(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Investigation of concentration of economic power. Hearings before the Temporary National Economic Committee, Congress of the United States, Seventy-fifth Congress, third Session [-Seventy-sixth Congress, third Session] pursuant to Public Resolution no. 113 (Seventy-fifth Congress) authorizing and directing a select committee to make a full and complete study and investigation with respect to the concentration of economic power in, and financial control over, production of goods and services .."

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 part of the decline in orders for steel that took 
place in the six or eight months following March, 1937, can be accounted for by 
the great volume of forward buying between October, 1936, and March, 1937, 
and the failure of actual requirements to come up to expectations after March. 

In s