendix VII.
" See Appendix VII.
13926
CONCENTRATION OF ECONOMIC POWER
VIII. The Statistical Findings
The constants were detennined by the least squares multiple correlation
technique.*' Final equations for the various relations are shown in Appendix VII.
The same statistical procedure also gives the percentage of the total variation
in the quantity of steel sold over the period studied that is accounted foi by the
economic factors included in the relations, and the amount that can be directly
attributed to the separate variations of each of the factors.'^ These percentages
are shown below in Table 1.'^
Table 1
Quantity of Steel Sold Measured
by
Percent of Variation in Quantity of Steel Sold
is
Directly attributable to variation in
Relation
Number
•s
is
|1
a
>>
ii
a a
ll
1
i
I
Production of Steel Ingots and
96
96
91
90
0
9
9
88
81
0
II
Production of Steel Ingots and
'_
III
IV
Estimated Steel Shipments
Estimated Steel Bookings
41
90
19
0
Two conclusions are indicated by Table 1:
(1) In each of the demand relations, the included factors accounted for
90 percent or more of the observed variation in the respective measure of
steel sales.
(2) Over the period studied only a small fraction (10 percent or less) of
the variation in steel sales was directly attributable to variation in steel prices,
while the major part of the variation was accounted for by included factors
other than the price of steel.
It should be emphasized, again, however, that these conclusions depend upon
the accuracy of three assumptions:
(a) That the demand relations set up are good approximations to the true
demand relations both as to factors included and the form of the relation.
(b) That the variables used more or less accurately measure what they are
supposed to.
(c) That the statistical technique yields approximately correct constant
values for the demand relation.
In appraising the second conclusion drawn from Table 1 it should be kept in
mind that the relative proportion of the total variation of steel sales attributable
to variation in the price of steel over any period will depend in part on the amount
of variation in steel prices relative to variation in the other factors. Over the
period 1922 to 1938 relative variation in steel prices was considerably less than
the relative variation in the other factors.
SI For an excellent description of the techniques followed see Schultz, op. cit.. Appendix C.
" For a rigorous definition of the term "variation" as used hern, and the details of the procedure used in
attributing variation in steel sales to the various "causative" economic factors see Schultz, op. cit., pp.
741-743. Simply stated, the percentage of variation directly attributable to any factor is the ratio of the
variation in the quantity of steel sold which would have taken place if only that factor had varied in the
way it did, to the variation in the quantity of steel sold that actually took place.
M Except for Relation II the period studied was 1922 to 1938; for Relation II, it was 1920 to 1938. It will
be noted that in all cases the sum of the percentages of variation directly attributable lo the separate factors
is not equal to the total variation accounted for by all the factors in the relation. The reason for this is as
follows: In obtaining the percentages attributable to any factor we assume that none of the other factors
varied. Actually, of course, this is not true; all of the factors varied, the changes in some factors tending
to increase the quantity of steel sold while other changes were tending to decrease sales. The net result of all
the simultaneous changes is the amount of variation accounted for by all the factors.
CONCENTRATION OF ECONOMIC PDWER 13927
A more useful measure of the importance of the price of steel is the elasticity of
demand coefficient. This coefficient is the ratio of the percentage change in the
quantity of steel sold to the corresponding percentage change in the price of steel,
other factors being fixed at some level.**
Table 2 below shows the values of the. elasticity of demand found in the four
demand relations when the values of the demand factors are at their average
levels for the periods studied."
Table 2. — Elasticity of demand for steel
Relation number: Elasikiiy
I +0. 12
II +0.52
III -0.21
IV -0.88
The values are consistent in this very important respect: they indicate that at
most a one percent decrease in the price of steel would cause {other factors remaining
the same) less than a one percent increase in steel sales {and conversely) . If this is
true, and if fluctuations in the other factors continue to be as great and as im-
portant as they have been in the past, the volume of steel consumption cannot be
stabilized by compensatory changes in the price of steel.
Which of the above values of the elasticity of demand is the most likelj'? The
values obtained from Relations III and IV are probably better than those from
I and II for the following reasons:
(a) On a priori grounds it seems reasonable that a change in the price of
steel would lead to a chi nge in the opposite direction in steel sales. *^ Rela-
tions I and II both indicate positive relations between steel prices and sales.
(b) As pointed out in Appendix VIII, steel ingot production is probably
not as accurate a measure of steel sales as the estimates used in Relations III
and IV.
(c) Industrial production (and its rate of change) is probably not as good
a measure of the composite of factors other than price of steel as the com-
bination of the two factors, industrial profits and supernumerary income.
The diff'erenoe in the values obtained from Relations III and IV can be due only
to the diff"erence between the estimates of steel sales used in each case, for the
relations are identical in other respects. From Chart 1 it is apparent that fluctua-
tions in the steel bookings figures used in Relation IV tend to lead industrial
profits, while steel shipments (used in Relation III) do so to much less degree.
The reasons for the lag of shipments behind bookings are discussed in Appendix
VIII. In Appendix VIII it is also pointed out that accounting profit figures tend
to lag behind the current profit situation that they supposedly measure. An
analysis of business historj- over the period 1919 to 1938 adds additional evidence
that the profits figures ordinarily reported have a significant lag.
In the graphical analyses that were made of the various demand relations, there
were clear indications that if the lags of shipments and industrial profits behind
bookings were removed, Relations III and TV w'ould both give about the same
results for the elasticity of demand, yielding a figure of 0.3 to 0.4. The evidence
and argument adduced in the preceding pages of this paper support the conclusion
that such a value^or one even lower — for the elasticity of demand for steel is not
a statistical happenstance, but a reality.
Although these findings are not absolutely conclusive in establishing this very
low elasticity of demand for steel, they certainly afford no basis for the view that
the price of steel is a practical medium for stabilizing production.
" A rigorous mathemsitical definition of the elasticity of demand is given in Appendix VII. It is also
shown there that the elasticity of demand may vary as the price of steel and the other factors influencing
the demand for steel vary.
55 These values of the elasticity have been computed at the arithmetic mean point of the factors influencing
the demand for steel.
56 See the 'iscussion of Section IV.
13928 CONCENTRATION OF ECONOMIC POWER
Appendix I. Basic Series Used in the Statistical Analysis
The basic series used in the statistical analysis are shown below in Table A-1.
These series are also shown in Chart 1, p. 21.
Table A-1. — Basic Series Used in Statistical Analysis
1910
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934.
1935.
1936.
1937.
1938
Steel
Ingot
Produc-
tion >
(thou-
sands of
gross
tons)
Steel
Book-
ings '
(thou-
sands of
gross
and net
tons)
Steel
Ship-
ments '
(thou-
sands of
gross
and net
tons)
20,783
27, 217
12,375
23, 705
29,173
24,154
29,639
30,847
28,827
32,560
36, 197
26,280
18, 431
10,385
15,607
16,222
21,050
28,766
31,620
18, 176
Indus-
trial
Produc-
tion*
(1923-
25 = 100)
Rate of
Change
of In-
dustrial
Produc-
tion »
Indus-
trial
Profits •
(bUlions
of
dollars)
6.419
4.468
-0.055
4.380
5.867
4.998
6.971
6.774
5.880
7.566
8.083
1.366
-3. 145
-5.375
-2.379
0.157
1.674
3.903
3.872
2.165
Supernu-
merary
Income '
(billions
of
dollars)
35.5
35.6
37.5
39.8
40.6
43.0
46.8
41.7
34.1
20!2
25.1
27:8
36.0
Com-
posite
Price of
Steel •
(cents
per
pound)
3.115
3.737
2.437
2.124
2.697
2.505
2.334
2.315
2.202
2.165
2.209
2.048
1.957
1.901
1.879
2.033
2.058
2.077
2.464
2.394
Sources:
■ Appendix II, T ble A-2, column (2).
» Appendix II, T ible A-2, column (5).
' Appendix II, 1 able A-2, column (7).
* Appendix VI, Table A-6, column (6).
• These are the link relatives of the figures in the previous column.
« Appendix V, Table A-5, column (5). The figure for 1937 which was actually used in the computations
was $3,959 billion. For this and the 1938 figure see Appendix V, p. 34.
' Appendix IV, Table A-i, column (8).
8 Iron Age, January 5, 1939, pp. 198-199.
CONCENTRATION OF ECONOMIC POWER 13929
Appendix II. Estimation of Domestic Bookings and Domestic Shipments
OF Steel for Steel Industry as a Whole
The details of the computations used in estimating domestic bookings and
shipments of rolled and finished steel from those of the U. S. Steel Corporation
are shown below in Table A-2. See also Appendix VIII, pp. 50-51.
Table A-2.^ — Estimation of bookings and shipments of steel for steel industry as a
whole, 1919-1988
1919
1920
1921
1922
1923
1924
1925
1926.
1927.
1928.
1929.
1930.
1931.
1932.
1933.
1934.
1935.
1936.
1937.
Production of Steel Ingots and Steel
for Castings (thousands of grcss tons)
United
States Steel
Corpora-
tion'
Steel in-
dustry as a
whole '
Ratio of
(2) to (1) 3
.0157
.1855
.8040
.2138
.2107
.3019
.4019
.3782
.4307
.5637
,5805
.4332
.5733
.7755
Domestic
Bookings of
United
States Steel
Corpora-
tion < (thou
sands of
{rross and
net tons)
Estimated
Domestic
Bookings:
Steel In-
dustry '
(thousands
of gross and
net tons)
25, 233
30, 212
7,609
30,391
25, 439
26, 214
30,557
29,138
28,488
33,761
39, 167
26, 977
17, 133
9,129
15,027
18, 777
22,751
30', 212
19,413
Shipments
of Rolled
and Fin-
ished Steel
Products,
Domestic,
by the
United
States Steel
Corpora-
tion » (thou
sands of
gross tons
and net
tons)
(6)
10,311
12, 453
6,832
10,708
13, 196
10, 493
12, 340
12,971-
11,860
12, 701
14:027
10,801
7,162
3,742
5,406
5.392
6,873
10, 182
11,588
6,025
Estimated
Domestic
Shipments:
Steel In-
dustry '
(thousands
of gross and
net tons)
20,783
27, 217
12,325
23,705
29, 173
24,154
29,639
30,847
■2S, 827
32,560
36, 197
26, 280
18,431
10, 385
15,607
16, 222
21,050
28,766
31,620
18, 176
Sources:
> Records of the United States Steel Corporation.
» American Iron and Steel Institute, Annual Statistical Report for 19S8, p. 15.
5 Column (2) divided by column (1).
' Records of the United States Steel Corporation.
» Column (4) multiplied by column (3).
• Records of the United States Steel Corporation. ,
' Columa (6) multiplied by column (3).
13930 CONCENTRATION OP ECONOMIC POWER
Appendix III. Revision of the U. S. Department of Commerce Estimates
OF National Income Payments
The estimates of the United States Department of Commerce for "national
income payments" — which are probably the best estimates of "consumers' cash
income" ' are based in large part upon another of their national income estimates
called "national income paid out".- In June, 1939, the estimates for the latter
were revised by the Department of Commerce,' but corresponding revisions were
not made in the income payment series. Thus it seemed desirable to revise the
figures for income payments in view of the basic revisions in the "income paid
out" series.
'I'he details of the revision are shown above in Table A-3.
The revised estimates of "national income paid out" are shown in column (1),
and the unrevised estimates in column (2~i. The unrevised estimates of "income
payments" are shown in column (3). The unremsed estimates of "income pay-
ments" are based on the unrevised estimates of "income paid out".'' The difference
between the two (shown in column 4) is equal to:
"Direct relief payments,"
plus "Benefit payments under Social Security Act,"
plus "Annuities and refunds to Federal employees,"
plus "Veterans compensation" (Soldiers' Bonus),
less "Employer contributions under the Social Security Act,'"
less "Employee contributions under the Social Security Act,"
less "Employee contributions under the Railroad Retii'ement Act,"
less "Contributions to Federal Retirement." '
For 1938 the sum of these figures was equal to -$17 million. « This figure
(-$17 million) was added to the 1938 revised figure 'for "income paid out"
(column 1) to give the revised 1938 figure for income payments (column 5).
For the years 1929 through 1937 the differences between "revised income pay-
ments" and "revised income paid out" were assumed to be the same as between
the corresponding ..unrevised figures. These differences are shown in column (4);
by adding these figures to those of "revised income paid out" (column 1) the
reoised estimates of income payments were obtained (column 5).
' Pee Appendix VIII, pp. 49-50.
s Robert R. Nathan, "Income in tlie United States, 1929-19,37." a bulletin of the United States Depart-
ment of Commerce, November, 1938.
3 Robert R. Nathan, "National Income in 1938 at 04 Billion Dollars," Stirvey of Current Business, June,
!939. After this report had been drafted, revisions in the income payment series were published by the
United States Department of Commerce, Survey of Current Business, October, 1939.
' See tbe sources referred to in footnotes a and c of Table A-3.
s See Cone, "Revised Estimates of Monthly Income Payments in the United States, 1929-1938", Sunev
of Current Business, October, 1938, p. 15.
« This figure was obtained in the following manner:
In the source referred to in footnote 3 "employer contributions under the Social Security Act" were piven
as $1,119 million for 1938. "Employee contributions under the Social Security Act" for 1938 were assumed
to be the same proportion of "Employer contributions" as in 1937. "Contributions to Federal Retirement"
were assumed to bo .$80 million; they had been $78 million in 1936, and $80 million in 1937 (See Cone, ibid.).
Similarly "Employee contributions to Railroad Retirement Fund" were assumed to be $fiO million; they
had been $81 million in 1937 (See Cone, ibid.). "Annuities and refunds to Federal employees" was assumed
to be $60 million, or approximately equal to the 1930 and 1937 figures of $58 million and $61 million respectively
(See Cone, ibid.).
Bv adding monthly figures in the Surveij of Current Busine.'is, Mm., 1939, p. 19, "Direct relief payments
were estimated at $1,06.' million, "Benefit payments under Social Security Act" at $,503 million and "Vet-
erans' compensation" zero.
CONCENTRATION OF ECONOMIC POWER
13931
Table A-3. — Revision of United States Department of Commerce Figures for
National Income Payments for Their June 1939 Revisions of National Income
Paid Out, 1929-1938
[Millions of dollars]
National Income Paid Out
National
Income
Payments,
Unrevised «
(3)
Unrevised
National
Income Pay-
ments Less
Unreviiied
National
Income
Paid Out «
(4)
Year
Revised >
(1)
Unrevised »
(2)
Revised
National
Income
Payments »
(5)
80,243
74,414
62,763
49,296
45, 565
52,057
55,814
64,207
70,694
65,021
78, 556
73,290
62,032
49,024
45,317
51, 510
55, 137
62,586
69,330
78,574
73,350
63,117
49,597
45,921
52,223
56,086
64.365
68,971
M,196
18
60
1,085
573
604
713
949
1,779
-359
-17
1930 -
63,848
1932
49, 869
1933 —
1934
52, 770
1935
56,763
1936
65,986
1937
70,335
1938
65,004
• Robert R. Nathan, "National Income In 1938 at 64 Billion Dollars," Survey of Current Business, June,
1939, p. 12.
' Robert R. Nathan, "Income in the United States, 1929-1937," a bulletin of the United States Depart-
ment of Commerce, November, 1938, p. 22.
' The figures for 1929 to 1935 are the sums of monthly figtires shown in the source referred to in footnote *.
The figures for 1936 and 1937 are from Fref^erick M. Cone, "Revised Estimates of Monthly Income Payments
in the United States, 1929 to 1938," Sur. iy of Current Business, October, 1938, p. 15. The figure for 1938 is
the sum of monthly figures in the Survey gf Current Business, March, 1939, p. 19.
* Column (3) minus column (2).
» Column (1) plus column (4).
12441)1-^-4 1—^)1. 20-
13932 CONCENTRATION OF ECONOMIC POWER
Appendix IV. Computation of Supernumerary Income
Details of the calculation of supernumerary income are shown below in Table
A-4. See also Appendix VIII, pp. 50-51.
Table A-4. — Calculation of Supernumerary Income, 1919-1938
Year
Kuznets'
come
Pay-
ments 10
Indi-
viduals I
(billions
of current
dollars)
(1)
United
States
Depart-
ment of
Com-
merce
Income
Pay-
ments '
(billions
of current
dollars)
(2)
Consum-
ers' In-
come '
(billions
of current
dollars)
(3)
National
Indus-
trial Con-
ference
Board,
Index of
the Cost
of Liv-
ing*
(Mar..
1935=
100.)
Mini-
mum
Cost of
Living
Per Per-
son * (cur-
rent dol-
lars)
(5)
Bureau
of the
Census
Mid-Year
Poptila-
tion Es-
timate
for the
United
States «
(thou-
sands of
persons)
(6)
Total
Income
Required
for Mini-
mum
Cost of
Living '
(billions
of dollars)
(7)
Super-
numer-
ary In-
come •
(billions
of dollars)
(8)
1919
$57. 499
67.056
55. 177
58.041
65. 854
66. 763
69. 921
72.823
73. 381
75.823
79.808
$57. 826
67. 437
55. 490
58. 371
66. 228
67. 142
70. 318
73. 237
73. 798
70. 254
80.261
74. 474
63. 848
49. 869
46. 169
52. 770
56. 763
65.986
70. 335
65.004
124.8
144.1
124.8
118.8
122.0
123.5
126.5
127.2
124.4
122.7
122.1
117.9
106.3
95.0
91.3
96.8
100.7
103.4
107.9
105.4
$281.8
325.4
281.8
275^5
278.9
285.7
287.2
280.9
277.1
275.7
266.2
240.0
214.5
206.2
218.6
227.4
233.5
243.7
238.0
105, 003
106, 543
108, 208
109, 873
111,537
113,202
114,867
116,532
118, 197
119,862
121,.'i26
123, 091
124,113
124, 974
125, 770
126, 626
127, 521
128,429
129,257
130. 215
$29.59
34.67
30.49
29.48
30.73
31.57
32.82
33.47
33.20
33.21
33.50
32.77
29.79
26.81
25.93
27.68
29.00
29.99
31.50
30.99
$28 2
1920
ZJ 8
1921
25.0
1922
28 9
1923
35.5
1924
35 6
1925
37.5
1926
39.8
1927
40.6
1928
43.0
1929
$80,261
74. 474
63.848
49. 869
46. 169
52. 770
56.763
65. 986
70. 335
65.004
46.8
1930
41.7
1931
34.1
1932
23.1
1933
20.2
1934
25.1
1935
27.8
1936
36. .1
1937
38. o
1938 » . —
34.0
' Simon Kuznets, National Income and Capital Formation, I919-I9S5, National Bureau of Economic
research, 1937, p. 24, row 10.
' These are the revisions of the United States Department of Commerce series for "monthly income
payments." See Appendix III, Table A-3, column 5.
» The figures for 1919 to 1928 are column (1) times 1.00568; the figures for 1929 to 1938 are the same as those
of column (2). The figures for 1919 to 1928 are -the result of linking the fipurer, in column (I) to those of
column (2) at 1929. That is, the figures in column (1) have been multiplied by the ratio of the 1929 figure of
column (2) to the 1929 figure of column (1). In so doing it was assumed that:
(1) The 1929 figure of column (2) was the correct figure for consumers' cash income.
(2) That Kuznets' figures for 1919 to 1928 were "in error" in the ratio of his 1929 figure to that of column (2).
• For the years 1919 to 1937 see Survey of Current Business, 19SS Supplement, p. 11. The 1938 figure in the
average of monthly figures for 1938 reported in Survey of Current Business, Mar., 1939, p. 20. The base of
these figures has been shifted from 1923 = 100 to Mar., 1935=100.
» Obtained by multiplying column (4) by $2.2582 (or by dividing column (4) by 100 and multiplying by
$225.82)> Essentially what has been done is to assume that the necessity or minimum cost of living per
person in March, 1935, was $225.82 (Appendix VIII, p. 61) and that it varied as did the National Industrial
Conference Board index of the cost of living.
« United States Department of Commerce, Bureau of the Census. Statistical Abstract of Vie United States,
19S8, p. 10.
' Column (5) multiplied by column (6).
' Column (7) subtracted from column (3). The resulting figure is the -estimate of supernumerary income,
or the amount of cash income available for disposal in the luxury goods market.
Appendix V. Estimation of Indiustrial Profits
For the years 1919 to 1937 the basic sources of data used in computing indus-
trial profits were the Statistics of Income reports of the United ^tates Bureau of
Internal Revenue. The formula used in computing industrial profits is given in
Appendix VIII. Table A-5 shows the details of the computations.
CONCENTRATION OF ECONOMIC POWER 13933
Table A-5. — Industrial Profits Estimated From the Statistics of Income, 1919-19S7*
[Thousands of dollars]
Year
Statutory
Net Income '
(1)
Total
Federal Tax >
(2)
Tax Exempt
Interest '
(3)
Dividends
Received
From
Domestic
CorpOTations*
(4)
Industrial
Profits"
(5)
1919
$8,415,872
5,873,531
457, 829
4,770,035
6,307,974
6,362,726
7,621,056
7,504,693
6, 610, 146
8,226,617
8,739,758
1,551,218
-3,287,646
-5, 643, 574
-2,647,367
94,1710
1,696,949
7, 326, 217
7.354,003
$2, 175, 342
1, 625, 235
701, 576
783, 776
937, 106
881,550
1, 170, 331
1,229,797
1, 130, 674
1, 184, 142
1,193,436
71i;704
398,994
285,576
423,068
696,048
736,125
1, 191, 378
1, 276, 184
$178,648
219,977
188, 789
394.042
496, 202
617,209
619,846
499:592
600,826
623,458
636,697
526,261
541,713
6H260
691,586
658,701
713,646
444,669
476,302
$8, 419, 078
1920
4,467,973
1921
—54,968
1922
4,380,301
1923
6, 867, 070
1924
4, 998, 385
1925
6, 970, 571
1926
6, 774, 488
1927
6,880,297
7 665 933
8,083,019
1930
1, 365, 776
-3,144,826
1932..
-5,374,900
1933
-2,378,849
1934 ..
156,828
1935.. -
1, 674, 370
1936
$2, 676, 598
2,682,227
3,902,910
1937
3,871,894
1 statistics of Income for 19S6, Pt. II, p. 47, from colomn headed "Net income less deficit."
' Ibid., column headed "Total Tax".
' Obtained from Mr. Edward White, Chief of the Statistical Section of the United States Bureau of
Internal Revenue, in a letter of July 14, 1939.
< Statisiict of Income for 19S6, Pt. II, p. 24, from column headed "Aggregate".
• The sum of column (1) plus column (3) minus the sum of column (2) plus column (4).
♦ 1937 figures were taken from a preliminary release (Pres^ervlce, No. 18-66) of the TTnited States Treas-
ury Department, August 23, 1939.
At the time the statistical computations for this paper were made, the aboVe
figures for 1937 were not available, so that the 1937 figure for profits, as well as
that for 1938 had to be estimated from other less accurate sources. The profits
figures reported periodically by the National City Bank of New York in its
monthly economic buUetin were used for this purpose.' These profit figures
cover reports of about 2,000 corporations, and show profits after depreciation,
interest, taxes, and other charges, but before dividends.
The general procedure used is as follows:
A. Estimating Profits for 1937:
(1) In its April 1, 1938, bulletin the National City Bank reported profit
figures for 1936 and 1937 for 2,300 corporations grouped by types
of business. These groups were then re-grouped into business
classes comparable to those used in the Statistics of Income for 1936.
(2) The ratio of the profit figure for 1936 for any group computed from
the Statistics of Income to that reported by the National City Bank
was then computed.
(3) The 1937 profit figure for that group as reported by the National
City Bank was then multiplied by this ratio, giving est^imated
profits for 1937 for this group.
(4) The sum of all such estimated group profits was the estimated figure
for industrial profits for 1937.
It is interesting to note that the resulting estimate of$3.96 billion is only about
2 percent larger, than the profit figure reported by the United States Bureau of
Internal Revenue for 1937. (See Table A-5.)
B. Estimating Profits for 1938:
(1) In its April 1, 1939, bulletin the National City Bank reported profits
for 1937 and 1938 for a group of over 2,400 corporations. These
corporations were grouped into classes corresponding to those
used in estii^ating profits for 1937.
' See the economic bulletins of the National City Bank- of New York for April 1, 1938. and April 1, 1939
139S4 CONCENTRATION OF ECONOMIC POWER
(2) Since the report for April 1, 1939, did not cover exactly the same cor-
porations as that for April 1, 1938, the group totals for 1937 of the 1939
report differed from those of the 1938 report. Thus for each group
the ratio of the 1937 profit figure shown in the 1938 report to that in
the 1939 report was computed. The 1938 figure for this group was
then multiplied by this ratio.
(3) The resulting group figure was then multiplied by the ratio computed
in step (2) of (A), giving estimated 1938 profits for that group.
(4) The sum of these group profit figures was the estimate of industrial
profits for 1938.
The estimated profits figures for 1937 and 1938 were (in billions) :
1937. -_ $3,959
1938 $2. 165
These were the figures actually used in the computations.
Appendix "VI. Computation of Industrial Production Index
It was decided that the index of industrial output most suitable for the purposes
of this paper was the Federal Reserve Board's Index of Manufacturing Production
with the iron and steel production subgroup removed.'
The Federal Reserve Board index is an aggregative type index with fixed weights.
The base of the index is 1923-25=100.
The procedure for removing the iron and steel subgroup was as follows:
Let gi, 52, • . ■, Q\y ■ ■ ■, ?N, represent the quantity of output of the various
products included in the index; q\ is the output of iron and steel;
and Wi, Wi, . . ., Wi . . ., w>n are the corresponding fixed weights.
Thus the value of the index of manufacturing output in any year, D, is:
N
(1) Md=-n — XlOO
where g'ji' is the output of the item i in the year D, and Yz {q{^^^^ -\- q-,^^'^* -\- q-^'>'^) is
the average output of the item i for the three base period years, 1923 to 1925.
Similarly, the value of the output index for the iron and steel subgroup for any
year D is
/o\ T gi""^i vinn
Thus the value of the index of manufacturing output with the iron and steel
subgroup removed for any year D is:
101
(3) Md' =
of '^qPwi j-giDu)iX 100
r>^X)wi(9i*"'2' + gi"" + 9i"") )-% «^i(gi"" + gi»24 4-gii«5)
Formula (3) may be rewritten:
N
MDX)/3Xl"'i(g,"" + gil«4 + gim5) _/pX/» «^l((7l"2Hgi"" + gi"")
(4) M^'= '-' ^
( HX)"'i('/"" + ?'"" + 9'"'") )-^ u;i(gi"w+gi"» + gi'»")
' For a description of the Federal Reserve Board's Indexes of production see the mimeographed release of
the Division of Research and Statistics, Federal Reserve Board, "Federal Reserve Index of Industrial
Production." reprinted from the Fderal Reserve Bulletin for February and March, 1927, with notes on sub-
sequent revisions. (Release dated Nov., 1937.)
CONCENTRATION OF ECONOMIC POWER
13935
The value of Ys'^Wi (3ii»2' + gi>92< + gii»") = 60,639,571 and the value of }i Wi
(?i"" + 3i"2^ + qi"") = 13,937,195.2
(5)
Thus formula (4) may be written:
A/d X (0.060640) -/pX (0.013937)
0.046703
Md' = -
u nn.c^no- w 60,639,571-13,937,195
where 0.046703 is equal to 1,000,000,600'
The details of the computation are shown below in Table A-6.
Table A-6. — Computation of Federal Reserve Board Index of Manufacturing
Production Excluding Iron and Steel, 1919-1938
[1923-1925 = 100]
Year
M
(1)
Ms
0.060640
(2)
I
(3)
Ix
0.013937
(4)
(2) -(4)
(5)
M
(5)X
0.046703
(6)
1919....
84
87
67
86
101
94
105
108
106
llS
119
95
80
63
75
78
90
105
109
84
5.094
5.276
4.063
6.215
6.125
5.700
6.367
6.549
6.428
6.792
7.216
5.761
4.851
3.820
4.548
4.730
5.458
6.367
6.610
5.094
82
99
46
82
105
106
113
104
119
130
94
60
31
53
60
79
HO
118
66
1.143
1.380
0.641
1.143
1.464
1.241
1.478
1.575
1.450
1.659
1.812
1.310
0.836
0.432
0.739
0.836
1.101
1.533
1.645
0.920
3.951
3.896
3.422
4.072
4.661
4.459
4.974
5.978
5.133
5.404
4.451
4.015
3.388
3.809
3.894
4.357
4.834
4.965
4.174
85
1920
83
1921
73
1922
87
1923
100
1924
95
1925
105
1926
107
1927
107
1928 .
110
1929
116
1930
95
1931
86
1932
73
1933
82
1934
83
1935
93
1936 .
104
1937 ..
106
1938 .. .
89
Sources:
Col. (1): Federal Reserve Index of Production: Manufactures. See footnote 2 in text. ■
Col. (2): Column (1) multiplied by 0.060640.
Col. (3): The subgroup index for iron and steel production. See footnote 2 in text.
Col. (4): Column (3) multiplied by 0.013937.
Col. (5): Column (2) minus column (4).
Col. (6): Index of Production: Manufactures excluding iron and steel.
Appendix VII. — Demand Relation Equations
Let Xp denote the quantity of production of steel ingots and castings (in thou-
sands of gross tons),
Zb, the quantity of steel bookings (in thousands of gross and net tons),
Xa, the quantity of steel shipments (in thousands of gross and net tons),
p, the Iron Age composite price of. finished steel (in cents per pound),
/, the index of industrial production excluding iron and steel (1923-1925 =
100),
/„ the link relatives of /,
S, supernumerary income (in biUions of dollars),
P, industrial profits (in billions of dollars),
and t, Time (in years measured from an origin depending on the period studied).
» These figures were obtained from Mr. F. A. Ooldenweiser, Director of Research and Statistics, of the
Division of Research and Statistics of the Board of Governors of the Federal Reserve System in a letter dated
June 29, 1939.
13936 CONCENTRATION OF ECONOMIC POWER
Then the four relations stated on page 24 may be stated mathematically in the
form:
Relation I: Xp—ap + b^p + c^I+dJ
Relation II: 2;p=ep+/p7) + gp/-|-/ip/r
Relation III: x,=a,+b,p + c^S + d,P+e^t
Relation IV: Xb = ab + bbP + CbS + dbP+ebt
A fifth relation also was studied.
Relation V: Xp=Ap^I°10'>*'
The various constants, Cp, bp, etc., were determined by the method of least
squares. The resulting statistical demand relations are shown below in Table A-7.
Table A-7. — Equations of Demand Relations
Period Studied
Relation
Number
Equation of Demand Relation
1919-1938' .___
1919-1938'. .._.
•1922-1938'
•1920-1938..
1919-1938*
•1922-1938*
1919-1938*
•1922-1938*
1922-1938*
I
I
I
n
m
in
IV
IV
V
ip= -63,800+7,478 p+893 7-91.1 t
ip 65,200+8.201 p+890 /
Xo 57,081+2.217 p+944 7-127 t
ip=»-71,700+8,605 p+857 7+87 I,
.T.=3,014+388 p+1,005 P+510 .?+144 t
!.= 11, 760-2,419 p+1,157 P+426 S+59 t
Jb =8,680+298 p+1,563 P+327 -S+270 t
Zb=33,480-10,254 p+1,863 P+258 S+143 t
0.158 p""' 72Mr
100.00201
1 Origin of time variable is January 1, 1929.
• Time variable has been excluded in the equation.
• Origin of time variable is January 1, 1929.
• Origin of time variable is July 1, 1930.
•Equations used in text Tables 1 and 2.
Given these equations, it is a simple task to measure the elasticity of demand,
or the elasticity of the quantity of steel sold with respect to thfe price of steel.
Let Xc be the quantity of steel sold as computed from one of the above equations.
dxc p
Then the elasticity of demand is equal to . . Thus the elasticity of
dp Xo
demand formula for the first of the above equations is (where e denotes elasticity
of demand) :
7,478 p
(a) e=
-63,800 + 7,478 p + 893 7-91.1 t
The elasticity formulae for the other equations are similarly defined, except for
Relation V, which is directly
e = 0.235
It is obvious therefore that, with the exception of Relation V, the elasticity is
not constant, but varies with the factors influencing demand.
In Table 2, the elasticities were computed by substituting the average values of
the demand factors in equations such as that above.
Appendix VIII. Definition and Measurement of the Economic Variables
Used
(1) The Quanliljj of Sted Sold. — The "quantity of steel feold" has been defined
on a previous page ' as the quantity of steel sold by steel producers. There are
two major problems in the measurement of the quantity of steel sold:
(a) No reliable data are available showing for the country as a whole the
physical quantity of steel sold, either as a gross figure or by separate types of
steel.
(b) As was emphasized in section III-A, even if sales figures for the separate
types of steel were available, there would still remain the problem of com-
bining them into an economically logical composite representing the total
> Seenipra, p. 11.
CONCENTRATION OF ECONOMIC POWER 13937
physical volume of sales. It should be obvious that the various types of steel
do not have the same economic importance (in a demand analysis) per pound.
Furthermore, the demand conditions for the different types of steel need not
(and, in general will not) be the same. Moreover, the types of steel tend to
change in character from year to year. The problem of finding some common
unit by which different items could be aggregated is one for which a thoroughly
satisfactory solution has never been reached.'' In the absence of an answer
to the problem, the only recourse is to adopt the usual aggregating proce-
dures used in making index numbers. It seems doubtful that this will result
in damagingly spurious information.
Six sets of data have been considered here in estimating the quantity of steel
sold.
(i) Production of steel ingots and castings in the United States as reported by
the American Iron and Steel Institute.' These figures have the advantage,
as a proxy measure of the quantity of steel sold, of representing almost com-
plete coverage of the production of steel ingots and castings in the United
States. The main disadvantage is that steel ingot production at best can
represent only finished steel production and not finished steel sales. Sales
differ from production by the amount of the net change in inventories of steel
in the hands of steel producers. At times of rapid economic change, the
fluctuations of these inventories are probably substantial enough seriously to
invalidate the use of production figures. Since there are no complete or
reliable figures available showing the net change in steel inventories in the
hands of producers, no adjustments can be made in the production figures.
(ii) Production of hot-rolled iron and steel in the United States. These.,
figures are reported by the American Iron and Steel Institute both as a total
figure in gross tons, and by separate gross-ton totals for about twenty differ-
ent types of hot-rolled iron and steel. ^ The advantages of these figures are
(a) They represent practically complete coverage of hot-rolled iron
and steel production in the United States.
(b) They are more nearly representative of finished steel than ingot
production figures. «
(c) The breakdown into separate types makes it possible to weight
the various types and thus obtain a more logical measure of finished
steel output. However, these figures have the same disadvantage as
steel ingot production figures, namely, that they are production and not
sales figures, and thus will be in error by the amount of net change in
hot-rolled iron and steel inventories in the hands of steel producers.
(iii) Index of Production: Steel Works and Rolling Mills computed by the
National Research Project of the Works Progress Administration.* The
main advantage of the series is that an attempt was made to combine the
different types of steel on the basis of a measure of their economic importance.
The series, however, has two substantial disadvantages:
(a) The production composite was computed for use in a study of
labor productivity in the steel industry. Thus the economic weights
used-^although perhaps satisfactory for a productivity study — have
practically no relation to the economic importance of the different types
of steel from the point of view of steel demand analysis.
2 See. for example, the excellent discussion in J. D. Black and B. D. Mudeett, Research H Agricultural
Index Nvmhers, (Social Science Research Council, Bulletin 10) 1938, and the sources quoted therein.
3 American Iron and Steel Institute, Annual Statistical Report for 19iS, 1939, p. 16. Th'-sc figures are re-
ported in total and by types on a gross ton basis. Five types of steel ingots are included (basic and acid
open-hearth, hcssemer, electric, and crucible). In 1934 and subsequent years the total for steel ingots and
castings included only that portion of the production of steel for castings used by foundries operated by com-
panies producing steel ingots.
Figures comparable to those of the American Iron and Steel Institute were also available for the United
States Steel Corporation.
* American Iron and Steel Institute, Annual Stalistical Report for 193S, pp. 21-23, and Annual Statistical
Report for 19S7, p. 22.
The following classification of hot -rolled iron and steel is made: plates; sheets; strip; black plate; hoops;
cotton ties and baling bands; merchant bars; concrete bars; structural shapes; sheet piling; rails; long splice
bars and tie-plate bars; skelp; wire rods; rolled forging billets; cross ties; blooms, billets, etc., for export;
strip and sheets for cold-reduced black plate and tin-plate (separate classification only for 1938); blanks or
pierced billets for seamless tubes (separate classification beginning in 1926); rolled steel car wheels (separate
classification beginning in 1931) ; and all other. Inasmuch as production of hot-rolled iron amounts (in most
years, 1919 to 1938) to less than 3 or 4 percent of the total, and since its fluctuations correspond closely to those
of hot-rolled steel, the above figures are very close approximations of hot-rolled' steel production. There is
a separate total for hot-roIIcd steel production, but no breakdown into separate types of products.
• In fact, a large proportion of the finished steel sold is hot-rolled steel.
' Works Progress Administration, National Research Project, Production. Employment, and Productijily
in 69 Manufacturing Industries (Report No. S-1), Pt. II, May, 1939, pp. 92-100. A detailed description of
the methods used in constructing the index is given.
13938 CONCENTRATION OF ECONOMIC POWER
(b) They are production and not sales figures,
(iv) Estimates of finished steel shipments by The Iron Age and Steel, two
leading steel trade journals.' In each case the estimates are based on reports
from steel producers. The estimates purport to show the total volume of
shipments of finished steel by separate types of steel to major consuming
industries. Although these estimates have the obvious advantages:
(a) The breakdown of the total into subtotals by separate types of
steel and by consuming industries makes possible the computation of a
logical economic composite measure,
(b) They are estimates of sales of finished steel and not of production
of steel,
their unreliability was too great to warrant using them. The coverage of
the series was low until recently, and varied considerably from year to
year. Moreover, the classification of types of steel and consuming indus-
tries was not the same from year to year so that it was difficult to make use
of the sub-group totals. :
(v) United States Steel Corporation subsidiaries' domestic shipments of rolled
and finished steel products. Inasmuch as steel production and sales of the
United States Steel Corporation subsidiaries have since 1919 comprised one-
third to.one-half of the total for the country as a whole it was thought possible
to estimate the nation's sales from those of the Corporation subsidiaries.
The series have the obvious disadvantage that they represent only sales of
the United States Steel Corporation. Although the types of steel sold by
the Corporation do not represent exactly the composite type for the industry
as a whole, and though the conditions of demand for the United States Steel
Corporation subsidiaries are not identical to those for the industry it was
decided that these disadvantages were not great enough to preclude use of
the figures. The series does have several important advantages.
(a) Its coverage is reasonably well defined in relation to the industry
as a whole.
(b) Only domestic shipment^ are included. Thus no adjustment for
exports is needed.
(c) The figures represent sales and not production.
(vi) United States Steel Corporation subsidiaries' domestic customers' book-
ings of rolled and finished steel products. These data consist of aU domestic
contracts for tonnage. They have the same disadvantages as the above ship-
ment figures, but they do have one important advantage over the shipm.ent
figures. Since there is some lag between the time an order (booking) is placed
and the time shipment is made,8 the shipment figures do not coincide in time
with the demand conditions under which the order is placed and the sale
made.
It was finally decided that three of the above six sets of data would not
prove useful in estimating the quantity of steel sold:
(a) The Iron Age and Steel estimates of finished steel shipments, be-
cause they were considered too unreliable.
(b) The W. P. A. National Research Project index of steel works out-
put because it was a production index, and because its construction was
not deemed suitable for a demand study.
(c) Production of hot-rolled iron and steel, because it followed steel
iugot production so closely ' that it possessed no great advantage over
the latter.""
Three separate estimates of the quantity of steel sold were used for further
experimentation:
(a) The American Iron and Steel Institute figures on production of
steel ingots and castings were used directly as representing sales. These
figures were considered as the least satisfactory of the three estimates.
However, it was decided to use them because of thfeir greater familiarity.
' In connection with this section see the detailed discussion in the memorandum of M. W. Worthing,
"Distribution of Steel Products to Major C insuming Industries," United States Steel Corporation, October
30. 1939.
• This lag varies greatly with the rate of operations and the order backlog of steel producers. For example
at the peak of the boom in 1937 shipments on some products were delayed for as much as six to seven months
after the placing of the order
• The year to year fluctuations in steel ingot production were almost identical to those of hot-rolled iron
and steel production except for a long-run smooth trend increase in the ratio of the former to the latter.
See chart A.
'• Preliminary experimentation Indicated that no improvement would be madp by computing a composite
measure in which the dIfTerent types of hot-rolled iron and steel were given weights corresponding to their
respective prices.-
CONCENTRATION OF ECONOMIC POWER 13939
(b) A second estimate was based on the United States Steel Corpora-
tion subsidiaries' figures for total shipments of rolled and finished steel.
The estimate was made in the following simple manner. It was assumed
that for any year the ratio of the Corporation subsidiaries' shipments
to those of the industry as a whole was the same as the ratio of the
Corporation subsidiaries' production of steel ingots and castings to that
for the industry. Estimated total shipments were thus obtained by
dividing the Corporation subsidiaries' total shipments by the latter of
the above two ratios.
This procedure was justified on these grounds:
(i) There is very little time lag between ingot production and actual
shipment.
(ii) It seemed reasonable to believe that the Corporation subsidiaries
weight losses (conversion losses) involved in converting steel ingots into
finished steel were not substantially different from those for the industry
Chart A
80
72
64
56
i ""
g 40
i 32
o
. 24
S
8
\
STEEL PRODUCTION
IN UNITED STATES
72
64
56
48 2
40 ?
32 i
24 I
16 1
i
8
/
X'
V
Gor
»ST
AN
•)
^
y
\
/
'^
\
NGS
1
/
\
1
/
s^
/I
x'
\
\
/
^
/
\\
/
/
\
HOT
f]
«3N
ROL
>IISH
AND
LED
-D
STCE
\\
y
r
/
'
Y
/
1
L
\
\
/
/
/
\
/
5 1 i 1 i i s 1 i 1 i 1 1 i i i §
\ \
\ 1 \
1 1 \
as a whole, or, at least, that the ratio of the Corporation subsidiaries'
weight losses to those of the industry did not change sharply from year
to year.
(iii) The ratio of the Corporation subsidiaries'" steel ingot production
to that of the industry was almost constant except for a smooth long run
decline,
(c) A third estimate was based on the United States Steel Corporation
subsidiaries' figures for domestic bookings of rolled and finished steel.
The method of adjusting this series "was the same -as that used for the Cor-
poration subsidiaries' shipments.
The above three estimates are shown on Chart 1, page 21, and in Appendix
I, Table A-1. The details of the estimations of bookings and shipments
are shown in Appendix II.
(2) The Price of Steel. — The difficulties of defining and measuring the price of
steel were indicated in sections III- A, and III-F:
(a) There are many different kinds of finished steel sold, and ^each kind
has its own price. Thus there is a problem similar to that encountered in
defining and measuring the quantity of steel sold. How should the different
types be combined in order to obtain an economically logical measure of the
13940 CONCENTRATION OF ECONOMIC POWER
price of steel? No absolutely conclusive answer to this question has ever
been given." However, since the prices of the different types of steel move
more or less closely together a fairly satisfactory solution can be obtained
by one of the well known averaging methods used in constructing price
index numbers.
(b) Prices tend to be different in different geographical regions. The
problem here is essentially the same as the first.
(c) Because of price concessions, published prices — which are the only
source of price data — do not always reflect the actual prices paid by steel
buyers.
There are three well known series computed for the composite price of
finished steel :
(i) That computed by Iron Age.
(ii) That computed by Steel.
(iii) That computed by the American Metal Market.
(i) The Iron Age Composite Price of Finished Steel is published
weekly in Iron Age.^^ The Iron Age composite is a simple un-
weighted arithmetic average of the following items: bars, plates,
shapes, hot-rolled strip, plain wire, heavy rails, black pipe, and
No. 10 gage hot- rolled sheets. '^ The prices used in each case are
the Iron Age weekly market quotations of base prices at Pitts-
burgh." The quotations are based on the Iron Age estimate as
to what is the "open market price." '^
(ii) Steel's composite price of finished steel is based on weekly
steel price quotations as reported by Steel. The composite is a
simple unweighted arithmetic average of the prices at Pittsburgh
of the following items: plates, shapes, bars, wide hot- rolled strip,
wire nails, plain wire, tin plate, black pipe. No. 24 gage hot-rolled
sheets, and No. 24 gage galvanized sheets." The weekly quotations
on which the composite is based represent the best judgment of the
editors of Steel as to the going market prices. '^
(iii) The American Metal Market composite price of finished steel
is based on daily price quotations published in the American Metal
Market.^^ The composite is a weighted arithmetic average of the
following finished steel products: bars, plates, shapes, pipe, wire
nails. No. 24 gage sheets, strips, and tin-plate.'^
The three composites are so similar in their movements as to make a
choice between them unimportant. The Iron Age composite price was
>i J. D. Black and B. D. Mndpett.-op. :il. See footnote 2 of this appendix.
12 Annual and monthly averages for the yeans 1903 to 1938 are given in the /ron ^fff, Jan. 5, 1939, pp 198-199.
13 Hot-rolled strip was not included prior to 1920. Before 1920 No. 24 gage hot-rolled annealed sheets is
used instead of No. 10 gage hot-rolled sheets. (Information from letter cited in footnote 15.)
1* Although steel prices vary from one geographical area to another (see section III-F.), the year to year
fluctuations in all areas tend to he more or less the same. Thus it is doubtful if there is any serious unrepre-
sentativeness in a composite based on Pittsburgh prices.
i» Mr. C. E. Wright, Managing Editor of Iron Age, in a letter dated April 5, 1939, writes: "We do not
change our base prices unless we feel tliat such a change has become more or less general. For example, when
a break in sheet prices occurred in October, 1938, we did not reduce our base price on the first news that a
concession had been made to one or two companies, but the following week, however, this concession had
become general. That is, all companies were making it to all customers, and therefore it became an open
market price."
1' This description of the Steel composite price was obtained from Mr. O. H. Manlove, Associate Editor
of Steel in a letter dated April 12, 1939. Prior to July, 1938, the price of No. 28 gage rather than of No. 24
gage galvanized sheets was used in the composite. After April, 1938, the editors of Steel added an extra
of 90 cents to the quotation for No. 24 hot-rolled sheets. Annual and monthly figures for the composite for
the years 1927 to 1938 are given in Steel, Jan. 2, 1939, p. 251.
" Mr. Manlove writes: . . "the quoted prices are the best judgment of the editors [of Steel], although in
the present market procedure with prices announced by quarters, there is comparatively little variation
from the published figures." . , , ,
1' Annual figures from 1899 to 1937, and monthly figures for the years 1914 to 1937 are shown In Metal
Statistics for 19S8, published by the American Metal Market Company, 1939, pp. 85-88. Recent figures
can be obtained from current issues of the American Metal Market, metal trades dally paper.
'» The weights are as follows:
Aars - - ■- - - 2.0
Plates - - 1-5
Shapes - - 10
Pipe(l-3) - - - - - 1-6
wire nails - — - }-0
Sheets 1-6
Strips. - - JO
Tin plate- - _0-j
Total - 10.0
These weights are rough estimates of the "relative Importance" of the different Items. Prior to Mar.
29, 1033, another system of weights was used. See Metal Statistics for 19S8, p. 85, and for 1932, p. 21.
CONCENTRATION OF ECONOMIC POWER 13941
selected for use in the statistical analysis because it was the most widely
known of the three.
(3) Consumers' Cash Income.^" — The measure of consumer income which
seemed most appropriate for the purpose of this paper, was one measuring the
actual amount of cash income received by consumers and disposable by them for
consumers' goods and services. The two basic series which most closely approxi-
mate this ideal measure are:
(a) The United States Department of Commerce series for "national
income payments." ^^
(b) The National Bureau of Economic Research series for "aggrecate
income payments to individuals" (basic variant) .^^
Neither of these series is completely appropriate; both include payment items
not actually received by consumers and exclude certain other items that are
received by consumers.*' However, both series are reasonably good estimates of
cash payments to consumers.
Inasmuch as the United States Department of Commerce series does not go
back beyond 1929, the National Bureau of Economic Research estimates were
used for the years 1919 to 1928. The latter series was linked to the former
at 1929."
(4) The Cost of Living.^^ — The majority of consumers' products-made-from-
steel — of which the passenger automobile is most important— are luxury goods
which can be thought of as being purchased with that part of consumers' cash
income remaining after the necessary or subsistence costs of living have been
paid. It is obviously unrealistic — as well as difficult — to draw a hard and fast
line at what can be called subsistence living costs, and to assume that only after
such costs are met can the consumer begin buying luxury goods. But on the
average some such relationship exists.
The subsistence costs of living per person were taken as $225.82 for the month
of March, 1935. This is equivalent to the $903.27 which the Works Progress
Administration set up as the minimum costs of living for a family of four in
March, 1935.2^ For other periods the subsistence costs of living were assumed
to vary from this level as did the National Industrial Conference Board index
of the cost of living vary from its March, 1935, level.*' After computing sub-
sistence costs of living for the country as a whole, subsistence living costs were
subtracted from consumers' cash income, leaving consume? s' supernumerary
income.^^ Supernumerary income, or the amount of cash income disposable for
luxury goods and services, was the income series used in the statistical analysis.
The series is shown in Chart 1. and Appendix I.
(5) Industrial Profits.^^ — The basic sources of data on industrial profits are re-
ports received by the United States Bureau of Internal Revenue from all com-
panies filing Federal corporation income and excess profits tax returns and per-
sonal holding company returns. These reports are compiled and published
annually in the Bureau's Statistics of Income reports.^'^ The profit figures which
were used in this study were based on these Statistics of Income reports,^' and were
computed from the following formula:
Industrial profits equal
Statutory net income **
" In connection with this section see Roos and von Szelislci, op. cit., pp. 39-41.
" Robert R. Nathan, "Income In the United States, 192&-37" (Bulletin of the United States Department
of Commerce, Nov., 1938); "National Income in 1938 at 64 Billion Dollars," Survey of Current Business.
June, 1939, p. 12; Frederick M. Cone, "Revised Estimates of Monthly Income Payments in the United
States, 1929 to 1938", Surrey of Current Business, October, 1938, p. 15.
" Simon Kuznets, National Income and Capital Formation, 1919-1986, National Bureau of Economic
Research, 1937.
2' See the discussions in the sources cited in footnotes 20 to 22.
»< Certain revisions, however, were first made in the Department of Commerce series; see Appendix III.
" See Roos and von Szeliski, op. cit., pp. 41-42.
>' Works Progress Administration, Division of Social Research, "Intercity Differences in Costs of Living
in March, 1935, .59 Cities" (Research Monograph XII).
" A detailed description of this index is presented in The Cost of Livinfi in the United States, 191i-]9S6,
pp. 13-42, published by the National Industrial Conference Board. Monthly and annual figures are
reported in the Survey of Current Business, 19S8 Supplement, p. 11, and current issues.
" The details of the computation of supernumerary income are given in Appendix IV.
" In connection with the discussion of this section see W. L. Crum. "Corporate Earnings on Invested
Capital", Harvard Business Review, v. XVI, No. 3, pp. 336 to 350.
" United States Treasury Department, Bureau of Internal Revenue, Statistics of Income for 1938, 1939,
Part II, and for previous years.
n The Statistics of Income reports represent almost complete coveraee of business profits. The number of
business concerns not filing returns under the various Federal revenue acts do a negligible proportion of the
nation's business.
" Statistics of Income for 19S6, p. 47. Statutory net income represents "net income less deficit." In other
S laces it is called simply "net income"; see, for example, p. 24. For 1936 and 1937 statutory net income
icludea "dividends received from domestic corporations."
13942 CONCENTRATION OF ECONOMIC POWER
less The total federal tax ^J
plus Tax-exempt interest on government obligations "
less Dividends received from domestic corporations (for 1936 and 1937)."
What was desired was a series showing for any year the real (economic) profit
situation, and thus the current real profit outlook of buyers of producers' products-
made-from-steel. The above profit figures are subject to some severe limitations
in this respect:
(a) Certain corporations included in the profit reports can be only of insig-
nificant importance as buyers in the capital goods market. The most
important of these groups is the group of "financial" institutions. Total
profits of this group — and others not relevant to the capital goods market —
however, are not a large enough proportion of the total, and do not vary
enough from the general movement of profits to distort the figures seriously.
(b) A much more serious limitation is that the tax accounting proce-
dures used by business tend to make their profit figures represent the
profit situation and outlook of a period somewhat prior to that for
which the figures are actually reported. The largest part of the receipts of
business comes from sales of goods and services. These receipts are based on
current sales prices more or less accurately reflecting current cost and demand
conditions. However, since production must precede the date of sale, and
since many of the costs of production are incurred at an even earlier date
(purchase of raw materials, equipment, etc.), the cost of production figures
used in profit calculations represent the cost situation of an earlier date. If
costs and prices have in the meantime changed drastically, a substantial part
of the profits or losses reported are what amounts to inventory and capital
profits or losses. Thus the profit figures reported tend to lag behind the cur-
rent cost and demand situation. The profit series used in this study is shown
in Chart 1 and Appendix I.
(6) Volume of Industrial Production. — The most widely known and probably
the best composite measure of the volume of industrial output is the Federal
Reserve Board's index of industrial production.*" However, it has a very serious
limitation for the purposes of this study. What is desired is a measure of the
industrial production of all commodities other than iron and steel (produced by
steel producers). Iron and steel production is the most important single compo-
nent of the Federal Reserve Board index, with a weight which gives it an aggregate
importance of almost 25 percent of the total for the index.^^ Thus relationships
observed between steel sales and this index would be in part spurious.
The detailed method of removing iron and steel production from the index is
described in Appendix VI. " The series with iron and steel production removed,
which is the one used in the statistical analysis, is shown in Chart 1 and A pen-
dix I.
Exhibit No. 1412
AN ANALYSIS OF CHANGES IN THE DEMAND FOR STEEL AND IN
STEEL PRICES, 1936-1939
This is an analysis prepared by the Special Economic Research Section of
United States Steel Corporation, composed of Messrs. Edward T. Dickinson, Jr.,
Ernest M. Doblin, H. Gregg Lewis, Jacob L. Mosak, Mandal R. Segal, Dwight
B. Yntema and Miss Marion W. Worthing. The work of this group was under
the supervision of Theodore O. Yntema, Professor of Statistics, University of
Chicago. This analysis was written by H. Gregg Lewis, who had the benefit of
suggestions from other members of the staff. It is issued by United States Steel
Corporation.
November 1, 1939.
M The total federal tax Includes the normal corporation income tax, war profits and excess profits taxes,
and the surtax on undistributed profits. See Statistics of Income for I9S6. p. 47.
" The amount of tax exempt interest on government obligations which is added to statutory net income is
the amount of wholly tax-exempt interest. See Statistics of Ivcome for 19S6, pp. 6 and 24.
" In 1936 and 1937 "dividends received from domestic corporations" were included in statutory net Income
for excess profits tax computation. In order to avoid double counting of corporation earnings it was there-
fore removed. See footnote 32. The detailed computations of industrial profits are shown In .\ppendix V.
" A full description of the index and values of the index and its sub-eroups back to 1919 may be obtained
from the Division of Research and Statistics of the Federal Reserve Board, Washington, D. C. Current
figures are reported In the Boaras monthly Federal Reseree Bulletin, the Survey of Current Buttness, and
numerous other places.
•' See Appendix VI. •
CONCENTRATION OF ECONOMIC POWER 13943
CONTENTS
r. Introduction
II. Summary of Findings
III. Nature of the Demand for Steel
IV. Short Run Variations in the Demand for Steel: June 1936 to October 1939
A. The Economic Background, 1930 to 1935
(1) Developments Affecting the Potential Demand for Producers' Durable Goods
(2) Developments Affecting the Potential Demand for Consumers' Durable Goods
(i) Automobiles
(11) Household Furnishings
(ill) Residential Construction
(3) Developments in the Field of Government Economic Policy
B. The Boom: July 1936 to March 1937
(1) Initiation of the Boom— July to December 1936
(a) Developments Affecting the Demand for Steel by the Consumers' Goods Industries
(1) Automobiles
(2) Household Goods
(3) Residential Construction
(b) Developments Affecting the Demand for Steel by the Producers' Goods Industries
(1) Railroads
(2) Machinery and Equipment
(3) Business Construction
(4) Public Construction
(c) The Behavior of the Price of Steel
(2) The Critical Period: January to March 1937
C. The Decline Begins: April to September 1937
D. Recession: October 1937 to June 1938
E. Recovery: June to December 1938
F. Recent Developments: January to October 1939
G. Conclusion
I. Introduction
A substantial increase in the level of steel prices early in the spring of 1937 was
followed shortly by a considerable decline in the volume of new orders of steel.
In June, 1938 a drop in steel prices almost immediately preceded several months
of rising sales by steel makers. Temporary bargain markets for certain steel
products occurring in the autumn of 1938 and recently in May of this year were
accompanied by temporary increases in steel purchases.
To the superficial observer, these events might indicate that reductions in the
price of steel greatly increase the tonnage sales and revenues of steel producers,
and conversely, that increases in the level of steel prices greatly reduce their
volume of business and revenues.
The primary question which the analysis of the following pages will attempt to
answer is this: How important was the level of steel prices in accounting for the
fluctuations in the demand for steel that occurred in the 1936-39 period? In
answering this question it will be necessary to examine the importance of other
aspects of price behavior, as well as the importance of other factors which might
account for the changes in demand.
II. Summary of Findings
An analysis of the period June 1936 to October 1939 leads to several conclusions
with respect to the influence of steel prices on the quantities of steel demanded in
the short run:
(1) Changes in the levels of steel buying during this period were largely
determined by:
(a) the current and anticipated levels of business activity, income and
profits;
(b) the expectations with respect to steel prices in the immediate
future as compared with current steel prices;
(c) the volume of steel inventory accumulated in the immediate past;
and
(d) the length of time required to fill new orders for steel.
(2) There is little evidence that the actual level of the buying price of
steel — within rather wide limits — is of importance in explaining the act^'al
level of steel buying, at least in the short run. That is to say, actual changes
in the price of steel, per se — apart from their efl'ect on buyers' expectations
as to the magnitude and direction of future changes — are of minor importance
in accounting for short run changes in the volume of steel purchases.
To illustrate, the reduction in new orders of steel following shortly after
the price advances of March 1937, and the increase in new orders which came
at about the same time as the price reductions of June 1938, can be more
reasonably explained by factors other than changes in steel prices. The
great increase in steel buying in September and October of 1939, which
occurred without reduction in steel prices, further illustrates the dominant
influence of factors other than price in the demand for steel.
13944 CONCENTRATION OF ECONOMIC POWER
(3) There is evidence, however, that widespread expectations that the level
of steel prices is going to advance substantially in the near future due to
rising labor and other costs, and that the advance will not be temporary, will
generally lead to substantial increases in the present volume of steel buying.
Such anticipations are strengthened if there have been price increases in the
nfear past, and if feelings of price inflation and rising business activity are
generally abroad in the economy. When the expectation becomes a certainty,
as when a price increase is announced prior to its effective date, the effect
will almost certainly be an increase in the level of buying from what it would
otherwise have been.
A substantial part of the great increase in new orders that took place in
the period November, 1936 to March, 1937 can be accounted for by such
protective forward buying in advance of price increases which had been
announced. This was especially true of the two months December, 1936 and
March, 1937 when announcements of price increases resulted in a greatly
increased volume of new orders.
(4) As a corollary to the above point, expectations of a stable level of
prices following a period of price advances will ten,d to reduce the volume
of buying. That is, anticipations of stable prices will tend to lead to a lower
volume of buying than anticipations of rising prices. Thus the cessation of
anticipated rising prices will in general be accompanied by a short run decline
of purchases.
For example, the decline in orders in January and February, 1937, is partly
explained by the fact that steel purchasers were relatively sure that steel
prices would not advance much before April 1, 1937. Part of the decline
in orders in the months following the price advances of March, 1937 can be
similarly explained.
Conversely, expectations of stable prices following a period of falling prices
tend to increase the volun i of steel purchases.
(5) Anticipations of faF ng steel prices tend to reduce the amount of steel
purchases in the short ru .
(6) A reduction in stee" prices that is expected to be temporary — that is,
a temporary low pr-ce market for buyers — in general wiU cause a short run
increase of buying. Cases in point are the increases in steel orders that
took place in October, 1938, and again in May, 1939.
For the most part, the effect on the volume of steel buying of such price antici-
pation factors as have been discussed above is merely to cause a short run shift
in the date of actual purchases, without changing the total amount bought over
a twelve or eighteen months' period from what it otherwise would hav^ been.
This follows from the fact that the total steel requirements of steel purchasers
over a twelve or eighteen months' period depend very largely on their current
and expected output of products made from steel. Increases (or decreases) of their
inventories of steel much beyond their expected production needs in the near
future are both expensive and risky.
Thiis, for example, short run heavy buying of steel in advance of expected
price increases, and in amounts considerably greater than current production
requirements, is likely to be followed by a substantial short run drop in purchases.
This reduction in buying will be aggravated if purchasers find their actual re-
quirements smaller than expected, and thus have unnecessarily large stocks of
steel on hand. A considerable part of the decline in orders for steel that took
place in the six or eight months following March, 1937, can be accounted for by
the great volume of forward buying between October, 1936, and March, 1937,
and the failure of actual requirements to come up to expectations after March.
In summary, while price changes and anticipated price changes affect the timing
of steel purchases, the large fluctuations in the total volume of steel production during
the 1936-1939 period cannot be attributed to changes in the levels of steel prices. On
the contrary, the evidence compels the conclusion that the influence of the level of steel
prices on the total consumption of steel was relatively unimportant.
III. Nature of the Demand for Steel
Considerations of central importance in any investigation of the demand for
steel are:
(1) Steel itself is a durable commodity; that is, it may, with some excep-
tions, be kept in stock for more or less long periods without serious physical
depreciation. Thus purchasers of steel may currently buy more steel than
they need for current consumption, building up a stock of steel for future
consumption. Conversely, the building up oi such a stock in the past enables
CONCENTRATION OF ECONOMIC POWER 13945
a steel purchaser (user of steel) currently to buy less steel than he consumes,
the balance of such consumption coming from depletion of his steel stock.
The size of these inventories will depend for the most part on the buyer's
present and expected production requirements of steel, on his anticipations
regarding near future prices of steel, and on his expectations regarding the
length of time it will take him to get delivery on near future orders of steel.
If the steel buyer expects that prices of steel shortly will be higher, or that
near capacity operations of steel producers may delay delivery on his orders
at a time when he expects his requirements will be high, he may currently
buy more than he needs for present consumption, stocking steel as protection
against future higher prices or delivery delay. On the other hand, if his
steel requirements turn out to be smaller than expected, he may find himpelf
with unnecessarily large inventories of steel on hand, and thus may consume
from stock and curtail his buying. This will be expecially true if the steel
buyer expects lower steel prices in the near future.
However, such changes in inventories for the most part exert only a short
run effect on steel buying, for, as already pointed out:
(a) Investment in inventories ties up capital, and thus is costly; and
(b) Drastic changes in inventories are risky; the buyer must be
relatively sure of the near future course of prices, and of his own pro-
duction requirements before he can afford to build up or deplete his
stock much out of line with his reasonably expected volume of consump-
tion of steel. Such anticipations are the more uncertain the further into
the future they extend.
This is not to say that changes in the size of steel inventories in the hands
of consumers are unimportant in the short run. Such changes are probably
of paramount importance in explaining the heavy buying of steel in the fall
and winter of 1936-37, and the drastic decline that followed.
(2) Steel is largely used in the production of durable producers' and
consumers' goods. (1) * As shown elsewhere, however, the demand for durable
goods is particularly subject to severe cyclical fluctuations. It is to be
expected that the demand for steel should likewise be severely affected by
general business cycles, since the demand for steel is ultimately derived from
the demand for the services rendered by durable goods.
(3) The demand for steel is inelastic. This means that a given percentage
change in the price of steel gives rise to a smaller percentage change in the
quantity of steel demanded. A reduction in the price of steel (if unassociated
"with other factors such as improving business conditions) therefore tends to
reduce the dollar volume of steel sales. This inelasticity of the demand for
steel follows from three factors:
(a) The low elasticity of demand for most of the finished products
which are made from steel;
(b) The low degree of substitutability of steel for other materials and
of other materials for steel. This low substitutability materially lessens
the responsiveness of the demand for steel to the changes in price; and
(c) The relatively small proportion of the value of the finished product
which is represented by the cost of steel. A large percentage change in
the price of steel can lead to only a small percentage change in the price
of the finished product even if the entire steel price reduction is passed
on to the final consumer. Since finished products made from steel
have, in general, a low elasticity of demand, changes in steel prices, can
have only a small effect on the output of finished products, and, therefore,
on steel sales.
These considerations taken together lead to the conclusion that, the large
changes in the demand for steel, which were associated with the changes in
the prices of steel during the period 1936-1938, were the result of (1) changes
in business conditions, (2) anticipation of future price changes, and (3)
anticipation, at times, of future delivery delays, rather than of actual changes
in the price of steel.
This conclusion is verified by the following consideration of the actual
course of events between June, 1936, and October, 1939.
IV. Short Run Variations in the Demand for Steel: June 1936 to October
1939
A. THE ECONOMIC BACKGROUND, 1930 TO 1935
To analyze the behavior of the demand for steel in the period June, 1936 to
October, 1939, an understanding of the economic developments in the depression
13946 CONCENTRATION OF ECONOMIC POWER
years 1930 to 1935 is necessary. These prior developments can be most fruitfully
appraised in terms of their later effects on the demand for steel if they are broken
down into three classes: (1) developments affecting the potential demand for
producei*' durable goods, (2) developments affecting the potential demand for
consumers' durable goods, and (3) developments in Government economic policy.
'(1) Developments Affecting the Potential Demand for Producers' Durable
Goods
(i) The population of the United States at the end of the year 1935
was almost five percent greater than at the end of 1929, and was increas-
ing at the rate of about seven-tenths percent per year. {2)
(ii) In the face of this substantially larger potential market for goods,
the productive capacity of American industry had actually declined
slightly. The accumulated excess of consumption of business plant and
equipment over production of these goods in the six years 1930 to 1935
amounted to more than three and a half billions of 1929 dollars. In
the six years 1924 to 1929 there had been a net excess of production
over consumption of these goods totaling almost twenty-one billions of
1929 dollars. (5)
(iii) Business expenditures (in 1929 dollars) on plant and equipment,
which at the bottom of the depression had faUen to about one-third of
their 1929 level, had in 1935 regained only about one-third of ihe depres-
sion loss and then stood at slightly over one-half of the 1929 level.'
Practically all of the recovery that had been made in capital expendi-
tures, moreover, had been gained in equipment expenditures. By the
end of 1935 business expenditures on equipment had regained almost
half of their deprespion loss, and were approximately two-thirds of the
1929 figures. (5) Yet business expenditures .on plant, "which in 1933
had fallen to one-quarter of the 1929 amount, had in 1935 regained
only one-sixth of the loss, and stood at less than forty percent of
the 1929 level. (5)
It is thus clear that up to the beginning of 1936 practically no recovery
at all had been made in long-term investment commitments in the busi-
ness plant of the country, and that recovery even in business equipment
expenditures had been small. Moreover, what recovery there had
been in equipment expenditures consisted largely of replacement outlays
made on a short-term basis as required by current demands on pro-
ductive capacity.
(iv) Behind this picture of small recovery in capital goods production
and in long term investments, lies another, the picture of a similarly
small recovery in business profits.
In 1932 corporate net earnings had dropped from a 1929 peak of over
$8 billion to a net deficit of over $5 billion; by 1935 they had recovered
only slightly more than one-half of the decline, and were only about
one-fifth of the 1929 level. (4) Thus, there had been Uttle incentive
for business men to make investment commitments in the years 1930
, to 1935. (5) But more important in interpreting later developments,
the fact that the current level of profits at the end of 1935 was so low,
and the previous six years on balance so heavily on the deficit side, {6)
made the outlook for long term investment in heavy capital goods an
exceedingly bleak one.
(v) Contrasted with the small recovery in profits and business capital
outlays was a somewhat more substantial recovery in consumers' outlay
and consumers' income. Consumers' real expenditures in 1935 were
only about thirteen percent below the 1929 level, and had regained one-
half of the depression loss. (7) Consumers' real income in 1935 was
similarly only about thirteen percent below its 1929 level. (7) By the
end of 1935, therefore, 1929 levels of activity in consumers' goods pro-
duction were not far distant.
It is clear that the recovery which had taken place in the years 1933
to 1935 was largely a consumption recovery, with capital expenditures
consisting mostly of the replacement outlays required by increasing
production of consumers' goods. A sustaining base of recovery in
long term investment commitments was absent.
(2) Developments Affecting the Potential Demand for Consumers' Durable
Goods. — Beside the five percent increase in population which had occurred
in the years 1930 to 1935, three important developments in the three major
categories of consumers' durable goods, automobiles, housing, and house-
furnishings had taken place:
CONCENTRATION OF ECONOMIC POWER 13947
(i) Automobiles (8). — With the decline in income after 1929, consumers
drastically curtailed their purchases of new automobiles. In the low
year, 1932, passenger car sales were only one-quarter of the 1929 level.
With consumer income increasing the years 1933 to 1935, automobile
sales recovered about one-half of the 1929 to 1932 decline; the total for
1935 was almost two-thirds of that for 1929. (£»)
The total consumer stock of passenger cars, which had increased
seven million in the six years 1924 to 1929, had grown less than half a
million in the six depression years 1930 to 1935. (10) Moreover, the
age of the stock had increased greatly. ' At the end of 1935 almost
half of the cars were more than six years old, whereas, in 1929 only
about one-seventh were, more than six years old. (10) Thus, at the
beginning of 1936 there loomed a very large potential "modernization"
demand for automobiles. With consumer income at that time already
not far from pre-depression levels, only small increases would be required
to bring forth that demand.
(ii) Household Furnishings. — Expenditures (in 1929. dollars) for
household furnishings fell less in the depression, and had recovered more
than for automobiles. At the bottom in 1933 they had fallen to about
55 percent of their 1929 level; in 1934 and 1935 more than half of the
loss had been recovered. (11) In the case of household furnishing there
existed not only a potential replacement demand, but also a large poten-
tial demand for "new invention" goods which had only begun to come
on the market before the depression — such goods as electric refrigerators
and other new electric appliances.
(iii) Residential Construction. — Developments in housing were in
marked contrast to those of other consumers' durable goods. Total
expenditures (in 1929 dollars) on residential construction at their Iqw
point in 1933 were only about one-sixth of the 1929 figure, and approxi-
mately one-tenth of the predepression peak in 1925 when the total was
over $5 billion. ' Practically no recovery was made in 1934, and in 1935
the total was only about 40 per cent of that in 1929 and only one-quarter
of the depression drop had been regained by that time. (12)
The result of this drastic decline in the level of housing expenditures
was the accumulation of a potentially large replacement and moderni-
zation demand. In the six years 1930 to 1935 the excess of consumption
of residential housing over construction had accumulated to a total of
eight and a half billions of 1929 dollars, about seven times the amount
spent in 1935, and almost three times the 1929 total. (12) Add to this
decline of the stock of housing, its greatly increased average age, the
five percent increase of population, and the great technical developments
in housing construction and it is clearly apparent that at the beginning
of 1936 there was an extremely large potential demand for housing.
Although relatively unimporatnt in the mid^t of the depression, the
relation between rents and costs became increasingly significant in the
upswing, particularly in early 1936. (13) From 1929 to 1933 the decline
in construction costs in most localities was not as great as the decline in
rentals. However, in 1934 and 1935 a substantial recovery was made
in rents while costs rose considerably less. Though the relation of rents
to costs at the beginning of 1936 was more favorable, the situation was
still precarious; a rather small rise in costs relative to rents would be
sufficient to stop recovery.
(3) Developments in the Field of Government Economic Policy. — In 1933 the
Federal Government inaugurated an unparalleled program of particlipation
in the economic activity of the country in the hope of hastening recovery
The most important of the developments were:
(i) A heavy deficit spending program in which the net contribution of
the Federal Government to disposable cash income increased from an
average of $160 million a month in the last ten months of 1933 to $310
million a month in 1935. (14)
This deficit spending was a very important factor in the 1933 to 1935
recovery of consumers' income and outlay, but uncertainty as to the
future course of spending, and the increase of the Federal debt, (15) con-
tributed to the absence of recovery in new lona; term investments.
(ii) Marked changes were made in Government monetary policy.
The Thomas inflation amendments to the Agricultural Adjustment Act
in 1933, the devaluation of the d6llar and the Silver Purchase Act of
124491— 41— pt. 26 24
13948 CONCENTRATION OF ECONOMIC POWER
1934, the Banking Act of 1935, along with a rapid and uncontrolled'flow
of gold into the United States, and rise of the Federal debt made business
men extremely apprehensive of rapid price inflation.
(iii) Government support of increasing wage rates, and the stimula-
tion of the growth of labor organizations by new Federal statutes
brought to business fears of rising wage costs and impending production
stoppages.
(iv) The N. R. A., the A. A. A., and other price-production experi-
ments made business men highly uncertain as to the future course of
Government industrial policy.
In summary, this was the background situation at the beginning of 1936:
(1) Consumers' income and outlay and industrial activity — 'except in
the heavy capital goods industries — were already approaching pre-
depression levels.
(2) With more and more industries nearing capacity operations, the
level of activity at which business expenditures for replacement of
equipment could no longer be deferred was not far distant. Greater
activity in the capital equipment industries would in turn increase con-
sumers' income and thus the demand for equipment.
(3) Similarly, a continued rise of consumers' income was almost cer-
tain shortly to bring forth the "modernization" demand for consumers'
durable goods (except possibly housing) whose purchase had been de-
ferred in the previous six years of depression.
(4) The belief was becoming wide-spread among business men that
the policies of the Federal Government would lead to price inflation.
(5) Investment — in housing and business plant — based on a long
term outlook had mtide little recovery.
The situation was dominated by the short run outlook, and was for that
reason vulnerable. All the elements of a short run boom were there, but
the requirement for a sustained recovery — recovery in outlays made on the
basis of the long term outlook — was lacking. Reasonable certainty as to
the future was absent. There was thiis a real danger that a" boom might
develop, reach runaway proportions, and collapse without any cushion of
long term industrial activity to fall back upon.
B. THE boom: JULY 1936 TO MARCH lfl37
Although business activity had declined somewhat in the winter of 1935-1936,
sentimeht among steel consumers by mid-spring was predominantly optimistic in
regard to immediate prospects. These feelings were undoubtedly strengthened
by the rapid and unbroken rise which advanced stock prices more than two-
thirds above their March, 1935 level (16) and the fact that the enactment of the
Veterans' Bonus in the winter promised a high level of activity in the last six
months of 1936. By May, most of the business activity indexes were moving
upward, some experiencing rather substantial contra-seasonal rises. But the
situation as yet had little of the characteristics of a boom, although there was
beginning to be considerable discussion of the possibility of inflation.
(1) Initiation of the Boom — July to December 19S6. — A combination of events
beginning in the three months May, June, and July, 1936, however, started an
upward movement of prices and industrial production which reached the status
of a boom by the end of the year:
(i) Payment of the Veterans' Bonus in June brought about a big increase
in consumers' income. {17) The result of the payment was to be felt in the
next six or eight months through its effect on sales of consumers' goods.
(ii) The agrioultural drought early in the 1936 crop season, and world-
wide re-armament accelerated by the Spanish Civil War led to a rapid rise
in the prices of basic agricultural products and industrial raw materials. (,18)
(iii) Of great importance in strengthening the feeling of price inflation in
the summer of 1936 was the prospect of higher wage rates arising out of
threats to organize such basic industries as steel and automobiles.
The addition of these developments to a situation already containing the basic
elements of a boom was all that was necessary to make the boom — and price
inflation — a reality.
The progress of the boom as it affected the steel industry in the last six months
of i936 is indicated in Table 1. Instead of the usual summer decline in orders,
f'aipments, and production, actual increases occurred, so that the third quarter
was somewhat better than the second. By the beginning of the last quarter
activity had definitely taken on boom proportions.
CONCENTRATION OF ECONOMIC POWER
13949
Table 1.-
-Composite Steel Price, U. S. Steel Corporation Subsidiaries Bookings
and Shipments, and Total Ingot Production
Iron Age
Composite
Price of '
Finished
Steel (cents
per pound)'
(1)
U. 8. Steel
Corp.
Domestic
Bookings of
Rolled and
Finished
Steel (OOO's
of gross -ind
net tons)»
(2)
U. S. Steel
Corp.
Domestic
Shipments of
Rolled and
Finished
Steel (OOO's
of gross and
net tons)'
(3)
Steel Ingot Production In the
United States
Date
Total
(OOO's of
gross
tons)'
(4)
Percent
of
capacity
(6)
Federal Re-
serve
Board
Index
Adjusted for
Variation
(1923-25
= 100) *
(6)
1929*
2.2m
2.048
1.957
1.901
1.8'9
2.033
2.058
2.062
Z040
2.021
2.028
2 028
2.033
2 091
2.081
2.096
2.116
2.116
2.199
1,264.9
923.9
554.9
274.1
433.8
520.1
619.0
771.0
604.9
926.3
877.1
818.4
1,055.8
1. 092. 4
R3&9
953.8
1. 144. 5
909.0
2,033.3
1,168.9
900.1
596.9
311.8
460.6
449.3
672.7
688.8
619.2
736.0
931.9
883.6
876.7
886.3
874.4
898.1
952.7
863.1
985.9
4,571
3,300
2,119
1,122
1,908
2.162
2,828
3.086
3,002
3,384
3,991
4,097
4.035
3,976
4,247
4,214
4,601
4,389
4,491
89.05
63.09
3&13
19.76
33.53
37.:38
48.64
52.39
54.63
67. 4C
69.99
69.68
70.75
67.6]
72.11
74.06
78.15
76.94
76.42
131
1930*
96
1931*
01
1832*
32
1933*
1034*
M
83
1935*
81
1936
Jan — . —
88
Frt> \
86
Mm
86
Apr
102
May" - -
107
Juno
116
July -
121
Aug - . .
