(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; monograph no. 1[-43]"

^^3d Sessfon*^} SBNATE COMMITTEE PRINT 



INVESTIGATION OF CONCENTRATION 
OF ECONOMIC POWER 



TEMPORARY NATIONAL ECONOMIC 
COMMITTEE 

A STUDY MADE UNDER THE AUSPICES OF THE DEPART- 
MENT OF COMMERCE FOR THE TEMPORARY NATIONAL 
ECONOMIC COMMITTEE, 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 AND DISTRIBUTION 
OF GOODS AND SERVICES 



MONOGRAPH No. 3 
WHO PAYS THE TAXES? 

(Allocation of Federal, State, and Local 
Taxes to Consumer Income Brackets) 



Printed for the use of the 
Temporary National Economic Committee 




UNITED STATES 

GOVERNMENT PRINTING OFFICE 

WASHINGTON : 1940 



TEMPORARY NATIONAL ECONOMIC COMMITTEE 

JOSEPH C. O'MAHONEY, Senator from Wyoming, Chairman 

HATTON W. SUMMERS, Representative from Texas, Vice Chairman 

WILLIAM H. KING, Senator from Utah 

WALLACE H. WHITE, Jr., Senator from Maine 

CLYDE WILLIAMS, Representative from Missouri 

B. CARROLL REECE, Representative from Tennessee 

THURMAN W. ARNOLD, Assistant Attorney General 

•WENDELL BERQE, Special Assistant to the Attorney General 

Representing the Department of Justice 

JEROME N. FRANK, Chairman 

•SUMNER T. PIKE, Commissioner 

Representing the Securities and Exchange Commission 

GARLAND S. FERGUSON, Commissioner 

•EWIN L. DAVIS, Chairman 

Representing the Federal Trade Commission 

ISADOR LUBIN, Commissioner of Labor Statistics 

•A. FORD HINRICHS, Chief Economist, Bureau of Labor Statistics 

kepresenting the Department of Labor 

JOSEPH J. O'CONNELL, Jr., Special Assistant to the General Counsel 

•CHARLES L. KADES, Special Assistant to the General Couiisel 

Representing the Department of the Treasury 



Representing the tx irtment of Commerce 
LEON HENDERSON, Economic Coordinator 
DEWEY ANDERSON, Executive Secretary 
THEODORE J. KREPS, Economic Adviser 



Monograph No. 3 
WHO PAYS THE TAXES? 

GERHARD COLM AND HELEN TARASOV 



•Alternates. 

n 



ACKNOWLEDGMENT 

This monograph was written under the general supervision of 

GERHARD COLM 

Professor of Economics 

New School for Social Research 

Fiscal Expert, Department of Commerce 

BY 

HELEN TARASOV 

of the Industrial Economics Division 

Department of Commerce 

The Temporary National Economic Committee is greatly indebted 
to these authors for this contribution to the literature of the subject 
under review. 

The status of the materials in this volume is precisely the same as that 
of other carefully prepared testimony when given by individual witnesses; 
it is information submitted for Committee deliberation. N^o matter wfiat 
the official capacity of the witness or author may be, the publication of his 
testimony, report, or monograph by the Committee in no way signifies nor 
implies assent to, or approval of, any of the facts, opinions, or recom- 
mendations, nor acceptance thereof in whole or in part by the members of 
the Temporary National Economic Committee, individually or collec- 
tively. Sole and undivided responsibility for every statement in such 
testimony, reports, or monographs rests entirely upon the respective 
authors. 

Joseph C. O'Mahoney, 
Chairman, Temporary National Economic Committee. 

in 



TABLE OF CONTENTS 



Page 

Letter of submittal . vii 

Part I. — Scope and main results of the study . 1 

A. Scope and limits 1 

1. Theoretical aspects . 1 

2. Technical aspects 2 

B. Sources and methods ^ 3 

1. Measurement of income 3 

2. Classification of taxes 3 

C. R6sum6 of results 6 

Part II. — Procedures 9 

A. Consumer income and consumption patterns in 1938-39 9 

1. Incomes 9 

2. Savings - 10 

B. Value added by manufacture 11 

C. Taxes : - 12 

1 . Tax classification 12 

(a) Personal nonshiftable taxes 12 

(b) Taxes on specific consumption 13 

(c) Business taxes 15 

(1) Corporate income taxes 16 

(2) Other business taxes 16 

2. Tax allocation 16 

(a) Personal taxes 17 

(6) Specific consumption taxes 18 

(c) Business taxes allocated to sectors of the economy. 20 

'3. Tax shifting 23 

(a) Corporate income and profits taxes 23 

(b) Taxes shifted to consumption 24 

(c) Taxes shifted to wages 25 

Part III. Results 27 

A. Final tax patterns 27 

1. Personal and consumer taxes 27 

2. Taxes and income 27 

3. Taxes as an influence on consumption 30 

B. Influence of taxes on business 32 

C. Influence of taxes on savings 32 

Appendices: 

I. The defense-tax bill 34 

II. Selective bibliography 36 

III. Tax capitalization 38 

IV. Distribution of taxes on real estate by tvpe of owners 39 

V. The method of study. 42 

Index 53 

LIST OF TABLES 

I. Savings and all taxes as percent of consumer income, 1938-39 .. 6 

la. Personal and specific consumption taxes, as percent of consumer in- 
come, 1938-39 12 

lb. All taxes as percent of consumer income, 1938-39 13 

II. The distribution of customs duties by type of import 14 

III. Specific consumption taxes 1938-39 19 

IV. Taxes paid by sectors of indu.stry 22 

V. Business taxes as percent of income produced, 1938-39 22 

VI. Estimates of State sales taxes paid by commodity groups ._ 25 

VII. Total taxes as percent of consumer income, if business taxes were 

shifted to wages, 1938-39 26 



VI TABLE OF ^CONTENTS 

Page 
VIII. Taxes as percentage of national income in United States and Great 

Britain 27 

IX. Income and tax percentages for 1938-39, cumulated 1.. 28 

X. Taxes and income, 1932 28 

XI. Taxes in relation to consumer income in 1938-39, if 1932 tax rates 

had been in force in 1938-39 29 

XII. Taxes and savings as percent of consumer income per consumer unit, 

1938r39 J 30 

XIII. Consumer savings by income brackets, related to consumer incomes 

in 1935-36 and 1938-39.--- 33 

APPENDIX TABLES 

A. Percentage of defense taxes paid by each income bracket 34 

B. Defense taxes as percent of consumer income 35 

C. Basic data on consumer incomes , 42 

D. Gifts, savings, and expenditures 43 

E. Tax receipts, fiscal year 1939 44 

F. Personal taxes by income brackets 45 

G. Specific consumption taxes and all consumption taxes, by income 

brackets - 47 

H. Taxes on business shifted to consumers and shareholders 50 

J. All personal taxes, including taxes shifted to .shareholders 51 

DIAGRAMS 

I. Personal and specific consumption taxes as percentages of consumer 

income, 1938-39 _ 4 

II. Original personal and specific consuirption taxes phis business taxes 

shifted forward, as percentages of consumer income, 1938-39 5 

III. Distribution and disposition of consumer income — United States, 

1938-39- _ - 7 

A. Consumer units and income by income brackets 7 

B. Consumer taxes and consumer savings 7 



LETTER OF SUBMITTAL 



The Honorable Joseph C. O'Mahoney, 

Chairman, Temporary National Economic Committee, 

United States Senate, Washington, D. C. 
My Dear Senator: I have the honor to submit herewith a report 
on Who Pays the Taxes? (Allocation of Federal, State, and Local 
Taxes to Consumer Income Brackets). It is an exploratory study, 
the results of which must be regarded as tentative. I believe, how- 
ever, that even these tentative results shed light on certain economic 
aspects of the existing tax system and may be useful in formulating 
a tax policy which takes into account the economic repercussions of 
taxation. 

The report has been prepared by Miss Helen Tarasov in consulta- 
tion with Gerhard Colm. The authors had the benefit of consultation 
with various members of the Industrial Economics Division of the 
Office of the Secretary, and of Marius Farioletti. 
Respectfully submitted. 

Ernest A. Tupper, 
Assistant Economic Adviser, Department of Commerce. 

VTI 



WHO PAYS THE TAXES? 

ALLOCATION OF FEDERAL, STATE, AND LOCAL TAXES TO CONSUMER 

INCOME BRACKETS 

Part I.— SCOPE AND MAIN RESULTS OF THE STUDY 

A. SCOPE AND LIMITS 

1. Theoretical aspects. — The distribution of the tax burden among 
people of various income brackets is an essential criterion not only of 
the justice of the tax system but also of its economic effects. The 
extent to which taxes curtail the purchasing power of the masses or 
absorb mcomes of the higher brackets or fall upon consumption or on 
savings cannot be known without studying the incidence of taxes on 
people in the various income brackets. No rational tax pohcy can be 
developed without knowledge of these basic facts. 

It is not surprising, therefore, that many attempts have been made 
to measure the impact of the tax system upon people of various income 
brackets. Usually these measurements have been based upon an 
analysis of the taxes paid directly plus. those indirectly included in 
the various expenses paid by famihes of a typical size, of a typical 
economic status, and residing in a specific locality.^ 

In this report an attempt has been made to give a break-down of all 
Federal, State, and local tax revenues, according to the income brackets 
of the tax bearers. Thereby the results of the sample studies are 
supplemented by an over-all picture. 

Before presenting the results it is necessary to emphasize that an 
analysis of the progressive and regressive elements in the tax system 
is essential for judging its equity and its economic effects, but they are 
by no means the sole elements that must be considered. Other factors 
to be considered can only be mentioned here: 

(a) For judging taxes, the use made of their yield should not be 
ignored. 

In this analysis, for instance. Social Security taxes are included 
without regard to the distribution of benefit payments. Business 
taxes, too, are treated without considering Government services 
rendered to business. For a comprehensive analysis of the effects 
of the fiscal system on the distribution of real income, the benefits 
derived from free services of the Government (education, pubUc 
hygiene, etc.) should not be neglected. 

(b) The type of tax should be considered. Among the levies which 
form the main part of the tax burden of the lower income groups are 
(besides Social Security taxes) poU taxes, excise taxes, and the part 
of property taxes which affects the costs of housing. A poll or head 

1 See, for instance, the very careful analysis of this sort published by the Twentieth Century Fund— 
Studies in CuTrerU Tax Problems, 1939. Research Director Carl Shoup, Associate Directors Roy Blough and 
Mabel Newcomer. For Great Britain, see Report of the Committee on National Debt and Taxation, 1927 
(so-called Colwyn Report). 



2 CONCENTRATION OF ECONOMIC POWER 

tax; a tax on a life necessity, such as housing; a tax on mass luxuries, 
such as tobacco ; or on what may be necessity and luxury at the same 
time, as gasoline, may fall upon people of the same income bracket and 
yet be quite different from the point of view of justice or social hygiene. 

(c) The indirect costs of various taxes, resulting from the cost of 
administrative expenses and economic frictions, should be considered. 

Various taxes imply not only different costs of collection but also 
varying costs of payment for the taxpayer and result in diverse 
degrees of tax resistance, a factor not to be overlooked in judging a 
tax system. The economic frictions also differ as between taxes. 
This aspect, too, is outside the scope of this study but should not be 
ignored in a comprehensive appraisal of a tax system and of reform 
proposals. 

Thus, although the aspect of progressiveness and regressiveness 
presented in this study is fundamental — it is probably the most 
essential single aspect for judging a tax system, nevertheless, it is not 
the only criterion which should be applied in deciding on tax policy. 

2. Technical aspects. — A study of the progressive and regressive 
elements in the tax system is further limited by difficulties inherent 
in the problem itself. Such a study belongs to the type of analysis 
which may be called constructional statistics. For this type of 
analysis the same accuracy cannot be claimed as for the usual sta- 
tistics based on factual counting. The National Resources Planning 
Board's estimates are the best available source of information on the 
distribution of consumer income, expenditures, and savings, even 
though they are based on a relatively small sample. The figures on 
tax revenue are taken from actual statistics, but the subdivision of the 
tax revenue often does not correspond to the break-down needed for 
this economic classification. In this respect, too, estimates had to 
take the place of actual statistics. Nevertheless, the study does give 
a picture of the comparative relation between the taxes and incomes 
of the various brackets. The reasonableness of the results has been 
tested by varying some of the basic figures within the range of prob- 
ability. Even assuming possible errors in certain estimates (especially 
with respect to the extrapolation of incomes, expenditures, and 
savings) the results wouljd not change fundamentally. Yet it is 
frankly admitted that this is an experimental approach that requires 
further refinement. It is hoped that the 1940 Census of Population, 
in conjunction with a more detailed analysis of the 1939 income-tax 
returns, will permit more accurate statistics on the distribution of 
incomes than the one used in the present report. 

A second difficulty arises from the fact that even the most complete 
statistical material can never measure the incidence of taxation. The 
incidence of taxes can be derived from figures on tax payments only 
on the grounds of theoretical reasoning and hypothetical conjecture. 
Whatever statistical refinement is achieved, therefore, the results 
necessarily rest on a number of assumptions. The incidence of 
taxation depends, among other factors, on the. degree of competition 
and monopoly prevailing in an economy, on the direction of Govern- 
ment regulation of prices and cost factors, on the general trend of 
economic growth or stagnation, on the development of labor produc- 
tivity, on the time which has elapsed since the introduction of new 
taxes or the increase (or decrease) in tax rates, and, finally, it depends 
on the use made of taxes and on the other fiscal policies (e. g., borrow- 
ing) pursued at the same time. 



