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
$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 determi n ing 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 imm une 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^
'^
^
/
a.
\
^
-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
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 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
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
2
?
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
^
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
.;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
—
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 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 $55.73 per weighted ton of products shipped.
Total costs are adjusted to 1938 interest, pension, wage, and tax rates, to 1938 price level, and to 1938
eflBciency.
Weighted tonnages are actual tonnages, adjusted for change in proportions of high and low cost products
and for the equivalent tonnage of average cost rolled and finished steel products represented by products
other than steel.
CONCENTRATION OF ECONOMIC POWER
13773
RELATIONSHIP BETWEEN TOTAL COSTS OF OPERATION
AND VOLUME OF BUSINESS - 1938 CONDITIONS
U. S. STEEL CORPORATION AND SUBSIDIARIES
1200
1100
1000
to
900 ^
1100
1000
^ 900
g 800
'=' 700
o 600
^ 500
''
■
]-'
■'
1927
^':
1928
*1929
1936
.^
i^^
^
./'
800 ^
700 °
600 o
500 2
»^
^
1934
^.<
-.
•'^
-1 ^°°
i 300
200
100
^^
^
^
300 2
100
n
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
MILLIONS OF WEIGHTED TONS OF ALL TONNAGE PRODUCTS SHIPPED
NOTt: TOTW. COSTS «OJU$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
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
<=> 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
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
8
19
23 5
9
20
25
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
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
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
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
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
/-
1100
1000
CO
900 ^
800 g
700 °
600 S
500 -
400 2
300 1
200
100
,'-'
'
'Y'-'
''
,^-
1928
'"■
.^ 'r936
1930
1933^
-i
.,''
IW
j931
^.'
''■
[.934
_^-
■-''
1932
1933
'-'
UNADJUSTED COSTS
EST. COSTS UNDER '38 CONDITIONS
'
L
^r
"
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!
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
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
ig
95.4
p
94.5
94.3
94.3
3C
94
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!
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
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
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
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 Disadvanta ge
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
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
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
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
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
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
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
I
c
5
i
a
t
5
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
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
\
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
-
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
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
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
Common Labor . ....
11 1
Total
202,108
100
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
60
50
« 40
a:
g 30
^ 20
10
40
30
3 20
o
10
ANNUAL DATA 19M
1.00
so «
.60 5
.40 g
20
60
50
40 „
30 i
20 ^
10
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
RE
ALEAfiN
NGS^
Ul
i 80
a
2 40
20
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
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.
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
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
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
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
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.
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.
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.
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.
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
Iron and Steel and Their Products
11
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 e conomically 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
9
9
88
81
II
Production of Steel Ingots and
'_
III
IV
Estimated Steel Shipments
Estimated Steel Bookings
41
90
19
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 
TVNews
OpenLibrary