123
s^:::, :..::;
121
S^o^v -----
ISO
1^
Dec
147
Monthly Average
2.077
999.6
•848.6
3.959
68.36
113
1937
Jnn
2.249
2.249
2.459
2.512
2.512
2.612
2.612
2.512
2.512
2.512
2.512
2.612
1,302.8
1, 270. 1
1,839.2
1, 167. 0
866.' 9
840.6
793.7
693.0
511.0
471. J
477.9
1, 143. 2
1,040.8
1,332.9
1,224.8
1,211.8
1,174.0
1,060.6
971.7
946.9
665.3
470.1
406.6
4.786
4,498
6,303
5,156
5,237
4,254
4.631
4,968
4,362
3.449
2,189
81.32
84.27
89.94
90.26
88.79
74.48
78.48
83.83
76.30
58.31
38.23
25.37
142
Feb
131
Mar
128
Apr
132
Mii :: ::: .:
136
Tnnf>
121
July - — -
141
Aug
144
Sept
125
Oct : .
101
Nov . ...
68
Deo
49
Monthly Average
2.464
922.7
965.7
4.193
72.38
118
1938
Jan L .
Z512
2.512
2.512
2.612
2.506
2.459
2.300
2.300
2.293
2.265
2.286
■2.286
426.8
376.9
503.8
541.8
3.38.3
428.6
485.0
518.1
663.5
729.1
778.0
665.0
484.1
380.4
476.4
450.6
399.6
443.6
410.6
626.3
525.7
636.1
628.7
670.3
1,733
1,704
2,012
1,925
1,807
1,638
1,982
2,647
2,658
3,118
3,572
3,143
29.15
31.74
33.85
33.44
30.39
28.40
33.42
42. 85
46.28
52.45
62.05
53.00
52
Feb -
61
Mar.
49
Apr .- -
60
mi: ::::
48
June
47
July
64
72
setrt"::.::::::::::::""
77
Oct
92
Nov
112
Dec
104
Monthly Average
2.394
636.2
602.1
2.320
39.79.
68
• Average of monthly figures.
' See: The Iron Age, January 6, 1939, p. 199.
' Source: U. S. Steel Corporation.
' Sec: The Irov Age, January 6, 1939, p. 191.
* See: Federal Reserve BvMetin, June 1937, July 1937, p. 675, July 1938, p.
p. 916.
3, July 1939, p. 694. October, 1939,
13950 CONCENTRATION OF ECONOMIC POWER
Table L — Composite Steel Price, U. S. Steel Corporation Subsidiaries Bookings
and Shipments, and Total Ingot Production — Continued
Jan..
Feb.
Mar.
Apr.
May,
June.
July.
Aug.
Sept.
Iron Age
Composite
Price of
Finished
Steel (cents
per pound)
2.286
2.286
2.286
2.286
2.236
2.236
2.236
2.236
2.236
U. S. Steel
Corp.
Domestic
Bookings of
Rolled and
Finished
Steel (OOO's
of gross and
net tons)
(2)
622.9
662.4
603.6
623.9
670.9
655.3
719.3
U. S. Steel
Corp.
Domestic
Shipments of
Rolled and
Finished
Steel (OOO's
of gross and
net tons)
734.6
602.7
678.1
715.0
639.0
662.6
618.0
Steel Ingot Production in the
United States
Total
(OOO's of
3,174
2,989
3,405
2,974
Percent
of
capacity
52.69
64.10
66.14
50.99
48.24
53.44
52.40
62.22
Federal Re-
serve
Board
Index
Adjusted for
Variation
(1923-25
= 100)
» The 1939 figures for The Iron Age Composite Price of Finished Steel are for the middle week af the respec-
tive months, and are from current issues of The Iron Aqe. The 1939 figures for columns (4) and (5) are from
current issues of The Iron Age. The 1939 figures for total steel ingot produrtion (column 4) do not include
steel for castings. The 1939 bookings figures are not strictly comparable to those of previous years since
they do not include "requirement contrncts"- that is, "price protection" contracts without actual
specifications.
The developments of the last six months of 1936 can best be analyzed in
terms of their effects on the demand for steel if they are studied in three groups:
(a) Developments affecting the demand for steel by the consumers' goods indus-
tries; (b) Developments affecting the demand for steel by the producers' goods
industries; and (c) Behavior of the price of steel.
(a) Developments Affecting the Demand for Steel by the Consumers' Goods
Industries
(1) Automobiles. — By the time production of the new 1937 models was
in full swing in October, 1936, automobile makers were already predict-
ing a 5,000,000 car year, i.e., one about as large as that in 1929. (19)
Automobile producers justified their anticipation on the basis of (1) the
unusually high level of sales in the spring, (.20) (2) the generally rapid-
rise of consumers' income and industrial activity and the widespread
optimisrti that business activity was going to continue upward, (3) the
boost to sales expected as a result of the Veterans' Bonus, and (4) a
recognition that consumers had a large depression "backlog of deferred
purchases" to make up.
Nor were these feelings unjustified by the events of the last quarter
of 1936. In October, November, and Decemoer, orders for the 1937
models poured in to producers at a 1929 pace. (21) Seasonally adjusted
retail sales of automobiles in November and December were up to the
1929 peak level and about forty percent greater than the corresponding
months in 1935.(^0)
By the end of December, automobile producers had already placed
~ orders for considerably more than half of the steel requirements on
their expected production of 1937 models. (.?,?) Inasmuch as the auto-
motive industry was the biggest steel consumer, these orders were un-
doubtedly a major factor accounting for the high level of steel bookings
jn the last four months of 1936.
(2) Household Goods. — The situation in the household goods industry
was not substantially different from that in automobiles. Though few
monthly figures on orders, sales, or production of household goods are
available, cjiiarterly data for sales of electric stoves and electric house-
hold refrigerators, and shipments of vacuum cleaners indicate that
sales of household goods in the las.t quarter of 1936 must have been
25% to 50% greater than in the corresponding quarter of 1935.(23)
CONCENTRATION OF ECONOMIC POWER 13951
Although the household goods industry is not one of the major con-
sumers of steel, some rather substantial tonnages were undoubtedly
placed in the last half of 1936, especially for lighter types of steel.
(3) Residential Construction. — Activity in residential construction'was
in marked contrast to that in automobiles and household goods. In
March, 1936, the value of residential construction contract awards
begJan a rapid upward climb. By mid-summer it appeared that the
long-delayed recovery in housing was at last under way; August con-
tract awards (adjusted for seasonal variation) were over 80 per cent
greater than for February. (:?-^)
However, the recovery was short-lived. Contract awards (seasonally
adjusted) reached a peak as early as September, and were about seven
per cent lower in the last quarter than in the third. An important
factor in cutting short recovery was the rise in building costs that took
place in the last half of 1936. (i5) Though residential construction
undoubtedly contributed to the high level of steel activity in the last
half of 1936, it was apparent before the end of the year that — barring
an unusual rise in rents, or a miracle — no further stimulus could be
. expected from that quarter. Already an important deflationary element
had entered the picture.
(b) Developments Affecting the Demand for Steel by the Producers' Goods
Industries
(1) Railroads. — In April, 1936, freight car loadings began an upward
rise, which increased in rapidity in the following summer and autumn.
By December, 1936, they had advanced to a level about twenty-five
per cent above the March figure and that for the previous December.
Although the level of freight car loadings in the last half of 1936 was
about twenty-five per cent under the 1929 amount, the railroads' stock
of serviceable freight cars had been so depleted (by failure to replace)
in the depression years that the freight car surplus in the last half of
1936 M^as actually more than forty percent under the 1929 average
surplus. (.^5)
Revenue passenger-miles in the last half of 1936 were about twenty-five
per cent above the first half of 1936.(^5)
As a result of increased freight and passenger traffic, net railway
operating income advanced rapidly. The average for the last quarter
of 1936 was over 30 per cent greater than for the fourth quarter of
1935. The railroads' net income for the same periods rose from $63
million to $126 million. (.^5)
With traffic nearing capacity levels, and with a greatly improved profit
outlook and current income situation, the railroads came into the market
with a tremendous upward spurt of orders in November and December,
1936.(^7) New orders of freight cars in December were greater than
the total for the year 1935. The number of new locomotives ordered
in November and December was greater than the combined total for
the three years 1932 to 1934. Orders for passenger cars and steel rails
similarly showed great increases. Undoubtedly the greatly increased
demand of the railroads for rails and rolling stock was an important
factor in the great rise in bookings of heavy steel in the last'quarter of
1936, and especially in December.
(2) Machinery and Equipment. — What was happening in the railroad
industry was happening more or less • similarly in industry jn general.
In April, 1936, industrial production began to move rapidly upward;
by December it had advanced more than twenty-five per cent over
March (on a seasonally adjusted basis), and had actuaUy reached the
average level for 1929.(5) In view of the decline of the stock of ma-
chinery and equipment that had taken place since 1929, wide areas of
industry were undoubtedly operating at or near capacity in the last half
of 1936. Thus, if demands on productive capacity were to be met, fur-
ther postponement of outlays for replacement of industrial equipment
was no longer possible.
An added incentive to make equipment outlays came from the fact
that the profit outlook was substantially improved. Profits of indus-
trial corporations for the fourth quarter of 1936 were more than forty
percent above the corresponding period of 1935, although still not much
more than half the 1929 level. (^5)
13952 CONCENTRATION OF ECONOMIC POWER
By mid-summer new orders of industrial equipment were increasing
rapidly, and by autumn activity in the machinery and equipment indus-
try was definitely in the boom stage. The year 1936 ended with a
tremendous spurt of new orders in December. {27)
Part of the increase of new orders was caused by rising equipment
prices in the last half of 1936, and the expectation that prices were going
to rise still further. (;gS) A substantial volume of orders for equipment
which was not required by current demands on capacity and which
otherwise might not have been purchased was placed as protection against
expected price advances. This was undoubtedly true in December when
widespread expectations of price advances early in the first quarter of 1937
caused many buyers to protect themselves by ordering in advance. (;85)
(3) Business Construction. — Activity in the business construction in-
dustry began to rise in the spring of 1936, and continued to advance
throughout the remainder of the year, but at a slower pace in the last
quarter. (;85) Although the total value of construction contracts
awarded for commercial and factory buildings and public utilities in the
last half of 1936 rras more than sixty percent greater than for the same
period in 1935, it was only slightly more than one-sixth of the total for
the full year 1929; recovery was undoubtedly still being impeded by
the low level of profits. However, the increase in 1936 contributed
importantly to the advance of activity in heavy steel production.
(4) Public Construction. — Developments in public construction were
markedly diflFerent from those in pri'vate building. The value of con-
struction contracts awarded for public buildings and public works and
the value of highway and grade crossing projects approved for con-
struction, which had risen very rapidly in the years 1934 and 1935,
continued to increase throughout most of the first half of 193Q.(29)
The peak, however, was reached early in the summer, and the last half
of the year was largely one of decline. The reasons for the decline were:
(i) With private industrial activity increasing, there was less pressure
to provide employment on public projects; and
(ii) Fears that continued public spending would lead to runaway
price inflations and demands for a baJanced Federal budget prom{)ted
a reduction of Federal construction outlays.
Though public construction expenditures even in the last half of 1936
were at a higher level than for 1935, the outlook for 1937 was one of a
continuing reduction. Thus another important deflationary element
was added to the prospects for 1937.
(c) The Behavior of the Price of Steel. — From the discussion of the previous
pages, it is apparent that the great upward shift of the demand for producers'
and consumers' goods in the last half pf 1936 undoubtedly would have caused
a large increase in the demand for steel irrespective of what had happened
to its price. As we shall see, however, the behavior of the price of steel
actually increased this demand.
During the first half of 1936 steel prices were for the most part unchanged.
In the last part of May, however, leading steel producers announced small
price advances on many steel products effective beginning July 1, for third
quarter shipments. The increases raised the average level of steel prices
about three or four per cent. (50) The effect of the price advances was to
strengthen expectations of further price increases. (S/) There followed a sub-
stantial amount of orders in June as protection against price increases. (S^)
There is no evidence that the price increases had any important effect on
steel purchases in the third quarter; July bookings were greater than the
total for June, and the total for the third quarter greater than that fo^^the
second. (33)
Prices for a few steel items were raised in September and October, 1936, but
the effects of the advances were slight, except to further the now widespread
anticipations of substantial increases over a wide range of steel products. (54)
These expectations were bolstered by (1) rapidly rising prices elsewhere in
the economy, (55) (2) the great increase in steel activity, (3) the "rising cost
of steel scrap, (5(5) and (4) expectations of an early rise in wage rates in the
steel industry. (57)
On November 16, 1936, a wage rate increase of approximately ten per cent
was granted to steel workers by the majority of steel producers. (57) The
effect of this advance was materially to incroi^f* ^^<^ T>at of steel production
CONCENTRATION OF ECONOMIC POWER 13953
Other increased costs had become effective in 1936, including the Social
Security Tax, and, in the case of the manufacturing subsidiaries of United
States Steel Corporation, vacations with pay to certain employees. Soon
after the wage advance of November, 1936, price announcements for ship-
ments during the first quarter of 1937 were made by the principal steel
producers. The announced price increases covered practically the whole
range of important steel products and raised the average level of steel prices
about six per cent. (38) The immediate effect of the price advance was the
placing in December of a very large amount of forward orders as protection
against the announced January price advances. (35) December bookings of
the subsidiaries of United States Steel Corporation were more than double
those in November, more than twenty-five percent greater than the peak
month in 1929, and were the largest monthly post-war bookings in the
history of the Corporation. (40)
Although a great increase in bookings might have been expected in De-
cember because of a simultaneous large upward shift in the demand for
products made from steel, (4^) the magnitude of the increase indicates clearly
the importance of speculative forward buying in that month.
An additional element that entered the picture in December, 1936, and
stimulated buying, was that deliveries — especially of steel sheets — were
being delayed by pressure on capacity. By mid-December deliveries of steel
sheets, for example, could ^not be promised within less than three or four
months. (4^) With producers of products made from steel expecting large
steel production requirements in the first six or eight months of 1937, some
forward tonnages were placed in December as protection against future
delivery delay.
(2) The Critical Period: January to March 1937. — Domestic bookings of steel
in January and February 1937 dropped approximately 35 per cent from the De-
cember level although they were still about 30 per cent above the level of the
previous October and November. (35) The most important reasons for the
decline were:
(i) Steel consumers had undoubtedly covered a large part of their first
and second quarter requireinents in 1937 -by orders in the last quarter of
1936, and especially in December. (43)
(ii) Further steel price advances, although anticipated as the outcome of
rising costs in the steel industry, were not expected much before the begin-
ning of the second quarter of 1937.(44) Thus in January and February
there was little price incentive for protective forward buying. This factor
alone is sufficient to account for a large part of the drop from December,
1936, when advance buying had been very large.
(iii) New orders of products made from steel similarly had declined from
a high forward buying level in December, 1936.(45)
(iv) Widespread strikes prevented substantial specifications of steel from
the automobile industry.
Although bookings had declined, the large backlog of orders, placed in the last
quarter of 1936, raised production and shipments of steel in January and Febru-
ary, 1937, above the corresponding December figures. (33)
In March, however, there was another upswing in orders of steel. Bookings
were almost fifty per cent greater than for February, but were about 10 per cent
under the peak figure for December, 1936.(33) Several factors were responsible
for the increase in orders in March:
(i) On March 2, 1937, Carnegie-Illinois Steel Corporation signed a labor
contract with the Steel Workers Organizing Committee which was followed
by formal contracts signed a few days subsequently by this and other manu-
facturing subsidiaries of United States Steel Corporation. Under these con-
tracts the basic common labor rate was increased about twenty percent
and there was established an eight-hour working day and a forty-hour work-
ing week, with time and a half compensation for all overtime in excess of
eight hours in any day, or in excess of forty hours in any week, such changes
becoming effective March 16, 1937. Other steel producers granted similar
wage advances. The signing of this first labor contract was immediately
followed by the announcement by various subsidiaries of United States
Steel Corporation in the first week of March, 1937 of price increases on most
steel products. Similar price increases were announced by other steel pro-
ducers. These price increases on the average raised second quarter prices
13954 CONCENTRATION OP ECONOMIC POWER
about ten percent above the level for the first quarter. (4^) The announce-
ment of Carnegie-Illinois Steel Corporation stated that the price increases
were necessary to cover the increased cost of production due to labor ad-
vances. A considerable part of the new orders in March consisted of forward
purchases placed as protection against these second quarter price advances.
(ii) At the same time the demand for products made from steel rose greatly
in March. Sales of automobiles and house-furnishings were much larger
than in January and February. (47) Rising industrial production and an-
ticipations of price increases also occasioned a large upward spurt of new
orders for machinery and equipment. (45)
(iii) Some large tonnages were also placed as protection against future
delivery delay. In March, delivery for some of the light hnished steels
could not be promised within less than approximately six months. (45)
Threatened production-stoppages due to possible labor difficulties bolstered
these fears of delivery delay.
In the meantime, however, several critical developments pointing to an almost
certain early collapse of steel buying were materializing:
(i) After a rapid rise in 1936, a decline of over 1.5 percent in real income
(adjusted for seasonal variation) occurred in January, 1937, and by March
it was .beginning to be apparent that consumers' real income was in the
leveling off stage. (50)
In addition, the Federal Government was reducing its expenditures while
at the same time considerably increasing its tax revenues; (14) the increase
was due largely to heavy Social Security Tax collections beginnuig in January
1937. As a result, the net contribution of the Federal Government to
consumers' disposable cash income dropped from an average of $370 million
per month in the last six months of 1936 t6 about $110 million per month in
the first four months of 1937. Such a reduction was large enough to have
significant deflationary effects. Since Government policy in the winter
and spring of 1937 was definitely toward a reduction of expenditures, the
outlook was for even further declinp.
The great upswing in sales of consumers' goods in 1936 and the first
quarter of 1937 had been abnormal. A substantial part of the sales un-
doubtedly consisted of "modernization replacement" purchases which
had been deferred in the previous depression years, and which were made
under the stimulus of the Veterans' Bonus and rapidly rising income. With
a considerable part of the deferred backlog of purchases already made up, by
the spring of 1937, sales of durable goods could not be expected to continue
at the high pace of the previous nine months unless consumers' income should
continue to increase very rapidly. If consumers' income should decline, the
level of sales was almost certain to drop.
Figures for automobile sales tend to support the above conclusion. The
peak of sales (on a seasonally adjusted basis) was reached as early as Decem-
ber 1936. (20) March, 1937, was about 5 per cent below December, 1936,
and June, 1937, about 20 percent below March.
Taking all these facts into consideration, the outlook at the beginning of
the spring of 1937 was very definitely toward a leveling off of, or a decline in,
consumers' expenditures. The immediate prospects for the consumers'
durable goods industries were thus definitely not bright.
(ii). A somewhat similar situation existed for the durable producers' goods
industries. Business expenditures for equipment in the twelve months,
March, 1936 to March, 1937, were about 50 per cent greater than the corre-
sponding outlays for 1935, and were at a level about equal to that in 1929. (51)
By the spring of 1937 it was probably true that a major part of the backlog
of replacement outlays deferred in the depression years had been made up.
Thus, unless industrial production continued to increase at a rapid rate so
that substantial outlays for expansion were required, expenditures on machin-
ery and equipment were almost certain to decline.
But by the end of March, 1937, it was becoming clear that industrial
production was not going to continue its rapid advance. The Federal
Reserve Board index of industrial production (adjusted for seasonal varia-
tion) reached a peak in December 1936 slightly above the 1929 average,
declined almost six per cent in January, 1937, rose slightly in February and
March and then levelled off and began to decline in the second quarter. (6)
Manufacture of non-durable goods (seasonallv adjusted) was definitely on
the deelir e bv the end of March, 1937. (6)
CONCENTRATION OF ECONOMIC POWER 13955
The situation was aggravated by the fact that piofits, after rising rapidly
in 1936, also began to level off early in 1937. Seasonally adjusted profits for
the first quarter were about one-sixth under the peak in the fourth quarter
of 1936. {26)
Additional disturbing elements in the first quarter of 1937 were the strikes
in the automobile, glass, rubber, machinery, and maritime industries, and
organizational activities in the iron and steel industry. The number of
strikes in progress in the United States trebled between December and
March, (52) and there was little indication that production stoppages were
going to decline.
(iii) The outlook for the railroads was also becoming much less optimistic.
Merchandise freight car loadings (seasonally adjusted) began to decline as
early as December, 1936, and miscellaneous loadings followed in February,
1937. (53) With traffic dropping off, though slowly at first, and with heavy
replacement expenditures already made in the last quarter of 1936 and the
first quarter of 1937, the pressure to provide further replacement was relaxed.
Railroad baying collapsed in the second quarter.
The collapse was hastened by the decline of the railroads' income. Net
income for each of the first and second quarters was far under the figure for
the last quarter of 1936. (25)
(iv) By April, 1937, there was a growing feeling that prices, which had
risen rapidly in the previous nine months, were very shortly going to sta-
bilize. (54) Factors important in dampening speculative anticipation of fur-
ther price inflation were:
(a) The excess of Government expenditures over receipts was being
rapidly reduced.
(b) On Deceml)er 21, 1936, the Secretary of the Treasury announced
the gold sterilization policy, whereby gold inflows were to be prevented
from contributing to inflation.
Cc) This was followed on January 30, 1937, by the Federal Reserve
Board announcement of a 33H per cent increase in enember banks'
reserve requirements.
(d) In January and February substantial increases in the yields of
government bonds occurred, and in March stock prices began to fall. (55)
(e) Rumors of an upward revaluation of the dollar in terms of gold
had become so strong in the first quarter of 1937, that President Roose-
velt was prompted to make a statement on April 9th that no change in
the United States' gold policy was contemplated.
(f) In March, prices of basic raw materials began to level off. (^.S)
With expectations of further price advances disappearing, a most important
stimulus to forward buying was being removed. Unquestionably many
orders both of steel and of products made from steel placed in the nine
months JuW, 1936 to March, 1937 had been of the speculative price pro-
tection variety. -
(v) The mere fact that advance buying both of steel and df products made'
from steel had been so heavy in the fourth quarter of 1936 and the first
quarter of 1937 pointed to a much lower level of orders throughout the two or
three following quarters. By the end of March buyers of steel had covered
a large part of their steel requirements for the next six or eight months.
Total domestic bookings of subsidiaries of United States Steel Corporation in
the six months, October, 1936 to March, 1937, were almost three per cent
above the six consecutive peak months of 1929, although their total ship-
ments in 1937 were more than seventeen per cent under the 1929 total. (56)
Total bookings in the six months referred to were equal to almost three-
quarters of the entire shipments made by the subsidiaries of the Corporation
in 1937.
With a large part of their steel requirements for 1937 covered by the end of
March, and with mills unable to promise delivery of all orders in hand in less
than six months, it was to be expected that steel consumers would be hesi-
tant about tying up more working capital and risking unduly large inventories
with further advance buying.
C. THE DECLINE BEGINS: APRIL TO SEPTEMBER 1937
The demand for steel began to decline rapidly in April, 1937. By September,
bookings had dropped more than sixty per cent below the March level. (66) But
production and shipments were maintained at a high level throughout the second
13956 CONCENTRATION OF ECONOMIC POWER
and third quarters largely on the basis of the large backlog of orders placed in the
winter of 1936-1937. Domestic shipments of subsidiaries of United States Steel
Corporation for the third quarter were about fifteen per cent under the peak
shipments of the first quarter; steel ingot production in the United States declined
less than five per cent in the same period.
The basic factors responsible for the decline in demand beginning in April have
already been indicated. In summary they were:
(1) Steel consumers had already placed orders for a large part of their
second and third quarter requirements through heavy advance buying during
the previous six months. Thus, even though their requirements had con-
tinued at a high level in the second and third quarters, they would have been
obliged to come into the market for but little additional tonnage.
(2) With consumers' real income and industrial production and profits
leveling off and beginning to decline in the second and third quarters, the
demand for consumers' and producers' durable goods rapidly feU off. Thus
steel requirements, both current and expected, were declining.
(3) In the previous boom months numerous steel consumers had optimis-
tically estimated that their steel requirements would continue to Increase in
the second and third quarters, and had placed orders on that basis. Thus,
when the expected increase in the orders for their finished products failed to
materialize and demand actually dropped off, they found themselves with
unnecessarily large inventories of steel on hand. (57)
(4) The general disappearance of expectation off further price inflation,
bolstered by the fact that after March there appeared to be a general stabiliza-
tion of steel prices, removed the price incentive for speculative forward
buying. (68)
As already pointed out, there is little reason to believe, and little evidence
to support the contention, that the decline in the demand for steel which
began in April 1937 was caused by the advances in the price of steel in March
and previous mouths. The decline would almost certainly have come at
about the time it did, even though no price increases at all had been made.
The changes in the other factors determining the demand for steel were too
great to be offset by any changes in steel prices.
D. recession: October, 1937 to june, 1938
By the end of September, 1937, the backlog of steel orders had been worked
off. (59) Thus, when general industrial activity collapsed abruptly early in
October, the demand for steel continued on its rapid downward pace; production
and shipments of steel by December had dropped precipitously to about one-third
of the September level in the case of production and 43% of the September level in
the case of shipments. (33)
The decline in general economic activity was most rapid; in the last quarter:
(i) Consumers' income (oh a seasonally adjusted basis) declined about
three per cent; (17)
(ii) Production of durable goods dropped about one-half; production of
non-durable goods, one-eighth; (6)
(iii) Industrial profits decreased about one-quarter; (S6) and
(iv) Stock prices fell one-fourth. (60)
Thus, the factors which were responsible for the spring and summer decrease in
demand for steel were greatly intensified, and new deflationary elements entered to
further the decline. The mistaken optimism of the winter and spring of 1937
turned to deep pessimism; advance buying and reduction of the 1930-1935
backlog of deferred purchases of durable producers' and consumers' goods turned
to postponement of buying. Producers of products made from steel, who in the
previous boom months had anticipated <iontinued rising sales, now found them-
selves with greatly excessive inventories of steel and steel products. (61) Before
a recovery in the demand for steel could be expected, these inventories had to
be reduced; in the meantime the decline in the demand for steel was doubly
aggravated.
Bookings, shipments, and production of steel continued downward until 'the
middle of June, 1938, except for slight increases in March and April. (33)
Meanwhile steel prices remained unchanged. (33) It has been contended that
a reduction of steel prices in the summer and fall of 1937 when demand was falling
off rapidly would have cushioned the decline. In view of the above aiialysis,
however, this seems very doubtful. The downward shift of the demand for
CONCENTRATION OF ECONOMIC POWER 13957
products made from steel and the inelasticity of their demand to price changes
were too great, and inventories of steel and steel products too excessive, to be
appreciably ofifset by any changes in the price of steel. Moreover, a reduction in
steel prices by leading producers might even have intensified the decUne in steel
buying by creating anticipations of further price decreases.
B. RECOVEBT: JUNE TO DECEMBER 1938
The decline of industrial activity, abrupt in the last quarter of 1937, continued
for the most part throughout the first half of 1938, but at a much slower pace,
especially in the second quarter. Slight — and possibly seasonal — increases in
bookings, shipments, and production . of steel occurred in March and April. (S5)
May, 1938, was the low point of the recession as measured by bookings of steel.
In the meantime the average level of steel prices remained practically un-
changed. (33) However, the period was marked by a growing hesitancy in regard
to the future course of steel prices, an increasing number of "rumors" of price
concessions, and the strengthening of expectations of price cuts. (62)
In the middle of February, a reduction of $4 a ton was made on cold-rolled
sheets in order to bring them into a more satisfactory relationship with hot-rolled
sheets. (63) N.o noticeable effect followed from the reduction except an intensified
uncertainty as to the future level of prices. (64)
Although steel producers in the latter part of February announced the continu-
ation of first quarter prices into the second quarter (with the exception of some
minor revisions in extras, and the recognition of the $4 break in the price of cold-
ro led auto sheets) , price uncertainty continued to grow. (65) In the second
quarter there were numerous rumors that price concessions were being made. (66)
This uncertainty undoubtedly caused a larger volume of postponed and hand-to-
mouth buy'ng than might otherwise have occurred.
In May, 1938, revisions in classifications of sheets and strip steel were announced
for the balance of the second quarter and the third quarter. (67) The reclassifi-
cations involved adjustments in extras for these individual products and were
accompanied by a reduction in the average price level of these products, brought
about by lowering the base price $2 per ton. (68) The announcements, however,
did not remove price uncertainty and rumors of price concessions. In the first
part of June, the price of galvanized sheet broke;$3 a ton, and further concessions
on the prices of other steel products were rumored to have followed. (68)
In the meantime, several important developments pointing to recovery of steel
demand were materializing.
(i) In the first place, the large inventories of steel and steel products
carried over from 1937 were being reduced to a level where steel consumers
would shortly have to enter the market in order to meet their current produc-
tion requirements. (69) Undoubtedly an important reason for the low-
demand for steel in the first half of 1938 was the fact that steel requirements
were being met from stocks in the hands of steel consumers.
(ii) In April, 1,938, the Federal Government announced a recovery program
consisting of:
(a) The desterilization of $1.4 billion of gold in the inactive fund;
(b) A 13>^ percent reduction of member bank reserve requirements by
the Federal Reserve Board; and
(c) A "four billion dollar pump priming" program. Beginning in
March, the excess of Government expenditures over receipts rose ra^Mdly.
The net contribution of the Federal Government to consumers' disposable
cash income rose from $221 million in the first quarter to $536 million in
the second, (14) and expectations were that the rapid rise would con-
tinue throughout the year. Moreover, the large public works program
announced in April promised large outlays in public construction in
the last six months of 1938.
(iii) In February, 1938, activity in residential construction began a rapid
upward climb. (^4) Residential construction contract awards (seasonally
V adjusted) in June were about 10 pier cent below the peak figure of the 1936-37
recovery. By the end of March, public works 'construction contract awards,
and the value of highway and grade crossing projects approved for construc-
tion, were also rising rapidly. (29) The outlook for heavy construction steels
was even better than it ha°d been in the 1936-37 boom.
(iv) Manufacturing production in the first and second quarters of 1938
leveled off after the rapid decline in the fourth quarter of 1937, and slight
seasonal increases occurred iif February, March, and April. (6) On a season-
13958 CONCENTRATION OF ECONOMIC POWER
ally adjusted basis, manufacturing production in May was only about three
per cent under the January figure. Production of non-durable goods (season-
ally adjusted) began to rise in May; some of its component items — such as
textiles and leather products — started to advance even earlier. With indus-
trial production thus leveling off, and with manufacturing inventories sub-
stantially reduced from the excessive height at the end of 1937, the outlook
for the last half of 1938 was an optimistic one.
(v) Profits of corporations in the second quarter of 1938 had risen above the
figure for the first quarter, so that the profit outlook was improving. (26)
'In the midst of rumored concessions in prices and uncertainty as to future prices,
Carnegie-Illinois Steel Corporation announced, on June 24, 1938, price reduc-
tions of $3 to $4 a ton on nearly all steel products, and eliminated the basing point
differentials at Chicago and Birmingham. (70) Other companies announced
similar price reductions and named new basing points in the major production
centers. On the average, announced steel prices were reduced eight to ten
per cent.
Undoubtedly, price concessions and uncertainty as to future prices had restric-
tive effects on buying in May and early June. The price announcements in the
latter part of June removed this uncertainty and served to bring out orders which
buyers had been holding pending the price announcements. This accounts, to
some extent, for the fact that bookings in May were about 35 per cent below the
average of March and April..
The price reductions came at the same time as the expected rise in industrial
activity began to materialize. Between June and December 1938:
(i) Consumers' real income (seasonally adjusted) rose over 5 per cent; (17)
(ii) Production of durable goods (seasonally adjusted) advanced over 80
per cent and regained 60 per cent of the recession loss; (6)
(iii) Production of non-durable goods (seasonally adjusted) rose 20 per
cent, recovering two-thirds of the recession decline. December was about
25 per cent above the low point in April (on a seasonally adjusted basis) ; (6)
(iv) The total value of construction contracts awarded (seasonally ad-
justed) almost doubled. (71) The December figure was approximately 50
per cent above the 1936-37 peak and within 16 per cent of the 1929 average;
and
(v) Profits of corporations more than doubled, recovering about one-half
of the recession drop. (S6)
With the resulting rapid upward shift in the demand for products made from
steel, the demand for steel- rapidly recovered. (33) Domestic bookings of sub-
sidiaries of United States Steel Corporation in the fourth quarter were two-thirds
above the recession low in the second quarter, but less than half the 1936-37 peak
figure in the first quarter of 1937. Steel ingot production in the United States rose
80 per cent from the second quarter to the fourth.
The fact that price reduction occurred at about the same time as steel orders
and general business activity began to improve has led some persons to believe that
the price reductions were primarily responsible for the recovery in steel demand.
Some have even contended that the steel price reductions were important factors
in the general rise of business activity in 1938. These views, however, are not
supported by the evidence. The basic factors responsible for the rise in steel
demand had begun to improve weeks and months in advance of the price reduc-
tions. Even if steel producers had not tit prices in June, 1938, steel buying
unquestionably would have followed about the same course as it actually did.
Steel prices held strong throughout the third quarter of 1938 except for a break
of $4 a ton on automobile sheets on July 29. (72) Although the price concessions
were temporarv, substantial tonnages were placed by the automobile companies
for their 1939 models.
In September, 1938 prices were reaffirmed for the fourth quarter but early in
October, when automobile producers came into the market with inquiries for large
tonnages, sheet and strip prices broke $4 a ton, followed by further concessions
of $2 to $4. Similar concessions were extended to other flat rolled products.
Although price concessions were withdrawn after a short time, all regular steel
consumers were given an opportunity to place orders at the low prices. (73)
Inasmuch as most buyers realized that the price reductions were temporary,
there occurred a large increase in bookings in October and November, J938. (74)
The effect of the temporary bargain market, however, was probably not to increase
CONCENTRATION OF ECONOMIC POWER 13959
the total buying of steel in the last quarter and the first few months of 1939, but
only to bunch purchases at the time of the price break. To the extent that
buyers took advantage of the price concessions, they reduced the volume of buying
in the following three or four months.
F. RECENT developments: JANUARY TO OCTOBER 1939
By November, 1938, business activity began to level off, and in the first two
quarters of 1939 industrial production and industrial profits declined slowly but
substantially while consumers' income remained virtually constant. (75) The
low point was reached in April and May, 1939, at a level about the same as that of
September, 1938. {76) Two important factors contributing to the decHne were:
(1) The nation-wide coal strike in April and May; and
(2) The great uncertainty regarding the foreign situation, especiaUy after
early March.
With the coal strike settled, and the foreign situation temporarily receding into
the background, business began to recover in May, and by the end of August had
reached the levels of November and December, 1938. {75) When war broke out
in Europe at the beginning of September, 1939, the advance was greatly acceler-
ated. Currently, activity in many lines of business is near the levels of the summer
of 1937. Undoubtedly a substantial part of the recovery since September 1st,
1939 has been of the speculative inventory-building-up variety, induced by fears
of rising prices and delivery delay.
The course of general business activity was closely paralleled by the demand for
steel. Seasonally adjusted bookings, shipments, and production declined until
May, 1939. {76) The decline was aggravated by the carry-over of steel pur-
chased during the price-concession period in the previous October. {77) Prices
of steel meanwhile continued steady, but with numerous signs of weakness by
April, which contributed to the hand-to-mouth buying of steel at that time. {78)
In the first part of May when automobile producers entered the market for
steel for their 1940 models, a sharp break in steel prices occurred, especially in
sheet and strip products. -{79) All buyers were given an opportunity to place
orders, and a large volume of purchases were made, at the low prices before the
concessions were withdrawn.
It seems likely that the price concessions of May, 1939, like those of October,
1938, served only to bunch steel orders at that time, without significantly affecting
the tonnage sales for 1939. Certainly the lowered price of steel has not had any
appreciable effect in lowering automobile prices. For steel producers, however,
such temporary price cuts, which do not significantly increase their total volume
of sales over a six of eight months period, can only mean decreased revenues.
With industrial production, consumers' income, and industrial profits rising
after May, 1939, the demand for steel recovered, and steel prices strengthened
somewhat. {80) Then, when war broke out on the first of September, bookings
bounded upward, and currently activity in the steel industry is almost at capacity
levels. Unquestionably a large part of the increase in the demand for steel since
the first of September has been caused by fear of rising steel prices and delivery
delay.
G. CONCLUSION
A careful examination of the record shows that changes in the demand for
steel during the 1936-1939 period were largely determined by the following factors:
1. The current and anticipated levels of business activity, income and
profits ;
2. The expectations with respect to steel prices in the immediate future as
compared with current steel prices;
3. The volume of steel inventory accumulated in the immediate past; and
4. The length of time required to fill new orders for steel.
Although price changes and anticipated price changes affect the timing of
steel purchases, the large fluctuations in the tot£l volume of steel production dur-
ing the 1936-1939 period cannot be attributed to changes in the levels of steel
prices. On the contrary, the evidence compels the conclusion that the influence
of the level of steel prices on the total consumption of steel was relatively unim-
portant.
13960
CONCENTRATION OF ECONOMIC POWER
Appendix
(1) Tin plate, the most important exception, is mainly used in the produc-
tion of perishable containers, i. e., cans which' are used a single time.
(S) Table 2. — Midyear population estirhates for continental United States
Year
Popalatlon
estimate
Percentage
Increase
Year
Population
estimate
Percentage
increase
1929
121,626,000
123,091,000
124,113,000
124,974,000
126,770,000
1934 .
126,626,000
127,621,000
128,429,000
129,252,000
130,216,000
0.68%
0 71
IMO
1.29%
0.83
0.69
0.64
1936
1931
J936 """
0 71
1932
1937
0 64
1933
1938 - ...
The population estimates are those of the U. S. Bureau, of the Census. See:
U. S. Department of Commerce, Bureau of the Census, Statistical Abstract of the
United States, 1938, (Washington, 1939), p. 10.
(5) Table 3. — Business gross capital formation, capital consumption, and net
capital formation, l'dS4 to 1938 (millions of 1929 dollars)
Year.
Flow of
producers'
durable
commodities
(1)
Business
coa^truction
(2)
Gross capital
formation
(l)+(2)
(3)
Business
capital con-
sumption
(4)
Net capital
formation
(3) -(4)
(5)
1924
4,838
6 368
6 761
6 993
6 083
6 891
6,791
4,012
2,601
2,779
3,811
4,683
6,080
6,987
6,243
3,408
4,026
4 325
4 467
4; 391
4,681
3,884
2,481
1,332
1,166
1,404
1,742
2,104
2,726
2,123
8,846
9,394
10,086
10,460
10,474
11 472
9,675
6,493
3,933
3,945
6,216
6,326
8,184
9,713
7,366
5,909
6,112
6,671
6,657
6,818
7,134
7,084
6,951
6,633
6,315
6,143
6,162
6,269
6,437
2 337
1926
3 282
1926
3,415
1927 '
1928 -
3,903
3,656
1929 :
1930.....
4 338
2,591
1931 :
-468
1932
-2,600
1933
-2, 370
1934
-928
1935
1936
1937
1938
Total 1924 to 1929
'
20,931
Total 1930 to 1936 .
—3,602
The above figures are estimates of Dr. Simon Kuznets of the National Bureau
of Economic Research. See: Simon Kuznets, National Income and Capital For-
mation, 1919-1985, (New York, 1937) pp. 40 and 48, and Simon Kuznets, Com-
modity Flow and Capital Formation in the Recent Recovery and Decline, 19S2-
1938, Bulletin 74 of the National Bureau of Economic Research, (New York,
June 25, 1939), p. 2.
The figures in columns (1) and (2) of the above table represent respectively
production of business equipment and business plant; column (3) is the sum of
columns (1) and (2). Cfolumn (4) is the total of depreciation, depletion, and
fire and marine loss of business plant and equipment. Column (5) is the excess
of column (3) over column (4) and represents the excess of production of business
plant and equipment over consumption.
CONCENTRATION OF ECONOMIC POWER
(4) Table 4. — Net earnings of corporations, 1919-1937
[millions of dollars]
13961
Year
Net earn-
ings
Year
Net earn-
ings
6.419
4,468
'65
4,380
5,867
4,998
6,971
6,774
5,880
7.666
1929
8,083
1930
1 366
1931 - ..-.
i 3 146
1922
1932
16.376
»2.379
167
1923
1933
1924
1934.
1925
1935
1,674
1926
1938
3,903
1927
1937
3,872
1928
1 =Deflcit.
The above profit estimates were based on federal income tax returns of all
reporting corporations. Net earnings are defined — for the years 1919 to 193& —
as net income less deficit of all reporting corporations less the total federal tax
plus tax exempt interest. For 1936 and 1937 net earnings are equal to net in-
come less deficit less the federal tax less dividends received from domestic cor-
porations plus wholly tax exempt interest. See: IT. S. Treasury Department,
Bureau of Internal Revenue, Statistics of Income for 1936, Part 2, pp. 24, and 47.
The figures for interest on tax-exempt obligations were obtained from Mr. Ed-
ward White of the Bureau of Internal Revenue in a letter dated July 14, 1939.
Figures for 1937 are from a preliminary release (Press Service, No. 18-55) of the
U. S. Treasury Department dated August 23, 1939.
(5) In addition to there being inadequate profit incentive there had also been
little necessity that business make capital expenditures in order to meet current
demands on productive capacity. In none of the six years 1930 to 1935 had
industrial production averaged more than four-fifths of the 1929 figure, and in
the low year, 1932, it had been only about one-half. See Table 5. Thus, in-
dustry's stock of plant and equipment was more than adequate to meet the
demands on it in these years. There was obviously no necessity for making
capital outlays to expand capacity, and the current level of operations was so
low that little replacement expenditures to maintain capacity were required.
(6) Table 5. — Federal Reserve Board Indexes of Production {1923-95=100)
Industrial Production
ally
adjusted
adjust-
ment
Manufacturing Production
(Combined Index
ally
adjusted
Durable Goods
Non-durable Goods
ally
adjusted
Monthly Average:
1929
1930
1031
1932.
1933
1934
1935.
1636
Jan
Feb
Mar
Apr
May
June
July
Aug
Sept...,
Oct
Nov
Dee
Monthly
13962 CONCENTRATION OF ECONOMIC POWIOU
(6) Table 5. — Federal Reserve Board Indexes of Production {1923-25 = 100) —
Continued
Industrial Production
Manufacturing Production
Season-
ally
adjusted
Without
seasonal
adjust-
ment
Combined Index
Durable Goods
Non-durable Goods
Season-
ally
adjusted
Without
seasonal
adjust-
ment
Season-
aUy
adjusted
Without
seasonal
adjust-
ment
Season-
ally
adjusted
Without
seasonal
adjust-
ment
1937
Jan
114
116
118
118
118
114
114
117
111
102
88
84
112
117
122
122
122
115
111
115
109
102
90
80
115
116
117
118
118
114
114
117
110
101
85
79
113
118
122
125
123
114
110
114
106
99
86
75
112
113
113
117
120
112
122
126
114
101
74
60
107
114
123
130
132
116
118
122
103
94
74
57
117
119
120
119
116
115
108
110
107
100
94
95
118
Feb
122
Mar
121
Apr. ..
120
May -...
116
113
July
104
107
Sept
109
Oct
103
Nov
97
Dec
90
Monthly Average..
110
110
109
109
107
107
110
110
1938
Jan
80
79
79
77
76
77
83
88
90
96
103
104
79
79
80
78
77
77
81
87
91
97
104
98
76
75
75
73
73
74
82
87
95
103
104
75
76
77
76
75
75
79
85
89
95
103
98
56
54
5*
53
51
50
58
64
69
83
94
92
53
54
57
58
56
53
58
63
66
79
92
85
93
94
93
91
93
95
102
108
107
106
110
114
94
Feb
95
Mar
94
91
May """ I"
93
94
July
97
104
109
Oct ...
109
113
Dec
108
Monthly Average..
86
86
84
84
65
65
100
100
1939
Jan
101
99
98
92
92
98
101
1102
99
99
100
95
94
98
97
■99
100
97
96
92
91
97
100
■104
98
98
100
96
94
97
95
'99
83
80
76
71
82
193
84
83
86
84
78
85
186
no
109
110
106
108
110
110
1 114
111
Feb
111
Mm"""
Apr
Apr.
106
107
June
108
July
106
Aug
1 Preliminary.
See: Survey of Current Business, 1938 Supplement, pp. 7-8, March
Reserve Bulletin, October, 1939, pp. 914-916.
pp. 14 and 19, and The Federal
(7) Table 6. — Consumers' Income and Outlay, 1929-1938
[Billions of dollars]
Consumers' income
Consumers' outlay
Year
(current
dollars)
(1929
dollars)
(current
dollars)
(1929
dollars)
1929
$78.6
73.4
63.1
49.6
45.9
62.2
,■16. 1
64.4
69.0
64.2
$78.6
76.0
72.4
63.8
61.4
65.8
76! 0
78.1
74.4
$73.3
69.1
56.3
44.1
42.3
49.7
52.2
58.9
62.6
$73.3
1930
70.6
1931 . . ..
62.9
1932
54.4
1933
55.3
CONCENTRATION OF ECONOMIC POWER
13963
The figures for consumers' income (in current dollars) are those of the U. S.
Department of Commerce for national income payments. See: Robert R.
Nathan, Income in the United States, 1929-1937 (U. S. Dept. of Commerce,
November 1938) pp. 29-30; and the Survey of Current Business, March 1939,
p. 19. The annual figures shown in Table 6 are sums of monthly figures reported
by the U. S. Dept. of Commerce. Consumers' income in 1929 dollars was ob-
tained by dividing consumers' income in current dollars by the National Indus-
trial Conference Board's index of the cost of living with base 1929 = 100. See:
Survey of Current Business, 1938 Supplement, p. 11, and March 1939, p. 20.
The consumers' outlay estimates are those of Dr. Simon Kuzneta of the Na-
tional Bureau of Economic Research. See: Simon Kuznets, National Income
and Capital Formation, 1919-1935 (New York, 1937) p. 85. The figures for 1934-
1937 were reported by Dr. Alvin Han=>en before the Temporary National Eco-
nomic Committee on May 16, 1939.
{8) For an exhaustive study of the demand for automobiles see: C. F. Roos
and Victor von Szeliski, "Factors Governing Changes in Domestic Automobile
Demand," in The Dynamics of Automobile Demand published by the General
Motors Corporation (New York, 1939).
(9) See: Roos and von Szeliski, op. cit. Chart 21 on page 60.
(10) See: Roos and von Szeliski, op. cit. Chart 15, p. 53.
{11) Table 7. — Consumers' Expenditures for Durable Household Goods, 1929-19S7
[MUlions of 1029 dollars]
Year
Consumers'
Expenditure
^^^oSfh^o^i^
Ooods
Year
Consumers'
Expenditure
for Durable
Household
Goods
1929
5,910
6,130
4,490
3,360
3.320
1934
4,010
4,670
- 8,900
6,330
1930
1931
1936
1932
1937
1933
These estimates are those of Mr. George Tofborgh of the Division of Research
and Statistics of the Board of Governors of the Federal Reserve System and re-
ported (June, 1938) in his memoranda on "The Prospects for Durable Goods.".
{12) Table 8. — Gross and Net Capital Formation Arising in Residential Con-
struction, 1924-1938
[Millions of 1929 dollars]
Year
Residential
Construction
Excess of Resi-
dential Con-
strueUon over
Consumption
Year
Residential
Construction
Excess of Resi-
dential Con-
struction over
Consumption
1924
4,589
.5, 218
4,757
4, 515
4,268
3,010
1,865
1.506
2,544
3,079
2,518
2,182
1,854
530
-646
-984
1932
600
648
695
1,193
1,965
2,193
1,949
-1,867
-1,899
1925
1933
1926
1934
-1,838
-1,214
1927
1935
1928
1939
1937
—226
1930
1938
1931
These are estimates of Dr. Simon Kuznets; see note (3) for source.
H— pt. 26 2.5
13Q64 CONCENTRATION OF ECONOMIC POWER
{IS) Table 9. — Indexes of Rent and Residential Construction Costs
[1929 = 100]
Monthly average:
1929 _
1933.... ..-.
1934
1935..
1036
Jan
Feb
Mar
June
July
Aug
Sept
Oct....
Nov
Dee
Monthly Average
1937
Jan
Feb
Mar
iKiV/.'.'.'.".'.'.'..'.'..
June -
July ,
Aug
Sept
Oct
Nov
Dec
Monthly Average
193S
Jan..
Feb
Mar
Apr..... ,
May
June
July
Aug
Sept....^
Oct
Nov
Dec
Monthly Average
1939
Jan..
Feb _. .
Mar...
tl&yV.y//.". :.'...:.
June
July
100.0
69.3
70.4
76.4
80.3
80.6
81.2
82.5
83.8
84.3
84.9
86.2
87.3
87.8
88.5
88.9
90.0
91.5
92.6
03.6
94.1
94.7
05.4
06.3
97.0
96.8
96.4
95.0
05.4
95.1
04.8
04.6
94.2
04.1
04.1
94.1
94.1
03.0
03.7
93.7
93.6
Construction Costs Indexes: Residences (E. H. Boeckh and Associates,
Inc.)
100.0
81.3
Si'.l
85.9
87.7
80.5
86.6
86.6
87.0
88.0
80.1
00.0
03.6
05.8
08.5
103.3
103.3
103.2
90.0
100.1
100.4
90.3
00.3
96.8
99.4
96.4
99.1
99.1
97.8
97.2
98.5
99. C
100.0
100.0
100.6
102 1
ino.8
99.3
99.3
100.1
99.3
99.3
100.6
101.4
101.1
New
York
100.0
04.6
73.0
67.9
60.6
70.4
70.4
71.2
71.2
71.9
71.9
72.6
72.6
72.6
72.7
73.2
74.4
75.8
76 6
76.5
76.8
83.4
84.9
85.1
66.5
84.1
8.3.6
83.0
81.8
85.1
84.3
84.3
84.5
84.7
84.6
85.4
85.8
86.6
86.8
85.5
85.5
85.3
85.3
85.5
86.0
86 0
Fran-
cisco
100.0
84.9
101.6
102.2
104.1
104.1
102.9
108.9
109.2
109.6
100.0
100.0
i(a<
113.4
108.6
110.6
112.2
112.2
116.2
116.2
123.1
122.5
116.2
117.7
116.2
116.2
116.6
116.6
116.7
116.7
116.7
116.7
116.7
116.7
116.6
118.1
118.1
118. 1
118.1
115.0
115.9
115.9
St.
Louis
100.0
70.6
80.4
83.7
82.6
8.3.8
84.0
84.0
85.5
85.5
85.0
84.0
84.0
88.1
88.0
80.2
89.8
80.7
01.3
90.6
90.1
92.3
91.6
90.9
90.3
89.2
91.6
91.6
91.6
91.4
90.5
90.0
92.1
92.1
93.4
93.1
92,2
100.0
81.1
88.7
85.6
83.8
85.9
87.0
87.0
86.4
84.7
84.2
84.2
84.8
85.3
87.8
94.1
96.1
98.0
103. 3
90.4
00.6
09.9
98.4
98.4
05.0
05.0
102.1
102.1
101.8
100.0
101.6
103.1
lot. 3
10^.3
103.0
106.1
104.3
102.4
103.5
102.4
102.4
103.9
105. 0
104..^
New
York
100.0
60.6
69.8
65.1
68.5
68.9
68.9
69.3
69.3
70.0
70.0
70.0
70.0
70.5
71.9
73.6
74.6
74.6
74.7
81.8
82.6
82.8
84.2
81.3
80.7
80.5
80.6
84.4
84.0
84.0
84.1
84.6
84.9
85.3
85.7
85.6
86.2
86.2
86.3
86.1
86.1
86.3
Si;. 7
86.7
86.8
San
Fran-
cisco
100.0
78.7
06.5
97.2
98.8
98.8
07.7
08.0
98.0
101.0
101.6
102.1
102.3
102.3
9.7
106.1
106.1
108.2
110.2
110.2
110.9
110.9
120.7
119 9
112.0
112.0
108.1
111. 3
112.4
114.5
112.4
112.4
112.1
112.1
112.2
112.2
112.2
112.2
112.2
112.2
112.4
113.6
113.6
113.6
113.6
113.6
113.6
113.8
St.
Louis
100.0
74.2
82.8
77.0
76.2
77.0
78.2
78.2
79.0
79. Q
78.3
77.2
77.2
77.7
78.4
78.4
81.6
82.6
82.9
83.3
83.2
84.3
83.5
82.9
85.6
84.6
83.0
82.9
83.8
87.6
87.6
87.6
87.3
87.?
87.8
89.3
89.3
90.5
91. S
91.8
96.4
96.4
96.0
06.0
96.0
95.6
94.4
The rent index is that of the National Industrial Conference Board with base
shifted from 1923 to 1929. See: Survey of Current Business, 193S Supplement,
p. 11, March 1939, p. 20 and September 1939, p. 20. The construction costs
indexes are those of E. H. Boeckh and Associates, Inc., and are based on actual
contractors' records of wages and material costs. The base of the Boeckh indexes
has been shifted to 1929. See: Survey of Current Business, 193S SuppleJtient,
p. 22, Ma.ch 1939, p. 22 and September 1939, p. 22.
CONCENTRATION OF ECONOMIC POWER
13965
(14) See: "Explanation of Method of Compiling Net Contribution," a memo-
randum of the Division of Research and Statistics of the Board of Governors of
the Federal Reserve System dated February 10, 1939.
(15) The gross federal debt outstanding at the end of March 1933 was 21.4
billion dollars. By the end of December 1935 it had increased 9.2 billions to a
total of 30.6 billion dollars. See: Survey of Current Business, 1936 Supplement,
p. 55, and 1938 Supplement, p. 65.
{16) The Standard Statistics Co., Inc. combined index of the prices of 420
industrial, railroad and public utility stocks stood at 64.6 (1926=100) in March,
1935. In March, 1936 the index stood at 108.7, ail increase of about 70 per cent
over March, 1935. See: Survey of Current Business, 1938 Supplement, p. 77.
(17) Table 10.
■Monthly indexes of consumers' income, 1936-1939
[1929=100]
Jan.-.- ' -
Feb
Mm- ..1 .
May - r.-
June..,-. - -..
July
Aug... ■.
Sept
Oct—..
Nov........ _
Dec
Monthly Average
1937
Jan
Feb :....
Mar..
Apr....
May
June
July. -
Aug
Sept
Oct
Nov
Dec...
Monthly Average
193S
Jan
Feb
Mar
Apr
May
June
July
Aug
Sept
Oct
Nov ,.:
Dec _
Monthly Average
1939
Jan.._ .
Feb....
Mar...
Apr
May*.
June
July _II.I.-..II
U. S. Dept. of Commerce
Index of Monthly In-
come Payments
Adjusted
for
variation
(1)
88.3
88.2
88.8
89.3
90.2
88.7
88.0
86.5
85.8
83.5
82.6
82.7
81.4
80.4
80.7
80.7
81.5
82.0
82.1
83.2
84.1
84.2
82.7
82.8
.83^.5
Unadjusted
seasonal
variation
76.9
72.1
76.2
77.9
75.1
92.1
86.3
75.9
83.3
84.6
79.2
86.6
83! 7
92.2
89.8
83.8
91.8
92.2
82.5
98.6
83.7
76.0
81.4
81.6
76.7
82.3
81.0
76.1
83.5
86.3
80.9
90.9
84.3
77.8
84.3
83.0
79.6
87.2
83.9
Cost of
Living
83.8
83.4
83.1
85.8
85.6
85.7
86.0
87.1
87.8
88.2
88.7
88.8
88.9
89! 4
88! 5
87.4
86.6
86.6
86.7
86.4
86.6
86.4
85.8
85.8
85.7
85.5
85.7
85.3
85.0
84.8
84.9
84.7
84.6
84.8
Real Consumers'
Income
Adjusted
for
seasonal
variation
(4)
90.2
91.0
92.3
92.4
92.7
107.3
101.9
95.9
95.0
96.4
9T.9
99.9
96.7
98.4
99.2
100.6
100.1
99.4
100.0
100.6
101.5
99.3
98.4
97.3
96.9
95.5
93.9
93.1
93.2
95! 0
95.6
95.8
97.3
98.1
98.2
99.3
97.4
97.8
98.7
Unadjusted
for
seasonal
variation
(6)
91.8
86.5
91.7
93.5
108! 4
101.4
88.8
97.1
101.3
94.0
116. 3
6.7
97.5
90.9
98.6
100.1
94.4
103.8
101.1
94.3
102.8
103.1
92.8
111.4
95.8
87.8
94.0
94.1
88.8
95.0
93.8
97! 3
100.7
94.6
106.1
91.5
99.4
97.8
94.0
103.1
98.9
13966 CONCENTRATION OF ECONOMIC POWER
Real consumers' income was obtained by dividing the indexes of monthly
income payments by the index of the cost of living. For the indexes of monthly
income payments see: Robert R. Nathan, Income in the United Slates, 1939-37
(Washington, November 1938) a bulletin of the U. S. Bureau of Foreign and
Domestic Commerce, and the Survey of Current Business, July 1939, p. 19, and
September 1939, p. 19. The cost of living index is that of the National Indus-
trial Conference Board with base shifted to 1929. See; Survey of Current Busi-
ness, 1938 Supplement, p. 11, March 1939, p. 20, and September 1939, p. 20.
Note.— The U. S. Department of Commerce inda^ of monthly income payments has recently been
revised. See Survey of Current Butineis, October, lfi39, pp. 15-10, and 19.
(18) Table 11. — U. S. Bureau of Labor Statistics indexes of wholesale prices,
1936-1939
[1926=100]
Date
Grains
Raw
Materials
Finished
Products
Semimanu-
factures
1936
Jan - —
78.9
78.3
75.6
73.9
70.6
73.0
102! 4
102.0
102.1
102.9
109.0
113.0
111.5
113.2
119.2
113.9
105.7
105.2
92.0
91.9
77.0
69.2
71.5
75.0
73.0
69.0
66.0
02.3
62.7
68.3
53.4
53.0
50.8
50.9
64.4
66.3
64.7
54.5
55.2
69.6
68.2
62.3
78.1
79.1
77.4
77.0
75.8
77.6
79.8
81.5
81.8
82.1
83.1
85.6
88.1
88.3
90.1
&8.7
87.1
86.1
86.5
84.8
84.4
80.7
77.2
75.4
74.9
73.6
73.2
71.3
70.7
71.4
72.3
71.4
72.0
70.9
71.6
70.9
70.9
70.9
70.1
68.5
68.9
67.7
67.8
82.4
82.2
81.3
81.6
80.5
SO. 7
81.6
82.4
82.3
82.0
82.6
83.8
84.9
85.4
86.4
87.4
87.5
87.7
88.8
89. 0
89.1
88.1
86.7
85.3
84.3
83.3
83.4
82.7
82.1
82.2
82.5
81.8
81.8
81.1
80.5
80.2
80.0
80.2
80.2
80.1
79.9
79.6
79.2
74.8
Feb
Mar .- -
74.4
74.6
£f£^:::::::::::::::::::::::::::::::::::::::::::::::::..
74.1
73.9
July -
75.2
74.6
75.9
Oct -
76.2
Nov -
78.6
Dec
82 3
1937
Jan .
85.4
Feb . . -
85.5
Mar . .. . :..:.—.: —
89.6
89.6
^^ :::::::::::::::::::::::::::::::::::::::::::::::::
87.6
June -■
86.8
July -
87.0
86.6
85.3
Ota.::::::::::::::::::::::::::::::::::::::::""::::":
82.6
Nov.. ...-
79.8
Dec
77.7
1938
Jan
76 9
Feb
76.1
Mar
75.6
Apr
76.3
Xy :: : :.
75.4
June . - .
74.1
July .
74.3
Aug . .
74.4
scm : : : .::::.. ..:
74.7
oepi. ................ ...... .^^^...
75.9
Nov -
76.2
75.2
1939
Jan
74.9
Feb
74.4
Mar -
74.6
Apr
74.4
ifii
74.3
June
74.1
July
74.4
I
See: Survey of Current Businett, 1938 Supplement, p. 12, March 1939, p. 20, and September
CONCENTRATION OF ECONOMIC POWER 13967
(19) The following statement appeared in The Iron Age, November 19, 1936,
p. 56:
"Most (automobile) companies have already made sales predictions for the
1937 model year, but if the piesent trend is continued these predictions will have
to be revised . . . and, if upward revision in sales budgets continues, the industry
may possibly eclips"e the 5,621,000 record number of units set in 1929."
See also: The Iron Age, Sept. 3, 1936, p. 53; Nov. 12, 1936, pp. 57 and 91; and
Jan. 7, 1937, pp. 73 and 98.
(SO) See: Roos and von Szeliski, op. cit., Chart 32, p. 74, and Chart 35, p. 81.
{21) See: The Iron Age, Nov. 12, 1936, p. 57; Nov. 19, 1936, p. 56; Dec. 10,
1936, p. 55; Dec. 17, 1936, p. 54; and Steel, Nov. 9, 1936, p. 33; and Nov. 16,
1936, p. 31.
{22) See statements in The Iron Age, Dec. 3, 1936, p. 62; Dec. 24, 1936, p. 47;
Dec. 31, 1936, p. 63, and Jan. 7, 1937, p. 99; and in Steel, Dec. 14, 1936, p. 24.
{23)
Table 12
Quarter
Billed Sales
of Electric
Ranges
($000)
(1)
Sales of Elec-
tric Household
Refrigerators
(number)
(2)
Vacuum Clean-
ers (Floor
Type) Ship-
ments of
(number)
:3)
1935
IV J
$3,542
4, 142
6,796
4.892
4, 852
6,941
9, 093
6,130
3,651
5,420
- 5,434
4,115
3,310
6.596
G, 359
166, 540
570,959
870, 600
392, 123
245,853
769, 705
936, 045
396, 137
267, 138
424,410
496,869
243,876
114, 047
600,280
803, 018
206, 206
274,818
1936
I
n
Ill :
IV --- -
325.611
352,956
368 590
li : :
III
282,261
281,409
277,436
' 228 139
IV
1938
I
li
lii
' 207,511
280,977
IV
1939
!...,„.
II
272, 202
(1) Compiled by the National Electrical Mamifucturers' Association from data furnishedby its i
Figures represent practically all of the output of electric ranges:
(2) Compiled by the Edison Electric Institute, b.Hsed on estimates of the National Electrical Manufaeturera'
Association covering reports of its members only.
(3) Compiled by the Vacuum Cleaners Manufacturers' ^iMociaiion, representing practically the entire
industry.
13968
CONCENTRATION OF ECONOMIC POWER
These three series are reported on monthly basis in ^^^^ Purvey of Current
Business, 1938 Supplement, p 141, March 1939, p. 51, and September 1939, p. 51.
(24) Table 13. -^Construction Activity Indexes
Federal Reserve Board Indexes of Value
of Construction ContracU Awarded
(1923-25=100)
Date
Adjusted for
seasonal variation
Unadjusted for
seasonal variation
Resi-
dential
Total
Resi-
dential
Total
87
117
87
117
1936
62
21
60
Jan - - - ---
26
52
22
45
Feb -
26
47
28
47
Mar.. - -—
30
47
35
53
fe;:;;:::::;:::;:::;:::;:;::::;::::::::::::::::::::;::::::::
32
46
52
38
39
56
60
Juhe .— -
59
46
65
July
62
46
65
Aug
59
47
60
Sept - - ---
43
57
41
54
Oct - - '
40
68
39
51
Nov -
Dec - <•■
45
66
38
53
37
65
37
56
1937
45
63
37
51
Jan - -
47
62
42
54
Feb - - ;-
66
47
56
Mar -- - -
44
63
51
61
i^:::::::::::::::::::::;:::::::::::::;;:::::;;::":::::::::
44
42
56
61
62
47
72
June - - -
44
67
45
75
July... ^
40
62
40
66
Aug ,
37
56
37
56
Sept
52
35
49
Oct - - :-- -
56
31
50
g^ :::::::::::::::::::::::::::::::::::::::::::: ""--
30
61
25
49
41
59
52
41
22
59
1938
42
Jan -
32
51
28
44
Feb - - - -
33
46
35
46
Mar.. - r
37
52
43
59
37
51
54
44
46
61
63
June - - -
49
59
49
65
July -
66
62
69
Aug..... - -
56
78
56
79
Sept
67
82
56
78
Oct - -
96
54
85
Nov
57
96
48
77
44
64
44
45
64
1939
86
70
Jan
73
61
63
Feb -
6J
6C
69
Mar -
5i
6
«
76
^:::::::;;:;:::::::;:;::::;:::v::::::::;::::::;::::::::::
6.
6.
i 6,
J 65
) 64
■73
June -
e
2 6
J 6!
73
July
1 'T
J '6-
177
Aug
1
T^
I Preliminary.
CONCENTRATION OF ECONOMIC POWER
13969
The Federal Reserve Board indexes are based on reports of the F. W. Dodge
Corporation for 37 states east of the Rocky Mountains. See: Survey of Current
Business, 1938 Supplement, p. 16, and March 1939, p. 21
from The Federal Reserve Bulletin, October 1939, p. 914.
(^5) Tablk 14. — Railroad activity
Data for
Federal Reserve
Board Indexes of
Freight Car Load-
ings (Combined
Index) (1923-25=100)
Adjusted
for
seasonal
variation
(1)
Unad-
justed
for
seasonal
variation
(2)
Freight
Car
Surplus
(OOO's
of cars)
Revenue
Carried
One Mile
(Class I
Railways)
(000,000's)
Net
Railway
Operating
Income
(Class I
Railways)
($000)
Monthly Average.
1S35
Mar.
Apr.
Api
May.
June..
July..
Aug..
Sept. .
Oct...
Nov..
Dec...
Monthly Average..
Jan..
Feb..
Mar.
May.
June.
July.
Aug..
Sept.
Oct..
Nov..
Dec.
Monthly Average.
Jan..
Feb..
Mar.,
Apr..
May.
June.
July..
Aug..
Sept. .
Oct-.
Nov..
Dec...
Monthly Average.
Jan..
Feb..
Mar..
Apr..
May.
June.
July..
61
1,491
1,341
1,370
1,385
1,377
1,593
1,709
1,855
1,659
1,475
1,436
1,787
21, 935
26,296
38, 130
34, 709
39,599
34, 103
26, 919
42, 157
57, 349
75, 455
54,224
46,021
41,408
2,030
1,797
1,921
1,856
1,902
2,164
2,438
2,429
2,200
1,977
1,817
2,127
1,981
1,648
1,64»
1,712
2,118
35,729
33.562
35, 152
41, 493
41, 797
60,259
61, 722
64,637
70,096
89,809
72,377
70,606
55,595
38,867
38,784
69,881
48,358
44,239
59,354
60,558
50,308
59,305
60,747
32,441
25,972
7,144
' 2, 122
14, 470
9,23T
16,497
26,001
38,387
' Deficit.
(1) and (2). Computed by the Federal Reserve Board from weekly data repofted by the Attceialwn of
American Railroads.
(3). Compiled by the Association of American Railroads, Car Service Dkiition. Data cover Class I rail-
roads and represent a daily average for the last half of the month.
(4), (5) and (6). Compiled by the Interstate Commerce Commission.
13970 CONCENTRATION OP ECONOMIC POWER
(25) Table 14. — Railroad activity — Continued
Federal Reserve
Board Indexes of
Freight Car Load-
ings (Combined
Index) (1923-25=100)
Freight
Car
Surplus
(OOO's
of cars)
(3)
Revenue
Passengers
Carried
One Mile
(Class I
Railways)
(000,000's)
(4)
Net
Railway
Operating
Income
(Class I
Railways)
($000)
(6)
Net
Income
(Class I
Railways)
($010)
(6)
Date
Adjuited
for ^
seasonal
variation
(1)
Unad-
justed
for
seasonal
variation
(2)
1938
Auk
62
64
68
69
69
63
71
75
70
64
229
169
144
175
221
1,976
l!662
1,664
1,028
45, 377
50,362
68,566
49, 665
49, 373
1,097
Sent
6,277
24 068
Oct :""
Nov
Dec
C2
C2
257
1,803
30,996
I 10, 338
1939
Jan
69
67
66
60
62
67
69
63
62
63
68
62
b7
70
218
209
202
265
211
175
166
1,790
1,555
1,618
1,681
1,725
32, 891
18, 591
34,317
15,267
25, 101
39,096
49. 012
' 8, 721
Feb
Mar
127,896
May
1 18, 594
' 1,685
July
These series are all reported in the Survey of Current Business, 19S8 Supplement,
pp. 84-86, March 1939, pp. 37-38, July 1939, p. 38, and September 1939, p. 37.
{S6) TABLfi 15.— Corporation profits*
Quarter
Federal Re-
serve Bank of
New York (168
industrials)
(millions of
dollars)
Standard Statistics Co.,
Inc., (119 industrials)
(1926 = 100)
Seasonally
adjusted
Seasonally
unadjusted
I
1936
$107. 9
147.6
124.7
199.8
170.0
269.2
217.2
283.5
248.3
309. 0
262. 4
194.9
74.0
84.4
61. S
187.0
153.3
49.7
55.6
61.5
92.7
7.S.2
96.0
83.6
130.7
109.6
113.5
104.4
92.8
38.7
35.8
32.2
79.0
'■<<. 7
46.9
II
62.6
Ill
54.9
IV -
81.3
I
1936
71.0
II
108.1
in
89.2
IV :
114.6
1937
I
103.6
11 .. .
T27.8
Ill
111.4
IV
81.4
I
1)38
36.6
II
40.4
Ill
34.4
IV :
69. S
I
1939
65. 0
li
1 ..■ ; 1 08.7
• Data arc for net Income after payment of Qxed char^Ci; ,:ud inxes. All of the abovn series arc rcpor'o) !u
the StiTvey of Current Butinest, 193S Supplement, p. 64, ^ Ir.r.^h 1039, p. 32, July 1939, p. 32, and September
1939, p. 32. See also Note 4.
' Preliminary.
(Jdqtuiiu)
CONCENTRATION OF
(jaqnmn)
[ s8Ai}oinojOT[ pBOJijea
ECONOMIC POWER
13971
(jaqmnn)
•-* W O COfOQOO
(000$) (I (Joop
-jno pno aoopaj) jnaui
nU'
(000$)
II oa poB OY (d 'H
OOZ-I) sjoiojM D|J?09i5i
CC00>0aCs»M»OO
?5SS§?3?
1 CO Oi C^ Ol CO <o
(000'000$)
,1 spooo iBOijpoja
01 (000$)
iUaajqoBj^ i^ni^jJOMpooAl
t^ « 02 C^ <0 lO u
(000$)
t ^jBjoa pas iBSnju^nao
j^coojQ -^ys
I SOOS^C^CoSoa^coScDas
(001 = 9261
jS^namdiqg 9^bj3av
/Iiq^noiv) ijodxg pnc
ousemoQ siooj, en;qoBi\r
OOOJOeO ^cs o
o cm' »o ic CO CO o r--' CO to t^ t^
iiiii
(00I=^S-K;6IiS?n9m
-diqg oSbjoav iSiqauoj^)
jaeradinba iijpnnoj
J2?5^S
(000$)
t S9UBJO pB9qJ9AO OIJ!)09ia
2iS?Jss|
(000$) t jnamdmba pas
;_5SSJoCSSss
(snox-?Joqg)
, aiBici I99;g pg^BOjjqBj
ooco«-H^ccMocoas=o-
gWOOMMO
(000$)i3njAi9qsi8a)g
ISBBil
O'OCi00T}<Q0-«*'00c0^OC
OCOM coc
(000$)
jo s.oOO) isieipa pajg
?S§?
SSSSSS
,S<!;S5.5<)-So;^o
:§ S
13972
(j^qmna)
SJ80 jaSnessBj pbojijbh
CONCENTRATION OF ECONOMIC POWER
W00«0 I g
(jaqmnn)
saAjpmoooT; peoJireH
(jaquma)
sreo iqapii peoairea
|§S||
(000$) (-"OOP
-}no puB joopni) jnaca
-djnba SniqoiiAiS jaAiOj
;sg8
ss:sis;s
iSi^i;
(000*)
Oa PnB OV id H
ooz-T) SJ050JV oij^ooia
r* lO cD«c o
(OOO'OOOJ)
spoof) [B0IJ}08[a
((100$)
yfj.'UiqOBJV aai^JOAipoo^
i^^^^Sl^issIs
SSSSSS:??
(000$)
Xjbioh paB iB3nju}nao
jrfAioj 'lUBaig :!.'duin<i
(00l=9Z6t I
S}n,»ni(liqg BSvio&v I
Xiqinojv) JJodxg puB
oijSdmoci sioox anigoBpy
•o o 00 >o ri r^ ec
(OOI = K;-ZZ61 s?nam
-diqS e3BjeAV ^ii{luoj^)
luomdsnba Ajpunoa
iiisi
t^ O -^ OS O Cfl »C CO 0
t~ O) — f- » to 1^ 00 t
(000$)
ssuBJo pBaqidAQ oix)09|a
!Sc3SSf5
ssiiggg
(000$) )ua[ndmba ptre
scnajsjis 3uinotJipnoo-Jiv
1 1 s 1 1 § §_S_i § i I
(suoj,-)joqg)
0}BId 1884S pe^BOuqBj
3J5?5
•>J'00OC>-Hdt-O'Ot~N«
(000$) anjAieqs l3»iS
3S3g
§8SS8g
(000$)
ojtuinjtia 90UJO 13«>S
iiiiiiS§i§gi
(j; -bs
JO b.ooo) sjaiioa laaiS
-iQO!Ct~ ooor
CONCENTRATION OF ECONOMIC POWER 13973
:2 »£■ » <» o -g ■Oo- SS S2 S •:; S -2
-a _-^ -- -r ^- g ^.^ £g g . . S a
^2 x« x= ^ ^ I ^5 if- I ^ ^ _:
ll If I II I li ^= = i - '^
•o£ OS c o « " fe §a fflS t; w ifl „ 3
il if i i i § !" ^U ' H II
^s l| I ^ ^ is^fi :H.a s 1^1 a ll
SIS S'^ S 2 ^ S^^ti Bt g sfeS g fcS
ga Sg S % S 6n2>.g' g_ o.sog o. "g"
iilii ili i P^i 1| I |il|i ii
6^-25 s aie d |»°|' •§! 1 ^-sg-ti of
as sal pa a ^pi s:? .§ s-^.a^^ i|
£^5323 £a£ £ ceS.^ ot^^- Sa2:2- ->
cS^'oS ■oS'c Tj 5"22g- "^^ds a§-o2.3 £§-
t-Sa5 a ° B a !:^„a -"S.S 22na Saj"?
«i|Qo fio02a'S|~?^s-s§S;^2fflOa1 »=3
-^i"i|"rrr-^=|-ia-|-i=i=|=r
13974
CONCENTRATION OF ECONOMIC POWER
(28) See comments in The Iron Age, Nov. 19, 1936, p. 113; Dec. 3, 1936, p.
118: Dec. 10, 1936, p. 114; Dec. 24, 1936, p. 85; and Dec. 31, 1936, p. 89; and in
Steel, Dec. 7, 1936, p. 114.
{S9) Table 17. — Business and Public Construction
Date
Value of Construction Contracts Awarded ($000)
Value of Highway
and Grade
Crossing Projects
Approved for
Construction
(Federal Funds)
($000)
Com-
mercial
buildings
Factory
buildings
Public
utUitles
Public
buildings
Public
works
Highway
Grade
crossings
Monthly Average:
1929
$77, 432
61,360
25, 925
10,227
8,281
12, 650
21)386
3; 624
10, 626
9,673
$43, 712
.58, 393
24,639
6,300
8,600
10, 616
$10, 065
11,651
15, 106
9,832
4,242
4,639
$77, 871
80,251
72,954
41! 626
52,087
$16, 667
241441
41,073
35. 172
44.273
52, 448
1933
1934
1936
Jan
10, 810
9,207
12,202
15, 197
13,919
15,021
15, 821
17, 326
13, 653.
16, 622
12,826
11, 976
7,065
7,761
6,496
6,284
9,815
9,468
14,564
10, 649
6,002
12.032
8,854
9,169
8,707
3, 885
6,475
7,319
5,419
9,146
13,810
4.422
12.493
11,198
10, 694
18, 104
4,132
4,843
6,991
3,079
6,404
7,672
3,190
14,188
13, 547
10, 931
6,764
16, 046
35.699
23,933
30, 779
33, 170
25,967
29,991
40,083
05,118
63, 653
75,117
69, 645
76, 387
58,302
60,076
52, 083
41,088
33, 746
26.208
24, 996
33, 711
47, 679
61,824
77, 361
78, 936
Feb
Mar
u^i :::::::::::::::
July .-
$1, 776
se^::::::::::::::::::::::::::
4,365
Oct
12.039
Nov
26.933
Dec
34, 972
13,707
9,013
9,306
8,149
48, 212
49, 659
1936
Jan
" 16, 650
12, 568
17, 261
24, 272
18, 785
21,910
28,641
21,963
20,065
21,989
22,986
23,156
8,956
13, 437
18,411
25,546
12, 895
10, 213
19, 140
14,980
18,838
18, 392
14,075
23,139
17,926
11, 939
18, 106
23, 753
12, 773
9,264
27, 512
17,946
15, 735
14,171
18,029
19,117
12, 496
4.342
8,971
6,264
13, 706
14, 669
6,657
11.246
6.730
7,991
5,041
4.207
68,948
34, 694
44, 191
49, 660
60. 792
70,717
99, 103
76,435
68.767
52. 861
65.839
42, 135
72, 030
80, 125
82.599
73. 596
62,963
53. 090
50.400
60, 476
49.123
46. 103
42,093
42,090
35. 973
Feb
39. 022
Mar
40.283
Apr
42.011
May
40, 561
July
34, 081
29.026
26, 576
oi?t.v::::::::::::::::::::::::::
23.615
Nov
20.233
Dec
17, 971
20,761
16,502
17, 189
8,517
59, 612
58, 724
1937
Jan
21, 463
22,295
30,007
28,540
26,610
24,488
29,112
29,590
25,333
18,' 855
16,613
37.028
12, 609
22, 248
30, 051
18, 539
36. 822
58, 601
37, 875
12.934
12, 640
13,504
20, 940
21, 788
32,364
20,256
20,985
10, 763
29, 863
49,992
27, 455
12,949
15, 602
17, 420
18, 286
6,533
4. 952
11,090
8,163
9,220
10,827
11, 255
9,982
6, 597
9,179
9,798
8,305
46,664
27,264
32, 221
44, 757
55,980
70, 064
52, 501
63, 103
62,873
45, 982
43, 983
46, 475
43,899
44,472
46, 856
47. 081
48,569
49,263
43,417
40,606
39.849
39. 112
39, 781
41,683
16 037
Feb
13.526
Mm "
12,859
Apr
13,688
May.. . .