CONCENTRATION OF ECONOMIC POWER 3 

With respect to the assumptions concerning incidence, a crude test 
of reasonableness has also been applied. The same computation has 
been made twice, once assuming that business-cost taxes are shifted 
forward to prices, then assuming that they are shifted backward to 
wages. With respect to the general patterns of progressiveness or 
regressiveness of the tax system the results of the two computations 
do not differ substantially. 

B. SOURCES AND METHODS 

1. Measurement of income. — The National Resources Planning 
Board publications on Consumer Incomes and Consumer Expenditures 
in 1935-36 were used as a basis for estimating the distribution of 
personal incomes and the patterns of expenditure and saving in the 
various income brackets. The Board's 15 income groups were con- 
densed to 10 classes of gradually increasing intervals for convenience 
in handling. Its concept of "consumer units," meaning both families 
and individuals, was adopted. Likewise, the Board's saving and gift 
figures were taken as a starting point, and its conclusions as to per- 
centages of expenditures directed to various types of goods and 
services by the different income brackets were used as the basis for 
allocating a wide range of consumer taxes. The Bd9,rd's estimates 
for 1935-36 were extrapolated to the year 1938-39 with the help of 
the Department of Commerce estimates of national income. The 
method applied for this extrapolation is discussed in part II, section 
A. 1. 

Business taxes were related to business activities measured by the 
income produced by production and services. These estimates were 
derived from the statistics of national income prepared by the De- 
partment of Commerce. 

These national income figures were conjoined to the Board's data 
in Structure of the American Economy in order to extend the latter 
figures to 1938-39. Extrapolation was naturally impossible in view 
of the uneven proportion of income produced by industries at different 
stages. The assumptions imposed by incomplete or defective data 
and the step-by-step technique of calculation are summarized below 
in part II B, and presented in greater detail in appendix V. 

Before any allocation to consumer or produced income of the 
various taxes could be made, however, it was necessary to supplement 
the ordinary fiscal classification of reveni by an economic grouping. 
Such economic classification is quite dinicult because one tax fre- 
quently, if not usually, contains elements of diverse economic signifi- 
cance. The property tax especially exemplifies the complex nature 
of taxes. 

2. Classification of taxes. — In classifying taxes according to their 
incidence three tax groups may be suggested : 

(a) Taxes on individuals (poll, income, estate and gift, non-business 
personal property) attributed to individual income taxpaj^ers under 
the presumption that they normally cannot be shifted. 

(b) Excise taxes on specific products (like tobacco, liquor) which 
can be attributed to the consumers of these specific products under 
the assumption that they can be and usually are shifted forward by 
the original payers. Property taxes on home-owners also fall into 
this category. These first two groupings should each be subdivided 
according to income classes, if variations in actual incidence and effect 
are to be observed. 



CONCENTRATION OF ECONOMIC POWER 



Diagram I 











^ 




Wi, 




m^ 


^ 


^^^^s 






i^ 


??5S? 




fc 


^e 


^^M 


^ 


« 




10- 


c 














-/o 


PERSONAL 


20- 




1 

irat cr 
\umtct 
■nma 

UCALU 

Tins) 


ILSS Ttnsi PIS 
CEISESi PAIS OF 


no UK 

13, HOST natau. 

US; mtS 3SL- 
T OP STATE AND 

csn&u. FRopjan 








■20 


AND 

SPECIFIC 

CONSUMPTION 

TAXES 

AS 


C) 30- 
















-30 


PERCENTAGES 
OF 


i 


















CONSUMER 


iS 


















INCOME 


^ 


















1938-1939 


r 

It! . 


































- 


IncQ— C1*A— » 


i^ 


















• I uiKl.r I5O0 


S 


















II ejOO • 1000 


C5 


















III 1000 • 1500 


^^0- 
















-40 


IT IJOO • MOO 


^ 


















-T 2000 ■ 3000 


TAXES AS PERCENT 

M Co 

1 1 






(nmu. ira a 

nsmiKZ OQRi 
ID mm tatia 

POU, TinS) tK 


CATS muTunu. 
B im aiR Tuxsi 

t-TnimS FEUOUL 








-30 
-20 


TI JOOO • JOOO 

nt 5000 ■ 10,000 

VIU 10,000 " 15,000 

n 15,000 • 20,000 

Z 20,000 ud ormt 


^ 
^ 






















IBB 




















t^TBnii. 


fo- 
ri 


n 








k:)W>» 


i 


S ^^M 


-/a 


I????3SS'" 


in n iz IT m Msnijxx 




/NCOME CLASSES, BY PiPCENT OF AGGREGATE INCOME 


J. nPAum w oamaa 
0D40-3IS 



CONCENTRATION OF ECONOMIC POWER 



Diagram II 




20 



30- 



40 



30- 



•qso 



/o 



icaatoB mramsi fzdbui. umricnims 
aiLB tiiPi mm no locil ucasni 



imasu. UD STUB laiminu- ncou 

T1£SS) DdOCSS OLD iOS IJCSDBABCB 
CONTBIBUnCSS; TEDSSiL UOt STATE ES- 

TiTi im can tuxsi fou tussi vd- 

THTWTW PKBSOUL FtOPERrT TAXES[ AKD 




ORIGINAL 
PERSONAL 

AND 
SPECIFIC 
20 C0NSUM!3TI0N 
TAXES 
PLUS 
BUSINESS 
30 TAXES 

SHIFTED FORWARD! 
AS 
PERCENTAGES 

OF 
CONSUMER 

INCOME 
1938-1939 



40 



II IE EZ Y 21 SIMIXZ 

INCOME CLASSES, BY PERCENT OF AGGREGATE INCOME 



I 


unlar 


liOO 


11 


»500 




1000 


III 


1000 




1500 


17 


1500 




2O0O 


V 


JOOO 




3000 


VI 


3000 




5000 


VII 


;ooo 




u,ooo 


VIII 


10,000 




15,000 


n 


i;,ooo 




20,000 


I 


20,000 


•nd 


tyin 



^ fS?*- 



u. BFurgoit or aamaet 
DD40-324. 



6 



CONCENTRATION OF ECONOMIC POWER 



(c) For taxes on business in general, no generalized assumption 
with respect to incidence as a whole can be made. These levies, 
such as pay-roll taxes and sales taxes, will be shifted to prices under 
certain business conditions; under other conditions they will be shifted 
backward to the factors of production (wage earners, producers of raw 
materials) or they will be partly borne by profits. For this group of 
taxes the ultimate tax bearer cannot be determined without a special 
analysis of economic conditions. In the first stage these taxes are 
therefore treated separately as business taxes and related to the various 
branches of industrj^ where the taxes were collected. For the final allo- 
cation of these taxes to individuals, the analysis must proceed under var- 
ious assumptions which will be discussed later. For the computation 
summarized in the following section, it has been assumed that — 

(1) Corporate income, profits, and capital-stock taxes are eventu- 
ally borne by stockholders, whose income would have been higher in 
the absence of such levies. These taxes were therefore finally dis- 
tributed on the basis of stockholdings by income brackets. 

(2) All other business- taxes are finally shifted to consumption. 
They were therefore allocated in proportion to the corresponding 
expenditures. 

C. RESUME OF RESULTS 

The main results of the study can be summarized in the following 
table: 

Table I. — Savings and All Taxes ' as Percent of Consumer Income, 1938-39 ^ 



Income classes 



I. Under $500 

II. $.500 to $1.000... 

III. $1,000 to $1.500.. 

IV. $1,500 to $2,000.. 
V. $2,000 to $3,000.. 

VI. $.3,000 to S-^OOO-. 
VII. $5,000 to $10,000. 

vin. $10,000 to $15,000 

IX. $15,000 to $20,000 
X. $20,000 and over. 

Total 



Total con- 
sumer in- 
come (in 
millions 
of dollars) 



2,363 

10, 038 

12,280 

10, 210 

12, 194 

7,743 

4,861 

2,238 

1,601 

6,333 



69. 861 



Taxes as percentage of income 



Federal 



7.9 
6.6 
6.4 
6.6 
6.4 
7.0 
8.4 
14.9 
19.8 
27.2 



9.2 



State 
and local 



14.0 
11.4 
10.9 
11.2 
11. 1 
10.6 
9.5 
10.6 
11.9 
10.6 



11.0 



Total 



21.9 
18.0 
17.3 
17.8 
17.5 
17.6 
17.9 
25.5 
31.7 
37.8 



20.2 



Savings as 
percentage 
of income 



-24.3 

-2.0 

5.2 

."5.8 

9.6 

16.8 

28.4 

32.3 

32.3 

38.3 



11.4 



1 Business taxes were assumed to be shifted to consumer income. 
i See diagram III. 

Sources: Income and savings derived by extrapolation of National Resources Planning Board figures in 
Consumer Incomes in the United States and Consumer Expenditures in the United States, Washington, 
1939. For original tax figures and their sources, see appendix table E. 



Federal taxes (which include Social Security taxes) impose an 
approximately flat-rate burden on incomes up to $10,000, with a very 
slight regression for incomes below $500, measured as percent of 
income, and with a slight progression above $3,000. Beginning at 
$10,000, the progression increases substantially. 



CONCENTRATION OF ECONOMIC POWER 
Diagram III 

DISTRIBUTION AND DISPOSITION OF 
CONSUMER INCOME 

UNITED STATES, 1938-1939 



Income Brockets 



I UNDER $500--- 
JI $50O-$l,0O0-- 

in $i,ooo-$i,5oo- 



A— CONSUMER UNITS AND INCOME 
(BY INCOME BRACKETS) 



Percent of Percent of . 

Consumer Units Aggregate t " 

Income Received 
17.0 3.4 



12 $1,500- $2,000 

I $2,OOO-$3,0OO 

21 $3,OO0-$5,0OO 

mi $5,000 -$10,000 

3nn $10,000 -$15,000- 

IZ $15,000 -$20,000 

X $20,000 AND OVER 

t Income tncludes Deoih and Corporote 



-29.5- 
-22.1 - 
-13.1 - 

- I 1 .2 - 

- 4.6- 

- I .5- 

- .4- 
-- .2 
-- .3 



14.4- 

17.6- 

14.6 - 

■ 17.4- 

II.O- 

-7.0- 

-3.2- 

--2.3 

- 9, 1 ■ 



•x^yy^AAy^yyx^y^^x^xyyA 



•^y^yyy^xyyA^yy^yy^yy^^y^yy, 



^^^^^^^^^ 



^^^^^^^ 



PER CENT 
10 15 20 



>^JJJJJjm 



^^^ 



V//////M 



W///////A 



'Z2Z 



l/yy///yyyyyy>'^ 



^9' 



15 20 

PER CENT 



I ond Pro'tis Toi9S. 



B— CONSUMER TAXES AND CONSUMER SAVINGS * 

Income Brackets 



m IZ 3t zzisnsniizz 




Average of 
All Brackets 






^ 




90 


- 




- 90 


60 


- 


■ 


- 80 


70 


- 




- 70 


60 


- 




- 60 


50 


- 




- 50 


40 


- 




- 40 


30 


- 




- 30 


20 


~ 




- 20 


10 


.'.•. 


- 10 


n 




.•••. 


Jo 



entoges of Consumer 



elude Business Toios shifted forward. 



g CONCENTRATION OF ECONOMIC POWER 

State and local taxes, on the other hand, hover m the vicinity of 
10 percent of the aggregate income of each income bracket except the 
lowest, where they amount to 14 percent of income. 

Even if this study presents merely reasonable estimates, it does 
indicate with a high degree of probability the contemporary tax 
pattern. It may be regarded as a scouting expedition into unmapped 
territory in order to delimit known and unknown regions and thereby 
at least blaze a trail for more thorough expeditions. And the study 
indeed shows wide blank areas to be filled in. 



Part II.— PROCEDURES 

A. CONSUMER INCOME AND CONSUMPTION PATTERNS IN 1938-39 

The National Resources Planning Board's studies of the distribution 
of incomes and the patterns of income utilization are based on a sample 
collected for the year 1965-36. This sample covered a large and 
representative number of families and individuals living alone, but 
excluded institutional residents from the general tables showing 
income and spending of consumer "units" at various income levels. 
The Department of Commerce estimates of the national income have 
been made for 1929 and subsequent years, but do not present a size 
distribution of incomes nor a break doAvn of income into consumption 
and saving. Therefore it was necessary to adapt the patterns of 
incorne distribution and income use of the year 1935-36 to the total 
income of 1938-39. 

1. Incomes. — For the purpose of distributing the much higher 
total of income payments in 1938-39 (Department of Commerce 
figures) among a slightly larger number of consumers than in 1935-36, 
it was assumed that the number of consumer units had grown in direct 
ratio to the rise in population; i. e., that family structure and the 
proportion of single individuals had not altered since 1935-36. The 
further and more arbitrary assumption was made that the number of 
consumer units in each income bracket as well as in the aggregate 
had increased in direct proportion to population growth. On the 
basis of this hypothesis, consumer income was extrapolated (with 
allowance for the number of consumer units per group) in proportion 
to the increase in total national income payments. In the absence 
of data, gifts ^ were also extrapolated proportionately, although no 
valid rationale supports this method, other than the well-known 
sluggishness in the change of expenditure habits. Because of changes 
in income and taxation, savings ratios were altered much more dras- 
tically. (See below, subsection 2.) Nonprofit institutions were 
ignored because their classification by income brackets would bear no 
relation to individual income classes; furthermore, since satisfactory 
information on their aggregate expenditures is lacking, there seems no 
proper niche for them. Inasmuch as they pay no direct taxes and 
frequently are allowed refunds on indirect ones, their elimination 
is not serious for research concerned primarily with tax burdens. 
The National Resources Planning Board investigation, which gave 
$724,000,000 as the income of such groups in 1935-36, also excluded 
such institutions from its major analyses. 

Before these income figures, as obtained by extrapolation, were 
accepted, another income calculation was made separately, using 
data of the Department of Commerce; income by income classes was 

' Gifts and personal taxes were obtained from Consumer Expenditures, tables lOA and 19A. The latter 
shows total gifts and taxes for families and individuals, while lOA gives taxes and gifts separately for families. 
The total for families was subtracted from the total for consumer units to get comparable individual taxes 
aid gifts. Taxes and gifts were assumed to be in the same proportion for individuals as for families. Taxes 
■when taken from the National Resources Planning Board study always mean personal taxes as defined by 
the Board, viz, individual income taxeis, poll taxes, and some personal property taxes. 

9 
254815—40 — t^o. 3 3 



IQ CONCENTRATION OF ECONOMIC POWER 

computed separately for wages and salaries, dividends and interest, 
entrepreneurs' withdrawals, rents and royalties. Government pay 
rolls, and relief, and totaled. The 1938-39 figures thus arrived at 
were adjusted in line with the deviation of comparable 1935-36 
figures from the corresponding National Resources Planning Board 
data to correct for conceptual differences. Since the final results 
were close to the figures found by the simpler method of extrapolation, 
the more cumbersome approach was deemed unprofitable. A further 
check was provided by the Commerce percentages on the share of 
national income going to wages and salaries, etc., which has risen very 
slightly; total relief expenditures in the fiscal years rose between the 
two periods but not so much as to be proportionately greater than 
would be requisite to conserve the 1935-36 level of aid, in view of the 
increase in the number of consumer units and the return of unemploy- 
ment to the 1935-36 level; so that all in all, it is very probable^ that 
the relative aggregate incomes of the brackets did not undergo much 
change. 

This check was particularly valuable because figures extrapolated 
from 1935-36 to 1938-39, two periods.of dissimilar business conditions, 
would be peculiarly liable to challenge. The depression may in many 
cases have caused a temporary shifting of consumers into lower income 
classes, where their expenditures might have continued, through 
recourse to borrowing or other forms of dissaving, at a level correspond- 
ing to their former higher incomes. Anticipations are decisive here, 
for it is the consumer expectation to return to "normal" that deter- 
mines his spending pattern. Sharp changes of standards of living 
are not welcomed, and optimism as to a future restoration of earning 
power may prevail for years, leading to progressive growth of indebted- 
ness, eventually reflected in property-tax delinquencies, repossessed 
articles, bad mortgages (between 1929 and 1933 one- third to one- 
half the outstanding mortgages were defaulted), and bad charge 
accounts written off by the respective creditors. Once the consumer 
becomes adjusted to the idea of his new and lower income level and 
its possible permanence he may seriously revise his budget. It is 
very probable that the deflationary process sketched above had about 
spent itself hi 1935-^36, and that the extrapolation of income figures 
into 1938-39 gives a figure reasonably clo^e to actuality. 

On the other hand, caution had to be used with regard to the high 
dissavings figure of 1935-36, when the cumulative effects of depression 
h.ui exhausted personal savings, capital values were still being drasti- 
cally written down, and, despite the revival of business optimism, 
unemployment was of a size to keep millions of consumers b'-^ow their 
customary income level. The possibly temporary rather than 
habitual nature of such dissavings — by people who had accumulated 
a backlog of capital, monetary and real, available for dissaving — 
should be given due 'weight, especially in analyzing expenditure 
patterns of the various income classes and allocating tax burdens. 
The trend toward loss of homes, both urban and rural, but affecting 
particularly the low income farm groups (cf. 9.1 percent increase in 
farm tenancy from 1925 to *1935) had been definitely reversed by 
1938-39 by the various housing and tenancy programs. 

2. Savings. — The correction for these features of the depression was 
introduced by a diversified approach in estimating the savings of 
various income brackets. In view of the above considerations, 
dissavings were cut down more than proportionately by assuming 



CONCENTRATION OF ECONOMIC POWER H 

that the lower income brackets used most of their additional income 
to avoid further increases of indebtedness; however, the continued 
existence of net dissavings in the two lowest income brackets (in- 
comes under $1,000) allows for the National Resources Planning 
Board's definition of "negative savings" as including increases in 
installment payments due, mortgages, charges, and notes due. For 
the lower and middle incomes (under $2,000), therefore, gifts and 
direct taxes were treated as reducing net income available for spending. 

Net consumer savings (savings less dissavings) were a'='sumed to 
have risen more than the national income, on the basis of rising 
incomes. The computation by income brackets produced a figure 
corresponding to the probable total savings indicated by several 
independent estimates. These check figures were based on — 

\l) The rate of increase in savings as evidenced in a comparison of 
a score of forms of saving; 

(2) The National Resources Planning Board's Structure of the Amer- 
ican Economy figure of $9,000,000,000 under the hypothesis of the 
national income's rising by units of $10,000,000,000. 

The checks were the more impressive in that in the highest income 
brackets expenditures rose more than did savings, since it was as- 
sumed that persons in these brackets paid taxes and made gifts out of 
that part of increased income, which would otherwise have been 
saved. (See table XII.) The conclusion that increased taxes were 
paid by reducing the percentage of savings is supported by the replies 
to private questionnaires sent out by Congressman Bruce Barton in 
late 1939 to a sample of families in the $6,000 to $40,000 income 
classes. He inquired where they would cut expenditures if income 
taxes were raised 10 percent. Only half of those replying admitted 
the necessity of making cuts, and of those 53 percent said they would 
lessen savings. Thus, 75 percent of the upper brackets intimated 
that they paid additional taxes out of savings. A further number 
(the percentages as to fields for cutting overlap, of course, since some 
of the families would split their economies over several fields of 
expenditure) planned to economize on renovation of houses and 
additions to them, or not to increase life insurance. Without passing 
any judgment here as to the general economic effects of lesser expendi- 
tures in these fields, it should be pointed out that these, too, are re- 
garded as savings in the National Resources Planning Board study, 
and the questionnaire's results therefore bear out the approach used 
in this monograph. 

B. VALUE ADDED BY MANUFACTURE 

A number of business cost taxes have an incidence which is very 
problematical and possibly variable. Such taxes may be absorbed 
by the business paying them, passed to another branch of business, 
shifted back to wages, or forward to the consumer, but it was neces- 
sary to make a preliminary allocation to the business initially paying 
them. 

This allocation was based upon the income produced by each of 
eight branches of industry, as given by the Department of Commerce. 

The National Resources Planning Board used the same figures in 
its Structure oj the American Economy, 1935-36 (although it intro- 
duced certain minor modifications described on p. 378 of the publica- 
tion). The branches are agriculture, mining, manufacture, utihties 
and transportation, construction, trade, finance, and services to the 



12 



CONCENTRATION OF ECONOMIC POWER 



consumer — plus a small item labeled miscellaneous. Government 
''income produced" was deducted, since it carries no tax burden. 
Employers' Social Security taxes were added to the figures of income 
produced by each sector. 

The Commerce figures segregate income produced on the industrial 
basis desired for this study. They had to be supplerriented for tax 
allocation purposes, however, by data from the Statistics of Income. 
This source classifies corporate returns according to the main business 
of a corporation, although it may conduct many diverse operations in 
other industries. These inconsistencies should at least be recognized 
and adjusted when possible. Furthermore, the Statistics of Income 
covers corporate enterprises only. Corporate production totals about 
two-thirds of the national income produced, but the proportions within 
different segments of the economy vary from 92 percent of manufac- 
turing to an insignificant part in agriculture. 

The preliminary allocation of taxes which follows does not pretend 
to establish the pattern showing final distribution of the tax burden. 
But in analyzing new taxes or changes in tax rates, it is important to 
know by whom the taxes are initially paid, even if they are shifted 
backward or forward. The process of shifting in itself creates fric- 
tions and may affect business methods.^ Table V, therefore, only 
indicates the approximate importance of the taxes which an industry 
must absorb, pass on, or shift back. It does not represent the tax 
burden of the particular industry. 



C. TAXES 

1. Tax classification. 

(a) Personal nonshiftable taxes. — Personal nonshiftable taxes have 
already been discussed in previous sections of the report. They in- 
clude individual income taxes, estate and gift taxes, poll taxes, em- 
ployees' old-age security contributions, and two-thirds of the personal 
property taxes. Tt is justifiable to presume from the very nature of 
these taxes and the fashion of their imposition that they cannot be 
shifted and that they are of a personal nature. 

Table la. — Personal and Specific Consumption Taxes in Percent of Consumer 

Income,' 2 1938-39 



Income classes 



I. Under $500 

II. $500 to $1,000 

m. $1,000 to $1,500... 

IV. $1,500 to $2,000... 

v., $2,000 to $3,000... 

VI. $3,000 to .$5,000... 

VII. $5,000 to $10,000- 

vin. $10,000 to $15,000 

IX. $16,000 to $20,000 
X. $20,000 and over. 



Personal taxes 



Federal 



Percent 
1.1 

.65 

.7 

.7 

.6 

.9 

1.7 

3.8 

6.7 

18.7 



State and 
local 



Percent 

0.2 

.2 

.3 

.4 

.5 

1.0 

1.5 

4.1 

6.4 

6.9 



Taxes on specific con- 
sumption 



Federal 



Percent 
3.0 
2.9 
2.8 
2.8 
2.7 
2.4 
1.9 
1.5 
1.3 



State and 
local 



Percent 

4.3 
3.5 
3.6 
3.9 
4.0 
3.7 
3.1 
2.7 
2.3 
1.6 



1 See diagram I. 

s Unlike tables I and lb, consumer income here does not include estate, gift, and inheritance taxes, nor 
corporate income and profits taxes, so that percentages are not directly comparable, particularly in the 
higher brackets. 

S«til«es: See table I. 



« Cf. Temporary National Economic Committee, Cliflord J. Hynniijg, Taxatidn of. Corporate Enterprise 
Washinijton, 1940, especially chapters on, payroll and corporate income taxes. 



CONCENTRATION OF ECONOMIC POWER 



13 



(b) Taxes on specific consumption. — Taxes on specific consumption 
goods with an inelastic denrand can be shifted to the consumers more 
easily and with fewer general repercussions than can taxes imposed 
on all production or sales. A tax on such a commodity can be shifted 
to its price because the demand is of such a nature that consumers 
are willing to increase their outlays upon it. Under ordinary circum- 
stances this increase is at the expense of outlays upon the whole range 
of other commodities or at the expense of saving. Outlays for all 
commodities at a given income level, however, can be increased simul- 
taneously only at the expense of saving, which is negligible in the in- 
come areas that account for most of the aggregate consumer expendi- 
tures. For this reason a general business-cost tax cannot be passed 
forward to price without an increase of total consumer expenditure, 
which would require conditions of general economic expansion. Other- 
wise an increase in prices resulting from a rise in business-cost taxes 
will cause sales to be curtailed and will result in decreased employment. 
This in turn may lead to a reduction in wages and consequently a 
backward shifting of the tax. It is on the basis of this argument that 
a distinction was made between specific consumption and general 
business-cost taxes. 

Table lb. — All Taxes as Percent of Consumer Income, 1938-39 i 



Income classes 



As percent of income ^ 



Personal taxes 2 



Federal 



State and 
local 



Taxes on consumption 2 



Federal 



State and 
local 



I. Under $500 - 

II. $500 to $1,000 

III. $1,000 to $1,500... 

IV. $1,500 to $2,000.-.. 
V. $2,000 to $3,000... 

VI. $3,000 to $5,000... 

VII. $5,000 to $10,000 . 

VIII. $10,000 to $15,000. 

IX. $15,000 to $20,000- 

X. $20,000 and over.. 



1.1 

.7 

.7 

1. 

1.0 

2.3 

4.8 

12.1 

17.3 

25.5 



0.2 
.2 
.3 
.5 
.5 
1.1 
1.8 
4.5 
6.6 
7.0 



6.8 
5.9 
5.6 
5.6 
5.3 
4.7 
3.7 
2.9 
2.5 
1.6 



13.8 

11.2 

10.6 

10.8 

10.6 

9.5 

7.7 

6.1 

5.2 

3..6 



Total. 



4.3 



9.6 



1 See diagram II. 

» Same as in tables I and la. 

' See note 2 to table la. 

Sources: See table I. 

Specifically, taxes can be regarded as falling directly on consumption 
when they are levied on selected finished products for final consump- 
tion, whether collected at the source, as manufacturers' excises, or 
directly from the final consumer, as on gasoline. But in allocating 
such taxes to consumer expenditures, it is important to get the net 
amount by first deducting excises passed on to intermediate forms of 
business (other than distribution) and shifted only indirectly to con- 
sumers, via services or transformed goods. As an example, gasoline 
and other automotive taxes paid by trucks and busses may be cited. 
If these levies are passed on to the consumer of transportation, they 
can be passed on only in the form of price for the service, and not 
directly. Such levies on taxicabs and on private cars used for business 
purposes (as by salesmen, realtors, doctors, etc.), ought also be rele- 
gated to the class of business taxes, but a lack of adequate data on 
these distinct uses of "passenger cars" precludes this accuracy. 
Similarly, the Federal tax on bituminous coal can be passed on to the 
buyer of such coal in full, even though it is included in the price so 



14 



CONCENTRATION OF ECONOMIC POWER 



that he is not conscious of it. But if a raihoad buys the coal, beafs the 
tax, biu"ns the coal for energy, and shifts the tax as one of many 
components in the price of a passenger ticket, it is a business and not a 
specific consumption tax. 