16,730
July
12:323
Aug
11,761
Sept "■"'"
12, 713
Oct
Nov
10, 731
Dec
10, 443
24, 754
26, 141
23, 144
8,742
48.489
43, 716
12,902
CONCENTRATION OF ECONOMIC POWER 13975
(89) Table 17. — Business and Public Construction— Continued
Date
Value of Construction Contracts Awarded ($000)
•
Value of Highway
and Grade
Crossing Projects
Approved for
Construction
(Federal Funds)
Com-
mercial
buildings
Factory
buildings
Public
utUities
Public
buildings
Public
works
Highway
42, 149
41, 407
40, 636
44,072
51, 158
51, 299
53,137
48,958
43, 373
38, 572
36,231
37, 677
Grade
crossings
1938
Jan
•15. 40
•13. 04
•20.20
•18.94
•19. 17
•18. 79
•28.15
•18. 33
•13.96
•24.17
•13. 72
•13.96
• 6. 59
• 4.92
•15.68
•11.47
• 8.65
•10. 68
• 9.69
•11.31
•10. 74
•13.79
•10.53
• 7.03
48, 451
5,149
10,694
9,373
43, 699
8; 689
13, 431
37,980
26, 167
21, 176
19, 726
44,312
50,125
26, 333
49,005
57, 631
78, 533
74, 832
65,827
88, 113
83, 162
92, 829
70, 692
114,075
10, 433
Feb
11,392
Mar
13, 577
May " :"■
:::::::;
July
12,836
11,416
serrti' ::::::::::: :::::::
12, 561
Oct :.::::....::
12,112
Nov
13. 930
Dec
12,794
Monthly Average
•17.99
•10.09
24. 071
70, 846
44,057
12,188
1939
Jan
29,509
18. 518
19,640
35, 336
21, 779
9,968
23,092
67,002
53, 115
.58,010
85,633
76, 141
73,607
78,960
36,294
35,968
34,969
35,600
40,769
41,024
37, 802
13, 572
Feb
13,613
Mar
12,906
Apr
12, 107
Sfay
10, 224
11.312
July
.
11,504
• In millions of Dollars.
The series showipg value of construction contracts awarded are compiled
by the F. W. Dodge Corporation and cover 37 states east of the Rocky Mountains.
The coverage of the data varies somewhat from year to year. See the 1938
Supplement of the Survey of Current Business, p. 169, for description of the series.
Figures for highways and grade crossing projects are those of the U. S. Depart-
ment of Agriculture, Bureau of Public Roads, and are condition figures as of the
end of the month for highway work administered by the Bureau.
All of the series of Table 17 are reported in the Survey of Current Business,
1938 Supplement, pp. 16-20; March, 1939, pp. 21-22 and September, 1939, pp.
21-22 and Standard Trade and Securities, Current Statistics, July 14, 1939, p. 21
(Published by the Standard Statistics Co., Inc.)
(30) See table 1, and Table 19 in note 38.
(Sl) Throughout July and August expectations of near future price advances
were becoming strong. See the statements in The Iron Age, July 30, 1936, p.
55; August 6, 1936, p. 71; Aug. 13, 1936, p. 89; Aug. 20, 1936, p. 73'; and Aug. 27,
1936, pp. 71 and 79.
(32) See the comments in The Iron Age, Mav 28, 1936, pp. 84 and 86; June
11, 1936, pp. 80, 82, 85, 88 and 90; and June 18. 1936, pp. 51 and 72. The follow-
ing statement appeared on page 102 of the June 25, 1936 issue:
"Of even more practical significance is the fact that fully half of the steel tonnage
booked during June has been prompted by price advances."
(33) See Table 1.
(34-) See Table 1, and note 38.. See also the statements in The Iron Age,
for Sept. 10, 1936, p. 93; Sept. 17, 1936, p. 93; Sept. 24, 1936, p. 75; Oct. 1, 1936,
p. 73; Oct. 22, 1936, p. 71; and Oct. 29, 1936, pp. 67-A and 70.. The price ad-
vances early in September were effective for the fourth quarter. Advance
announcement of the price increases undoubtedly caused some protective forward
buying in the last half of September.
(35) See note 18, Table 11.
13976 CONCP]NTRATION OF ECONOMIC POWER
(36) Table 18. — Composite Price of No. 1 Heavy Melting Steel Scrap
[Dollars per gross ton]
Date
1936
1937
1938
1939
13.47
. 14.12
14.75
14.69
13.39
12.81
13.29
15.04
16.45
16. es
16.31
17.10
18.33
19.27
21.25
21.02
18 54
17.28
18.79
20.43
18.73
15.89
13.34
13.46
14.00
13.86
13.40
12.40
11.54
11.32
13.29
14.51
14.34
14.21
14.74
14.88
Feb;;v:::v:::::::::::::::::::::;::: ::::::::: ::::;
Mar :: :. ... . :::.
Apr ^
May
July...-
Aug .
15 46
Sept
19 25
ort": :::::::;::::::: :::::::::::::::::::::::::::::::::::::::
Nov
Dec
Monthly Average
14.76
18.03
13.54
■
The above steel scrap composite price is that computed by The Iron Age,
and is an average of Pittsburgh, Chicago, and Philadelphia quotations. See:
The Iron Age, January 5, 1939, p. 205. The figures for 1939 are for the middle
week of the respective months, and were obtained from current issues of The
Iron Age.
(37) Employee Representation Plans had been set up in 1934, 1935 and
1936 in many of the leading steel plants. In the summer and autumn of 1936,
the number of employees enrolled in these labor organizations had grown greatly.
Negotiations concerned with the question of wages Were carried on between
steel employers and E. R. P. representatives during most of the summer and
fall. These negotiations and the possibilities of future labor difficulties led to
widespread expectations of a rise in steel wage rates. The developments of the
summer and autumn of 1936, can be followed in statements in The Iron Age,
Aug. 6, 1936, p. 71; Oct. 22, 1936, p. 71; Oct. 29, 1936, pp. 67-A and 70; Nov.
12, 1936, pp. '2-78 and 91; Nov. 19, 1936, p. 76, and other issues from July
to Novemoer.
(38) Appendix T,able 19 below shows the announced prices and announced
eflFective dates at Pittsburgh and Chicago of several steel products of United
States Steel Corporation.
See also: Table 1, and the comments in The Iron Age, Nov. 26, 1936. dh. 77,
88; Dec. 3, 1936, p. 105; and Dec. 17, 1936, p. 85.
CONCENTRATION OF ECONOMIC POWER 13977
Table 19. — Announced Prices with Effective Dates (Subsidiaries of United States
Steel Corporation)
PITTSBURGH
Date
An-
nounced
Date
Effec-
tive
For
Ship-
ments
to
Struc-
tural
Shapes
Plates
Soft
Steel
Bars
Hot
Rolled
Sheets
Hot RoUed
Annealed
Sheets
24 Qa.
Galva-
nized
Sheets
24
Ga.
Hot
Rolled
Strip
Tin
Plate
Steel"
Billets
St'd
Steel
Rails
100#
100#
100#
100#
100#
100#
100#
Ba.'?e
Box
100#
Gross
Tons
Gross
Tons
12- 1-38
12- 1-38
12- 1-38
12- 1-38
1- 1-39
9-20-38
6-24-38
5-18-38
2-18-38
1- 1-38
10-15-37
8-5-37
7-20-37
7- 1-37
5-12-37
4- 6-37
3- 5-37
3- 5-37
12- 1-36
1- 1-37
1- 1-37
12- 1-36
10- 1-36
9- 5-36
5-26-36
5-23-36
4- 1-36
3-12-36
1- 1-36
12- 2-35
3-31-39
6-30-39
3-31-39
12-31-38
9-30-38
9-30-38
6-30-38
9-30-38
3-31-38
12-31-37
12-31-37
$2.10
$2.10
$2.25
$2.15
Not Sold
as Such
After May
18. 1938
$3.60
$2.15
$34.00
11-10-38
$5.00
::: ::
""5.'35
9-20-38
6-24-38
2.10
2.10
2.25
2.25
2.10
2.10
2.25
2.25
2.25
2.25
2.45
2.45
2.15
2.15
2.30
10 Oa.
Base
2.40
3.50
3.50
3.80
3.80
2.15
2.15
2.30
2.40
34.00
34.00
37.00
37.00
40.00
5-18-38
42 50
2-18-38
$3.15
42.50
10-15-37
8-5-37
7-20-37
2.26
2.25
2.25
2.25
2.25
2.25
2.45
2.45
2.45
2.40
2.40
3.15
3.15
3.80
3.80
2.40
2.40
2.40
37.00
37.00
37.00
42.50
42.50
7- 1-37
2.40
2.40
3.15
3.15
3.80
3.80
5-12-37
4- 6-37
9-30-37
9-30-37
2.25
2.25
2.45
2.40
■5.'35
37.66
42.50
3- 5-37
42.50
3- 5-37
U-24-36
6-30-37
3-31-37
3-31-37
^30-37
^31-37
12-31-36
12-31-36
9-30-36
9-30-36
6-30-36
6-30-36
3-31-36
3-31-36
2.25
2.05
2.25
2.05
2.45
2.20
2.40
3.15
3.80
2.40
2.15
37.00
34.00
1 1-24-36
2.15
2.80
3.40
11-18-36
4.85
12- 1-36
30.00
1.95
1.95
1.95
2.60
2.60
2.50
3.20
"T26
^5-36
5-26-36
5-23-36
1.90
1.90
1.90
1.90
1.90
1.90
2.05
1.95
1.95
1.95
32.00
30.00
3-17-36
1.85
1.85
28.00
29.00
3-12-36
1.80
1.80
1-14-36
1.85
1.85
11-27-35
1.80
1.80
1.80
1.80
Prior Pri
36 Preva
ling.._.
1.85
1.85
2.40
3.10
1.85
5.25
29.00
36. 37W
12- 1-38
12- 1-38
12- 1-38
12- 1-38
1- 1-39
9-24-38
9-23-38
9-20-38
8-31-38
6-24-38
5-18-38
2-18-38
1-1-38
10-15-37
8- 5-37
7- 1-37
5-12-37
4- 6-37
3-5-37
3-5-37
1- 1-37
12- 1-36
1- 1-37
1- 1-37
10- 1-36
9- 5-36
5-26-36
5-23-36
12-31-35
12- 2-35
1- 1-36
3-31-39
6-30-39
9-30-39
12-31-38
2.10
2.10
2.25
2.15
Not Sold
as Such
After May
18, 1938
3.50
2.15
5.00
34.00
40.00
40.00
11-10-38
5.00
2.10
2.15
1
9-20-38
12-31-38
9^30^8
9-30-38
6-30-38
9-30-38
3-31-38
12-31-37
'6^30^37
y-30-37
6^30^37
3^31-37
9-3C-37
9-30-37
12-31-36
12-31-36
9-30-38
9-30-36
3.50
2.15
5.35
34.00
2.10
2.10
2.30
2.30
2.10
2.10
2.30
2.30
6-24-38
2.15
2.40
10 Ga.
Base
2.50
3.50
2.15
2.40
6.35
34.00
37.00
37.00
5-18-38
42.50
2-18-38
3.25
3.90
42.50
12- 3-37
5.45
10-15 37
2.30
2.30
2.30
2.30
2.30
2.30
2.30
2.30
2.50
2.50
2.50
2.50
3.25
3.25
3.26
3.25
3.90
3.90
3.90
3.90
37.66
37.00
37.00
37.00
42. -50
8- 5-37
7-20-37
5-12-37
4- 6-37
2.50
2.60
2.60
"5.' 45
""5.' 45
42.50
42.50
42.50
2.30
2.30
3- 5-37
2.50
3.25
3.90
37.00
2.10
2.10
12- 1-36
39.00
11-24-36
2.10
2.10
2.25
2.96
3.50
2.25
11-18-36
4.95
10- 1-36
2.05
2.05
2.05
2.70
2.70
2.60
3.30
9- 5-36
1.95
1.95
1.95
1.95
1.95
1.95
2.10
2.00
2.05
2.05
5-26-36
5-23-36
1.90
1.95
29.00
29.00
11-26-35
9-30-36
5,35
5.35
Prior Pri
ce Preva
iling..._
1.85
1.85
1.90
i.95
2.50
3.20
1.95
27.00
36. 37 ^^
13978 CONCENTRATION OF ECONOMIC POWER
(39) See the comments in The Iron Age, Nov. 26, 1936, p. 77; Dec. 3, 1936,
pp. 62, 105; Dec. 10, 1936, pp. 93, 96; Dec. 17, 1936, p. 85; and Jan. 7, 1937, pp.
66, 99, 183. The gist of these statements i^ that bookings in December, 1936
were probably the largest monthly bookings in peace-time steel history, and that
the major part of the orders were placed as protection against announced Jan-
uary 1, 1937 price advances.
(40) See Table 1. Peak bookings in 1929 totalled 1,590,793 tons in March.
The December, 1936 total was exceeded in only three months of United States
Steel Corporation history: In November, 1912 with 2,062,939 tons, in March,
1916 with 2,064,472 tons, and in November, 1916 with 2,357,161 tons.
(41) See notes 20, 23, 27 and 29.
(45) Delivery promises of the Carnegie-Illinois Steel Corporation on most sheet
mill products in December, 1936 averaged about 3K to 4>^ months. That such
conditions were general in the industry may be seen from statements in The Iron
Age, Dec. 3, 1936, p. 97; Dec. 10, 1936, p. 93; Dec. 17, 1936, p. 89; and Dec. 31,
1936, p. 67, where it is pointed out that backlogs on some products extended
completely through the first quarter of 1937.
(43) See the remarks verifying the statement in The Iron Age, Dec. 3, 1936,
p. 62; Dec. 10, 1936, p. 96; Dec. 24, 1936, p. 47; fend Jan. 7, 1937, pp. 66, 99, and
183, and other sources cited in note 39. See also notes 22, 40, 57 and 61.
(44) Steel prices are ordinarily changed only at the beginning of each quarter,
with the announcements of changes coming usually in' the last month of the
preceding quarter. See also the comments in The Iron Age, Feb. 25, 1937, p. 91.
(46) See Table 16 in note 27, note 21 and the statement in The Iron Age, Feb.
25, 1937, p. 114.
(46) See Table 1, and Appendix Table 19, in note 38. See also the remarks in
The Iron Age, March 11, 1937, pp. Ill, 114; April 1, 1937, p. 95, and April 15,
1937, p. 89.
(47) See notes 20 and 23, and Survey of Current Business, 1938 Supplement, p.
141.
(48) See note 27. See also The Iron Age, Mar. 11, 1«37, p. 133; Apr. 1, 1937,
p. 118; and May 6, 1937, p. 116.
(49) The delivery situation is summarized in statements in The Iron Age,
Mar. 25, 1937, p. 85; April 15, 1937, p. 89; and in Steel, Mar. 15, 1937, p. 103;
Mar. 22, 1937, p. 75; and Mar. 29, 1937, p. 83.
(60) See note 17. If Veterans' Bonus payments in June 1936 are taken into
account, consumers' real income in March, 1937 was almost seven percent below
the level of the previous June (on a seasonally adjusted basis).
(61) Although there are no monthly figures on business outlays for capital
equipment, rough interpolation of column (1) Table 2, note 2, by the series in
Table 16, note 27, indicates that business capital equipment expenditures in the
period March, 1936 to March, 1937 were roughly half again as great as in 1935
and approximately equal to 1929 figure.
(52) The U. S. Bureau of Labor Statistics estimated that in December 1936
there were 258 industrial disputes in progress, in March 1937, 760. Man days
idle due to labor disputes for the same period increased from 2.1 million to 3.3
million. See: Survey of Current Business, 1938 Supplement, p. 39.
(63) See: Survey of Current Business, 1938 Supplement, p. 85.
(64) For example, see the comments in The Annalist, March 26, 1937, p. 482;
April 9, 1937, p. 554; April 16, 1937, pp. 604, 608; April 23, 1937, p. 650; April 30,
1937, pp. 681 , 682; May 7, 1937, p. 722. The fall of security and basic commodity
prices which began in February to April, 1937, led to the belief that the speculative
part of the boom was coming to an end, and that prices of securities a)id basic
commodities were going to adjust to a more stable position.
(66) See: Survey of Current Business, 1938 Supplement, pp. 56, 75 and 77.
(66) See Table 1. Total domestic bookings of United States Steel Corporation
in the six consecutive peak months of 1929 amounted to 8,215,225 tone.
(67) No data are available showing changes in inventories of steel in the hands
of buyers (producers of products made from steel). That inventories were
becoming excessive, however, is borne out by the following statements:
After stating that steel consumers were building up inventories of finished
products and work-in-process beyond the level required by current sales The Iron
Agexn its April 22, 1937 issue, page 90, commented:
"This leads steel sellers to the conclusion that we arc approaching a time when
ultimate consumers must be given time to digest a surplus."
CONCENTRATION OF ECONOMIC POWER
13979
Throughout the spring and summer similar statements pointing to the growth
of excessive inventories appeared in increasing volume and emphasis. See The
Iron Age, May 6, 1937, p. 103 (Some stove manufacturers have enough steel to
meet all of their remaining 1937 requirements.); May 20, 1937, p. 93 (Steel
jobbers are trying to reduce their heavy stocks.) and p. 84 (Large steel consumers
are well taken care of owing to large stocks on hand and heavy orders placed but
not yet received.) ; June 3, 1937, pp. 97, 98; June 10, 1937, p. 103; July 22, 1937, p.
109; Aug. 12, 1937, p. 107; Aug. 19, 1937, p. 91; Aug. 26, 1937, p. 87; Sept. 2, 1937,
p. 88 (". . . fresh business is undoubtedly being affected by the presence of
greater than expected inventories at consumers' plants."); Sept. 9, 1937, p. 107;
Sept. 16, 1937, p. 105 (Steel consumers are liquidating their stocks before placing
replenishment orders.); Sept. 23, 1937, p. 101 (Inventories almost universally
•are excessive; with sales of products made from steel slowing down, liquidation of
heavy steel inventories is proceeding slowly.).
Table 20 below shows the value of inventories of materials, supplies, work-in-
process, and other inventories excluding finished products in the hands of various
steel consumers on January 1 and December 31, 1937 as reported by the U. S.
Bureau of Census (See: U. S. Department of Commerce, Bureau of the Census:
Census of Manufactures, 1987, preliminary release dated April 15, 1939, with title
"Inventories, January 1 and December 31, 1937").
Although inventories of steel are not reported separately, it is doubtful if they
varied significantly from the broad picture shown by the figures below.
Table 20. — Value of Inventories of Materials, Work in Process, Etc. in the Hands
o/ Various Steel Consumers January 1, 1937 and December 31, 1937
Industry
Value of Inventories of Mate-
rials, Work In Process, etc;
($000)
Jan. 1, Dec. 31, Per cent
1937 1937 Increase
Agricultural Implements (including tractors)
Aircraft and parts — -
Boiler-shop products
Bolts, nuts, washers, and rivets made in plants not operated in connection
with rolling mlllB •
Cars, electric and steam railroad not built in railroad repair shops
Cash registers, adding and calculating machines,- and other business ma-
chines except typewriters
Cranes, and dredging, excavating and road building machinery
Cutlery (not including silver and plated cutlery) and edge tools.
Electrical machinery, apparatus, and supplies.
Engines, turbines, water wheels, and windmills
Locomotives, railroad, mining, and industrial not made in railroad repair
shops
Machinery, not elsewhere classified
Machine-shop products -
Machine tools — —
Motor vehicles not including motorcycles
Motor- vehicle bodies and motor- vehicle parts —
Refrigerators and refrigerating and ice-making apparatus
Ship and boat building, steel and wooden, including repair work
Stamped and pressed metal products; enameling, japanning, and lacquer-
ing
Steel barrels, kegs, and drums. _--
Structural and ornamental metal work, made in plants not operated in
connection with rolling mills
Textile machinery and parts. _
Tin cans and other tinware, not elsewhere classified
Tools, not including edge tools, machine tools, files, or saws
Washing machines, wringers, driers and ironing machines for household
use
Wire drawn from purchased rods
Wirework not elsewhere classified
Foundry products (gray-iron and malleable iron).
Fnrgings iron and steel, made in plants not operated In connection with
steel works or rolling mills
Hardware not elsewhere classified
Heating and cooking apparatus except electric
Machine-tool accessories and machinists' precision tools
Pumps (hand and power), pumping eQuipment and air compressors..
$101, 472
24, 968
17,958
9,080
26,930
21,400
17, 770
8,872
187, 236
31, 839
11,947
110, 919
62, 135
34, 673
"173, 139
125, 804
26,020
24,962
27, 019
6,178
31, 749
13,758
49, 593
12,764
6,115
16,015
15,640
21, 577
12,304
27, 161
51,096
14, 163
21, 508
$112,388
43,348
21, 941
37, 163
27, 307
24, 895
11,455
253, 693
40,260
21, 053
135, 485
71,817
44,551
214, 808
128,211
32, 765
31,091
31, 345
0,814
16, 251
76,963
■15, 488
16,936
17,460
22, 946
13,522
30, 740
61, 484
16, 235
27,567
11%
74
22
124491— 41— pt. 26-
13980
CONCENTRATION OF ECONOMIC POWER
(58) See Table 1. In the last week of April steel producers announced the
continuation of second quarter prices through the third quarter. The Iron Age,
April 29, 1937, p. 99, made the following comment:
"The principal incentive for forward buying having been removed by the an-
nouncements of a continuation of present steel prices through the third quarter,
an easier situation prevails in the steel market."
See also: The Iron Age, April 22, 1937, p. 94; April,^9, 1937, pp. 103, 105, 106,
108; May 6, 1937, p. 116; May 13, 1937, p. 140; July 15, 1937, p. 95 (No price
advances expected for fourth quarter.).
As early as the middle of July, third quarter prices for most steel products were
reaffirmed for tlie fourth quarter. (See The Iron Age, July 22, 1937, p. ^9).
This completely removed the price incentive for forward buying.
(59) By the end of September delivery promises of the Carnegie-Illinois Steel
Corporation, for example, had dropped from the March, 1937 peak of six to seven'
months for most sheet mill products to two to four weeks, which is normal when
there are no mill backlogs.
(60) The Standard Statistics Co. Inc., index of prices of 348 industrial stocks
stood at 126.2 (1926=100) in September 1937, and at 95.2 in December 1937.
See The Survey of Cxirrent Business, 1938 Supplement, p. 77.
(61) There are no series available showing separately (1) inventories of steel in
the hands of consumers, and (2) inventories of finished and semi-finished products
made from steel in the hands of producers. However, this limitation is not serious;
excessive inventories of finished and semi-finished products made from steel, as
effectively reduce the demand for steel as excessive inventories (in the hands of
steel consumers) of steel itself.
Table 21 below shows the year end value of inventories of various steel con-
suming industries. These inventories include raw materials (including steel), fuel,
supplies, work-in-process, finished products, etc. These figures are the revised
estimates of Dun and Bradstreet, Inc. based on their Surveys of Inventory Trends.
The mid-year estimates for 1939 are from Dun's Review, Sept. 1939, pp. 8, 10
and 12.
Table 21. — Inventories 1935-1938 for Selected Manufacturing, Wholesaling and
Retailing Lines
Manufacturing:
Iron and Steel Products
Hardware -.-■.- -
Foundries '.
Electrical Apparatus and Appliances
Machine Shop Products
Agricultural Machinery
Automobiles
Automobile Accessories and Parts
Wholesaling:
Hardware
Automobile Equipment
Plumbing and'Heating Supplies.
Machinery and Erjuipment
Electrical Goods and Appliances
Retailing:
Farm Im plements
Radio, Electric and Gas Household Appli
ances...
Hardware
Hardware and Farm Implements
Motor Vehicle Dealers...
Automobile Accessories (Independents)
Automobile Accessories (Chains)
1935 Year
1936 Year
1937 Year
1938 Year
End In-
End In-
End In-
End In-
ventories
ventories
ventories
($000)
($000)
($000)
(.$000)
$89, 093
$111,366
.$131,412
$109,072
203, 838
238, 491
290,959
261,863
34, 584
329, 773
395, 728
557, 976
423,504
210,678
143, 689
172, 427 '-
227, 604
185, 497
310, 559
366, 460
436, 087
326, 193
197, 704
257, 015
341, 830
266, 969
120, 328
139, 581
140, 977
134,210
99,730
114,690
125,012
122,012
57,019
62, 151
67, 745
63,816
68,692
79,683
94, 823
90, 556
59, 659
80,540
99,870
77, 899
34,080
46, 272
53, 676
63, 569
73,407
73, 935
92,419
8.3,065
196, 562
202, 022
218. 184
214, 475
44, 937
49,913
55, 902
55. 119
399,685
398, 267
561, 557
463, 846
29,843
32,299
36, 498
37, 702
29, 843
33, 461
37,811
37,055
$116,000
266,000
418,000
172,000
240, 000
229,000
139,000
127,000
67,000
92,000
86,000
52,000
89,000
221,000
56,000
5.35,000
40,000
46,000
See also note 57 and the source quoted there, and The Iron Age, May U, 1939, pp. 57, 58, 58- A. and 58-B
for other inventory figures. The inventory situation as it affected the demand for steel is also discussed in
The Iron Age, Oct. 7, 1937, p. 97 and p. 105 (One fact stands out prominently— stocks in the hands of many
consumers and jobbers are large enough to last for many weeks at the current rate of consumption.); Oct.
28, 1937, p. 91; Nov. 11, 1937, p. 113 (Airto companies have a large carryover of steel not used in the produc-
tion of 1937 models.); and p. 119 (Consumption of stocks going on slowly.); Jan. 20, 1938, p. 93; Feb. 24,
a 1938, p. 56 (.\uto producers still have large inventories of steel.).
CONCENTRATION OF ECONOMIC POWER 13981
{62) See the comments in The Iron Age, Jan. 27, 1938, p. 75; Feb. 3, 1938, p. 91;
Feb. 10, 1938, p. 97; Feb. 17, 1938, pp. 98, 99; Feb. 24, 1938, p. 85; May 5, 1938,
pp. 86-87; May 12, 1938, p. 106; May 19, 1938, p. 68; May 26, 1938, p. 95; June
2, 1938, p. 92; June 23, 1938, pp. 72-75; and Jan. 5, 1939, pp. 73, 106, 110.
{63) See: The Iron Age, Feb. 17, 1938, p. 54.
(64) See: The Iron Age, Feb. 17, 1938, pp. 54, 98-99; and Feb. 24, 1936, p. 85.
{65) See: The Iron Age, Feb. 24, 1938, p. 85.
{66) See: The Iron Age, May 5, 1938, pp. 86-87; May 12, 1938, p. 106; May
19, 1938, p. 68; June 2, 1938, p. 92.
{67) See: The Iron Age, May 26, 1938, pp. 88, 95.
{68) See: The Iron Age, June 16, 1938, p. 81, and June 23, 1938, pp. 72, 75.
{69) See the statements in The Iron Age, Mar. 10, 1938, p. 95; June 16, 1938,
.p. 81; Jan. 5, 1939, pp. 73, 106; July 21, 1938, p. 70. See also note 61.
{70) See especially the article by C. E. Wright, "Basing Point Changes May
Alter Industrial Map of the United States," in The Iron Age, July 7, 1938, pp.
84-D, et seq., and in the same journal, June 30, 1938, pp. 60-63. See also: July
7, 1938, p. 86; July 14, 1938, p. 80; July 21, 1938, p. 70.
{71) See notes 24 and 29.
{72) See: The Iron Age, Aug. 11, 1938, pp. 42, 64.
(75) See: The Iron Age, Oct. 13, 1938, pp. 306, 308-309; Oct. 20, 1938, p. 66;
Jan. 5, 1939, pp. 110, 176.
{74) See Table 1, and the comment in The Iron Age, Oct. 20, 1938, p. 66, and
Jan. 5, 1939, pp. 110, 176.
(75) See Tables 5, 10, 14 and 16.
{76) See Table 1, and the comments in The Iron Age, Jan. 5, 1939, p. 221; Jan.
26, 1939, pp. 74 and 76; Feb. 2, 1939, p. 96; Feb. 23, 1939, p. 76; Mar. 2, 1939,
pp. 70 and 84; Mar. 23, 1939, p. 82; April 13, 1939, p. 94.
{77) See, for example, the statements in The Iron Age, Feb. 23, 1939, p. 76;
Mar. 2, 1939, p. 70; and Mar. 30, 1939, pp. 46-48.
{78) See The Iron Age, April 13, 1939, p. 94; April 27, 1939, p. 88; and May
4, 1939, p. 98.
{79) See the discussion in The Iron Age, May 11, 1939, pp. 96 and 98; May 18,
1939, pp. 124 and 126-127; May 25, 1939, pp. 61-62; and June 15, 1939, pp. 69
et seq.
{80) See Table 1, and Tfie Iron Age, issues from June to August, 1939.
Exhibit No. 1413
AN ANjClYSIS OF THE DEMAND FOR STEEL IN THE AUTOMOBILE
INDUSTRY
This is an analysis prepared by the Special Economic Research Section of United
States Steel Corporation, composed of Messrs. Edward T. Dickinson, Jr., Ernest
M. Doblin, H. Gregg Lewis, Jacob L. Mosak, Mandal R. Segal, Dwight B. Yntema
and Miss Marion W. Worthing. The work of this group was under the supervision
of Theodore O. Yntema, Professor of Statistics, University of Chicago. This
analysis was written by Jacob L. Mosak and has had the benefit of suggestions
from other members of the staff. It is issued by United States Steel Corporation
November 1, 1939.
contents
I. Purpose.
II. Summary of the Findings.
III. The Automobile Industry.
IV. Steel Consumption.
V. The Demand for Automobiles:
A. Major J'actors Aflectins the Year to Year Changes in Sales of New Automobiles
(1) Consumer Income.
(2) Potential New Owners and the Maximum Ownership Level.
(3) Replacement Demand.
(4) The Price of Automobiles.
B. Secondary Factors .\ffectiiig Retail Sales
(1) Used Car Allowances.
(2) Financing Terms.
(3) Operating Costs.
(4) Dealers' Used Car Stocks.
VI. The Relation of Steel Prices to the Demand for Automotive Steel.
I. Purpose
The purpose of this study is to review ttie factors which determine the de-
mand for steel as 'a raw material in the automobile industry, and to appraise
particularly the relative significance of the price of steel as jne of these factors.
13982 CONCENTRATION OF ECONOMIC POWER
II. Summary of the Findings
The quantity of steel consumed as a raw material by the automobile manu-
facturing industry depends directly upon (1) the number of cars produced; (2)
the quantity of steel used per car; and (3) the production of replacement parts
and related equipment. The price of automotive steel can afifect the quantity
sold only by influencing one or more of these three factors.
The demand for automobiles has recently been the subject of an exhaustive
study by C. F. Roos and Victor von Szeliski.i They found that the number of
new passenger cars sold in any year was dependent on (1) national income; (2)
the number of cars in operation; (3) the age distribution of cars in operation;
(4) the scrapping rate; (5) the price; and (6) other factors including used car
allowances, financing terms, operating costs and dealers' used car stocks. After
making proper allowance for the influence of national income, number of cars in
operation, age distribution and scrapping rate, they concluded that the elasticity
of demand for new passenger cars was approximately 1.5; or in other words, that
a 1% reduction in the price of cars would cause a 1.5% increase in the number of
cars sold.
A change in the price of steel could affect the number of automobiles sold only
if it were passed on to the ultimate consumer as a reduction in the price of cars.
The cost of steel in a car is, however, only approximately 13% of the f. o. b. price,
or approximately lO% of the delivered price of the car. A reduction of 10% in
the price of automotive steel would therefore permit a reduction of only 1% in
the delivered price of the car if it were passed on entirely to the consumer. Since
this 1% reduction in automobile prices would give rise to only a 1.5% increase
in the number of cars sold, it follows that a 10% reduction in the price of steel
would give rise to an increased consumption of steel of but 1.5% through its eflFect
on the number of cars sold. Thus changes in the price of steel are far from effec-
tive in raising the consumption of steel through increasing the volume of car
sales.
From the evidence available, the price of steel appears to have an even less
significant effect on the quantity of steel used per car. The increased use of
steel per car over the past fifteen years is mainly attributable to the increased
popularity of enclosed models and heavier cars, and intense competition among
motor companies in the improvement of their product. The conclusion that the
price of steel is a minor influence in the quantity of steel consumed per car is
supported by the absence of a relationship between the weight of passenger cars
and the price of steel after the series have been adjusted for secular trends.
Replacement parts and accessories have a value only about one-seventh as
great as new cars produced. It seems reasonable to conclude that the elasticity
of demand for these products is not greatly different from the elasticity of de-
mand for new cars. But even if p>arts and accessories had an appreciably higher
elasticity of demand, because of their relatively small value they could not have
much effect on the elasticity of demand for all automotive steel.
The total effect of a reduction in price upon the consumption of steel by the
automobile industry, through increase in the volume of car production, increase
in the use of steel per car, and increase in the production of replacement parts
and accessories, is therefore relatively slight. Combining these elements, the
elasticity of demand for automobile steel is probably not in excess of .2 or .3.
In other words, a 10% reduction in the price of such steel could not increase the
consumption of steel in the motor vehicle industry by more than 2 or 3%.
The price of steel is therefore a minor factor in determining the quantity of steel
consumed as a raw material in the automobile industry.
III. The Automobile Industry
A brief review of the growth of the automobile industry to its present-day
position is ttesic to an understanding of the nature of the demand for automobiles
and of the derived demand for steel. From a stage of infancy in 1900 when pro-
duction amounted to only 4,000 vehicles, the manufacture of automobiles rapidly
matured into an industry of first importance with a production of 4 million ve-
hicles in 1923. Between 1923 and 1929 production averaged about 4.4 million
vehicles annually, of which more than 85% were passenger cars. During the
same period the dollar volume of production averaged nearly 3 billions annually,
of which about 2.5 billions represented passenger car production. In the recent
peak year, 1" T7, production stood at 5 million motor vehicles having a whole-
sale value of 2,970 million dollars. Of this amount about 80% represented pas-
' C. F. Roos and Victor von SzeliskI "Factors Oovernlng Changes iu Domestic Automobile Demand",
The Dynamics of AtUormbUit Demand, General Motors Corporation, N. Y., 1939.
CONCENTRATION OF ECONOMIC POWER
13983
senger cars.^ In the census year 1935 the industry gave employment to 164,000
people and had a total wage bill of 217 million dollars.^
Complementary to the motor vehicle industry is the important motor vehicle
bodies and parts industry. In 1935 this industry employed 60% more wage
earners and had a 50% larger wage bill than did the motor vehicle manufacturing
industry proper. If the two industries are added together the total number of
employees in the automotive industry in 1935 was 425,000 and the total wage
bill was 545 miUion dollars.^
However the dollar volume of output in the two industries cannot be added
together, since that would involve duplication. It has been estimated that
between 75 and 80% of the production of motor bodies and parts ordinarily
represents original equipment entering into the production of new cars. If this
portion of output is subtracted from the total it appears that the dollar output of
Chart 1
PRODUCTION OF AUTOMOBILES AND OF PARTS AND ACCESSORIES,
AND TOTAL NUMBER OF CARS REGISTERED IN UNITED STATES
140
1926
= 100
120
REGISTRATIONS
k.
100
^,^^
n^
,^
/
^
\'*"'
'"y
^-^
\
80
>
r
V
/
\
1
//
Y'*
.PARTS AND ACCESSORIES
60
/^
f
\
-\™o - -
UCIION
1
A
^
NJ ^
^
40
Z'
t
UTOMC
BIcN
r
KUDUC
TICN
\,
20
^
,0-^
0
60 X
UJ
o
z
40 -
20
: Rae^'ch and Auto^ob-le Manijf»r(u'''>g Aaocai^n
replacement parts at producers' prices averaged only 287 million dollars in the
census years, 1923, 1925, 1927 and 1929. If in addition there are added certain
other accessories such as skid chains, stamped automotive parts and storage
batteries, the total production at producers' prices of replacement parts and ac-
cessories averaged 407 million dollars.^
It has been estimated that in 1938 the total employment offered directly and
indirectly in motor vehicle bodies and parts, tires and petroleum refining plants,
sales and servicing, road building, truck and bus driving, in the raw material
supply industries and in automobile insurance and finance companies was well
over 6 million persons." This is about 14% of the total gainfully employed in the
United States. It has been further estimated that approximately 14 billion dollars
of new investment went into the furnishing of motor transport directly and in-
directly in the period 1923 to 1929.' This is about equal to the amount that was
> statistical Abstract of the United Slates, 193R, page 370.
» United States Department of Commerce, Census of Manujactures, 1935, page]29.
* Ibid.
' S. Kuznets, Commodity Flow and Capital Formation, National Bureau of Economic Research, New York,
1938, pages 66 and 88.
• Automobile Facts and Figures, Automobile Manufacturers Association, 1938, page 47.
' Wendell D. Hance, "The Role of the Automobile Industry", paper presented before the American Eco-
nomic Association, December 28, 1938, typescript, pages 3-5. See also. Automobile Facts and Fitures, Auto-
mobile Manufacturers Association, 1938, page 51.
13984
CONCENTRATION OF ECONOMIC POWER
invested during the same period in railroads, electric power and telephone com-
panies combined. 5
Like other durable goods industries, however, the automotive industry has
not experienced uninterrupted gro\vth but has been subject to great fluctuations in
production and employment. This is shown by Chart 1 and by the figures on
production, employment, and payrolls for selected years given in Table 1 :
As is to be expected, the fluctuations in the consumers' stocks of automobiles,
as evidenced by automobile registrations, are much smaller in amplitude than the
fluctuations in production (Chart 1). This follows directly from the fact that
the automobile is a durable good. Once bought, it yields services over a period of
five to fifteen years, and no further purchases are necessary, except for replacement.
Table 1.
-Production, Employment and Payrolls in the Motor Vehicle Industry for
Selected Years
Year
Produc-
tion (mil-
lions of
cars)
Bureau of
Labor Sta-
tistics In-
dex of Em-
ployment,
1923-5 = 100
(monthly
average)
Bureau of
Labor Sta-
tistics In-
dex of Pay-
rolls, 1923-
5=100
(monthly
average)
1921... . .....
1.7
5.6
1.4
5.0
52.9
111.3
60.5
128.3
1929 . .
1932
1937..
Sources: Statistical Abstract of the United States, 1938, p. 370.
ment, pp. 32 and 41, and November, 1938. pp. 14, 18.
Survey of Current Business, 1938 Supple-
Replacement purchases, however, can be postponed, since the old car can almost
always be made to serve another year if economic conditions are adverse. This
difference in fluctuation is clearly seen' by comparing the figures in Table 2 on
total number of cars registered and on gasoline consumption with those shown
above on production and employment in Table 1.
Table 2. — Total Number of Car§ Registered and Gasoline Consumption for Selected
Years
Total Num-
ber of Cars
Registered
(millions)
Gasoline
Consump-
tion (mil-
lion barrels)
1921.
1929.'
1932.
10.5
26.5
24.1
29.7
31.0
31.2
13.2
Sources: Automobile Manufacturers Association, Automobile Facts and Figures, 1938, pp. 15 and 16, and
1929, p. 14; Survey of Current Business, 1938 Supplement, p. 121.
Since most of the production of motor bodies and parts represents original
equipment entering into the production of new cars, it is natural that the two
series should have similar fluctuations, as shown in Table 3.
Table 3. — Value of product in the aulomohile industry for selected years
[Millions of dollars]
Year
Motor
Vehicles
Bodies and
Parts
1929 .
$3, 710
1.568
1,097
2,391
$1. 551
1931... . . ..
945
1933 . .
756
1935 ■.
1,551
Source: Census of Manufactures, 1935, pp. 1180 and 1156.
' Charts and Tables for Use in Hearings on Savings and Inreslment before the Temporary National Economic
Committee May 16 to iG, 1939, Investment Banking Section, Securities Exchange Commission. Supplementary
Table I-a, "Plant and Equipment Expenditures", introduced by Lauchlin Currie, M&i' 13, 1939.
CONCENTRATION OF ECONOMIC POWER
13985
It appears, however, that even if that portion of the production of all accessories
and parts which is used for replacement purposes is segregated, the fluctuations
in this segregated series still conform to the changes in automobile production
rather than to those in the total number of cars registered.' This indicates that
the factors affecting the demand for replacement parts are roughly the same as
those affecting the demand for automobiles.
IV. Steel Consumption
The automotive industry has been the largest single consumer of steel for five
of the last six years, taking between one-quarter and one-sixth of the total finished
steel output. For the period 1923-1938 it consumed on the average almost 5
million tons of steel per year.'" About 125 different kinds of steel are used in the
modern automobile. Chief among these are sheets. The automobile industry
in 1937 took about 45% of total production of sheets, and in addition it used
Chart 2
AUTOMOBILE STEEL CONSUMPTION AND AUTOMOBILE PRODUCTION
IN UNITED STATES
CO
Z c
O =
i
8
7
6
4 d
■s.
2
1
0
;
1
V^CONSUMPirON
A (GROSS IONS)
/
y
/"
\
1
/
/
\
J
\
>
/
\
A
y
^^
u
kJ
\
/
s\
/
y
./
V
(
\
\
/
/
\;/
PRODUCT
ON
iilllllllliillllii
about 55% of the total production of strip and 45% of the total production of
bars.'i
Large quantities of steel are bought annually for the production of automobile
accessories and parts for replacement. As has already been noted, the annual
production of these parts amounts to about one-seventh of the value of passenger
cars produced. In addition, steel is bought for the production of tools and dies,
for repairs and maintenance and for capital investment in plant and equipment
of the automobile industry. '^
The latest available estimates of automotive steel consumption show that the
steel shipments to the automotive industry have fluctuated in fairly close con-
formity with automobile production. From Chart 2 it appears that in every
year 1923-1938, except in 1929, the two series moved in the same direction,
although, it should be noted, the changes were not proportional to each other.'^
• See Table 8 in the Appendix.
'« See Table 9 in the Appendix.
" Automobile Facts and Figures, 1938, page 47.
■2 The automobile industry is, of course, also responsible for a great deal of steel consumption in the build-
ing of highways, bridges, filfing stations, and plant and equipment in the petroleiun and rubber tires indus-
tries. This indirect steel consumption is not dealt with here.
'3 See Table 9 in the Appendix. The available data on steel consumption by the automobile industry
are not entirely satisfactory, since the coverage and classification of steel purchases in the magazine Steel
differ from those in Iron Age, and the coverage and classification differ from year to year in the sarpe maga-
zine. However, for the purposes for which the data are used here these limitations are not prohibitive.
13986 CONCENTRATION OF ECONOMIC POWER
An important feature of automotive steel consumption has been the ever
continuous tendency to increase the use of steel in automobile construction. '<
The increased use of steel has resulted from the larger size of modern cars, the
adoption of all-steel tops and bodies by automobile producers and the trend
toward heavier cars and enclosed models. From 1925 to 1930 the percentage
of open model cars decreased from 30% to 0.5%. '^ In recent years, however, this
tendency has been somewhat offset by the use of greater glass area and of plastics.
Experiments are now being conducted to determine the practicability of using
plastics even for automobile bodies. If successful, this may have an appreciable
effect on automotive steel consumption.
V. The Demand for Automobiles
The authoritative study of the nature of the demand for automobiles made by
Messrs. Roos and von Szeliski *' relates to retail sales of passenger cars within
the United States for the years 1919-38, and does not include commercial cars
and trucks, nor exports of any vehicles. This, however, is not a serious limita-
tion, for, as already noted, passenger-car production represents about 80% of
total motor vehicle production, and domestic sales of passenger cars constitute
about 75% of total sales." The following analysis draws heavUy upon this
study.
A. MAJOR FACTORS AFFECTING THE YEAR TO YEAR CHANGES IN SALES OF NEW
AUTOMOBILES
As with all durable goods, the demand for new automobiles is derived from
the demand for services which they yield. Since the automobile yields trans-
portation services for a period from five to fifteen years, the demand for these
services may be satisfied by running the old cars another year or by buying used
cars instead of new cars. Thus the consumption of automobile services is dis-
associated from the purchases of new cars, and, as is to be expected, the former
is far more stable than the latter. This is illustrated in Chart 1 and in Table 8
in the Appendix, in which automobile production is contrasted with registrations.
The number of cars in use held up relatively well even in the trough of the de-
pression, while the sale of new cars suffered a great decline.
(1) Consumer Income. — The level of consumer income is, of course, a major
factor in determining the number of new car sales. When the level of income
is high, sales will be high; conversely, when national income is low, car sales
will be low. This is illustrated in Table 4 and in Chart 3, which relate new' car
sales to consumer disposable income (Table 4, Columns 2 and 3, and Chart 3).
"< See Table 7.
" Steel Facta, American Iron and Steel Institute, December, 1930, No. 16, page 3.
!• Op., cit. supra,
" S L. Horner, "Statement of the Problem," The Dynamics of Automobile Demand, page 7.
CONCENTRATION OF ECONOMIC POWER
13987
Table 4.
-Retail Passenger Car Sales, Disposable Income, Minimum Cost of
Living and Supernumerary Income, 1919-1938
Year
(1)
Retail Pas-
senger Car
Sales '
(thousand
units)
(2)
Disposable
Income 2
(billion
dollars)
(3)
Necessi-
tous Liv-
ing Costs >
(billion
doUars)
(4)
Super-
numerary
Income ♦
(bUlion
dollars)
(6)
1919.
1,591
1,657
1,471
2,088
3,351
3,172
3,252
3,495
2,705
2] 652
1,903
1,096
1,526
1,928
» 2, 531
» i 749
« 1, 850
$61.38
66.29
53.60
56.45
64.98
66.02
69.46
72.94
72.53
74.92
78.50
71.21
60.29
46.67
45.23
52.38
55.65
63.06
68.97
64.20
$21.92
25.44
22.22
21.38
22.30
22.98
23.86
24.18
24.02
24.08
24.30
23.68
21.52
19.42
18.82
20.10
20.90
21.66
22.76
22.66
$39.46
1920
40 85
1921
31.38
1922
35.07
1924 - .- - .
43.04
45.60
1926 -..
48.76
1927
48.51
1928
50 84
1929
54.20
1930
47.63
1931
38.77
27.25
1933 —
1934.. 1
32.28
1935
34 65
1936
41 40
1937
46.21
Source: Data used by Roos and von Szeliski in The Dynamics 0} Automobile Demand.
1 For 1919-1925 the data represent factory production, less exports and foreign assemblies, less assumed
changes in dealers' stocks; for 1926-1929 the data are estimated by General Motors from retail sales of General
Motors passenger cars and new car registrations of other passenger cars; for 1930-1938 the data are estimates
of the Automobile Manufacturers Association.
> Figures for 1929-1938 are Department of Commerce income payments series, plus eatrepreneurial sav-
ings, less Federal income, gift, estate and inheritance taxes; figures for 1919-1928 are a backward extension
of 1929-1938 figures, on basis of data provided In Kuznets' National Income and Capital Forviation, 1919-
1935, less Federal and direct taxes.
' Estimated at $200 per capita for 1923 and varying with the National Industrial Conference Board Index
of the Cost of Living for other years.
' Equals column (3) minus column (4).
» Model year (12 months) ending in October.
A somewhat more refined analysis is possible if consumer income is adjusted
by deducting necessitous expenditure. Before distributing his income among
different commodities or between savings and spending, the individual consumer
must first allocate a portion to meet his necessary living costs. Messrs. Roos
and von SzeHski estimate the t;ubsistence or necessitous living cost to have been
$200 per capita for 1923 and to have varied in other years with the National
Industrial Cfonference Board index of the cost of living.
13988
CONCENTRATION OF ECONOMIC POWER
Deducting these costs from disposable consumer income, they obtain estimates
of "supernumerary income" which is available for expenditures on automobiles
and other goods (Table 4, Column 5).
It will be observed from Charts 3 and 4 that for a given amount oJ disposable
or supernumerary income the sales of automobiles are characteristically much
higher when income is rising than when it is falling. The reason for this has long
been known. The demand for automobiles, as for most durable goods, depends
not only on the level of income but on psychological factors such as the state of
confidence. Declining business activity and decreasing income give rise to un-
certainty and to fears that income will decline still further. As a result, even
relatively high levels of income may be associated with a low volume of sales.
Conversely, increasing business activity and income lead to increased confidence,
and to the allocation of an increasing proportion of present and future income to
automobile purchases.
The rate of change of the national income may be taken as an approximate
index of this psychological factor. If the rate of change, as well as the level of
Chart 3
2,500
1,000
500
0
RELATION OF PASSENGER CAR SALES TO
DISPOSABLE INCOME
IN UNITED STATES
l*»
19J7.
'?*
I9».
•»" .9«.
•I924
IJSt
-"»
>»? .1927
.1922
,,33
.,-34
.1919
.1920
\
.193?
loz.
c
50 5b 60 65 70
BILLIONS OF DOLLARS OF DISPOSABLE INCOME
Souaw: "ThtOrnamaal
national income, is taken into account, the relationship to sales is seen to be
much closer than appears from the use of national income alone. '* Although the
introduction of the rate of change of the national income as a factor in the analysis
does help to explain variations in the demand for automobiles, it is not an
entirely satisfactory measure of consumer confidence, since other phenomena
have ajjpreciable infiuence on the psychological reactions of automobile buyers.
A given rate of change may have little effect on sales at one time and a great
effect at another. The rate of change in income is, therefore, at best only an
approximate index of these psychological factors.
(2) Potential New Oivners and the Maximum Ownership Level. — The concept of
Messrs. Koos and von Szeliski of potential new ownership and of the maximum
ownership level is an important contril)ution to the analysis of the demand for
automobiles, since it not only explains the relationship between income and auto-
mobile sales but also presents a logical analysis of the major forces determining
the sales of automobiles.
Briefly summarized, their explanation is as follows:
(a) At any given time under given economic conditions there is a maximum
number of cars that will be kc[)t in operation. In the long run, changes in this
'» S. L. norncr, op. cil. supra. .
CONCENTRATION OF ECONOMIC POWER
13989
maximum ownership level depend, of course, on the growth of population, on the
development of highways, and on technical progress. From year to year, how-
ever, the level changes in response to the economic status of consumers, and to
other factors such as price and durability of the car.
(b) The number of potential new owners is equal to the difference between the
maximum ownership level and the existing consumers' stock of cars.
(c) The number of new owner sales is proportional to the number of potential
new owners, and to factors dependent on income, price, trade-in allowance,
volume of installment credit, and similar factors.
(d) Thus the demand factors are made to enter twice in the analysis of auto-
mobile sales, first in determining th maximum ownership level under any given
set of economic conditions, and secoi d in determining the nature of the reaction
of sales to changes in the number of i otential new owners.
(e) By means of these concepts, the relationship between income and auto-
mobile sales is explained. Assuming other factors to remain constant, the maxi-
mum ownership level at any given time depends on the level of income at that
Chart 4
2.500
2,000
1.500
1,000
500
0
RELATION OF PASSENGER CAR SALES TO
SUPERNUMERARY INCOME
IN UNITED STATES
1929.
.1937
■i^;""
..«.-^
.!«•
1935.
.930. r„,
1922
.1933
■13.
1919.
• 1938
•1920
.1932
Jni
3,500
3.000
o
i^
2,500
o
2,000
o
z
<
1,500
1,000
3
O
500
0
25 30 35 40 45 50 55
BILLIONS OF DOLLARS OF SUPERNUMERARY INCOME
" Tht Oywno cl AuOmMe Dr^t.-'d "
time, but the stock of cars in operation depends on previous income. The num-
ber of potential new owners is the difference between the maximum ownership
level and the stock of cars in operation and depends therefore on the difference
between this year's and previous years' income. Thus, while it is the change in
income from Yoxst levels which determines the number of potential owners, it is
the current level of income which determines the relation between the volume of
sales and the number of potential new owners.
(f) The maximum ownership level is a potent force in determining the volume
of new car sales. A sudden increase in income may increase the maximum owner-
ship level to a figure far above the number of cars in operation and thus lead to
a very large increase in car sales. This is undoubtedly what happened in 1937.
On the other hand a sudden decline in income may decrease the maximum owner-
ship level to a figure below the number of cars in operation. In that case there
will be an actual liquidation of part of the stock of cars in operation. This is
what happened during the depression years 1930-32.
(3) Replacement Demand. — Messrs. Roos and von Szeliski's theory with respect
to replacement demand is as follows:
(a) Not only do consumers adjust the number of cars in operation towards the
maximum ownership level, but they also adjust the quality of the cars in operation
towards some optimum level by means of replacement.
13990
CONCENTRATION OF ECONOMIC POWER
(b) Replacement demand depends on the pressure for replacement, and on such
economic factors as price, income and trade-in allowances.
(c) The age distribution of the cars in operation combined with experience
tables for scrapping furnishes a measure of the pressure for replaceinent. The
studies of car survival which have been made since 1926 show that car life during
the last fifteen to twenty years has slowly increased. Griffin's study of 1926 shows
50% of the cars surviving about seven years, whereas the most recent study based
on 1933-37 registrations shows 50% surviving about nine years. From these
frtudies may be computed the percentage of an original group of cars that is
scrapped after the first, second, third, etc., year of service. By application of
these percentages to the figures giving the age distribution of the cars in operation
in any year a measure of the replacement pressure during that year is obtained.
This index represents the theoretical scrapping rate.''
(d) Theoretical scrapping, however, merely indicates normal replacement pres-
sure. It is not equal to actual replacement since this varies with economic cir-
cumstances. In times of prosperity people scrap more cars than is indicated by
theoretical scrapping. The converse is true in periods of depression. Thus in
1929 actual scrapping was about one-third higher than theoretical scrapping,
whereas in 1933 it was about 60% lower than the theoretical rate.''" Replacement
sales therefore depend not only on theoretical scrapping but on income and price
and other economic factors.
(4) The Pi ice of Automobiles. — The almost continuous reduction (until 1933) of
car prices (Appendix Table 12) undoubtedly contributed significantly to the
great development of the automobile industry. The effect which price changes
have on year to year changes in sales is, however, a more difficult question.
One of the major difficulties encountered is the fact that manufacturing speci-
fications change so frequently. Since price changes do not occur separately but
in conjunction with changes in car models, it is impossible to segregate satis-
factorily the influence of price changes on car sales.
A second difficulty in analyzing the effect of price changes is the fact that there
have not been suflBciently wide fluctuations in car prices to warrant very reliable
conclusions. A lonp-run decline in car prices has been associated with a long-run
increase in sales. But since year to year changes in price have not been large,
it is difficult to discover what effect they have had on year to year changes in gales.
Another major problem in any statistical analysis of the effects of changes in
car prices upon volume of sales is the construction of a price index. Automobiles
have improved so rapidly in quality, and the changes in design and construction
have been so frequent, that it is next to impossible to construct a satisfactory price
index. The Bureau of Labor Statistics index of wholesale prices, which is an
average of prices of different makes, has been shown to be seriously in error.'^
The index used by Roos and von Szeliski is the average delivered price of the
lowest priced cars freely available in volume (Ford, Chevrolet, and Plymouth).
Their assumption underlying the use of this index is that this average price deter-
mines the number of cars sold, and that the prices of other cars merely determine
the distribution of sales among the various makes.''
Roos and von Szeliski conclude that price has not been a very important factor
in determining automobile sales. The usual measure of the responsiveness of
quantity sold to price is the elasticity of demand, or the ratio of the percentage
change in the quantity sold to the percentage change in price. It was found that
this elasticity was not constant but varied from year to year with changes in
economic conditions, particularly with changes in income and in the maximum
ownership level.
The figures below give the statistical estimates on the elasticity of demand for
the years 1919-1938:
Table 5. — Elasticity of Demand for Automobiles with Respect to Price, 1919-19S8
1919-.
1.03
1926
1.26
1933
1.30
1920._
1.04
1927.
1.33
1934
1.34
1921. _
1.04
1928
1.37
1935
1.34
1922-.
1.05
1929
1.41
1936
1.33
1923...
1.05
1930
1.51
1937
1.38
1924...
1. 15
1931
1.46
1938
1.53
1925...
1.22
1932
1.44
Source:
Roos and von Szeliski, 2
he Dynamic) of Automobile Demand
p. 94.
'• See Roos and von Szeliski, op. cit. supra, pages 47-53.
" See Roos and von Szeliski, op. cit. supra, page 52, chart 15.
" A. T. Court, "Hedonic Price Indexes", The Dynamics of Automobile Demand, pages 99-103.
'» This index is available only since 1920. For ttie years 1919-25 the Bureau of Labor Statistics Index was
used.
CONCENTRATION OF ECONOMIC POWER 13991
These are presented as the best estimates of the elasticity of demand. The
authors point out that the elasticity may be anywhere between .65 and 2.5,
and that 1.5 is probably a good representative figure. It will be noted that
there has been a long-run tendency for the elasticity to increase except in periods
of depression.
For comparison, there are given below their findings on the income elasticity
or the responsiveness of demand to changes in supernumerary income.
Table 6. — Elasticity of Demand for Automobiles with Rested, to Income, 1919-
1938
1919
_.-.-. 1.55
1926
2.08
1933
2.19
1920
1.65
1927
2.20
1934
2. 19
1921.._. ...
1.56
1928
2.25
1935
2.20
1922...
1.61
1929
2.39
1936
2.25
1923
...... 1.69
1930-.
2.62
1937
2.40
1924... •
1.80
1931
2.57
1938
2.58
1925
1.94
1932
...... 2.44
Sonrce: Roos and von Szeliskl, The Dynamics of Automobile Demand, p. 89.
Since the income elasticity is considerably higher than the price elasticity it
is evident that the influence of price on sales is not suflSciently powerful to over-
come the effect of the wide swings in income during the business cycle. Even
if the price elasticity were equal to the income elasticity, it would require a
50% reduction in price to offset a 50% decline in income. Since cash costs
for raw materials, tools, wages, salaries, taxes, and other out of pocket expenses
constitute about 90% of the wholesale price of a car it is obvious that such
reduction in price unaccompanied by a reduction in costs would be disastrous.^*
B. SECONDART FACTORS AFFECTING RETAIL SALES
The factors listed above account for all but a small part of the annual variation in
retail sales of automobiles. There are, howpver, some secondary factors which
are highly correlated with- those we have considered and whose iflfleucnce it is
therefore diflBcult to segregate from the rest. In certain years these forces may
be of particular importance in stimulating or discouraging sales.
(1) Used Car Allowances. — The used car allowance is one such factor. Since
the net cash cost to the buyer of a new car is the difference between the new
car price and the used car allowance, it is obvious that the size of the the allow-
ance must affect new car sales. In the statistical study made by Roos and von
Szeliski it was, however, impossible to measure this effect.
(2) Financing Terms. — Financing terms have become progressively easier.
The percentage of installment contracts running more than 12 months rose
from 14.5% in 1928 to 68% in 1937. The percentage of down payments that
were under standard terms (33>i% on new cars, 40% on used cars) rose from
6.1 % to 23.3% during the same period. Beginning in 1935 the easing of financing
terms was particularly marked.^* This undoubtedly has had an important
effect on the increased volume of purchases of new cars since 1935.
(3) Operating Costs. — Operating costs are particularly important as a long-run
trend factor determining the maximum level of ownership. The better the quality
of the car and the lower the annual operating and maintenance costs, the greater
is the number of cars that will be kept in operation.
Operating costs were not included as a separate variable in the statistical
study. The variation in operating costs has, however, closely paralleled the varia-
tion in automobile prices, and it is difficult therefore to separate their respective
effects on the volume of automobile sales. Roos and von Szeliski have made
estimates of the elasticity of demand for automobiles 2« when the effects of oper-
ating costs are included and also when they are excluded. It is the latter series
which has been used in the section on prices (Table 5).
(4) Dealers' Used Car Stocks. — Dealers' used car stocks undoubtedly have im-
portant effects on sales of new cars in some years. When used car stocks are large,
dealers tend to push the sales of used cars with greater vigor, and they lower their
trade-in allowances. Both of these have a depressing effect on sales of new cars.
Unfortunately no data are available on the number and value of used cars in
dealer hands, and it is difficult to get any statistical evidence on this point.
However, even a superficial examination of the automotive journals indicates that
this is at times a major problem.
" See S. M. Dubrul, "Significance of the Findings", Dynamics of Automobile Demand, pages 123-39.
'* Roos and von Szeliski, op. cU. supra, page 68.
" Roos and von Szeliski, op. cit supra, pages 92t 94.
13992 CONCENTRATION OF ECONOMIC POWER
VI. The Relation of Steel Prices to the Demand for Automotive Steel
Since automotive steel is purchased primarily as a raw material for the produc-
tion of cars, consumption of steel by the automotive industry bears a direct relation
to production of cars. This is illustrated by the data on automobile production
and steel consumption ^6 in Chart 2. In almost every year from 1923 to 1938 the
two series moved in the same direction and had approximately the same relative
magnitude of fluctuation. For the period as a whole, however, there was a long-
run tendency for the consumption of steel to increase by a greater percentage than
automobile production. This is shown by the upward trend in steel consumption
per automobile (Table 7).
In contrast to the close relationship between steel consumption and automobile
production, there is little, if any, relation between steel consumption and steel prices.
This is illustrated graphically in Chart 5, in which both the composite feteel price
and the simple average of sheet and strip prices are plotted." The consumption
Chart 5
z
o
1-
u.
o
V)
z
o
-1
i
AUTOMOBILE §TEEL CONSUMPTION AND STEEL PRICES
IN UNITED STATES
9
8
7
6
5
4
3
2
1
_ _
9
8
7
6
5 1
4 1
q:
3 ^
</)
' 1
1
^AUTOMOBILE SI
EEL
^
■^
y
■s.
V
^NSU
<PIH
/
f
\
V
/
sj
t¥
roMO
IHU
\
1
X
•*«*»
n
***=,
-=^
•-'^
as*
i
^
.—
^
-.
STE£
. PRK
f£
ktt,
^iilllillliiiiiiis
series (Appendix Table 9) and price series (Appendix Table 10) move in the same
direction about as often as they move in the opposite direction.
This conclusion is further supported by Chart 6, a scatter diagram of auto-
motive steel consumption against automotive steel prices. In this chart, per-
centage changes in annual consumption and annual average price are shown in
relation to each other. Since the scatter diagram fails to indicate that lower steel
prices have been associated historically with greater quantities of steel purchased
and vice versa, it justifies the view that price has been less important than other
influences in determining the consumption of automobile steel.
As was pointed out earlier in the discussion, the quantity of steel consumed as a
raw material in the automobile industry depends directly upon three factors:
(1) The number of cars produced;
(2) The quantity of steel used per car;
(3) The production of replacement parts and accessories.
A change in the price of steel can, consequently, affect the consumption of steel
only as (1) it leads to a correspondin"g change in the price of automobiles and tlius
2« See Note 13 regarding limitations of data.
2' See Table 10 in the .'vppondix for these prices. This average of hot and cold rolled sheet and hot
rolled strip prices is henceforth referred to as the automotive steel price. A simple average of the prices of these
steels was taken because (1) they are the most important kinds of steel used in making automobiles and
(2) they arc used in about e^ual amounts per car.
CONCENTRATION OF ECONOMIC POWER
13993
to an inverse change in the number of automobiles sold, or (2) it induces the
manufacturers of automobiles to change the quantity of steel used per car, or
(3) it causes a change in the production of replacement parts and accessories.
Considering first the influence of steel prices on automobile prices and produc-
tion, we find it impossible to determine directly from an examination of steel
prices whether a change in the price of steel used by the automobile industry has
had any direct effect on the price of automobiles. (See Chart 7 and see Table 12
in the Appendix.) The statistics show merely that both series of prices moved
in a long-run downward trend until 1933, and thereafter moved horizontally or
upward. It will be observed also that the movements in these price series were
not greatly different from the movements of a broad price index of manufactured
goods during the same period. From this statistical evidence it is impossible to
infer whether changes in the price of steel did or did not cause changes in automobile
prices.
It is possible, however, to discover the extent to which steel prices might affect
car prices by an analysis of the cost of steel in an automobile. As is shown in
Chart 6
o
180
S
<^
160
fe
^
140
2
o
1?0
'n
CO
100
RELATION OF AUTOMOBILE STEEL PRICE TO
AUTOMOBILE STEEL CONSUMPTION
1933 •
1925.
.,.,
193S
1926.
.1936
19.U.
I937.
1927.
• 1929
1931.
.1924
•1930
- .1,3.
90 IQO 110 120
AUTOMOBILE STEEL PRICE IN % OF PREVIOUS YEAR
130
Table 11 in the Appendix, the cost of steel used in the manufacture of a Chevrolet
sedan is approximately 13% of the F. O. B. price or approximately 10% of the
delivered price of the car.^^ A reduction of 10% in the price of automotive steel
would therefore permit a reduction of only 1% in the dehvered price of the car if
the reduction were passed on entirely to the consumer. The effect of changes in
car prices on car sales has been studied by Roos and von Szeliski, and their esti-
mates of the elasticity of demand for automobiles appear in Table 5 of this
memorandum. Their figure of 1.5 for the elasticity of demand for automobiles
means that a 1% change in the price of automobiles gives rise to a 1.5% change
in the number of automobiles sold. Since a 10% reduction in the price of steel,
if passed on entirely to the buyer, is equivalent to only a 1% reduction in the price
of the car, it would increase the volume of car sales and therefore the consumption
of steel by only 1.5%. Even if the maximum limit of 2.5 estimated by Roos and
von Szeliski be taken as the elasticity of demand for automobiles, the effec of a
10% drop in the price of steel, if passed on to the consumer, would be to increase
«« In heavier, higher priced automobiles the quantity of steel used does not increase proportionately with
price, *nd consequently the ratio of steel co.' t to retail price falls below 10%.
13994
CONCENTRATION OF ECONOMIC POWER
car sales and automotive steel consumption by only 2.5%." It is evident, there-
fore, that a reduction in the price of steel is not an effective means for raising steel
consumption through increases in car sales and car production.
The effect of a reduction in the price 04' steel on the quantity of steel used per
car is, so far as can be determined, practically negligible. During the past
fifteen years a number of factors have contributed to an increased use of steel
per car. These are:
(1) The increased popularity of enclosed models and heavier cars;
(2) Engineering improvements in the production of both steel and auto-
mobiles;
Chaet 7
PRICES OF AUTOMOBILE STEEL,
AUTOMOBILES AND MANUFACTURED GOODS
!N UNITED- STATES
4
cr.
<r> 2
0
1
0
100
1 80
1 ^°
z
40
20
\
30 2
20 e
10
0
\
^N
PP
AUT
ICE
OMO
'ER
BILE
POU
MD
^
V
V
y
;2;
s:
4
>
AUl
PR
OMOBILE
CE PER
ST
=ou
EL
40
^
^
k
192
1
6= 100
100
80 1
60 Q
40
20
"^
^
'n
d
s
AUVOM
PR!
)BIL
:e^
x
^
^
^
MANUF
^CTl
RED
GOC
)0S
.
PR
IVMOL
CE
)
s
^.,: Aon ^5^.
iiiilissliisllil
(3) Intense competition among motor companies pointed toward improve-
ment of their product.
There is, however, no logical or statistical basis for inferring that the price of
steel has been a significant factor in causing the increased use of steel per car.
Unfortunately, the statistical series available on the steel consumption per
automobile are none too satisfactory. Attempts to derive such a series from
the estimates of steel consumption by the automobile industry appearing in the
" It might appear that a reduction in the price of automobiles made possible by a general decrease in all
cost components would bo an effective means of stimulating automobile purchases. The elasticity figures
here quoted, however, do not apply to price reductions accompanied by a general deflation in the prices of
the cost factors throughout the economy. The effects of general deflation in prices and costs are too compli-
cated and complex to explore here.
CONCENTRATION OF ECONOMIC POWER 13995
trade magazines, Iron Age and Steel, led to the conclusion that these estimates
were too inaccurate for this purpose.'" Resort was had, therefore, to estimates of
the average dry shipping weight of all passenger cars as ap index of the quantity
of steel per car.
Table 7. — Average Dry Shipping Weight of all Passenger Cars, 1920-1938
[Weighted by new car registrations]
Weight in
Weight in
Weight in
Year:
Pounds
Year:
Pounds
Year:
Pounds
1920.-_-
2373
1927
---. 2486
1934....
2691
1921
2223
1928
2587
1935
2791
1922....
2324
19^9
.... 2623
1936....
2881
1923-—
.... 2348
1930
.... 2598
1937....
2871
1924
2338
1931
2682
1938
2918
1925-..
.... 2356
1932
.... 2769
1926
2412
1933
2599
Source: Datafrom private sources. Approximately the same data for 1925 to 1938 is obtained from Auto-
mobile Facts and Figures (1939), p. 48, by dividing average price per car by average price per pound.
Chart 8
110
RELATION OF AUTOMOBILE STEEL PRICE TO
PASSENGER CAR WEIGHT
mi*
1925«
,1928 i,j5
1938. I»6
• 19M
193*.
. •l924
1930
.937'
19M»
90 100 110 120
AUTOMOBILE STEEL PRICE IN % OF PREVIOUS YEAR
^
> Aj* an^' Aijtorroi>ife f^-aflJtrt-jting Ai^ocit
Chart 8 compares the average car weight with an index of typical automotive
steel prices " in order to determine what relationship exists between them. In
order to eliminate the trend in both series, annual percentage changes instead
of the original figures are plotted against each other. This chart does not dis-
close any relationship of car weight to steel prices. This lends support to the
conclusion that the quantity of steel used per car depends primarily upon technical
factors and long-run trends in taste rather than upon changes in the price of steel.
In the discussion thus far there has been no consideration of the elasticity of
the demand for steel in the accessories and parts industry. It is reasonable to
suppose, however, that the elasticity of demand for steel in this industry is about
the same as in the automobile industry. In any event, since the value of the
production of parts and accessories is only about one-seventh of the value of the
production of cars, the elasticity of demand for steel in both industries taken
»» see note 13 regarding the quality of the data.
1 Weights from Table 7, prices from Table 10 in the Appendix.
124491 — 41— pt.
13996 CONCENTRATION OF ECONOMIC POWER
toegther is not much different from the elasticity of demand for steel in the au-
mobile industry aloiie.'^
To sum up: Tlie elasticity of demand for steel used as a raw material in the
automobile industry is very low, probably no higher than .2 or .3. A 10% reduc-
tion in the price of automotive steel, if passed on to the ultimate consumer, would
lead to approximately a 1.5% increase in the consumption of steel directly attrib-
utable to increased car sales, and possibly to a very slight additional increase
directly attributable to increased weight per car. Considering these combined
effects, a 10% reduction in the price of automotive steel would probably not
increase its consumption by more than 2% or 3%.
If automotive steel sold for $60 a ton before such a 10% price reduction, a maxi-
mum increase of 3% in automotive steel consumption resulting therefrom would
mean that the steel producer would sell 1.03 tons of steel where he formerly sold
one ton, but for such increased quantity he would receive, under the reduced prices
of $54 a ton, only $55.62 as compared with his earlier sales return of $00 for a
smaller quantity. Consequently a reduction in steel prices would decrease the
gross revenues, increase the operating expenses and greatly reduce the net income
of the steel industry.
Automobile sales have slumped in depression because of reduced consumer
income and fear and uncertainty of the future. Against these tidal forces a reduc-
tion in the price of automotive steel could have no significant effect.
Appendix
Table S.-^Production of Cars and of Automobile Parts and Accessories, and Total
Number of Cars Registered, 1921-1933
Production of Auto
Parts and Accessories
at Producers' Prices '
(millions
of dollars)
Production of Cars at
Producers' Prices '
(millions
of dollars)
Total Number of Cars
Eegistered >
(Millions)
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
38.4
55.2
80.9
76.6
100.9
100.0
95.5
92'. 7
74.1
62.0
48.2
51.8
$1, 147
1,596
2.278
2,034
2,524
2,679
2,175
2,557
2.800
1,643
1,124
627
757
100.0
81.2
95.4
104.5
61.3
42.0
10.5
12.2
15.1
17.6
19.9
22.0
23.1
24.5
26.5
26.5
25.8
24.1
23.9
47.7
55.5
68.6
80.0
90.5
100.0
105.0
111.4
120. 5
120.5
117.3
109.5
108.2
Sources: ^ . t. ^
' S. Kuznets, Commodity Flow and Capital Formation, National Bureau of Economic Research, New
York, 1938, p. 137. The estimates of auto parts and accessories e.xcludo original equipment used in the
production of new cars. Data not available for years subsequent to 1933.
2 Automobile Fads and Figures, Automobile Manufacturers Association, 1938, p. 16.
32 The cost of steel in tools and dies is a negligible part of the total cost of an automobile and is. therefore,
disregarded in this study.
CONCENTRATION OF ECONOMIC POWER 13997
Table 9. — Automobile Production and Steel Consumption, 1923-19S8
Year
Production of Motor
Vehicle.s '
Automotive Steel Con-
sumption '
Steel Con-
sumption
per Motor
Vehicle
Produced'
(thousands
of units)
1929=100
(thousands
of gross ■
tons)
1929=100
(gross tons)
1923
4180
3738
4428
4506
3580
4601
5622
3510
2472
1431
1986
2870
4120
4616
5016
2655
74.4
66.5
78.8
80.1
8l'.8
100.0
62.4
44.0
25.5
35.3
51.1
73.3
82.1
89.2
47.2
4182
2981
5486
4895
6963
6565
4406
3149
1864
3530
4101
6016
6712
6977
3619
63.9
45.5
74.7
74^8
106.4
100.0
67.3
48.1
28.5
53.9
62.7
91.9
102.2
106.6
55.3
1.000
797
1924
1925 .. .. .. .. .
1.103
1926... .. ...
1927
1928. .. .. ..
1929
1930
1931 .
1932
1933
1934
1935 .. . ..
1936 -.
1937
1938.. ..
1 statistical Abstract of the United States, 1938, p. 370. 1938 figure from Automobile Facts and Figures, 1939,
p. 4.
2 Computed by apportioning individual hot-rolled product totals on the basis of Tfon Age distribution
reports and by allocating jobber shipments to ultimate consumers. See M. W. Worthing, Distribution
of Steel Products to Major Consuming Industries, United States Steel Corporation, October 30, 1939.
3 Considered less reliable as an index of steel used per car than the average dry shipping weight shown in
Table 7.
Table 10.— Steel Prices, 1923-1938
[Cents per poundl
Year
Hot-rolled
Strip
Hot-rolled
Sheets No.
10 Gage
Cold-rolled
Sheets
"Automotive
Steel" Price
(average of
the three
prices) i
Iron Age
Composite
Steel Price
1923
3.04)S
2.57
2.27
2.30
2.26
1.93
L68
1.54
1.43
1.58
1.85
1.85
1.91
2.35
2.25
2.96«i
2.79
2.45
2.37
2.20
2.04
2.12
1.99
1.86
1.71
1.62
1.85
l!92
2.35
2.25
5. 19«S
5.00
4.39
4.30
4.17
4.03
4.06
3.64
3.13
2.80
2.48
2.96
2.95
3.02
3.49
3.31
3.73«i
3.45
3.04
2.99
2.88
2.67
2.69
2.44
2.18
1.98
2^22
2.22
2.n
2.60
2. 697<
2 505
1924
1925
2 334
1926
2 315
1927
2 202
1928 ■
2 165
1929
2.209
2.048
1930
1931 - .
1932
1933
1934 ...
1935...
1936
2.077
1937 ___
1938
Source: The Iron Age, Annual Statistical Supplement, reprinted from the January 5, 1939, issue, pp. 8-10.
1 A simple average of the prices of these three steels was taken as the average automotive steel price because
(1) they are the most important kinds of steel used in making automobiles and (2) they are used in about
equal amounts per car. See Table 11
13998 CONCENTRATION OF ECONOMIC POWER
Table 11. — Net Cost of Steel Products Used in Construction of Chevrolet Sedan, 1938
[F. O. B. Flint Price $595.0O-New York Delivered Price $730)
Steel Products Used
Analysis of Steel Cost July 14,
1938
Product
Weight in
pounds
Price at Detroit
Cost of product
for one auto
144.0
162.0
23.0
15.8
45.0
17.0
278.0
554.0
531.0
98.9
509.2
0.2
23.6
0.9
284.2
0.6
0.6
19.0
6.0
1.0
6.0
2.3
7.8
70.8
$20. 50 Q. T.
20. 50 Q. T.
3. 85 per 100#
2.32 perl00#
3.15 perl00#
37.32 O. T.
45.00 Q.T.
3.35 perlOO#
2.925perl00#
3.77 per loo*
2.65 perlOO#
30.29 periOO#
4. 575 per 100#
4. 575 per 100#
3.60 perl00#
38.42 Q.T.
99. 54 Q. T.
11.75 Q.T.
3. 47 per 100#
9. 58 per 100#
3.47 porlOO*
61.32 Q.T.
3. 82 per 100#
3.77 perlOO#
$1 31
Pig Iron— Malleable
1 48
0 89
Plates
0.37
Bars — cold drawn
1.42
Sheet bars
0.28
Billets — forging - ...
5.68
Sheets— cold rolled
18.56
Sheets — hot rolled - ..
15.53
3.73
13.49
0.06
1.08
0.04
9.95
0.01
0.03
0.10
Nails -
0.21
0.10
0.21
Wire rods
0.05
0.30
2.64
Total products listed - .
2800.1
> $77. 42
« $77.42 is about 13% of $595 and about 10% of $730.
Source: Data on steel products used were taken from MiU and Factory, June, 1938, p. 96, and were supple-
mented by conversations with persons in the automobile industry. Data on prices were prepared by United
States Steel Corporation.