Most Federal excises were treated as taxes on specific consumption ; 
exceptions were made for a third of automotive taxes, four-fifths of 
those on utilities, and small portions of a few other levies on products 
which are partly bought by industry itself either for transformation 
(processed vegetable oils) or sale of its service (toilet preparations). 
The distribution of automotive taxes was based on figures of theAuto- 
mobile Manufacturers' Association, which show that for the last few 
years a fairly steady proportion of one-fourth of total Federal auto- 
motive excises (including gasoline) were paid by trucks. This ratio 
was raised to one-third to cover busses and cabs, but even the corrected 
amount may still have overvalued the direct consumer share. ^ Of 
the electrical energy excise, one-fifth was regarded as a consumers' tax 
on the basis of the exact ratio between sales to residential consumers 
plus half the sales to farms (many of which are supplied by publicly 
owned plants) and total sales. (See Statistical Abstract, 1938.) 

Many State and local imposts were likewise included as specihc 
consumption taxes. Among them are gross -receipts taxes on amuse- 
ments, selective sales and other taxes on tobacco, all alcohol taxes, 
motor-vehicle licenses (except for the share of about a third imputed 
to transportation) and other taxes. 

The proper treatment of customs was somewhat puzzling. It 
would have been correct to allocate to this classification customs duties 
paid by people importing products for their own use and duties on 
finished consumption goods intended for sale to ultimate consumers, 
while duties on commodities to be subjected to further manufacture 
would then be classed as business taxes. However, data are not pub- 
lished in this form, so a rough distribution was made among trade 
and agriculture,* and specific consumption, under which were grouped 
duties on foods, tobacco, and alcoholic beverages, as shown in table II, 

Table II. — The Distribution of Customs Duties by Type of Import 



Item 



1938-39 



Total customs duties 

To agriculture (30 percent) '_-. 

On foods 

On tobacco 

On spirits, wines, and liquors 

Total- 

On trade 



$408, 127. 000 



$318, 352, 000 



122, 438, 000 
131, 150,000 
25. 243, 000 
43, 589, 000 



95, 056, 000 
102, 636, 000 
23. 927, 000 
34, 206, 000 



322, 420, 000 
85, 707, 000 



255, 825, 000 
62, 527, 000 



I See text, pp. 14 and 22f. and footnote 4 on this page. 

Source: Annual Report of the Secretary of the Treasury, 1939, pp. 385-388. 



3 The same source estimates 62 percent of passenger-car mileage as being for business purposes, which 
would indicate consumer-paid motor-vehicle taxes to be overestimated in the present study. However, 
it was felt preferable to give a minimum estimate for such business-paid taxes on the basis of actual figures 
rather than an estimate which does not account for actual gasoline tax and registration fee costs. 

* The attribution of 30 percent of customs receipts to agriculture is admittedly a procedure of expediency. 
Detailed studies tracing the shifting of tariff duties have been made, but only for speciflo-products. (Cf. 
Lippert S. Ellis, The Tariff on Sugar, Raleigh Foundation, Freeport, 111., 1933; Haldor R. Mohat, The 
Tariff on Wool, Tariff Re.search Committee, Madison, Wis., 1935.) 

The 30 percent was the figure arbitrarily selected as the proportion of customs receipts to be allocated 
to the Department of Agriculture in order to compensate farmers for their higher costs due to other tarifls. 
In the final allocation of all taxes as carried out in this study, the precise distribution of customs receipts 
among branches of industry (aside from duties considered to be inevitably levies on specific consumption) 
is of no moment; it is Important only for the intermediate allocation of business taxes to branches of industry. 



CONCEMTRATION OP ECONOMIC POWER 15 

A further problem of tax distribution inherent in the levy itself was 
in the analysis of general property taxes. These taxes are a conglomer- 
ation of imposts on real and personal, tangible and intangible property. 
Coverage, assessment methods, tax rates, and exemptions vary from 
community to community without fiscal logic, and data are most 
unsatisfactory, since they are either so comprehensive as to embrace 
all variants of property taxes, regardless of comparability, or else 
strictly limited, usually on a basis that does not permit extension of 
estimates. 

The part of property taxes imposed oh residential real estate was 
treated as a specific tax on consumption, i. e., on housing. Therefore, 
an estimate had to be made of what part of the general property tax 
is derived from the taxation of residential housing. Preliminary to 
that, even the broad distinctions between personal and real property 
taxes, between the share of taxes paid by farmers, by home owners, 
and by business^ had to be made by way of approximation. For the 
purpose of the present study, the Department of Commerce, in coop- 
eration with the National Association of Real Estate Boards, conducted 
a questionnaire survey in early 1940 to obtain data on the above 
problems, covering a representative sample of urban and a few rural 
communities.® Barely half the Boards completing these schedules 
were able to answer the question — and mostly on the basis of expert 
estimates — as to the relative proportions of general property taxes 
that were paid on real property generally, on residences owned and 
rented, and on all business real property. However, certain local 
studies of a thorough nature made under the auspices of the W. P. A.,® 
and Census data combined in various ratios, all served as checks on 
the results derived from the sample study. 

The Commerce survey bore out the surmise founded on the spot 
studies j^hat the share of taxation on residential property to be imputed 
to the owner-tenant, to the owner who rents out his property (busi- 
ness), and to his tenant, depends largely on local circumstances and 
housing conditions. Data on farm homes and taxes are more com- 
plete, but their use is equally difficult because they show levies and 
not actual tax collections. Tax delinquencies in the "good" year 1939 
still amounted to about 10 percent of taxes levied on general property 
and always vary sharply as between communities and types of prop- 
erty. (In 1936 delinquencies were near 14 percent.) 

The total tax on residences was computed to be over $1,200,000,000, 
or 34.3 percent of all real-property taxes. This is a minimum figure 
and is lower than the median and average of the sample survey, which, 
however, contained a predominance of middle-sized and small cities, 
with only 2 big cities represented among the metropolises, at least 
as regards the completed schedules. In view of other calculations 
and spot studies, and the trend toward homestead exemption, the low 
value of farm homes, lowered assessments and, often, limitations on 
tax rates, it was felt inadvisable to select any higher percentage. 

(c) Business taxes. — Business taxes, as distinct from specific con- 
sumption taxes, include all taxes which are directly or indirectly 
imposed on business as such and not on specific consumers' goods. 
To this category belong corporate income, capital stock, and license 
taxes; the part of property taxes imposed on business, and all kinds 

« See the questionnaire and a preliminary summary of results in appendix i v . 

• See bibliography; nlso Work Projects Administration, Division of Social Research, Urban Housing. 



IQ CONCENTRATION OF ECONOMIC POWER 

of taxes imposed on specific cost factors, such as pay roll taxes ; and, 
finally, taxes on products used by business, as on minerals, on im- 
ported raw and semi-manufactured materials, and on trucks, railroads, 
and utilities. (See table IV.) 

(1) Corporate income taxes: Corporate income taxes are a problem- 
atical item for economic classification. There would be good 
reason for including them among the direct taxes on individuals. 
It might be assumed that coi;porate income taxes cannot usually be 
shifted and do result in a curtailment of corporate profits available 
for distribution to stockholders or for the accumulation of surplus. 
They thus lead to a reduction of dividends or else of the equity value. 
In both cases the impost falls on the stockholder. It may even be 
that under some circumstances corporate income taxes can be shifted. 
In industries subject to regulation (e. g., railroads) the regulating 
authority often considers income taxes in determining what a reason- 
able price is, or measiffes a reasonable rate of return by profits after 
taxes. In the final allocation (diagram III) of all taxes to individual 
consumer units, corporate income and profits taxes have in fact been 
treated as taxes on the stockholder. (See subsection 3 (a).) However, 
there are several good reasons for regarding these taxes, at least in 
the intermediate stage, as falling on business. In an attempt to show 
the total taxes paid by business, the corporate income taxes should 
not be excluded. 

If corporate income tax rates are graduated, the direct allocation to 
stockholders of various income brackets is very difficult, because the 
progressiveness of the corporate income tax does not necessarily 
correspond at all to the income brackets of dividend receivers; i. e., a 
large and profitable corporation paying the highest corporate income 
tax rate may have its dividends going to small holders of one or two 
shares, in whose salaries dividend income plays a minute role. A 
further consideration is that preferred stock dividends must be paid 
regardless of the profit earnings, and if they are paid in full, the pre- 
ferred stockholders in effect pay none of the corporate income and 
profits levies, which fall entirely upon the common stockholder. 

(2) Other business taxes : The inclusion of part of excise revenues 
has already been sufficiently discussed under taxes on specific con- 
sumption. Further, all State and local utility taxes, miscellaneous 
State selective sales taxes, local sales and excise taxes, local "other" 
taxes, and general sales taxes were in the first instance regarded as 
levies on business. 

Although general sales taxes are frequently shifted to the consumer, 
the process is not inevitable. Even if, in the eyes of the law, the 
buyer pays the tax on a purchase, he may actually be paying a- price 
lowered sufficiently to compensate him for his seeming tax burden, 
so tha4, the tax in reality is paid by the seller, or perhaps shifted even 
further back to his supplier. It is because of the uncertainty of their 
incidence that sales taxes are attributed to this group in this inter- 
mediate stage of allocation. 

All non-residential real property taxes are regarded as business 
taxes, since agriculture is included as one of the industrial sectors. 

2. Tax allocation. 

Tax allocation, being a step beyond simple classification, offers 
further problems for solution. Yet, as was said in part I, that is the 
very distribution required for adequate analysis of tax significance. 



CONCENTRATION OF ECONOMIC POWER 17 

(a) Personal taxes. — The allocation of taxes to the 10 consumer 
income groups involves problems of varying significance and difficulty. 
The National Resources Planning Board gave figures for personal 
taxes and gifts in 1935-36, which were used for deriving a figure for 
gifts in 1938-39; but in allocating taxes for the same year original 
data were used. Gifts were regarded in the National Resources 
Planning Board study, and also in the present one, as reducing the 
amount available for expenditures in the lower brackets, for savings 
in the higher brackets. (See pp. 9, 43.) 

For 1938-39, individual income taxes (both Federal and State) 
were allocated in line with the percentages paid by the corresponding 
income brackets shoAvn by the Treasury publication Statistics o/ 
Income, Although the tax returns are legally for individuals, while 
the National Resources Planning Board study uses consumer units, 
the margin of error, involved in this simple method of allocation is 
small, since in actual fact income tax returns in the lower brackets 
are mostly made out for families or individuals living alone. 

There are a considerable number of separate returns for husband 
and wife in the upper brackets, so that the application of tax payment 
percentages from Statistics of Income to family incomes may lead 
to a slight underestimation of tax progression.'^ On the other hand, 
the $5,000 intervals in consumer incomes over $5,000 provide con- 
siderable leeway for combined incomes to fall into a proper bracket, 
while rates up to the final limit of $20,000 are not steep enough to 
cause serious distortion in allocations. Moreover, no further adjust- 
ment is made for tax evasion than was made by the Board in develop- 
ing its basic figures, so that any exaggeration in progression shown is 
automatically compensated, at least in part.^ 

The slight revenue of poll taxes was distributed according to the 
number of consumer units in each income class over $500, on the as- 
sumption that such levies had the effect of disenfranchising the poorest, 
as has long been complained. Employees' old-age insurance contribu- 
tions were allotted to the six brackets below $5,000 on the basis of a 
Social Security Board analysis of such pajonents.^ Estate and gift 
taxes, on the other hand, were allocated to income classes above 
$5,000. This upper limit for social security taxes and low limit for 
estate and gift taxes, each one rather abrupt in itself, makes for a 
smooth transition when the two taxes are applied in combination. 
It should be noted that death taxes were added onto total income 
(because excluded from the National Resources Planning Board 
consumer incomes) to get the figure on consumer incomes which has 
been used in estimating personal tax percentages. 

The teclmical difficulties of allocating aggregate death taxes to 
income brackets are obviated by high rates of specific exemption and 
additional deductions allowed, which make it plausible to presuppose 
that very few receivers of income under $5,000 would pay this tax. 
The share of these few would be too small to consider. The same ar- 

' See appendix V. p. 44f. 

' Aggregate tax payments were distributed in proportion to the percentage paid by each income group 
as shown in Statistia of Income, so that if separate returns were made by husband and wife, the income of 
each would fall into a lower income bracket than justified by the joint income of the two as a "consumer 
unit." Therefore, a portion of individual income taxes paid is allocated to a lower bracket than is proper 
on the basis of family income. 

» Social Security Bulletin, April 1939, J. R. Arnold amd M. J. Wasserman, "Old-Age Insurance: Covered 
Workers and Average and Median Taxable Wages in 1937," p. 4. The article shows a small aggregate of 
payments for an open-end class over $3,000. See appendix V, p. 46, regarding adjustments. 



18 CONCENTRATION OF ECONOMIC POWER 

guments apply to gift taxes but with even greater force, since taxable 
gifts are made primarily by those large estates which are above the 
specific estate tax exemptions. 

Finally, personal property taxes were distributed; the part presumed 
to be collected on tangibles was allocated according to partial expendi- 
tures on durable consumer goods; the part levied on intangibles was 
assumed to be paid in line with the stockholdings of various income 
brackets. The proportions of such holdings were developed by inter- 
polation of figures derived by the Twentieth Century Fund as is 
described more fully in part C, section 1, of this chapter. These taxes 
are definitely paid directly by the property owner, unlike the estate 
and gift taxes which, for the purpose of the study, are only presumed 
to be thus paid, and probably come out of the additional capital or 
windfall gain furnished by the legacy. 

(b) Specific consumption taxes. — Specific consumption taxes are 
likewise shown as a percentage of income, although they Inust first 
be proportioned to expenditures in order to obtain iheir distribution. 

In allocating them to income brackets, the 1935-36 ratios of 
expenditures on each type of commodity were taken as the base to 
which the appropriate taxes were allotted. It is assumed that even 
over time the relative percentages spent on various products scarcely 
change; this assumption is fully confirmed by numerous studies show- 
ing the sluggishness of change in consumer habits over decades (and 
across oceans).^" Since the proportion of families and single individ- 
uals in each income group was assumed unchanged from 1935-36, 
data from the Consumer Expenditures (tables 22 A and 31 A) study 
were used, as they show aggriegate expenditures of all consumer units, 
in millions of dollars and in percentages. 

In allocating specific consum.ption taxes to incom.e brackets, the 
iro.plicit assumption was that taxable expenditures within a field 
rem.ain in constant ratio to total expenditures for that field of goods 
or services. Exceptions to this were specifically provided for in the 
allocation of personal property taxes and liquor taxes, and the possi- 
bility of sizable error due to this procedure is particularly stressed in 
the allocation of autoro.otive and residential property taxes and is 
indicated for certain other fields. The possibility *of error, however, 
did not lead the author to make adjustm.ents which would, perforce, 
have been based on even m.ore hj^pothetical assum.ptions than was the 
actual method. The accuracy of the results (and itm.ust be repeated 
that the study has been aim.ed at developing an approach and dis- 
covering gaps in ro.aterials, rather than at mathematical precision) 
then depends on whether the errors were additive or multiplicative, 
and if the former, to what extent they canceled each other. Where 
compensation was to be logically presupposed, it has been pointed 
out; so, likewise, have the more important sources of possible error 
been indicated. For example, the unsatisfactory breakdown of 
excises probably flattened the curve of regression. (See table III.) 

>oSee W. H. Lough, High-Level Consumption, 1935. 



CONCENTRATION OF ECONOMIC POWER 
Table III. — Specific Consumption Taxes, 1938-39 



19 



Income classes 



I. Under $500 

II. $500 to $1,000 

III. $1,000 to $1,500... 

IV. $1,500 to $2,000... 
V. $2,000 to $3,000... 

VI. $3,000 to $5,000... 

VII. $5,000 to $10,000.. 

VIII. $10,000 to $1.5,000 

IX. $15,000 to $20,000 

X. $20,000 and over. 

Total ....... 



Taxes on specifle consumption plus 
business taxes shifted to con- 
sumption 



Millions of 
dollars 



1,716 

1,999 

1.672 

1,939 

1,099 

554 

200 

124 

328 



10, 117 



As percent 
of expendi- 
ture 



16.9 
17.4 
18.0 
18.4 
18.7 
18.8 
18.9 
19.2 
19.2 
19.0 



18.2 



As percent 
of income i 



20.6 

17.1 

16.3 

16.4 

16.0 

14.4 

11.8 

9.9 

8.8 

5.7 



14.8 



' Corporate income taxes are excluded in this table. See table lb 'or consumption taxes related to 
income including corporate income taxes. 

Sources: Annual Report of the Secretary of the Treasury, 19.39; Bureau of the Census, State and Local 
Government Special Study A^o. 7, Revised, "State Tax Collections for the Fiscal Year 1938-1939"; and 
estimated. 

The unavailability of statistics on taxicabs and business use of cars 
has already been commented on, and their exclusion would probably 
reduce the proportion of taxes paid by the four upper brackets. 
Likewise, a breakdown between expenditures on new cars by various 
price classes, as against spending on gasoline, might well sharpen the 
regression considerably. 

All taxes on food (including custom.s, the Sugar Act levy, and 
Federal excises) were distributed on the basis of total foQd expenditures 
by each class. Nuisance and luxury taxes and manufacturers' excises 
were sim.ilarly prorated on the basis of expenditure on household 
necessities and recreation, although an exception was m.ade for certain 
luxuries, such as box seats, etc., which were imputed to the four top 
brackets only. Tobacco taxes, of course, were likewise allocated on 
the basis of tobacco consum.ption. 

The alm.ost com.plete absence of data regarding consum.ption of 
alcoholic beverages (they were included as •'beer, etc." under "food" 
in the National Resources Planning Board schedule) means that it 
is im.possible to allocate accurately over one-half billion dollars 
worth of Federal liquor taxes, and about $250,000,000 worth of State 
excises. In view of their fiscal unportance this problem cannot be 
disregarded . 

Distribution of such taxes in proportion to food expenditures would 
imply that, whatever one's family-size income, expenditure level, or 
geographic location, a fixed proportion of food outlays goes for alco- 
hohc beverages — an assumption wholly unjustified in a nation where 
wine and beer are not normally regarded as part of the meal. To 
assume that expenditures on liquor are made solely by people ''who 
can afford luxuries" is likewise contrary to fact. Consideration was 
therefore given to the facts that a large proportion of low incomes goes 
for food,^^ that a greater proportion of I'arger famiUes is found in the 

11 See Department of Labor, Study of Consumer Purchases, 1935-36 series; also Maurice Leven, Harold 
Q. Moulton, and Clark Warburton, Americe's Capacity to Consume, The Brookings Institution, Wash- 
ington, 1934; and W. H. Lough, High Level Consumption, 1935, pp. 236 ff. 



20 CONCENTRATION OP ECONOMIC POWER 

lower-income groups, with correspondingly larger outlays on foods ;'2 
that the National Resources Planning Board combined home and 
restaurant expenditures in the published tables, so that a distortion 
might exist in the direction of higher food expenditures by high income 
brackets than should be accounted for by food alone — i. e., that highly 
taxed liquor might be a greater proportion of food outlays in the higher 
incomes than in other income groups; that, however, higher-income 
groups do not necessarily increase their liquor consumption in the same 
proportion that other items of expenditures are raised with a larger 
income. In view, of all these circumstances, an arbitrary compromise 
was adopted, with half of liquor taxes being proportioned to food ex- 
penditures and half to recreation. 

Tentatively, taxes on residences, excluding farm land and business 
property, were classified by income groups on the basis of amounts 
spent for housing. Under the assumption that all residential property 
taxes are paid by the user, whether owner or tenant, the tax on housing 
rose proportionately as aggregate housing expenditures increased with 
a rise in total expenditures. 

This procedure may have led to an overestimate of the indirect taxes 
paid by the upper income brackets. However, the error was at least 
partly compensated by the trend during the last several years to grant 
total or partial property-tax exemptions to homes below a certain 
value. Such an allocation presupposes that residential property taxes 
are passed on to ultimate consumers, and completely ignores the possi- 
bilities of tax capitalization. A special appendix III describes this 
alternative. It may be mentioned here that regardless of the final 
disposition of the property-tax burden, it remains an annual outgo for 
the home owner, to be paid out of other income, and an increase in 
such taxes has the identical economic effect that any other tax pay- 
ment by the same consumer unit might have, until a new owner ac- 
quires the property at a lower price. 

The question of taxes passed on in rentals to low-income groups, 
and ignorance as to the aggregate proportions of homes owned and 
rented in such groups, further complicates the allocation. The results 
of the 1940 Housing Census will be invaluable in clarifying these 
matters, although some very interesting WPA studies have already 
shed light on this matter for particular localities. 

(c) Business taxes allocated to sectors of the economy. — The inter- 
mediate allocation of business taxes to the various branches of industry 
involved no question of income, but rather of industrial classification 
of taxes and, for general taxes on business, an investigation as to the 
relative shares paid by each. These general taxes include corporate 
income and profit levies, capital-stock taxes, employers' Social Secur- 
ity taxes. State corporate and general business taxes, and the business 
share of Federal excises and customs and of State and local general 
property, excise, and license taxes. 

Statistics of income for corporations were used as a basis for the 
allocations of income and profits taxes by industries. Since the 
capital-stock tax is in reality an alternative to the excess-profits tax, 
even if not so treated in the Statistics of Income, it was included with 
excess-profits taxes. ^^ The income produced (see table V) includes 

>' Consumer Incomes in the United States, pp. 20-22-97. It should be further noted here that a larger 

gortion of food expenditures in the lowest brackets are "imputed"£xpenditures on food grown and consumed 
y farm families. 

" Cf. Temporary National Economic Conmiittee, Clifford J. Hynning, Taxation of Corporate Enterprise, 
Washington, 1940. 



CONCENTRATION OF ECONOMIC POWER 21 

the net income (^ corporate and noncorporate enterprises, while 
only corporate income taxes are reported as business taxes. Income 
taxes paid on inc me derived from proprietorships and partnerships 
are directly allocated to the individual incomes of the businessmen. 
The latest data available, for 1937, were used in calculating the 
percentages of these taxes paid by the various sectors of industry, 
perforce assuming little change in the proportion paid by each sector, 
and the absolute amounts for 1938-39 were attributed to the appro- 
priate industrial sectors. State corporate income and profits taxes, 
which amounted to little ($113,000,000) were distributed in the 
samt oroportions. 

T( distribute nonresidential real-property taxes pro rata among 
the industriial sectors would be of doubtful propriety, since income 
produced by each sector may bear no relation to its property (viz, 
services). On the other hand, a straight allocation to wealth would 
likewise be unreasonable, since wealth estimates are very questionable, 
and not related to rates assessed by each locality on real wealth located 
m its jurisdiction. 

A compromise figure for 1935-36 was reached by means of a rather 
intricate process involving national wealth, income produced after 
the exclusion of agriculture, and other data available. The 1938-39 
distribution was made with the same ratios. The figure of $1,186,000,- 
000 in business real-property taxes fits in well with the Dun & Brad- 
street ^^ total of $581,000,000 for manufacturing and trade alone, of 
which part was personal property taxes. Tax exemptions on the 
property tax, however, vary widely and reduce the trustworthiness 
of classified tax incidence estimates. With many States, especially 
in the South and now in New England, too, ofi^ering tax reductions 
and exemptions as a special inducement to manuiacturers, any pro- 
portionate distribution of tax totals is thrown askew. Since each 
State and locality has its own practices in these respects, evaluation 
of totals becomes even more. a matter of hypothesis. 

The relatively small revenue of personal property taxes on business 
(totaling $273,000,000) was prorated among agriculture, mining, 
manufacture, utilities construction, and services, in accordance with 
their holdings of wealth less the amount assumed to be real property. ^^ 

The speciail business taxes which affect primarily a particular field 
of industry can be distributed rather simply. The following alloca- 
tions are self-explanatory: Federal, State, and local automotive 
levies on trucks, busses, and cabs were attributed to transportation 
(included with utilities). ^^ Four-fifths of the Federal excises on 
electrical energy, excises on communications, special property taxes, 
which are usually levied on utilities, State gross-receipts taxes on 
utilities, and the Federal employment tax on carriers were all classed 
as taxes on utilities and transportation. State severance taxes and 

" A Survey of Taxes Paid by Business in 1938, Dun & Bradstreet, New York, December 1939. 

'5 S<'e Nntional Resources Planning Board, Structure of the American Economy, pp. 374-77. 

•6 This distribution makes no allowance for such taxes paid by trucks owned by other types of enterprises 
and used within the business. Such a figure would lessen the ratio of taxes to value added by utilities. 
Although parfial figures, covering fewer than one-fourth of the 4,000,000 trucks registered, are available on 
the number of trucks in various lines of business, no attempt was made to allocate the business taxes on 
motor vehicles any further. Over a third of the nongovernment trucks for which the breakdown is made 
are express, haulage, or utility or railroad-owned trucks. Yet 60 percent of all truck trips are under 10 miles 
in lengtn, so that it may be assumed that the mileage of commercial trucks is far higher than is their pro- 
portionate number. (See Automobile Facts and Figures,, pp. 42 and 71.) This fact would considerably 
raise their share of gasoline taxes paid. Furthermore, interstate trucks pay much larger and more varied 
registration fees, so that their shaie in such levies would be much more than proportionate. Since the 
business-tax estimates are an intermediate stage not directly affecting final tax shifting, it was thought ad- 
visable not to attempt a very questionable distribution of gasoline and motor-vehicle taxes on the basis of 
these relationships. 



22 



CONCENTRATION OF ECONOMIC POWER 



the Federal bituminous coal fee naturally were allocated to mining. 
Federal stamp and various bank taxes and State documentary and 
insurance company levies were all attributed to finance. "Other" 
State licenses were attributed to services, since such imspecified 
licenses are frequently of a regulatory nature. 



Table IV.— Taxes Paid by Sectors of Industry 










Federal 


state and local 






e* 






B 






a 

03 


t» 


t>> 


Sector of industry 


8a 

.as 

it 

8,« 


g 

73 


o 


i 


8 



E 


s 

a 
S 


CO 

"i 
e 

2? 