Table 12.-
^Automobile Prices, Automotive Steel Prices and Prices of Manufactured
Goods, 1919-38
Retail Passenger Car Price >
Index of
Wholesale
Prices of
Manufactured
Qoods >
(1913=100)
Year
(dollars per
car)
(cents per
pound)
Automotive
Steel Price >
(cents per
pound)
1925
•$1007
943
977
911
843,
798
767
723
630
664
658
687
704
779
42.70
39.1
39.3
35.2
32.1
30.7
28.4
26.1
24.2
24.7
23.6
23.8
24.6
20. 7
3.04^
2i.99
2! 67
2.69
3.44
V>.18
1.98
1.89
2.22
2.22
2.28
2.73
2.60
162.4
1926
157.6
1927 .
152.0
1928
153.1
151.5
140.4
123.?
1932 .
111.6
1933-
113.8
1934
126.0
130.9
1936
130.3
1937
139.8
1938.
130.9
Sources:
1 Taken from Autojnobile Facts and Figures, A. M. A. 1939, p. 48. Kepresents delivered price at
factory of the cheapest four or five passenger closed models of each make, weighted by new car
rcKistration.
> From Table 10.
» Taken from F. C. Mills, Prices on Recession and Recovery. National Bureau of Economic
Kus<iarch, 1936. Appendix III, pp. 491-2. Fieures for 1936-1938 from National Bmocu of Econo»ic
Uc:Jc:ircU.
CONCENTRATION OF ECONOMIC POWER 13999
Exhibit No. 1414
AN ANALYSIS OF THE DEMAND FOR STEEL IN THE RAILROAD
INDUSTRY
This is an analysis prepared by the Special Economic Research Section of the
United States Steel Corporation, composed of Messrs. Edward T. Dickinson, Jr.,
Ernest M. Doblin, H. Gregg Lewis, Jacob L. Mosak, Mandal R. Segal, Dwight B.
Yntema and Miss Marion W. Worthing. The work of this group was under the
supervision of Theodore O. Yntema, Professor of Statistics, University of Chicago.
This report was written by Mandal R. Segal and has had the benefit of suggestions
from other members of the staff. It is issued by United States Steel Corporation.
November 1, 1939.
contents
I. Purpose.
II. Findings.
lU. The Railroad Industry.
- IV. Factors Affecting Steel Consumption by the Railroads.
A. Freight Traffic.
B. Passenger Traffic.
O. Financial Condition.
1. Operating Revenues and Profits.
2. Source of Funds (or Capital Investment.
V. Demand for Steel by the Railroads.
A. Maintenance and Capital Expenditures.
1. Maintenance Expenditures.
2. Capital Expenditures.
B. Steel Consumption by the Railroads.
C. The Influence of the Price of Steel on Its Consumption by the Railroad Industry.
VI. Summary. (,
I. Purpose
The purpose of this study is to review the factors which influence the railroads'
demand for steel, to analyze their relative importanc?, and in particular to ap-
praise the role of price in the consumption of steel by the railroads.
II. Findings
(1) The railroads' demand for steel is derived from the demand for railroad
services, which in turn depends primarily upon the national income and the com-
petition from alternative means of transportation.
(2) Due to the durability of railroad equipment, changes in the demand for
railroad services generate much greater fluctuations in the demand for equipment
and hence in the consumption of steel by the railroads.
(3) Capital expenditures for rolling stock and other equipment requiring steel
are ultimately dependent on the demand for rail transportation; but in the short
run they are determined by th§ adequacy of existing facilities for current traffic
needs and by the funds available for such outlays. Whep the demand for railroad
services declines, there is less need for capital expenditures on rolling stock and
other equipment which require steel in their production, and also railroads have
less funds available for such expenditures.
(4) The consumption of steel for maintenance purposes is closely related to the
volume of traffic currently handled by the roads.
(5) Price is of minor importance in determining the consumption of steel by
the railroads. This is emphatically true of aggregate steel purchases over a con-
siderable period of time. It follows from two facts:
(a) the demand for transportation services for freight and for passengers
certainly has a low elasticity, probably leas than unity. This means that a
given percentage change in rates probably will have a smaller proportionate
effect on the volume of railroad services. The effects of rate changes, how-
ever, may vary among railroads and different sections of tlie country.
(b) the cost of steel is a comparative!}^ small fraction of the total cost of
transportation service, and consequently has little effect on the price (i. e.,
rates) or volume of railroad services.
(6) Even with respect to cyclical timing of steel purchases, price is not a major
influence. Capital equipment is purchased when it is needed and can be profitably
employed, and when funds are available. Maintenance is determirfed partly by
the requirements of traffic and partly by the revenues w-hich can be used for this
purpose. In a depression, purchases are reduced to conform to current needs.
In their dire financial straits the railroads are in no position to speculate on possible
fluctuations in steel prices by making purchases not required by current needs.
14000 CONCENTRATION OF ECONOMIC POWER
(7) The low consumption of steel by the railway industry in the last decade
has been due to the declining trend in rail transportation, the severe depression
in business and the poor financial conditions of the railroads. No conceivable
reduction in price of steel to the railroads could have counteracted these forces to
a substantial extent.
III. The Railroad Industry
Any attempt to determine the railroads' demand for steel must start with some
analysis of the railroad industrj'. As the chief agency for transportation, rail-
roads have been indispensable in the national economy. They are also of great
importance as sources of employment, investment and consumption.
The railroads have recently been in a stage of transition from an expanding to a
declining industry. The periods of experimentation between 1830 and 1850, of
rapid expansion from 1850 to 1890, and of sub.sequent development of particular
areas and of feeder and cross lines have, since the World War, been followed by an
era of shrinkage in railroad mileage, severe competition from other transportation
agencies, and declining trends irroperating revenues. In recent years the financial
position of most railroads has been seriously impaired; approximately one-third
of the railroad mileage of the United States is in bankruptcy,' and in 1938 only a
few roads earned all their fixed charges.
IV. Factors Affecting Steel Consumption by the Railroads
The decline in railroad operations has been accompanied by a decline in the
railroads' consumption of steel. The railroads had long been the leading customers
of the steel industry, but during the past ten years the purchases of steel by the
railroad industry have declined absolutely and relatively. Whereas in 1923, the
railroads consumed approximately 25.4% of k\l the finished steel produced and
ranked first as a consuming group, by 1932 they consumed approximately 10%
of the finished steel produced and ranked third as a consuming group. In 1938
they consumed only 6.1% of the total finished steel produced in the United States.''
Railroads use a variety of steel products for many different purposes. Their
purchases range from rails, plates and structural shapes to bolts, nuts, washers
and rivets. In general, steel is now consumed by the railroads in the form of new
locomotives and cars and as material for the maintenance of way, structure and
equipment.
The demand for steel by the railroads is derived from the demand for freight
and passenger tra'nsportation services, which exhibit marked cyclical fluctuations.
It is this fact which accounts for the great fluctuations in the purchases of steel
by railroads. These fluctuations are further intensified by the durability of the
steel products used by the railroads, and by the limitations of impaired financial
condition and the reduced operating revenues experienced in periods of depression
by the railroads. These relationships are set forth in more detail in the follow-
ing analysis.
A. FREIGHT TRAFFIC
Freight traffic rose gradually during the 1920's, reaching a peak of 450 billion
ton-miles of railroad revenue freight in 1929.^ In the depression low of 1932 it
fell to 235 billion ton-miles. By 1937 freight traffic had recovered to 363 billion
ton-miles, but for the year ended June 30, 1938, it declined again to 313 billion
ton-miles,*
It has long been recognized that freight car loadings are closely related to indus-
trial production. This is to be expected, for frg'ight car loadings include mainly
industrial products and industrial raw material such as coal, coke, forest products
and ore. Chart 1 shows this close relationship between industrial production
and miscellaneous and merchandise freight car loadings.* By statistical analysis
the National Resources Committee found that fluctuations in freight traffic were
almost completely explained by fluctuations in industrial production. «
The evidence points strongly to the conclusion that the demand for rail freigtlit
transportation has a rather low elasticity. First, the very close dependence of
rail freight traffic on business activity lends support to this conclusion.
1 See infru, Table 2.
' See Appendix, Table 7.
' Soe Appendix, Table 8.
< Se"^ Appendix, Table 8.
» See also Appendix, Table 9. Miscellaneous and merchandise G- c. 1.) freight car loadings were used,
since they cover industrial products more directly than the combine^ index of freight car loadings' which
includes agricultural products and livestock.
• Industrial Committee of the National Resources Committee, Patterns of Resource Use (1939). p. 99. It
was found that ninety-seven percent of the variation in freight traffic was explained by variations in indus-
trial production and a time trend.
CONCENTRATION OF ECONOMIC POWER 14001
Second, the demand for freight transportation is derived from the demand for the
goods to be transported and must have a low elasticity because transportation
charges are not usually a large fraction of total value of goods carried. The
availability of substitute means of transportation tends to increase the elasticity
of demand for rail freight service, but it should be noted that the possibilities of
substitution are limited by technical obstacles. Furthermore, from a broader
point of view the substitution of one transportation agency for another does not
greatly affect the elasticity of demand for total transportation or for the steel
consumed by all transportation agencies.
Although the movements of the two series have been much alike, freight car
loadings have not kept pace with industrial production, especially since the
depression low of 1932. The failure of the freight car loadings to recover as rapidly
as industrial production since 1932 has been due largely to the diversion of traffic
Chart 1
INDUSTRIAL PRODUCTION AND M'D'SE AND MISC. CAR LOADINGS
l-tUtHAL RESERVE INDEXES ADJUSTED FOR SFASONAL VARIATION
1923 . 1925 = 100
180
180
160
140^
120 1
100 i
80 2
a
60 ^
40
20
0
160
w 140
UJ
« 120
z 100
S 80
o
5 60
40
20
A
1
NOUSTRIAL
C>.
>y^
>JSS
®^
r^
z\
V
'z
^
^
^v
\
^
A
;?
^
7
\
L
U
-^
)
kvx
V.
G
iR LOADING.
>
M'D'SE (LCU.^
CAR LOADINGS
&>«>.' >^<W
1924 1926 1928 1930 1932 1934 1936 1938
from the railroads to trucks, waterways, and pipe lines. This shift from the
railroads to the other commercial forms of transportation is evident from the
following table:
Table \.— Distribution of Freight Traffic by Commercial Agencies, 1926 and 1937
Agency
Amount Millions
of Revenue Ton-
Miles
Percent of Total
Ton-Miles
1926
1937
1926
1937
Steam Railways
447, 443
23,530
90,038
9,543
21,700
1,3131
Nil/
362,815
43. 380
93,244
16,883
44, 793
"3
Percent
75.4
3.9
15.2
1.6
3.7
0.2
Percent
64 6
Intercity Trucks
7 7
Great Lakes
16 6
Electric Railways!
Airways /
0.1
Total..
593, 567
561,815
100.0
Source: Report of Committee, Appointed Sept. 20, 1938, by the President of the United States to Sub
Recommendations upon the General Transportation Situation (1938), pp. 44-6.
14002 CONCENTRATION OP ECONOMIC POWER
These data show that the proportion of the commercial freight business handled
by the railroads declined from three-fourths of the total in 1926 to less than two-
thirds in 1937. While the railroads handled 84.6 billion ton-miles of freight less
in 1937 than in 1926, the intercity trucks, the waterways, and the pipe lines
increased their freight ton-miles by 19.9, 10.5 and 23.1 bilUon ton-miles, respec-
tively. In percentage terms, there was a decrease of 18.9% in the freight ton-miles
handled by the railways between 1926 and 1937.
B. PASSENGER TRAFFIC
Passenger traffic has exhibited a severely decUning trend since 1923, when
revenue passenger miles had reached an all-time peak (excluding the period of
federal control) . By 1937 passenger traffic had dropped 35% from its 1923 level,
in spite of an increase of 16% in population and a rise of 24% in consumers'
income (in 1936 dollars).'' This marked downward trend in passenger traffic has
been due mainly to the expansion of other forms of transportation, primarily the
automobile; from 1923 to 1937 -automobile passenger car registrations increased
88%* and passenger car mileage rose more than 50%.
Railroad passenger traffic has been greatly influenced by consumers* incomes.'
A recent study of the National Resources Committee found that consumers'
income was- the most important factor in explaining the variations in passenger
traffic 1" and that railroad fates were, much less important. Its findings indicate
that the demand for passenger service was inelastic in that a one per cent reduc-
tion in rates would induce about a one-half of one per cent increase in passenger
traffic." This inference may not necessarily be the same for all railroads or for
different sections of the country but appears to be warranted for total passenger
transportation.
Consumers' income and competition from other transportation media, then, are
the most important factors explaining fluctuations in passenger traffic. In the
depression of the early thirties rail passenger traffic fell to extremely low levels
because the cyclical decline in consumers' incomes was reinforced by the continued
shift of traffic to other media of transportation. In 1933, revenue passenger miles
were less than 50% of their average level during the twenties; in the 1937 recovery
that foUowed, although consumers' income (in 1936 dollars) surpassed its 1929
level, rail passenger traffic failed by 21% to reach its 1929 volume." In the last
few years the volume of passenger traffic would probably have been even lower
had it not been for marked improvements in the quality of service.
C. FINANCIAL CONDITION
1. Operating Revenues and Profits. — Since 1926 the financial condition of rail-
roads has become progressively more critical. This is mainly attributable to a
decUne in the volume of traffic which became marked beginning with 1930. Total
operating revenues reached a peak of 6.4 billion dollars in 1926 and declined to a
low of 3.1 billion dollars in 1933. Although in 1937 revenues recovered to 4.2
billion dollars, 35% above the 1933 level, they fell again in 1938 to 3.6 billion,
only 15% above the depression low.'*
This drastic decline in total operating revenues, combined with the smaller
reduction in operating expenses and the slower decline in taxes and fixed charges,
caused larger deficits." Net income in 1929 amounted to about 900 million dol-
lars, and dropped to a deficit of 140 million in 1932. There were deficits again
in 1933 and 1934, and an insignificant net income in 1935. After moderate earn-
ings in 1936 and 1937, the deficit reached over 120 milHon dollars in 1938. Divi-
dend rates" in 1929 and 1930 were 6% of the capital stock outstanding, but in
1932 they were only 1.12%, and even in the best recent year not higher than
2.11%."
' See Appendix, Table 10.
« See A.ppondlx, Table 11.
• See Appendix, Table 10.
'• Cf. National Resources Oommlttee, op. eit., pp. 00-1 and 99.
" Ibid.
'« See Appendix, Table 10.
>• See Appendix, Table 12.
i< See Table 2 in text and Appendix, Table 12.
" See Appendix, Table 13.
CONCENTRATION OF ECONOMIC POWER
14003
Table 2. — Railroad Operating Revenues, Net Income and Mileage in Receivership,
1928-38
Year
Operating
Revenues i
{thousands
of dollars)
Net Income »
(thousands
of dollars)
Percent of Total
Railroad Mile-
age Operated
by Receivers
or Trustees «
1928
$6,111,738
6, 279, 621
5, 281, 197
4, 188, 343
3, 126, 760
3, 095, 404
3, 271, 667
3, 451, 929
4, 052, 734
4. 166, 069
3, 565, 000
$786. 824
898, 807
623,907
134, 762
139, 204
« 5, 863
16,887
7,539
164,630
. 98,068
123,000
2.02%
1929
1930
1931
1932
8.71
16.24
16 64
1933
1934
1935
26 87
1936
27 67
1937
1938
Sources:
' Figures (or operating revenues and net income cover Class I railroads only, while mileage in receiver-
ship applies to all railroads.
» Based on Reports of Interstate Commerce Commission and Association of American Railroads, Bureau
of Railway Economics (A Review of Railway Operations in 19SS).
• Based on Reports of Interstate Commerce Commission.
• Deficit.
The uufortunate financial condition of the railroads is forceably demonstrated
by the growing percentage of mileage operated by receivers or trustees. Table 2
shows an uninterrupted rise from 2.02% in 1928 to 30.98% in the middle of 1938
in spite of some recovery in operating revenues and net income between 1933 and
1937.
In part, the financial difficulties of the railroads have been due to the inflexi-
bility of certain costs. From 1921 to 1929 an average of 11.0 cen^s of every doUar
of operating revenue went to pay fixed charges; then from 1930 to 1937, this
average rose 63% to 17.9 cents." In dollar terms, the expenditures for interest
on debt remained fairly stable in the depression; from 518 million dollars in 1931,
this expenditure rose to slightly over 525 million dollars in 1932, and graduaUv
feU to 491 million in 1937.i'
Hourly wage rates have exhibited similar inflexibility. The average hourly
wage in 1929 was 66.6 cents; after a drop to a low of 62.9 cents in 1933, it rose to
68.6 cents in 1935 and 70.9 cents in 1937.18 Since approximately 45% of the
average operating revenue dollar is expended for wages,'" this comparative inflex-
ibility in wage rates has adversely aflFected the earnings and general financial
position of the railroads.
2. Source of Funds. — During the relatively prosperous period of the twenties,
about 70% of the funds used by the railroads for capital investments were ob-
tained from income, approximately 25% from the issue of securities, and the
relatively small remainder from reduction of working capital. During the period
from 1931 to 1937, not only was income available for investment greatly reduced,
but the capital market was practically eliminated as a source of funds for capital
investment; in fact, during this period more securities were retired by the railroads
than issued. The following table shows the amount and the sources of the funds
for capital purposes from 1921 to 1937:
" Report of the Emergencv Board to the President (October, 1938), p. 16.
" See Appendix, Table 13.
» Ibid.
" Report of the Emergencv Board to the President (October, 1938), p. 30.
14004 CONCENTRATION OF ECONOMIC POWER
Table 3. — Funds for Capital Purposes, Class I Railroads and Their Lessor Com-
panies, 1921-1937
Period
From Income
From Reduction
in Working Cap-
ital
From Securities
Total
(millions
of dol-
lars;
Average
Annual
Total
Millions
of dol-
lars
Percent
Millions
of dol-
lars
Percent
Millions
of dol-
lars
Percent
(millions
of dol-
lars)
1921-30
1931-37_ -
$6, 003. 4
1, 402. 4
69. 10%
8G.29
$555. 1
414.6
6. 39%
25.51
$2, 129. 5
U91.8
24. 51%
•11.80
$8, 688. 0
1, 625. 2
$888.8
232.2
Source: Compiled from Tables submitted to the Temporary National Economic Committee by J. W.
Ballinger III. Record Vol. Ill, p. 404.
On an annual basis, funds for capital investments derived from income, from
reduction in working capital, and from the sale of securities averaged approxi-
mately 868 million dollars during the twenties, while the annual average from
the same sources during the period fram 1931 to 1937 amounted to about 232
million, only about 27% of the level during the more prosperous years.
Apart from the expenditures for capital investment during the period from 1921
to 1937, a further amount of 24.7 billion dollars was necessary to finance mainte-
nance (excluding depreciation and retirement). Of this amount about 18.2 billion
was spent during the 1921 to 1930 period and 6.5 billion in the 1931 to 1937 period.
On an annual basis, expenditures in the earlier period were 1.8 billion as compared
with 0.9 billion dollars during the depression years.^"
V. Demand for Steel by the Railroads
A. maintenance and capital expenditures
As a result of the decline in traffic and revenues, the railroads curtailed their
expenditures for maintenance and capital equipment. Not only has there been
less need for equipment and materials, but there have -been less funds available
to purchase these materials and equipment. Table 4 shows the drastic curtail-
ments in the subdivisions of the two major groups of expenditures:
Table 4.
-Class I — Railroad Maintenance and Capital Expenditures, 1923-1938
(Millions of dollars]
Maintenance Expenditures
Capital Expenditures
Year
Ways
and
struc-
tures
Equip-
ment
Total
Ways
and
struc-
tures
Equip-
ment
Total
1923 -
$813.7
792.7
816.4
866.8
868.6
837.9
855.4
705.5
530.6
351.2
322.3
365.3
394.0
454.8
495.6
420.2
$1, 465. 2
i;259!8
1,283.1
1.219.1
1.166.9
1,202.9
1,019.3
817.0
618.9
598. 7
637.9
681.9
783.0
826.7
676.5
$2,278.9
2, 052. 7
2, 070. 2
2, 149. 9
2,087.7
2, 004. 8
2, 058. 3
1,724.8
1, 347. 6
970.1
921.0
1,003.2
1,075.9
1,237.8
1, 322. 3
1,096.7
$377. 4
381.1
410.1
513.2
482.9
452.4
532.4
644.3
288.8
130.8
120!?
109.0
139.9
186.9
111.5
$681.7
493.6
338.1
371.9
288.7
224.3
321.3
328.3
73.1
36.4
15.5
92.0
79.3
159.1
322.9
115.4
$1,059.1
1924
874.7
1925
748.2
1926
885.1
1927 -'
771.6
676.7
1929 -
853.7
1930
872.6
1931
361.9
1932 ...
167.2
104.0
212.7
1935
188.3
1936
299.0
1937
509.8
226.9
Source: Reports of Carriers to Bureau of Railway Economics and published by Association of American
Railroads.
JO Testimony of J. W. Ballinger, III, before the Temporary National Economic Committee May 17, 1939.
Due to the diflerences ia methods of computation, these figures differ slightly from those given in text in
Table 4.
CONCENTRATION OF ECONOMIC POWER
14005
1. Maintenance Expenditures. — Of the two groups of expenditures, total main-
tenance outlays constitute about 40% of operating expenses, and, in general,
absorb one-third of total operating revenues. If the annual average from 1921 to
1928 is taken as 100, the index for maintenance expenditures fell to 44.3 in 1933
and recovered to 63.6 in 1937.2' In actual money terms, maintenance expendi-
tures declined from over 2 billion dollars in 1929 to less than a billion dollars in
1932 and 1933; they made some recovery in 1937, reaching 1.3 billion dollars.
Maintenance expenditures decreased more in the depression than revenue
ton-miles as shown in Chart 2. From 1929 to 1933 such expenditures dropped
Chart 2
RAILROAD TRAFFIC AND EXPENDITURES
CLASS I RAILROADS
900
800
700
600
500
400
300
CO
UJ
d 200
Z
z
o
»-
bu 100
O 90
60
^ 70
O 60
- 50
5 40
30
20
10
91)
e.o
7.0
6.0
5.0
4.0
3.0
CO
-1
§
0.8 V)
0.7 Z
0.6 2
0.5 d
ca
0.4
03
02
0.1
**"
_
^
s
k
;vEr
UETO^
-Ml
ES
y
\^
\J
*>
*w
\
-"
V
\
\
MA
EXP
\
NTE
END
DOU
ITU
JWS
CE
?ES
y
,>
\
^T
^*
h^
^A
V
/
1
V
//
UCAPII
AL EXPENDITUR
ES
lOLLA
«S)
11
1
rN
ll
[
/
^
Scum. Aaoca
iliiiiisliiiili;
3
%
55%, while traffic in terms of revenue ton-miles shrank by only 48% from 1929
to its 1932 low.22 Moreover, traffic resumed the upswing one year before these
expenditures turned up again. Roughly the course of both series is similar, for
the decline in railway services meant a reduction in the need for railroad mainte-
nance at about the same rate. The fact that maintenance expenditures decHned
more rapidly than traffic is explained by practices which Commissioner Splawn
of the Interstate Commerce Commission has called "continued skimping." ^^
In the upswing beginning in 1934, maintenance expenditures on the roads were not
expanded at the same rate as traffic represented by ton-miles. In 1937 traffic
was some 20% below the 1929 level, while these expenditures were 35% lower.
n See Appendix, Table 14.
" Ibid.
» Quoted in the Report of the Emergency Board to the President, (193S), p. 12.
14006
CONCENTRATION OF ECONOMIC POWER
The Interstate Commerce Commission in the Fifteen Percent Case, 1937-1938.
summarized the evidence on maintenance as follows:
"During the period of recovery in railroad traffic and earnings since 1932,
maintenance continued to be governed mainly by immediate needs dependent on
volume of traffic. A year or so ago, when the outlook seemed more encouraging,
a number of roads expanded their maintenance programs, but curtailed them again
sharply in latter part of 1937, when operating costs increased and traflSc declined.
The cumulative burden of deferred maintenance dating from the depression
therefore is still present. * * * Deferred maintenance of structures appears
to be particularly extensive."'^
2. Capital Expenditures. — The decline in the demand for railroad services
affected gross capital expenditures far more drastically than expenditures for
Chart 3
1000
900
800
700
600
500
400
300
to
% 200
-J
§
S 100
to 80
S 70
2 eo
d 50
=^ 40
30
20
10
GROSS CAPITAL EXPENDITURE?
CLASS I RAILROADS
1000
900
800
700
000
500
400
300
CO
20O 1
O
o
'^ ^
80 (rt
eo 2
50 d
30
If
20
)0
'
—
^
\
\
/
r
-
^>
.-*'
s
^.
WAYS AN
STRUaUR
3
ES
\
/'
\
/
\
\
\
\
\
\,
i
'/
\
>/
\
\
\
EQU
IPM
NTl
k 1
\;
V
i
SblKi: AsilxiX
maintenance (Chart 2 — Sec Table 4 and Appendix, 'J'able 8). Capital expendi-
tures t;hrank from an a\erage of 843 million dollars in the eight-year period 1923-
1930 to only 259 million dollars in the 1931-1938 period, a decline of nearly 70%.
These expendituriis fell from 873 million dollars in 1930 to 362 million in 1931, a
drop of almost f)0% in one year.
The capital outlay.s for new equipment exliibited greater lluctuatious than those
for way and .structures, as shown in Chart 3. Expenditures for capital equipment
" Fifteen J^crcerU Canf. {I'U l-'arlf. ISV. 1. C. (.;. U26, 41. «i). as » rcsiU*. of Iheatv ^h:lri> niiruiihueuts Id
niiiiiUonancc, ihero hfi.sboe,ii c<iusi<lerabln unflerin!iiDtt!n:\iK!e on tho rikilroiwis in lorms of past i>erformrtiices.
In » QU(«tionn).iro uimlyMS liisi yeiir tLo JnUiisUle Comiucrco Couinu&sion found that "at the close of l»3a
rhoto w;ls dcferri;il uiajntfu.uicfc ou ih;; railway propcrUes rtmouuting to $238 u:illioc bssuuutiR a tralli<' a.s
Jar^u at thai of 1«37 to be in prospect " Cf. I. C. C, ftr.jncla( Ki<jL.ir'.mr':U; of iMUria'j: til..tv;:JiLUt No.
3911, Matrl. 1639
CONCENTRATION OF ECONOMIC POWER
14007
amounted to some 300 million dollars annually during 1926 to 1 930.^5 In 1932
they were not much more than 10% of that amount; in 1933 they were 95% below
their 1929 level.^o They recovered to the 1929 volume in 1937 as a result of the
needs of the increased traffic, which forced railroads to make up in part for their
previous curtailment in expenditures of this type.
Reductions in capital expenditures for roadway and structures were relatively
moderate when compared with those for new equipment. Even these reductions,
however, were startling in their absolute magnitude; expenditures fell from ap-
proximately 500 million dollars in the predepression period to about 100 million
in the years of deepest depression. The rise in 1937 to less than 200 million
dollars was likewise much less spectacular than the corresponding change in
capital expenditures for equipment.
The difference in behavior of the two classes of capital expenditures is explained
by the difference in the means of adjustment to reduced demand for railroad
services. But utilizing idle stocks of cars, railroads were able to eliminate almost
completely their expenditures for new cars. When traffic increased as was the
case in 1936-7 and the idle stock was depleted, it became necessary to buy large
quantities of capital equipment.-' On the other hand, it was necessary thj-ough-
out the depression to make continuous use of the entire roadway and structures,
and expenditures for such facilities could not be curtailed as drastically as expen-
ditures for rolling equipment.^*
B. STEEL CONSUMPTION BT THE RAILROADS
The drastic reductions in expenditures for maintenance and capital equipment
by the railroads involved curtailment of the industry's consumption of steel
(Chart 4 — See tables 4 and 5). The available statistic? do not allow a clean-cut
division between maintenance and capital expenditures for steel, but the following
table on iron and steel consumption shows the general trend and the fluctuations
in the railroads' total consumption of finished steel and also in the railroads'
"direct" purchases of iron and steel products. 2»
Table 5.— Iron and Steel Consumption
by Railroads, 192S-19S8
Iron and Steel
Products as
Estimated
Iron and Steel
Percent of
Consumption
Products
Total RaU-
of Hot-Rolled
Year
Purchased '
road Pur-
Iron and Steel
(million.s of
chases of Fuel
Products >
dollars)
Materials
(thousands
and
of gross tons)
Supplies »
1923
$465 0
26. 7%
27.2
8 424
1924 -
365.6
7; 196
1925..
419.3
30.1
7,809
1926
507.3
32.5
7,656
1927
432.6
31.0
6,232
1928
397.5
31.3
6.119
1929
437.8
329.7
32.9
31.7
7 28.S
1930
4,679
1931..
202.1
29.1
2.710
1932
1,050
1,317
1933...
110.7
23.8
1934
139.8
26.0
2,271
1935
156.9
26.5
1,751
1936 ,.
273.8
34.1
3 645
1937
1938
162.2
26.1
1,289
Sources:
1 Data for years 1923 to 1929 and 1933 to 1938 from Association of American Railroads, Bureau of Railway
Economics. A Review of Railway Operations. Data for 1930 to 1932 from "/Jaitoay /Ipe." These figures
cover "direct" purchases only and are not comparable with the tonnage figures in the last column.
• Computed by apportioning individual hot-rolled product totals on the basis of Iron Age distribution
reports and by allocating jobber shipments to ultimate consumers. See M. W. 'Worthing, Distribution of
Steel Products to Major Consuming Industries, United States Steel Corporation, October 30, 1939.
» See Table 4 in text.
M See Table 4 in text.
>' See Appendix, Table 16.
>s See Appendix, Table 16.
" "Direct" purchases only cover purchases made directly by the roads and do not include the value of
material and supplies purchased indirectly for the railways by contractors who carry on construction work,
build equipment and do other work for the railways.
14008 CONCENTRATION OF ECONOMIC POWER
Like railroad revenues, dollar purchases of iron and steel products reached their
peak in 1926, fell to a post-war low in 1932 when purchases were 77% below 1929,
and recovered in 1937 to the level of 1930. Even then, they were still 30%
below the 1926 peak. For the five years beginning with 1926, purchases averaged
421 million dollars, and for the five years 1931-5 they averaged only 146 million
dollars, a decline of 65%. In tonnage terms, the railroad consumption of steel
amounted to 7800 thousand gross tons in 1925, fell 87% to a low in 1932, and
recovered to approximately 50% of the 1925 level in 1936 and 1937.
The "direct" purchases of iron and steel products decreased more rapidly in
the depression years, 1931 to 1933, and increased more rapidly from 1934 to 1937
Chart 4
RAILROAD STEEL CONSUMPTION AND RAILROAD
MAINTENANCE AND CAPITAL EXPENDITURES
1
L
»
\
/
V
i
\
/
\
\
\
r
ST
NSU
(TO
Ea
Mn
INSI
ION
\
.
\
i
=»t.
/
MA
AN
EXf
S
NTE
DC
•ENC
DOL
v\
NAr
APIT
ITU
AL
RES
\
1
J
vl
\
\
\
i
/
/
\
9
8
6 2
s i
4 <2
o> o ^ <y» G^ o^ <y^ o\ ffi
Scum, fmn Agt, Airwtican tion and 5te«/ inOAute, jnrf Astocittion ol AmttKan RsJfoaJi
than total purchases of fuel, materials and supplies. These differences in rate
of fluctuation are partly attributable to the fact that these iron and steel prod-
ucts are durable goods and their purchases can be more easily postponed than
such items as fuel, which must bear a direct relation to traffic. The Interstate
Commerce Commission found that, in part as a result of such deferability, one-
fourth of the total undermaintenance in 1938 consisted in lack of rail renewals.^"
Full renewals could have meant increased expenditures on rails of 43 million
dollars, if there were a return to the 1937 traffic, which was still approximatelv
20%, below the traffic level of 1929.3'
2' Expenditures on heavier rails reached a peak of $47.2 million in 1928, fell 75% to a depression low o f
$12.0 million in 1932 and recovered to $31.8 million in 1936. In general, rail purchases from 1923 to 1930
averaged $40.0 million per year but fell 40% to an average of $23.1 million from 1931 through 1937 (Report of
the Carriers to the Bureau of Railway Economics).
3' Cf. I. C. C, Financial Requirements of Railways, Statement No. 3911, (March, 1939).
CONCENTRATION OF ECONOMIC POWER
14009
C. THE INFLUENCE OF THE PRICE OF STEEL ON ITS CONSUMPTION BY THE RAILROAD
INDUSTRY
It has been shown that fluctuations in steel consumption are closely related to
the volume of railroad services, to business fluctuations and to the financial con-
dition of the railroads. In a setting of these basic factors, what role can the
price of steel play in its consumption by the railroads?
Some understanding of the effect of changes in the price of steel on the railroads'
consumption of steel can be had from Table 6 and Charts 5 and 6. Table 6
shows railroad expenditures for equipment and ways and structures in terms of
1929 dollars. The figures in this table were computed by dividing the actual
expenditures given in Table 4 by the appropriate railroad price indices for equip-
ment and road construction.^^ This result is a quantity index which shows
physical volume of investments measured in terms of 1929 dollars, rather than
the sums of money expended from year to year. If there were marked influences
of steel prices on the quantity of steel included in the total physical volume of
investment and therefore on total investment, rising quantities of total invest-
ments would tend to be associated with decreasing steel prices.
Table
-Railroad Expenditures at 1929 Prices, 1923-1935
[Millions of 1929 dollars]
Year
Equipment Expenditures
Ways and Structure Expendi-
tures
Total
Mainte-
nance
Expend-
itures
Total
Capital
Expend-
itures
Grand
Mainte-
nance
Capital
Total
Mainte-
nance
Capital
Total
Total
1923
1924
1925-
1926.
1927 _-
1928.
1929
1930
1931
1932
1933-
1934
1935
$1,376
1,288
1,355
1,372
1,239
1,248
1,203
1,024
894
752
727
702
704
$640
505
364
398
293
240
321
330
80
44
19
101
82
$2,016
1,793
1,719
1,770
1,532
1,488
1,524
1,354
974
796
746
803
786
$761
742
787
835
847
833
855
743
594
429
406
446
481
$353
357
395
494
471
450
532
573
323
160
112
147
133
$1, 114
1,099
1,182
1,329
1,318
1,283
1,387
1,316
917
589
518
593
614
$2, 137
2,030
2,142
2,207
2,086
2,081
2,058
1,767
1,488
1,181
1,133
1,148
1,185
$993
f&
892
764
690
853
903
403
204
131
248
215
$3,130
2,892
2,901
3,099
2,850
2,771
2,911
2,670
1,891
1,385
1,264
1,396
1,400
Source: Derived from Table 4 in text and from price indexes in S. Fabricant, Capital Consumption and
Adjustment, National Bureau of Economic Research, 1938, pp. 178-179.
Chart 5 compares percentage changes in composite steel prices with percent-
age changes in various types of physical investments of railroads. In Chart 5d
the physical volume arising from maintenance expenditures is evidently without
any gross relation to price changes. In the case of. total volume of physical invest-
ments, rising prices, as shown in Chart oa are associated with slightly rising
volumes. The same is true for the other two components, the volume of total
way and structure expenditures (Chart 5c) and total equipment expenditures
(Chart 5b). Chart 5e shows unmistakably, with respect to the physical volume
of capital expenditures, that the more steel prices rose, in comparison to the pre-
ceding period, the greater increases there were in the physical investment of the
railroads, and' thus in the quantity of steel consumed by the railroads. If any-
thing, these charts show, as a whole, that decreasing steel prices have not been
associated historically with rising quantities of physical investments.^^
A more direct analysis of the relation between steel prices and steel consumption
by the railroads is shown in Cliart 6 (based on Table 5 and Appendix Table 17).
This chart compares annual changes in steel consumption by the railroads with
changes i^ the composite steel price, which was used in the absence of an accurate
average price per ton for railroad steel.^^ If the annual purchases of steel by the
railroads were primarily influenced by changes in steel prices, a general correlation
between greater purchases and lowered prices should be evidenced. Chart 6
3! Taken from S. Fabricant, Capital Consumption and Adjustment, National Bureau of Economic Research
1938, pp. 178-179.
" See appendix. Table 17.
'• It should be noted that the data with respect to railroad steel consumption are subject to some inaccu-
racies, and that the changes in the average price per ton of railroad steel may differ somewhat from the changes
in the composite steel price.
14010 CONCENTRATION OF ECONOMIC POWER
fails to show such a relationship and warrants the inference that other factors of
much greater importance than price influence the quantities of railroad steel
purchases.
The foregoing analysis supports the vie>v that price considerations play rather
subordinate roles among the factors which determine the voUime of steel purchases
in the tailroad industry, as well as total steel consumption by the railroads. Steel
Chart 5
H_ 120
z
o
cc 100
UJ
a.
z
^ 80
z
UJ
i 60
i «>
^ 120
X
9 100
i
1-80
60
RELATION OF STEEL PRICES
TO RAILROAD PHYSICAL INVESTMENTS
1924 - 1935
100
80
60
40
fa)
I9ii
1526
1929
1934
1924
930*1.
27 19*3
1935
1931
^193.
TO
AL
(d)
1925
1925
1928*
193S
.934
1924 1<
19*30
bl933
• 1931
1929
1932
MAINTENANCE
80 §
(b)
192S
.929
1934
925*
1924^
"°192
1928.
1933*
^ ,932
1935
'.93.
EQUIF
WENT
(C)
1925
1924
1926»
1927
1929
• 1935
19-34
1930
1928
•
1933
1931
• 1932
VAYS
AND
ITRUC
rURES
,934
CAf
TAL
T
1930
•
1926*
1924
I92S
1927
• 1935
,933
1931
1^
•
85 90
95 100 105 110 115 85 90 95 100 105
STEEL PRICES IN PER CENT OF PREVIOUS YEAR
no
40
115
Source: Iron Aoe, Aisocatlon of American Railroads and Interstate Commerce Gynmlsshn
for maintenance is dependent to a high degree upon technical factors and to that
extent the volume of traffic dictates quantities required. Capital expenditures
for steel, on the other hand, appear to be dominated by such factors as availability
of sufficient funds and the necessity of making replacements and improvements
which were postponed during the depression. When necessitated by increased
demand for transportation, capital expenditures are likely to be made at a rapid
rate. Thus, general business conditions seem to determine the timing and extent
CONCENTRATION OP ECONOMIC POWER
14011
of those expenditures, which rise rapidly in a boom in spite of rising steel prices
and shrink during the depression to an unimportant fraction of their former level
irrespective of steel price decreases.
The comparative inefifectiveness of a reduction in steel prices to increase the
consumption of railroad steel «an be further demonstrated by a comparison of the
total cost of steel purchases with total railroad revenues. Ultimately the cost of
steel purchases is reflected in the cost to the consumer of railroad services. A
comparison of railroad revenues (here assumed to represent the cost to the con-
sumer of railroad services) with estimated total expenditures for railroad ateel,
during the period from 1923 to 1938 ^^ shows that for the sixteen-year period
expenditures for steel averaged about 5% of railroad revenues. If a reduction of
10% were made in the price of railroad steel and if this reduction were passed
on in the form of a rate reductidn, it would constitute a reduction of approxi-
mately one-half of one per cent in the cost to the consumer of railroad services.
As shown elsewhere herein, the increase in the volume of railroad services as a
result of rate reduction is low. (The elasticity of demand in the case of passenger
Chabt 6
5- 220
£ 120
52 100
RELATION OF RAILROAD STEEL CONSUMPTION
TO COMPOSITE STEEL PRICE
1924 - 1938
• 1936
1934*
1933,
"^=-
\nf
1937*
»». .1927 "
6
,1930
-.93,
193l'
1932
1938*
90 100 110
STEEL PRICE IN PER CENT OF PREVIOUS YEAR
180
160
220
120
Souirt. /ran Age an) Aintrian Inn ml StttI InsCtute
tr^fl5c was found to be approximately one half.*»). Hence a 10% reduction in
steel prices, if passed on to the consumer of railroad services, would serve to
increase the total services rendered by railroads only slightly and to increase the
steel consumed by the railroads probably by less than 1%. In addition, trans-
poration rates are more or less inflexible due to the institutional method by which
they are set up, and any changes in the cost of steel purchases,which is "a small
part of total railroad costs, could hardly be expected to be passed on in the form
of lower rates in the short run.
The inability of steel producers to increase railroad consumption of steel
substantially by a price reduction is apparent; and consequently, reductions in
steel prices by themselves could be expected to have little effect on railroad rates,
on total volume of railroad services, and on total steel consumption by the
railroads.
" The average computed here is only a rough approxiraation, for there are definite limitations in the stee'
expenditures data -ised. The figures were computed by multiplying estimates of railroad steel consumption
(which are subject to some inaccuracies) by the ;omposite steel price index in lieu of an average price per ton
for railroad steel. Even if the average price of railroad steel were slightly in excess of the composite steel
price index, It would not, however, materially change thetesults.
»« See: Supra, p. 6, and National Resources Committee, op. cit., pp. 90, 91 and 99.
124491 — 41— pt. 26 28
i4ai2
CONCENTRATION OF ECONOMIC POWER
VI. Summary
Steel costs form only a small proportion of the total costs of transportation,
and a change in steel prices could have practically no effect in increasing rail
transportation by reducing its cost to the ultimate consumer. Consumers'
income, industrial production and competition from other transportation media
have been the important factors influencing the volume of railroad services and
the consequent consumption of steel by the railroads. The demand for steel by
the railroads is inelastic; a given percentage price reduction will not induce as
large a percentage increase in the consumption of steel. The low consumption
of steel by the railway industry in the last decade has been due to the declining
trend in the demand for rail transportation, to the severe depression in business,
and to the financial difficulties prising therefrom. No conceivable reduction in
steel prices could have significantly altered the conditions arising from these forces.
Appendix
Table 7. — Estimated Consumption of Hot-Rolled Steel by the Railroads, 1928-1938
Year
Consumption
(thousands of
gross tons)
Percent of
Total Con-
sumption by
All Industries
Year
Consumption
(thousands of
gross tons)
Percent of
Total Con-
sumption by
All Industries
1923
8,424
7,196
7,809
7,656
6,232
6,119
41679
25. 3%
25.6
23.4
21.6
19.0
16.3
17.7
15.9
1931
2,710
1,050
1,317
2,271
1,751
3,645
('•,184
289
14 1%
1924
1932
10 0
1925
1933
7 9
1926
1934
12 0
1927
1935
7 3
1928
1936
10.8
1929
1937
11.4
1930
1938
6. 1
Source: Computed by apportioning individual hot-rolled product totals on the basis of Iron Age distri-
bution reports and by allocating, jobber shipments to ultimate consumers. See M. W. Worthing, Distri-
btUion of Steel Products to Major Consuming Industries, United States Steel Corporation, October 30, 1939.
Table 8.— Freight Traffic, 1921-1938
Revenue Ton-Miles
Year
Revenue Ton-Miles
Year
All roads
(miUions)
Class I roads
(millions)
All roads
(miUions)
Class I roads
(millions)
1921 -
309,533'
342. 188
416, 256
391,945
417,418
447,444
432,014
436,087
450. 189
306,840
339,285
412, 727
388. 415
413,814
443, 746
428, 737
432,915
447, 322
1930
385,815
311,073
235,309
250,651
270, 292
283,637
341, 182
362,815
1 313. 109
383,450
1922
1931 -
309.225
1923 - . .
1932 -
233, 977
1924 ..^c... .
249, 223
1925....
1934
268,711
1926..
1935
1936
282, 037
1927
339 246
1928
1937
360 620
1929
1938
290,154
' Year ended June 30, 1938. Other figures are for the calendar year.
Source : From Reports of Interstate Commerce Commission.
CONCENTRATION OF ECONOMIC POWER
14013
Table 9.
-Indexes of Industrial Production and Freight Car Loadings, 1921-1938,
Monthly Averages {1923-1925 = 100)
Industrial
Produc-
tion '
Freight Car Loadings »
Year
Industrial
Freight Car Loadings '
Year
Miscella-
neous
Merchan-
dise (1. c. 1.)
Produc-
tion!
Miscella-
neous
Merchan-
dise a c. 1.)
1921
67
85
101
95
104
108
106
72
84
97
97
106
109
108
111
lis
87
94
99
105
105
105
104
105
1930
96
81
64
76
79
90
105
110
86
99
78
55
58
64
69
82
86
67
97
1923
1932
1933
72
1934
1935
1926 „
1927 ....
1928 . -.
1937
67
1929. ...
Sources: Monthly data of these figures (as used in Chart 1) are found in Reports of the Division of Re-
search and Statistics of the Board of Governors of the Federal Reserve Board.
1 Released by Board of Governors of the Federal Reserve System, Division of Research and Statistics.
2 Federal Reserve Bulletin, June, 1937, pp. 624-527, and current numbers.
Table 10. — Passenger Miles, Consumers' Income, and Population, 1920-1937
Revenue
Passenger
Miles
(billions)
Consumers' Income
Midyear Es-
Year
(Billions
of current
dollars)
(Billions
of 1936
doUars)
timates of
Population
(miUions)
1920
46.8
37.3
35.5
38. <
36.1
35.9
35.5
33.6
31.6
31.1
26.8
21.9
17.0
16.3
18.0
18.5
22.4
24.7
$66.8
53.9
57.9
66.5
67.3
70.9
73.3
73.2
75.8
79.3
71.9
59.4
45.9
44.6
51.8
55.9
63.9
70.0
$47.0
42.3
48.8
54.8
55.4
57.0
58.-3
58.9
62.2
65.3
60.4
54.6
46.7
48.2
54.0
56.9
63.9
106.5
1921
108.2
1922
109,9
1923
111.5
1924
• 113.2
1925
114.9
1926
116.5
1927
118.2
1928 -.- ,
119.9
1929
121.5
123.1
1931 -
124.1
1932
125.0
1933 :
125.8
1934
126.6
1935-. ...--1
127.5
1936
128.3
1937
129.3
Source: Report of Industial Committee to National Resources Committee, Patterns of Resource Use
(1939).
Table 11. — Registrations of Passenger Automobiles,
1921-1938
Year
Number
(thousands
omitted)
Index
(1921 = 100)
Year
Number
(thousands
omitted)
Index
(1921 = 100)
1921. .
9346
■ 10863
13480
15461
17496
19237
20219
21379
23122
100.0
116.2
144.2
165.4
187.2
205.8
216.3
228.8
247.4
1930 .......
23059
22366
20644
21532
22563
24178
25450
25262
246.7
1922
1931 :..
239.3
1923 . ....
1932
223.5
1924
1933
220.9
1925
1934
230.4
1926
1935
241.4
1927
1936
258.7
1928
1937
272.3
1929
1938
270.3
boorce: Automobile Facts and Figures, 1939, p. 16.
i4ai4
CONCENTRATION OF ECONOMIC POWER
Table 12. — Total Operating Revenues, Total Operating Expenses, Taxes, and Net
Railway Operating Income, 1921-1938 (Class I Railroads)
Total
Operating
(thousands
of dollars)
Total Operating
% of total
revenues
Amount
(thousands
of dollars)
Net Railway Operat-
ing Income
Amount
(thousands
of dollars)
1921.
1922
1923
1924
1925.
1926
1927.
1928.
1929.
1930.
1931.
1932.
1933.
1934.
1935.
1936.
1937.
$5,516,598
5, 559, 093
&,289, 580
fi,'921, 496
6, 122, 510
6, 382, 940
6, 136, 300
6, 111, 736
6, 279, 621
5, 281, 197
4, 188. 343
3, 126, 760
3, 095, 404
3, 271, 587
3,451,929
4, 052, 734
4, 166, 069
3, 565, 000
$4, 562, 668
4, 414, 522
4, 895, 167
4, 507, 885
4, 536, 880
4,669,337
4, 574, 178
4, 427, 995
4, 506, 056
3, 930, 929
3, 223, 575
2, 403, 445
2, 249, 232
2,441,823
2, 592, 741
2,931,425
3, 119, 065
2,722,000
82.71%
79.41
77.83
76.13
74.10
73.16
74.54
72.45
71.76
74.43
76.97
76.87
72.66
74.64
75.11
72.33
74.87
76.35
$275, 876
301, 035
331,916
340, 337
358, »16
388, 923
376, 110
389,432
398,683
348, 554
303,628
275. 135
249,623
239, 625
236, 946
319, 753
325, 665
341,000
$600,937
760, 187
961, 955
973, 937
1,121,076
1, 213, 090
1, 067, 985
1, 172, 864
1, 251, 698
868,879
525,628
326, 398
474, 296
462, 652
499, 819
667, 347
690, 204
372, 846
10.9%
13.7
15 3
10.4
18.3
19.0
17.4
19.2
19.9
16.5
12.5
10.4
15.3
14.1
14.5
16.5
14.2
10.5
Sources: Report of Committee Appointed September tO,
Recommendations Upon the General Transpo)
Reports of Interstate Commerce Commission
J the President of the United States to Submit
Recommendations Uponthe General Transportation Situation (19SS), Table 12, p. 78. Based on
sion.
' 1938 figures from Association of American Railroads, Bureau of Railway Economics, A Review of Rail-
way Operations in 19S8.
Table 13. — Average Hourly Earnings, Average Dividend Rates on Capital Stock
Outstanding, and Interest on Debt, 1929-1937 (Class I Railroads)
Year
Average
Hourly
Earnings >
(cents)
Average Dv-
idend Rates'
(percent)
Interest on
Debt*
(millions of
dollars)
1929
66.6(S
67.8
68.9
63.6
62.9
63.6
69! 1
70.9
5.99%
6.01
3.99
1.12
1.16
1.64
2.11
2.07
511
1930
609
1931
618
1932
625
1933
524
1934
1935
611
1037....
' Committee of Public Relations of the Eastern Railroads, A Yearbook of Railroad Information (1935 and
1938 editions), p. 62.
' Quoted in Report of the Emergency Board to the President, (1938), p. 9. The capital stock outstanding
Includes stock of railroads In trusteeship or receivership. Also based on data from Statistics of Railways
in the United States, Interstate Commerce Commerce, 1937.
« Committee of Public Relations of the Eastern Railroads, op. cit., p. 38.
„ - CONCENTRATION OF ECONOMIC POWER 14015
Table 14. — Maintenance Expenditures and Revenue Ton-Miles, 1921-1938 (Class
I Railroads)
Mainteuance Expendi-
tures
Ton-Miles of
Revenue
Freight
(millions)
Year
Maintenance Expendi-
tures
Ton-MUes of
Revenue
Freight
(millions)
Year
(millions
of dollars)
Index Av-
erage 1921-
1928=100
(mUlions
of dollars)
Index Av-
erage 1921-
1928=100
1921
1922
$2,008
1,981
2,279
2,053
2,076
2,150
2,088
2,' 005
2,058
96.6
95.3
109.6
98.7
99.8
103.4
100.4
96.4
99.0
306, 840
339, 286
412, 727
388,415
413,814
443, 746
428, 737
432, 915
447,322
1930
1931
1932
1,725
1,348
970
921
1,003
1,076
1,238
1,322
1,097
82.9
64.8
46.6
44.3
48.2
61.7
69.5
63.6
62.8
383, 460
309, 225
1924
1925
1926
1927
1928 -
1929
1933
1934.
1935
1936
1937...
1938
249, 223
268,711
282, 037
339, 246
360,620
290,164
Source: Association of American Railroads, Bureau of Railway Economics. A Review of Railway Opera
tions in 1938 (p. 20), and Interstate Commerce Commission, Statistics of Railways in the United States
Table 15.
-Railroad Equipment Expenditures, Available Freight Cars, and Car-
loadings, 1921-19SS
Year
Average
Freight
Cars Owned
(thousands)
Average
Service-
able Cars
(thousands)
Surplus Cars
Lowest
Reported
(thousands)
Carloadings
(thousands)
Equipment
Expenditure
(mUlions
of dollars)
1921
2,317
2,304
2,303
2,331
2,365
2,345
2,329
2 298
2,267
2,270
2,229
2,160
2,072
1,969
1,863
1,770
1,723
1,713
1%
2,113
2,146
2,168
2,190
2,191
2,164
2,132
2,128
2,053
1,922
1,779
1,674
1,584
1,526
1,643
1,496
80
4
14
99
112
81
135
104
119
393
636
545
318
208
112
104
139
966
1,000
1,098
1,113
1,124
1,209
1,129
1,197
1, 202
085
776
651
687
646
734
826
847
726
311
1922
246
1923
682
1924 . . . -
494
338
1926
372
1927..
289
1928 ......
224
1929
321
1930
328
1931.
73
36
1933
16
1934
92
1936
79
1936
169
1937
323
1938- .
115
Source: Taken from testimony of Dr. Lauchlin Currie before the Temporary National Economic Com-
mittee on May 16, 1939, Volume 3, Table IV, p. 357. For note^ and methods see Dr. Currie's appendix.
14016 CONCENTRATION OF ECONOMIC POWER
Table l&.— Freight and Passenger Miles, Mileage Operated, Equipment in Service,
and Installalion of Equipment and Rails of Class I Railways, 1926-1937
1?
s
o
Equipment in
Service, End of
Year
Rails Laid in Replace-
ment (thousands of
gross tons)
New Equipment
InstaUed a
^R
•£S
1%
Year
f
ai
sf
II
til
_
•a
S
c
1
SO
2
1
1
c
ti
11
fe
p-l
^
p^
Ph
t-
^
CO
«
ft.
1926
486
35.5
394. 9
62 8
2,349
54.8
3,818
2,210
1,608
472
474
33.6
31.6
399.2
403.5
61.4
59 5
2,325
2,298
53.8
53. 1
3,819
3,806
2,125
2,080
1,690
1,725
1928
1929
490
31.1
406.5
57 5
2,277
52.3
3,610
1,958
1,652
1930
4^0
26.8
408.0
56.5
2,277
52.1
2,764
1,517
1,157
1931
338
21.9
408.2
55.1
2,201
50.7
1,715
985
730
1932
257
17.0
407.1
53. 2
2,145
49.4
798
395
403
90
2,815
58
1933
273
16.3
404.9
50.8
2,035
46.5
403
459
14
1,936
7
1934 -
•296
18.0
402.4
48.2
1, 938
43.8
1,165
631
534
90
23, 948
270
1935
310
18.5
400.3
46.5
1,836
41.6
1,159
583
576
139
6,987
225
1936
372
22.4
397.8
45.0
1,758
40.6
1,701
921
780
98
37, 554
159
1937
395
24.7
395.6
44.4
1,744
40.3
1,975
1,030
945
441
69,118
576
Source: Interstate Commerce Commission, Statistics of Railways, 1937, and earlier volumes.
' Total of revenue and non-revenue ton miles.
' Total of steam and electric locomotives.
5 Data prior to 1932 not available. Data for domestic shipments of locomotives (U. S. Bureau of the
Census) and freight and passenger cars (American Railway Car Institute) indicate the following changes
between 1929 and 1937; locomotives, — 36.5<j;, freight train cars, — 10.97ci and passenger train cars, —57.6%.
See: Survey of Current Business, 1938 Supplement, p. 162.
Table 17.— Composite Price of Finished Steel, 1923-1988
Composite
Steel Price
(cents per
pound)
2.697
1924 ■.... 2.505
1925 2.334
1926 - 2.315
Source: Iron Age (January 5, 1939), p. 198.
1932.
1933-
1934.
Composite
Steel Price
{cents per
pound)
1.957
1.901
1.879
2.033
_ 2.058
2.077
2.464
2.394
Exhibit No. 1415
AN ANALYSIS OF THE DEMAND FOR STEEL IN THE CONTAINER
INDUSTRY
This is an analysis prepared by the Special Economic Research Section of
United States Steel Corporation, composed of Messrs. Edward T. Dickinson,
Jr., Ernest M. Doblin, H. Gregg Lewis, Jacob L. Mosak, Mandal R. Segal,
Dwight B. Yntema and Miss Marion W. Worthing. The work of this group
was under the supervision of Theodore O. Yntema, Professor of Statistics, Uni-
versity of Chicago. This analysis was written by Mandal R. Segal who had the
b'^nefit of suggestions from other melnbers of the staff. It is issued by United
States Steel Corporation.
November 1, 1939.
CONTENT.?
I. Purpose.
II. Findings.
III. The Container Industry.
IV. Steel Consumption by the Container Industry.
V. Relation of Price to Demand tor Tin Plate:
A. Demand for Food or Packer Cans.
B. Demand for Oencral Line Cans:
(1) Beer Cans.
(2) Lubricant Oil Cans.
(3) Paint and Varnish CanS.
(4) Other General Line Cans.
C. Quantity of Tin Plate per Can.
VI. Summary.
CONCENTRATION OF ECONOMIC POWER 14017
I. Purpose
The purpose of this study is to analyze the factors determining the quantity
of steel consumed by the container industry ' (of which tin can manufacture is
the greater part) and particularly to determine the importance of price as a
factor in this consumption.
Steel (mainly in the form of tin plate) is the principal raw material in the
production of tin cans, which are used mainly to pack consumers' goods, such as
food, oil, beer, and paints and varnishes. Thus, the consumption of steel by
the tin container industry depends upon:
(1) The output and consumption of all "canned" goods, i. e., all goods that
are put in containers;
(2) The proportion of this output that is put in tin cans and not in containers
made of other materials; and
(3) The amount of steel used in the individual container.
A change in the price of steel can affect steel consumption by the tin container
manufacturing industry only in so far as it affects these three factors.
II. Findings
(1) The demand for products packed in tin cans is dependent primarily upon
consumer income.
(2) There has been an upward trend in tin can consumption, which is explained
by the increasing use of tin can§ to pack new food and other products. This
adaptation of tin cans to additional uses in the past ten years has also contributed
to the relative cyclical stability of tin can consumption.
(3) Public tastes and preferences have been among the most important factors
limiting the substitution of tin cans for other containers. The physical and
chemical characteristics of the products to be packed have also tended to restrict
substitution and adaptation of tin cans to new uses. Existing equipment and
distribution facilities constitute another deterrent to substitution.
(4) Other investigations have shown that the demand for food is generally
inelastic. The available data also indicate that fluctuations in the total consump-
tion of canned food products have had little net relation to fluctuations in canned
food prices or to fluctuations in the ratio of canned food prices to other food
prices. From these facts it seems reasonable to infer that the demand for
canned food has a low elasticity.
(5) The price of tin plate is about one-tenth of the retail price of representative
food products packed in tin containers and is probably less than one-tenth of the
retail price of most products packed in general line cans. A 10 percent reduction
in the price of tin plate, if entirely passed on to the consumer of the canned
product, would therefore reduce the retail price of the canned product only about
1 percent. In the case of canned food products, this would result in a saving
to the consumer of only a small fraction of one cent per can. Consequently, any
reduction in the price of tin plate, even if the consumer of the canned commodity
received the full benefit of such price reduction, could affect the final price of the
canned commodity only slightly and could have no appreciable effect on the con-
sumption of canned goods.
(6) In the past there has been no discernible relation between the consumption
of representative individual canned food products and the price of tin plate. This
simply reflects the major importance of other factors and the very minor impor-
tance of the price of tin plate in determining the consumption of canned foods.
(7) Physical and chemical requirements of the.products to be packed determine
almost entirely the weight, composition and amount of the steel used in tin
containers.
These findings show that tin plate prices have only a negligible effect on the
immediate factors determining steel consumption by +he container industry,
namely:
(a) the output and consumption of all "canned" goods;
(b) the proportion of this output that is put in tin cans and not in other types
of containers; and
(c) the amount of steel used in the individual container.
To sum up, the consumption of steel by the container industry is derived
from and dependent upon the consumption of the products packed in tin con-
tainers. A change in the price of tin plate can increase its consumption only by
1 Although the container industry might be'defined as comprising all types of containers, including many
made of materials other than steel, this study is limited to "lightcontainers" made from light steel products,
and is devoted primarily to tin cans, in the manufacture of-which roughly three--qaarters or more of the
steel taken by the container industry is consumed.
14018 aONOENTRATION OF ECONOMIC POWER
the reduced cost being passed on to the ultimate consumer of the canned product
and "only to the extent that the consumption of canned products is increased
thereby. Taking into account the low elasticity of demand for canned products,
the limited degree of substitutability attributable to price, and the relative un-
importance of the price of the tin container in the price of the packaged product,
it is evident that the demand for tin plate is very inelastic. Fundamental fac-
tors such as consumers' income and public tastes determine the demand for
canned commodities and, hence, the consumption of steel by the tin-can indus-
try. The price of the steel consequently is of minor significance in determining
this consumption.
III. The Container Industry
The demand for steel by the container industry is a derived demand, i. e., it
is dependent upon the factors affecting the demand for containers. Conse-
quently, any study of the container industry's demand for steel must first deal
with the factors which have a bearing on the demand for products packed in tin
containers.
The active growth of the canning industry in this country began in the period
of the 1860's and 1870's; by the turn of the century canneries were started all
over the country for fruits, vegetables, and fish.^ Since then the tin container
manufacturing industry has progressed rapidly. Whereas in 1904 this industry
had a value of product 0^41 million dollars, by 1937 the product totalled 350
million dollars. In the latter year more than 16 billion cans were manufactured
and the "tin can and other tinware" industry paid out 37 million dollars in wages
and 245 million dollars for materials, fuel, and purchased electric energy. ^ It
has been estimated that approximately 100 cans per year are made in the United
States for every person in the country.^
Tin plate is suitable for containers for a wide variety of products on account
of its cheapness, strength, lightness, nontoxicity, pleasing and sanitary appear-
ance, ease of fabrication, durability, bright reflective surface and suitability for
coating and decorating by lithographing. The use of the tin can has therefore
expanded in the food packing field and has been recently extended into new
fields, such as motor oil, paints and varnishes, pharmaceutical and toilet articles,
beer and tobacco.
In contrast to most other industries consuming steel, the tin can industry has
maintained a comparative cyclical stability, with small fluctuations even in the
depression period. This is due to the fact that canned goods are mostly perish-
able consumers' products, i. e., production is followed more or less immediately
by consumption and the factor of durability is usually not involved since most
cans are used only once.
A second characteristic of the tin can industry is the relatively steady upward
trend in can production. Basically, this trend is caused by new and more di-
versified uses for tin cans. These, in turn, are associated with public tastes and
preferences and technological advances in the steel and container industries.
The products of the can industry can be grouped under two heads:
(a) Food or Packer cans, which include most of the tin containers for food-
stuffs, comprise approximately 55 percent of the total value of all products of
the tin can industry. More than 11 billion of the 16 billion cans produced in
1937 were of this type.^ In 1935 the packer, or food, cans « were distributed
among products as follows: 57 percent were used for vegetables, soups, fruits
and juices; 25 percent for evaporated milk; 15 percent for fish and meat; and
the remaining 3 percent were used for food specialties and dry and sweetened
milk.^ Of the foods canned, there are many varieties, including 46 classes of
canned vegetables, 33 classes of fruits, 10 classes of juices, 27 classes of fish and
shellfish, 41 classes of specialites, 23 classes of meats, 37 classes of soups and 8
classes of ready-made entrees.*
> National Canners Association, The Canning Industry, March 1, 1939.
' Census of Manufactures, 1937, "Tin Cans and Other Tinware", release October 12, 1938.
^'International Tin Research and Development Council, Tin Plate and Tin Cans in the Vnited Stales
(1936), p. 6.
» See -Appendix, Table 7.
• Although there are 27 different sizes of sealed cans used by commercial canners, in general the packers
cans are standardized in size and shape.
' Evaporated Milk Association, Some Fads About Ecaporated Milk and Other Dairy Products, p. 19.
• American Can Company, The Canned Food Handbook, pp. 14-16.
CONCENTRATION OF ECONOMIC POWER
14019
(b) General line cans totalled nearly 5 billion in number in 1937 and represented
45 percent of the value of the product of the tin can industry." These general line
cans are of innumerable sizes and shapes ranging from tiny pill boxes to large oil
cans and are used for countless varieties of commodities which do not require heat
treatment in the packaging process.'" Among the most important products packed
in general line cans are oil, beer, paints and varnishes, bulk milk (which is placed
in large dairy milk cans), tobacco, toilet and pharmaceutical articles, and wax.
It should be noted that beer, ice cream, dairy milk, and coffee cans are classed in
this category of general line cans in spite of the fact that the products contained
in such cans are food products.
IV. Steel Consumption by the Container Industry
The consumption of st^el as a raw material in the manufacture of tin cans and
other light containers has shown a substantial upward trend since 1923. In the
period from 1923-1929 this industry consumed an annual average of 1.42 million
gross tons of steel but in the period from 1932-1938 its consumption averaged
1.95 million gross tons, an increase of about 37 percent." The industry's relative
position as a steel consumer has also risen; in 1923 it took only 3.6 percent of the
total finished rolled steel produced in this country, but since 1932 it has taken
nearly 9 percent of the total output. In 1938 it ranked third among consuming
industries, accounting for 9.1 percent of the total production of finished steel. The
upward trend and increasing importance of the container industry as a consumer
of steel can be seen from the following table: '^
Table 1. — Estimated Consumption of Steel by the Container Industry, 192S-19S8
Year
Consumption
of Hot-Rolled
Iron and Steel
(thousands of
gross tons)
Percent of
Total Hot-
Rolled Stee
Production
Year
Consumption
of Hot-Rolled
Iron and Steel
(thousands of
gross tons)
Percent of
Total Hot-
Rolled Steel
Production
1923
1,205
1,210
1,427
1,348
1,408
1,619
1,707
1,670
3.6%
4.3
4.3
3.8
4.3
4.3
4.2
5.7
1931.
1,415
1,037
1,759
1,557
2,039
2,455
2,874
1,908
7.4%
1924..
1932....
1925
1933
1926
1934
8 2
1927
1935
8.5
1928
1936
7 3
1929
1937
7 8
1930
Source: Computed by apportioning individual hot-rolled product totals on the basis of Iron Age dis-
tribution reports and by allocating jobber shipments to ultimate consumers. See M. W. Worthing, The
Distribution of Steel Products to Major Consuming Industries, United States Steel Corporation, October 30,
At least 70 to 80 percent of the steel consumed by the container industry is
used in the form of tin plate and terne plate in the manufacture of packer and
general line cans. In the production of tin cans the most important raw material
is tin plate which consists of about 98 percent steel and 2 percent tin.'^ The cost
of tin plate represents approximately 60 percent of the value of the tin can."
Terne plate, while similar to tin plate, contains some lead andjis primarily used,
therefore, for non-food products such as lubricating oil.
Due to the relative stability and upward trend in the consumption of tin plate,
mills engaged in its production were operating at from 60 to 90 percent of capacity
during the depression while the steel industry as a whole was producing at from
« General line cans and packages include tin boxes and pails. In general the total value of product of
packers and general line cans listed here was 335.0 million dollars, which was 93.4 percent of the total value
listed for the tin can and other tinware industry. The remainder, i. e., 18.5 million dollars, represented
finished tinware other than cans.
10 Beer is an exception, requiring heat treatment.
" See Table 1 in text.
» The series of "Tin Plate an4 Terne Plate" taken from the Census oj Manufactures shows a similar
course. See Appendix, Table 9. Differences in the two tables are probably due to varying methods of
computation and different coverage and classification.
'3 National Canners Association, Tlie Story of the Tin Can, and American Can Company, A Word about
Tin Cans, p. 1.
i< See Appendix, Table 8.
14020
CONCENTRATION OF ECONOMIC POWER
15 to 60 percent of capacity. '^ The upward trend in the consumption of tin plate,
most of which is used in the production of tin cans," can be seen in the following
table:
Table 2. — Consumption of Tin Plate in the United States, 1923-1935
Total Con-
sumption
(thousands
of gross tons)
Tin Plate Used for Packer
Cans
(Thousands
of gross tons)
(Percent of
total)
1,291
1,383
1,348
1,541
1,293
1, 579
1,560
661
719
702
715
738
925
51%
52
52
58
55
47
59
Source: International Tin Research and Development Council, Statistieal Yearbook, 1939, p. 134. These
figures differ from the estimates given in Table 1 due to differences in^overage and construction.
V. Relation of Price to Demand for Tin Plate
Total consumption of steel by the container industry, most of which is in the
form of tin plate, is compared with an index of the price of tin plate from 1923 to
1938 in Chart 1." This chart shows the marked upward trend in consumption
of steel by the container industry during this period. It is evident from the chart
that the relatively large fluctuations in steel consumption must be due to influences
other than changes in the price of tin plate.
In order to determine the relative importance of the influence of the price of tin
plate on its consumption in a more detailed manner, the two major branches of
the container industry, i. e., packer and general line cans, are analyzed to discover
the extent to which the tin plate price affects:
(1) the output and consumption of all canned goods, i. e., all goods that are
put in containers;
(2) the proportion of tliis output that is put in tin cans; and
(3) .the amount of steel in the tin cans.
A. dem'and for food or packer cans
The relative stability in the production ot food cans, which constitute 55 percent
of the total value of all tin containers and consume approximately the same per-
centage of the total quantity of tin plate taken by the container industry, is due
to the fact that food in general is a necessity. P'ood expenditures are the first and
greatest expense item in the budget, representing approximately 35 percent of the
expenditures of the middle income class 's and higher proportions in the lower in-
come groups.'' While consumers' income declined 28.5 percent from 1929 to the
low of 1932, food production fell only 10.3 percent. In the recovery through 1936,
consumers' income rose 36.8 percent, while food production gained only 3.4 per-
cent.^" Thus, total food production has been remarkably stable, showing rela-
tively less fluctuation than has consumers' income (Chart 2). The consumption
of canned foods ^i has fluctuated in the cycle in close relation to consumers' income
but with somewhat greater amplitude than total food production (Chart 2).
■5 International Tin Research and Development Council, Bulletin 4, Tin Plate and Tin Cans in the
UnUed States. 1936, p. 6. . ^ ^^ ■. ^ c-. .
'6 International Tin Research and Development Council, Tin Plate and Tin Cans m the United, Stales
(Bulletin 4, Oct., 1936), p. 77. In addition, .see Iron Age (March 10, 1936), p. 58A.
i: For supporting data, see Appendix, Table 10. . „.., ^ . ^ r- ■
IS Works Progress Administration, Division of Social Research, Inter-city Differences in Cost of Living
(March, 1935) 59 cities. ^ . ^ u , 4
■ » A recent survev by the United States Department of Labor found that: "Although food accounts for
approximately two-fifths of the total expenditures of families at the bottom of the income scale, only one-fifth
or less of the expense of families in the highest income level is accounted for by food." (United States De-
partment of Labor, Bureau of Labor Statistics, How Urban Families Spend Their Incomes, Release, July 10,
19.38.)
" See Appendix, Table 11. , .. ^ u i
21 This close relationship between canned foods and consumers' incomes was recently borne out by several
studies by the Division of Statistics of the Natio.nal Canners Association. These studies found that the
chahges in emplovment and consumers' buying power are cf great importance m determmmg both the prices
and consumption of the h-ading canned vegetables: cf. National Canners Association, padon Affecting Fro-
duciion and Distribution of Canned Peas, Corn, Tomatoes, and Snap Beans, and also CanniJ/ S^el Corn (igs^i.
CONCENTRATION OF ECONOMIC POWER
14021
"Vhe relative importance of consumers' income in the determination of the con-
sumption of canned foods is indicated in a study by the National Resources Com-
mittee reported in "Patterns of Resource Use." 22 There it is shown that 88
percent of the variation in the consumption of canned fruits and vegetables is
explained by two factors, consumer income and a time trend. Furthermore, if
the residual variation in consumption not accounted for by these factors is plotted
against an index of canned food prices, or against the ratio of canned food prices
to general food prices, no relationship is evident. Thus the price of canned foods
appears not to have been an important factor in determining their consumption. ^^
Chart 1
STEEL CONSUMPTION BY
CONTAINER INDUSTRY AND TIN PLATE PRICES
inn ' '>~»
9.0
ao
7ja
6.0
so
AD
ao
1 -
0
" 1.0
u. 09
0 03
0.7
s «
i "
03
02
0.1
1
900
800
700
600
900
400
300
200
cc
Ml
00
^" i
60 i
50
40
30
20
10
CO
s
STl
><SU
<T0
\
.EL
VI PT
MS)
ON
"7
/
/
\
-
/
i!—
^
^i
=-
—
— '
W
f^
h—
h-
—
....
BTr
P
RICES
I
e vd Amenan Imn vrd StK-l Institute
The role of the price of steel entering into the tin can in which food is packaged
is even less important. It can only affect the consumption of canned products
by affecting their price. But the cost of steel used in tin cans constitutes only
a small part of the retail prices of the canned products'. In an analysis of eleven
leading canned vegetables and fruits, it was found that in only four cases was the
cost of tin plate more than 10 percent of the final retail price of the product; in
two cases, the cost of tin plate was less than 5 percent of the final price; while
in the remaining five products, the cost of tin plate varied from 5 to 10 percent
of the retail price. The following table lists the products, their retail prices, the
32 Op. cU., p. 103.
" The investigations by Henry Schultz, Theory and Measurement of Demand, Chipago, 1938, show that
the demand for most agricultural products is inelastic.
14022
CONCENTRATION OF ECONOMIC POWER
Chart 2
CONSUMPTION OF CANNED FOODS,
FOOD PRODUCTION AND CONSUMER INCOME
"900
800
700
600
500
400
300
200
CO
ce.
OD
X 80
s ^°
Z 60
50
40
30
20
10
1
12
70
&
50
40
30
CO
20 1
-1
§
■0 S
8 to
M
5 -d-
00
4
3
2
1
___^
^
r=»^
^
-=
s=
=
:J_
consumeF^
^
J
^
A
IlNCOMEp
(1935DpLLARSP
M
c
c
/
IN
ANN
0.
)EX
ED
OF
FOO
TIO
3
1^
-
\
^
-^
/
A
-<*
r"-
~*n^<
—
•■»
•J
^^Ts
^
'
'^Tdex oN
/
mnn
f
PRODUCTION
P««;«« Ovrn/t^t. ■■P^»^ cfResouxe Us»"3nJ U.S.B.LS.
5
itimated tin plate cost, and the estimated proportion that such tin plate cost
orms of the total selling price.
Table 3.— Tin Plate Costs in Relation to Retail Prices of Canned Goods, 19S8
Commodity
Retail
Price >
(cents)
Estimated
Tin Plate
Costs •
(cents)
Tin Pkte
Cost as Pro-
portion or
Retail Price
(percent)
8.8^
7.4
U.2
1L6
16.1
18.0
21.3
22.4
13.4
29.5
26.7
1.22e
0.88
1.22
1.22
1.22
1.48
1.48
L48
0.87
L22
0.87
13.9%
11.8
10.9
4. Corn— No. 2 Standard
10.6
6. Peas— No. 2 standard „
6. Peaches— No. 2Hcan- .■...
8.1
8.0
7. Pears— No. 2]4 can . .
6.9
8. Pineapple— No. 2}^ can . ' ^
6.6
0. PinV KfilTTion — IfioJ! Cfln, Yftll
6.6
10. Asparagus— No. 2 can
11. Canned Sal on— 16 oe. o^,.tall ;..
4.1
3.4
Sources:
> United States Bureau of Labor Statistics, Bulletin 635, Indexes of Retail PH«e of Food In 51 CUies, (1938).
• Calculated by assuming a 60 percent steel proportion (see supra, p. 6, and Appendix, Table 8) in the
price of tin cans, which in 1938 were as follows:
No. 1 TaU can— 16 oz _ $14.56 per 1,000 cans
No. 2W Can— 1 lb. 1? oz.. 24.71" "
No. 2 Can -1 lb. 3 oz ,.. 20.32" "
Almanac of the Canning Indxutry, 1939, p. 236, published by The Canning Trade.
CONCENTRATION OF ECONOMIC POWER
14023
Any practical reduction in the price of tin plate, if fuUy reflected in the price
of the canned food products, could reduce the final product price only slightly.
Even a 10 percent reduction in the price of tin plate, if passed on to the consumer,
would by itself lower the retail prices of most of these canned food products by
less than 1 percent, or approximately one-tenth of one cent per can.
Chart 3
RELATION OF PACKS OF TOMATOES, PEAS
AND CORN TO PRICE OF TIN PLATE
1921 - 1938
TOM/
TOES
1922
1934
I
1925
1927
930* 1935
• 1929
• 1923
1937
192.
1933 •
1931
1938*
,•1528
t-1936
la24
1926
180
PE
\s
1922
•
1935
1937
1933 •
19i8
• 1930
1938»
iWs*
1923
1934*
1921
•
1332 «
1926'
^1931
1925
1927
CO
^N
1925
1935
*937
193!
1933*
1923
1929
.934
•
.19^
1930*
1936
^19?-?
1926
192!
1932*
.3?a
VIZI
80 F5 ^tO 95 100 i05 HO 115 - !?0
TIN PLATE PRICE *n PER CENT 0" PREVIOUS VLAK
20 w
180 2
140 ^
o
100 h_
UJ
O
60 a:
UJ
a.
20 ^
220 -
t/>
130 o
o-
140
100
60
20
Source: Almonx Of Ccnnirg InJusty cr.o Iron Aqc
The relative unimportance of the price of tin plate, so far as consumption of
of canned food products is concerned, is confirmed by Chart 3 where percentiige
changes in the production of canned tomatoes, peas and corn arc compared with
percentage changes in the price of tin plate.-* Since changes in the production
of these products have not been associated historically with changes in price
of tin plate, they must have been due mainl\ to the inilueuce cf other facturf .
" See also Apijeudli, Table 12.
14024 CONCENTRATION OF ECONOMIC POWER
To summarize: The cost of tin plate is a very small element in the price of
canned foods and therefore changes in the price of tin plate can have but very
slight effect on the price and consumption of canned foods.
There is still the question of the possible effect of the price of tin plate on the
proportion of canned food output that is put in cans rather than in containers
made 'of glass.^' Among the factors important in determining this proportion are:
(1) the relative prices of the different types of containers; (2) public taste, custom
and convenience (as well as the possibility of changing these by advertising) ; (3)
technical limitations on substitution; and (4) the existing bottling and canning
equipment in the consumers' goods industries.