» 05 

" n 


o 
a 

o 




















S 








o 




9 


« 


o 




a> 










Q 


m 


o 


w 


Q 


O 


O 


K> 


■ m' 


o 


Agriculture 


X 


X 


X 














X 


Mininn . .. . . _ 


X 
X 
X 
X 
X 
X 
X 
X 


X 
X 
X 
X 
X 
X 
X 
X 




X 
X 
X 


X 
X 
X 
X 
X 
X 
X 






X 




X 


Manufacture 


X ' 


X 


X 


Utilities*-- 


X 


X 


X 


Construction 


X 


Trade 


X 


"x 

X 




X 


X 
X 
X 




X 


Services 


X 


Finance 






X 


Miscellaneous 






X 





















' Includes capital stock taxes. 
> Includes State payments. 



' Includes both real -ind personal property. 
* Including transpoi tation. 



Table V. — Business Taxes as Percent of Income Produced, 1938-39 

[All money figures in millions of dollars] 





Income 
produced 


All business taxes ' » 


Taxes as percent of income 


Sectors of industry 


Federal 


State 
and local 


Federal 


State 
and local 


Total 


Agriculture . . . . . 


6,623 
1,120 
13,660 
6,810 
1,852 
9, 268 
8, 534 
7,685 
1,320 


104 

109 

1,165 

732 

55 

474 

131 

237 

8 


1,226 

128 

669 

1,012 

15 

759 

307 

270 

35 


1.8 
9.7 
8.5 
10.8 
3.0 
5.1 
1.5 
3.1 
.6 


21.8 

U.4 

4.9 

14.9 

.8 

8.2 

3.6 

3.5 

2.7 


23.6 


Mining 


21 1 


Manufacture -. 


13.4 


Utilities' 


2,5.7 


Construction 


3.8 


Trade... 

Services 


13.3 

5.1 


Finance 


6.6 


Miscellaneous 


3.3 






Total 


55,872 


3,015 


M,421 


5.4 


7.9 


13.3 







I Part of the Federal excise on toiletries was also included here to cover their commercial purchase for 
sale in the shape of service. 

> See table IV for the precise taxes included as business taxes. 

' Including tran.sportation. 

* After the completion of this study, a report by the Tax Policy League came out on Tax Yields 19S9. 
This publication shows taxes at $13,794,718,327, lower by almost $200,000,000 than the estimate of the present 
study. The discrepancy is due entirely to the difference in methods of estimating general property tax 
receipts. 

Sources: See appendix table E for tax sources in detail. Income produced i.s from the Survey of Current 
Buaineis, June 1939, and the Department of Commerce release of Jan. 28, 1940, on income payments to 
Individuals. 



Customs receipts on products other than food and tobacc were 
divided between agriculture and trade, with 30 percent of total 
customs attributed to agriculture. This percentage was selected 
because of the implication that agriculture is discriminated against 
to this amount, found in the legislative provision allowing at least 



CONCENTRATION OF ECONOMIC POWER 23 

30 percent of customs revenues to be used by the Secretary of Agri- 
culture for compensating the farmer. To trade were further attrib- 
uted "other" State selective sales taxes, chain-store licenses, local 
sales and excise taxes, and the larger portion of State general sales 
taxes. A share of these was treated as paid by manufacturers; the 
proportion was obtained from Dun & Bradstreet's A Survey of Taxes 
Paid by Business in 1938 {op. cit.) which indicated a ratio of 1 to 1.2 
for such taxes paid by manufacturers and by trade (retail and whole- 
sale). Although this ratio includes both sales and excise taxes, which 
are separately allocated in the present study, the ratio was applied, 
because it provided a correction for not allotting any of the State 
and local licenses to manufacturing. 

3. Tax shifting. 

The pattern induced by the shifting of taxes depends on various 
nonfiscal considerations. The directions in which various business 
taxes are shifted probably fluctuate constantly, according to business 
conditions, competition, demand, general income levels, and even 
fashion. Although most manufacturing industries and much agri- 
culture compete on a national market, their State distribution is 
likewise important in analyzing the fashion in which taxes are 
shifted. First, some industries, either by their nature (services; 
residential construction, some perishable foodstuffs), or through 
pressure and legislative restraint (milk in agriculture, the independent 
grocer in States with an anti-chain-store bias) have a limited local or 
regional field of action, with little outside competition, and it is the 
local residents who are primarily affected by tax shifting. Secondly, 
there are taxes, such as those on transactions, severance of natural 
resources, etc., which are imposed locally on all business of the com- 
munity or region, and their incidence and transfer may not affect 
customers or employees elsewhere of companies active in such a 
region. But the figures by States and by industrial branches are not 
always congruent, and estimates must be resorted 'to to make them 
agree fully. In view of these limitations, it is obvious that no deter- 
minations of tax incidence can be absolute for any length of time; so 
that it was felt more useful to show hypothetical patterns that might 
be expected to appear under the extreme assumption of all taxes 
shifting to the consumer. This assumption was varied once by assum- 
ing a shift of all business cost taxes backward to wages. (See sec.c 
following.) 

(a) Corporate income and profits taxes. — Corporate income taxes 
(as well as excess-profits and capital-stock taxes) were allotted to 
income brackets in direct ratio to the stockholding of the respective 
brackets.'^ 

Such a procedure implies that securities held in all brackets do not 
differ with respect to the profitability of the respective corporations, 
and therefore yield a tax proportionate to security holdings. While 
there is no data showing direct correlation among the income brackets 
of security holders, on the one hand, and the profitability of the 
respective corporations and the taxes imposed, on the other, yet the 
assumption seems to be valid when one deals with aggregates, par- 

" Derived from figures cited by Colm and Lehmann in "Economic Consequences of Recent American Tax 
Policy," Social Research Supplement I, N. Y., 1938,p.46. Their estimates in turn were made on the basis 
of income-tax statistics and estimates by the Twentieth Century Fund, The Security Markets,, N. Y., 1935. 



24 CONCENTRATION OF ECONOMIC POWER 

ticularly in view of the broad intervals (of $5,000) into which these 
income classes are grouped.^* 

The assumption that business taxes in general are shifted to con- 
sumers in the form of prices does not apply to corporate income and 
profits taxes, which, in view of their dependence on earnings, are not 
ordinarily regarded as cost factors. Nevertheless, the possibility of 
their being passed on in exceptional cases must be kept in mind. To 
the extent that they are passed on, the progressiveness of the tax 
structure in the upper part of the income pyramid is materially reduced. 

In the final allocation of business taxes the corporate income taxes 
are added to the personal incomes of the respective brackets as well 
as to the taxes borne by these income brackets. All undistributed 
corporate profits should be added to and corporate losses deducted 
from personal incomes as the basis for a final allocation. In view of 
the fact that undistributed profits and losses were almost equal, both 
were ignored; the corporate taxes were added to the income basis 
because otherwise these taxes would have been related to an income 
figure which included corporate profits only after deduction of taxes. 

(6) Taxes shifted to consumption. — Other business taxes, however, 
were considered as definitely shifted to the ultimate purchaser and 
were allocated to consumer income in the same fashion as were the 
"levies on specific consumption." That is to say, they were made 
proportionate either to total expenditures or to spending on specific 
goods or services. Most of them, however, due to the very fact that 
they could not be directly allocated to any branch of consumer spend- 
ing in the first place, were distributed in ratio to total consumers' 
expenditures. For example, even the taxes known to be imposed on 
railroads could not be allotted in proportion to travel, since 90 percent 
of American railroad business is freight, and there is no information 
as to the proportions of such taxes entering the cost of manufactured 
food, of farm products for export and for home consumption, of heavy 
raw materials for use in producers' machinery, in durable consumption 
goods, in fuel, for passenger travel. Similarly, the exact incidence 
of the many other taxes falling on industry at various stages of pro- 
duction can scarcely be traced. 

The sole exception to this was the distribution of general sales taxes. 
Just as the Dun & Bradstreet survey (op. cit.) allowed a rough alloca- 
tion between manufacturing and trade, so, many studies, both local 
and general, give a comprehensive and fairly consistent picture of the 
sales-tax yield on various products identifiable with fields of expendi- 
ture shown in the National Resources Planning Board study. 

The average proportions used are shown in table VI, derived from 
a series of local, State, and Work Projects Administration analyses 
of sales-tax receipts. These figures are corroborated by estimates in 
several States and agree in general with Jacoby's estimates.^® 

'8 The Department of Commerce initiated a sample study of the composition of estates, the results of 
which wUl shed some light on the relationship between the size of estates, on the one hand, and the type 
of corporations in which securities are held, on the other hand. 

i» Neil Jacoby, Retail Sales Taxation, Chicago, 1937. 



CONCENTRATION OF ECONOMIC POWER 25 

Table VI. — Estimates of State Sales Taxes Paid by Commodity Groups 



Conunodity group 



Average 
percent- 
ages 1 



Taxes paid (in milliens of 
dollars)— 



1937 J 



1938 3 



Foods. 

Apparel 

Building 

Furniture 

Automobiles 

General merchandise 

Amusements _ . 

Miscellaneous 

Utilities 5 

Total 



23.5 
6.0 
4.2 
3.8 

11.9 

16.5 
2.3 

23.8 
8.0 



$101. 3 
25.9 
18.1 
16.4 
51.3 
71.1 
9.9 
102.6 
34.5 



$105. 
27. 
18. 
17. 
53. 
74. 
10. 
107. 
36. 



$103. 9 
26.5 
18.6 
16.8 
52.6 
72.9 
10.2 
105.3 
35.4 



100.0 



431.0 



450.7 



442.3 



' Based on: State of Utah (with assistance of the WPA) Report of the Subcommittees on the Homestead 
Exemptions Proposal and Taxation, October 1936, p. 117, "Sales Tax Collection by Groups." The per- 
centages of the 4 States given (California, Kentucky, Ohio, and Utah) were averaged after adjustment for 
the exclusion of utilities and amusements from the California and Ohio tax. 

2 U. S. Bureau of the Census, Finamtal Statistics of States, 1937. 

3 Tax Policy, January 1940. 

* U. S. Bureau of the Census, State and Local Government Special Study No. 7, Revised, "State Tax 
Collections. Fiscal Year 1938-39." 

s The sales tax on utUities was excluded from the final tax allocations. The percentage is probably too 
high, as some States do not tax utilities; on the other hand, "miscellaneous" would include sales taxes in some 
States on other commodities which do not belong in a retail sales analysis, so that the deduction of 8 percent 
is probably a fair one. 

But sales taxes need not always be passed on to the consumer; they 
may be, in reality, absorbed by the seller (or passed backward to the 
producer) who, if compelled by law to charge the consumer for the 
tax, may simultaneously lower his price. Such is the conclusion 
reached in a special study made in the Department of Commerce, ^^ 
which indicates that prices are only slightly affected even by the 
general sales tax when it is legally transferred to the consumer. Of 
course, it is always possible that without the sales tax prices would 
have declined. 

Jacoby's statement that retail sales taxes at large reach less than 
60 percent of consumer expenditures agrees with the relationship 
found between consumer expenditures and sales-tax collections 
(making allowance for the increase in use taxes, lessened evasion, and 
in some cases stiffening of rates) and with the various calculations of 
net sales-tax burden hovering at about IK percent. The present 
calculations also apparently support conclusions as to the extreme 
regressiveness of such taxes when they are passed on to the consumer. 

(c) Taxes shifted to wages. — To take the opposite extreme, if it were 
assumed that all business taxes except those on corporate income and 
profits and on agricultural real property were shifted backward to 
wages, the general pattern of taxes as a percentage of income would be 
conserved but with an even more regressive bent. The lowest tax 
percentage would be about 12 percent of the consumer income of the 
$5,000 to $10,000 bracket. The steep regression would begin with 
taxes taking 25 percent of incomes under $500, assuming that the 
taxes were shifted in proportion to the Social Security Board per- 
centages of taxable wages in each income bracket. ^^ Since these data 

so Warhurst, H. P., The Effect of the General Sales Tax Levies on Retail Sales 19SS-S5, Bureau of Foreign 
and Domestic Commerce, Marketing Research Division, April 1937. 
" See appendix V, p. 46, with regard to the distortion involved in this method. 

264815 — 40— No. 3 4 



26 



CONCENTRATION OF ECONOMIC POWER 



go to an open-end class of incomes over $3,000, the regression is dis- 
torted by a too high percentage for the $3,000 to $5,000 class and too 
low a percentage (dr the next bracket. But the general tendency is 
undoubtedly correctly indicated in percentage table VII. 

Table VII. — Total Taxes as Percent of Consumer Income, If Business Taxes Were 
Shifted to Wages, 193&-39 



Income classes 


Taxes as 
percent of 
income ' 


Income classes- 


Taxes as 

percent of 

income 


I Under $500 


25 
17 
17 
17 
15 
16 


VII. $5,000 to $10,000 


12 


II. $500 to $1,000 


VIIT. $10,000 to $15,000 


23 


Ill $1,000 to $1,500 


,IX. $15,000 to $20,000 


31 


IV $1,500 to $2,000 , 


X. $20,000 and over 


38 


V $2 000 to $3 000 


Average . 




VI $3,000 to $5,000 


19 









1 Excluding real-property taxes on farms. 

Sources: For income, see table I. All taxes and their sources are shown in appendix table E. 

It should be noted that taxes on agricultural real property are not 
included in the computation, since they can scarcel}'^ be passed back 
to wages in any important amount. In computing table VII, it has 
been assumed that such taxes would be capitalized. 