Although exact information on prices of the two types of containers is not
available, a rough approximation based on the 1935 Census of Manufactures shows
that the average price of glass containers of food products was slightly more than
2^ a piece and the average price per food can was slightly less than 2jf.28 These
prices are, however, not strictly comparable, since they cover all types and sizes
of food containers, many of which are not in direct competition,^? and since they
do not take into account the fact that the tin can is used only once whereas the
glass container may be a semi-durable commodity used more than once. How-
ever, when the glass food container is not returned,^^ the total original cost must
enter into the final product price of the glass-contained product just as the total
initial cost of the tin can must enter into the selling price of the canned goods.
Although in general the price of tin cans for food is slightly lower than the price
of glass containers, there are more important factors determining the use of cans
for most foods.
First of all, food products in cans have come to be accepted and even pre-
ferred because of certain advantages which this form of packaging possesses.
Cans are not breakable; they are readily fabricated in desired sizes, weigh less,
require less space, and are ea' ier and cheaper to ship' and handle than bottles
or glass containers. Juices wl oh "cloud up" do not lose their sale:; value in cans.
It is not necessary to return tl ; cans and no deposit or bookkeeping is involved.^'
Habit and custom are im^jortant determinants of preferences for different
types of containers. Most vegetables, fruits, evaporated milk and fruit juices
have found acceptance in cans, while bottles are preferred for fresh milk and
certain beverages. Of late there has been a growth in popularity of cans over
bottles for the packaging of beverages and juices for consumption in the home;
experiments are being carried on in the canning of ginger ale, coca-cola, and other
liquids. On the other hand, bottles are being used to package certain high priced
vegetables and fruits.^" Thus, public taste and custom,^' which are of major
importance in determining the proportion of foods packaged in. cans, tend to
limit the substitution of cans for glass containers and vice versa in response to
price changes.
The lines of competition among the various types of containers have also
recently been extended by the use of frozen foods. In the past few years there has
been a marked increase in the packaging of frozen fruits, which reached a total of
117.8 million pounds in 1938.^^ Part of the frozen food pack is packaged in cans
of ordinary food sizes (No. 10 can's) and in larger tin containers (10-50 pound
and 1-5' gallon cans), but the majority of the frozen foods is found in 1 pound
cartons (cups and boxes). In general, however, these frozen foods have not cut
into the sale of vegetables and fruits packed in tin containers, because (1) the
frozen foods compete mainly with fresh foods, (2) they are higher priced than
canned foods, (3) many homes are not equipped for storage, and (4) many vege-
tables and fruits have not yet been successfully frozen.
In addition to the public taste, the chemical and physical properties of foods
give rise to technical limitations on the kinds of food put in tin cans.
Substitution of bottles for cans, or vice versa, is restricted by the costliness of
such a shift. The change from the use of bottles to tin cans involves the cost of
2s While paper container.s are useful for many purposes, they seem to compete more directly with^lass
than tin. In any event, competition with tin plate is greatest from glass and the discussion is confined
primarily to that problem. Many of the same considerations, however, would apply to competition from
other types of containers, including paper.
2« Census of Manufactures, 1935, pp. 842 and 958.
'■ See infra, p. 15.
" Beverageand milk bottles are the exceptions here. Their original costs are approximately 2.4d and 3.6)!
respectively. Because it is refilled a number of times, the average cost per bottle is relatively low.
» For other advantages of the tin can, cf. 'Bnlletin 4, of International Tin Research and Development
Council, Tin Plate and Tin Cans in the United States (1936), pp. 8-9.
M Western Canner and Packer, Yearbook and Statistical Number (1939), p. 174.
" Some change in public taste may be effected by advertising campaigns, as was true in the case of the
beer can. See infra, pp. 15-16.
32 Western Canner and Packer, op. cit., p. 190.
CONCENTRATION OF ECONOMIC POWER
14025
scrapping old bottling equipment and investing in new canning equipment.
This change will be made only if large savings are realizable, or if consumers'
preferences require the change.
B. DEMAND FOR GENERAL LINE CANS
Many consumers' products, such as oil, paints and varnishes, tobacco, chemi-
cals, pharmaceutical and toilet articles, spices, beer, and coffee, are packaged in
general line cans. As previously stated, this group of cans accounts for approxi-
mately 45 percent of the total quantity of tin plate taken by the tin can industry.
In analyzing the demand for general line cans, special attention will be devoted
to beer, oil, and paint and varnish cans. These are the most important of the
general line cans and illustrate the types of considerations covering the demand
for such containers.
(1) Beer Cans.- — Of the general line cans, beer cans are today among the most
important. In 1937 more than 630 miUion beer cans were produced with a value
of 14 million dollars or approximately 4 percent of the total value of all the tin
cans and tin utensils produced in that year.^^ Since the sale of beer was re-
legalized so recently, and beer cans have been in use for only the past four years,
any analysis of the beer industry's consumption of tin cans necessarily must be
tentative.
As shown in Table 4, beer consumption has had an upward trend since re-
legalizatioh in 1933; it also appears to respond to changes in consumers' income.
Table 4. — Production of Fermented Malt Liquors, and Consumers' Income, 1934-
1938
Year
Production of Fermented
Malt Liquors '
Consumers' Income '
Number of
barrels
Index
(1934=100)
Millions of
dollars
Index
(1934=100)
1934
37, 678, 313
45. 228, 605
51,812,062
58, 748. 087
56,340,163
100
120
138
156
150
52, 057
55, 814
64, 207
70, 694
65, 021
100
1935
107
1936 "
123
1937
136
1938
125
Sources:
1 United States Treasury Department, Bureau of Internal Revenue Alcohol Tax Unit, Statistics on Fer-
mented Mall Liquors and Cereal Beverages (Fiscal year data).
2 Survey of Current Business, June, 1939, p. 12.
With the increasing consumption of beer, there has been a shift to beer pack-
aged in bottles and cans. Packaged beer rose from 20 percent of the total in
1933 to 45 percent in 1938. Of the total beer packaged, the proportion in cans
has varied only from 13 percent in 1935, the year of its introduction, to 17 percent
in 1937. Table 5 shows the percentage of all beer in packaged form and the
percentage of packaged beer in cans.
Table 5.— Percentage of Beer Packaged in Bottles and Cans, 1933-1938
Year
Percentage
of Beer
Packaged
Beer in Cans
as Percentage
of Total Pack-
aged Beer
1933 -
20%
25
30
30
44
45
(■)
'' 13%
1934
1935 _
1936
1937
1938 (4 mos.)
Source; Based on data in Metal Containers, August,
1, 1938, and Steel, January 6, 1936.
' Canned beer was not introduced until 1935.
8, Canco, October, 1935, Sales M^ nagement, July
Census of Manufacturers, 1937, p. 2.
14026 CONCENTRATION OF ECONOMIC POWER
The percentage of beer put in cans or bottles depends upon several factors.
First, there is the comparative cost of container and transportation for bottled
and canned beer. Since the first cost of a beer can to the brewer is about the same
as the cost of the better grade of bottle," re-use of the bottle '^ makes it a cheaper
packaging medium. On tlie other hand, transportation costs are higher for
bottled than for canned beer. Thus, wide geographical distribution of a parti-
cular brand of beer involving considerable transportation costs, danger of bottle
breakage and recovery difficulties in return of bottles for re-use tend to make
cans less costly and more desirable than bottles. Consequently, in the export
trade ^^ and where the beer is shipped over long distances in the domestic market,
the tin can has 'made Ifl-rge inroads in the bottled beer field.
Secondly, public taste, convenience and custom are important considerations
in the use of the tin can. Tin cans save space, cool more quickly than bottles,
and do not require a deposit or bookkeeping.*' On the other hand, bottled beer
was estabhshed before Prohibition, whereas canned beer was introduced in 1935.
In restaurants, taverns, and hotels the public has become accustomed to bottled
beer, and where the return of the empty bottle is not a problem, these intangible
factors have a marked influence pu the type of container which will be used. Since
the introduction of tin beer cans, the can companies have advertised widely *^
the benefits of beer in cans, and pubUc taste and custom have been partly altered
in favor of cans.
The cost of tin plate in the tin beer can constitutes on the average less than
15 percent of the retail price of a can of beer.*' Consequently a 10 percent reduc-
tion in the price of tin plate, if passed on entirely to the consumer, would reduce
the price of a can of beer by only about fifteen hundredths of one cent. Since
the cost of the tin plate is such a small part of the cost of a can of beer, and the
use of cans for beer depends largely on geographical distribution and consumer
acceptance, a change in the price of tin plate can have only a slight effect on the
use of beer cans.
(2) Lubricant Oil Cans. — The lubricant oil industry is an important consumer
of tin cans, using them as containers for its products in both domestic and export *"
markets. Several oil companies now market their lubricants exclusively in cans."
The demand for tin cans as containers for motor oil depends, of course, on the
demand for such oil. Lubricant oil consumption has shown an upward trend
since 1921, which has been mainly due (1) to the increase in the number of cars,
total motor vehicle registration having almost tripled from 1921 to 1937, and
(2) to the 70 percent increase in the average mileage per car in this same period.
The following table shows the increase in the lubricant oil consumption and
motor vehicle registrations and the average mileage run during the period from
1921 to 1937:
Table 6. — Lubricant Consumption, Motor Vehicle Registration, and Average Car
Mileage, 1921-19S7
Year
Consumption
of Motor Oil '
(thousands of
barrels)
Total Motor
Vehicle Regis-
tration 2 (thou-
sands of ve-
hicles)
Mileage
Run » (aver-
age per car)
1921 .... . .
10.463
19, 937
26,501
23,844
29,705
6,080
9,221
11, 526
9,756
12, 653
6,896
1929 . ..
8,752
1933 . .. . .
8,864
1937 .
« 10, 304
' National PelroUum News, January 11, 1939, p. 24.
' Automobile Facta and Fii/uren, 1938, p. 16.
« O. Terborgh, Passenger Automobiles Memorandum, April 12, 1937, p. 2.
♦ 1936 figure u.sed here.
" See Appendix, Tables 14 and 15.
" In 1937 only 17% of the beer sold as packaged beer was sold in can.-., the balance of 83% being sold in
bottles. Since, during such year, only 673 million beer bottles were produced as contrasted with 030 million
beer cans, it would appear that each beer bottle was probably used on an average of Ave times (see Table 5
and Appendix, Ta()le 15).
" According to one estimate, eighty percent of the packaged beer in the e.xport trade is in cans. (Steel,
January 3. 1938, p. 211.)
" International Tin Research and Development Council, Information Circular No. 1, Canned Beer
(January, 1936).
" Sales Management, July 1, 1938.
" See Appendix, Tables 8 and 14.
«» R. Skoinp, The Evolution of a 7'in Can, p. 16.
*' International Tin Research and Development Council, Tin Plate and Tin Cans in the United States,
Bulletin No. 4 (1930). p. 115.
CONCENTRATION OF ECONOMIC POWER 14027
This great increase in the consumption of lubricant oil has brought an increase
in the use of oil cans." As in the beer industry, the consumers in the past few
years have learned the advantages of oil in sealed cans and have accordingly
increased their consumption of oil in such containers.^' Customers prefer such
tin cans due to the assurance that the oil is free from contamination or substitu-
tion, and retailers favor canned oil because it is convenient and suitable for
attractive displays and it permits greater diversification in grade of oil." In
part, however, the tendency to favor oil in cans is offset by declines in consumers'
income, for a drop in income leads to the purchase of the cheaper grades of oil,
which are sold in bulk.'*5
(3) Paint and Varnish Cans. — Consumption of general line cans by the paint
and varnish industry has shown^ an upwarS trend subject, however, to wide
• fluctuations. Production of paints and varnishes rose steadily from 1919 to a
peak in 1929, declined 45 percent to a low in 1932, and by 1935 recovered to
87 percent of the 1929 level.'" The report of the National Resources Committee
in Patterns of Resource Use found that the consumption of paints and varnishes
was closely related to consumers' income.*^
Since the cost of tin plate is only a fraction of the cost of the finished product
in a thi container and since substitution of tin cans for other types of containers
is a minor factor, tin plate prices are relatively unimportant in their effect on
the consumption of paints and varnishes.
(4) Other General Line Cans. — Other industries which use general line con-
tainers also reflect this same general pattern of upward trend and dependence
on consumers' income. Automobile pohshes, waxes, and anti-freeze solutions
sold in cans reflect the growth in the number and use of automobiles. The
tobacco and chemical industries,*^ which use tin containers, both show a high
degree of correlation with consumers' income.
The demand for general line cans necessarily follows closely the production .of
the products packed in the cans, which, in turn, is generally dependent prima,rily
upon consumers' income. The cost of tin plate, which constitutes a small portion,
usually less than 10 percent, of the retail price, is therefore only a minor influence
in determining consumption of tin plate by these industries. Possibilities of
substituting general line tin cans for other typgs of containers by a price reduction
tend to be limited due to other factors such as fixity of public taste and diflSculties
of technical adaptation.
C. QUANTITY OF TIN PLATE PER CAN
The quantity of tin plate used in any specified tin can depends almost entirely
on the product to be packed. The specifications for the composition of the tin
plate, the manner of processing and the weight and amount of tin plate in the
tin cans depend upon (1) variable corrosion, deformation and the other technical
matters encountered in the packing of the different food stuffs and other materials
and (2) needs of the canners, retailers, and consumers.*' These technical factors
control the quantity of tin plate used in a can designed for a particular use.
VI. Summary
The price of tin plate is of minor significance in determining its consumption.
Since the demand for products packed in tin cans is generally not very elastic,
and especially since the cost of tin plate is such a small proportion of the cost of
the final product, a reductio i in the price of tin plate> would be ineffectual in in-
creasing the consumption of steel. More fundamental forces such as consumers'
income, public tastes, and technical considerations determine the demand for
canned commodities and, hence, the consumption of tin plate by the canning
industry.
" Steel, January 2, 1939, p. 149, and International Tin Research and Development Council, op. cit., p. 115.
" Sales Management, July 1, 1938, p. 50.
" International Tin Research and Development Council, op. cit., pp. 115-6.
"Si«e;, April 11, 1938, p. 23.
*o National Resources CMnmittee, Patterns of Resource Use, PreUminary Edition for Technical Criticism,
March, 1939. p. 122. See also Appendix, Table 16.
" The amount of variation in the consumption of paints and varnishes explained by consumers' income
!S95 percent. Ibid.
*8 ThejNational Resources Committee found that tobbacco consumption was related to consumer income
and a. time -trend and that the amount of variation explained was 96 percent; also that the consumption
of chemicals was related to consumers' income and the Federal Reserve Board's index of industrial pro-
duction, and that the amount of variation explained was 93 percent. Cf. National Resour^is Committee,
op. cit., pp. 104 and 121. ' Also see Appendix, Tables 17 and 18.
" International Tin Research and Development Council, op. cit., especially pp. 67 and 88, and also th'
'--' ' General Report, -1 937, by the Council.
124491— 41— pt. 26 29
14028 CONCENTRATION OF ECONOMIC POWER
Appendix
Table 7.— Production of Tin Cans and Utensils, 19S7
Type of Can
Number
(millions
of cans)
Value
(millions
of dollars)
Percent of
Total
Value
Packers Cans:
1,800.1
9, 592. 2
' 4, 190. 1
630.9
1.0
1.6
$16.1
188.9
130.4
14.1
0.5
5.0
4.8%
Sanitary cans (including sweetened condeased milk cans)
General Line Cans:
50.4
Special general line cans:
4.2
0.2
1.6
Total tin cans, utensils, etc
1 16,215.9
» 335. 0
100.0
1 The number of cans listed here is only the number reported whereas the value is for all cans, including
8 million dollars worth of cans for which numbers were not reported.
• Includes tin cans, utensils, etc., made as secondary products in other industries but does not include
$18,537,673 of finished tinware, other than cans.
Sources: Censiis of Manufactures, 1937, "Tin Cans and Other Tinware."
Table 8
SECTION A— STEEL COSTS IN THE TIN CAN INDUSTRY, 1935
Value of Products of the Tin Can and Other Tinware Industry K. $287, 582, 216
Cost of Steel Materials Used in the Tin Can Industry 2 170, 665, 145
Steel Costs as Proportion of Total Value of Products of the Tin
Can Industry 3 59. 3%
Sources: ' Census of Manufactures, 1935, "Tin Cans and Other Tinware," p. 958. The coverage here
is the same as for the steel materials and represents 98.4 percent of the aggregate value of products for the
industry.
2 See Section B below.
3 Computed by dividing cost of steel materials by value of product.
SECTION B— STEEL MATERIALS CONSUMED BY THE TIN CAN INDUSTRY, 1935
Quantity
Reported
(net tons)'
Tin plate--.
Teme plate -
Black plate.
$159,938,020
5,911,536
4,815,589
,404,948
61, 193
63,745
All products-
Source: Census of Manufactures, 1935, "Tin Cans and Other Tinware," p. 958.
1 The reported tons cover 95 percent of the cost figures given, for in some cases only the value was reported .
Table 9. — Tin plate and terne plate production, 1919-1937
Year
Total Produc-
tion of Tin
Plate and
Terne Plate
(thousands of
pounds)
Total Value of
Tin Plate and
Terne Plate
(thousands of
dollars)
Year
Total Produc-
tion of Tin
Plate and
Terne Plate
(thousands of
pounds)
Total Value of
Tin Plate and
erne Plate
(thousands of
dollars)
1919
2, 539, 224
1, 725, 781
3, 296, 551
3,661,837
3, 753, 474
$175, 776
96,181
162, 476
188, 610
190, 426
1929
4, 377, 488
3, 268, 025
3, 962, 956
4, 216, 002
6, 022, 908
209.675
1921
1931
140, 984
148, 770
1925 ....
192, 323
1927
1937
278, 380
Source: Census of Manufactures, "Tin Plate and Terne Plate."
CONCENTRATION OF ECONOMIC POWER 14029
Table 10. — Prices of tin plate and consumption of steel by the container industry,
1923-1938
Tin Plate Prices '
Consumption
oT Hot-Rolled
Iron and Steel
(thousands of
gross tons) 2
Year
Tin Plate Prices i
Consumption
of Hot-Rolled
Year
(Dollars per
base box)
Inde.x
(1929=100)
(Dollars per
base box)
Index •
(1929=100)
Iron and Steel
(thousands of
gross tons)'
1923
1924
1925
1926...
1927
1928
1929
1930 _
$5,42
5.50
5.50
5.50
5.48
5.25
5.35
5.19
101
103
103
103
102
98
100
97
1,205
1,210
1,427
1,348
1,408
1,619
1,707
1,670
1931
1932
1933- --
1934
1935-.
1936
1937
1938
4.94
4.69
4.43
5.25
5.25
5.25
35.22
35.31
92
88
98
98
»105
'103
1,415
1,037
1,769
1,557
2,039
2,455
2,784
1,908
Source:
1 Iron Age, January 5, 1939, p. 198. These figures are average prices for the year at Pittsburgh.
' Computed by apportioning individual hot-rolled product totals on the basis of Iron Age distribution
reports and by allocating jobber shipments to ultimate consumers. See M. W. Worthing. The Distribution
of Steel Products to Major Consuming Industries, United States Steel Corporation, October 30, 1939.
3 For a number of years prior to 1937, published prices were subject to a trade discount of 7^i percent.
Since January 1, 1937, the published prices have been net. From January 1, 1938, to November 10, 1938,
there was a refund of 250 per base box. A nominal list price comparable to those in the years preceding
1937, which takes these two factors into account, would have been $5.64 in 1937 and $5.51 in 1938. The
index figures shown in the table for 1937 and 1938 are based upon these nominal list prices.
< Strip was included in container industry tonnage in 1938.
Table U.
-Consumers' income, food production, and canned food consumption,
1919-1936
Consumers'
Income
(biUions of
1936 doiiars) '
Index of
Food Pro-
duction
(1923-
1925 = 100 »
Consumption of Canned
Foods
Year
Amount
(millions of
pounds) 3
Index
(1923-
1925=100)
1919
$46.6
47.0
42.3
48.8
54.8
55.4
57.0
58.3
58.9
62.2
65.3
60.4
54.6
46.7
48.2
54.0
56.9
63.9
94
84
83
94
99
103
98
97
96
98
97
93
90
87
92
99
79
90
(*)
(*)
1891
2620
2890
2931
3847
3493
3590
3822
4096
4218
3404
3282
3568
3973
4753
(')
(<;
59
1920
1921
1922
81
1923
90
1924
1925
1926
1927
1928
1929
127
1930
131
1931
106
1932
102
1933
1934
123
1935
1936 .
0)
Sources:
' National Resources Committee, Patterns of Resource Use, March, 1939, p.
' Federal Reserve Board, Division of Research and Statistics, June, 1937.
5 National Resources Committee. Patterns of Resource Use. March. 1939. p.
* Data not given.
14030 CONCENTRATION OF ECONOMIC POWER
Table 12.— Food Packs and Tin Plate Price, 1920-19S8
Food Packs < (thousands of cases)
Price of Tin
Plate > (dol-
lars per base
box)
#3 Tomatoes
#2 Corn
#2 Peas
1920 -
U, 368
4,017
11,538
14,672
12, 519
19, 770
9,455
13, 137
8,639
14, 146
16,998
9,573
16,028
11,986
22, 376
26,985
24,209
26, 076
22,960
15,040
8,843
11,419
14, 106
12,131
24.320
19,069
10,347
14, 497
17, 487
15,692
19,415
9,358
10,193
2l',471
14,621
23,541
20,470
12, 317
8,207
13,042
13,948
19.315
17,816
17, 709
it! 943
18,530
22,035
13,286
10, 367
12, 893
15, 742
24,699
16, 553
23.467
25,459
$7.53
1921
6.90
4.73
5.42
5.50
5.50
5.50
6.48
5.25
5.35
J930 . .
6.19
(931 .. ,
4.94
(932
4.69
1933
. 4.43
1934
5.25
1935
1936 -
5.25
5.25
1937 . ..
'5,22
1938
«5.31
I Data on packs from The Canning Trade, Almanac of the Cannino Industry, 1939, p. 208.
» Irmi Age, January 5, 1939, p. 200. These figures are averages for the year and represent tin plate prices
« A list price comparable to those in the years preceding 1937 would have been $5.64 in 1937 and $6.51
in 1938. See note 3 to Table 10.
Table 13.
-Prices of Tin Plate and Average Retail Prices of Canned Foods in 51
Cities Combined, 1 923-1938
Canned Foods ' (cents per can)
Tin Plate >
(dollars
Yeu
Salmon
Red(16oz.
can)
Peas
(No. 2 can)
Cora
(No. 2 can)
Tomatoes
(No. 2 can)
Milk
Evaporated
(14!^ oz.
can)
1923
30. US
30.1
32.2
36.6
32.7
33.7
- 31.1
32.1
32.2
24.3
19.3
21.1
22.1
25.3
25.6
25.7
17.3(f
17.9
17.9
17.1
16.5
16.0
16.5
16.0
14.0
12.7
12.8
16.6
17.1
16.1
16.2
15.1
15.4»f
16.0
17.6
16.3
15.7
15.9
15.8
15.3
13.3
10.6
10.1
11.5
12.5
12.0
12.9
11.6
12. 5<
12.8
13.2
11.8
11.9
11.7
12.8
12.1
10.1
9.3
9.1
10.5
10.1
9.4
9.3
8.8
10.9(»
10.2
10.2
10.3
10.-3
10.0
9.8
9.1
8.2
6.8
6.5
6.7
7.0
7.6
7.6
7.1
$5.4»
1924
5.50
1925
5.50
1926
5.60
1927
6.48
1928— -
6.25
1929 . .
6.35
1930
5.19
1931
4.94
4.69
1933
4.43
5.25
5.25
5.25
•5.22
»5.31
Sources:
1 Retail Prices of F^od, 1923-36, United States Dept. of Labor, Bureau of Labor Statistics, pp. 80-90. Also
Retail Prices, Jan.. 19,33, p. 12. and Jan. 1939, pp. 12, 13.
" Tin plate prices are average prices for the year at Pittsburgh. Cf. Iron Age, January 5. 1939, p. 200.
> A list price comparable to those in the years preceding 1937 would have been $5.04 in 1937 and $5.51 in
1938. See note 3 to Table 10.
CONCENTRATION OF ECONOMIC POWER 14031
Table 14. — Cost of Containers and Retail Price of Beer, New York City
Type of Container
Cost of Con-
tainer (cents
per can)'
Retail Price of Beer
(cents per can)
Locally pro-
duced
Produced
elsewhere
Cans
2. 136fi
>2.100
10.0)!
10. 0 to 12. 5^
»12.S
Bottles
Source:Informationsuppliedbytrade, October, 1939.
Prices quoted in quantities of 1,000 units.
Represents better class of bottles, re-used a number of times. For a cheaper grade of re-used bottles the
price quoted was $19.90 per 1000 and for "one-trip" bottles it was $17.55 per
can and bottle in Table 15.
' Prices quoted for case of 24 cans or bottles. Bottle deposits excluded.
Compare average value of
Table 15. — Beer Bottles and Beer Cans, 1937
Number
Produced
Value of
Product
Average
Value per
Container
Beer bottles
673,053,408
630,896,567
$13, 382, 049
14,108,829
i.m
%2\t
Beer cans.
Source: Census of Manufactures, 1937, "Glass" and "Tin Cans and Other Tinware".
Table 16. — Production of Paints and Varnishes, and Consumers' Income, 19 19-1985
1920
1921
1922
1923
1924
1925
1926
1927
Index of
Varnish
and Paint
Production
(1929=100)
49.9
54. G
44.6
67.5
71.0
78.7
79.6
87.6
Consumers'
Income
(billions of
1935 dollars)
46.6
47.0
42.3
48.8
54.8
55.4
57.0
58.3
58.9
1930.
1931.
1932.
1933.
1934.
1935.
Index of
Varnish
and Paint
Production
(1929=100)
96.2
100.0
81.6
70.6
5.5.6
70!7
Consumers'
Income
(billions of
1936 dollars)
62.2
65.3
60.4
54.6
46.7
48.2
M.O
66.9
Source: National Resources Committee, Patterns of Resource Use, Preliminary Edition for Technical
Criticism, 1939, p. 122.
Table 17. — Tobacco Consumption and Consumers' Income, 1919-1936
Year
Tobacco
Consump-
tion (1923-
1925=100)
Consumers'
Income
(billions of
1936 dollars)
Year
Tobacco
Consump-
tion (1923-
1925 = 100)
Consumers'
Income
(billions of
1936 dollars)
1919
82
87
85
96
105
112
lis
$46.6
47.0
42.3
48.8
54.8
.55.4
57.0
58.3
58.9
124
134
131
123
111
116
128
135
151
62.2
65.3
60.4
54.6
46 7
1920
1921
1930
1922
1931
1923
1932
1924
48 2
1925
54.0
56.9
1926..
1927
Source: National Resources Committee Patterns of Resource Use, Preliminary Edition for Technical
Criticism, 1938, p. 104.
14032 CONCP]NTRATION OF ECONOMIC POWER
Table 18. — Consumption of Chemicals, Industrial Production, and Consumers'
Income, 1 923-1935
Year
Consump-
tion of
Chemicals
(1929 =
100)1
Average
Industrial
Production
and Con-
sumers'
Income
(1923-5
= 100) »
Consumers'
Income
(billions of
1936
dollars)
Year
Consump-
tion of
Chemicals
(1929 =
100)'
Average
Industrial
Production
and Con-
sumers'
Income
(1923-5
= 100) '
Consumers'
Income
(biUions of
1936
dollars)
1923
1924 -
1925
1926..
1927
1928
1929
64.1
58.1
66.6
76.4
75.2
92.2
100.0
100
97
103
106
106
112
118
54.8
55.4
57.0
58.3
68.9
62.2
65.3
1930...
1931
1932
1933
1934
1935 -.
89.7
78.2
61.3
77.8
87.5
90.7
102
90
74
81
88
96
60.4
54.6
46.7
48.2
54.0
56.9
Sources: National Resources Committee, Patterns of Resource Use, Preliminary Edition for Technical
Criticism, 1939, p. 121.
' The production series given in summary was adjusted for imports and exports as given in the Statistical
Abstract of the United StaUs.
2 Consmer income is reduced to a 1923-25 base; this is combined with the Federal Reserve Board index
of industrial production, each series being weighted equally.
Exhibit No. 1416
AN ANALYSIS OF STEEL PRICES, VOLUME AND COSTS-
CONTROLLING LIMITATIONS ON PRICE REDUCTIONS
Thia analysis was prepared by the United States Steel Corporation in connec-
tion with its studies in preparation for the hearings on the steel industry before
the Temporary National Economic Committee. The work was under the super-
vision of Theodore O. Yntema, Professor of Statistics, University of Chicago.
October 30, 1939.
TABLE OF COIfTENTS
I. Purpose of the Study.
II. Summary of Findings.
III. U. S. Steel Corporation Costs in Relation to Volume.
IV. Composition of Costs.
V. Relation of Costs and Volume to Prices.
VI. Weighted Tonnages and the Operating Rate.
Appendix I— Computation of Average Worliing Capital— 1938.
Appendix II— Computation of Percentage Increase in Volume to be Expected from Percentage Decrease in
Price, assuming Elasticity of Demand for Steel of 1.
Appendix III— Actual Operating Deficit in 1938 and Estimated Addition to Deficit if Prices Had Been
Further Reduced.
Appendix IV — Weighted Tonnage which Must Be Shipped to Break Even at Various Price I/Cvels.
Appendix V— Computation of Increases in Volume Needed to Offset Decreases in Price.
Appendix VII— Operating Rate Required to Earn 5% on Tangible Investment if Break-even Point were
at over 90% of Capacity.
Appendix VII— Effect of Inter-company Transactions on Profit and Loss Statement.
Tables
Table 1— Elements of Total Costs— 1938 Conditions^
Table 2— Estimated Additions to 1938 Deficit— How Deficit Would Have Increased if Prices Had Been
Reduced and Volume Had Increased to Same Relative Extent.
Table 3— Increases in Volume Needed to Compensate for Decreases in Average 1938 Prices, Compared to
Maximum Probable Resulting Increases in Volume.
Table 4— Increases in Volume Needed to Compensate for Decreases in Average 1938 Pri(tes, Compared to
Maximum Probable Resulting Increases in Volume.
Table 5— Compilation of Total Costs Exclusive of Inter-company Items and Casts Connected with Extra
neous Transactions.
Table 6— Estimate of Inter-company Transactions.
Table 7— Total Costs and Volume of Busine.ss- Unadjusted.
Table 8— Analysis of Operating Costs into Components.
Table 9 — Taxes and Volume of Business.
Table 10— Estimated Ta\es (Other than Social Security and Federal Income) in Prior Years if 1932-1938
Rates Had Prevailed.
Table 11— Taxes and Volume of Business, under Present Tax Laws.
Table 12— Adjustment of Payroll to 1938 Wage Rates.
Table 13— Estimate of Social Security Taxes at 1938 Rates.
Table 14— "Other Expenses" Adjusted for Changes in Price Level.
Table 15- Costs Adjusted to 1938 Interest, Tax, Pension, and Wage Rates and 1938 Price Levels, but Unad-
justed for Secular Trend.
CONCENTRATION OF ECONOMIC POWER 14033
Table 16 — Deviation of Adjusted Total Costs from Average Costs for Volume Involved.
Table 17— Adjustment of Adjusted Costs for Time Trend.
Table 18— Total Costs of Operations and Volume of Business— 1938.
Table 19 — Deviation of Payrolls, Adjusted to 1938 Wage Rates, from Average Payroll for Volume Involved.
Table 20— Adjustment of Payrolls, Adjusted to 1938 Wage Rates, for Time Trend.
Table 21— Payrolls, Adjusted to 1938 Wage Rates and EflSciency, and Volume of Business— 1938.
Table 22 — Deviation of Adjusted "Other Expenses" from Average for Volume Involved.
Table 23 — Adjustment of Adjusted "Other Expenses" for Time Trend.
Table 24 — "Other Expenses," Adjusted to 1938 Price Levels and Operating Conditions, Related to Volume
of Business— 1938.
Table 25— Elements of Total Costs Adjusted to 1938 Conditions.
Table 26— Elements of Total Costs— Revised— 1938 Conditions.
Table 27— Cash and Non-Cash Costs— 1938 Conditions.
Table 28— Percentage of Fixed to Total Costs at Various Rates of Operation— 1938 Conditions.
Table 29— Sales and Other Operating Revenues— 1938.
Table 30 — Transportation and Miscellaneous Revenues Related to Volume of Business.
Table 31— Transportation and Miscellaneous Revenues per Weighted Ton of Products Shipped.
Table 32— Estimated Additions to 1938 Deficit— How Deficit Would Have Increased if Prices Had Been
Reduced Various Percentages.
Table 33— Increases in Volume Needed to Compensate for Various Reductions in Average 1938 Prices,
Compared to Maximum Probable Resulting Increases in Volume.
Table 34 — Estimate of Annual Sales and Revenue at Prices Prevailing in Second Half of 1938.
Table 35 — Increase in Volume Needed to Compensate for Various Reductions in the Prices Prevailing in
the Second Half of 1938, Compared to Maximum Probable Resulting Increases in Volume.
Table 36— Weighted Tonnages of All Tonnage Products and Unewighted Tonnages of Rolled and Finished
Products Shipped.
Table 37— Average Weighted Equivalent of Unweighted Tonnages.
Table 38— Ratio of Tons of Rolled and Finished Products to Weighted Tons of All Tonnage Products.
Table 39— Percentage of Tons of Various Classes of Products Shipped to Total Tons of Rolled and Finished
Products Shipped— 1927-1938.
Table 40— Ratio of Unweighted to Weighted Tons of Rolled and Finished Products.
Charts
Chart 1 — Total Costs (Unadjusted) and Volume of Business.
Chart 2— Taxes and Volume of Business.
Chart 3— Depletion and Depreciation Related to Volume of Business.
Chart 4 — Payrolls Adjusted to 1938 Wage Rates and Related to Volume of Business.
Chart 5— Total Costs Adjusted to 1938 Interest, Tax, Pension and Wage Rates and 1938 Price Levels, but
Unadjusted for Changed Operating Conditions. Related to Volume of Business.
Chart 6— Percentage Deviations from Average— Total Costs Adjusted to 1938 Interest, Tax, Pension and
Wage Rates and 1938 Price Level but Unadjusted for Changed Operating Conditions.
Chart 7 — Relationship between Total Costs and Volume of Business— 1938 Conditions.
Chart 8— Percentage Deviations from Average — Payrolls Adjusted to 1938 Wage Rates.
Chart 9— Payrolls, Adjusted to 1938 Conditions, Related to Volume of Business.
Chart 10— "Other Expenses" Adjusted to 1938 Price Level and Related to Volume of Business.
Chart 11 — Percentage Deviations from Average — "Other Expenses", Adjusted for Price Changes.
Chart 12 — "Other Expenses" Adjusted for Changes in Commodity Prices and for Changed Operating
Conditions, Related to Volume of Business.
Chart 13— Composition of Total Costs in Relation to Volume of Business.
Chart 14 — Transportation and Miscellaneous Revenues Related to Shipments.
Chart 15— Estimated Additions to 1938 Deficit— How Deficit Would Have Increased if Prices Had Been
Reduced and Volume Had Increased to Same Relative Extent.
Chart 16— Relation Between Sales and Costs— Effect of Reduction from Average 1938 Prices.
Chart 17— Increases in Volume Needed to Compensate for Various Decreases in Average 1938 Prices-
Compared to Probable Resulting Increases.
Chart 18— Relation Between Sales and Costs Effect of Reduction from Second HaK of 1938 Prices.
Chart 19 — Increases in Volume Needed to Compensate for Various Decreases in Price Prevailing in the
Second Half of 1938 — Compared to Probable Resulting Increases.
Chart 20— Relation of Actual Tons of Rolled and Finished Steel Products Shipped and Weighted Tons of
All Tonnage Products Shipped.
Chart 21— Ratio of Tons of Rolled and Finished Products to Weighted Tons of All Tonnage Products at
Different Rolled and Finished Shipment Levels.
Chart 22 — Per Cent of Weighted Tons of Rolled and Finished Products Shipped Represented by Un-
weighted Tonnages of Rolled and Finished Products Shipped.
I. Purpose of the Study
The success of mass produi^tion methods in American industrial practice has
given much emphasis, to the importance of volume in reducing costs. Failure to
appreciate, in the case of the steel industry, the limitations of the extent to which
increased volume means reduced costs per unit of product has frequently been
made the basis of the charge that steel prices are higher than they should be and
that the absence of price competition, rather than the limitation imposed by costs,
is the reason why steel prices are not lower than they are. It is further said
that if steel prices were lowered in times of recession more steel would be sold
and payrolls and employment would be better maintained. The fact of the matter
is that steel prices are not high. In recent years they have barely covered costs.
This is evidenced by the fact that in the last ten years, 1929 to 1938, the United
States Steel Corporation realized a return of only 1.91% on the combined invest-
ment of the bondholders and stockholders. From 1930 to 1938 the rate of return
averaged less than 1% per year.
However, the contention that steel prices are too high is not made in ignorance
of the small margin of profit currently being realized on sales. The theorj- of
14034
CONCENTRATION OF ECONOMIC POWER
those who would lower steel prices is that such reductions greatly stimulate sales
and that volume and costs are so inter-related in the production of steel that, on
the mass production principle, the cost per ton would be drastically reduced if
volume were increased. Hence, it is argued that if prices were reduced, volume
would be stimulated to a point where costs would drop sufficiently to make steel
production profitable at the reduced prices. It has been the purpose of this
study to ascertain the increases in volume that would have to take place to offset
various decreases in steel prices by the United States Steel Corporation sub-
sidiaries, taking into consideration the effect of increased volume on costs, and to
estimate the financial gain or loss which would result from price reductions.
II. Summary of Findings
1. The total annual costs of the United States Steel Corporation and its sub-
sidiaries may be divided into (1) those which must be met regardless of the ton-
nage of steel produced and (2) the additional costs of producing each additional
ton of product. Taking the coats shown by the profit and loss statements of the
Corporation from 1927 to 1938 and adjusting to 1938 wage, interest, and tax rates,
and to 1938 prices and other operating conditions, this study shows that under
1938 conditions the costs of the first or fixed type amount to $182,100,000 per year
while those of the second type, the additional costs, are approximately $55.73 per
weighted ton of product shipped.' Thus the total costs for any volume of busi-
ness at 1938 wage, interest, and tax rates, and 1938 material prices, may be
estimated by multiplying the weighted tons shipped by $55.73 and adding
$182,100,000 to the result. It should be noted, however, that while these costs
are exclusive of all non-operating income and expense, they cover all operations
of the Corporation's subsidiaries and hence do not represent merely the cost of
producing steel. Furthermore, even weighted lonnages shipped do not reflect the
full volume of business, since some goods and services are sold by the Corpora-
tion's subsidiaries which are not measured in tons. Nevertheless, the other
operations rise and fall with increases and decreases in shipments of products to a
sufficient extent that the total costs maintain approximately the relationship to
shipments just described.
2. Within the range of actual experience the additional costs, at 1938 wage and
tax rates and 1938 material prices, arising with the shipment of each additional
ton remain constant at $55.73. This is true when production averages as high as
90.4% and as low as 17.7% of ingot capacity for the entire year. While the addi-
tional costs per ton tend to remain constant, the average costs per ton decrease
as the volume of shipments rises since the fixed costs of $182,100,000 are spread
over a greater number of tons.
3. An approximation of the elements composing the total costs of the United
States Steel Corporation and its subsidiaries is shown below:
Table 1.
-Elements of Total Costs 1938 Conditions — United States Steel Corpora-
tion and Subsidiaries
Item
Costs that must
be met regard-
less of operat-
ing rate
Additional
cost for each
additional
ton ' of
product
shipped
Interest -
$8, 300, 000
7, 700, 000
24, 200, 000
62, 100,.000
2,500,000
47,800,000
$0.00
0.00
Taxes other than Social Security and Federal Income
1.43
PayroU
29.10
1.16
nt.hPT r.axh jr.rptmfPtt
21.67
Total Cash Costs _
Depreciation and Depletion
152, 600, 000
29, 500, 000
53.36
2.37
Total costs .
$182,100,000
$55.73
' Tonnage weighted to adjust for variation from normal proprotions;of high and low cost products and for
shipments of products other than steel.
■ By weighted tons are meant tonnages of each type of product shipped (except cement and certain liquid
and gaieous coke-oven-by-products), converted, on an average mill cost basis, to equivalent tons of average
cost rolled and finished steel products. In this way adjustment is made for (1) variaticns in the types o'
steel products constituting the total shipments and (2) shipments of products other then steel.
CONCENTRATION OF ECONOMIC POWER
14035
4. The elasticity of demand for a product is measured by the ratio of the
relative resulting increase in volume to the relative decrease in price. Assuming
that, as all analyses indicate, steel has an elasticity of demand no greater than 1,
and that competitive meeting of prices would prevent increased participation in
the going volume of business, the relative increase in shipments attributable to a
price reduction by the Corporation would not be greater than the relative decrease
in prices, and the total sales in dollars would not be increased. Under these
circumstances, the smallest general drop in steel prices that could possibly have
raised the 1938 volume of steel sold to 1937 levels would have been 40.9%. A
price reduction of this proportion, even if successful in restoring 1937 volume,
would have resulted in a cash loss for the year of $219,920,000 and a total loss to
the Corporation, including depreciation and depletion, of $280,704,000. Such a
cash loss, if continued, would have exhausted the average 1938 working capital
in less than two years and annual total losses on this scale would wipe out the
entire equity of preferred and common stockholders existing on December 31, 1938
in about four and a half years.
5. Any reduction in steel prices below the average prices prevailing in 1938
would have served only to increase the 1938 deficit of the Corporation which
amounted to $8,758,572, exclusive of non-operating income and expense. Even
if the elasticity of demand for steel is assumed to be as high as 1, the estimated
additions to the 1938 deficit which would have resulted from various price re-
ductions would have been as follows:
Table 2. — Estimated Additions to 19S8 Deficit — How Deficit Would Have Increased
if Prices Had Been Reduced and Volume Had Increased to Same Relative Extent —
United States Steel Corporation and Subsidiaries
Reduction in average price
Estimated addi-
tion to deficit
Reduction in average price
Estimated ad-
dition to deficit
$3, 900. 000
20, 500, 000
43,300,000
15%
$68,700,000
97,400,000
cm
20% ::...:::::..:.:.
IW
6. Because of the low elasticity of demand for steel, the increase in volume
resulting from a reduction in price is certain to be less than the increase needed
to ofifeet the adverse effects of the lower price on profits. Regardless of the volume
of operations taken as a starting. point, the percentage increases in volume neces-
sary to compensate for various percentage reductions in the average 1938 prices,
so that profits or losses would neither be increased or decreased by the price
reduction, are as shown below. Also shown are the percentage increases in
volume which would result if the elasticity of demand for steel were as high as 1.
Table 3. — Increases in Volume Needed to Compenstate for Decreases in Average
19S8 Prices, Compared to Maximum Probable Resulting Increases in Volume —
United States Steel Corporation and Subsidiaries
Reduction in price
Increase in
volume re-
quired to oflE-
set price
decrease
Greatest
probable re-
sulting in-
crease in
volume
Reduction in price
Increase in
volume re-
quired to off-
set price
decrease
Greatest
probable re-
sulting in-
crease in
volupae
1%
19! 6%
48.8%
M
15%
96. 7%
190. 3%
17. 7%
25.0%
5%
20%
10%
7. At average prices prevaihng in the second half of 1938, subsequent to the
June 24, 1938 price reduction, the percentage increases in volume required to
offset further price reductions would be even higher.
14036 CONCENTRATION OF ECONOMIC POWER
Table 4. — Increases in Volume Needed to Compensate for Decreases in 2nd Half
1938 Prices, Compared to Maximum Probable Resulting Increases in Volume —
United Slates Steel Corporation and Subsidiaries
Reductioa in price
Increase in
volume re-
quired to
offset price
decrease
Greatest
probable
resulting
increase in
volume
Reduction in price
Increase in
volume re-
quired to
offset price
decrease
Greatest
probable
resulting
increase in
volume
24^0%
63.1%
1.0%
5.3%
11. 1%
138.4%
342. 6%
17.7%
w
2Qcy
25.0%
,/fc
8. No percentage of capacity as ordinarily computed can be named as the
break-even point for any given level of steel prices, since the exact percentage at
which the Corporation would break even will also depend upon the type of
products composing the total tonnage sold and these vary considerably from month
to month and from year to year, even when total shipments are approximately
the same. On a weighted tonnage basis, the break-even point for the Corporation
at prices prevailing in the second half of 1938 would be at about ten and a half
miUion weighted tons. Depending on the type of products sold, this weighted
tonnage would be equivalent to from 50% to 55% of capacity.
9. If prices were reduced to the point where the Corporation would break even
only if operations averr.ged around 90% of capacity, which some critics have
suggested as the proper break-even point, the Corporation would have to operate
at the impossible annual rate of 130% of capacity to make a 5% return on its
investment in tangible assets.
III. Costs of United States Steel Corporation in Relation to Volume
Costs must of necessity fall into one of two categories. Some items of cost are
the same regardless of the amount of steel and other products produced providing
tlvere is not a complete shutdown. These costs are known as "fixed costs" or
"overhead costs". There are other items of cost termed "variable costs" or
"incremental costs" or "additional costs" which are not the same regardless of
volume but increase with increases in the vc-ame of steel produced and sold.
These costs can be diminished by cutting down the production of steel. The
fixed costs, on the other hand, cannot be diminished except by complete shutdown,
but they can be spread over a greater number of units of products by increasing
production. Both the variable and the fixed costs are composed both of costs
which embrace current cash outlays, and of depreciation and depletion, which
do not.
The cost pattern of the United States Steel Corporation and its subsidiaries,
set forth in the summary of findings in Section II above, has been derived from
the profit and loss statements of the Corporation as set forth in a financial report
covering the years 1917 to 1937, submitted to the Federal Trade Commission on
February 17, 1939, in connection with the Commission's studies relative to the
Temporary National Economic Committee. A profit and loss statement for 1938
on a comparable basis was prepared by the Comptroller's Department of the
Corporation. These profit and loss statements are basically the same as those
contained in the annual reports of the Corporation, except that some of the items
have been reclassified in order to render all of the yearly statements comparable
and to comply with certain requests of the Federal Trade Commission. Only
the figures from 1927 onward were used in this study because the data for making
the necessary adjustments were not available for prior years.
The total operating costs as shown on the profit and loss statements referred
to consist of the cost of goods sold and operating exj^enses of transportation and
other incidental operations, plus the selling, general and administrative expenses,
and taxes other than Federal income and profits, plus proNisions for depreciatitJn
and depletion. To the total of these operating costs was added the ai.iount of the
idle plant expenses, and from t.iat total was subtracted the amount of the dis-
counts from purchases. These last two items were classified in the report under
"Other Income Deductions" and "Other Income," respectively, but it was felt
that they were sufficiently connected with operations to be included in thv. oper-
ating costs for the purpose of this study. Also added in was the amount of the
interest on bonds, mortgages and other long-ieriu debt. Items classified under
CONCENTRATION OF ECONOMIC POWER 14037
"Other Income" and "Other Income Deductions," with the exception of the idle
plant expenses and purchase discount*, were considered to be extraneous non-
operating income and expense and were not included in the computation. Federal
income and profits taxes were excluded for a different reason. These taxes are
dependent on profits, not on costs or volume of business, and hence will differ
with different prices that may prevail. Since the purpose of this study was to
chart a cost curve which could be related to estimated sales and revenues at
various price levels, these profits taxes have been omitted from the costs and any
profits or losses shown in this study are before provision for Federal income and
profits taxes. The various components of total costs, as extracted from the
above mentioned report, are shown in columns 2, 3, 4 and 5 of Table 5.
Table 5. — Compilation of Total Costs Exclusive of Inter-Company Items and Costs
Connected with Extraneous Transactions ' — United States Steel Corporation and
Subsidiaries
(l)
Year
(2)
Operating
Costs 2
(3)
Add: Idle
Plant
Expenses
(4)
Less: Pur-
chase Dis-
counts
(5)
Add: Bond
Interest
(6)
Less: Esti-
mated Inter-
Company
Transactions
(Table 6)
(7)
Total Costs
After Elimi-
nation of
Inter-Com-
pany Items 3
1927
1,211,837,563
1, 246, 408, 662
1, 298, 859, 622
1, 078, 516, 167
733, 392, 971
424, 275, 927
557, 631, 125
606, 196, 210
769, 564, 100-
1,035,128,989
1,269,970,378
765, 723, 200
1, 822, 861
2, 550, 146
1, 679, 794
2, 516, 129
2, 904, 457
3, 084, 823
3,160,905
,2,694,390
2, 089, 259
1, 396, 989
1,136,149
2, 440, 185
1, 000, 490
909,916
1,153; 639
831, 634
275, 219
525, 447
716, 245
939, 641
1, 455, 169
1, 815, 869
993, 387
26,063,504
25, 746, 009
14, 944, 870
5, 640, 096
5, 469, 624
5,313,461
5, 164, 453
5,051,052
4, 959, 780
4,918,431
5, 141, 088
8, 262, 327
371, 721, 399
389, 268, 104
434, 445, 063
360, 637, 726
201, 536, 082
71, 173, 131
150, 992, 224
170, 362, 492
236,439,013
308, 210, 336
373, 94?, 630
160, 898, 7:3
1928
884 526 797
1929
880 055 337
1930
724 881 027
1931
1932
539, 399, 336
1933 -
1934
1935.
539, 234, 485
1936 _
1937.. ._
1938-- --
731,778,904
900, 484, 116
614, 533, 572
Total
10, 997, 504, 914
27, 476, 087
11, 600, 542
116, 674, 695
3, 229, 631, 953
7,900,423,201
' The figures herein contained, except the estimated inter-company items, have been taken from the
consolidated profit and loss statements submitted to the Federal Trade Commission on 2/17/39. The
figures are reconcilable with those appearing in the Annual Reports, but are not exactly the same, as some
reclassification was necessary in order to comply with the Commission's request and also to put the state-
ments for all years on a comparable basis.
2 As per statement to Federal Trade Commission consists of (1) cost of goods sold and operating expenses
of transportation and other incidental operations; (2) selling, general administrative expenses, and taxes
other f an Federal income and profits taxes; (3) provisions for depre' iation and depletion. Additions to
bond sinking fund reserve are included in depreciation in 1927 and 1928.
' Excluding income and expense connected with non-operating transactions.
The costs that would be obtained by adding together these components, how-
ever, would involve a certain amount of duplication, for on the profit and loss
statements both the sales and revenues and the cost of goods sold and operating
expenses of transportation and miscellaneous operations include transactions
between the subsidiary companies of the United States Steel Corporation. Since
the inter-company sales of any one company constitute the costs of the other,
and since inter-ccmpany profits are eliminated from inventory valuations in
making inventory adjustments, both costs and sales and revenues are inflated,
from a consolidated viewpoint, by the amount of the inter-company items, but
the profit shown is unaffected by this method of handling.^ These inter-company
transactions are segregated in the accounts of the Corporation and its subsidiaries
and in the annual reports so far as the sales of products and some of the miscel-
laneous operations are concerned, but the accounting systems of the common
carrier transportation companies make no provision for this segregation. Con-
sequently, the inter-company business done by the common carrier transportation
companies must be estimated. Estimates of the net sales and operating revenues
with all of the inter-company items eliminated had already been prepared by the
Comptroller's Department of the Corporation. By subtracting these net sales
and operating revenues derived from outside sources from the total of net sales
and operating revenues shown in tlie profit and loss statements submitted to the
Federal Trade Commission, the estimated amount of all inter-company items was
' See Appendix VTT for illustration and more complete discussion.
14038
CONCENTRATION OF ECONOMIC POWER
ascertained. This computation is shown in Table 6. Since the relationship
between the physical volume of outside business and the cost of doing all outside
business was what was desired, the inter-company items had to be deducted from
the costs which were taken from the report ot the Federal Trade Commission.
The deduction of the inter-company items estimated in Table 6 is made in column
6 of Table 5. Column 7 of Table 5 shows the total operating costs on an integrated
basis for each of the years 1927 to 1938.
Table
-Estimate of Inter-Company Transactions-
tion and Subsidiaries
■United States Steel Corpor
Year
Net Sales and
Operating
Revenues '
Net Sales and
Operating
Revenues less
Inter-Company
Items 3
Estimated
Inter-Company
Items
1927 .. - -
1,318,334,399
1,381,843,104
1,502,211,063
1,175,046,726
725,248,082
354,693,131
521,943,224
588,835,492
776,348,013
1,099,931,336
1,395,649,630
766, 673, 763
946, 613, 000
992, 575, 000
1,067,766,000
814, 409, 000
523, 712, 000
283, 520, 000
370, 951, 000
418,473,000
539, 909, 000
791,721,000
1,021,602,000
605,775,000
371,721,399
1928 -. --
389, 268, 104
1929
434, 445, 063
1930 - -. ..
360, 637, 726
150, 992, 224
170,362,492
1935 - - —
236, 439. 013
1936 ..
308,210,336
1937 -
373, 947, 630
1938 -
160,898,753
Totals .
11, 606, 657, 953
8, 377, 026, 000
3, 229, 631, 953
1 As per Statement to Federal Trade Commission, 2/17/39.
' As per data prepared by Comptroller's Department of U. S. Steel Corporation— Does not include
Miscellaneous Non-Operating Income.
To ascertain the relationship between costs and volume, it was necessary to
compare the above costs with the related volume figures. Since the cost figures
were obtained from the profit and loss statements, they represent the cost of goods
sold rather than the cost of goods manufactured and hence are comparable with
the number of tons shipped each year rather than with the number of tons pro-
duced. In relating costs to shipments, however, it is necessary to make some
adjustment for changes in the proportions of high and low cost products constitut-
ing the total. For instance, if the shipment figures showed that in each of two
years 10,000,000 tons had been shipped, the total costs would be greater in the
year in which costly products, such as sheets and tin plate, constituted most of
the tonnage than in the year when lower cost products, such as rails and heavy
plates, predominated. Adjustment for such a condition has been made by letting
each ton of rolled and finished steel product which is of a type whose 1933-1937
average mill cost was less than the average cost of all rolled and finished steel
products, count proportionately less than -a full ton, while tons of products of a
class which is on the .average more costly than the average cost of rolled and
finished steel products, have been made to count proportionately more than a
full ton. By weighting in this way the number of tons of all tonnage products
shipped each year has been converted into eguivalent tons of average cost rolled
and finished steel products.' Thus in any years in which the 1933-1937 average
proportions of high and low cost products were sold the actual tonnage is the same
as the weighted tonnage, while the total tonnage for years having an abnormal
proportion of high or low cost products has been converted to a tonnage figure
that is comparable to the normal years from a cost standpoint. By a similar
weighting process, the total tonnage of rolled and finished products shipped has
been further adjusted to include the equivalent tons of steel represented by the
products other than steel which are sold on a tonnage basis by the Corporation's
subsidiaries.
• The total tons shipped are gross tons except with respect to a few products Included on a net tonnage
lasls, and the average cost per ton is the a^irage for this mixed tonnage.
CONCENTRATION OP ECONOMIC POWER
14039
Table 7. — Total Costs (Unadjusted) and Volume of Business — United States Steel
Corporation and Subsidiaries
Year
Total Costs
(Table 5)
Millions of
Weighted
Tons of
Product?
Shipped
Year
Total Costs
(Table 5)
Millions of
Weighted
Tons of
Products
Shipped
$867,002,039
884.526,797
880,055.337
724.881.027
539,399.338
361,225,861
13.0
14.0
15.1
U.9
8.1
4.4
1933
$414,438,812
442,862,915
539,234,485
731,778,904
900,484,116
614.-533,672
6.2
1928
1934 .
6.1
7.6
1930
1936
11.0
13.2
1938
7.8
In Tabl^ 7 the total costs as shown in Table 5 are compared with the weighted
tons of all products shipped. In Chart 1 these costs are plotted in relation to the
weighted tons shipped. It is apparent from this chart that there is a close rela-
Chart
1
TOTAL COSTS (UNADJUSTED) AND VOLUME OF BUSINESS
U. a STEa CORPORATION AND SUBSIDIARIES
1100
° 8 8 8 i 8 8 8 8 8 i i ^
MILLIONS OF DOLLARS
\
,,
''''
1 ^
g ^
Q 700
o 600
- 500
2 400
200
100
0
\
1937
y
1927
mi
IK^!*
1936
.''
1930
"?
,^
■''
'
y.
ilwi
1»3J
'.w'i
"l
?»''
■'
(
HOTt: COSTS**
J 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1
MIUJONS OF WEIGHTED TONS OF ALL TONNAGE PRODUCTS SHIPPED
E una 80ND INTtREST. BUT BEfORE fEOtRM. JNCOME TM *ND EXCLUSI'/t OF NON-OPERATING ITEMS »N0 INTER-COMPA
8
NY TIWNSACTIONS
tionship between total costs and tonnages shipped and that this relationship is
of such a nature that it appears as a straight line in the graph. Fitting a straight
line in such a. way that the sum of the squares of the vertical deviations of each
of the points from the line is at a minimum, it appears that the average relation-
ship between costs and volume for the period 1927 to 1938 was such that the total
costs consisted of $54.51 per ton sold, plus fixed costs of $120,530,000. However,
this relationship represents the relation prevailing on the average under all the
changing conditions that took place between 1927 and 1938. An inspection of
the chart reveals clearly that the costs in the earher years were generally lower
than average, while in the later years they have been above this average. This
shows that the present-day costs of -producing steel are higher than they formerly
were. To ascertain what the relationship between cost and volume would be
under 1938 conditions it was necessary to adjust the cost figures for each of the
prior years for changes in interest, tax and wage rates, material prices, and other
conditions that have since taken place.
Since different changes have taken place in the various types of expenses, the
first step in attempting an adjustment of the data to 1938 conditions v.as to break
down the total costs into their principal components. Table 8 gives this break-
down. The components shown are (1) taxes other than Federal income and profits
14040
CONCENTRATION OF ECONOMIC POWER
taxes and social security taxes, (2) interest, (3) depreciation and depletion, (4)
payroll, (5) pensions, (6) social security taxes, and (7) other expenses. Payroll
figures do not exactly reflect the salary and wage cost in the goods sold because
some payroll goes into inventory and some of the goods sold are taken from the
previous year's inventory. To the extent that payroll is more or less than the
wage and salary cost in goods sold in any particular year because of production
exceeding or being less than shipments, the "other expense" component is under-
stated or overstated. The amount of inventory fluctuation is relatively small,
however, and the effect on the ultimate cost computation is negligible. With
the breakdown of costs computed, the adjustments applied to each component
must be considered.
Two of the components, interest and pensions, are in no way dependent upon
the volume of business performed. These items are completely fixed. Hence,
to convert to 1938 conditions, all that had to be done was to substitute in each
year the 1938 amount of interest and pensions paid.
Tablk 8. — Analysis of Operating Costs Into Components — United States Steel
Corporation and Subsidiaries
Taxes
Other
Total Costs
than
Deprecia-
tion and
Deple-
tion'
Year
Before In-
come Taxes
as per
Federal
Income
and Prof-
Interest
Payroll '
Pensions'
Social Se-
curity
Taxes
Other E\-
penscs
Table 6
its Taxes
and Social
Security '
1927
867, 002, 039
34,817,116
26, 063. 504
58, 906, 007
410, 289, 135
2,414.226
334,512,051
36, 015, 942
37, 739, 322
2, 924, 879
352, 602, 172
1929
880,055. .37
14, 944, 870
63, 274. 163
406, 886, 492
3,354,504
353, 855. 986
1930
724,881 J27
« 36, 047, 026
5, 640, 096
58, 550, 120
367, 945, 736
3, 772, 053
2.52,925,996
1931
5.39, 399, 336
33, 162, 707
5, 469, 624
47, 317, 895
253, 178, 649
5. 241, 466
195. 028. 995
1932
361, 225, 861
31,943,315
5, 313, 461
40, 319, 794
131,602,678
6,904,978
145,141,635
:933
414,438.812
33, 288, 485
5,164,453
43, 584, 499
160, 746, 223
7, 163, 032
164, 492, 120
lb34
442, 862, 915
32, 615, 83!
5, 051, 052
44, 579, 309
207, 564, 103
7, 223, 546
145, 829, 074
1935
539, 234, 485
»34,691,33C
4, 959, 780
47, 801, 389
246, 508, 043
7, 362, 723
197,911.220
1936
731,778,904
37, 999, 606
4,918,431
55, 466, 762
328, 070, 724
7.642,026
4, 081, 587
293, 599, 768
1937
900,484,116
45, 132, 333
5,141,088
59, 589, 159
426, 330, 944
7, 380, 254
13,415,904
343, 494, 434
1938
614, 533, 572
34, 602, 915
8, 262. 327
48, 532, 841
275, 364, 898
7, 743, 046
11,309,216
228, 718, 329
7, 900, 423, 201
428,055,928
116, 674, 695
635, 159, 241
3,614,488,117
69, 126, 733
28, 806, 707
3,008,111,780
1 Total taxes as per profit and loss statements in annual reports, less reserves for Federal income and prof-
its taxes as per report to Federal Trade Commission and social security taxes as per profit and loi s jtatements
in annual reports.
> Does not include cost of dismantling, moving and rearranging facilities, which is Included in deprecia-
tion and depletion in the profit and loss statement in the annual reports from 1936 on.
3 Excluding construction payroll.
* Excluding certain taxes charged against special income.
5 As per 1936 annual report— iron ore taxes not segregated from other overhead in 1935 annual report
profit and Joss statements.
» Figures for 1928 to 1938 as per annual reports and the S. E. C. registration statement. Form A-2; figure
for 1927 «upplied by Comptroller's Department of the Corporation.
Since tax rates have changed considerably over the period under consideration,
it is clear that some adjustment in the tax figures was necessary. In Table 9
the taxes, other than social security and Federal income and profits taxes, are
Table 9. — Taxes and Volume of Business — United Stales Steel Corporation
and Subsidiaries
Year
Taxes Other
than Social
Security and
Federal In-
come and
Profits Taxes
Millions of
Weighted
Tons of
Products
Shipped
Year
Taxes Other
than Social
Security and
Federal In-
come and
Profits Taxes
Millions of
Weighted
Tons of
Products
Shipped
34,817,116
36, 015, 942
37, 739, 322
36, 047, 026
33, 162. 707
31,943,315
13.0
14.0
15.1
11.9
8.1
4.4
1933 . ...-
33,288,485
32,615,831
34, 691, 330
37, 999, 606
45, 132, 333
34,602,915
6.2
fi. I
7.6
1936
U.O
1931
1937
13.2
1932
1938
7.8
CONCENTRATION OF ECONOMIC POWER
14041
compared with the weighted tons of products shipped. The relationship be-
tween taxes and volume was then ascertained by plotting taxes against shipments
as shown on Chart 2. An inspection of the chart shows that the points for 1927
to 1931 fall approximately along one line, while the points for 1932 to 1938 fall
along another. The simplest estimate of what the taxes would be at various
rates of ojjeration under 1938 conditions was to compute the average relationship,
by the least squares method referred to above, for the period 1932 to 1938.
It will be noted that the 1938 taxes fall very close to the line and very close
to the points for other years when volume was about the same. Hence it is
evident that the tax rates which affect the Corporation with regard to taxes
other than Federal income and social secuiity taxes, have not been materially
Chart 2
TAXES AND VOLUME OF BUSINESS
U. S. STEEL CORPORATION AND SUBSIDIARIES
60
60
50
50
,,
L —
5
40
__,
«i
I'?,
'nV
L'?— -.
a'i--
1938 »?.
•-^ -
11-
—
15*27
o
30
^-^ '
19M
30 o
^,,-'—
z
z
o
20
20 2
zi
-J
z
10
10
0
0
C
) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
MILUONS OF WEIGHTED TONS OF ALL TONNAGE PRODUCTS SHIPPED
NOTt: TAOS INaUOE AU TAXES KCtPT SOCIW. SECURITY AND FtDERAl (NCWrlE AHO fKOrn% TAXES
altered throughout the 1932-1938 period. The taxes paid from 1927 to 1931,
however, are clearly much below what would be paid today at the same operating
rates. For this reason in this study the actual taxes paid for each of the years
1932 to 1938 have been used in computing the estimated total costs under 1938
conditions, but the taxes for the years prior to 1932 have been estimated on the
laasis of the average 1932 to 1938 tax-volume relationship. This relationship
indicates that total taxes, other than Federal income and social security, amount
to $24,217,000, plus $1.43 for each weighted ton of product sold. The calcula-
tion of the estimated taxes for the years 1927 to 1931 is shown in Table 10.
The final estimate of the t;. <es, other than social security and Federal income
taxes, that would be paid under present tax laws at various volumes of production
is shown in Table 11.
14042 CONCENTRATION OF ECONOMIC POWER
Table 10. — Estimated Taxes (Other Than Social Security and Federal Income) in
Prior Years if 19S2-19S8 Rates Had Prevailed— United States Steel Corporation
and Subsidiaries
[Taxes =$24,21 7,0004-$I.433 per weighted ton shipped <]
Year
Millions of
Weighted
Tons Shipped
FiMd (Mil-
lions of
Dollars)
Variable
(Millions of
Dollars)
Tota'. Taxes
(Millions of
Dollars)
1927
13.009
13.994
15.089
11. 935
8.131
24.217
24.217
24.217
24.217
24.217
18.642
20.053
21.623
17. 103
11.652
42 859
1928
44 270
1929
45.840
1930
41. 320
1931 . .-
35. 869
1 1932-1938 average relationship of taxes in relation to volume.
Table 11. — Taxes and Volume of Business, Under Present Tax Laws — United
States SteeV Corporation and Subsidiaries
Year
Taxes Other
than Social
Security and
Federal In-
come
Millions of
Weighted
)Tons of Prod-
ucts Shipped
Year
Taxes Other
than Social
Security and
Federal In-
come
Millions of
Weighted
Tons of Prod
ucts Shipped
1927
> $42, 859, 000
■ 44, 270, 000
'45,840,000
• 41, 320. 000
■35,869.000
31, 943, 315
13.0
14.0
15.1
11.9
8.1
4.4
1933
33, 288, 485
32, 615, 831
34, 691, 330
37,999,606
45, 132, 333
34, 602, 916
6.2
1928
1934 .._
1935
6.1
1929
7.6
1936
11.0
1931 . .
1937
13.2
1932.
1938
7.8
' Amouats for years prior to 1932 estimated on basis of 1932-1938 average relationship between cost and
volume (see Table 10).
On Chart 3 have been plotted the depreciation and depletion charges for each
of the years against the weighted tonnage shipped. A computation of the line
of average relationship by the least squares method referred to above shows that
the total depreciation and depletion charges have averaged $29,500,000, plus
$2.37 for each weighted ton produced. ' An inspection of the chart with the line
drawn in shows that all of the points fall rather close to the Une and deviations
that exist are more or less at random and do not seem to be the result of conditions
changing with the passage of time. Hence, the actual depreciation and depletion
charges for each year have been used without adjustment.
Table 12. — Adjustment of Payroll to 19S8 Wage Rates — United States Steel Cor-
poration and Subsidiaries
Year
PayroU
Averse
Hourly
Earnings
Ratio of 1938
Average Hourly
Earnings to
Tliose of Year
of PayroU
Estimated
Payroll at 1938
Rates
410, 289, 135
400.000,492
406, 886. 492
367, 945, 736
253, 178. 649
131, 602. 678
160, 746, 223
207, 564, 103
246, 508, 043
328,070,724
426. 330, 944
275, 364, 898
1.682
.687
;687
.691
.614
.596
.705
.731
.737
.864
.902
1.3226
1.3130
1. 3149
1. 3130
r.3054
1. 4691
1. 51,34
1.2794
1. 2339
1.2239
1. 0440
1.0000
542, 648, 410
1928. - -
625, 200, 646
535,015,048
483,112,751
1931
330, 499, 408
1932 —
193. 337, 494
1933 - -
243. 273. 334
1934
265, 557, 613
1935
304, 166. 274
1938
401. 525, 759
1937
445, 089, 506
1938
275, 364, 898
CONCENTRATION OF ECONOMIC POWER 14043
Wage rates have increased considerably from the low point of the depression
and even from the prosperity year of 1929. Hence, the need for adjustment of
the payroll figures to 1938 conditions is obvious. This adjustment is shown in
Table 12, v/here the payrolls for each of the respective years are multiplied by the
ratio of the 1938 average hourly earnings to the average hourly wage prevailing
in the year of the payiolls. Chart 4 shows the relation to volume of the payrolls
for each of the respective years adjusted to 1938 wage rates. A line of average
relationship has been constructed by the least squares method. An inspection of
the chart shows that, in general, payrolls for the later years are lower than what
they would have been for the earlier years if 1938 hourly rates had prevailed.
This indicates that, to some extent at least, the increases in hourly wage costs
have been offset by increased productivity per man hour. Adjustment for this
factor has been taken care of, however, by the adjustment for long-term trend
in the cost volume relationship which will be discussed later.
Closely connected with the adjustment of payroll is the estimate of social
security 'taxes. These taxes did not exist prior to 1936, and the rates have been
Chart 3
DEPLETION AND DEPRECIATION RELATED TO VOLUME OF BUSINESS
U. S. STEEL COKHUHATION AND SUBSIDIARIES
70
^ 60
<
o 50
o
- 40
^ 30
o
d 20
s
10
0
1
1
70
60 1
50 §
40 fe
30 1
20 1
10
1928
^'
>'l9S9
'"
n-R
-
,.
J935
193_Oj
-T927
1937
'9«,-
193^
1
r
']931
,,
„,-
,-'
'''
(
3 1 2 3 4 5 6. 7 8 9 10 11 12 13 14 15 16 17 1
MILLIONS OF WEIGHTED TONS OF ALL TONNAGE PRODUCTS SHIPPED
8
increasing since then. The best method of estimating what the social security
taxes would have been for each of the respective years under 1938 conditions
would seem to be to take the ratio of the 1938 social security taxes to the 1938
payroll and apply this ratio to the adjusted payroll figures for each of the respec-
tive years. In ascertaining the ratio of social security taxes to payroll, the total
payroll, including construction payroll, has been used, since the„total social security
tax paid is based on the entire payroll, including that charged to construction as
well as that charged to operations. The pajToll figures used in the cost computa-
tion, however, consist only of operating payroll and do not include construction
payroll, which is properly treated on the books of the Corporation as a capital
expenditure. The estimated amount of social security taxes for each of the
respective payroll figures is shown in Table 13.
The classification "other expenses" shown in the breakdown of total costs is
simply the residual amount after deducting each of the other components. It is
impossible to obtain a satisfactory breakdown of this item because, in order to
keep the costs of different operations and products properly segregated, the
accounts of- the United States Steel Corporation are classified on an entirely
14044 CONCENTRATION OF ECONOMIC POWER
different basis from that used in this study. "Other expenses" does consist,
however, largely of goods and services purchased from others, including freight
paid for transporting materials, although freight paid on shipments of finished
goods to customers is treated as a deduction from sales. To some extent at least,
therefore, these expenses J^ave been affected from year to year by changes in the
commodity price levels.
Table 13. — Estimate of Social Security Taxes at 1938 Rates — United States Steel
Corporation and Subsidiaries
1938 Social Security Tax -f- Total 1938 Payroll ' = 1938 Average Rate]
$11,309,216 -^ $282,209,332 = .040074 J
Year
Payroll at 1938
Wage Rates
Social Security
Taxes at 1938
Rates
Year
Payroll at 1938
Wage Rates
Social Security
Taxes at 19as
Rates
1927
$542, 648, 410
525, 200, 646
.535,015,042
483,112,751
330. 499, 408
193, 337, 494
$21, 746, 092
21, 046, 891
21, 440, 193
19, 360, 260
13, 244, 433
7, 747, 807
1933...
1934 _
1935
$243, 273, 334
265, 557, 513
304,166,274
401, .523, 759
445, 0S9, 506
275. 364, 898
$9 748 932
1928
1929
10,641,952
12 189 159
1930
1936
"* 1 6, 090 743
1931
1937.
1938
17 836,517
1932 .
11,034,973
' Including Construction Payroll.
It might be said that no adjustment should be made for the changing prices of
materials and services purchased on the theory that such prices are generally
loifrer when steel production is low and higher in periods of prosperity when steel
production is high, and that a computation of the relation of volume to cost should
include the normally expected change in. material prices. However, it was felt
that it was advisable to make a broad general adju.stment which would roughly
eliminate the effect of the low prices wnich prevailed in the low volume years for
the following reasons:
1. The purpose of the study is to ascertain the extent to which a price decrease
as of 1938 could be compensated for by increased volume as distinguished from
reducing wage rates, or having the benefit of reduced material costs, either of
which would permit of some reduction in the price of steel.
2. Comparison of the "other expenses" for 1931, 1935 and 1938, three j-ears of
approximately similar volume, shows that under 1938 conditions prices of materials
did not decline as much as they formej-ly did when steel operations were low.
3. An arbitrary reduction in steel prices, the advisability of which is being
considered in this study, .would be less likely to be accompanied bj' a decline in
material prices than would a decline in steel prices brought about by natural
economic forces.
4. Adjustment for material price changes tends to raise the cost figures for the
low volume years, thus decreasing the slope of the total cost line and increasing
the estimate of the extent to which increased volume would decrease unit costs.
Hence, adjustment rather than nonadjustment presents a case more favorable to
the critics of steel prices.
On the other hand, since many of the things the Corporation must purchase
did not drop in price to the same extent that commodities in general did, overad-
justment had to be avoided. The basis of adju.stment used has been the U. S.
Bureau of Labor Statistics index of wholesale commodity prices other than farm
and food products. In order not to overadjust, only half of the "other expense"
items have been adjusted by multiplying them by the ratio of the 1938 index to
the index prevailing in the year in which the "other expenses" were incurred. To
the half of the "other expen.ses" thus adjusted has been added the remaining half
without adjustment. The calculation of the adjustment and the final adjusted
figures are shown in Table 14.
CONCENTRATION OF ECONOMIC POWER
Chabt 4
14045
MILLIONS OF DOLLARS 5
oissiiii 1
ADJUSTED TO 1938 WAGE RATES AND RELATED TO VOLUME OF BUSINESS
U. S. STEEL CORPORATION AND SUBSIDIARIES
^,'''
600
00
a.
500 <
o
400 ^
o
300 w
o
200 li
100
0
1927
1928
,.^'<^
^'
1930
-'''
*1937
,935,
■93^.
^'^
1936
^
1933
-"
"was
'^
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
MILLIONS OF WEIGHTED TONS OF ALL TONNAGE PRODUCTS SHIPPED
NOTE: CONSTRUCTION PAYROUS EXCLUDED
Table 14.
'Other Expenses" Adjusted for Changes in Price Level — United States
Steel Corporation and Subsidiaries
Millions of
Weighted Tons
of Products
Shipped
4.4. _
1932
6.1
1934
6.2
7.6
1933
1935
7.8-
8 1
1931
11.0
1936
11 9
1930
13 0
1927
13.2
1937
14 0
1928
15 1
1929
Total
......
"Other Ex-
penses" Un
adjusted
145, 141,
145, 829,
164, 492,
197,911,
228, 718,
195, 028,
293, 599,
252, 925.
334, 512,
343, 494,
352, 602,
3,13, 855,
(Dj
570. 818
914,537
246, 060
955,610
359, 165
514, 498
799,884
462, 998
256,026
747, 217
301,086
927, 993
,055,i
(E)
70.4
78.4
71.2
77.9
81.7
75.0
79.6
85.2
94.0
85.3
92.9
91.6
(F)
Index to
That of
Year of
Costs
1.1605
1.0421
1. 1475
1. 0264
0.9589
0. 8691
0. 9578
0. 8794
(Q)
50% of Costs
Adjasted for
Price Change
(DXF)
377, 354
784, 644
359, 165
222, 543
875, 401
265, 369
362, 212
499, 484
039, 175
802, 077
(H)
Adjusted
'Other Ex-
pen.se"
(D+Q)
156,789.252
148, 898, 776
176, 623. 414
202, 740, 254
228,718,330
203, 737, 041
297, 475, 285
247, 728, 367
312, 618, 238
336, 246, 701
331, 340, 261
334. 730, 070
2, 9;i7, 645. 989
B. L. S. Index of Wholesale Prices— all commodities except food and farm products.
14046
CONCENTRATION QF ECONOMIC POWER
a
2
i^sSiSi
s§IS.sl
5
a
o
1
1
6
^
^■-"
X
1
« -
g
,
sg^Sisiiigsg
a
^
ll
liilSg^'giiii
«■
s
1
^
§s"§iHi§§s'ii§
1
■1
«
>>
iil§giSiiSi§
i"
11
^
9-"
t,-o cTcj 2"« 2 2 S ^ c3 53
i
11
■§
(^
1
i
si
isssiiisiiii
« 2
1
a§giiis"H'iiis
1
°3
Ig-gigio-iHisl
11
■3g
11
gl »iliiii§iSi§
S
ll
it
1?
2gSsi^lgllSS
i
•c5
'^1
o :$ ^"^"« !j S S S S IS S
g^
o c
o-O
1
S"5;
e
pa
„ o
*^ eg
■SH
1
ssssssssssss
s
,j
S
1
II
tC tC t~." h-T r-" t-" J^ t~r t-^ t-" t~ t-^
w2
1^
1 .
i
:-«
III!
^u^%m^%%n%%
oc
5
5 w
II
miMmmm
1°
dH^i^^^^^p^-^
1
fig
«
mmmmmm
111!
1-1
a
00 oo" oo 00 oo ao CO 00 00 oc CO 00
§
•S
2S
^.2ga
So
05
J
iiiiililliii 'i
j2
5
O
>-
- 1
1
1^11
i
ill!
"w
s
3
"1
!
iiil
a
UCQ
Sa
nisi
1
23
H
: 3
2|
iffli
h3
n
looioc^P
; ^
^?
:"*!?t
■«"<o<Of-r-
"^
CONCENTRATION OF ECONOMIC POWER
14047
In Table 15 the various components of cost for each of the respective years
adjusted to 1938 interest, tax, pension and wage rates and to 1938 price levels
have been added together to give total cost figures. These total cost figures
represent what would have been the cost for the volume sold in each of the
respective years if 1938 conditions of wages, interest, taxes, pensions and prices
had prevailed. These total costs, however, do not take into consideration the
long-term downward trend of costs in relation to volume due to changes in
methods of production within the industry itself and hence do not quite represent
what the cost would have been if the same volume had been sold in 1938. An
instance of this was referred to in connection with the adjustment of payrolls
where it was seen that the payrolls for earlier years would have been greater in
relation to volume produced than present-day payrolls if present-day hourly rates
were paid. Hence it became necessary to make a further adjustment of total
costs for any general increase in efficiency and other changes in operating conditions
that may have taken place during the period.
Table 16. — Deviation of Adjusted Total Costs ' from Average Costs for Volume
Involved — United States Steel Corporation and Subsidiaries
Year
MiUions of
Weighted
Tons of
Products
Shipped
Actual
Adjusted
Costs
Average
Adjusted
Costs for
Volume >
Deviation
Deviation
as Percent
age of
Average
Cost
4.4
6.1
6.2
7.6
7.8
8.1
11.0
11.9
13.0
13.2
14.0
15.1
446.1
518.3
, 522. 5
617.6
614.3
646.7
824.6
866.1
994.8
919.9
1005. 1
1016. 3
429.7
629.1
535.0
616.8
628.5
646.1
81.5. 6
868.2
932.5
944.2
991.0
1055. 3
-1-16.4
-10.8
-12.5
+0.8
-14.2
4-0.6
-(-9.0
-2.1
+62.3
-24.3
+14.1
-39.0
+3.8
1934 — - -
-2.0
1933 — -
-2.3
1935
+0.1
1938
—2.3
1931
+0.1
1936
+1.1
-0.2
-(-6.7
-2.6
1928
+1 4
1929
—3.7
' Total costs adjusted to 1938 interest, tax, pension, and wage rates and to 1938 price levels.
• As indicated by line of average relationship, Chart 5.
In Chart 5 the total costs adjusted to 1938 interest, tax, pension and wage rates
and 1938 price levels are plotted against the number of weighted tons shipped.
The line of average relationship computed by the least squares method has been
drawn upon the graph. In Table 16 have been set forth the actual adjusted costs
and next to them have been placed what those total costs would have amounted to
for the same volume if they were actually located on the line. In the third column
the deviations or difi'erences between the two are set down, and in the fourth
column, the deviations are reduced to percentages of the average cost for the
volume in question as indicated by the line of average relationship. These per-
centage deviations are plotted by years on Chart 6. An inspection of this chart
will show that many of the deviations are more or less at random, but that to a
certain extent costs in later years tend to be lower than average, while costs in
earlier years tend to be higher. This general trend, computed mathematically
by the least squares method, shows the extent to which the deviations from the
average are correlated with the passage of timie. The computation shows that
costs tended to be 2.17% above average at the beginning of the period and 2.15%
below average at the end. Thus the gradual increase in efficiency in the eleven
years preceding 1938 has been associated with a total drop in costs roughly equal
to 4.32% of average costs. This represents an average decrease in costs equal to
.393% of average costs per years. Thus, the costs under 1938 conditions can be
estimated by subtracting from the actual adjusted costs for any particular year
the amount of the average cost for that year's volume multiplied by .00393 times
the number of years intervenmg between the year in which the costs were incurred
and 1938. This adjustment is shown in Table 17.
14048
CONCENTRATION OP ECONOMIC POWER
Chart 5
TOTAL
AN[
1200
1100
1000
1 900
g 800
0 600
- 500
B 400
1 300
200
100
0
COSTS ADJUSTED TO 1938 INTEREST, TAX. PENSION AND WAGE RATES
) 1938 PRICE LEVELS, BUT UNADJUSTED FOR CHANGED OPERATING
CONDITIONS RELATED TO VOLUME OF BUSINESS
U.S. STEEL CORPORATION AND SUBSIDIARIES
—
..^'
1100
1000
900 1
1929
/
^■1929
1927
>1937
1936
^
1930
^^
■935.
k:.
600 0
500 ^
400 2
300 g
200
100
0
1934
.<
^1938
199
'>^
/
^'
J
^^
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
MILLIONS OF WEIGHTED TONS OF ALL TONNAGE PRODUCTS SHIPPED
Chart
PER
z
0
i
Z
LU
a.
CENTAGE DEVIATIONS FROM AVERAGE - TOTAL COSTS ADJUSTED T
INTEREST, TAX, PENSION AND WAGE RATES AND 1938 PRICE LEVE
BUT UNADJUSTED FOR CHANGED OPERATING CONDITIONS
U. S. STEEL CORPORATION AND SUBSIDIARIES
D 1938
L,
10
8
6
z
4 S
<
4 0^
6
8
10
8
6
4
.2
0
2
4
6
8
10
—
—
—
""""
■
glllssiilisi
CONCENTRATION OF ECONOMIC POWER
14049
Table 17.-
-Adjustment of Total Costs i for Time Trend-
Corporation and Subsidiaries
■United States Steel
A
Year
B
Millions of
Weighted
Tons of
Products
Shipped
C
Average
Adjusted
Costs for
Volume 2
D
Years
Prior
to 1938
E
Fractional
Decrease
(.00393XD)
Adjust-
ment
(CXE)
G
Actual
Adjusted
Costs
H
Final
Adjusted
Costs
(G-F)
1927
13.0
14.0
15.1
11.9
8.1
4.4
6.2
6.1
7.6
11.0
13.2
7.8
932.5
991,0
1,055.3
868.2
646.1
429.7
535.0
529.1
616.8
815.6
944.2
13
10
9
8
7
6
5
4
2
1
0
. 04323
. 03930
. 03537
.03144
.02751
. 02358
.01965
.01572
.01179
.00786
. 00393
.00000
40.3
38.9
37.3
27.3
17.8
10.1
10.5
8.3
7.3
6.4
3.7
0.0
994.8
1,005.1
1,016.3
866.1
646.7
446. 1
522.5
518.3
617.6
824.6
919.9
614.3
954.5
1928
966.2
1929 .
979.0
1930
838.8
1931
628.9
1932
436.0
1933..
1934
512.0
510.0
1935
610.3
1936
818.2
1937
916.2
1938
614.3
> Total costs adjusted to 1938 interest, tax, pension, and wage rates and to 1938 price levels,
s As indicated hy line of average relationship, Chart 5.
The resulting adjusted cost figures thus represent the original costs for each of
the years from 1927 to 1938, adjusted to 1938 interest, tax, pension and wage
rates, to 1938 price levels and to 1938 general efficiency. The final adjusted
figure for 1938 is, as it should be, the same as the unadjusted figure for that year,
except that the social security taxes allocable to construction payroll have been
eliminated. Table 18 sets forth these costs in relation to volume. The figures
Table 18. — Total Costs of Operation and Volume of Busi7iess — 19S8 Conditions ' —
United States Steel Corporation and Subsidiaries
Millions of Weighted
Tons of Products
Shipped
Costs, 1938
Conditions
Year on which
Estimate
Based
Millions of Weighted
Tons of Products
Shipped
Costs, 1938
Conditions
Year on which
Estimate
Based
436.0
510.0
512.0
610.3
614.3
028.9
1932
1934
1933
1935
1938
1931 j
11.0_ ..--
818.2
83,8.8
954.5
916 2
96(i. 1
979.0
1936
e.i
11.9. ._
1930
6.2
13.0
1927
76
13.2
1937
78
140
1928
8 1
15.1
1929
....
' Total costs adjusted to 193S interest, pension, wage and tax rates, to 1938 price levels, and to 1938
efficiency.
are plotted on Chart 7. A computation of the line of average relationship fitted
to those points by tiie least squares method shows that the normal relation of costs
to volunie under 1938 conditions is $55.73 per weighted ton shipped, plus .$182,-
100,000, as ^vas stated in the summary. The smallness of the deviations of the
actual costs, adjusted to 1938 conditions, from the normal cost line shows the
faithfulness with which the Corporation's costs follow this pattern.
IV. Composition of Costs
If it were not for the fact that the total costs obtained by adding together the
adjusted components of cost that were developed in connection with the deriva-
tion of the total cost pattern had to be adjusted for the gradual increase in effi-
ciency and other changes in conditions over the period, it would be possible to
obtain a breakdown of the fixed and variable total costs simply by listing the fxed
and variable elements of the various adjusted components. Since the total custs
were adjusted for these changes in operating conditions as indicated by the long-
term trend in the cost-volume relationship, an approximate breakdown could
be made only by making a similar adjustment for time trend for each of the indi-
vidual compoiMMits in which a time trend exists.
14050 CONCENTRATION OF ECONOMIC POWER
Since the 1938 amounts of interest and pensions were used for all volumes, no
time' trend adjustment had to be made with regard to these items. An inspection
of the movement of taxes and of depreciation and depletion with increases in
volume, as shown on Charts 2 and 3 respectively, reveals that no appreciable time
trend exists with regard to these items. The social security taxes will vary di-
rectly with any adjustments that are made in the payroll figure. It is apparent
that the bulk of the adjustment made in the total cost figures for time trend arose
from the time trends involved in the payroll and "other expense" items. Clearly,
then, it was these two items which had to be adjusted for time trend in order to
obtain an approximate breakdown of total costs.
Chart 4 shows the payroll for each of the various years adjusted to 1938 wage
rates and plotted against the volume of goods shipped in the year in which the
payroll was incurred. An inspection of this chart shows that, with the exception
of 1929, the payrolls in earlier years, would have been greater in relation to volume
than the payrolls in 1937 and 1938, if 1938 wage scales had been in efifect. In
Table 19 the deviations of the adjusted payrolls for each of the years from the
average payroll for the volume involved, as indicated by the line of average rela-
tionship on Chart 4, are computed. Also shown are the percentages by which
the actual payrolls, adjusted to 1938 wage rates, are in excess of or less than the
average payroll for the rate of operations. On Chart 8 these percentage devia-
tions are plotted by years.
Table 19. — Deviation of Payrolls, Adjusted to 1938 Wage Rates, from Average Pay-
roll for Volume Involved — United States Steel Corporation and Subsidiaries
Year
Millions of
Weighted
Tons of
Products
Shipped
Actual Pay-
roll Ad-
justed to
1938 Rates
Average
Adjusted
Payroll for
Volume '
Deviation
Deviation
as Per-
centage of
Average
r4.4
6.1
0.2
7.6
7.8
8.1
11.0
H.9
193.3
2G5.6
243.3
304. 2
275.4
330. 5
401.5
483.1
192.4
250.3
253.7
301.5
308.3
318.5
417.4
448.0
485.5
492 3
519.6
657.1
+0.9
+15.3
-10.4
+2.7
-32. 9
+12.0
-15.9
+25. 1
+57.1
-47.2
+5.6
-22.1
+0.5
+6.1
-4.1
+0.9
-10.7
+3.8
1936 -
-3.8
1930
+7.8
1927
1,3.0 i 542.6
+11.8
1937
13.2
14.0
15.1
445.1
525.2
535.0
-9.6
1928
+1.1
1929
-4.0
> As indicated by line of average relationship, Chart 4.
It is apparent from the chart that there has been some decline with the passage
of time in the amount of labor required to produce a given quantity of steel. Com-
putation of the least squares trend shows the extent to which the deviations from
the average are correlated with the passage of time. As indicated on the chart,
payrolls tended to be 7.19% above average in 1927 as compared to 7.22% below
average in 1938. This represents a gradual increase in productivity per man-
hour during the eleven years preceding 1938 that has resulted in a decrease in the
payroll recjuired for any given volume at 1938 wage rates equivalent to 14.41% of
the averas;e pavroU for that volume. This amounts to an average decrease in
payroll of 1.31% of average payroll per year. Thus the ]iayroll under 1938 con-
ditions can be estimated by subtracting from the actual payroll for any particular
year, adjusted to 1938 wage rates, the amount of the averace adjusted payroll for
that year's volume, multiplied by .0131 times the number of years intervening
between the year in which the payroll was paid and 1938. This adjustment is
shown in Table 20.
CONCENTRATION OF ECONOMIC POWER
Chart 7
14051
RELATIONSHIP BETWEEN TOTAL COSTS OF OPERATION
AND VOLUME OF BUSINESS - 1938 CONDITIONS
U. S. STEEL CORPORATION AND SUBSIDIARIES
1200
HOC
1000
9^ I
80) 1
700 °
600 o
500 ^
400 £
300 g
200
100
0
1100
1000
- 900
g 800
'^ 700
o 600
^ 500
5 400
i 300
200
100
^
r^
■
^'
r
1927
<:
.-928
^1930
^
1935^
^1
1934.
,«ni3
' '
■-.
^^
^X
/
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
MILLIONS OF WEIGHTED TONS OF ALL TONNAGE PRODUCTS SHIPPED
NOTE: TOT«. COSTS AOJUSTtD TO 1938 INTtREST, TAX. FWSION, UiD WAGE RATES; TO 1938 PRICE LfVElS. AND TO 1938 DMUINCV
Chart 8
PERCENTAGE DEVIATION FROM AVERAGE -
PAYROLLS ADJUSTED TO 1938 WAGE RATES
U. S. STEEL CORPORATION AND SUBSIDIARIES
1
—
—
-
V"-"
^-J
^_^
^-
1
1
i
1
^ \
I \
\
3
R ;
\
\ \
% \
\ \
\ \
14052 " CONCENTRATION OF ECONOMIC POWER
Table 20. — Adjustment of Payrolls, Adjusted to 1938 Wage Rates, for Time Trend —
Uniterir States Steel Corporation and Subsidiaries
(B)
Millions of
Weighted
Tons of
Products
Shipped
Average
Adjusted
Payroll for
Volume '
(D)
(E)
Fractional
Decrease
(F)
Adjust-
ment
(CXE)
Actual
Adjusted
Payrolls
(H)
Final
Adjusted
Payroll
(O-F)
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
13.0
14.0
15.1
11.9
8.1
4.4
6.2
G. 1
7.6
11.0
13.2
7.8
485.5
519.6
557.1
448.0
318.5
192.4
253.7
250.3
301.5
417.4
492.3
308.3
.117882
. 104784
.091686
. 0785S8
. 065490
. 052,'<92
. 039294
.026196
.013098
.000000
65.7
46.9
29.2
15.1
16.6
13.1
542.6
525.2
535.0
483.1
330.5
193.3
24.3.3
265.6
304. 2
401.5
445.1
275.4
472.7
457.1
469.3
436.2
301.3
178.2
226.7
252.5
292.4
390.6
438.7
275.4
' As indicated by line of average relationship, Chart 4.
Table 21. — Payrolls, Adjusted to 1938 Wage Rates and Efficiency, Related to Vol-
ume of Business — • United States Steel Corporation and Subsidiaries
Millions of Weighted Tons
of Products Shipped
Payroll-
1938 Con-
ditions
Year on
which
Estimate
Based
Millions of Weighted Tons
of Products Shipped
Payroll-
1938 Con-
ditions
Year on
which
Estimate
Based
178.2
252.5
226.7
292.4
275.4
301.3
1932
1934
1933
193f
im
1931
390.6
436.2
472.7
438.7
457.1
460.3
6.1
6.2
13.0
7.fi
13.2
14.0
7.8
8.1
15.1
1929
In Table 21 are shown the payrolls, at 1938 wage rates and at 1938 efficiency
as indicated by the time trend adjust;iients, for each of the various tonnages
shipped during the 1927-1938 period. On Chart 9 these final adjusted payroll
figures are plotted in relation to volume. A straight line, fitted to the points by
the least squares method, indicates that the total payroll under 1938 conditions
will amount to $02,100,000 plus $29.10 per weighted ton of product shipped.
Table 22. — Deviation of Adjusted ^ "Other Expenses" From Average for Volume
Involved — United States Steel Corporation and Subsidiaries
Year
Millious of
Weighted Tons
of Products
Shipped
Actual "Other
Expenses"
Adjusted for
Price
Average "Other
Expenses"
for Volume
Involved '
Deviation
De\:iation a.<!
Percentage of
Average
1932
4.4
6.1
6.2
7.6
7.8
K. 1
11.0
11.9
13.0
13.2
14.0
15.1
156. 8
148.9
176.6
202.7
228. 7
203. 7
297.5
247.7
312.7
336.3
331. 4
334.8
143. 1
175.8
177.7
204.6
208 4
214.2
269. 9
287.2
308.3
312.2
327.6
348.7
+ 1,3.7
-26.9
-1. 1
-1.9
4 20.3
-10.5
+27.6
-39.5
H-4.4
+24.1
+3.8
-13.9
+9.6
1933
-0.6
1935
-0.9
1938
+9.7
1931
-4 9
1936
+ 10.2
1930
— 13. 8
1927 .
+ 14
+7.7
1928
+1.2
1929
-4.0
' "Other Expenses" adjusted for changes in the general price level as indicated by the B. T,. S. index of
wholesale prices for commodities other than food and farm product*: assuming 50% of items affected.
' As indicated by line of a\ onige relationship, Chart 10.
CONCENTRATION OF ECONOMIC POAVER
14053
Chart
PAYROLLS, ADJUSTED TO 1938 CONDITIONS,
RELATED TO VOLUME OF BUSiUESS
U. S. STEEL CORPORATION AND SUBSIDIARIES
600
.^'
f 400
o
<o 300
z
o
=! 200
s
100
0
1930,
:k'«h
,''
im
1934
•l933
1938
1931
/-
^^
1932
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
MILUONS OF WEIGHTED TONS OF ALL TONNAGE PRODUCTS SHIPPED
NOTI: PAYDOIU ARE AOJUSTtO TO 1938 WACC RATES AND EfnCJENCr
600
tf>
cc
500 <
_i
o
400 °
u.
O
300 in
z
o
20b d
s
100
Chart 10
700
600
"OTHER tXPLNSES", ADJUSTED TO 1938 PRICE LEVEL,
RELATED TO VOLUME OF BUSINESS
U. S. STEEL CORPORATION AND SUBSIDIARIES
600
500 <
s
300 to
o
200 ^
s
100
0
< 500
! «o
o
V) 300
z
o
d 200
100
0
IPJ7,
i^.
1928
'1929
'''
1938^
• 1931
"'
--
1930*
^.
i9.
1934
59M
19*35
"'
^'
(
NOTE; AOJUSTVE
BY THE B
) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 15 17 1
MILLIONS OC WEIGHTED TONS OF ALL TONNAGE PRnnilCTS SHIPPED
NT TO 1938 PRICE LEVEL MADE ON ASSUMPTION THAT ONLY HAl/ OF ITEMS FLUCTUATE WITH THE GENERAL PRICE l£»a
LS. INDEX OF WHOLESALE PRICES FOR COMMODITIES OTHER THAN FOOD AND FARM PRODUCTS
8
14054
CONCENTRATION OF ECONOMIC POWER
A similar time trend computation and adjustment has been made with regard
to the item "other expenses." Chart 10 shows the "other expense" items adjusted
for changes in the general price level, as per Table 14. The computation of the
average relationship shows that these cost items tend to amount to $58,600,000
plus $19.21 per weighted ton of product shipped. The deviations from this line
are in some instances fairly large and in general it would appear that "other
expenses" tended to be below average in the earlier years and above average
toward the end of the 1927-1938 period. In Table 22 are shown the deviations
and the percentage deviations from average of the "other expense" items adjusted
for price changes. Chart 11 shows the percentage deviations, plotted by years.
Computation of the general trend shows that they are correlated with the pas.sage
of time in such a way that "other expenses" at 1938 price levels tend to be 5.21 %
below average in 1927, as against 5.26% above average in 1938, representing an
increase over the eleven year period prior to 1938 amounting to 10.47% of average,
or .95% of average per year. The adjustment of the figures for each of the
respective j'ears for this time trend is made in Table 23. In Table 24 the final
adjusted figures are shown in connection with the volume of shipments involved.
In Chart 12 the "other expense" items, adjusted for both price changes and time
trend, are shown in relation to volume. Computation of the line of average
relationship shows that undei 1938 conditions "other expenses" tend to amount
to $21.40 per weighted ton plus $50,900,000.
Table 23.
-Adjustment of Adjusted "Other Expenses" for Time Trend — United
States Steel Corporation and Subsidiaries
(A)
Year
(B)
Millions of
Weighted
Tons of
Products
Shipped
(C)
Average
Adjusted
"Other Es-
penses" •
(D)
Years Prior
to 1938
(E)
Fractional
Increase
(.009521 X
D)
(F)
Adjust-
ment
(CXE)
(O)
Actual
Adjusted
"Other Ex-
penses"
(H)
Final
Adjusted
"Other Ex-
penses"
(Q+F)
1927_.
1928
13.0
14.0
15.1
11.9
8.1
4.4
6.2
6.1
7.6
11.0
13.2
7.8
308.3
327.6
348.7
287.2
214.2
143.1
177.7
175.8
204.6
312^2
208.4
11
10
9
8
7
6
5
4
3
2
1
0
. 104731
.095210
. 085689
. 076168
.066647
. 057128
. 047605
. 038084
. 028563
. 019042
. 009521
.000000
32.3
31.2
29.9
14!3
8.2
8.5
6.7
5.8
5.1
3.0
0.0
312.7
331.4
334.8
247.7
203.7
156.8
176.6
148.9
202.7
297.5
336.3
228.7
345.0
362 6
1929
364.7
1930
1931
218.0
1932- ...
165.0
1933
185.1
1934_.
155.6
1935,.
208.5
1936 .
1937....
339.3
1938
228.7
' For volume involved, as indicated by line of average relationship, Chart 10.
Table 24. — "Other Expenses", Adjusted to 19S3 Price Levels and Operating Con-
ditions, Related to Volume of Business — United States Steel Corporation and
Subsidiaries
Millions of Weighted Tons
of Products Shipped
Other Ex-
penses—
1938 Con-
ditions
Year on
which Esti-
mate Based
Millions of Weighted Tons
of Products Shipped
Other Ex-
ir3r^^on-
ditions
which Esti-
mate Based
4.4
165.0
155.6
185.1
208.5
228.7
218.0
iiiiii
11.0
302.6
269.0
345.0
364! 7
1936
6.1,..- .
11.9
1930
6.2...
13.0
1927
7.6
13.2
1937
7.8
14.0
1928
8.1
15.1
1929
CONCENTRATION OF ECONOMIC POWER
Chart 11
14055
PERCENTAGE DEVIATION FROM AVERAGE -
OTHER EXPENSES ADJUSTED FOR PRICE CHANGES
U. S. STEEL CORPORATION AND SUBSIDIARIES
20
15
z
o
10
rfi
5
o
UJ
0
^
5
10
15
20
25
Chart 12
"OTH
FOR (
500
CO 400
<
§ 300
o
z 200
o
^ 100
0
ER EXPENSES", ADJUSTED FOR CHANGES IN COMMODITY PRICES AND
CHANGED OPERATING CONDITIONS, RELATED TO VOLUME OF BUSINESS
U. S. STEEL CORPORATION AND SUBSIDIARIES
^^-
400 ^
<:
300 2
o
200 z
o
100 ^
1936
19.
19:3
'''
r,','^'
193«,
iT931
.^'
•
.930
_
"!-
ISM
,'''
"1
(
NOTE: BASIS Of
OF DCVWt
) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1
MILLIONS OF WEIGHTED TONS OF ALL TONNAGE PRODUCTS SHIPPED
ADJUSTMtNTS- (11 B LS. INOU Of WHOLESALE PRICES fOR COMMODITIES OTHER THAN FOOO AND f»RM PROCuaS AND
ONS FROM AVERAGE
8
(2) TIME IRtNB
14056 OONOENTRATION OP ECONOMIC POWER
Adding together the fixed and variable elements of the final adjusted com-
ponents of cost gives the following result:
Table 25. — Elements of Total Costs Adjusted to 1938 Conditions* — United States
Steel Corporation and Subsidiaries
Itpm
FLxed
Variable
Interest - - . ...
$8,300,000
7. 700, 000
24,200,000
29,500,000
62,100,000
2,500,000
50,900,000
$0. 00 per ton
Pensions -
0.00 " "
Taxes - . -
1.43 " "
2.37 " '•
29.10 " "
1.16 " "
21.40 " "
$185, 200, 000
• Elements of cost individually adjustee^ to 1938 interest, tax, pension, and wage rates; to 1938 price levels;
and for time trend.
The total fixed and variable costs obtained by this calculation are thus seen to
be substantially the same as the results obtained by adjusting total costs, instead
.of the individual components, for time trend. It would seem that the time trend
adjustment made with respect to total costs would represent the more accurate
figure of the two. First of all, it includes any slight time trends that may exist
with regard to the other components. Secondly, as has already been mentioned
in connection with the discussion of the derivation of the total cost curve, the
payroll figures are not comparable with shipments to the extent that production is
greater or less than shipments. Consequently some of the deviations on which
the payroll time trend was estimated may be the result of inventory fluctuations.
Similarly, since the "other expense'" classification was simply the residual amount
after deducting from total costs the other components of cost, the amount of
this item for each year has been overstated to the same extent that payroll has
been understated, or vice versa, because of inventory fluctuations, and some of the
deviations on which the "other expense" time trend was computed may also hare
been the result of the inventory situation. Since total costs were extracted from
the profit and loss statements and represent the sum of any offsetting errors in
"other expenses" and payroll, they are comparable with shipments and the
deviations cannot be attributed to inventory fluctuations, except to the minute
extent that the situation just described resulted in payroll adjustments being
applied to a small portion of costs which should have been classified as "other
expense," and vice versa.
Assuming, then, that the total costs of $182,100,000 plus $55.73 per weighted
ton of product shipped, developed by adjusting total costs for time trend, repre-
sents the more accurate figure, the approximation of the components of total
costs under 1938 conditions may be revised as follows:
Table 26.
■Elements of Total Costs — Revised 1938 Conditions-
Steel Corporation and Subsidiaries
■United Slates
Interest .-- ..-
Pensions. _-- - .-.
Taxes other than Social Security and Federal Income
Depreciation and Depletion
Payroll
Social Security Taxes
Other Expenses
Total Costs
$8, 3fKl, 000
$0. 00 per ton
7, 7m>, 000
0.00 •' "
24,200,000
1.43 " "
29. 5(»0, 000
2.37 " "
62, 100, n(X)
29.10 " "
2, .W 000
1. 16 " "
47, 800, 000
21.67 " "
$182, 100, 00'
$,-5. 73 per ton
The entire adjustment has been nadc in tli "other expense" figures because
the adjustment does not involve a relativoh- large amount in any event, and
because the "other expense" r'a.^sification n^prescnts a conglomeration of items
and is the least accurate claspiMv^tion anyiow. Chart 13 graphically portrays
the relation of the various eleuicnt? of co.'^t to volume. The differences between
the two estimates of fixed and variaMlo cosi-^^ shown above are so small that the
CONCENTRATION OF ECONOMIC POWER
Chart 13
14057
COMPOSITION OF TOTAL COSTS OF OPERATION
IN RELATION TO VOLUME OF BUSINESS
U. S. STEEL CORPORATION AND SUBSIDIARIES
7
1200
6 7 8 9 10 11 1? 13 14 lb 16 17 18 "mtWEST
killIONs of weighted tons of all
tonnage products shipped
NOTE: 19271938 CXPtRltNCE AOJUSTtD TO 1938 CONDITIONS
graphing of either upon the chart would show the same result. The "other ex-
pense" item has been labeled "goods and services purchased, etc." on the chart
to indicate its general nature.
Segregating the cash from the non-cash items, the cost.pattern of the Corpora-
tion under 1938 conditions is as follows:
Table 27. — Cash and
Non-Cash Costs 1938 Conditions-
Corporation and Subsidiaries
■ United States Steel
Item
Fixed
Variable
$8, 300, 000
7, 700, 000
24,200,000
62, 100, 000
2, 500, 000
47, 800, 000
0.00 " ■'
1.43 " "
Payroll
29.10 " "
1.16 " "
21.67 " "
Total Cash Costs
152,000,000
29, 500, 000
53 36 per ton
Depreciation and Depletion
2 37 " "
Total Costs
$182, 100, 000
$55. 73 per ton
14058 COXOENTRATION OF ECONOMIC POWER
The question frequently arises as to what portion of the total costs of the United
States Steel Corporation represents fixed costs and what portion represents
variable costs. These percentages will differ, of course, with the volume of
shipments, because by definition, fixed costs remain the same regardless of the
quantity produced, while variable costs increase with each additional ton of
product. The percentages of fixed to total costs at various operating rates are
shown in Table 28. The computation and significance of the "Average Equiv-
lent Percentage of Unweighted Rolled and Finished Capacity" shown in the
table are explained in Section VI, "Weighted Tonnages and the Operating Rate."
Table 28. — Percentage of Fixed to Total Costs at Various Rates of Operation — 1938
Conditions — United States Steel Corporation and Subsidiaries
[Total costs = $,'i5.734 per weighted ton of product shipped + $182,100,000. Fixed costs = $182,100,0001
Millions of VV'eighted Tons of Products
Shipped
Average
Equivalent
Percentage of
,Unweighted
Rolled and Fin-
ished Capacity
(See Table 37)
Fixed Costs
(Millions of
Dollars)
Total Costs
(Millions of
Dollars)
Percentage
Fixed to Total
2.45...
10
20
30
40
50
60
70
80
90
100
182.1
182: 1
182.1
182.1
182.1
182. 1 .
182.1
182.1
182.1
318.6
414.5
509.3
605.1
700.4
795. 2
891.0
986.3
1031.6
1177.0
57.2
4.17
43.9
6.87
35.8
7.59 — --.
30.1
9.30
26.0
11.00
22.9
12.72
20.4
14.43 -....
18.5
16.14
16.8
17.85
15.4
V. Relation c Costs and Volume to Prices
Knowing the relation between volume and total cost, it becomes important to
determine the way in which decreased prices fit into the picture. To determine
whether price reductions would pay or would involve prohibitive losses, it is
necessary to compare the increase in volume and the total revenues which might
be anticipated if prices were reduced with the total costs of producing the tonnage
involved. The Corporation's shipments could be increased through price. reduc-
tion, if at all, only by increasing the total amount of steel consumed, since com-
petitive meeting of prices by other producers would prevent increased participa-
tion in the going volume of business. It is important to bear in mind that the
rise to be expected in the total volume of steel consumed as a result of a drop in
prices is not very great in the steel industry. Steel is not sold directly to the
ultimate consumer. It reaches him only as a part of the finished automobile,
refrigerator, typewriter, apartment house, tin can, or safety pin, as the case may
be. In other cases, steel is used only as part of the machinery and equipment
used in making the products which reach the man in the street. No matter how
low the price, steel can be sold only if products which are produced from steel or
by the use of steel are being sold. In the case of products produced from steel,
the cost of the steel is usually so small a fraction of the total cost of the product
that a reduction in steel prices, even if passed on to the ultimate consumer, would
not result in a sufficient decrease in the price of the finished product to cause an
appreciable increase in its sale. As far ag steel for production equipment is con-
cerned, it goes without saying that regardless of the price of steel no one will
invest in productive machinery unless he feels the prospects in this particular
line of business justify such investment.
Analysis of the iniluence of price as a factor affecting steel consumption in the
automobile, railroad and container industries reveals that a decrease in the price
of steel can increase the consumption of steel only t. > a limited extent by promoting
the use of more steel per unit or permitting steel to be substituted for some other
competing material. Any substantial increase in the consumption of steel in
these industries could be brought about only by increasing the consumption of
the finished product or service rendered. Consequently, the price elasticity of
the demand for steel depends primarily upon the price elasticity of demand for
the finished product and the relative cost of steel to the price of the finished product.
The elasticity of demand is measured by the ratio of the relative resulting increase
GONOPNTRATJON OF ECONOMIC POWER 14059
in volume to the relative decrease in price. For example, in the case of automo-
biles, an exhaustive study by Messrs. Roos and von Szeliski * found that the price
elasticity of demand for automobiles was about 1.5, which means that a 1%
reduction in price would increase the number of automobiles sold about 1.5%.
Since the cost of steel in the -form sold by the steel producers is about one-tenth
of the retail price of a representative low-priced automobile, it follows that a
reduction of 10% in the price of steel, even if the saving in cost is passed on to
the ultimate consumer, can efiFect at most only a 1 % reduction in the price of the
delivered automobile. A price reduction of that amount, as has been stated,
could bring about but a 1.5% increase in the number of automobiles sold and in
the amount of steel used in the automobile industry. The increased consumption
arising out of the extent to which steel might be substituted for some other ma-
terial, or the extent to which the use of steel per automobile might be increased
if steel prices were reduced, would probably not increase the elasticity by more
than .1. Taking into account all factors and making a liberal allowance for
possible error, the elasticity of demand for automotive steel is not in excess
of .2 or .3.
The price elasticities for the finished products or services in the container and
railroad industries have not had the benefit of as definite measurement as that
made of the demand for automobiles by Messrs. Roos and von Szeliski, but the
evidence indicates that the demand for these products is considerably less elastic
than the demand for automobiles. Moreover, analyses of all the factors influencing
steel consumption in these industries, even assuming an elasticity of demand for
the finished products and services as high as 2, show definitely that the price
elasticity of the demand for steel in each of the respective industries is considerably
less than 1. A mathematical analysis of the correlation between the amount of
all steel sold and the various factors, including price, which influence the quantity
sold, reveals that a negligible portion of the fluctuations in the quantity sold ^re
attributable to price and that a steel price change, other factors remaining un-
changed, will not result in as great a percentage change in the volume of steel
sold. This confirms the individual analyses of the principal steel consuming
industries already mentioned.
While the above analyses indicate clearly that the elasticity of the demand for
steel is considerably less than 1, it has been assumed for the purposes of this
study, in order to use a figure that is beyond question, that the elasticity of
demand for steel is as high as 1. An elasticity of 1 means that any decrease "in
price will result in a proportional increase in volume which will keep the total
sales in dollars unchanged. This is to say that for small changes in price a given
percentage decrease, such as a 5% reduction in price, will result approximately in
the same percentage increase in volume sold.^
It is evident that unless the elasticity of demand for the product exceeds 1 by
a substantial margin, the theory that price reduction in and by itself would
produce profits through increased volume is utterly fallacious, not only for the
United States Steel Corporation, but for any business or any industry. Since,
with the elasticity of demand equalling 1, the total sales receipts would remain
the same, for the theory to work the total costs of producing the greater volume
would also have to be the same or less. For example, no increased payroll could
be incurred to produce the greater volume. Such a condition could exist only
when all costs were "fixed" or "overhead" and none were "additional" or "vari-
able." Only then would the cost per unit go down, relatively, as fast as the
volume went up. Application of the theory of increased profits through price
reduction could thus only produce loss to the enterprise which adopted it. The
actual amount to be lost "by the United States Steel Corporation through reducing
prices, however, and the amount by which the increase in volume to be e.^cpected
as a result of reducing steel prices falls short of the increase needed to offset the
price reduction, can be estimated only by including in the computation the rela-
tionship which exists between costs and volume.
The sales of the subsidiaries of the United States Steel Corporation to outside
customers, less discounts, returns and allowances, amounted in 1938 to $560,508,-
302.96. This amount is net, after deduction of freight paid on shipments to cus-
tomers,. The transportation and miscellaneous revenues, after excluding the
estimated amount of inter-company profits, amounted to approximately $45,267,-
000. The weighted tonnage of all products shipped amounted to approximately
* C. F. Roos and Victor von Szeliski, "Factors Governing Changes in Domestic Automobile Demand,"
The Dynamics of Automobile Demand. General Motors Corporation, N. Y., 1939.
» For computation of percentage increases in volume which would result from various percentage decreases
in price if elasticity of demand were 1, see Appendix II.
124491— 41— pt. 26 31
14060
CONCENTRATION OF ECONOMIC POWEll
7,800,000 weighted tons. The sales and revenues per weighted ton for 1938,
therefore, are made up as follows:
Table 29. — Sales and Other Operating Revenues 1988 — United States Steel Corpora-
tion and Subsidiaries
Item
Total
(MUlions)
Per Weighted
Tons of Prod-
ucts Shipped
Sales
$560,508
45.267
5.80
Total sales and revenues
$605. 775
$77 66
Sales and revenues for 1938 thus averaged $77.66 per weighted ton of tonnage
products shipped.
It should be borne in mind, however, that any profit or loss estimates based
on these figures suppose 1938 average sales per weighted ton. To the extent that
prices are not always proportional to costs, to the extent that some products sell
for a higher or lower -■■price than others costing the same to produce, the 1938
average sales per weighted ton is afifected by the quantities of each particular
product sold, as well as by the prices prevailing. Hence the profit or loss esti-
mates contained in the summary are accurate for 1938 average sales per weighted
ton, but different products might be sold in another year which would constitute
the same weighted tonnage when weighted on a cost basis, but which would carry
diflFerent profit margins and thus result in a slightly diff'erent average sales per
weighted ton figure, although the actual 1938 prices for each product had not
changed. The possibihty that an over-all price reduction in 1938 would have
materially changed the proportion of sales constituted by each type of product
would, however, seem rather remote.
Furthermore, in computing the total sales and revenue at various shipment
levels, it has been assumed that the revenues from transportation and miscella-
neous operations rise and fall approximately as product shipments go up and
down. That this has been the general relation of transportation and miscellane-
ous revenues to the volume of steel and other manufacturing business done may
be seen from Chart 14. The data on vhich this chart is based are given in Table
30. The transportation and miscellaneous revenues, after deducting estimated
inter-company items, amounted in 1938 to $5.80 per weighted ton and it is assumed
for the purposes of this study that they would increase sufficiently so as stiU to
amount to $5.80 per ton if the volume of products sold were increased. It should
be noted, however, that the 1938 transportation and miscellaneous revenues were
higher than average in relation to the volume of products shipped and that in
recent years there has been a tendency for these revenues to fall off somewhat
per ton of product shipped when the higher levels of production are reached, as
is indicated by the position of the points for 1930, 1936 and 1937. Hence it is
rather likely that at the higher operating rates the total sales and revenues
wotild be silghtly less than estimated in this study. To the extent that trans-
portation and miscellaneous operating revenues might fail to increase proportion-
ately with increases in shipments, the losses entaUed through pride reduction
woiUd be greater than those described.
OONOENTRATION OP ECONOMIC POWER
Chart 14
14061
TRANSPORTATION AND MISCELLANEOUS REVENUES RELATED TO SHIPMENTS
U. S. STCEL CORPORAHON AND SUBSIDIARIES
90
</> 80
f:
S 50
i ^
5 30
^ 20
10
0
>1929
y
90
80 w
70 5
60 %
50 ^
40 1
30 -
20 ^
10
0
U27
:?
y
y
y
y
/
1936
1930
•l937
1938,
193S
y
1934
y
^
1931
1S32
y
w
y
/
^
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
MILLIONS OF WEIGHTED TONS OF ALL TONNAGE PRODUCTS SHIPPED
HOTt: m. IMTER-eOWWY fIDIS EXCtUOED
Table 30.-
■Trans'poTtation and Miscellaneous Revenues ' Related to Shipments —
United States Steel CorporaticM and Subsidiaries
Millions of Weighted Tons of Products Shipped
Year
Net Sales and
Revenues
(Millions of
Dollars)
Net Sales
(Millions of
Dollars)
Transporta-
tion and Mis-
cellaneous
(MUlions of
DoUars)
4.4
1932
1934
1933
1935
1938
1931
1936
1930
1927
1937
1928
1929
283.5
418.5
371.0
539.9
605.8
523.7
791.7
814.4
946.6
1021. 6
992.6
1067. 8
261.7
388.5
345.3
500.8
560.5
488.7
738.5
760.8
861.4
C64.4
903.9
972.3
21.8
6.1
6.2
30.0
25.7
7.6
39.1
53.2
63.6
85.2
13.2
57.2
88.7
15.1
95.5
' Estimated amount of inter-company items excluded.
14062
CONOENTRATION OF ECONOMIC POWER
On the other hand, the greatest extent to which it is likely that fluctuations in
the revenues from transportation and miscellaneous operations per ton of products
sold will change the picture presented may be seen from the fact that the trans-
portation and miscellaneous revenues have averaged only 7.5% of total sales and
revenues over, the period 1927-1938 and that the transportation and miscel-
laneous revenues per ton of product sold have never in that period amounted to
more than 75 cents above or $1.65 below the 1938 figure. The figures for
each year, together with the deviations from the 1938 figure, are given in Table 31.
The effect of such possible deviations on the profit computations that have been
based on these sales and revenue figures is further minimized by the fact that the
cost of these operations would also be likely to fall at least partly as much as the
revenues derived therefrom.
Taking $77.66, then as the total sales and revenue per weighted ton at average
1938 price levels, the practical effect of the comparative inelasticity of demand
for steel may be examined.
Table 31. — Transportation and Miscellaneous Revenues Per Weighted Ton of
Products Shipped — United States Steel Corporation and Subsidiaries
Year
Millions of
Weighted
Tons of
Products
Shipped
Transporta-
tion and
Miscellaneous
Revenues
( Millions of
Dollars)
Per
Weighted
Ton
Deviation
from 1938
1027
13.0
14.0
15.1
11.9
8.1
4.4
6.2
6.1
7.6
11.0
13.2
7.8
85.2
88.7
95.5
53.6
35.0
21.8
25.7
30.0
39.1
53.2
57.2
45.3
$6.55
6.34
6.32
4.50
4.32
4.95
4.15
4.92
6.14
4.84
4.33
6.80
1928
1929.
+0.62
1930
-1.30
1931.
-L48
1932
— 0 86
1933
-1.65
1934
-0.88
1935
-0.66
1936
—0 96
1937
— 1 47
1938
—0.00
In 1937 approximately 13,200,000 weighted tons of products were shipped by
the Corporation's subsidiaries. In 1938 this tonnage dropped to 7,800,000,
a decrease of 5,400,000. To bring the 1938 volume up to the 1937 level a 69.23%
increase would have been needed. Since steel has at best an elasticity of 1, it
would have been necessary to drop the price at least to a point where 13,200,000
tons would have brought in the same amount of dollars as the 7,800,000 tons did
at 1938 prices. Since the- total sales, as distinguished from other revenues,
amounted to $560,508,000 in 1938, the $71.86 sales per weighted ton would have to
have been reduced to $560,508,000 divided by 13,200,000, or $42.46 per weighted
ton. This represents a price decrease of 40.9%. Disregarding the possibility
that other operations might not expand with increased shipments and adding
in the full $5.80 per weighted ton realized from other operations under actual
1938 conditions, the total sales and operating revenues would then have amounted
to $48.26 per weighted ton. Since the variable cash costs amount to $53.36 per
weighted ton, the Corporation would have sustained a cash loss of $5.10 on every
weighted ton sold in addition to failing to recover any part of the fixed cash costs
of $152,600,000. Assuming that the price reduction would have been successful
in restoring the 1937 volume, 13,200,000 weighted tons would have been sold.
At a loss of $5.10 a ton, the variable cash costs that would not have been covered
by the sales price would amount to $67,320,000. Adding in the $152,600,000
fixed costs not covered by the sales price, the Corporation would have had a
cash loss for the year of $219,920,000. This is without making any provision for
the depreciation and depletion of the fixed assets of the Corporation and its sub-
sidiaries. The average working capital of the Corporation for the year 1938
amounted to $397,241,615." Hence, the drain of such a cash loss upon the Cor-
poration would have exhausted its working capital in less than two years. If
the amount of the depreciation and depiction of assets at this volume of operations,
amounting to $60,784,000, is added to the cash loss, there is a total loss of
$280,704,000. Annual losses at this rate would wipe out the combined equity
of the preferred and common stockholders as of December 31, 1938, in about
four and a half years.
»For computation of average working capital, see Appendix I.
CONCENTRATION OP ECONOMIC POWER 14063
Probably such drastic decreases, however, are not contemplated by those who
criticize the steel industry for failing to reduce prices. While a smaller decrease,
in price could not have been expected under any circumstances to raise the 1938
volume to the 1937 level, it may nevertheless be contended that some price
reduction and some resulting stimulation in volume were desirable. The facts
are that any further price reduction in 1938 would have served but to increase
the loss sustained. Chart 15 shows the actual deficit in 1938, amounting to
$8,758,572 after the deduction of bond interest but before the Federal income and
profits taxes and exclusive of non-operating income and expense. Added thereto
are the amounts of additional deficit that would have been incurred if various
percentage reductions in price had been made, assuming that steel has an elasticity
of demand of 1. For instance, even a 1 % reduction in price would have increased
the 1938 loss by well over $3,000,000, while a 10% reduction in price would have
resulted in an additional loss, after allowing for the maximum probable increase
in volume, of over $43,000,000. To the extent that the assumed increases in
volume might have failed to result from the decreases in price, of course, the
additional losses would have been greater. The figures on which the chart is
■ based,' together with the additions to the 1938 deficit which would have taken
place if prices had been reduced and no increase in volume resulted, are shown in
Table 32.
Table 32.— Estimated Additions to 1938 Deficit— How Deficit Would Have In-
creased if Average 1938 Prices Had Been Reduced Various Percentages — United
States Steel Corporation and Subsidiaries
Percentage
Reduction
in Price
Estimated Ad-
ditional Loss,
Assuming ElaS'
ticity of De-
mand for Steel
ofl
$3,900,000
7,900,000
12, 000, 000
16, 200, 000
20, 600, 000
24,900,000
29, 300, 000
33,900,000
38, 500, 000
43,300,000
Estimated Ad-
ditional Loss, if
No Increase in
Volume Re-
sulted from
Price Reduc-
tion
$5, 600,
11, 200,
16, 800,
22, 400,
28,000,
33, 600,
39,200,
44,800,
50,400,
66, 100,
Percentage
Reduction
in Price
Estimated Ad-
ditional Loss,
Assuming Elas
ticity of De-
mand for Steel
ofl
97,
100, 000
100,000
200,000
400,000
700,000
200,000
800,000
500,000
400,000
400,000
Estimated Ad-
tional Loss, if
No Increase in
Volume Re-
sulted from
Price Reduc-
tion
$61, 700, 000
67, 300, 000
72,900,000
78, 500, 000
84, 100, 000
89, 700, 000
95, 300, 000
100, 900, 000
106, 600, 000
112, 100, 000
The relationship between steel prices, volume and costs for the United States
Steel Corporation and its subsidiaries is such that the probable increase in volume
which would result from a price decrease is never as great as the increase which
would be required to compensate for the reduced amount received per ton. For
instance, the relationship between the average 1938 prices and costs is illustrated
in Chart 16. This chart reveals that the total sales and revenues at 1938 prices
would cover all of the costs only if production amounted to 8.31 milHons of weighted
tons of product or more. The dashed sales line shows the total sales and revenue
that would be realized if the prices at which the products were sold were reduced
10% without making any red'uction in the rates charged in the transportation
and miscellaneous operations. It should be noted that with this reduction in
price, production would have to reach 12.36 millions of weighted tons to break
even.8 This represents an increase in volume of 48.8% required to compensate
for the decreased prices. The same relationship holds not only with regard to
the break-even point, but also to the netting of any particular amount of profit
or loss. For instance, if production amounted to 6,000,000 weighted tons, the
loss at 1938 average prices would be about $50,500,000. If 1938 prices were
decreased 10%, volume would have to be raised 48.8%, or to about 8,925,000
weighted tons, in order not to increase the loss.
In Table 33 are shown and on Chart 17 are graphed the percentage increases in
volume that would be required to compensate for various percentage decreases
from the 1938 average prices as compared with the maximum probable increase in
volume that would result from a price decrease.' This probable increase i ';ased
on the assumption that the elasticity of the demand for steel is 1. As hac been
mentioned above, aU the evidence indicates that this is a most optimistic estimate
and that the real figure is considerably smaller.
' For method of computation, see Appendix in.
» See Appendix IV for computation of break-even points.
» See Appendix V for method of computation.
14064
OONOENTRATION OF ECONOMIC POWER
Chart 15
ESTIMATED ADDITIONS TO 1938 DEFICIT
HOW DEFICIT WOULD HAVE INCREASED IF PRICES HAD BEEN REDUCED
AND VOLUME HAD INCREASED TO SAME RELATIVE EXTENT
U. S. STEEL CORPORATION AND SUBSIDIARIES
%
672
2 3 4 5 6 7 8 9 10 n 12 13 14 15 16 17 18
PERCENTAGE REDUCTION IN 1938 AVERAGE PRICE
OONOENTRATION OF ECONOMIC POWER
14065
Chart 16
RELATIONSHIP BETWEEN SALES AND COSTS
EFFECT OF REDUCTION FROM AVERAGE 1938 PRICES
U. S. STEEL CORPORATION AND SUBSIDIARIES
1200
1 1 1 1 1 1 1
-
^
2
r
V
1
1200
1100
c=i SIM£S AND REVENUES
<1938 PRICES)
c«=» SALES AND REVENUES
11938 PRICtS USS 10»)
1100
1000
900
800
700
600
500
400
300
200
100
0
1000
900
800
700
600
/
F
1
1
'^
1
B
EAK
EVi
i"
^
"
r
1
t
V::
is
1
is
1
1
•5«
m
m
0r
11
\
\
^
%
/
11
~
liW
i
'■■.
i
1
if-
iii'
■Si
1
/
<|
■S
TO
AL
m
1
"
'f^
400
/
/
i
1
M
M 0i:
:;;:;
'M
1
300
200
100
0
/
/(
%
1
;i
1
SI
M
IBM
1
■:-SI
i
1
4
'/
;■■■
B:- M
M'
1
1
1
/
r
';
...
., „
If
2^
i
-:
m
0 1 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 17 18
MILUONS OF WEIGHTED TONS OF ALL
TONNAGE PRODUCTS SHIPPED
NOTE: COSTS ARE BASED ON 19271938 DPERIENCE. ADJUSTED TO 1938 CONDITIONS
14066
CX)NCENTRATION OF ECONOMIC POWER
Chart 17
INCREASES IN VOLUME NEEDED TO COMPENSATE FOR
VARIOUS DECREASES IN 1938 PRICES
COMPARED TO PROBABLE RESULTING INCREASES IN VOLUME
U. S. STEEL CORPORATION AND SUBSIDIARIES
240
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
PERCENTAGE DECREASES FROM AVERAGE 1938 PRICES
NOTt: PROSABU RESULTING INCREASES IN VOLUME EfSED IM ASSUMPTION THAT ELASTIQTY OF DEW^NO EQUALS 1
TaSle 33. — Increases in Volume Needed to Compensate for Various Decreases in
Average 1938 Prices — Compared to Probable Resulting Increases in Volume —
United States Steel Corporation and Subsidiaries
Percentage Reduc-
tion in Price
Percentage
Increase in
Voltune
Needed
Probable
Percentage
Increase,
Assuming
Elasticity of 1
Percentage Reduc-
tion in Price
Percentage
Increase m
Volume
Needed
Probable
Percentage
Increase,
Assuming
Elasticity ofl
3.4
7.0
10.9
15,1
19.6
24.5
29.8
35.5
41.8
1.0
2.0
3.1
4.2
5.3
6.4
7.5
8.7
9.9
10
48.8
56.-4
64.8
74.2
84.8
96.7
190.3
453.6
5858.2
11.1
12.4
12 .
13.6
14.0
16.3
15
17.7
20
25.0
25
33.3
42.9
Since the average 1938 prices represent the average results of prices in effect
before and after the June 24, 1938 price reduction, a similar analysis to that made
above has been made with the total sales and revenue line reduced to what it
would have been if the prices prevailing in the second half of 1938, after the price
reduction, had prevailed the entire year.
OONCENTRATION OF ECONOMIC POWER 14067
The total sales and revenue line at the prices prevailing in the second half of
1938 was computed by adjusting the sales component of the total sales and
revenue per weighted ton computed above, on the basis of the selling value of
rolled and finished steel products sold during the second half of 1938, as compared
to the average selling value of these products for the entire year. The f. o. b.
selling value of rolled and finished steel products sold each month was obtained
from the Comptroller's Department of the Corporation. The total selling value
of rolled and finished steel products for the last six months of the year, divided by
the weighted tonnage of such products shipped during that period, gives the
average amount per weighted ton received after the June 24, 1938 price reduction.
The selling value for the entire year, divided by the weighted tonnage shipped
during the entire year, furnishes the average amount received per weighted ton
of rolled and finished steel products for the year as a whole. The $71.86 referred
to above as the sales per weighted ton of tonnage product^ shipped represents the
amount received for all products sold per weighted ton of all tonnage products
shipped. Multiplying this figure by the ratio of the average steel price computed
above for the last half of 1938 to the average price for 1938 as a whole, gives
an estimate of the total sales per weighted ton of products if the same propor-
tionate reductions in the prices of rolled and finished steel products that were in
effect in the second half of 1938 had been applied to all products sold and had been
put into effect at the beginning of the year. The computation, made in Table
34, shows that the sales per weighted ton would then be $67.33. Adding to this
the $5.80 per weighted ton arising from miscellaneous transportation operations,
the estimated total annual sales and revenues per weighted ton at prices prevailing
after the June 24, 1938 price reduction would be $73.13.
Table 34. — Estimate of Annual Sales and Revenue at Prices Prevailing in Second
Half of 1938 — United States Steel Corporation and Subsidiaries
Period
F. 0. B. Selling
Value of Rolled
and Finished
Steel Products
Shipped "
Weighted Tons
of Rolled and
Finished Steel
Products
Shipped I
SeUing Value
Per Weighted
Ton
$230, 750, 804
246,835,450
3,087,392
3,798,057
$74. 74
July to Dec , 1938
64 99
Year 1938
$477. 586. 254
6,885,449
$69 36
Item
1938 Average
Per Weighted
Ton
Adjustment
Second Half
1938 Average
Per Weighted
Ton
Sales
$71.86
5.80
64.99/69.36
Total Rftlps ftnd RflVftniifif!
$77. 66
$73 13
1 Before yearly adjustment lor returns and allowances, amounting to $4,732,196 and 3,504 unweighted tons.
Table 35. — Increases in Volume Needed to Compensate for Various Decreases in
2nd Half, 1938, Prices Compared to Probable Resulting Increases in Volume —
United States Steel Corporation and Subsidiaries
Percentage Reduc-
tion in Price
Percentage
Increase in
Volume
Needed
Probable
Percentage
Increase,
Assuming
Elasticity of 1
Percentage Reduc-
tion in Price
Percentage
Increase in
Volume
Needed
Probable
Percentae
Increase
Assuming
Elasticity of 1
4.0
8.4
13.1
18.3
24.0
30.2
37.2
44.8
53.5
1.0
2.0
3.1
4.2
6.3
6.4
7.5
8.7
9.9
10
63.1
74.1
86.7
101.3
118.3
138.4
342,6
2984.4
11
12 4
12
13 6
15
17 7
20
25 0
9
14068
aONOENTRATION OF ECONOMIC POWER
Chart 18 shows the price, cost, and volume relationship under prices prevailing
in the second half of 1938. Here again the revenues at higher volumes have been
overstated to the extent that the transportation and miscellaneous revenues
might drop below $5.80 per weighted ton if shipments were increased. Never-
theless it may be noted that the break-even point at these prices is 10.47 millions
of weighted tons, as compared with 8.31 millions of weighted tons at the average
1938 prices. Similarly, a 10% decrease in the second half of 1938 prices would
be offset only if volume were increased 63.1% to 17.1 millions of weighted tons.
This would mean that the Corporation would have to operate at an average of
over 90% of capacity for the entire year in order to break even. Table 35 and
Chart 19 show the percentage increase in volume required to compensate for
various decreases in the second half of 1938 prices 'o as compared to the increase
in volume that would result from the drop in prices if the elasticity of demand for
steel were as high as 1."
In connection with the break-even point, it has been contended that in a truly
competitive arrangement the break-even point will be only a little bit short of
capacity. It has already been shown that if the prices prevailing in the second
half of 1938 were reduced by 10%, the break-even point for the Corporation and
its subsidiaries would be moved up to over 90% capacity. The effect of this high
break-even point on the possible return on investment that might be realized by
the Corporation may be seen by calculating the rate of operation that would have
to be maintained to realize an average return as modest as 5% on the entire
tangible investment. The average investment of the Steel Corporation, consist-
ing of the combined interests of stockholders and bondholders, after eliminating
all intangible values, for the year 1938 amounted to $1,586,523,686. A 5%
return on this investment would require a profit before bond interest of over
$79,300,000. Deducting bond interest of $8,300,000, this leaves a profit after
bond interest of over $71,000,000 to be obtained. Even if Federal income taxes
are disregarded, to realize such a profit at prices 10% below second half of 1938
prices, which would be low enough to put the break-even point at around 90% of
capacity, the Corporation would have to operate throughout the year at an aver-
age rate of over 130% of capacity, an obvious impossibility. '^
VI. Weighted Tonnages and the Operating Rate
The operating rate for the subsidiaries of the United States Steel Corporation,
as ordinarily computed and as published in the annual reports, is obtained by
dividing the total number of tons of rolled and finished steel products produced
by the estimated capacity in tons for producing rolled and finished steel products.
This figure has its limitations as an indicator of steel producing activities. For
instance, the production of a million tons of sheets, with all the processing that
they require, would involve considerably more activity than the production of
the same quantity of rails, which are more simple to produce, yet either would
result in the same operating rate.
>» See Appendix V for method of computation.
" While the assumption of any elasticity of demand for steel greater than 1 is highly unrealistic, it is
interesting to note that even if steel had an elasticity of 1.5 or 2, the percentage increase in volume needed to
offset a price reduction would still greatly exceed the percentage increase in volume which would then
result from reducing prices:
Decrease in Price
1%-
2%-
3%.
4%.
5%
10%
15%
30%
Resulting Increase in Volume
1.5%
3.1%
4.7%
6.3%
i?:?l
'27.6%
39.8%
54.0%
70.8%
If Elasticity =2
2.0%
4.1%
6.3%
8.5%
11. 1%
23. 5%
77. f
104.1
Increase in Volume Needed to
Offset:
Decrease in
Average 1938
Steel Prices
3.4%
7.0%
10.9%
15.1%
19, 6%
48.8%
96.7%
190. 3%
453.6%
5858.2%
Decrease in
Second Half
1938 Steel Prices
8.4%
13.1%
18.3%
24. 0%
63.1%
138.4%
342. 6%
2984.4%
(No increase
sufficient)
' 'See Appendix VI for- computations involved.
OONCENTRATION OF ECONOMIC POWER
Chart 18
14069
RELATIONSHIP BETWEEN SALES AND COSTS
EFFECT OF REDUCTION FROM 2nd HALF 1938 PRICES
U. S. STEEL CORPORATION AND SUBSIDIARIES
1100 -
1000 -
900 -
CO
^ 800 -
-J
o 700-
0 600-
1 500-
i ^-
300-
200?
100
0
I 1 1 1 1 1 1
//
1
° 8 8 8 i § 8 8 8 i 1 i ^
MILLIONS OF DOLLARS
( 1 SALES AND REVENUES
• Jno H«.f 19M PRICtS.
c»=. SALES AND REVENUES '
/
->
-
■^
y
n
-^■:>
1
^^^';M'-''
\|
//
f^
v/
Y
4
yf
■
/
i ^
1 i
it
/
7
h 1
ill!
/
/
^^
f
r
TO
TAL
COSTSJ .
/
I
m
■:.
'%
%
lift
i
M
i!
1;
1
i
i'y. 'M
i
1
%
1
1
11
II
m
M
■':-^^
is
Si
•
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
MILUONS OF WEIGHTED TONS OF AU
TONNAGE PRODUCTS SHIPPED
NOTE. COSTS ARE BASED ON 19271938 EXPERIENCE. ADJUSTED TO 1938 CONOmONS
Table 36. — Weighted Tonnages of All Tonnage Products and Unweighted Tonnages
of Rolled and Finished Products Shipped — United States Steel Corporation and
Subsidiaries
Year
Weighted Tois
All Tonnage
Products
Shipped
Actual Tons
Rolled and Fin-
ished Products
Shipped
Year
Weighted Tons
All Tonnage
Products
Shipped
Actual Tons
Rolled and Fin
ished Products
Shipped
1927
13,008,520
13,994,239
15,088,968
11,934,595
8,130,577
4,352.016
12,993,283
13.973,129
15, 234, 355
11, 624, 294
7, 676, 744
3,974,062
1933
6, 160, 338
6,096,937
7,631,783
11,012,458
13, 186. 548
7,758,891
1928
1934
5 911 760
1929
1935
. 71356! 185
10 784 716
1930.
1936
1931
1937
12.789.841
6,659,253
1932
1938
In making this study of the relation of cost to volume it was apparent that the
cost of producing a million tons of rails, for instance, would be lower than the cost
of producing a million tons of sheets. For this reason, as has already been men-
tioned, the tonnages of each type of product shipped were weighted on the basis
of the average miU cost for that type of product. Both the subdivisions of rolled
and finished steel products and the main classifications of tonnage products other
than steel were weighted by the ratio of their average mill cost to the average
14070
OONCJ-ENTRATION OF ECONOMIC POWER
Chart 19
INCREASES IN VOLUME NEEDED TO COMPENSATE FOR
VARIOUS DECREASES IN 2nd HALF 1938 PRICES
COMPARED TO PROBABLE RESULTING INCREASES IN VOLUME
U. S. STEEL CORPORATION ANO SUBSIDIARIES
440
400
360
ui
§ 320
J 280
^ 240
q:
o
5 200
o
S 120
40
].
n
-
1
'
'
„ IN HLfi
IN VOL
SES
UME
_nnO 1-
.mmM
nl [•
t±x:::t:i::..l..
PRODABU
. RESULTING
INCREASES
IN VOLUME
1 2 3 4 5 6 7 8 9 10 U 12 13 14 15 16 17 18 19 20 21
PERCENTAGE DECREASES IN PRICES
NOTE: Pm>8ABL£ RESULTING INCREASES IN VOLUME BASED IN ASSUMPTION THAT ELASTICITY Of DEMAND EQUALS I
mill cost of all rolled and finished products. In this way the actual tonnages
shipped each year were converted to equivalent tons of average cost rolled and
finished steel products. Accordingly, the break-even point at various price levels
has been stated in terms of weighted tonnages anH it is essential to keep in mind
the relationship between these weighted tonnage Lgures and the regular operating
rate figures if the significance of the weighted tonnage figures is to be properly
appreciated.
Table 36 shows the unweighted tonnages of rolled and finished steel products
shipped each year, together with the corresponding weighted tonnages of all
tonnage products. The weighted and unweighted tonnages are plotted in rela-
tion to each other on Chart 20. Inspection of the chart shows that the points
for each year fall very nearly in a straight line but that as the unweighted tonnages
decrease, the weighted tonnages do not decrease quite so rapidly, indicating that
in slack years the sales of heavy, low-cost steel products fall ofif relatively more
than the sales of the lighter, high-cost items and of the products;;other than steel.
Computation of the line of average relationship by the least squares method
shows that the weighted tonnage is generally 95.36% of the unweighted tonnage
plus 742,000 tons. This would indicate that if the shipments of rolled and finished
steel products reached the January 1st, 1939 capacity of approximately 17,940,000
tons, the corresponding weighted tonnage snipped would amount to about
17,850,000 weighted tons. On the basis of the average relationship existing be-
tween weighted and unweighted tonnages at various shipment levels, it is possible
to compute the weighted tonnages and the percent of capacity operated based on
weighted tonnages, which would be equivalent to various operating rates as ordi-
narily computed. The results of this computation are tabulated in Table 37
OONOENTRATION OP ECONOMIC POWER
Chart 20
14071
REL/
SH
18
g 14
g 12
|.«
1 '
2 6
i *
2
0
ITION OF ACTUAL TONS OF ROLLED AND FINISHED STEEL PRODUCTS
IPPED AND WEIGHTED TONS OF ALL TONNAGE PRODUCTS SHIPPED
U. S. STEEL CORPORATION AND SUBSIDIARIES
/-
16
14 §
12 £
8 ^
6 Q
2 ^
0
1928
,x'''
iT929
"fr-
1937. /
-V
.>*
1931
y,^
""«f
Z'
f^a.
"?.
^/
X
<^
-^
CAPACirr. JAN.
__ 1 1 _
1.193
1
C
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
ACTUAL TONS - ROLLED AND FINISHED PRODUCTS
Table 37.-
- Average Weighted Equivalent of Unweighted Tonnages — United State
Steel Corporation and Subsidiaries
Actual Tons
Rolled and
Finished
Products
(Millions of
Tons)
Percent of
1/1/39
Capacity
Average
Equivalent
Weighted
Tonnage of All
Products
(Millions of
Tons)
Percent of
Estimated
1/1/39 Weighted
Capacity
.90
5
1.60
9.0
1.79
10
2.45
13.7
2.69
15
3.31
18.5
3.59
20
4.17
23.4
4.49
25
5.02
28.1
5.38
30
6.87 .
32.9
6.28
35
6.73
37.7
7.18
40
7.59
42.5
8.07
45
8.44
47.3
8.97
50
9.30
52.1
9.87
55
10.15
56.9
10.76
60
11.00
61.6
11.60
65
11.86
66.4
12.56
70
12.72
71.?
13.46
75
13.58
76.1
14.35
80
14.43
80.9
15.25
85
15.28
85.6
16.15
90
16.14
90.4
17.04
95
16.99
95.2
17.94
100
17.85
100.0
It should be noted, however, that all of the points on Chart 20 do not fall
exactly on the average line. These variations from year to year constitute one
reason for using weighted rather than unweighted tonnages in making this study.
To the extent that the relationship between weighted and unweighted tonnages
in ar\y year is different from the average relationship for the volume involved,
the equivalent weighted tonnage figures given in Table 37 will be inaccurate.
Since this variance in relationship betweei weighted and unweighted tonnages
arises from differing proportions of each type of product constituting the total
sales, the amount of the variance becomes more limited as operations approach
14072 OONOENTRATION OF ECONOMIC POWER
capacity, since at capacity the proportions are fixed by the capacity for producing
each type of product.
Table 38. — Ratio of Tons of Rolled and Finished Products to Weighted Tons of
All Tonnage Products — United States Steel Corporation and Subsidiaries
Year
Unweighted
Tons Rolled
and Finished
Products
Shipped
Weighted Tons
All Tonnage
Products
Ratio
Unweighted
to Weighted
1927
12.99
13.97
15.23
3!9V
5.81
5.91
7.36
10.78
12.79
6.66
13.01
13.99
15.09
11.93
8.13
4.35
6,16
6.10
7.63
11.01
13.19
7.76
0 9985
1928
0 9986
1929
1 0093
1930
0 9740
1931
0 9446
1932
0 9126
0 9432
1934
0.9689
1935
0.9646
1936
0. 9791
1937
0.9697
1938
0.8582
Since the variations from average do not show up very clearly in Chart 20,
the percentage of weighted tons of all tonnage products shipped represented by
unweighted tons of rolled and finished products is computed in Table 38. The
resulting data are graphed on Chart 21. Here again it becomes apparent that
the actual tonnages tend to be smaller than the weighted tonnages when operating
rates are low. Deviations of the individual points from the line of average rela-
tionship, however, are considerable. In 1938, for instance, the ratio of the
unweighted tonnages shipped to the weighted tonnages was even smaller than
would be expected on the basis of the average relationship on which Table 37 is
based. Thus the 1938 shipments of rolled and finished steel products, amounting
to 6.66 millions of unweighted tons or 37.1% of January 1st, 1939, capacity,
would normally be expected to equal 7.09 miUions of weighted tons of all tonnage
products, or 39.7% of weighted capacity. As a matter of fact, however, the
Chart 21
RATIO OF TONS OF ROLLED AND FINISHED PRODUCTS TO WEIGHTED TONS
OF ALL TONNAGE PRODUCTS AT DIFFERENT ROLLED AND FINISHED
SHIPMENT LEVELS
U. S. STEEL CORPORATION AND SUBSIDIARIES
5
1928
^
-
-•
1935
^
.936
1930
1927:
1937
,^
19^
-
>»l^
--1
31
-^
1938
'
^
1 2 3 4 5 6 7 -8 9 10 11 12 13 14 15 16 17 18
UNWEIGHTED TONS OF ROLLED AND RNISHED PRODUCTS SHIPPED
CONCENTRATION OF ECONOMIC POWER
14073
weighted tonnage in 1938 amounted to 7.76 millions of weighted tons, or 43.4%
of weighted capacity. This indicates that no percentage of capacity as ordinarily
computed can be named as the break-even point, since the exact percentage at
which the Corporation would break even will depend upon the type of products
composing the total products sold, and these vary considerably from year to
year, even when total shipments' are approximately the same.
Some of the year to year variations in the relation between the weighted ton-
nages of all tonnage products and the unweighted tonnages of rolled and finished
products shipped are accounted for by variations in the percentage of light and
heavy steel products shipped, and others by the amount of products other than
steel which are sold. In order that the causes of the variations may be observed,
the percentage relationships between the unweighted tonnages of each class of
products shipped and the total tonnage of rolled and finished steel products
shipped are given in Table 39. Notice should be taken of the fact, however, that
the apparent increase in the tonnage of sundry materials and by-products sold in
1938 is partly the result of including in sales, cost of goods sold, and sundry
materials and by-products, shipped, the sales of crushed slag. In prior years the
net proceeds from the sale of this item were included in miscellaneous income.
The total tonnage of sundry materials and by-products in 1938 would amount to
6.3% of the tonnage of rolled and finished steel products shipped, if computed
on the same basis as in prior j^ears.
Table 39. — Percentage of Tons of Various Classes of Products Shipped to Total
Tons of Rolled and Finished Products Shipped— 1927-1938— United States Steel
Corporation and Subsidiaries
Class of Product
Semi-Finished
Rails. ._
Plates
Heavy Structural
Shapes
Merchant Bar, H. R.
Strip, Hoops,-etc.-..
Sheets, Black and Tin
Plate
Cold Rolled Strip, Wire
and Wire Products-
Copper Products and
Insulated Wires and
Cables
Tubular Products
Fabricated Structural
Shapes
Angle Bars, Tie Plates
and All Other Rail
Joints
Total 1 to 10
-\xles
Wheels.
Nuts, Rivets, Spikes,
Bolts :
Foundry Products
Special Track Work. . .
Sundry Iron and Steel
Products
Totaintoie
All Rolled and Fin-
ished Steel
Pig Iron, Ingots, Ferro
and Scrap
Limestone, Coal, Coke
and Iron Ore
Sundry Materials and
By-Products
Total Tons of All
Products Sold on
Tonnage Basis
Percentage of Total Tons of Rolled and Finished Products Shipped
1927 1928 1929
0.3
0.6
0.5
. 0.4
0.2
0.7
2.7
100.0
1.8
100.0
2.5
100.0
2.8
38.4
3.6
1931 1932 1933
135.4 il45.5 144.8 ,133.5 134.0 137,6 144.9 145.2 139.3 141.2
1936 1937 1938
10.9
.5.1
6.8
7.7
17.3
22.0
9.6
43.7
15.2
If on, same basis as other years, would be 6.3.
14074
QONCENTRATION OF ECONOMIC POWER
In order that the efifect of variations in the proportionate amounts of each
type of rolled and finished products, as distinguished from variations in the
proportionate amount of products other than rolled and finished steel products,
may be further shown, Table 40 gives the unweighted tonnages of rolled and
finished products shipped each year, together with the weighted tonnages of
rolled- and finished products only, exclusive of other tonnage products. The
graphing of these data on Chart 22 shows clearly that the more costly light prod-
ucts constitute a higher proportion of sales in slack periods, since the percentage
of weighted tons of rolled and finished products represented by unweighted tons
falls off considerably as the operating rate goes down. It is again noticeable that
there is also considerable variance in the relationship between weighted and un-
weighted tonnages of rolled and finished products shipped in years in which the
unweighted tonnages were approximately similar. For example, the chart shows
that in the year 1934 the actual tonnage was much nearer to the amount of the
weighted tonnage than would be expected at such a low shipment level. The
explanation may be obtained by an inspection of Table 39, which shows that in
1934 rails, plates and other heavy, lower cost products constituted almost as great
a percentage of the total shipments as in years when the operating rate v/as much
higher.
Table 40. — Ratio of Unweighted to Weighted Tons of Rolled and Finished Products —
United States Steel Corporation and Subsidiaries
Year
Actual Tons
Rolled and
Finished
Products
Weighted
Tons Rolled
and Finished
Products
Ratio Un-
weighted to
Weighted
1927
12.99
13.97
15.23
11.62
7.68
5.97
5.81
5.91
7.36
10.79
12.79
6.66
12.75
13.62
14.66
11.51
7.86
4.23
6.97
5.89
7.36
10.47
12.49
6.89
1. 0187
1928
1.0258
1929
1.0395
1930
1.0103
1931.
'932 :
0.9766
0. 9397
li,33
0. 9732
1934
1.0039
1935
0.9996
1936
1.0297
1937
1.0243
1938
0.9672
All in all, then, it may be said that the weighted tonnage figures arrived at in
this study are a fairly accurate indication of the true break-even point. There is
not a sufficiently constant relationship between weighted and unweighted ton-
nages, however, to make possible an accurate conversion of the break-even point,
expressed in weighted tonnages, to a break-even point expressed in terms of per-
centage of capacity operated, as this percentage is ordinarily computed. The
conversion made in Table 37 can serve as a rough guide and nothing more. It
indicates that the break-even point, under prices prevailing in the last half of
1938, of approximately ten and a half million weighted tons would represent about
55% of capacity on the ordinary basis. If the preponderance of light products
that prevailed during 1938 recurred, the equivalent unweighted tonnage would
be somewhat lower, reducing the break-even point to about 50%. Regardless of
this variation, the entire study indicates beyond question that the nature of the
demand for steel and the cost pattern of the United States Steel Corporation and
its subsidaries are such that the increased volume of sales that might result could
not compensate for a general price reduction, and that general lowering of prices
is not the road to prosperity, undiminished employment, and undiminished pay-
rolls in the steel industry.
Appendix I. Computation of Average Working Capital-
-1938
Item
Jan. 1, 1938
Dec. 31, 1938
Total
$480,737,119.87
117,331,070.84
$510, 338, 510. 30
79, 261, 328. 94
$991, 075, 630. 17
196, 592, 399. 78
Working Capital
$363,406,049.03
$J31, 077, 181.36
$794, 483, 230. 39
Average Working Capital
$397, 241, 815. 19
«
CONCENTRATION OF ECONOMIC POWER
Chart 22
14075
PER CENT OF WEIGHTED TONS OF ROLLED AND FINISHED PRODUCTS SHIPPED
REPRESENTED BY UNWEIGHTED TONNAGES OF ROLLED AND FINISHED
PRODUCTS SHIPPED
U. S STEEL CORPORATION AND SUBSIDIARIES
105
100
z
Ui
" 95
a.
90
85
80
105
100
UJ
95 "
ee.
UJ
a.
90
85
80
1934^
1935
.,36'
^''
.93.
930
1927
■^'^
_,-
>9«
^--
193*
"'
■^
'
1932
(
) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
UNWEIGHTED TONS OF ROLLED AND RNISHED PRODUCTS SHIPPED
Appendix II. Computation of Percentage Increase in Volume to be
Expected from Percentage Decrease* in Price, Assuming Elasticity
OF Demand for Steel op 1
Let:
p = price
v = volume
s = total sales in dollars
d = fractional decrease in price
i = fractional increase in volume
Then:
(1) pv=s
If the elasticity of demand for steel is 1, a decrease in price will be accompanied
by a corresponding increase in volume sufficient to keep the total sales in
dollars unchanged.
Hence:
(2) (p-pd) (v + vi)=s
■ Substituting from (1):
(3) (p-pd) (v+vi) = pv
Factoring:
(4) pv(l-d) (l + i)=pv
Dividing by pv (1-d) :
pv
(5) l + i=
Simplifying:
(6) l + i=
pv (1-d)
1
1-d
Subtracting 1 from each side of the equation:
1
(7) i= 1
1-d
12449] — 41— pt. 26 —32
14076 aONOENTRATION OF ECONOMIC POWER
Simplifying:
1
(8) i =
1-d
1-d
d
1-d
1-d
Substituting various values for d :
d
1
d
i
d
i
1 <■
:
d
i
0.01
0.0101
0.11
0. 1236
0.21
0.2658
1 0.31
0. 4493
0.41
0. 6949
.02
.0204
.12
.1364
.22
.2821
.32
.4706
.42
.7241
.03
.0309
.13
.1494
.23
.2987
.4925
.43
.7544
.04
.0417
.14
.1628
.24
.3158
.34.
.5152
.44
.7857
.05
.0526
.1,";
.1765
.25
.3333
.35
.5385
.45
.8182
.06 _
.0638
.16
.1905
.26
.3514
.36
. .5625
.46
. 8519
.07
.0753
.17
.2048
.27
.3699
.37
.5873
.47
. 8868
.08
.0870
.]8
.2195
.28
. 3889
.38
.6129
.48
.9231
.09
.0989
.19
.2346
.29
.4085
.39
.6393
.49
.9608
.10
""
.20
.2500
.30
.4286
.40
.6667
.50
1.0000
Appendix III. Actual Operating Deficit in 1938 and Estimated Addition to
Deficit if Prices Had Been Further Reduced
Actual Operating Deficit — 1938:
Actual Deficit Before Provision for Income Taxes but after
Interest, as per Statement to F. T. C $4, 787, 454
Non-Operating Income:
"Other Income (Net)", as per Statement to
F. T. C $2,524,320
Add: Idle Plant Expenses included therein 2, 440, 185
Subtract: Discounts on Purchases included
therein
4, 964, 505
993, 387
3,971, 118
Actual Operating Deficit 8, 758, 572
Estimated Addition to Deficit if Prices Had Been Further Reduced:
Given:
1938 volume = 7.8 millions of weighted tons.
Sales at 1938 prices = $71.86 per weighted ton.
1938 transportation and miscellaneous revenues = $5. 80 per weighted
ton.
Costs = $55,734 per weighted ton + 182.1 millions of dollars.
Let:
d = fractional reduction in price.
i = resulting increase in volume.
X = the additional loss (in millions of dollars).
L=average loss at 1938 average prices and volume (in millions of
dollars) .
Li = average loss at reduced price and changed volume (in millions of
dollars).
Then:
X = L,-L
Since los.ses equal costs minus sales and revenues:
L = [55.734 (7.8) + 182.1]-[71.86 (7. 8) + 5.80 (7.8)].