This rough allocation of taxes to wages (5ould be carried out with 
far greater detail by the use of Department of Labor statistics, 
published in the Monthly Labor Review of various months, on 
annual earnings in various industries. However, since only a few 
sample industries were covered by these studies, extrapolation or 
guesswork would be necessary in any case for extending these results 
to total consumer income. 

A complete shifting backward of all business cost taxes can hardly 
be realistically assumed. The backward shifting consists rather of a 
constant pressure, like that of any other increase in business costs, 
against increases in wages which would be possible in the absence of 
these taxes; i. e., the effect lags, and a direct shift to wages is rare. 
The tax patterns of table VII, therefore, probably show an exag- 
gerated regression. 

The close correspondence of income and expenditures in the lower 
brackets leads, perforce, to a certain intercorrelation of results. But 
the use of the assumption of taxes shifted to wages is intended not as 
an arithmetical check but rather for a test of the basic assumptions, 
and, as such, its results validly support the approach used in the 
study. 



Part III.— RESULTS 
A. FINAL TAX PATTERNS 

The tax pattern, despite apparesntly sharp changes of rate, shifts 
slowly, especially as regards the tax structure within national-income 
patterns. Generally speaking, the total American tax pattern is 
regressive at its lower end, nearly proportionate through all the middle 
brackets from $1,000 to $5,000, and increasingly progressive from 
there on. 

1. Personal and consumer taxes. — Personal taxes have become some- 
what more important in the total American tax system, furnishing 
28.2 percent of. all Federal, State, and local tax revenues in fiscal year 
1939, as against 23.2 percent in the fiscal year 1936. As a percentage 
of coilsumer income, they rose from 4.3 percent in 1935-36 to 5.7 
percent in 1938-39/ but consumption taxes, as defined in this study, 
did not decline; in' fact, their percentage also rose, although very 
slightl}'' from 14.2 to 14.5 percent of consumer income. The in- 
creasing, reliance on personal taxes must be imputed primarily to the 
more important share oi the Federal Government in taxation and to 
the higher revenues, especially from income taxes, resulting from 
business improvement. Federal receipts more than tripled in volume 
between 1932 and 1938. It is of interest to .note that even in 1935, 
two-fifths of all British taxes (including local rates) were direct, 
despite import and heavy excise duties, although the total tax yield 
(excluding social insurance payments) was well over such American 
receipts of 18.5 percent of consumer income. 



Table VHI.— 


Taxes as Percentage of National Income in United States and Great 
Britain ' 




> 


Great Britain— 1935 


United States 


Typ« of taxes 


Taxes 


Taxes as 

percent of 

income 


1935-36 


1938-39 




Taxes 


Taxes as 

percent of 

income 


Taxes 


Taxes as 

percent of 

income 


Personal 

Consumer 




t?.81,500.000- 
539, 200, 000 


9.2 
13.6 


$2, 555, 000, 000 
8, 475, 000, 000 


4.3 
14.2 


$3,972,000,000 
10,118,000,000 


5.7 
14.5 



• See International Labor Review, XXXIV: 5, November 1936, Ursula Hicks— "Some Effects of Financial 
Policy on the Distribution of Income in Great Britain Since the War," pp. 594-617, especially p. 599. 

• Excluding social vecurity payments for Great Britain. 

For United States, see table I and text. 

2. Taxes and income. — The large share of taxes paid in 1938-39 by 
the lower-income groups is strikingly depicted in table IX. A com- 
parison of column 1, which shows the income of each income bracket 

' Put in another relationship, Federal income, estate and gift, excess profits, and capital-stock taxes rose 
from 2.9 percent of national income in fiscal year 1928 to 4.6 percent in fiscal year 1938, after falling to 1.9 
percent in fiscal year 1933. 

27 



28 



CONCENTRATION OF ECONOMIC POWER 



as a percentage of aggregate income, and column 2, whicli shows the 
taxes paid by each income bracket as a percentage of the total taxes 
paid, with both series cumulatively arranged, shows this clearly. It 
will be noted that except for the lowest ^ and the two highest brackets, 
which appear to have paid more than their straight proportionate 
share of taxes, all other income brackets paid an amount propor- 
tionate to income. Diagram III illustrates these ratios. 



Table IX.- 



-Income and Tax Percentages for 1938-39, ^ Cuttiulatecl From, Lo,west 
Brackets 



Income classes 



Percent of in- 
come found 
in each 
income 
bracket ' 



Percent of all 

taxes paid by 

each income 

brack«t ' 



I. Under $500 

■II. $500 to $1,000...- 

III. $1,000 to $1,500.. 

IV. $1,500 to $2,000-. 
V. $2,000 to $3,000.. 

VI. $3,000 to $5,000.. 

VII. $5,000 to $10,000. 

VIII. $10,000 to $15,000 

IX. $15,000 to $20,000 

X. $20,000 and over. 




3.7 

16.6 
31.5 
44.4 
59.5 
69.2 
75.4 
79.4 
83.0 
100.0 



• See diagram II. 

' See appendix table C for sources of data and for basic income figiires used, which were taken from column 
6; i. e., income including various so-called death taxes, but not including corporate income and profits 
taxes shifted to consumers. This table therefore does not coincide with income percentages in diagram III. 

3 All taxes paid are shown with sources in appendix table E, and their distribution by income brackets 
appears in appendix tables F to J. 

A comparison of the 1938-39 tax pattern with that of 1932 is of 
interest. In 1932 total tax payments were a more nearly constant 
percentage of income as between income brackets; i. e., a curve drawn 
across a graph similar to diagram III would have been flatter, espe- 
cially at the extremes, denoting a more proportional system. At that 
time, the lowest income bracket paid neither liquor nor social-security 
taxes, whUe income-tax rates in the higher brackets were less steeply 
graduated. 

Table X. — Taxes and Income, 1932 ^ 





Consumer 

income ' 

(in millions 

of dollars) 


Percentage of income 


Income classes 


All personal 
taxes' 


Total taxes,' includ- 
ing shifted taxes 




On in- 
come 


On con- 
sumer 


On in- 
come 


On con- 
sumer 


Up to $750 .- ... 


$5, 676 
12,885 
7,225 
8,610 
5,409 
4,758 
4,930 


0.2 

.5 

.7 

1.3 

2.3 

4.2 

16.0 


6.2 
6.6 
6.6 
6.4 
6.0 
4.9 
3.1 


0.2 

.5 

.9 

1.6 

4.1 

6.8 

22.8 


15 4 


$750 to $1,500 -- - 


14.1 


$1,500 to $2,000 


13.6 


$2,000 to $3,000.... 


11.0 


$3,000 to $5,000 


11.9 


$5,000 to $15,000 


9.5 


Over $15,000 


5.9 






Total 


49, 493 


2.7 


5.9 


4.0 


12.5 







> Compare with table lb. 

' Computed on the basis of National Resources Plarming Board — Consumer Incomes in the United States 
1935-36, and Department of Commerce— /ncome in the United States, 1929-37, November 1938. 

3 From Bureau of Internal Revenue — Internal Revenue Collections from Specific Sources, 1918-38; U. S. 
Bureau of the Census— Statistical Abstract of the United States, 1938; U. S. Bureau of the Census— i^flancfol 
Statistics of State and Local Oovernments, 1932. 

* This regression is due to the p jsence of consumption in excess of income. 



CONCENTRATION OF ECONOMIC POWER 



29 



Despite these changes, the basic relationship between taxes and 
consumer income at different income levels appears to alter slowly, 
partly because the relatively minor role of most individual taxes means 
that any rate or administrative changes affect aggregate percentages 
less than might be expected. 

Table XI. — Taxes In Relation To Consumer Income, 1938-39, If 1932 Tax Rates 
Had Been In Force In 1938-39 



Income classes 



All taxes at 

1932 rates as 

percent of 

income in 

1938-39 1 



1938-39 taxes 

as percent of 

consumer 

income ' 



I. Under $500 

n. $500 to $1,000... . 

in. $1,000 to $1,500.. 

IV. $1,500 to $2,000.. 

v. $2,000 to $3,000.. 

VI. $3,000 to $5y000- . 

Vn. $5,000 to $10,000. 

Vni. $10,000 to $15,000 

IX. $15,000 to $20,000 

X. $20,000 and over. 

Total 




22.0 
17.9 
17.2 
17.6 
17.0 
16.3 
15.0 
17.8 
22.0 
31.3 



18.5 



1 Corporate income taxes are excluded. 

Sources: Computations based on figures from sources cited in table X. 
Internal Revenue Code of the United States. 



See also tables I and XI, and 



An attempt was also made to apply tax rates of 1933 to 1938-39 
income /igures to discover what the curve would have been had the 
tax system r-emained unchajiged. It was assumed, for the purpose 
of comparison, that expenditure patterns would not have varied. 
The results are summarized in table X, which shows that the curve 
would have been approximately horizontal between the $1,500 to 
$2,000 and $10,000 to $15,000 brackets. There would have been a 
slight progression as between the three lowest brackets (below $1,500), 
as wed as between the two top brackets (i. e., $15,000 and over). 
Under actual 1938-39 tax rates the progression began at $10,000 and 
rose more steeply. At the same time, actual tax rates were regres- 
sive among the lowest-income brackets. In this comparison, cor- 
porate income taxes were excluded. 

The chief causes of the difference in the tax system of 1932 as 
apphed in 1938-39 and the actual system of 1938-39 were the far 
greater role of State and local taxes, which are predominantly regres- 
sive, the absence of taxes on alcoholic beverages and of social-security 
taxes, and the very low level of manufacturers' excises. The imposi- 
tion of these new taxes accounts for the difference in the average 
percentage of income taken by taxes (18.5 percent actually against 
11,6 percent with the hypothetical application of 1932 rates). 

In order to put this study on a basis comparable with various 
analyses of tax burdens of "typical" famiUes, especially the one by 
Newcomer,^ taxes were computed on an average consumer unit basis 
by changing aggregates to averages for each of 10 consumer income 
classes. It should be noted here that the rise in taxes on expenditures 
per consumer unit is not as rapid as the rise in saving as is shown in 
table XII. 



' studies in Current Tax ProbUmt, Twentieth Century Fund, 1937. 



30 CONCENTRATION OF ECONOMIC POWER 

Table XII. — Taxes and Sayings as Percent of Consumer Income Per Consumer 

Unit, 1938-39 i 



Income class 


Income per 
unit 


Expendi- 
ture per 
unit 


Expenditure 
as percent 
of income 


Savings as 

percent of 

income 


All taxes, 
except on 
corporate 
income as 
percent of 
income 


I 


$346 

847 

1,381 

1,929 

2,689 

4,121 

7,749 

13,000 

20, 333 

51, 259 


$420 
833 
1,249 
1,722 
2,294 
3,152 
4,842 
6,679 
9(348 
15, 446 


121.4 
98.3 
90.4 
89.3 
85.3 
76.5 
62.5 
51.4 
46.0 
30.1 




22 


II 




17.9 


III 


5.2 
5.8 
9.6 
2 17.0 
29.4 
35.8 
36.9 
42.2 


17 2 


IV 


17 6 


V 


17.0 


VI 


16.3 


VII 


15.1 


VIII 


17.8 


IX : 


22.0 


X .- 


31.3 






Total - 


1,705 


1,382 


81.1 


3 11.7 


11 7 







' See also table I, and appendix table D. Tables I and XII do not quite correspond to each other, as cor- 
porate income and profits taxes have been omitted from income and tax percentages in the present table. 
2 Savings begin to approximate taxes only at the $3,000 to .$5,000 income level. 
' Note that average per consumer unit savings and taxes are identical. 

Sources: Computed from data in National Resources Planning Board Consumer Expenditures in tht 
United States; for tax data see table E. 

More significant, perhaps, is the fact that savings begin to exceed 
taxes slightly only when the income of the average consumer unit is 
between $3,000 and $5,000. For the whole $3,000 to $5,000 group 
the excess of saving over taxes is slight (17 percent of income to 16.3 
percent of income taken by taxes), so that it is probable that the 
excess first appears nearer the $5,000 than the $3,000 level. A fur- 
ther point of interest is that the average consumer unit in the lowest 
income bracket (under $500) pays out in taxes exactly the same pro- 
portion of its income as does a unit in the next to the highest bracket 
with an income between $15,000 and $20,000.* 

The ratio of taxes paid to income in the different income brackets 
does not, of course, describe the tax system completely. For one 
thing, the same percentage of income may be taken by taxes at two 
different periods and yet the nature of the tax structure and therefore 
its economic effects may have altered considerably. For example, 
direct individual taxes may account for a larger share of the taxes 
paid by a particular income bracket without changing its total pay- 
ments. Again, the percentage of taxes paid may remain constant 
over time; yet the absolute amount will have risen or declined if in- 
comes received by the bracket changed proportionately. This latter 
effect might take place irrespective of whether tax rates changed. 

3. Taxes as an influence on consumption. — Two types of tax influence 
on consumption must be distinguished: First, there is the effect of 
"nonshiftable" income taxes, personal and business taxes in curtailing 
disposable income and in changing the structure or amount of spend- 
ing. Second, there is the invisible influence of excise and business 
taxes on consumption. 

The effects of taxation on individual spending are extremely 
difficult to trace. Judging by surveys, a quarter of the people who 
pay no direct taxes are unaware that they pay any taxes at all. Cer- 

* It should be stated that the preceding table excludes corporate income and profits taxes from both in- 
come and taxes, thus sharpening regressiveness. (Taxes would be 37.8 percept instead of 31.3 percent of 
income in the top bracket if corporate income taxes were included in proportion to dividends.) 