L, = [55.734 (7.8) (l + i) + 182.1]-[71.86 (1-d) (7.8) (1 + 0 + 5.80 (7.8)
, (l + i)l
Removing brackets and simplifving (2):
L = 55.734 (7.81 + 182.1-71.86 (7.8) -5.80 (7.8)
L= 182.1 -21.926 (7.8)
L= 11.077
Removing brackets and simplifving (3):
L, = 55.734 (7.8) (1 + i) + 182.1 -71.86 (1-d) (7.8) (l + i)-5. 80 (7.8)
(1 + i)
L, = 49.934 (7.8) (1 + i) f 182.1 - (71. 86-71. 86d) (7.8) (1 + i)
(1)
(4)
(5)
(6)
(7)
(8)
CONCENTRATION OF ECONOMIC POWER
Factoring:
(9) Li = 7.8 (1 + i) [49.934- (71. 86-71.86d)J+ 182.1
Simplifying:
(10) L, = 7.8 (1 + i) (49.934) -71.86 + 71.86d) + 182.1
Li = 7.8 (1 + i) (71.86d-21.926) + 182.1
Substituting (6) and (10) in (1):
(11; X = 7.8(l + i) (71.86d-21.926) + 182.1-11.077
Simplifying:
(12) X=(l + i) (560.508d- 171.023) + 171.023
(13) X = 560.508d-l71.023 + 560.508di- 171. 023i+ 171.023
(14) X = 560.508d + 560.508di-171.023i
(15) X = 560.508d + i (560.508d- 171.023)
Additional Loss if Elasticity of Demand for Steel is as High as 1:
If the elasticity = 1:
(16) ''="i4d ^^®^ Appendix II)
Substituting (16) and (15):
(17) X = 560.508d+j4d(560.508d- 171.023)
Simplifying:
-^^560.508d-560.508d^ + 560.508d^- 171.023d
14077
(18)
(19)
(20)
1-d
._560.508d- 171.023d
^389.485d
1-d
Substituting various values for d:
d
X
d
X
d
X
d
X
0.1
3 9
7.9
12.0
16.2
20.5
24.9
29.3
33.9
38.5
43.3
48.1
53.1
58.2
63.4
68.7
74.2
.02
.07
.12
.17
79.8
.03—.
.08
.13
.18
85.5
.04 __.
.09
.10 _...
.14...
.15
.19
r.
.05
.20
Additional Loss if Prices Had Been Reduced and No Increase in Volume Resulted:
If no increase in volume resulted:
(21) i=:0
Substituting (21) in (15) the* general equation:
(22) X = 560.508d
Substituting various values for d:
d
\^
d
X
d
X
d
X
0.01
5.6
1112
16.8
22.4
28.0
0.16
33.6
39.2
44.8
50.4
56.1
61.7
67.3
72.9
78.5
84.1
0.16-.
■g9.7
.02.
.07
.12
13
.17. __
18
95.3
.03
08
100.9
.04
.09
.14
19
106.5
.05
.10 -
.15
20
112.1
Appendix IV. Weighted Tonnage Which Must Be Shipped to Break Even
AT Various Price Levels
Given:
Costs = 55.734 V+ 182.1, where:
Costs are in millions of dollars, and
V = millions of weighted tons shipped
Transportation and Miscellaneous Revenues under 1938 conditions =
5.80 V
Let: P = Sales per weighted ton shipped
V| = V at the break-even point
At the break-even point Sales and Revenues will equal costs; therefore:
P V,+ 5.80 V, = 55.734 V, + 182.1
CONCENTRATION OF ECONOMIC POWER
•55.734 V, = 182.1
14Q78
Transposing:
P V, + 5.80 V,-
Factoring and simplifying:
Vi (P-49.934) = 182.1
Dividing by (P-49.934):
v_ 182.1
^' P-49.934
Substituting various values for P:
Price conditions
P
(P-49.934)
V,
Average 1038
71.86
67.33
64.67
60.60
2). 926
17. 396
14. 736
10.666
8 305
2d half 1938
10 468
Average 1938 less 10%
12 357
2d half 1938 less 10%
17 073
Appendix V. Computation of Increases in Volume Needed to Offset
Decreases in Price
Given:
Total costs^ 182.1 millions of dollars +$55. 734 per weighted ton.
Sales, average 1938 prices =$71.86 per weighted ton shipped.
Sales, 2d half 1938 prices = $67. 33 "
Transportation and miscella-
neous revenues ■= $5.80 " " " "
Let:
V = volume before price decrease, in weighted tons.
¥!•= volume, in weighted tons, needed to offset decrease in price.
S = sales, exclusive of other revenues, per weighted ton shipped before
decrease in price,
d ■= fractional decrease in price.
X = fractional increase in volume needed to offset d.
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
Then:
= SV+5.80 V- (182.1 + 55.734 V)
V,- 55.734 V,
_Vi-V_Vi V_ Vi
X- V ~V V~V ^
Profit or loss before price decrease
Profit or loss after price decrease=(S-dS) Vi + 5.80 V,- 182. 1 + 55.734 V,
If the increase in Vi over B is to offset the decrease in S, (2) and (3)
will be equal. Hence:
SV+5.80 V- 182.1-55.734 V=(S-dS) V, + 5.80 V,- 182.1-55.734 V
Adding 182.1 to both sides:
SV+5.80 V- 55.734 V=(S-dS) Vi + 5
Factoring and simplifying:
V (S-49.934) = V, [(S-dS) -49.934]
Removing brackets:
V (S- 49.934) =Vi (S-dS- 49.934)
Vi^ S- 49.934
V S-dS-49.934
Substituting in (1) : .
S-49.934
^ S-dS-49.934 ^
CONCENTRATION OF ECONOMIC POWER
14079
Substituting percentage decreases from the average 1938 prices.
21.926
At average 1938 prices:
71.86-49.934
"71.86-71. 86d-49.934 ^"^ 21.926- 71.86d ^
Substituting various values for d:
d
21.926-71.86d
21.926
, 21.926 ,.
21.926-71 .86d
''^ 21.926-71.86d "
0 01
21. 207
20. 489
19. 770
19.052
18.333
17.614
li 177
15.459
14. 740
14.021
13. 303
12.584
ll!l47
10.428
9.710
8.991
8.273
7.554
6.835
6.117
5.398
4.680
3.961
3.242
2.524
1.805
1.087
0.368
1.0339
1.0701
1.1091
1. 1509
1.1960
1. 2448
1.2977
1.3554
1.4183
1.4875
1.5638
1.64S2
1.7424
1.S47S
1.9670
2. 1026
2. 2581
2.4387
2.6503
2. 9026
3.2079
3. 5844
4.0619
4. 0850
5. 5355
6. 7631
8.6870
12. 1474
20.1711
59. 5815
.0339
.02
.0701
.1091
.1500
.1960
.2448
.07
.3977
08
.3554
09
.4183
10
.4875
.11
.5638
.6482
.7424
.8478
.9670
16
1.1026
17
1,2581
18 ....
1.4387
19
1.6503
1.9026
2.2079
2.6844
3. 0619
24
3. 6850
25
4. 6355
.26 ..
27 .
5.7631
7.6870
28
11.1474
19.1711
59. 5815
Substituting percentage decreases from
At 2d half, 1938, prices:
67.33-49.934
"~67.33-67.33d-49.934
Substituting various values for d :
1 =
half, 19S8, prices:
17.396
17.396-67.33d
d
17.396 -67.33d
17.396
17.396
17.396-67.33d
^^■l7.396-67.33d "
16.723
16.049
15.376
14. 703
14.030
13.356
12.683
12.010
11.336
10.663
9.990
9.316
8.643
7.970
7.297
6.623
5.950
5.277
4.603
3.930
3.257
lioio
1.237
0.564
1.040
1.084
1.131
1.183
1.240
1.302
1.372
1.448
1.635
1.631
1.741
1.867
2.013
2.183
2.384
2.627
2.924
3.297
3.779
4.426
5.341
6.735
9.108
14.063
30. 844
.040
.084
.03
04
.18S
05 . ..
240
.06
.302
.372
.448
09
.635
10
.631
.741
.867
1.018
1.183
1.384
1.627
1.024
.18
.19...
.20 - 1
2.297
2.779
3.426
22 mil I"I
4.341
5.735
.23 ...
8.108
13.063
.26
29.844
14080 CONCENTRATION OF ECONOMIC POWER
Appendix VI. Operating Rate Required To Earn 5% on Tangible Invest-
ment IF Break-Even Point at Over 90% of Capacity
Tangible Investment: «^«"- -?. l^^^ Dec. 31, 19S8
Assets $1,918,729,289 $1,711,279,006
Less: Current Liabilities 117,331,071 79,261,329
Total Investment $1,801,398,218 $1,632,017,677
(Bondholders' & Stockholders' Interests)
Less Intangibles 260,368,522 1
Tangible Investment $1, 541, 029, 696 $1, 632, 017, 676
Average Tangible Investment, 1938 $1,586,523,686
5% Return on Average Tangible Investment $79, 326, 184
Less: Bond Interest 8,262,327
Return after Bond Interest $71,063,857
The break-even point would be at over 90% of capacity if prices were 10%
lower than the average prices prevailing in the 2d half of 1938 (see Appendix IV).
At these prices total sales and revenues would amount to $66.40 per weighted ton
of tonnage products shipped.
At these prices, profits, before income taxes, in millions of dollars =66.40 —
182.1-55.734 V. (See Appendix IV.)
To realize 71.0 million dollars, ignoring income taxes:
71.0=66.40 V-182.1-55.734 V
10.666 V = 253.1
253 1
V = YK-?^ = 23.73 millions of weighted tons
The trend of the relationship between tons of rolled and finished products
shipped and weighted tons of all tonnage products shipped indicates that the
weighted capacity figure for all tonnage products equivalent to the 17.9 millions
of unweighted tons constituting the rolled and finished capacity as of January 1,
1939, would be 17.85 millions of weighted tons. Hence, at such prices operations
23 73
would have to reach the impossible rate of yfs^R' or 133% of capacity, to realize
a return as modest as 5% on tangible investment.
Appendix VII. Effect of Inter Company Transactions on Profit
AND Loss Statement
The consolidated profit and loss statements of United States Steel Corporation,
submitted to the Federal Trade Commission do not state the sales and revenues
and the cost of goods sold and operating expenses of transportation and miscel-
laneous operations on a purely integrated basis. This is to say that the sales
shown are not merely sales to purchasers other than subsidiaries of the Corpora-
tion, but include sales between subsidiary companies. Thus tlie sale for $80 of a
product costing $60 to another subsidiary which resold it to outside purchasers for
$100 would result in sales of $180 on the profit and loss statement. Cost of goods
sold would similarl}' involve a duplication and would amount in all to $60, the cost
to the first company, plus $80, the cost to the second company, or $140. This
would leave a gross profit of $40. Considering the consolidated companies as a
single unit, what Has really happened is that goods costing $60 were sold for $100,
leaving a profit oi $40. thus the profit is neither overstated nor underslatcd by
this method of handling. From an integrated viewpoint, both sales and costs
are inflated, however, by the amount of the inter-company .sales.
This relationship is not disturoed by sales to suhsid-.iries for use in constructing
capital equipment since such sales arc treated as outside sales and the full sales
price is charged to capital recount by the purchasing company and on the con-
solidated balance sheet of the Corporation. Only goods sold for conversion or
resale are treated as inter-company items.
The fact that goods purchased by one subsidiary from another are retained in
inventory or that goods sold were sold from inventory does not affect the relation-
ship between inter-company sales and integrated costs, since adjustment is always
CONCENTRATION OF ECONOMIC POWER 14081
made to defer the taking up as profit on the consolidated profit and loss statement
the profit shown by one subsidiary on sales to another subsidiary when the goods
sold are still in the inventory of the second subsidiary. This profit is taken up
as realized profit only when the goods are sold outside the organization. The
eff'ect of this adjustment for inter-company profit in inventory may be seen in the
following series of illustrations.
Case I. Company A sells to afl^liated Company B for $100 a product which
cost it $80. Company B sells this product for $120. The sales and cost of goods
sold set-up would be as follows:
Item
Co. A
Co. B
Consoli-
dated
Inter-Co.
Sales
Integrated
Basis
Sales
$100
80
$120
100
20
$220
180
$100
100
$120
Cost of goods sold
80
20
40
Case II. In addition to the transaction in Case I, Company A sells to Company
B for $80 a product which cost it $60. Company B has this in inventory at the
end of the year, and the profit shown by Company A on the second sale to Com-
pany B is treated as unrealized on the consolidated profit and loss statement
because the product has not been sold outside the consolidated organization. The
sales and cost of goods sold set-up would then be as follows:
Item
Co. A
Co. B
Inter-Co.
Profits in
Inventory
Consoli-
dated
Inter-Co.
Sales
Inte-
grated
Basis
Sales
$180
$120
$300
$180
$120
Cost of goods manufactured or pur-
$140
$180
80
$320
60
$180
$140
$20
6(1
Cost of goods sold
140
100
■ 20
260
180
80
40
20
20
- 40
Case III. In addition to the transactions in Cases I and II, Company B sells
for $75 a product which it bought from Company A the previous year for $65.
It had cost Company B $60 to produce this article, with the result that $5 inter-
company profits tied up in inventory was deducted from consolidated gross
profits in the previous year by addition to cost of goods sold in a manner similar
to the $20 adjustment for inter-company profit in inventory in Case II. The sales
and cost ol goods sold would then be as follows:
Item
Co. A
Co. B
Inter-Co.
Profits in
Inventory
Consoli-
dated
Inter-Co.
Sales
Inte-
grated
Basis
Sales
$180
$195
$375
$180
$193
65
180
$5
60
320
380
60
60
Cost of goods manufactured or pur-
140
180
180
140
Total goods available for sale...
140
245
80
5
20
200
60
140
165
15
320
180
140
40
30
15
55
;5
The fact that goods purchased from another subsidiary are usually further fabri-
cated before being resold does not change the relationship. It merely adds the
cost of such fabrication to the cost of goods sold bv the purchasing subsidiary.
14082
CONCENTRATION OF ECONOMIC POWER
rience the integrated cost will be the cost to the first subsidiary plus the cost of
fabrication incurred by the second subsidiary. It will still be less than the cost
shown on the consolidated profit and loss statement by the amount of the inter-
company sales.
This may be illustrated by a situation where the first subsidiary sells a product
costing $60 to the second subsidiary for $80. The second subsidiary fabricates
it further at a cost of $10 and sells it to an outside purchaser for $120. The profit
and loss set-up is shown below, the column headed "Consolidated", showing the
figures that would appear on the consolidated profit ."nd loss statement and the
column headed "Integrated" showing the figures on an integrated basis:
Item
Co. A
Co.B
Consoli-
dated
Inter-Co.
Sales
Integrated
Basis
Sales --
$80
$120
$200
$80
$120
Cost of goods purchased
80
10
80
70
80
60
60
70
90
150
80
70
Gross Profit
20
30
50
SO
It should be noted that the costs shown on the consolidated profit and loss
statements are always in excess of integrated cost by the amount of the current
year's inter-company sales. Hence it is proper to deduct from costs of goods sold
as shown on the profit and loss statement the amount ot the inter-company sales
and revenues in order to arrive at a cost figure which is comparable with the
tonnages shipped to outside purchasers.
Exhibit No. 1417
AN ANALYSIS OF STEEL PRICES, VOLUME AND COSTS CONTROL-
LING LIMITATIONS ON PRICE REDUCTIONS
This analysis was prepared by the United States Steel Corporation in connec-
ton with its studies in preparation for the hearings on the steel industry before
the Temporary National Economic Committee. The work was under the super-
vision of Theodoie O. Yntema, Professor of Statistics, UnivQrsii)y of Chicago.
OCTOBEB 30, 1939.
Importance of Peice-Volume-Cost Relationship
The success of mass production methods in American industrial practice has
given much emphasis to the importance of volume in reducing costs. Failure to
appreciate, in the case of the steel industry, the limitations of the extent to which
increased volume means reduced costs per unit of product has frequently been
made the basis of the charge that steel prices are higher than thq|>y should be arid
that the absence of price competition, rather than the limitations imposed by
costs, is the reason why steel prices are not lower than they are. It is further
said that if steel prices were lowered in times of recession more steel would be sold
and payrolls and employment would be better maintained. The fact of the matter
is that steel prices are not high. In recent years they have barely covered costs.
In the last ten years, 1929 to 1938, the United States Steel Corporation realized
a return of only 1.9% on the combined investment of the bondholders and stock-
holders. From 1930 to 1938 its rate of return has averaged less than 1%.
However, the contention that steel prices are too high is not made in ignorance
of the small margin of profit currently being realized. The theory of those who
would lower steel prices is that such reductions would greatly stimulate sales and
that volume and costs are so interrelated in the production of steel that the cost
per ton would be drastically reduced if volume were increased. Hence, it is
argued that if prices were decreased even belc^v present costs, volume would be
stimulated to a point where costs would drop suflSciently to make steel production
profitable at the reduced prices. In determining whether or not this is so, two
all important factors must be considered. They are (1) the amount by which
unit costs do in fact shrink as volume increases and (2) the amount by which
volume is in fact increased by lowering prices.
CONCENTRATION OF ECONOMIC POWER 14083
Costs and Volume
It would have been possible to obtain some idea of the way in which the total
costs of the United States Steel Corporation and its subsidiaries increase with
increases in output simply by comparing total costs shown on its consolidated
annual profit and loss statements over a period of years with the tonnage of rolled
and finished products shipped each year. Such an analysis would not be wholly
accurate, however, for several reasons.
In the first place, the total costs for the same number of tons shipped would
differ depending upon the type of product whose tonnage predominated. For
instance, if the shipment figures showed that in each of two years 10,000,000 tons
had been produced, the costs would naturally be higher in the year in which high
cost products, such as sheets and tin plate, constituted the larger portion of the
tonnage, than in a year when lower cost items, such as rails and heavy plates,
predominated. In the following analysis this difficulty has been adjusted for by
computing for each year a tonnage figure representing what the equivalent gross
tonnage, from a cost standpoint, would have been if the total shipments were
constituted of normal proportions of high and low cost products. Similarly, the
total tonnage of rolled and finished products shipped has been adjusted to include
the equivalent gross tons of steel represented by the products other than steel
which are sold on a tonnage basis by the Corporation's subsidiaries.
Secondly, the cost figures taken from the profit and loss statements represent
costs in different years when different wage rates, interest rates, tax rates and
prices prevailed. Hence, it is necessary to adjust the cost figures for years prior
to 1938 to conditions prevailing in 1938 in order to ascertain what the relation
between costs and volume would be as of 1938, the most recent full year and the
one in which material prices, pension payments, and wage, interest, and tax rates
are most representative of present-day conditions.
Since diffarent adjustments are necessarily required for different types of costs,
the total costs of the Corporation and its subsidiaries, exclusive of inter-company
items and costs connected with extraneous non-operating transactions, for each
of the years 1927 to 1938 were broken down into (1) interest, (2) pensions, (3)
depreciation and depletion, (4) taxes other than social security and Federal income
and profits taxes, (5) payroll, (6) social security taxes, and (7) other expenses.
Each of these costs as they appeared in past years has been separately adjusted
to the levels of 1938. ^ Federal income taxes have been omitted from the com-
putations because they depend upon profits rather than volume and it was the
purpose of this analysis to work out a cost-volume relationship which can be
compared to total sales and revenues at various price and volume levels.
The total costs obtained by adding together the adjusted items for each year
represented what the costs would have been in those years if 1938 prices and
interest, wage, and tax rates had prevailed. A further adjustment was then made,
based on the downward trend of costs in relation to volume over the period of
time involved, to take into consideration the extent to which the same tonnages
could have been produced in 1938, because of increased eflficiency, at lower cost
than they could have been produced in prior years at 1938 prices, pensions and
wage, interest, and tax rates.
' The adjustments applied to each of the cost factors enumerated were as follows:
fl) Interest— The interest eost, not being dependent on volume, was converted to 1938 conditions by
substituting the 1938 interest charge in the figures for each year.
(2) Pensions— The cost of pensions, like interest, was converted to 1938 conditions by substituting the
1938 figure.
(3) Depreciation and Depletion— Since there has been no important change in the Corporation's account-
ing policy, no adjustment has been made in the depreciation and depletion figures.
(4) Taxes Other than Federal Income and Profits Taxes and Social Security Taxes— Since analysis of the
relation of taxes to volume discloses that the taxes for 1933 onward follow one pattern while the taxes for
previous years were considerably lower, adjustment was made for the changed tax laws by substituting in
prior years the taxes for the volume involved which were indicated by the 1932-1938 line of average relation-
ship between taxes and volume.
(5) Payroll— The payrolls for each of the respective years were adjusted to 1938 rates on the basis of the
proportionate change in the average hourly earnings between the year in which the payroll was incurred
and 1938.
(6) Social Security Taxes— Social security taxes at 1938 rates for the various amounts of payroll were
estimated by applying the 1938 ratio of the?e taxes to payroll.
(7) Other Expenses— This item consists largely of goods and services purchased from others. An approxi-
mate adjustment for the changing prices of these items which the Corporation must purchase has been
made on the basis of the Bureau of Labor statistics index of wholesale prices for commoditiesother than
food and farm products. While low operating rates are usually accompanied by somewhat lower material
prices, material costs were adjusted to 1938 price levels in this study in order to ascertain the chan?"' 'n
unit costs which would be attributable to changes in volume alone.
14084 OONOENTRATION OF ECONOMIC POWER
The final adjusted costs and the weighted tonnages to which they are related
are as follows:
Table 1. — Total Costs and Volume of Business — 1938 Conditions ' — United States
Steel Corporation and All Subsidiaries
Millions of Weighted
Tons of Products
Shipped
Costs— 1938
Conditions
(Millions of
Dollars)
Year on
which Esti-
mate is
Based
Millions of Weighted
Tons of Products
Shipped
Costs-1938
Conditions
(Millions of
Dollars;
Year on
which Esti-
mate is
Based
436.0
510.0
512.0
610.3
614.3
628.9
1932
1934
1933
1935
1938
1931
818.2
954.5
916.2
966.2
979.0
1936
6.1
11.9. -
1930
13.0
1927
7.6
13.2
1937
7.8
1928
8.1
15.1
1929
! Total costs are adjusted to 1938. interest, pension, wage, and tax rates, to :938 price level, and to 1938
efficiency.
These costs have been plotted in relation to the volume of shipments on Chart
1. Insertion of the straight line established by the respective points shows the
average relationship between volume and cost under 1938 conditions. The
smallness of the vertical deviations of the various points from this line of aver-
age relationship indicates how closely the costs of the United States Steel Cor-
poration and its subsidiaries follow this pattern.
Chart 1
RELATIONSHIP BETWEEN TOTAL COSTS OF OPERATION
AND VOLUME OF BUSINESS - 1938 CONDITIONS
U. S. STEEL CORPORATION AND SUBSIDIARIES
1200
HOC
1000
900 %
800 1
700 ^
600 o
500 ^
Ann °
1100
1000
^ 900
i "°
° 700
o 600
^ 500
2 400
i 300
200
100
0
^
1927
^
iis
^
1930
^
J
35.-
r<^.
1934
^..
-.
^
^
^
300 g
200
100
0
_^
^
(
NOTE: TOTAL CO-
) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1
MILLIONS CF WEIGHTED TONS OF ALL TONNAGE PRODUCTS SHIPPED
STS ADJUSTtO TO 193S IMTERtST, T«. PWSBN. AND WASE RATES; TO 1938 PRICE IfVELS AND TO l?38 EFTCiENCY
8
Mathematical computation of the cost relationship indicated by the cost line
shows that the total costs of the Corporation and its subsidiaries amount to
$55.73 per weighted ton,^ plus $182,100,000.' The $182,100,000 represents the
' The total weighted tonnage shipped does not include the volume of goods and services sold on other
than a tonnage basis. This additional cost of $55.73 represents the additionrtl cost of all goods and services
shipped or sold per weighted ton of tonnage products shipped.
' A similar analysis of the unadjusted costs shows that the average relationship prevailing during the
period 1927-1938 between costs and volume was such that total costs tended to be $.'>4.5I per weighted ton
plus $120,530,000. If these figures were substituted for those based on the adjusted costs, the calculations
which are to follow would be changed considerably, but the general conclusions as to the increase in volume
needed (o oflset a price decrease would remain unaffected.
OONOENTRAXr. X OF ECONOMIC POWER 14085
portion of the costs which remains the same, regardless of the vokime of steel
produced, as long as the Corporation's subsidiaries are operating at all. Such
costs are frequently termed ""xed costs" or "overhead costs." The $55.73 per
weighted ton represents the a iditional cost of the increased operations incidental
to each successive weighted ton of product sliipped. Such costs are termed
"variable costs," "incremental costs," or "additional costs." In the case of the
Corporation these additional costs remain constant at $55.73 per ton through
the range of volume within which the Corporation's suljsidiaries operated from
1927 to 1938. While additional costs might possibly vary if facilities were
pressed to the absolute limit, or if operations fell to a point even lower than they
did in the depth of the depression in 1932, the 1927-1938 experience includes
annual rates of operation varying from 17.7% to 90.4% of ingot capacity.
Although the ij:< rease in costs with each additional ton of steel produced tends
to remain constant, this does not mean that the average cost per ton of steel
remains the same. Since the average cost amounts to $55.73 per ton plus the
pro rata portion of fixed costs and since the $182,100,000 of fixed costs can be
distributed over more units as production is increased, the average cost of
operations per ton of steel shipped will obviously decrease as volume rises.
It should be noted here, however, that while the costs mentioned are exclusive
of all non-operating income and expense, they cover all operations of the United
States Steel Corporation and its subsidiaries and, hence, do not represent merely
the cost of producing steel. Furthermore, even weighted tonnages shipped do
not reflect the full volume of business, since some goods and services are sold by
the Corporation's subsidiaries which are not measured in tons. Nevertheless,
other operations rise and fall with increases and decreases in shipments of prod-
ucts to a sufficient extent that the total costs maintain approximately the
relationship to shipments just described.
An analysis of the elements composing the fixed and the additional costs
discloses that they are composed as follows:
Table 2. — Elements of Total Costs 19S8 Conditions — United States Steel Corpora-
tion and all Subsidiaries
Item
Costs that
must be met
regardle5s
of operating
rate
Additiona.
cost lor each
additional
weighted ton
of product
shipped
$8, 300, 000
7, 700, 000
24, 200, OOO
62, 100, 000
2, 500, 000
47, 800, 000
0.00
1.43
Payroll
29 10
Social Security Taxes
1 16
Other Cash E.xpenses
21.67
$152, 600, 000
29, 500, 000
2:37
Total Costs
$182,100,000
$55 73
The relationship of the various elements of cost to volume is graphically portrayed
in Chart 2. It is worthy of note that not only does payroll constitute the largest
item of additional costs but it is also the largest element of the overhead or fixed
costs, amounting to over one-third of that item.
Prices and Volume
If the relationship between volume and costs is known, it becomes possible to
determine the increase in volume that would be needed as the result of a price
reduction in order to ofiFset the decreased amount received for each unit of product
sold, assuming no reduction in wage rates, material prices, and other such factors.
Similarly, to the extent that it is possible to estimate the increase in volume that
would be likely to result from a price decrease, it is possible to'estimate what is to
be gained or lost by reducing prices.
In connection with the increase in volume that would be attained through a
price reduction, it must be remembered that if any substantial increase results, it
will have to result from an increase in the total amount of steel consumed, since
competitive meeting of prices will generally prevent any sustained increase in the
14086 OON01]NTRATION OF ECONOMIC POWER
participation of the United States Steel Corporation and its subsidiaries in the
going volume of business. On the other hand, the rise to be expected in the total
volume of steel consumed as a result of a drop in prices is not very great in the
steel industry. Steel is not sold directly to the ultimate consumer. 'It reaches
him only as a part of the finished automobile, typewriter, apartment house, tin
can, or safety pin, as the case may be. In other cases, steel is used only as part
of the machinery and equipment used in making the products which reach the
man in the street. No matter how low the price, steel can be sold only if products
which are produced from steel or by the use of steel are being sold. In the case of
products produced from steel, the cost of steel is usually so small a fraction of the
Chart 2
COMPOSITION OF TOTAL COSTS OF OPERATION
IN RELATION TO VOLUME OF BUSINESS
U. S. STEEL CORPORATION AND SUBSIDIARIES
1200
1100
NOTE. 19271938 EXPERIENCE ADJUSTED TO 1938 CONDITIONS
total cost of the product that a reduction in steel prices, even if passed on to the
ultimate consumer, would not result in a sufficient decrease in the price of the
finished product to cause an appreciable increase in its sale. As far as steel for
production is concerned, it is evident that regardless of the price of steel no one
will invest in productive machinery unless he feels the prospects in his particular
line of business justify such investment.
Analysis of the influence of price as a factor affecting slefiT consumption in the
automobile, railroad and container industries reveals that a decrease in the price
of steel can increase the consumption of steel only to a limited extent by promoting
the use of more steel per unit or permitting stool to be substituted for some
other competing rnaterial. Any substantial increase in the consumption of steel
in these industries could be brought about only by increasing the consumption of
the ^nished product or service rendered. Consequently, the price elasticity of the
I
CONCENTRATION OF ECONOMIC POWER 14087
demand for steel depends primarily upon the price elasticity of demand for the
finished product and the relative cost of steel to the price of the finished product.
The elasticity of demand is measured by the ratio of the relative resulting
increase in volume to the relative decrease in price. For example, in the case of
automobiles, an exhaustive study by Messrs. Roos and von Szeliski ♦ showed that
the price elasticity of demand for automobiles was about 1.5, which means that a
1% reduction in price would increase the number of automobiles sold about 1.5%.
Since the cost of steel in the form sold by the steel producer is about one-tenth of
the retail price of a representative low-priced automobile, it follows that a reduc-
tion of 10% in the price of steel, even if passed on to the ultimate consumer,
could effect at most only a 1% reduction in the price of the delivered automobile,
which, as has been stated, could bring about but a 1.5% increase in the number of
automobiles sold and in the amount of steel used in the automobile industry.
The increased consumption arising out of the extent to which steel might be sub-
stituted for some other material, or the extent to which the use of steel per auto-
mobile might be increased if steel prices were reduced, would probably not increase
the elasticity by more than .1. Taking mto account all factors and making a
liberal allowance for possible error, elasticity of demand for automotive steel is
not in excess of .2 or .3.
The price elasticities for the finished products or services in the container, and
railroad industries have not had the benefit of as definite measurement as that
made of the demand for automobiles by Messrs. Roos and von Szeliski, but the
evidence indicates that the demand for these products is considerably less elastic
than the demand for automobiles. Moreover, analyses of all the factors influenc-
ing steel consumption in these industries, even assuming an elasticity of demand
for the finished products and services as high as 2, show definitely that the price
elasticity of the demand for steel in each of the respective industries is considerably
less than 1. A mathematical analysis of the correlation between the amount of
all steel sold and the various factors, including price, which influence the quantity
sold, reveals that a negligible portion of the fluctuations in the quantity sold are
attributable to price and that a steel price change, other factors reniaining un-
changed, will not result in as great a percentage change in the volume of steel
sold. This confirms the individual analyses of the principal steel fconsuming in-
dustries already mentioned.
While the above analyses indicate clearly that the elasticity of the defnand
for steel is considerably less than 1, it has been assumed for the purposes of
this study, in order to use a figure that is beyond question, that the elasticity
of demand for steel is as high as 1. An elasticity of 1 means that any decrease
in price will result in a proportional increase in volume which will keep the total
sales in dollars unchanged. This is to say that for small changes in price a
given percentage decrease, such as a 5% reduction in price, will result approxi-
mately in the same percentage increase in volume sold.
The Effect op Price Reductions
It is evident that unless the elasticity of demand for the product exceeds 1 by
a substantial margin, the theory that price reduction in and by itself would
produce profits through increased volume is utterly fallacious, not only for the
United States Steel Corporation, but for any business or any industry. Since,
with the elasticity of demand equalling 1, the total sale receipts would remain
the same, for the theory to work the total costs of producing the greater volume
would have to be the same or Ijess than the amount required to produce the
original volurpe. No increased payroll could be incurred to produce the greater
volume, for example. Such a condition could exist only when all costs were
"fixed" or "overhead" and none were "additional" or "variable". Only then
would the cost per unit go down, relatively, as fast as the volume went up.
Application of the theory of increased profits through price reduction when the
elasticity of demand for the product is 1 or less could thus result only in loss to
the enterprise which adopted it. The actual amount to be lost by reducing
steel prices, however, and amount by which the increase in volume to be ex-
pected as a result of reducing steel prices falls short of the increase needed to
offset the price reduction, can be estimated only by including in the computation
the relationship which exists between costs and volume.
The sales and revenues of the United States Steel Corporation's subsidiaries
in 1938 amounted to $77.66 per weighted ton of products shipped. Of this
amount $71.86 represented the amount received from the sale of goods and
$5.80 represented revenues from transportation and miscellaneous operations.
« C. F. Roos and Victor von Szeliski, "Factors Governing Changes in Domestic Automobile Demand."
The Dynamics of AtUomobilt Demand, General Motors Corp'n, N. Y., 1939.
14088
CONCENTRATION OF ECONOMIC POWER
On Chart 3, the line representing total costs as volume varies is compared to
the corresponding amount of total sales and revenues at the 1938 amount per
weighted ton. This relationship indicates that the break-even point at 1938
average prices was at 8.3 million weighted tons. This is equivalent to an oper-
ating rate of 40% to 45% of capacity, depending upon the type of products
constituting the total products shipped.
A price decrease of 10% from the average 1938 prices, assuming that the
amount per weighted ton received from transportation and other operations
remains unchanged, would result in total sales and revenues as represented by
the dashed line in the chart. ^ The result of such a price reduction would be to
Chart 3
RELATIONSHIP BETWEEN SALES AND COSTS
EFFECT OF REDUCTION FROM AVERAGE 1938 PRICES
U. S. STEEL CORPORATION AND SUBSIDIARIES
1200
I I I I I TH"
1200
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
MILLIONS OF WEIGHTED TONS OF ALL
TONNAGE PRODUCTS SHIPPED
NOTE; COSTS ARE B«EO ON 1927.1938 EXPERIENCE, ADJUSTCO TO 1933 CONDITIONS
rai.se the break-even point to 12.36 million weighted tons, an increase of 48.8%,
which should be contrasted with the 11% increase in volume which is the maxi-
mum that might be expected from a 10% price decrease. The same relationship
holds not only with regard to the break-even point, but also with regard to the
netting of any particular amount of profit or loss. For instance, if production
amounted to 6,000,000 weighted tons, the loss at 1938 average prices would be
about $50,500,000. If average 1938 prices were decreased 10%, volume
would have to be raised 48.8% to atout 8,925,000 weighted tons in order not to
increase the loss.
In this discussion it has been assumed that the revenues from transportation
and miscellaneous operations would increase sufficiently as the production of
' The reduction has been made in the $71.86 representing 1938 sales per weighted ton, but not in the $5.80
per weighted ton representing transportation and miscellenaous revenue.
CONCENTRATION OF ECONOMIC POWER
14089
steel increased so as still to amount to $5.80 per ton. The general tendency of
these revenues has been to rise and fall approximately as product shipments go
up and down, but in recent years there has been a tendency for them to fall off
somewhat per ton of products shipped when the higher levels of production are
reached. To the extent that revenues from transportation and miscellaneous
operations might fail to increase proportionately with increases in shipments, the
increases in volume needed to compensate for the various price decreases would
be even greater than those stated.
The impracticabihty of attempting to raise volume from recession to prosperity
levels by means of price decreases may be seen from the effect of the price reduction
that would have been needed to bring the 1938 volume of the Corporation's sub-
sidiaries up to 1937 levels. Such^an increase would have been from 7,800,000 to
13,200,000 weighted tons, a rise o*f 69.23%. Since steel has at best an elasticity
of demand of 1, such an increase could have been brought through price reduction,
if at all, only by a price decrease of at least 40.9%, reducing sales per weighted ton
from $71.86 to $42.46.^ Disregarding the possibility that other operations might
not expand with increased shipments and adding in the full $5.80 per weighted ton
realized from other operations at actual 1938 volume, the total sales and operating
revenues would then have amounted to $48.26 per weighted ton. Since the vari-
able cash costs amount to $53.36 per weighted ton, the Corporation's subsidiaries
would have sustained a cash loss of $5.10 on every weighted ton sold in addition
to failing to recover any part of the fixed cash costs of $152,600,000. Assuming
that the price reduction would have been successful in restoring the 1937 volume,
13,200,000 weighted tons would have been sold. At a loss of $5.10 a ton, plus
$152,600,000 fixed costs not covered by the sales price, the Corporation and its
subsidiaries would have had a cash loss for the year of $219,920,000. This is
without making any provision for depreciation and depletion of fixed assets. The
drain of such a cash loss upon the Corporation and its subsidiaries, if continued,
would have exhausted their 1938 average working capital in less than two years.
If the amount of the depreciation and depletion of assets at this volume of opera-
tions, amounting to $60,784,000, is added to the cash loss, there is a total loss of
$280,704,000. Annual losses at this rate would wipe out the combined equity of
the preferred and common stockholders as of December 31, 1938, In about four
and a half years.
While a smaller decrease in price could not have been expected to raise the 1938
volume to the 1937 level, it may nevertheless be contended that some price reduc-
tion and some resulting stimulation in volume were desirable. At 1938 average
prices, however, the increase in volume which would take place even if the elasticity
of demand in steel were as high as 1 would be far short of the percentage increase
needed to prevent additional loss. The increases in volume needed to offset
various percentage reductions in price and the maximum probable increase in
volume which would result are as follows:
Table 3. — Increases in Volume Needed to Compensate for Various Decreases in
Average 1938 Prices — Compared to Probable Resulting Increases in Volume —
United States Steel Corporation and Subsidiaries
Percentage Reduc-
tion in Price
Percentage
Increase in
Volume
Needed
Probable
Percentage
Increase, As-
suming Elas-
ticity of 1
Percentage Reduc-
tion in Price
Percentage
Increase in
Volume
Needed
Probable
Percentage
Increase, As-
suming Elas-
ticity of 1
J
3.4
7.0
10.9
15.1
19.6
24.5
29.8
35.5
41.8
1.0
2.0
3.1
4.2
5.3
6.4
7.5
. 8.7
9.9
10
48.8
56.4
64.8
74.2
84.8
96.7
190.3
453.-6
5, 858. 2
11 .
12
13 :
14.
15
20
25
25 0
8
33 3
g
30
42 9
The divergence between the needed increase in volume and the maximum probable
resulting increase, based on an elasticit,v of 1, is illust ated in Chart 4.
The following tabulation sets forth what the additional loss in dollars to the
United States Steel Corporation and its subsidiaries would have been if the 1938
« If the elasticity of demand were 1, the total sales in dollars would remain unchanged by price reductions
since volume would increase to a compensating extent. Hence dividing the 1938 sales of .$560,508,000 by
13,200,000 will give the average price at which steel would have to have been sold to have shipped 13,200.000
weighted tons in 1938.
14090
OONCENTRATION OF ECONOMIC POWER
average prices had been reduced various percentages. Separate estimates have
been made showing the additional loss if the maximum probable increase in volume
resulted and if no increase in volume resulted from the price reduction. The
actual loss would have to fall somewherebetween these two limits.
Chart 4
INCREASES IN VOLUME NEEDED TO COMPENSATE FOR
VARIOUS DECREASES IN 1938 PRICES
COMPARED TO PROBABLE RESULTING INCREASES IN VOLUME
U. S. STEEL CORPORATION AND SUBSIDIARIES
240
220
?
180
n
^
160
z
140
s
12U
:^
Ul
ion
80
Q-
60
40
20
n
r
r
n
n
'
„r
P
I
V
REA
OL
ED
SES
JM
ED
n
I
.-.-
.;.:
''
■
,/:'
..jjM-
probable
resulting
'increases
in volume
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 13 19 20 21
PERCENTAGE DECREASES FROM AVERAGE 1938 PRICES
NOTE; PROBA>iLE RESULTING INCREASES IN VOLUME BASED IN ASSUMPTION THAT ELASTICITY OF DCMAND EQUALS
Table 4. — Estimated Additions to 1988 Deficit — How Deficit Would Have Increased
if Average 1938 Prices Had Been Reduced — United States Steel Corporation and
Subsidiaries
. Percentage Re-
duction in Price
Estimated
Addition to
Deficit Assum-
ing Elasticity
of Demand for
Steel of 1
Estimated
Addition to
Deficit if No
Increase ip Vol-
ume Resulted
from Price
Reduction
Percentage Re-
duction in Price
Estimated
Addition to
Deficit Assum-
ing Elasticity
of Demand for
Steel of 1
Estimated
Addition to
Deficit if No
Increase In Vol-
ume Resulted
from Price
Reduction
1...
$3,900,000
7,900,000
12,000,000
16, 200, 000
20, 500. 000
24, 900, 000
29, 300, 000
33,900,000
38, 500, 000
43, 300, 000
$5,600,000
11,200,000
16, §00, 000
22,400; 000
28,000,000
33, 600, 000
39, 200, 000
44, 800, 000
50, 400, 000
56, 100, 000
11
48, 100, 000
53, 100, 000
58, 200, 000
63, 400, 000
68, 700, 000
74, 200, 000
79, 800, 000
85, 500, 000
91. 400. 000
97,400,000
61,700,000
67, 300, 000
72, 900, 000
78, 500, 000
84,100,000
89 700 000
2... .
12
3
13
4
14
5
15
6...:
16
7
17
95 300 000
8
18
100 900 000
9
106,500,000
112,100,000
10
GONOENTRATION OF ECONOMIC POWER
14091
The 1938 operating deficit, before Federal income tax, and the additional
losses that would have resulted from price reductions 'if volume increased to the
same relative extent that prices decreased, are illustrated in Chart 5.
Since the average 1938 prices represent the average results of prices in effect
both before and after the June 24, 1938, reduction in the price of steel products,
Chart 6 has been constructed to show the relationship between annual sales a,nd
revenues and annual costs at various levels of production if the prices prevailing
in the second half of 1938 had prevailed over the entire year J At such a price
Chart 5 ' .
ESTIMATED ADDITIONS TO 1938 DEFICIT
HOW DEFICIT WOULD HAVE INCREASED IF PRICES HAD BEEN REDUCED
AND VOLUME HAD INCREASED TO SAME RELATIVE EXTENT
U. S. STEEL CORPORATION AND SUBSIDIARIES
96 — 872
nt
ESTIMATED ADDITIONS TO
DEFICIT IF PRICES HAD BEEN
REDUCED AS IND.CATED
ACTUAL 1938 DEFICIT I I ■ I ■ I
■■■■■■mill
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
PERCENTAGE REDUCTION IN 1938 AVERAGE PRICE
level the break-even point for the United States Steel Corporation and its sub-
sidiaries would be at about ten and a half million of weighted tons, which is equiv-
alent to an operating rate of 50% to 55% of capacity, depending on the type of
products predominating. A 10% reduction in prices at this level would result
in sales and revenues indicated by the dashed line in Chart 6. The total sales
and revenues would then reach the amount of the total costs only if operations
were around 90% of capacity. If the break-even point were this high, the Cor-
poration's subsidiaries would have to operate at the impossible annual rate of
over 130% of capacity to earn a return, before income taxes, as modest as 5% on
their investment in tangible assets.
' The amount of sales and revenues per weighted ton if the prices prevailing in the second half of 1938
had- prevailed throughout the entire year has been estimated by reducing the sales per weighted ton propor-
tionately to the extent to which the selling value per weighted ton of rolled and finished steel products
shipped during the second half of 1938 was less than the average selling value of rolled and finished products
for the entire year.
124491— 41— pt. 2€
-33
14092
CONCENTRATION OF ECONOMIC POWER
The percentage increases in volume required to ofifset a given percentage de-
crease in the average price prevailing in the second half of 1938, as shown in Table
5 below, would be even greater than those needed at the average level of prices
for 1938 as a whole.
Table 5. — Increases in Volume Needed to Compensate for Various Decreases in
Snd Half, 1938, Prices Corn-pared to Probable Resulting Increases in Volume —
United States Steel Corporation and Subsidiaries
Percentage Reduc-
tion in Price
Percentage
Increase in
Volume
Needed
Probable
Percentage
Increase,
Assuming
Elasticity of 1
Percentage Reduc-
tion in Price
Percentage
Increase in
Volume
Needed
Probable
Percentage
Increase,
Assuming
Elasticity oil
4.0
8.4
13.1
18.3
24.0
30.2
37.2
44.8
53.5
1.0
2.0
3.1
4.2
5.3
6.4
7.5
8.7
9.9
10
63.1
74.1
86.7
101.3
118.3
342! 6
2984.4
11.1
2
11
12.4
3
12 —
13.6
4
13
14.9
5
14 -
16.3
g
15
17.7
7
20 .—
25.0
g
25
33.3
9
Chart 6
RELATIONSHIP BETWEEN SALES AND COSTS
EFFECT OF REDUCTION FROM 2nd HALF 1938 PRICES
U. S. STEEL CORPORATION AND SUBSIDIARIES
1200
1200
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
^ MILLIONS OF WEIGHTED TONS OF ALL
TONNAGE PRODUCTS SHIPPED
NOTE. COSTS ARE BASED ON 1927- 1938 DPERIENCE, ADJUSTED TO 193S CONOITIONS
CJONOENTRATION OF ECONOMIC POWER 14093
The divergence between the needed increase in volume and the maximum probable
increase to result from a reduction in the prices prevailing in the second half of
1938 is graphically portrayed on Chart 7.^
The inter-relationship between prices and sales, costs, and volume for the
United States Steel Corporation and its subsidiaries, then, is such that the increase
in volume required to offset a price decrease is far greater, than the price reduction
could be expected to stimulate. Hence, in times of recession the Corporation
and its subsidiaries, unhappily, do not have the alternative of lowering their
prices, moving their goods and employing their workers. Reduction of prices
beyond that necessary to meet competition and keep their shaje of the going
business, can result only in severely increasing the financial losses which are
incurred in such periods. Instead^ of prosperity, undiminished employment, and
.undiminished payrolls in the steel industry, only bankruptcy, unemployment and
dwindling payrolls would result from further price reduction.
» While the assumption of any elasticity of demand for steel greatipr than 1 is highly unrealistic, it Is inter-
esting to note that even if steel had an elasticity of 1.5 or 2, the percentage increase in volume, needed to
offset a price reduction would still greatly exceed the percentage increase in volame which would then result
from reducing prices:
Resulting Increase to Volume
Increase to Volume Needed to
Oflset:
Decrease in Price
If E'c3ticlty=
If Elasticity =
Decrease in
Average 1938
Steel Prices
Decrease to 2d
Half 1938
Steel Prices
1%
1.6%
3.1
4.7
6.3
8.0
17.1
27.6
39.8
54.0
70.8
.0%
6.3
8.6
11.1
23.5
38.4
56.3
77.8
1011
3.4%
7.0
10.9
15.1
19.6
48.8
96.7
190.3 .
463.6
5858.2
4.0%
2 " "
8.4
3
13.1
4
18.3
5
24.0
10
63.1
15
138.4
20
342.6
25
2984.4
30
(No increase
sufficient)
14Q94
OONCJENTRATION OF ECONOMIC POWER
Chakt 7
INCREASES IN VOLUME NEEDED TO COMPENSATE FOR
' VARIOUS DECREASES IN 2nd HALF 1938 PRICES
COMPARED TO PROBABLE RESULTING INCREASES IN VOLUME
U. S. STEEL CORPORATION AND SUBSIDIARIES
440 ; _
2 3 4 5 6 7 8 9 10 U 12 13 14 15 16 17 18 19 20 21
PERCENTAGE DECREASES IN PRICES
NOTE: PROBABLE RESULTING INCREASES IN VOLUME BASED IN ASSUMPTION THAT C'JVSTICITY OF DEMAND EQUALS I
'Exhibit No. 1418" is included in Hearings, Part 27, appendix, p. 1419.
CONCENTRATION OF ECONOMIC POWER 14095
Exhibit No. 2180
THE DISTRIBUTION OF STEEL TO MAJOR CONSUMING INDUSTRIES
This is an analysis prepared by the Special Economic Research Section of United
States Steel Corporation, composed of Messrs. Edward T. Dickinson, Jr., Ernest
M. Doblin, H. Gregg Lewis, Jacob L. Mosak, Mandal R. Segal, Dwight B.
Yntema and Miss Marion W. Worthing. The work of this group was under the
supervision of Theodore O. Yntema, Professor of Statistics, University of Chicago.
This analysis was written by Marion W. Worthing and has had the benefit of
suggestions from other members of the staflF. It is issued by United States Steel
Corporation.
October 30, 1939.
description o^ the problems
The table presented on pages 2 and 3 of this memorandum shows yearly esti-
mates of the tonnage of steel production destined for each of the major consuming
industries from 1923 to 1938. The estimates presented in this table have been
prepared because statistics covering the full distribution of either production or
shipment from year to year are not available.
Estimates of the amount of steel distributed to the major consuming industries
have been published every year since 1922 or 1923 by Iron Age and Steel maga--
zines. These estimates are based on annual reports by individual steel producers
of shipments of the major classes of steel products to each type of consumer.
Although each trade journal sends out its own requests for data to the steel pro-
ducers, the information asked for and received by each publication is similar.
The figures published by each of these magazines have been used in this study.
Because of variable coverage of firms from year to year, and because of ambiguities
and variations in classifications, the tabulations resulting from the combination of
the reports of the contributing steel companies do not represent accurately the
distribution of steel by consuming industries.
In an attempt to remedy some of the defects in the original shipment tables, a
series of adjustments have been made which will be presented in detail later in this
report but which may be described here briefly as follows:
1. Distribution percentages derived from the original shipment data were ap-
plied to hot-rolled steel -production in order to secure comparability of the totals
among the different years, i. e., the percentage distribution of the shipments of each
class of product to the major industries was applied to the production of the most
nearly comparable hot-rolled products.
2. Production indicated for jobbers, warehouses, and similar distributors was
reallocated to the appropriate consuming industries on the basis of the informed
opinion of a number of persons concerned with the sale of merchant products.
14090
OONOENTRATION OF FCONOAIIC TOWER
3~
ioot-coooa)r-0":it^air-as»0'«fQ
^ ^ ^ ^ ^* ^ r^ ,_r r^ ,-r r-T ,-H M Cf Cf r
5(Nt--OJ(NO'-'OOSCOCOt
S?3gg
oofeS<
5 05OO00O J
:^^ OsaiCOC^C^COOOr-HOOOiiOC
"^ ^^ CD <© r^ CO '^ c^ c^*" CO CO »o lo '
isslBSIISSsKis
Tj4-(tiiOCDcOt^t-CD-'
^:SS3f
« !
DOTS°iS«S2ot^S3
h CD O -^ C^ ^"" CO 1*" CD to CD M
>6503CTsosoio>coo>d05050iCiaac
CONCENTRATION OF ECONOMIC POWER
14097
3 s
3^
SSoSSSSS
5-cf-rci-iNmv"*"<
CT lO CD iO »0 UO CO -^ cf r-T CO IN CO « to* ■■
^ -•-— "--^-M-c
;§i3§iisl
iiglBiiliiililis
J(5cOCOt^-^»00>'-tCOC5»0»0
?S§K§S:^SSS
CO^Oi<NC^C»c3iOOC»OJC
^8§
Cos
li-s
® o ^
.9-,: "5
|2 g
■- S "a
-Si I
ill
11 i
W;'3 '^
ll I
Q ~ • • II:
:s§ I?
:§l is
I ^
14098 CONCENTRATION OF ECONOMIC POWER
SUMMARY AND EVALUATION OF RESULTS
The final series presented in the accompanying tables represent the annual
domestic hot-rolled steel production estimated to have been consumed eventually
by the major steel consuming industries, either as hot-rolled steel or in further
finished forms. The products destined for export are included in a separate
group in these tabulations but the consumption of imported iron and steel is not
included in the table.
The total estimated consumption of hot-rolled steel by industry groups is
shown as computed on the basis of both Iron Age and Steel data. It is impossible
to determine whether the results based upon Iron Age or those based on Steel are
more reliable. Because of the confidential nature of the data from reporting
companies, there is no chance to judge the accuracy of the methods used in
making either magazine's compilations. By comparing the movements of these
consumption series with other measures of activity in the various industries, it
is possible that some evaluation may be made of particular estimates. This
process, however, is diflicult, time-consuming, and subject to error.
Although the final series retain some of the deficiencies which were inherent
in the basic data, it is nevertheless reasonable to state that the adjusted figures
are in most respects more useful and more trustworthy than the original ones.
In the first place, by aUpcating total hot-roUed steel production to industry
classes on the basis of reported shipments, the resulting series become comparable
from year to year. Further, because the secondary distribution of steel does not
follow the same pattern as that typical of direct shipments from the mills to the
consumers, it is more realistic to make an approximate allocation of jobber
tonnages to industries that customarily buy from those sources than to ignore
such distribution. In many instances a comparison of the estimated distribution
calculated from figures based on Iron Age and Steel show consistent results.
Where the discrepancies between the figures are large the variation can be ac-
counted for in some instances by differences in industry classifications used by
the compilers of the original data.
Adjustment Procedures
NOTE ON THE DISTRIBUTION DATA PUBLISHED BY IRON lGE AND STEEL
The compilation of steel shipment data was started by both Iron Age and Steel
in "about 1922. Since it was not possible for the trade journals to get very com-
plete coverage in the earlier years, the figures for the first two or three years are
rather heavily weighted with data for the subsidiaries of United States Steel
Corporation. Even now, a few small companies do not report their shipments.
The coverage was improved rather rapidly but at no time has it been complete
nor has it been the same from year to year. Although the published tables are
often described as representing a specified percentage of the steel industry's
ingot producing or finishing capacity, it is not possible to use these percentagee
as bases for raising the tonnage figures to an assumed 100 percent because the
shipments of certain products were more completely reported than others. Until
1938, each publication was anxious to print the distribution table in its annual
statistical number, the first issue in January. Since shipment data for the
entire preceding calendar year could not be collected before this publication date,
the steel producers reported estimated shipments for December or for both
November and December. It is probable that these estimates, did not deviate
seriously in relative distribution by industries from the shipments actually going
to consumers. In 1937 and 1938 the tabulations were delayed so as to allow for
more complete reports. Every year, each publication has made it a practice to
publish some data for preceding years, presumably for purposes of comparison.
Uafortunately, the components of these figures and percentages for the various
yftars differ in many instances so thpt they are not strictly comparable. The only
figures used in making the present computations were the basic tables showing
tonnages both by product and by industry.
ADJUSTMENT OF DISTHlSUTION TO PRODUCTION DATA
Since the published shipment tables were not comparable in their original form
and did not cover the entire industry and since there are no data available on
total steel shiprrents, it seemed best to use the proportions indicated by the
shipments to allocate hot-rolled steel production to the ultimate users.' Aoord-
' Practically all the stoel products wblcB aro shipped may bo classified as hot-rolled products or «r« tke
result of the further finishing of hot-rolled products.
OONOENTRATION OF ECONOMIC POWER 14099
'agly, the percentage distribution by industry of each class of products for which
shipments were reported separately was applied to the appropriate hot-roiled
iron and steel production. The resulting series have the advantage of being
derived from totals which are comparable from year to year, which represent
practically the entire production in the country, and which contain no duplica-
tions. It is reasonable to expect that, to the extent that the various shipment
figures reported in different years are representative of all shipments, the per-
centages going to diflferent industries derived from the reported shipments should
be adequate for allocating production data to ultimate consuming classes. Pro-
duction and shipments for any given year are, of course, unlikely to be identical
but they are probably not very dififerent. The main discrepancies between them
arise from variations in the inventories of finished steel held by steel producers at
the end of each year. There are no figures available by which the condition of
these inventories can be ascertained, but it seems, from the opinion of various
persons concerned with such matters, that relatively little finished steel is cus-
tomarily stored and the amounts that are kept on hand as stock do not ordinarily
change radically from year to year.
INDUSTRY CLASSIFICATIONS
It is probable that the greatest errors inherent in the shipment data arise from
inconsistent industry classifications of consumers on the part of the reporting steel
companies. Some companies made careful reports of their shipments based on
standard systems of customer classification, but at least a few others make rather
inadequate ones. Certain companies do not appear to have had or do not now
have records suitable for accurately reporting shipments by industry groups.
Fortunately, data for these companies that have been reported most incompletely
have usually been segregated from the others in the published tables. The trade
journals have endeavored to check back on the reports and have asked for recals-
sifications when they have detected obvious mistakes, but it has been impossible
to eliminate all errors. Since the present adjustments have been made for much
broader industry groups that thoselshown in the original published tabulations,
it is hoped that many of the misclassifications indicated above do not affect them.
The combinations of industry subgroups into the larger classes were made with
the advice of the persons on the respective staffs of Iron Age and Steel responsible
for the tabulations. A list of the industry sub-groups is shown in Appendix B.
PRODUCT CLASSIFICATIONS
Relatively few difficulties arose from the product classification, primarily
because there has been little variation in' the product classes over the entire
period, except that in some of the earlier years merchant and concrete bars were
combined and galvanized sheets were included with all other sheets. In the
adjustment of each published product total, the most nearly comparable hot-
roUed production was carefully chosen and the selection of products for these
classes was checked with the editors of Iron Age and Steel. A list of the product
classes currently used by Iron Age and Steel and the group of hot-rolled products
related to each class in the computations are shown in Appendix C.
JOBBERS, WAREHOUSES, AND OTHER DISTRIBUTORS
From 15 to 20 percent of the finished steel shipped each year goes to jobbers,
warehouses, and other distributors * who, in turn, resell the steel to the same
types of consumers as those who take their shipments direct from the mill.
Because certain products, at least, are distributed through jobbers to the various
consuming industries in different proportions from those typical for direct salec,
it is desirable to reallocate the jobber shipments to consuming groups. In the
absence of any statistical basis, allocations have been made, product by product,
on the basis of advice from a number of people familiar with the uses. and distri-
bution channels of the various products. The methods employed in these adjust-
ments and reasons for their use are described below.
Since the proportion of rails and track accessories handled by jobbers is insig-
nificant, the actual tonnages reported in all years were, for convenience, assigned
to miscellaneous industries. For the most part, the shipment of plates, shapes,
merchant bars, black plate for tinning, strip, and all other sheets other than galva-
nized which are distributed by jobbers appear to go to much the same consuming
industries in about the same proportions as do the ishipments resulting from mill
* For simplicity all types of distributors will be referred to here as jobbers.
14100 GONOENTRATION OF ECONOMIC POWER
sales. Consequently, the jobber tonnages for these products were allocated to
the consuming industries in the same proportions as direct sales.
Three product groups, galvanized sheets, pipes and tubes, and wire products,
appear to have different distribution pattersn through jobbers than in direct
sales. Special inquiries were made concerning the jobber sales of these products
and the following methods of allocation were adopted, based on the information
collected :
1. Galvanized Sheets:
(a) 50% to the construction industry for roofing ventilating, etc.
(b) 10% to agriculture for bins, tanks, etc.
(c) 40% to miscellaneous industries, including sheet metal shops, air-
conditioning, etc.
2. Pipes and Tubes:
(a) Total production of line pipe and oil country goods was allocated
to the oil, gas, and water industry. Whenever the estimated
production of skelp and rounds necessary for conversion to line
pipe and oil country goods differed from the tonnage of skelp and
rounds allocated to the industry on the basis of the original figures,
the difference was adjusted by adding or subtracting the required
tonnage from the amount originally allocated to jobbers.
(b) The remaining jobber tonnage was split between mechanical tubing
and a combined tonnage of standard pipe, boiler tubes, and all
other pipe.
(c) Mechanical tubing was allocated 40% to automotive, 40% to ma-
chinery, and 20% to miscellaneous industries.
(d) Other pipe was distributed 50% to construction, 25% to agriculture,
and 25% to miscellaneous industries.
3. Wire Products were Allocated:
(a) 35% to agriculture.
(b) 40% to construction (including repairs).
(c) 25% to miscellaneous.
Appendix A. Miscellaneous Notes on Procedure
1. In 1938 both Iron Age and Steel included railroad buildings and bridges under
construction instead of under railroads as was formerly the case. In order to make
these classes reasonably consistent with previous years, transfers of plates, shapes,
and merchant bars were made to railroads from construction based on the pro-
portions of these products Used for railroad buildings and bridges in recent years.
2. All concrete bars handled through jobbers have been transferred to construc-
tion in the first adjustments.
3. For 1923, 1924 and 1925 shipments through jobbers reported by Iron Age
were allocated to industry classes by the compilers.
4. Miscellaneous industries should not be compared through all the years unless
furniture and furnishings, etc., are added.
5. Before 1933 the Iron Age did not report galvanized sheets separately. For
previous years 65% of the jobber tonnage of sheets was allocated as galvanized
sheets and the remainder as ordinary sheets.
6. All shipments of plates, shapes, bars and pipes originally allocated to the
container industry were transferred to construction.
Appendix B. Industry Sub-Groups Included in Major CoNSUMiNa Industry
Groups
Automotive — Automobiles, trucks, parts, etc. (tractors in some years).""
Railroads — Trackwork, cars and locomotives, parts, railroad buildings and bridges.
Agriculture — Implements, equipment, other farm uses, tractors in some years.
Construction— fabricators, building contractors, concrete reinforcing companies,
building hardware and trim companies, concrete bar jobbers, highways, boiler
and tank makers, power developments, containers made from heavy steel
products.
Shipbuilding — Ships, boats, barges.
Containers light — Made from light steel products, predominantly tin cans.
Machinery— Machinery, hand tools, electrical machinery and equipment.
Oil, gas and water — Oil, gas and water.
Mining, lumbering and quarrying — Mining, lumbering and quarrying.
Furniture and furnishings — Furniture and stove makers, domestic appliances,
refrigerators, office equipment.
CONCENTRATION OF ECONOMIC POWER 14101
Miscellaneous — Bolt, niit and rivei makers, forgers, pressed and formed metal
manufacturers.
Appendix C. Hot-Rolled Product Combinations Used in Adjusting Iron
Age and Steel Shipment Data
Rails — Heavy, light, girder, and high tee.
Track accessories — Long splice bars, tie plate bars, and cross ties.
Structural shapes — Heavy structural shapes and sheet piling.
Plates — Sheared and universal.
Bars — Merchant bars (carbon and alloy) and light shapes.
Concrete bars — Concrete bars, including those reroUed from old material.
Black plate — Black plate for tinning onlv (combined with strip and strips for
tinning in 1938).
Galvanized sheets — Galvanized sheets a I'l galvanized formed products.
Sheets — Hot-rolled sheets minus galvanized sheets and galvanized formed prod-
ucts plus all black plate other than black plate for tinning.
Strip — Strip, hoops, bands, and cotton ties.
Pipes and tube — Skelp and blanks or rounds for piercing.
Wire products — Wire rods.
All other — Rolled forging billets, blooms and billets for export,' car wheels and
other hot-rolled products.
Exhibit No. 2181
INDEXES OF MILL-NET YIELDS ON PRODUCTS SHIPPED BY UNITED
STATES STEEL CORPORATION SUBSIDIARIES
This is an analysis prepared by the Special Economic Research Section of
United States Steel Corporation, composed of Messrs. Edward Tt Dickinson, Jr.,
Ernest M. Doblin, H. Gregg Lewis, Jacob L. Mosak, Mandal R. Segal, Dwight
B. Yntema and Miss Marion W. Worthing. The work of this group was under
the supervision of Theodore O. Yntema, Professor of Statistics, University of
Chicago. This analysis was written by Marion W. Worthing and has had the
benefit of suggestions from other members of the staff. It is issued by United
States Steel Corporation.
November 1, 1939.
Purpose
The purpose of this memorandum is to describe the construction of several
index numbers which have been prepared from mill-net yield ^ figures for the
major steel products shipped to the domestic market by subsidiary companies of
the United States Steel Corporation.^ These index numbers have been built
from different combinations of mill-net yields — the combinations being adapted
to the type of analysis for which each was designed. The first section of the
memorandum is devoted to a general description of the methods used in con-
structing the indexes and the reasons for their preparation. In the appendix
material, the handling of special problems which arose during the computations
is discussed and technical procedures are outlined.
' The mill-net yield is the amount of money actually received by the steel company for its products,
i. e., the delivered price minus the freight. v
» Excluding in most instances shipments of the Columbia Steel Company.
14102
CONCENTRATION OF ECONOMIC TOWBR
The Indexes
In the following paragraphs the indexes which have been constructed are
described briefly.
1. GENERAL YEARLY INDEXES
The over-all yearly index was based upon mill-net yields for nearly all the
major steel products of the United States Steel Corporation subsidiaries. These
major products were then divided into two sub-groups — light and heavy * — and
a separate index was computed from yields for products in each group. Each
of these three yearly indexes was computed by two somewhat different methods.*
As a result, six separate yearly indexes were prepared which were based on yield
data from all the major steel producing subsidiaries. They are shown in Table 1.
2. GENERAL MONTHLY INDEXES
Three indexes were computed on a monthly basis: (a) AU products; (b) Heavy
products; (c) Light products. These indexes are shown in Tables 2, 3, and 4.
The group of indexes referred to here and in paragraph 1 above might be said to
represent United States Steel Corporation subsidiaries' mill-net yields in general.
To the extent that fluctuations in mill-net yields are representative of fluctuations
in delivered prices, these yield series indicate changes in^rices.
3. IRON AOE INDEX
In Table 5 the Iron Age composite price of finished steel has been reduced to
index form for purposes of comparison.
Table 1. — Comparison of Annual Mill-Net Yield Indexes for Mills of United States
Steel Corporation Subsidiaries Computed from Yields at Separate Mills and Com-
puted from Average Yields at All Mills, 1920-1938
[1926=100]
All Products
Heavy Products
Light Products
Year
Based on
yields at
separate
miUs
Based on
average
yields for
all mills
Based on
yields at
separate
mills
Based on
average
yields for
all mills
Based on
yields at
separate
mills
Based on
average
yields for
all mills
1920
125.4
107.9
86.0
102.7
108.1
101.1
100.0
96.5
93.6
94.9
8210
79.4
77.2
89.5
91.2
89.1
98.8
».8
124.8
107.4
85.7
101.9
107.9
10L2
100.0
96.4
93.2
94.4
87.8
8L2
78.8
76.6
89.0
90.8
99! 6
99.7
123.6
105.8
82.6
102.6
108.4
101.1
100.0
97.1
95.0
96.4
89.2
82.8
81.5
80.7
90.0
92.2
90.2
101.1
102.2
122.7
105.2
82.2
10L6
108.2
101.3
100.0
96.9
94.3
95.9
88.7
82.0
80.8
80.0
89.6
9L8
89.6
103.0
103.6
129.2
112.2
92.9
102.9
107.6
101.1
100.0
95.7
91.9
93.1
87.3
81.1
76.9
72.8
88.8
90.1
87.9
96.0
96.9
129.2
1921
112.2
1922
92.9
1923
102.9
1924
107.5
1925
101.1
1926
100.0
1927
95.8
92.1
1929... .
92.8
1930
86.8
1931
80.5
1932
76.3
1933
72.7
1934
88.6
1935
89.7
1936
87.3
1037
95.4
1938
95.3
Source: All indexes computed from mill-net yields of major steel products shipped I
by mills of subsidiary companies of the United States Steel Corporation.
• The major products falling into the "heavy" classification are: semi-flnished goods,
rails, and pipe. "Light" products Include sheet, strip, tin plate, and wire products.
< A description of these methods appears in the Appendix.
the domestic market
plates, shapes, bars.
I
CONCENTRATION OF ECONOMIC TOWER
14103
Table 2. — Monthly Indexes of Mill-Net Yields for Steel Products Shipped to the
Domestic Market by Mills of United States Steel Corporation Subsidiaries, 1912-
19S9
[Average month of 1926=100]
Year
Jan.
Feb.
March
April
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
1912
57.9
57.3
57.5
57.6
58.1
68.4
58.3
58.9
59.6
60.8
60.9
61.9
1913
63.3
64.3
65.2
65.7
66.0
66.4
66.2
65.9
65.3
64.2
62.9
1914
61.2
60.8
61.2
61.1
60.6
59.9
59.0
59.2
59.4
59.6
69.5
58.8
1916
58.2
67.3
58.0
68.7
58.1
71.0
58.8
74.2
59.1
77.3
59.6
79.2
60.1
82.3
60.4
84.0
61.0
86.0
62.1
63.1
90.8
64.5
1916
93.6
1917
98.8
103.5
107.3
111.9
115.4
118.7
123.3
127.5
131.5
134.4
134.2
132.8
1918
139.3
137.6
136.1
135.8
137.9
138.2
139.4
139.6
140.6
142.1
142.6
140.2
1919
136.1
134.7
132.1
125.4
123.9
122.0
121.9
121.5
121.2
121.5
120.8
121.6
1920..
122.2
122.2
122.3
123.5
124.0
123.8
124.4
124.8
125.1
126.1
125.8
126.6
1921
126.1
85.4
126.5
84.7
126.6
82.8
118.7
82.7
114.6
82.9
113.6
83.6
106.1
84.2
98.7
85.1
94.6
86.2
90.4
87.3
86.9
1922
90.0
1923
91.9
112.6
101.9
94.1
112.7
102.3
96.0
113.0
102.9
98.4
112.6
103.0
100. 4
111.0
102.3
101.9
108.3
100.7
103.6
107.3
100.2
105.1
105.3
100.3
106.2
104.2
99.7
107.7
102.0
99.7
110.4
101.4
99.6
110.6
1924
101.2
1925
99.6
1926
99.8
100.0
99.6
100.2
100.1
99.8
99.8
99.5
99.9
99.6
99.9
99.8
1927
98.7
98.0
97.0
96.6
95 9
96.4
96.3
96.3
95.9
94.9
95.2
93.5
1928
93.4
93.4
93.3
93.4
94.3
93.8
92.9
92.4
92.7
92.9
93.9
93.7
1929
94.2
94.2
93.9
94.3
94.2
94.3
95.0
96.4
94.6
94 3
94.3
94.0
1930
92.4
91.6
91.2
89.9
88.9
88.0
86.6
86.0
85.0
83.3
82.0
1931
82.2
83.2
82.3
81.8
81.4
80.4
79.9
79.8
81.9
80.0
81.3
80.2
1932
78.6
77.0
79.1
76.0
79.3
76.6
78.7
75.0
77.7
74.5
79.2
74.6
79.5
73.6
79.3
76.0
79.0
77.2
78.8
79.4
78.2
82.6
77.9
1933
83.6
1934
87.1
88.1
87.4
87.1
87.4
91.8
92.9
91 9
93 3
92.5
89.9
1935
92.1
92.0
91.9
91.9
92.0
91.2
90.5
90.8
90.0
89.6
88.8
89.6
1936
89.0
89.1
87.6
86.4
87.1
88.2
87.3
88.8
89.6
90.0
90.6
1937..
91.4
92.3
93.3
95.8
98.0
99.8
101.6
101.9
103.4
105.7
104.8
105.3
1938
105.4
93.2
105.1
94.1
105.9
95.8
104.3
95.1
104.4
94.8
102.7
92.1
97.9
91.4
96.2
95.9
93.7
91.6
92.2
1939
Source: Computed from mill-net yields for the major steel products shipped to the domestic market by
mills of United States Steel Corporation subsidiaries. Complete notes are presented In the text of the
memorandum.
Table 3. — Monthly Indexes of Mill-Net Yields for Heavy Steel Products Shipped
to the Domestic Market by Mills of United States Steel Corporation Subsidiaries,
1912-19S9
[Average month of 1926=100]
Year
Jan.
Feb.
March
April
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
1912
58.2
64.1
67.5
65.1
57.6
66.0
67.7
60 1
58.2
67.0
58.9
67.1
58.6
66.9
59.3
67.1
60.3
67.2
61.6
66.6
61.7
65.5
62.7
1913
63.8
1914
62.2
61.8
62.3
61.6
60.9
60.3
59.6
59.9
59.8
59.7
59.6
59.1
1915
69.0
58.4
58. 3
58.8
00.2
59.6
60.0
60.3
60.9
62.6
63.8
65.3
1916
68.5
69.6
71.8
75.1
78.4
80.6
84.0
85.4
87.9
90.1
93.0
96.4
1917
102.0
105.1
109.3
114.4
118.3
120.6
124.0
128.4
132.9
134.9
134.9
136.3
1918
140.1
137.5
135.0
137.3
138.1
i.''7. 8
140.2
140.9
140.3
141.8
141.2
141.0
1919
137.4
136.9
132.5
125.3
122.8
l.iO. 7
120.6
119.7
119.0
118.4
117.7
119.0
1920
120.8
121.1
120.6
121.4
122.0
121.8
122.0
122.7
122.8
123.6
123.3
123.2
1921.....
123.1
123.5
123.8
114.6
111.4
110.7
104.7
97.0
93.1
87.9
85.3
83.0
1922
81.6
81.3
78. 9
78.9
79.0
79.5
80.5
81.3
82.4
84.7
85.9
87.5
1923
112! 9
92.6
113.5
95.1
114.1
113^6
100.8
111.9
102.6
108.8
104.0
107.8
105.1
105.8
106.1
104.6
107.5
101.7
109.7
100.7
110.7
1924
100.1
1925...
101.7
102.1
102.3
102.6
102.3
101.1
100.8
101.2
100.5
100.4
100.0
100.2
1926.
99.9
99.6
99.2
99.8
99.7
99.8
100.4
100.6
100.4
100.2
100.3
99.3
1927
-99.3
99.4
98.5
97.7
97.3
97.3
96.7
96.4
95.3
93.5
93.9
92.4
1928
93 9
94.1
93.6
94.4
95.5
9.'.0
93.7
93.5
93.8
94.1
95.3
95.0
1929
96.9
95.9
94.8
96.6
95.3
95.5
96.2
97.2
95.8
96.0
95.6
94.7
1930
94.9
94.3
93.4
91.0
90.5
86.8
85.9
84.9
83.7
83.4
82.3
1931
83.3
84.4
83.3
82.9
82.3
80.8
80.2
80.5
82.0
80.1
81.9
79.9
1932
79.1
80.8
80.8
81.1
79.3
81.0
81.6
81.9
81.2
81.9
81.5
81.1
1933
80.8
79.8
80.3
78.6
77.9
78.1
76.1
•'3.0
81.4
83.4
85.4
85.8-
1934
86.6
87.7
86.4
89.2
92.4
94.1
9,3..?
94.3
94.0
91.3
1935
93.2
93 3
93.0
93.0
93.3
92.3
91.8
91.7
90.9
90.3
91.1
1936
90.1
90.9
90.1
87.5
87.2
89.7
87.9
88.6
89.9
90.5
90.9
92.1
1937
93.4
63.9
94.8
97.9
99.9
101.5
103.9
104.7
106.2
108.0
106.9
106.9
1938
107.1
107.1
109.0
106.6
106.0
104.5
99.4
97.9
97.4
95.7
96.0
95.6
1939
95.1
96.7
97.5
95.5
96.7
94.9
93.7
Source: Comr
Uted fn
)m mill
net yiel
(Is forth
e major
heavy s
teel pro
lucts sh
ippedt
) the do
mestic n
aarket
by mills of United States Steel Corporation subsidiaries. Complete notes are presented in the text of this
memorandum.
14104
CONCENTRATION OF T<X)ONOMIC POWER
Table 4.— Monthly Indexes of Mill-Net Yields for Light Steel Products Shipped to
the Domestic Market by Mills of United States Steel Corporation Subsidiaries,
1912-19S9
[Average month of 1926 = 100)
Year
Jan.
Feb. March April May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
1912.
1913.
1914.
1915.
1916.
1917.
1918-
1919.
1920.
1921.
1922.
1923.
1924.
1925.
1927^
1928.
1929.
1930.
1931.
1932.
1933.
1934.
1935-
57.9
62.1
59.8
57.3
138.6
134.7
125.7
132.7
112! 5
103.1
100.2
98.0
92.8
92.4
89.4
80.9
78.1
72.4
87.7
91.1
87.8
89.3
103.5
91.0
57.3
07.5
101.4
138.7
133.2
125.0
133.2
92.1
97.6
111.7
103»5
101.5
96.4
92.5
92.2
8l!7
77.1
71.6
87.8
00.7
87.1
90.5
102.9
91.0
57.9
64.1
59.6
57.9
07.8
104.3
139.0
132.2
126.5
133.0
91.1
98.5
111.3
104.8
101.1
95.2
93.1
92.8
88.8
81.2
77.5
72.3
87.2
90.8
84.7
91.4
102.4
93.9
73.1
107.6
138.8
126.6
128.2
127.9
111.0
194.5
101.6
95.5
92.2
92.9
75.7
70.6
88.1
90.7
85.2
93.2
101.7
94.8
68.2
64.7
60.1
59.2
75.9
110.0
138.6
126.8
128.8
121.9
91.3
100.4
92.9
87.1
80.5
75.9
70.6
87.8
90.7
87.1
95.8
102.6
92.7
.';7.9
65.2
59.5
00.0
77.0
115.7
140.0
125.0
128.5
120.2
92.1
101.1
108.0
100.5
100.3
95.3
92.5
93.1
86.8
80.0
77.1
70.3
86.5
90.1
86.6
98.0
100.8
88.8
58.4
65.3
58.3
60.5
79.8
122.5
92.2
103.3
106.7
99.3
99.3
95.9
92.0
93.5
86.5
79.8
77.1
70.2
91.2
89.3
86.9
98.9
96.2
88.7
58.6
64.5
58.3
60.9
81.8
126.6
137.6
125.7
129.9
102,7
105.0
99.0
98.2
96.2
91.1
93.4
86.3
79.2
76.1
71.5
91.5
89.8
87.8
98.7
94.3
59.0
63.8
59.1
61.3
83.1
129.8
142.2
120.2
130.5
98.1
94.2
107.0
104.0
98.8
99.3
96.9
91.5
93.1
85.2
81.8
76.5
72.0'
90.4
87^8
100.1
94.2-
61.5
85. 1
134,5
143,8
128.5
132.0
96.2
93.3
108.6
103.4
99,0
99.0
96.8
91.6
92.5
75.0
74,6
92.1
88.7
.88,5
103.0
91.4
59.8
62.2
59.7
62.0
87.2
133.8
146,2
127,6
131.5
94.3
94.5
112.4
103.3
99.0
99.6
96.9
92.1
92.9
83.2
80.8
74.4
79.3
90.7
88.3
102! 4
86.3
127.1
139.4
127.4
133.6
95.0
95.6
111.2
103.7
99.3
100.7
95.1
.92.3
93.3
81.9
80.6
74.1
80.8
87^7
103^5
88.3
' Source: Computed from mill-net yields for the major light steel products sTiipped to the domestic market
by mills of United States Steel Corporation subsidiaries. Complete notes are presented in the text of this
memorandum.
Table 5.-
■The Iron Age Composite Price of Finished Steel as an Index, 1912-1989
[Average month of 1926=100]
Feb. March April May June July Aug. Sept. Oct,
1912.
1913.
1914.
1915.
1916.
1917.
1918.
1919-
1920.
1921.
1922.
1923.
1924.
1925.
1926.
1927.
1928.
1929.
1930.
1932.
1933.
1934.
1935.
1936.
1937.
61.5
76.5
62.7
14612
153.3
145.6
162.7
129.1
84.6
103.7
118,6
104,1
100,3
98.5
92.3
94,7
93,4
85.4
81.2
81.4
84.0
97.1
lOS. 5
98.7
95.2
151.2
153.3
145.6
152.7
123.1
82.2
109.5
117.5
104.6
100.0
95.9
93.7
94.7
92.8
85.6
97.1
108.5
98.7
60.2
77.1
63,6
61,0
105,7
161,5
153,3
141,8
164.2
115,9
82,2
115,5
115.0
104.7
100.3
86.5
94.1
94.7
92.7
61.9
77.3
62.5
62,1
112,8
177.5
153,3
130,9
170,9
114,0
85,1
119,9
112,0
102,0
100,1
96.2
94.0
96.0
90.6
85,3
82.4
78.5
85.9
74.6
61.5
61.9
118.8
197.1
153,3
130,5
168,0
114,5
87.7
119,7
100.2
100.6
99.5
95.7
93.0
96.2
9,b.'i
82.4
62.4
116.2
216.2
153.3
130.5
166.3
109.0
90.1
118.6
108.0
95.5
93.0
96.6
82.4
78.6
91.5
65.0
72.0
60.5
63.5
114.0
102.2
91,5
118,6
105.8
87.2
84.4
82.7
81.1
90I3
108.5
39,4
96.6
67.1
70,2
62,5
65,3
115,9
226.7
153,3
130.5
172.7
96.0
87.0
118.6
103.5
100.0
95.4
93.3
95,6
86,3
84,0
82.7
90.3
108.5
99.4
69.5
68.7
63.5
67.3
119,4
218,1
153,3
129,8
170,8
93,0
101,6
118.6
101.8
97.7
100.0
95,0
93.3
95.4
85,9
84,0
82,7
81,6
90.5
108.5
99.0
71.7
67.3
62.5
70.6
123.4
149.8
153,3
131.8
162.5
91.3
103.6
118.6
101.0
98.3
100.0
93,7
94.9
85.6
84.0
89.1
91.4
108.5
97.4
73.4
65.0
60.3
76.4
130.5
148.8
153.3
133,2
154.2
88.8
103,3
118.6
101.6
99.8
100.0
92.1
94.2
74.5
63.2
59.0
83.8
141.6
148.6
149,5
134,3
133.1
86.4
102.4
118.0
103.6
100.6
100.0
92.1
94.5
95.2
84.8
82.5
82,0
84,0
89,1
95.0
108.5
98.7
Source: Computed from data published in the/ron Age, January
in the weekly issues through August, 1939.
1939, p. 199, and from data published
CONCENTRATION OF ECONOMIC TOWER 14105
General Statistical Methods
At this stage it is in point to discuss the reasons for reducing mill-net yield data
to index number form and also the methods by which this may be accomplished.
An index number may be defined as a device for summarizing the fluctuations of
a number of individual groups or series of figures. The need for such a summary is
indicated in the following discussion. Although mill-net yields from the sale of
steel plates, to take an example at random, are important among yields for steel
products, the fluctuations in yields for steel plates could not safely be assumed to
-be indicative of the changes in mill-net yields fgr steel products in general. The
same thing may be said of the yields for any other single steel product. Clearly,
some kind of summarization is necessary because a single yield series is not suffi-
ciently comprehensive to be representative of a broad classification of goods.
Therefore, in order to have some way of measuring the relative changes in steel
mill-net yields in general, it is necessary to analyze either all the individual mill-
net yields for all steel products, i. e., plates, structural shapes, rails, sheets, tin
plate, etc., or, at least, the mill-net yields for a representative sample of all steel
products. But if each of the many groups of yields were to be considered singly,
it would be impossible to arrive at any definite conclusions about their movements
taken as a whole. Therefore, such groups of figures or series are often combined or
averaged into a single series of figujes so as to obtain a composite, month by
month or year by yeat, as the case may be, of their movements as a group.
This combined or average series of figures is the basis for an index nuhaber.
Since the actual numerical value of any item in a combined series has little or no
meaning in itself, the members of that series are usually expressed as relatives.
That is, one item of the combined series, or an average of several, is selected as a
base,gfid its relative is said to be equal to 100; then the other members of the series
are expressed as a percent of the base. When the same base is used for an entire
series, the series is said to have been related to a fixed base.
Clearly, the mill-net yields of individual products difi'er in importance depending
upon such criteria as the amounts of the different kinds of products usually shipped
or upon their value. In averaging the yields used in an index, therefore, it is
necessary that each figure -be multiplied by an appropriate weighting factor, for
in that way the relative effect of the movements of any one of the component
.series is limited to a predetermined proportion of the whole. .
Index numbers are ordinarily described by the method used in making the
combinations. The actual methods of construction are various, depending on
the kind of average used and the way in which the magnitude of group change
is expressed. All of the mill-net yield indexes of which the construction is
described in this memorandum are of the general type known as ratios of aggre-
gates. This means that the index number for any month or year was obtained
by computing the sum of the products ' of individual mill-net yields and the
corresponding selected weighting factors. The result of this process, called an
aggregate, was then divided by a similar aggregate for the selected base period
in order to express the aggregate as a relative.^
It is usually undesirable to use the same set of weights throughout a long
period. With the passing of time, certain products become less important and
others more important. Extreme cases of this kind occur whenever a product
ceases to be produced or a new one is marketed and becomes important. Further-
more, allowances must be made for changes in the form in which the data are
reported. Most of these difficulties can be overcome by using several weighting
periods and then chaining the aggregates for these periods, into a single series.
A weighting period is simply a group of years or months for which the same set
of weights is used in computing the aggregates. A different group of weights is
determined for each weighting period. Then, in order to make a continuous
series, the aggregates for each weighting period are chained together. Chaining
is accomplished by computing aggregates for one overlapping month or year.
That is, if July, 1926, were selected as the point for chaining, an aggregate for
July would be computed as of the original weighting period and another aggregate
as of the second one. The original series could then be extended by computing
1 The word "product" is used here in the mathematical sense. If Yi, Yj, Y3, etc., are yields for a series
of products and Wi, W2, Wj, etc., are their respective weighting factors,, the sum of the paired products
would be Yi, multiplied by Wi. plus Yj multiplied by Wj, plus (Y3W3), etc.
» The mathematical formula for this index is:
S (WkYi)
2 (WkYo)
in which Wk is the weighting factor, Yo the mill-net yield for the base period, and Yi, the mill-net yield
for the given period.
14106 CONCENTRATION OF ECONOMIC POWEfR
the next item — the figure for August — which would be larger or smaller than
the July aggregate in the original series in the same proportion that the August
figure was larger than the July figure in the second series.
When it becomes necessary to break these series into weighting periods, the
breaks should come at points where adjustments can be made for as many as
possible of the changes which are bound to creep into the series. Any period
may be used as a base for computing relatives in this kind of an index. The
actual selection of a particular base depends on what period the maker of the
index wishes to emphasize or to consider as normal. The base for anj index
may be easily shifted to another year or month by dividing all the present index
numbers by the index number for the selected year or month.
Data From Which the Indexes Were Computed
1. mill-net yields
The basic materials for these indexes were monthly figures on the tonnages shipped
and the mill-net selling values of more than 75 classes of steel products.^ These
tonnages and amounts were used to compute average mill-net yields. Roughly
75 per cent of the total tonnage of steel shipped each year by United States
Steel Corporation subsidiaries was covered by these figures. Reports from the
Carnegie-Illinois Steel Corporation and the National Tube Company were tabu-
lated for individual mills. The American Steel and Wire Company, the American
Sheet and Tin Plate Company, and the Tennessee Coal, Iron and Railroad
Company, on the other hand, reported on a company basis only. Shipments
of the Pacific Coast subsidiary were not included in the computations.
The classes of products which could be used were limited by the form in which
records were originally reported and subsequently kept, on file by the subsidiary
companies. Since the use of all the yield data available would have entailed
a great amou' it of clerical labor, only the most important products, from a tonnage
standpoint, ; nd those products made at the most important mills were used.
Practically £.J aUoy steels have been excluded from these figures. Details con-
cerning the use of the mill-net yield data in the indexes are included in the Ap-
pendix,
2. DATA USED IN PREPARING WEIGHTS
All of the weights were based on tonnage figures reported in the United States
Steel Corporation's "Annual Summaries of Domestic Shipments" in which the
total tonnages of the various steel products shipped to the domestic market by
all United States Steel Corporation subsidiaries are shown. The classes of
products reported therein differ in several respects from the classes of pi'oducts
by which the monthly mill-net yield data were reported. In the "Annual Sum-
maries of Domestic Shipments," all steel products are accounted for, but the
actual product classes which are used vary considerably from year to year.
Although more kinds of products are represented in the annual summaries, the
monthly data were usually reported only for certain important subclassifications,
the makeup of which tended to be relatively homogeneous. The mechanics
involved in computing the weights used in the mill-net yield indexes are described
in the. Appendix.
Appendix
Notes and explanations relating to special statistical procedures or problems
involved in the construction of the indexes are presented in this section of the
report.
WEIGHT computations
Although the products for which mill-net yields were available represent only
part of the tonnage actually shipped, the weights used in constructing the indexes
were related to almost all of the products shipped. That is, the weights were based
not only upon the tonnage shipped of the particular products for which mill-net
yields were available, but also upon the tonnages of the other products for which
yields were not available in detail. The reason for using all these products in the
weighting computations is that it is desirable to give a single component series as
much weight as it deserves from a_ proportional view point. Therefore, if the
movements of a given series are not only representative of the particular product,
but are also reasonably representative of the movements of another series, it may
be used to represent the second series in instances where no yields are obtainable.
« Data for the 76 product classes were not, however, all available at any one time.
CONCENTRATION OF ECONOMIC POWER 14107
Furthermore, it should be noted that the more homogeneous are the products
included in a single series, the more likely it is that the fluctuations in the miU-net
yields are genuine and do not result in part from varying quantities of subclassifi-
cations of goods of different qualities.
Several methods were used' to determine what products might be most suitably
included in a single weight group. The yearly movements of the series for which
mill-net yields were reported in the annual shipment summaries were compared
with each other in some instances. Whenever the movements of several series
were found to be similar, shipments of these products were combined for weighting
purposes. In other cases, knowledge of the product or the way it was customarily
priced determined into what weighting group its influence should be thrown.
BELECTION OF WEIGHTING PERIODS
The yearly indexes prepared in this study were chained at 1926, and 1936, so
that three weighting periods were used 1920-1926, 1926-1936, 193&-1938. The
weighting periods used for the monthly series were as follows:
January, 1912— August, 1920
August, 1920— July, 1926
July, 1926— July, 1936
July, 1936— June, 1938
June, 1938— July, 1939
The monthly series were chained in the same years as the yearly series so that
their periods would be similar. Before the actual month was selected at which
the shift was to be made, monthly aggregates for both series were computed. The
month selected for chaining in each case, e. g., Jul^ 1926, was one for which the
difference between the aggregates of the two series seemed to be about tb*^ eame^as
the average for the year.
As a base period for all of these series the aggregate for 1926, or for the average
month of 1926, was chosen for several reasons. The year 1926 may be described
as one of relatively normal business conditions, i. e., neither a boom nor a depres-
sion was in progress. Mill-net yield levels were relatively stable throughout the
year. Furthermore, 1926 is often used as the liase for year index numbers.
The weights used for the period 1912-1920 inclusive were computed on the ba^is
of shipments during the year 1920. Weights for the other periods were based on
average shipments for several years. For the period 1920-1926 the weights were
based on 1924, 1925, and 1926 shipments, and those from 1926 through 1936 on
average shipments for 1934, 1935, 1936, and 1937. Redistribution of the latter
weights to different product classes were made for the 1936-1938 and 1938-1939
periods.
Weights for the 1926-1936 period were based on shipments in the years at the
end of the veighting period so that the index would be weighted by a combination
of products typical of recent years.' The same principle was applied to the 1920-
1926 weights. The years 1924-1926 and 1934-1937 were averaged in each case
so that the weights would not be greatly influenced by variations in shipments
caused by cyclical changes in the volume of business. The selection of 1920 ship-
ments for the purpose of weighting the 1912-1920 period was in accordance with
these procedures. The break was made in the series in 1936 because so many
new product classifications became available at about that time. It seemed, how-
ever, when it came to constructing weights for the 1936-1938 or 1938-1939 periods
that the weights computed on the basis of 1934-1937 shipments could not be im-
proved upon. Therefore, the weights for product groups derived from shir^ments
for 1934, 1935, 1936, and 1937 were merely redistributed to the larger nudiber of
classifications available in the years 1936-1938. In order to use data for 1939, it
was necessary to shift the weights again because of the change in the form in which
the data were reported.
COMPUTATION OF TWO GENERAL INDEXEt
In an early section of this memorandum it was noted that the general yearly
indexes had been computed by two slightly different processes. A description
of the methods and reasons for their use follows:
' Weighting by quantities sold in years near the end of each weighting period probably results in a slightly
lower index than weighting by quantities sold in years near, the beginning of each weighting period. This
small downward bias tends to offset the small upward bias Inherent in some of the mill-net series due to the
gradual introduction and growth of higher priced specialty steels In some of the classes of products. This
upward bias was kept ^ low as possible by using only standard products and excluding specialty producti
from the series used in the index.
124491- 41— pt. 26 34
14108 CONCENTRATION OF BOONORIIC POWRR
In computing the first general yearly index, separate mill-net yields were used
for each product at each mill in so far as the data were available in that form.
For the products which were reported only on a company basis, the average mill-
net yield for the company as a whole had to be used. The weight arrived at for
each product was allocated among the mills according to the importance of each
miU.
Theoretically, the same plan could have been used in the computation of the
monthly indexes, but some practical difficulties interfered. In the first place,
some of the monthly series were incomplete, i. e., occasionally no public shipments
were made during a month or series of months. Such lapses in the data were
naturally frequent among products which are primarily shipped to oth6r plants
of the company for further finishing. Also, plants would occasionally be shut
down for repair. Gaps in the series could not be ignored in the proces.s of comput-
ing the index because the effect of the omission of the amount represented by the
mill-net yield multiplied by the appropriate weight would usually be sufficient
to cause an unreal movement in the final index. If such breaks in the data were
relatively infrequent, it was usually possible to substitute a reasonable figure in the
blank, but when the number of months in which no shipments occurred became
frequent, there was less justification for substitution and the computation pro-
cedure had to be altered.
At the same time, another difficulty arose with respect to the mill-net yield
data. So far as it was possible, the individual series were confined to homogeneous
product groupings, but in the case of steel product groups homogeneity is only a
relative term. Each separate shipment, even though nominally related to the
same base price, is subject to different additions and deductions. When the total
shipments for any month are very small, the yield for these shipments will not
always be representative of the average either for that mill or that general class of
prv,ducts. A substantial variation in the yield for one month brought about by
a small tonnage shipped at an abnormally high or low yield can, therefore, cause
unwarranted variations in the index.
Because of these two major difficulties which arise from the use of the monthly
data at separate mills, it seemed advisable to combine the tonnage figures and the
selling values at all the mills for each product and then compute the average
mill-net yeild for each product. The monthly indexes from 1920 on were com-
puted fr^m average mill-net yields for all reporting mills.
The general yearly indexes were Qomputed by both of the methods just described.
A comparison of the indexes computed .by these methods reveals relatively insig-
nificant differences (see Table 1).
THE 1912-1920 MONTHLY INDEXES
The segments of the monthly indexes covering the period January, 1912, to
August, 1920, were computed by a modified procedure. The statistical methods
employed were identical with those throughout the rest of the series, but
the data on which the computations were based were for selected products and
selected mills. Tiiat is, mill-net yields for an abbreviated list of representative
products were weighted on the basis of total shipments for the year 1920. Dis-
cussion with respect to combinations of data for groups of products or groups of
mills is not applicable, of course, to the 1912-1920 figures.
ADJUSTMENTS OF PRODUCT CLASSIFICATIONS
A further adjustment had to be made in some of the miU-net yields in order to
make the data comparable. As it has been stated previously, even a relatively
homogeneous product classification may be composed of sub-groups of products
some of which could reasonably be reported separately. The result of this situa-
tion is that changes in product classifications have often occurred which did not
actually involve the addition or deletion of a particular type of product from those
shipped. Thus it is, to use a specific example, that the classification "bars,
rounds, etc." for a number of years might include, "deformed concrete bars",
"bars, rounds, special quality", and "seamless tube rounds" in addition to the
products usually described collectively as "bars and rounds." Then the method
of reporting shipments would be changed and each of these sub-classifications
might be recorded separately. It is plain that after the separation "bars and
rounds" would no longer be the same class of products. Therefore, within a single
weighting period comparability in the product class for "bars and rounds" was
maintained by combining the data for these sub-classifications with the main
classification; definite splits in classes wore made only with the adoption of a new
weighting period.
CONCENTRATION OF ECONOMIC POWER 14109
SUBSTITUTED FIGURES
Even when yearly data were used for computing index numbers, it was necessary
in a few instances to fill in a yield for one mill in a blank year. The usual pro-
cedure for estimating these yields was to assume that the yield in a given year at
that mill would be related to the yield in an adjacent year in the same proportion
that similar yields for the same product in a nearby mill were related.
The records of shipments kept at the mills contain deductions for returned goods.
In some cases, the entire returned tonnage was subtracted from the tonnage
shipped, but only part of the value of the returned shipment was deducted.
'Ordinarily, the tonnage involved in irregularities of this sort would be relatively
small and the effect on the mill-net yield would consequently be insignificant.
The result of this practice in the case of coke tin plate, however, was to cause
an abnormally high mill-net yield for December, 1938. In order to avoid this
variation, a mill-net yield was computed for that month excluding the effect of
returned goods.
EXHIBFT No. 2182
IMPROVED QUALITY OF STEEL AS A PRICE REDUCTION
[This statement was prepared by the United States Steel Corporation in con-
nection with the hearings on the steel industry before the Temporary National
J]coDomic Committee.]
November 1, 1939.
A comparison of the price of any steel product ten or fifteen years ago with the
price of the product known by the same name today gives an incomplete and in-
accurate impression, as it does not take into account the many substantial econ-
omies which buyers of steel have been able to effect by reason of its improved
quality. These improvements have not, in general, been compensated for by
price increases, and have, therefore, amounted to price reductions. So great
have these improvements in quality been, that in some cases it is misleading even
to call the products by the same name.
Many people who are unfamiliar with steel may think of it as a single product,
of different sizes and shapes, but of more or less uniform quality. This may have
been relatively true in the early days when customers had to adjust their needs
to the limitations of steel. Today, however, there is no such product as common
steel, but, instead, practically aH steel is "custom made" to conform to the special
requirements of each buyer. The ability to obtain steel perfectly suited to their
requirements has enabled manufacturers to take advantage of modem methods
of production to effect reductions in their costs and to design new models which
could not have been produced with the steel of former years.
The automobile is an outstanding example of the custom made character of
steel. In the 1939 model of one of the leading automobiles there are no less than
32 distinctly different t5T)es of steel, each specially made to order and designed,
by chemical content or special treatment, to meet the particular requirements of
the part for which it is to be used. The steel specifications for a given part, for
different makes of automobiles are usually different.
Reductions in real price through quality improvements, which are found m
almost all types of steel products, are illustrated by the following examples in each
of the principal general types of steel products.
Sheet Steel
The average base price of steel sheets used in making automobile bodies has
been reduced approximately 30% in the last fifteen years, despite the fact that
duixiig this period all of their service properties have been tremendously improved.
The modern sheets, made by a radically new method of processing, can be
produced 33j4% thinner and 20% wider than was possible fifteen years ago and
can now be made in coils hundreds of feet long as compared with a former maxi-
mum length of 100 inches. The deep drawing qualities have been increased 30%
to 40% permitting deeper stamping of much more sharply rounded shapes. Deep
dr.awing qualities, of common sheets although limited and inferior formerly
14110 CONCENTRATION OF ECX)NOMIO POWER
entailed an extra charge. Today there is no extra charge for the truly remarkable
drawing properties of automobile sheets.
The modern one-piece automobile steel top is stamped out in one operation
from a single sheet of steel. Formerly, it was necessary to make the top of wood
and fabric entailing numerous hand operations.
A front fender can now be stamped out of one sheet in the same way in one
operation. It was formerly made of two sheets of steel separately formed and
then joined together.
Painting the surface of the earlier automobile sheets to develop an acceptable
finish required elaborate preparation, glazing, and many separate coats with
intermediate treatment between each two coats. Today these operations are
unnecessary, because of the fine grain and dense polished surface of modem sheet
steel. As a result, the time required to apply the finish to a part has been reduced
from a minimum of 48 hours to 6 hours. The following photomicrographs, (pp.
Uin-14112) of a 1939 and of a 1924 type of sheet, made in 1939 by 1924
methods, show clearly the improved surface.
Enameled articles, such as electric refrigerators, sinks and bathtubs, made of
sheet steel are stamped into shape and enameled. Improvements in the modern
steel enameling sheets have permitted the manufacture of deeply formed modern
designs of these products, with a nearly perfect and non-chipping enamel finish
impossible a few years ago. Despite these cost saving improvements, the base
price of steel enameling sheets has been reduced approximately 30% in the last
fifteen years.
The present day high quality silicon steel sheets used in the electrical industry
have made possible more efficient generation and use of electrical energy and thus
contribute significantly to the use of larger generators and motors than were used
fifteen years ago. The resulting savings are well illustrated by reduction in the
losses of electric power in the steel cores of generators from .77 watts per pound
in 1924 to .52 watts per pound in 1939, an improvement of about 32>^%. These
improved sheets are sold at subsiantiaily the same price as the cprresponding
sheets fifteen years ago.
Tin Plate
Approximately 50% of all tin plate sold today is so-called "cold reduced" tin
plate. This new and improved type, introduced in quantity about 1931, is sold
at a lowfer price than the hot rolled tin plate of 1924.
The improved workability and resistance to corrosion of cold reduced tin plate,
which could not have been obtained by 1924 methods, have been important factors
in the canning industry. Its high corrosion resistance has made it possible
to can prunes and some types of cherries which could not be canned successfuily
with hot rolled tin plate. Cans containing other types of acid fruits which could
not be kept longer than eighteen months without risk of becoming unsalable, can
now be kept as long as seven years. Its improved workability has materially
facilitated fabrication.
All modem tin plate has a much more uniform coating of tin due in part to
improved smoothness in the surface of the black plate ob which the tin is applied.
This improved surface, which is important in aU uses of tin plate, is illustrated
by the following photomicrographs (p. 14113).
The purchaser of modern tin plate has less waste in trimming the sheets to the
size and shape required for his purposes since modern tin plate is much more
accurate in its dimensions than was the tin plate of fifteen years ago. In^'eased
uniformity of thickness has contributed to the economical use of high speed
machines with automatic feeders in the can-making industry, by eliminating the
necessity of frequent adjustments in the machines and has also resulted in a reduc-
tion in sheet damage.
It is estimated that, due to the improved quality of modem tin plate, the average
weight of tin plate used for any given purpose has decreased about 10%, which,
at present prices, would mean a reduction of about 20 cents per base box.
Formerly, a relatively high number of sheets of the tin plate were damaged in
transit to the purchaser's plant due to twisted and bent edges. The modern,
much more compact and better protected packages, made possible by the almost
perfect uniformity of the sheets, have greatly reduced these losses.
In 1936 the United States Steel Corporation developed and introduced a new
low«r^^riced type of tin plate, electrolytically plated, called "Ferrostan". It is
being used to replace the more expensive tin plate in cans for baking powder,
coffee and similar "dry pacH" and has made possible the economical packaging
of various additional commodities for convenient distribution.
CONCENTHATinN OF EOONOMIC POWER
14111
14112
CONCENTRATION OF EOONOMIC POWKH
KCTi
ft.CC
CONCENTRATION OF ECONOMIC POWER
14113
14114 CONCENTRATION OF ECONOMIC POWER
In 1933 the United States Steel Corporation developed and introduced a light
coated terne plate as a substitute for tin plate at a substantially reduced price.
Although never used where the surface would be in contact with food, since its
coating contains a substantial amount of lead, it has been extensively used in
cans for paint, varnish and oil.
Structural Steel
Tlie use of an improved series of beams, known as wide flange beams, developed
in their modem efficient form during the last twelve to thirteen years, has per-
mitted the following estimated construction savings:
Higher tier buildings 20%
Lower tier buildings and heaviest mil] buildings 15%
Short bridges where wide flange beams are substituted
for built up girders 15% to 20%
Large bridges and light mill buildings 5% to 10%
Bulkheads, seawalls and similar retaining walls can now be built of modern
improved steel sheet piling more than twice as strong as was possible with the
steel sheet piling of five or six years ago. Despite its doubled strength, this im-
proved piling, known as Z piling, costs only 7% more per square foot than the best
section available five or six years ago. Before the introduction of Z piling, in
order to obtain equivalent strength it was necessary to reinforce and strengthen
the older type of section, involving additional steel and increased fabrication costs.
A new type of steel foundation pile, introduced in 1935, has made possible
savings ranging from 10% to 40% over competitive forms of construction for
bridges, buildings and other structures, and has made possible the reclamation
of many building sites.
In 1934 the United States Steel Corporation introduced a new type of steel
bridge floor for use with concrete or asphalt but weighing less than the reinforced
concrete then used. This has been a great step toward more economical bridge
building, since the weight of the bridge flooring, and not that of the traffic, which
is always light by comparison, determines the size of the foundations and super-
structure. Its lighter weight makes possible lighter and less expensive founda-
tions and superstructures in new bridges, and has often extended the useful life
of old bridges. It has also been used to increase the load carrying capacity of old
bridges with only minor, inexpensive changes or additions to the existing struc-
ture. Under favorable conditions, it costs no more than leinforced concrete slabs.
Late in 1937, an all steel, open lattice-work type of floor, for use without
concrete or asphalt, was introduced. Its exceedingly light weight makes it
particularly fitted for the lifting spans of draw bridges and its open lattice-work
character makes it very useful in states where snow removal is a problem. Though
it costs approximately twice as much as reinforced concrete slabs, it permiis more
than offsetting economies, in machinery and power required for the lifting' of.
draw bridges, and in reduced maintenance and snow removal costs.
One or the other of these types of floors has already been used in nearly 500
old or new bridges, including its use in New York City in reflooring three main
bridges and flooring one new bridge.
Stainless Steel
Stainless stee , although only introduced on a large scale during the last ten
years, has already replaced other types of steel and other competitive products
in many uses where ability to withstand high temperatures or resistance to
corrosion is a factor. It has considerably red^iced upkeep and replacement costs
when it has been used, because of its long life due to its high strength and cor-
rosion resistance, and because its hard, permanently bright surface does not
require repeated painting and is not worn away by scouring.
Stainless steel has been quite generally adopted in the dairy industry for
vessels, pipes and other pasteurizing and bottling equipment. It has been found
to be the outstanding commercial metal completely insoluble in milk cooling from
the pasteurizing temperature and its surface can easily be kept clean and sanitary.
It has beei; extensively used recently in the construction of tank trucks for
carrying milk.
It has also been extensively used in the food processing and packing industries,
for reasons of hygiene and economy. It has been installed in the pineapple
canning plants in Hawaii, as it has proven to be the only commercial metal
which successfully resists the action of pineapple juice.
CONCENTRATION OF ECONOMIC POWER 14115
Enamel and glass ware equipment, considered a great step forward when
introduced in the dairy, food processing and packing industries, have now been
largely replaced by stainless steel because their great weight and fragility made
transportation costly and repairing impracticable.
Due to its resistance to corrosion and its high strength relative to its weight,
it has gained a constantly wider and more varied use in the building industry
generally, for such parts as doors, hinges, window moldings and restaurant counters.
It is particularly adapted for the making of pipes and tubes for use in the chemical
industry and in other industries employing equipment subject to severe cor-
rosive influences.
Modern drums of stainless steel are now replacing wooden barrels in the ship-
ment of beverages and syrups, furnishing extra strength, durability and perfect
protection of the contents. The ability to cleanse completely the inside of such
containers is making it possible to use the same containers almost indefinitely.
Current developments indicate the possibiUty of adopting extremely thin and
strong stainless steel strip for use in airplane wings and bodies, permitting welding
instead of riveting, therebyXreducing wind resistance and improving operating
efficiency.
Although its initial cost per unit weight is somewhat higher than that of the
produces which it has replaced, its increasingly wide use has been due to the
fact that the initial cost has been more than offset by the subsequent reduction
in expense of upkeep and replacement and by the hghter weights required because
of its higher strength.
Alloy Steels
The science of predicting and controlling the properties of alloy steels which
will result from "heat treating" has made very great progress in the last fifteen
and even in the last five years. By heat-treatment is meant the modification
of properties through the agency of heating the steel to a fairly high temperature
and applying thereafter a controlled cooling rate (as in quenching) and a reheating
if desired. The steel itself, however, must fit the heat-treatment and vice-versa.
It is now possible to predict and'control such properties as hardness, toughness or
resiliency with a remarkable accuracy, absolutely unknown fifteen years ago.
This is very important in all mass production industries, such as in the auto-
mobile industry, where it is common to heat treat parts made of alloy steel.
Modern manufacturing methods in these industries are based upon the fact
that alloy steels of given specifications, though made at different mills, at different
times and under varying conditions, when heat treated in the routine of production
will have standard properties.
The inability to predict these properties accurately made it necessary, formerly,
to provide for a comparatively wide margin of safety through the use of excess
weight or extra alloy content. Equal or superior efficiency can now be obtained
by the use of lighter weight parts made of less expensive steel with a lower aUoy
content.
A reduction of 15% to 20% in the average weight of automobile transmission
gears, for example, has resulted from these improvements and analogous v/eight
reductions of varying degrees, have been made in the many other parts made of
alloy steel, without sacrifice of service value.
Alloy steels are generally sold in the form of bai-s to manufacturers, who usually
forge, cut and machine them into the desired part and then heat treat them to
provide varying degrees and types of properties. It was, until recently, con-
sidered almost impossible to control the depth to which steel would harden when
heat treated. As a result a relatively brittle core to a part was often unavoidable
in obtaining a surface of the required hardness. Steel can now be produced which,
when heat-treated, will have the desired hardness on the wearing surface and a
far lesser degree of hardness in the center, thus reducing the brittleness and
prolonging the life of the part.
The ability to control the size of the small crystals or grains of which steel is
composed has also had a distinct share in the improvement of alloy steels as
different size grains produce different properties. No technology of grain control
existed even twelve years ago.
The greater refinement in the properties of allo^' steels has increased the number
of standard alloy steels listed by a leading producer from 39 in 1924 to 168 in
1938. This has obviously given the users of these steels far greater latitude of
choice in sheeting the steel best fitted to their needs.
Although in general the prices of alloy steels have been substantially reduced
since 1924, the large number of standard types makes it impractical to indicate
14116 CONi'F-NTRATTON OF ECONOMIC POWFvR
the exact measure of price change in each. However, as examples, the price of
two common types -has been reduced approximately 15% and the price of two
other common types has been reduced approximately 11%.
High Tensile Strength Steels
Great savings have been eflfected by the railroads through the use of a new
type of steel developed by the United States Steel Corporation and commerci-
ally introduced in 1934 under the trade name of COR-TEN.
By the end of 1938, there were in service 19,249 freight cars and over 1,000
passenger cars made in whole or in part of COR-TEN. COll-TEN has a yield
point 50% to 100% higher than plaia carbon structural steel, combined with
four to six times its resistance to atmospheric corrosion. It is possible, therefore,
to build railroad cars of this material at a substantial weight reduction without
decreased strength or durability.
It is possible to construct freight cars made of COR-TEN weighing from 10%
to 22% less than cars of conventional construction made of plain carbon steel
with all the subsequent operating economies which this reduction in weight
entails. Despite its improved properties, the cost of a COR-TEN freight car
has been found to be no greater than that of the conventional carbon steel freight
car when full advantage is taken of the superior properties of COR-TEN to
effect weight reductions.
Even if the cost of a freight car made of COR-TEN were greater than the cost
of a corresponding car made of plain carbon steel, the operating economies
resulting from its use are so important that its widespread use would be economi-
cally justified.
It is estimated that it costs $18 a year to haul a ton of weight over an average
annual operating mileage of 11,000 miles per car. The average reduction of 2.4
tons in dead weight in each of these 19,249 freight cars, due to the use of COR-
TEN, has therefore resulted in an annual average saving to the railroads of
approximately $43.20 per car, a total saving of approximately $830,000 per year
throughout the life of these cars. If the saving is considered in terms of addi-
tional carrying capacity, each of these freight cars, if loaded to the limit, has an
added annual revenue-paying capacity of 2.4 tons. Over an annual operating
mileage of 11,000 miles per car, this represents a carrying capacity of 26,400
ton-miles more than that of a conventional car built of plain carbon steel. The
19,249 COR-TEN freight cars can, therefore, be said to have provided a total
additional annual revenue-paying capacity of 508,173,600 ton-miles.
The experience of railroads in effecting such economies through the use of
COR-TEN has led to its increasingly wide use in many other industries with
attendant large savings to the users.
The performance of this material has resulted in the development of other
similar and competitive steels.
Steel Pipe and Tubes
The service qualities of all steel pipe and tube have, in general, greatly increased
since 1924. A few typical examples of the type of improvement are sufficient to
indicate this progress.
One of the chief beneficiaries of the improvements in steel pipe and tubes has
been the oil industry. By the use of new seamless oil well casing pipe and oil
well drill pipe with numerous new types of special threaded joints, it is now
possible to drill wells to dfepths of approximately three miles as opposed to the
maximum depth of 5,000 feet in 1924.
In order to suspend lengths of casing pipe as long as 13,000 feet and weighing
up to 300,000 pounds, it has become necessary to produce steels with tensile
strengths up to 125,000 pounds and with resistance to external pressure up to
10,000 pounds.
Drilling such deep wells makes necessary a string of drill pipe weighing up to
200,000 pounds which is hung in tension and rotated at speeds varying from 100
to 400 revolutions per minute. This pipe is therefore subjected to tensile stress
from hanging, torsional stress from the rotary movement, and alternate bending
stresses at the tool joints.
The deep drilling made possible by these improved types of casing and drill
pipe has permitted successful redrilling at deeper levels of oil fields, notably in
California, which had been a«bandoned for failure to produce at the former maxi-
mum drilling depth, and has also opened up new reservoirs of oil in Texas,
L^-tisiana and Colorado.
CONCENTRATION OF ECONOMIC POWER 14117
In modern oil refining plants, highly alloyed seamless steel tubes, with from
10,000 to 60,000 hours of useful life, have replaced tubes which rarely lasted more
than 2,000 to 6,000 hours. Today, it is possible to design tubular apparatus for
use in refining to carry unit stresses between 10,000 and 25,000 pounds at 1,000°
F., greatly increasing the safety factor in operation and making an enormous
saving in weight of equipment.
Improvements in the steam boiler industry have been largely associated with
improvements in boiler tubes, superheater tubes and steam pipe through which
steam is conveyed. In 1924, due to the fact that the industry was limited to
the use of low carbon steel boiler plate and tubes, few steam plants operated at
pressures above 500 pounds and at steam temperatures above 700° F. Through
the use of modern seamless boiler tubes of alloy steel, boilers are now constructed
for pressures up to 2,500 pounds and temperatures of 1,000° F. These tubes will
resist the most severe oxidation and corrosion which boiler service conditions
impose.
All of these increases in quality have been effected with a lowering of price to
the public, except that boiler tubes have increased in price because of the extra
cost necessary to produce the very great increase in quality required for high
temperatures and high pressures which were totally unknown fifteen years ago.
Rails
Railroad engineers fully appreciate that the railroads are now receiving rails
substantially improved in quality and dependability. Although, within the last
15 years, the speeds of passenger and freight trains have increased,- on the aver-
age, approximately 70% and 55%, respectively, and wheel loads have increased
between 40% and 80%, it is the consensus of opinion, among producers and
buyers of rails, that the modern rail will last longer, wear better and result in
fewer failures in service than the rails produced 15 or 20 years ago. The price
is approximately the same.
In considering the price of rails, it should always be borne in mind that even
so-called worn out rails are sold by the railroads as prime scrap*. The average
monthly price of this type of scrap at Pittsburgh for the period 1924-1938 was
$14.30 a ton or 34% of the average monthly price of rails during that period. At
present the price of old rails as scrap ranges from $22 to $25 a ton, or approxi-
mately 60% of the present price of rails.
Lower material, labor and upkeep costs have resulted from increasing the stand-
ard length of rails from 33 feet to 39 feet at no increase in price despite the fact
that changes in length or weight entail comprehensive and costly changes in steel
works and rolling mill equipment. Some railroads order 45 foot rails and serious
consideration is being given to the advisability of using rails up to 100 feet or more
in length.
The increased dependability of modern rails is due to the improved quality of
the steel as verified by increasingly severe testing and inspection standards.
Formerly one specimen of rail from each of the 25 to 30 ingots in each heat were
tested by dropping a 2,000 pound wedge shaped steel weight on the top of each
specimen from heights ranging from 17 feet for the lightest rails to 20 feet for the
heaviest rails. Recently the maximum height of drop was increased to 22 feet.
If all specimens passed this test, one specimen was broken and its interior structure
examined. If any specimen failed to pass these tests, the top rail in each ingot
was clafisified for less severe service and placed in a lower price category.
Some railroads now require that a specimen from each ingot be subjected to
both tests (i. e.. drop test and fracture inspection) and that, if on examination of
the test fracture, a specimen is rejected, other tests be made from specimens cut
progressively from the rails of that ingot. Under the requirements of this pro-
gressive test, rails failing to meet the test are not accepted even as lower priced
rails but are rejected and scrapped.
Each of these tests is supervised by separate inspectors representing the rail
producer and the railroad. The examination of the cross section of the rail and
its approval or rejection is, in many cases, a matter of judgment with respect to
which the railroad inspectors have, as a practical matter, fairly arbitrary powers.
The requirements of this test and the standards of straightness and smoothness
have become so exacting, that rails, which would have been considered good aver-
age rails fifteen or twenty years ago and completely acceptable, are rejected today
by both the mill engineers and the railroad inspectors.
Special treatments have recently been devised and introduced which appear to
remove one cause for one of the most troublesome types of rail failures, known as
the '-'transverse fissure". This is a eeparajion of the metal inside the head of the
14118 CONCENTRATION OF ECONOMIC POWER
rail developed by flexure in service and is not apparent until failure of the rail.
The gradual introduction since 1932 of rails subjected to one or the other of these
special thermal treatments has remarkably reduced failures due to transverse
fissures. The figures of the American Railway Engineering Association indicate
that in rails rolled during the five years 1932-1936 there were only 9 such failures
in the fi rst year of service as compared with 343 such failures in the first year of
service in rails rolled during 1927-1931.
One of the treatments against transverse fissures, the Brunorizing process,
developed and introduced by United States Steel Corporation, produces rails
which are also much more resistant to shock at low temperatures, an imjjortant
safety factor in parts of the country where the rails must stand much zero or
sub-zero weather.
The increased inspection and control in the rail mills in the making of the steel
in the open hearth furnace, in the rolling and in the cooling and other special re-
quirements in connection with rail making have all increased the manufacturing
costs of rails and, contrary to the general trend, more men are required to operate
a rail mill today, than fifteen or twenty years ago
Wire Products
Improvements in wire products have been due in part to the increased perfection
of rods from which wire is made, resulting from scientific selection of ores, classifi-
cation of raw materials and other improvements in steel making. Improved
modern wire making machinery and practices have also contributed to this im-
provement. Typical illustrations from a few of the numerous wire products are
sufl^cient to indicate the general improvement in quality.
The resiliency of spring wire, for example, has generally increased in the order
of magnitude of 50%. Resistance to fatigue, due to repeated flexing, as in the
case of springs usea in automobile engines, has been greatly increased, thereby
prolonging the life of the spring. Although the price of spring wire has been re-
duced since 1924, it is estimated that its reliability and service has increased from
100% to 500%, depending updn the type of wire.
The uniformity in size, form and physical properties of nails, rivets, bolts and
other similar products, requiring the forming c.i a head by pressure exerted on the
end of the wire at high speed has greatly improved due to the improved quality
of the wire and wire bars from which they are made. Today practically perfect
Leads are produced, while formerly irregularities in the wire resulted in considers
ble irregularity in the shapes of heads which caused failures in service.
The uniformity of nail wire is such that since 1928 it has been a standard require-
ment that the number of nails per pound must not vary beyond 5% of the stand-
ard number. As any slight variation in the thickness of the wire would cause
considerable variation in the number of nails, such a narrow limit of variation was
formerly impossible. The' importance of this is obvious to a buyer of nails who
buys nails by weight. Likewise, the increased accuracy in the gauge of wire is
extremely important to the purchaser of wire by weight.
The life of galvanized wire has been greatly increased through the improved
quahty of the galvanizing. It is now a standard requirement that such wire
must bend ai )und its own diameter without roughening the galvanized coating, a
requirement which could not have been met commercially a few years ago.
It is probahle that few people realize the important part wire has played in the
development of the modern automobile tire. Nevertheless, the bead of automobile
tires has for many years been strengthened by the use of wire. Improved methods
in the manufacture of wire have resulted in the production of wire for this purpose
having a tensile strength of approximately 300,000 pounds per square inch, which
is from 25% to 50% higher than the tensile strength of wire used 10 to 20 years ago.
In addition, this wire is now required to meet physical tests for toughness, such as
torsion and elongation, both before and after, it is subjected to the vulcanizing
process. These improved physical properties have permitted a reduction in the
size and number of wires necessary for strengthening the tires and have also
materially contributed to the increased life and dependability of modern tires.
General Improvement in Quality
In general, the quality of the steel used in all steel products today is far more
uniform and dependable and better suited to the needs of the users than fifteen
years ago due to genera] improvements in steelmaking.
The basic mateiials from which pig iron is made are now selected and graded for
uniformity which makes a much superior and more uniform iron.
CONCENTRATION OF ECONOMIC POWER 14119
Chemical control of practically all processes, other than the strictly mechanical,
has become universal. Pyrometry for the measurement and control of tempera-
tures has been developed and applied in all important processes.
Slag control in the open hearth furnaces has permitted the production of the
several kinds of steels for various uses with a dependability and an accuracy that
was not even approximated several years ago.
The increasingly greater approach to laboratory standards of precision and qual-
ity in the commercial production of millions of tons of steel every year, has required
the training and maintenance of a large staflF of metallurgical engineers, upon whom
practical operating men have necessarily come to rely more and more in the turn-
ing out of a modern marketable product.
In short, during the last fifteen years, steel making has been changed from an
art to a science with the resulting general improvement in all steels and with ac-
companying benefits to the users of steel.
Exhibit No. 2183
Fig.
RELATION OF INDUSTRIAL PRODUCTION. EXCLUDING
IRON AND STEEL, TO STEEL SALES
28/
X
.
Average
relation
27
lb/
35^ y^'^^
/ 23
• ■30
i
■38. y
^•■33
•22
*iai9
• ■20
\z --"
15 20 25
ESTIMATED DOMESTIC STEEL SHIPMENTS
30 35
[MILLION TONS)
NEC 35966 SUB
14120 CONCENTRATION OP ECONOMIC POWER
Exhibit No. 2184
CONTRAST IN PRODUCTION-PROFIT COMPUTATIONS
MR T W LIPPERT, METALLURGICAL EDITOR, IRQN AGE
(Based on data for two large componies)
U. S. STEEL CORPORATION
CONCENTRATION OF iDCONOMIC POWER
Exhibit No. 2185
14121
Table I. — Reconciliation of Total Costs Before Bond Interest and Inter- Company
Items in "Analysis'' and Registration Statement, 19S6-S7 — U. S. Steel Cor-
poration
1936
Cost before Bond Interest and Inter-Company Items as pjr
"Analysis (Table 5, columns (2) (3) &>Xi)":
Total -v-
$770,713,718
Operating Costs
Idle Plant Expense.
Purchase Discount.
1,269,970,378
1, 136, 149
-1,815.869
1, 035, 128,
1, 396,
-1,455,
769, 564, 100
2, 089, 259
-939, 641
Break-down of Total Costs as per Registration Statement:
Total - -
770,713,718
General administrative and selling expenses
Payments under pension plan to U. S. Steel and Carnegie
Pension Fund
Taxes (other than Federal income and surtax)..
Taxes (state and Federal social security and railroad re-
tirement)
Idle plant expenses
Depreciation and depletion...'
Plant and Organization siyvey expenses
Discount on Purchases
Reversal of provisions under Railroad Retirement Act
of 1935 -
Cost of goods sold, operating expenses of trafasportation
common carriers and miscellaneous operations
,415,904
, 136, 149
, 589, 159
, 756, 776
,815,869
.744,729
,772,329
7, 642, 026
37, 220, 467
4,081,587
1, 396, 989
55,466,762
1,379,829
-1,455,169
7, 362, 723
34, 541, 519
2, 089, 259
47, 801, 390
285,003
-939,641
Sources: "Exhibit No. 1416" and Registration Statement (Form At2) submitted to the Securities and
Exchange Commission.
Table II. — Comparison of Break-Down of Lumped Costs in the "Arfalysis" and in
Registration Statement, 19S5-S7 ' — U. S. Steel Corporation
1937
1936
1935
Breakdown of Lumped Costs as per "Analysis":
Total..
$1, 143, 773, 008
$929,880,828
$680,858,276
Payroll ^...
Other Expenses
426,330,944
343,494,434
373,947,630
328,070,724
293, 599, 768
308,210,336
246, 508, 043
197,911,220
236, 439, 013
Inter-Company Items
Breakdown of Lumped Costs as per Registration Statement:
Total
1,143,773,008
929,880,828
680,858,276
General administrative and selling expenses...,
Plant and Organization survey expenses
44,668,352
1, 756, 776
.-744,729
110, 294, 824
'7,650,787
10,213,951
16,892,758
.953, 720, 009
39,447,790
1,379,829
1,396,989
-1,455,169
34, 613, 494
285,003
2, 089, 259
939 641
Idle Plant Expenses
Discount oii Purchases
Reversal of provisions under Railroad Retirement Act of
1935.
Maintenance and repairs (including provisions for blast
furnace reUning, oven wall relining and rebuilding and
- 85,589,705
5, 492, 640
7,860,403
17, 461, 037
772, 707, 604
59,097,850
2, 809, 295
5, 249, 623
14, 010, 938
563, 642, 455
Depreciation-, depletion and amortizing of fixed assets
(credited directly to property plant and equipfnent)....
Rents and Royalties
Additions to Reserves Charged to Cost of Goods Sold,
Etc.-TotalJ :......
Residual'
' "Lump costs" is used to designate the total "other Expenses," "Payroll," and "Inter-Company Items"
in the "Analysis." It is the same as total costs less taxes, depreciation, pensions and bond interest (spe-
cifically segregated items in the "Analysis," Table 8), all before deducting inter-company items.
' There may be some duplication between reserve additions and maintenance and repair items. The
effect of such duplication would be to lower the residual.
14122
CONCENTRATION OP ECONOMIC POWE(R
Table II-A. — Additions to Reserves Charged to Cost of Goods Sold, Etc., 1935-37^-
U. S. Steel Corporation
1937
1936
1935
Total
$16,892,758
$17,461,037
$14,010,938
General contingent reserves currently provided by charges to opera-
tions and held for purposes arising from operating activities
Accident and hospital reserves currently provided through charges
to operations and held to cover expenditures resulting from operat-
9, 036, 960
4.463,981
273, 525
930,016
1,048,068
527
396,168
730,814
9,658,240
3,879,490
1,247,012
791, 799
1,361,740
-184
489, 773
219,256
7,228,531
3,780,359
1, 071, 976
503 125
Reserve.s raised by current accruals for purposes of absorbing ex-
traordinary expenses in specified operations ..
Other reserves provided and held for specific purposes or for other
genera! contingencies for which they may be available
Insurance reserves ' - -
826,642
23,840
309,663
-89
—9 970
Reserves for sundry marketable securities .
Inventory valuation reserves
Reserve for U. S. Steel Corp. stock for employes' subscription. .
Reserves account outside real estate, real estate mortgages and
43,244
-645
.-29,900
-46, 170
-55, 734
-84, 185
149, 764
55 666
Reserves account house and land sales installment contracts and
mortgages under employes' home owning plans
Reserves for accounts and notes receivable not colIectYble within
71,631
' Detail for "Additions to Reserves" item of Table II.
> Provided by charges to operation, together with accretions thereon, are available for absorbing fire,
windstorm and paM of the marine losses of subsidiaries. The subsidiaries are self-insurers against such
losses and generally do not insure with outside insurance companies. Specific funds have not been segre-
gated for these reserves.
Table III. — Taxes Other Than Federal Income and Social Security Taxes,
1927-38— U. S. Steel Corporation
Year
Total
Charged to
Costs as per
"Analysis"
Capital
Stock
Tax
Total Ex-
cluding
Capital
Stock
Lake
Superior
Iron Ore
Properties
All Other
Properties
Excise and
Miscella-
neous
Differ-
ence"
1927
$34,817,116
36,015,942
37, 739, 322
36,047,026
33, 162, 707
31, 943, 315
33, 288, 485
32,615,831
34, 691, 330
37,999,606
45, 132, 333
34,602,915
$34, 817, 116
36,015,942
37, 739, 322
36,047,026
33, 162, 707
31,444,131
32, 289, 301
30, 983, 323
32, 954, 959
36,a90,842
43,203,328
32, 602, 890
1
1928
1929
1930
1931
» $14, 215, 651
12,681,884
« 12, 735, 967
13,859,408
13,828,911
15, 187, 324
18, 197, 478
12, 837, 735
» $19, 858, 097
U8, 950, 835
•17,599,925
17, 399, 374
18,604,455
20,209,831
24, 685, 087
19,207,090
< $93, 984
1 105, 617
« 288, 591
•289,094
480, 272
407,906
320,763
558, 066
-$1,004,925
1932
1933
1934
1935
1936......
< $499, 184
999, 184
1, 632, 508
' 1, 736, 371
1, 808, 764
1,929,005
2,000,025
-294,205
+1,664,818
-564,653
+41,321
+385, 781
' Difference between accrued tax liabilities and amounts charged to costs. Breakdowns in 1932-36
Annual Reports show accrued liabilities for taxes; Reports for other years show actual charges to costs.
> As shown in 1932 Annual Report; no breakdown given in 1931 Report.
' Approximate amount.
* Allocated to 1932 although paid under the National Industrial Recovery Ac
_L-t which riecame effective in
June 1933. ~ ■ - ' ° '■ ..- . , ^ ^
• As shown in 1933 Annual Report. 1932 Report shows $19,087,813 for other than Lake Snperi ir Iron Ore
Properties.
» As shown In 1934 Annual Report; 1933 Report shows $13,314,519 for Iron Ore properties an J $17,869,925
for other properties.
' As shown in 1936 report; 1935 report shows $1,759,922.
Source: Break-down from Annual Reports.
CONCENTRATION OF ECONOMIC POWER
14123
Table IV. — Taxes Other Than Federal Inconie and Social Security Taxes, 1927-
1938 — Recompvied "Fixed" and "Variable" Costs — U. S. Steel Corporation
Estimate of
"Fixed"
Cost per year
Estimate of
"Variable"
C... per Ton
r=
$24,217,000
$1,433
.96
iDCluding 1937:
Totcl
29,950,000
0. 691
2,000,000
27,950,000
Other -t - -
6.69i
.714
Excluding 1937:
Total
31,090,000
0.514
Capital Stock Taxes (1938)
2,000,000
29, 090, 000
Other' -
0. 514
.788
' This relation approximates the one for 1927-31 shown in Chart 2, p. 13 of the "Analysis "
Source: Based upon'Annual Reports and "Analysis."
" Table V. — Maintenance and Repairs, 1927-38 ' — U. S. Steel Corporation
Charge? to Costs
From Current Expen.
- Charges to
1 Costs in
Other
I ExcesE of
Credits to I Expendi-
Reservc tiires *
Weighted
Tons of
Produi ts
Shipped
(Millions)
1927.
1928
1929
1930.
1931.
1932
1933
1934.
1935
$113.
106,
107,
875, 264
684,913
235, 214
.';i2, 158
419, 408
112,622
169, 688
894,512
978, 426
126, 667
304,054
448, 926
$91,035,
85, 707,
85, 900,
7.1,411,
21 [924!
40,"'974,'
47, 567,
70,378,
161 $22, 840, 103 $85, 626,
302 20,977,611 79J'86,
446 21,334,768 I 79,411,
765 21,100.393 i 70.192,
14,080,750 I 43,300.
7, 187, 828 1 20, 628:
8,279,732 ' ""
11,919,714
12, 410, 8-32
15, 747, 807
860
30, 135,
39, 183,
45, 447,
67, 262,
,267 1
,199 ;
5,219,
3, 038.
1,296,
1,754,
1, 791,
2, 119,
3, 116,
3,621,
2, 100,
$138,
3, 537,
1,799,
1, 139,
962.
974,
1, 567,
720,
-122,
1. 342,
13.0
14.0
15.1
11.9
6.2
6.1
7.6
11.0
13.2
7.8
■ Includes a portion of idle plant expenses.
' Expenditures excluding extraordinary replacements charged to depreciation and replacement reserves.
Source: Annual Reports, except weighted tons which are from "Analysis."
Table VI. — Stripping and Development Expenses, 1927-88 — L'. S. Steel Corporation
■ Year
Charged
to Costs
Charged to
Costs in Ex-
cess of E,x-
penditures
Weighted
Tons of
Products
Shipped
(Millions)
1 Charged to
,,„ 1 Charged Costs in Ex-
, ^ ^-"^ ; to Costs cess of Ex-
i penditures
Weighted
Tons of
Products
(MiUwns)
1927
1928
1929—
1930
1931
1932
$4,431,341
5, 073, 611
6,218,468
5, 224, 575
2,963.640
482, 575
-$1,085,560
.599.479
1. 713, 361
-949, 275
-1.52,656
-797,763
13.-0
14.0
15.1
11.9
1933 -. : $2,103,857 1 $1,006,073
1934.. 1 2, 382, .389 ' 953,003
1935 ' 2,698,050 ' 1,421,606
1936. 5,431,607 3,291,631
1937. : 7.557,125 .^ 290, 267
1938 ! 2.622,625 477.360
1 1
6.2
6.1
7.0
11.0
13.2
7 8
Source: Annual Reports, except for weighted tons which is from the "Analysis.'
124491— 41^pt. 26-
14124
CONOENTRATION OF p]<'ONOMIC POWER
Si I
s
1
El
83
\
/ \
/
\
"\
.<^
I
— g
— g
,3 S5
CONCENTRATION OF ECONOMIC POWER 14125
Exhibit No. 2187
THE NET REGRESSION OF VOLUME ON PRICE
RELATIONSHIP I
' 1
1 1
i i 1
1 i 1
1 ■
1 — [~l-
i
1 "
1 i ?*'
.26 I
:
! ' 1
RELATIONSHIP E
-
.. ...
20.
/^
y
..^
" i
c» t.O tj 8.4 te
RELATIONSHIP!
RELATIONSHIP m
\
—^
^3
*"
.„
i*
♦s
•?^
'^
" 1 i
v
\
.«*
p
.sr
„ f«
S8
*^,
7?^
C£HT$ PES L».
emmsire poiec of Finisnee tnct.
It t.o u t.4 t.e t3
CENTS PCT t».
■ M/ar o' FmtDKO srm ,
14126
CONCBNTRAaiON OF ECONOMIC POWER
i 1
1 1 - 1
fc fe • ^
g ° g "
g 8 ^!
if
SUPPLEMENTAL DATA
The following telegrams have no connection with subjects dealt
with in this volume; they are included in the record to authenticate
certain statements made by Mr. Eugene Orvis in his prepared state-
ment, admitted to the record as "Exhibit No. 1293" and included
in Hearings, Part 16, appendix, p. 9330.
[Telegram]
Kansas City, Mo., October 10, 1939.
James Brackett,
Secretary, Temporary National Economic Committee,
Federal Trade Commissio-i. Building, Washington, D. C:
We have been advised that the committee desires certain information from
Great Lakes Pip-^ Line Co. relative to the prepared statement presented at the
oil-industry hearing by Eugene L. Orvis. The owners of Great Lakes Pipe Line
Co. are Continental Oil Co., Midcontinent Petroleum Corporation, Skelly Oil Co.,
The Texas Corporation, The Pure Oil Co., Sinclair Refining Co., Cities Service
Oil Co., and Phillips Petroleum Co. The members of the traffic committee of
Great Lakes Pipe Line Co. and the company bv which each is employed are
C. A. L. Walker, with Great Lakes Pipe Line Co.; Charles Ervin, with the Texas
Corporation: Henry Hauseman, with the Pure Oil Co.; C. A. Hoimos, with Cities
Service Oil Co.; P. H. Kii.ns, with Continental Oil Co.; C. R. Musgrave, with
Phillips Petroleum Co.; J. M. ODay, with Sinclair Refining Co.; H. W. Roe,
with Midcontinent Petroleum Corporation: and A. F. Winn, with the Skelly
Oil Co.
This traffic committee is an advisory group whose purpose is to a.ssist the pipe-
line management to correlate the cumbersome pipe-line operation with the
requirements of all shippers and the regulations of the Interstate Commerce
Commission, '^he recommendations of the committee are in no wise binding
upon the pipe-line management or upon any member of the committee nor may
any shipping representative speak for the pipe-line company. We ha\e examined
the minutes of the traffic committee of Great Lakes Pipe Line Co. and they do
not contaiti the paragraphs quoted by Mr. Orvis on page 10 and continuing to the
top of page 11 of his prepared statement; however, the matters appearing in the
purported minutes were discussed on several occasions with the traffic managers
of the shipping companies prior to item 60 on page 13 of Mr. Orvis' prepared
statement being incorporated in the tariffs of the pipe-line company.
The quotation at the bottom of page 21 and extending over to the top of page
22 of Mr. Orvis' prepared statement purports to be an excerpt from the minutes
of a meeting held June 23, 1938, of officials of tiie companies owning Great Lakes
Pipe Line Co. No such meeting v.as held. The quotation is an excerpt from a
letter written by the traffic manager of the pipe-line company to the traffic
managers of the various shipping companies.
Haery Moreland,
Vice President, Great Lakes Pipe Line Co.
[Telegram]
Kansas City, Mo., October 10, 1939.
James Brackett,
Secretary, Temporary National Economic Committee,
Federal Trade Commission Building, Washington, D. C.
Further referring to my telegram to you dated October 9, the paragraphs quoted
by Mr. Orvis on page 10 and continuing to the top of page 11 of his prepared
statement are not a part of the minutes of the traffic committee of Groat Lakes
Pipe Line Co. but are undoubtedly a memorandum expressing the views of the
14127
14128 COxNCBNTRATION OF ECONOMIC POWDR
traffic committee on the subjects therein referred to and as stated in my earUer
telegram were taken Into consideration in drafting itepti, 60 of Great Lakes Pipe
Line Co. tariffs which is correctly quoted on page I'S of Mr. Orvis' prepared
statement. The copy of a letter which you have dated January 22, 1937, trans-
mitting the minutes of the meeting of the traffic committee of Great Lakes
Pipe Line Co. held at Kansas City, Mo., January 19, 1937, is correct. The minutes
as attached thereto are also correct except that the heading of the third para-
graph on the first page of said minutes referring to Messrs. Kuhns, Okay Winn,
and Stewart should read "Members absent" instead of "Members present."
The excerpt which you have of the minutes of the meeting of the traffic committee
held at Kansas City, October 9, 1936, is correct; however, the reference to case
2610 attached to this excerpt is not attached to the recorded minutes of the
committee.
The reference to case 2610, however, is undoubtedly to the proposed rate of 6.9
referred to in the excerpt. The copv of letter which you have dated Januarv 22,
1937, signed by C. A. L. Walker, referring to W. T. L. application D-37-175 to
Messrs. Kuhns and others is correct. We have been unable to find in the files of
Great Lake^ Pipe Line Co. copy of the letter purported to have been written bv
C. A. L. Walker to R. E.Stewart, dated May 11, 1937; or by Mr. Walker to
Continental Oil Co. and other oil companies, dated May 22, 1937; or by R. E.
Stewart to Mr. Walker and others, dated May 7, 1937; however, we have no reason
to doubt the correctness of the copies of these three letters which you have, and
are not questioning their authenticity. We desire to add that the question of
whether the intrastate railroad or interstate railroad rate is applicable from the
terminals of the pipe-line company is a legal t}uestion and the determination of
that question has no connection with the published rates of the pipe-line company
which appear in its tariffs.
Harry Moreland,
Vice President, Great Lakes Pipe Line Co.
INDEX
Page
A. A. A J3647, 13689, 13948
Adams, Henry Carter 13716
Agricultural Adjustment Act 13947
Agriculture, United States Department of 13676, 13719-13720, 13975, 14119
Agricultural Economics, Bureau of 13676,13720,14119
Bureau of Public Roads 13975
Secretary of 1 3676
Almanac of Canning Industry and Iron Age 14023
American Can Co 14018-14019
American Car & Foundry Co 13S9S
American PJconomic Association 13983
American Iron and Steel Institute 13848, 13850, 13852, 13854,
13863, 13872, 13903, 13929, 13937, 13986, 14008, 14011, 14020
American Locomotive Co 13898
American Metal Market Co 13940
American Railway Car Institute : 14016
American Railway Engineering Association 14118
American Rolling Mill Co 13903
American Sheet & Tin Plate Co 13808, 14106
American Statistical Association 13898
American Steel & Wire Co 13910, 14106
"Annalist, The" 13770-13771
Appert, Richard H 1 13649-13650, 13671
Asia 13722
Association of American Railroads 13969, 14003-14005-14009, 14015
Association of National Advertisers 13910-1391 1
Atlas Portland Cement Co 13749
Automobile Manufacturing Association 13819,
13983-13985. 13987, 13994-13996
Austria 13863
Ballinger, J. W 14004
Banking Act 13948
Bean, Dr. Louis : 13597,13617,13632,
13710, 13719-13720, 13723, 13729, 13732-13733, 13735-13741
Examination by, of U. S. Steel Corporation demand analyses. 13720-13732
Bensheimer, J 13616
Bethlehem Steel Corporation 13903
Birmingham, Ala 13900, 13958
Bituminous Coal Act ' 13712
F31ack, J. D_ 13937,13940
Boeckh, E. H - 13964
Borah, Senator William E., in memoriam 13648
Buffalo . 13896, 13899
Bureau of Business Research 13617
Bureau of Corporations, United States 13746
Bureau of Internal Revenue, Alcohol Tax Unit 14025
Bureau of Railwav Economics 14003-14004. 14007-14008, 14015
California _" 14116
Cailiuann, Dr. Riidolf._, . 13616
Canada . 13903
Carnegie-Illinois Steel Corporation - - 13910,
13953-139.54, 13958. 1397S, 13-.'-^0, 14106
"Cartels" 1 13616
Cash costs 13600-13602
II INDEX
Page
Census of Manufacturea 13872, 13890,
13895, 13914, 13979, 13983, 14018. 14024-14025, 14028, 14030
Census, United States Bureau of. 13850, 13854, 13914, 13960, 13973, 13979, 14016
Chamberlin, Edwin... 13651, 13908
Chevrolet 13990, 1 3993
Chicago 13805, 13837, 13899, 13958, 13976
Cincinnati 13896
Cities Service Oil Co 14127
Clark, Prof. J. M 13698
Cleveland Facing 13853, 13896, 13899, J3900
Colorado 13900, 14116
Colorado Fuel and Iron Corporation 13900
Columbia Steel Corporation 13749, 13900, 14101
Commerce, United States Department of . 13914,
13930-13932, 13941, 13960, 13963, 13973, 13979, 13983, 13987
Commercial and Financial Chronicle 13770
"Comparative Asscmblv Costs in the Manufacture of Pig Iron" 13899
Cone, Frederick M...-" 13931, 13941
Congress of the United States 13672
Continental Can Co 13898
Continental Oil Co 14127-14128
Cor-Ten 14116
Costs :
Cash 13600~lo602
Operating, analysis of . 13666-13670
Relationship between, and sales 13802-1361 6
United States Steel Corporation's analysis of 13617-13631, 13695-13718
Costs, prices and demand, relationship between 13655-13665
Court, A.- T 1 13990
Cowles Commission for Econorinic Research 13587, 13604, 13650
Crum,'W. L 13941
Currie, Dr. Langhhn 13640-13641, 13984, 14015
Czechoslovakia 13S63
Das Deutsche Ivartelbrect 13616
deChazeau, Dr. Melvin 13615-13617, 13629, 13642, 13648-13652
13654, 13661, 13663, 13670, 13672, 13732, 13740, 13907, 13918
Demand for -tccl, United States Steel Corporation's analvses of.. 13632-13648,
13720-13741
Discussion of, bv Dr. Theodore Otte Yntema 13732-13741
Examination of, by Dr. Louis lican 13720-13732
Demand, prices and costs, relationship between 13655-13665
Der Unlauten Wettbewerg..- 1361(j
Detroit 13896, 13899
Dickinson, Edward T., Jr 13913, 13942, 13981, 13999, 14016, 14095, 14101
Doblin, Ernest M 13913, 13942, 13981, 13999, 14016, 14095, 14101
Dodge, F. W., Corporation . 13969, 13975
Dubril, S. M : 13991
Dun & Bradstreet, Inc. . 1398U
Eastern Railroads, Public Relations committee of 14014
Econometric Society 13898
"Economics of Overhead Costs, Tiie"... 13698
Edison Electric Institute 13967
Electric Overhead Crane Institute . 13973
]•: urope - .-- 13722
Evaporated Milk Association. ^ 14018
fJzekiel, Dr. Mordecai. 13617,
13671, 13676, 13684, 13709-13710, 13730, 13739 13741
Fabricant, S 14009
Fairless, Benjamin F _ . 13585. 13587
Federal Reserve Bank of New York . . 13970
Federal Reserve Board 13659-13660, 13924. 13934, 13942. 13949, 13954-
13955. 13957, 13962, 13969, 14001, 14013, 14027, 14029, 14032
Board of Governors of . 13935, 13963, 13965, 14013
Division of Research and Statistics 14029
"Federal Reserve Index of Indu.strial Production". 13934-13935
Federal Trade Conuni.Sfsion 13586, 13078, 13788, 14036 14038, 14080, 14127
Ford Motor Co. 13898,13990
INDEX III
Page
Fordham University 1 3650
Foreign and Domestic Commerce, United States Bureau of 13966
Foundry Equipment Manufacturers Association 13973
General Motors Corporation 13590,
13733, 13898, 13915, 13963, 13982, 13987, 14059, 14087
German cartel law 13616
Germany 13863
Goldenweiser, F. A 13935
Government of the United States,. 13641-13642, 13645, 13659, 13670, 13688
Great Britain '. 13863
Great Lakes Pipe Line Co 14127-14128
Hance, Dr. Wendell D 13983
Hansen, Dr. Alvir 13963
Harvard School of Business . 13720
Harvard University 1 3693
Harvard University Press 13700, 13908
Haskins and Sells Foundation 13698
Hatfield, 13698
Hauseman, Henry — 14127
Hawaii 1 13900
Hitler, Adolf 13645
Holmes, C. A 14127
Holmes, Justice Oliver Wendell 13740
Homer, S. L 13986, 13988
Hosmer, W. A 13700
Houston 13896
Hydraulic Institute 13973
Indiana 13899
Inland Steel Co 13903
In Memoriam Senator William E. Borah 13648
Internal Revenue, United^ States Bureau of 13932-13933, 13941, 13961
International Harvester Co 13898
International Tin Research and Development Council 14018,
14020, 14024, 14026-14027
Interstate Commerce Commission 13705,
13716, 13969, 14003, 14005-14006, 14009, 14012, 14014 14016, 14127
Iowa 13899
"Iron Age" 13033,
13695-13696, 13717, 13794, 13797-13799, 13801-13803, 13807-13809,
13814-13815, 13818, 13836, 13863, 13884-13885, 13888, 13890, 13895,
13924, 13928, 13940, 13949, 13967, 13974-13976, 13978-13981, 13985,
13992-13995, 13997, 14007-14009, 14011-14012, 14019-14020, 14029-
14030, 14096-14102, 14104, 14120.
''Iron and Steel and Their Products" 13910
Iron and Steel Works Directory 13903
Ironton, Utah . 13900
Japan ..• 13722,13863
"Jobs for All" . - 13677
Jones & Laughlin Steel Corpc.ttion 13903
Justice, DepartmeAt of 13616, 13671, 13838, 13840-13842, 13844-13845
Kentucky ' 13899
Kuhns, P. H . 14127
Kuznets, Simon 13932, 13941, 13960, 13963, 13983, 13987, 13996
Labor Statistics, United States Bureau of 13668, 13882-13883, 13966,
13978, 13990, 14020, 14022, 14030, 14044, 14046, 14052-14053
Labor, United States Department of 13731, 14020, 14030
Lackawanna Facing 13833
Lake Erie 13899-13900
Lake Michigan 13899
Lake Superior.,.-. 13899, 14122
Lewis, Harold Gregg . . 13049-13650,
13738, 13913, 13942, 13981, 13999, 1401-6, 14095, 14101
Lippert, T. W 13695-13696, 13712, 14120
Los Angeles..--' 13896, 13900
Louisiana 14116
Manlove, G. H . 13940
Maryland State Planniig Commip^jion 13900
IW INDEX
Page
McGraw-Hill Book Co 13617, 13907, 13918
Midcontincnt Petroleum Corporation 14127
Miller, Governor Nathan 13586
Mills, Dr. Frederick C 13647, 13685, 13998
Milwaukee 13896
Minnesota 13899
Moreland, Harry 14 127-14 1 28
Moody's Industrials 1376S
Moore, Henry L 13698, 13913
Mosak, Jacob L 13913, 13942, 13981, 13999, 14016, 14095, 14101
Mudgett, B. D 13937, 13940
Musgrave, C. R.-- 14127
Nathan, Robert R 13930-13931, 13941, 13963, 13966
National Association of Cost Accountants 13910-13911
National Bureau of Economic Research 13899,
13932, 13941, 13960, 13963, 13983, 13994, 13996, 13998, 14009
National Canners Association 14018-14020
National City Bank, New York :___ 13933
National Electrical Manufacturers' Association 13967, 13973
National Income and Capital Formation 13932, 13941, 13960, 13963
National Industiial Conference Board Index 13786,
13888, 13932, 13941, 13963-13964, 13966, 13987
National Industrial Recovery Act 14122
National Machine Tool Builders Association 13973
National research project, W. P. A 13937
National Resources Committee 13725, 13913.
14000-14022, 14011, 14013, 14020, 14022, 14027, 14029-14031
National Steel Corporation 13674, 13903
National Tube Co 14106
Newark 13896
New Jersey 1 3900
New York'. 13896, 13911, 13915, 13918, 13996
New York Stock Exchange 13770-13771
N. R. A 13694, 13948
O'Day, J. M 14127
Ohio_.-. : 13896, 13899, 13900
Oil Well Supply Co J 13749
Olds, Irving S . :... 13585, 13650
Organization Service Corporation 13973
Orvis, Eugene . 1 41 27-14 128
Pennsylvania 13900
Persons, Dr. Warren 13663
Philadelphia 13896, 13900
Phillips Petroleum Co 14127
Philo Verlag und Buchhandlung G. M. B. G 13616
Pittsburgh 13797, 13803, 13808, Facing 13833, 13834
13837, 13887, 13896, 13899, 13900, 13940, 13976, 14030. 14117
Plymouth Motor Corporation 13990
President of the United States 1 401 4
Price reductions, effects of . . . J 13597-13606
Prices:
High, failure of to produce business 13686-13688
Reduced, through concerted action of producers, efTect of 136SS-13603
Prices, demand and co.st'^, relation.'ihip between.. 13655-13665
Pure Oil Co 14127
"Railway Age" . 13973, 14007
Rechtsanwalt am Landgericht 13616
Republic Steel Co 13900, 13903
Roe, H. W .... ... 14127
Rollmann & Maver .. i:^616
Hoos, C. F ". 13589, 13733, 13915-13916, 13922.
13941, 13963, 13967. 13982, 13986-13991, 13993, 14059. 140S7
Roosevelt, President F. D 14003. 14005
Russia 13863
Sales, relationship between, and costs -.. 13602-13616
vSanders, 1369.S
INDEX V
Page
San Francisco 13896, 13900
Schultz, Henry 13917, 13925, 14020
Securities and Exchange Commission 13700. 13702, 13705, 13984, 14040
Segal, Mandal R 13913, 13942, 13981, 13999, 14016, 14095, 14101
Sherman Act 13078
Silver Purchase Act 13947
Sinclair Refining Co 14127
Skelly Oil Co . 14127
SiniondsSaw & Steel Co --- 13898
Skemp, R 14026
Social Science Research Council 13937
Social Security Act - 13930
Social Security taxes 13600-13601, 13953-13954, 14044, 14057, 14083, 14122
Spanish Civil War 13948
Splawn, Commissioner M. W 14005
Sprague, Prof. O. M. W 13693
St. Louis 13896
Statistical Yearbook 14020
Standard Statistics Co., Inc . 13909, 13965, 13970, 13980
Standard Trade and Securities 13909
"Statistical Investigation in the Demand for Iron and Steel" 13913
"Statistical Yearbook" 14020, 14024
Steel Workers Organizing Committee 13953
Stratton, 13907, 13918
Stewart. R . E 14128
Studies, United States Steel Corporation:
Analyses, demand. See Demand for steel. United States Steel
Corporation's analyses of.
Summary and description of ^ 13587-13616
Sweezv, Paul M 13918
Taitel", Martin 13622, 13671, 13694, 13709-13712, 13718, 13740-13741
Tariff Commission, United States 13644, 13660
Tenn(>s.-^ee Coal, Iron and Railroad Co 13900, 13910, 14106
Terborgh, George 13963, 14026
Texas-!---,-- - 14116
Texas Corporation, The 14127
Timken Roller Bearing Co., Inc 13898
Topeka & Santa Fe Railroad Co 13898
Treasury, United States Department of 13033, 13941, 13961, 14025
United States Steel Corporation 1 3585-13587,
13595-13600, 13603-13604, 13606-13615, 13617-13618, 13620,
13622-13627, 13629, 13630-13633, 13636, 13638-13639, 13644,
13649-13651, 13654, 13658, 13661, 13666, 13667, 13670, 13672-
13673j 13676, 13679, 13683-13685, 13687-13688, 13690-13691,
13G93, 13695, 13697, 13699-13700, 13702-13705, 13707-13709,
13711, 13713-13715, 13717, 13720, 13722, 13724, 13728-13731,
13740, 13746-13747, 13751-137.57, 137.59, 13761-13773, 13786-
13788, 13796-13798, 13800, 13802, 13804-13806, 13821, 13834-
13836, 13848-13849, 13851, 138.53. 13857, 13859-13860, 13868-
13871, 13874, 13878, 13880-13881, 138S3-13889, 13891-13896,
1.3899-13900, 13903, 13906, 13910, 13913, 13929, 13937-13939,
13942, 13949-13950. 139.53, 13055, 13958. 13976, 1397S, 13981,
13997-13999. 14007, 14016,- 14019, 14029, 14032-14038, 14058-
U059, 14061-14065, 14068, 14073-14074. 14080. 14082-14095,
14098, 14101. 14106, 14110, 14114, 14116, 14121-14123.
Analytics of:
Cost in relatio!! to volume 13617-13631
Demand for steel 13632-13648, 1 3720-13741
Special Economic Research Section of 13913, 13981, 13999, 14016
Studies made bv. See Studies, United States Steel Corporation.
University of Chicago 13587,13601. 136.50, 13694, 13732, "13738,
13913, 13942, i3981, 13999, 14016, 14032, 14082, 14095, 14101.
University of Pittsburgh 13617
University of Rochester ^ 13720
Univeroitv of Virginia - 13617
U. S. S. R 1386.3
VI INDEX
Page
Utah 13900
Vacuum Cleaners' Association 13967
Vandergrift mill 12808, 13834
von Szeliski, Victor 13589, 13733,' 13915-13916, 13922, 13941,
13963, 13982, 13986-13991, 13993, 14059, 14087
Walker, C. A. L 14128
West Virginia 13899-13900
Wliceling Steel Corporation 13903
White & Case. 13650
White, Ed ward . 1 3933
White, Hugh E 13586
Whitman, Roswell, H 13913
Winn, A. F . 14127
Winn, O'Kay I - - - : 14128
Wisconsin 13833
Wooden, Walter B.... . 13586, 13616, 13643
Woodworking Machinery Manufacturers Association 13973
Work Projects Administration „. 13694
Works Progress Administration -^.. _ 13937, 139.41, 14020
Division of Social Research ,---^_ 14020
World War I 13678, 13680, 14000
Worthing, Miss M. W ' 13895,
13899, 13913, 13918, 13938, 13942, 13981, 13997. 13999, 14007.
14012, 14010, 14019, 14029, 14095, 14101
Wright, C. E 139-!0. 13981
Yntcma, Dwight B 13913, 13942. 13981, 130Q9, 14016. 14095. 14101
Yntema, Dr. Theodore Ottc 13.586-
13587, 13604, 13615, 13617, 13620-13622, 13626. 13628, 13630-
13635, 13637-13640, 13642-13644, 13648-13649, 13652 13654,
13659-13660, 13662-13663, 13075, 13682, 13087, 13697, 13709,
13712-137M, 13717-13718, 13732, 13740-13741, 13913, 13942,
13981, 13999, 14016, 14032. 14082, 14095, 14101
Discussion by of United States Steel Corporation demaiid analvses--- 13732-
13741
Youngstown Sheet & Tube Co .. 13674. 13903
BOSTON
PUBUCUBBABV
k