CX)NCENTRATION OF ECONOMIC POWER 31 

tainly these persons are not consciously influenced in their consumption, 
pattern by tax considerations, but the influence nevertheless exists. 
If taxes raise prices noticeably, buyers tend to direct their purchases 
toward other articles, though they may be unaware of the cause of 
the price rise. If they purchase the same quantity as before, they 
will have to curtail expenditures elsewhere, unless they are in one of 
the upper-income groups where the alternative of curtailing saving 
exists. The very conscious attempts by smokers to evade the New 
York City cigarette tax of 1939, which led to a shift in the type of 
tobacco purchased, is an illustration in point. Taxation probably 
affects purchasing to a lesser extent if it is passed on to the consumer 
in a form other than increased nominal prices; a slight decrease iii 
size or quality due to heavier taxes without change in nominal price 
is a shift of the tax burden to the consumer but one which may not 
cause him to modify his spending pattern. 

If taxes are passed backward, reducing wages and salaries, however, 
the effect on expenditures (of the lower income brackets in particular) 
becomes more clearly ascertainable because more information is 
available concerning the effects of changes in income than of changes in 
prices. 

Personal taxes, of course, affect primarily the sum total of expendi- 
tures and savings for each income class. Because of the high exemp- 
tions allowed, their direct effect is confined to the expenditures of the 
four top income, brackets. The reduction in the absolute volume 
of expenditures will affect total consumption and the revenue from 
consumption taxes, and this, in turn, may lead to further shifts in 
spending and saving. As income decreases, expenditures, even on 
necessities, include less taxable fields — older and less expensive auto- 
mobiles, less valuable furniture subject to lower personal-property 
taxes, and housing in districts with lower assessments. This may 
be counterbalanced by the decreased use of untaxed services. 

An illustration of this tendency appears in the slight "hump" of 
taxes as a percentage of expenditures for the income bracket between 
$1,500 and $2,000 (table III). It is caused by the increase in the 
proportion of expenditures njade on automobiles and recreation, both 
subject to more taxes than other spending, and also by the rise in 
income and personal-property taxes paid. 

If monetary income is sharply and regularly reduced by taxation, 
the pattern as well as the total amount of expenditures may differ 
from what it would be at a higher level of disposable income. 

The effects of business taxes depend on the type of tax, the type of 
customers and employees, and the nature of their income and expendi- 
tures patterns, and the individual products or income groups to which 
the taxes are passed. Thus, if a dominant local chain store succeeds 
in passing on a new levy in the form of a general price increase the 
effect is mostly regressive, affecting the cash customers unable to drive 
to another district or to order by telephone; but such a tax might 
also be concentrated on certain semiluxury goods little bought by the 
low-income clients. Similarly, a tax on garages and parking lots 
would be, if traiisf erred to the consumer, mildly progressive, since it 
would affect automobile owners accustomed to paying for special 
parking facilities. 

In the case of a reduction of wages through the passing on of 
business taxes to employees, the effect on expenditures would depend 



32 CONCENTRATION OF ECONOMIC POWER 

on the amount of the decrease or the prevented increase and the 
income-expenditures patterns of the employees affected. A very 
shght decrease with no corresponding fall in prices might, if regarded 
as temporary in nature, merely reduce savings or increase dissavings; 
a sharper wage decrease might force a rearrangement of expenditures, 
with corresponding changes in consumption goods bought and, there- 
fore, in the regressiveness. For example, an increased proportion of 
expenditures on food at the expense of recreation in the two lowest 
brackets might actually reduce regression, since food taxes are fewer 
and lighter than are amusement and other nuisance levies. 

Unfortunately, very little material is available on the quality of 
goods purchased by different income classes. And the quantity of 
figures based on the family expenditure studies give no inkling of 
State and local differentiations in purchases and, therefore, in invis- 
ible taxes collected from each class. 

B. INFLUENCE OF TAXES ON BUSINESS 

Although this study is concerned with the incidence of taxes on 
consumer incomes, it was necessary to make an intermediate allocation 
of certain taxes to various sectors of industry, as has been" previously 
discussed in part I, section B2, and part II, section C, 1(c) and 2(c). 
The tentative allocations thus made are presented in table V. The 
intermediate character of this table precludes any finality for its 
results. For thorough studies of business taxes and their incidence 
on business, reference is made to the following two studies, which were 
also used as a basis for th6se allocations: 

Temporary National Economic Committee, Clifford Hynning, Taxation of 
Corporate Enterprise, Washington, 1940. 

Dun & Bradstreet, A Survey of Taxes Paid hy Business in 19S8, New York, 
December 1939. ^ 

C. INFLUENCE OF TAXES ON SAVINGS 

The form of savings, both personal and business, and the influence 
of taxation on them is a question requiring careful analysis. It should 
be noted here that business savings are not depicted anywhere in the 
graphs, nor, indeed, considered in the whole study. 

ThQ contrast between the distribution of consumer income and that 
of consumer saving has already been stressed. The amount of saving 
within a given income bracket apparently can be considered as 
constant over time as other aspects of expenditure. The Study of 
Consumer Purchases series indicates that the form of consumer 
dissaving and saving varies with the income level and at the same 
time indicates a certain constancy within the lower-income classes. 

Table I and diagram. III indicate the quantitative relationship 
between incomes on the one hand and taxes and savings on the other. 
It can be assumed that changes in taxation in the lower and middle- 
income brackets do not affect savings substantially. A certain 
increase in prices may result in increased dissavings, yet this wiU be 
only a temporary result. Saving constitutes such a relatively small 
proportion of income in the lower brackets that a change in taxes wiU 
affect spending rather than saving. Saving in the lower and middle 
brackets is also a relatively rigid item in the budget of consmners in 
these brackets. In some cases, savings belong practically to what 
should be called "life necessities." 



CONCENTRATION OF ECONOMIC POWER 



33 



Table XIII. 



-Consumer Savings by Income Brackets, Related to Consumer 
Incomes in 1935-36 and 1938-39 



Income classes ' 



Consumer savings 
Cin millions of 
dollars) 



1935-36 1938-39 



Percent of total 
savings made by 
each group 



1935-36 1938-39 



Savings as percent 
of income in each 
group 2 



1535-36 



I. Under $500 

11. $500 to $1,000 

III. $1,000 to $1,500. .. 

IV. |1, 500 to $2,000 — 
V. $2,000 to $3,000... 

VI. $3,000 to $5,000... 

VII. $5,000 to $10, 000- 

VIII. $10, 000 to $15, 000. 

IX. $15,000 to $20, 000 

X. $20, 000 and over.. 

Total.... 



-800 

-636 

-2 

441 

1,069 

1,176 

1,218 

679 

473 

2,360 



-575 

-197 

642 

593 

1,167 

1,299 

1,380 

726 

517 

2,425 



-13.4 

-10.6 



7.4 

17.9 

19.6 

20.4 

11.3 

7.9 

39.5 



-7.2 

-2.5 

8.1 

7.4 

14.6 

16.3 

17.3 

9.1 

6.5 

30.4 



-38.8 
-7.3 

5.0 
10.1 
17.7 
29.7 
38.7 
39.3 
47.7 



-24.3 

-2.0 

5.2 

5.8 

9.6 

17.0 

29.4 

35.8 

36.8 

42.2 



5,978 



7,977 



100.0 



100.0 



10.0 



11.7 



> Same as in table I. 

' Excluding corporate income tax: Comparable to each other but not to Diagram III B. 

Sources: See table I. 

In the upper brackets, the situation is quite different. Here, saving^ 
is miore a residual use of income, remaining after taxes have been paid 
and current expenditures made. Here a change in taxes presumably 
affects saving more than current spending. 

For the economic effects of taxes, it would be very important to 
distinguish the extent to which taxes imposed on the various income 
brackets may affect savings in the form of life-insurance investment, 
in real-estate, n stocks, Government securities, etc. The Depart- 
ment of Commerce has initiated a project designed to discover the 
investment habits of various wealth brackets by means of an analysis 
of the composition of estates. This survey will form a basis for a 
later analysis of the effect of taxes on savings of yarious quality. 



Appendix I.— THE DEFENSE TAX BILL 

The defense tax proposals of May and June 1940 presented an 
opportunity to apply the methods developed in this study in fore- 
casting the probable incidence of new levies on consumer income. 
The following computation is presented mainly as an illustration for 
the possible use that can be made of this method. 

1. The taxes proposed in H. R. 9966 and 10039 can be summarized 
as follows: 

(1) Supertax of 10 percent on all income taxes. 

(2) Lowering of personal exemption for individual income tax to 
$800 for single persons, $2,000 for married couples. 

(3) Graduated increase in surtax rates on incomes $6,000 to 
$100,000. 

(4) All corporate income-tax rates raised by 1 percent of corporate 
income. 

(5) Supertax of 10 percent on 2, 3, and 4. 

(6) Supertax of 10 percent on estate ta^xes. 

(7) Supertax of 10 percent on Federal excises. 

(8) Increases in levies on gasoline, beer, liquor, cigarettes, transfer 
tax on stocks and bonds. 

(9) Reduction of exemption on amusements. 

A yield of $1,047,000,000 was expected from these changes. This 
proposed revenue was distributed to 1938-39 consumer incomes in 
order to ascertain what changes, if any, would be produced in the 
tax pattern. It was found that new taxes affect the total tax pattern 
to some extent. 

Table A. — Percentage of Defense Taxes Paid by Each Income Bracket 



Income bracket 



I. Under $500 

n. $500 to $1,000... 

in. $i,nooto $1,500., 

IV. $1,500 to $2,000.. 

V. $2,000 to I'^.OOO.. 

VI. $3,000 to $5,000.. 

VII. $5,000 to $10,000. 

VIII. $10,000 to .$15,000 

IX. $15,000 to $20,000 

X. $20,000 and over. 

Total 



1938-39 
taxes 



Percent 
3.7 
12.8 
15.0 
12.9 
15.1 
9.7 
6.2 
4.0 
3.6 
17.0 



100.0 



Defense 
taxes under 
H. R. 9966 



Percent 

2.2 

9.2 

12.2 

11.1 

13.8 

9.4 

6.8 

4.9 

4.6 

25.6 



100.0 



Defense 
taxes under 
H. R. 10039 



Percent 

1.4 
6.2 
9.5 
8.9 
11.2 
10.3 
6.4 
6.1 
6.2 
33.8 



100.0 



Sources: See table I and bills cited as H. R. 9966 and H. R. 10039; since this study was completed, H. R. 
10039 has been enacted into law with only very minor modifications. 



If it is intended to measure the modification of the existing tax 
patterns by a new tax proposal, the expected tax payments would be 
related to the total income of the consumer units of the various in- 
34 



CONCENTRATION OF ECONOMIC POWER 



35 



come brackets. If the intent, however, is to measure the relative 
curtaiknent of spending and saving of consumers in the various 
brackets by the new tax, the tax payments should be related to the 
income disposable for spending or saving {i. e., the income after de- 
duction of all personal taxes previously imposed). The results of 
these procedures are shown in table B. 

Table B. — Defense Taxes ' as Percent of Consumer Income * 



Income bracket 




I. Under $500. 

II. $500 to $1,000 

III. $1,000 to $1,500.. 

IV. $1,500 to $2,000.. 
v. $2,000 to $3,000.. 

VI. $3,000 to $5,000.. 

VII. $5,000 to $10,000. 

VIII. $10,000 to $15,000 

IX. $15,000 to $2Q,000 

X. $20,000 and over. 

Average., 



' See text, appendix I, p. — , for the taxes. 

2 Assumed to be the same as in 1938-39. See table I, appendix table C, and text, appendix I, pp. , 

for allocation of these taxes. 

3 Income after deduction of all personal taxes previously imposed. 

The definition of disposable income is total consumer income less 
all direct personal taxes (other than corporate income taxes). The 
proposed defense taxes on corporate income were included in disposable 
income in calculating the ratio of defense taxes to disposable income. 

2. Method of allocation. — Gasoline and cigarette taxes were allocated 
in direct ratio to expenditures on these commodities. The new tax 
on beer was proportioned to food expenditures, on liquor to recreation. 

The expected new amusement revenue was distributed in ratio to 
recreation expenditures of the six income classes under $5,000. The 
increased yields from estate and transfer taxes were allocated to the 
four brackets above $5,000,^ in proportion to 1938-39 estate taxes and 
to savings, respectively. The supertax on "other excises" was made 
proportionate to expendittires. Increases in the corporate-income tax 
were assumed to be shifted to shareholders in direct ratio to stock- 
holdings. The supertax on individual income taxes was, of course, 
presumed to fall in accordance with 1938-39 income-tax payments. 
The revenue to be produced by lower personal exemptions was as- 
sumed to be paid by the five income brackets below $5,000, in part by 
new taxpayers, in greater part by former taxpayers. These payments 
were allocated in proportion to the total estimated exemptions (in 
1937) of the two groups of taxpayers. The increase in revenue from 
higher surtax rates was allocated by applying the percent increase 
in the surtax rate to the surtax paid by the affected income bracket 
in 1937. The sum of the increased yields from brackets under 
$20,000 was deducted from the total expected yield to get a similar 
fi'^ure for the open-end class of incomes over $20,000. This figure 
agreed with the probable average surtax increase of the corresponding 
classes, and thus provided a check for the method used. 



Appendix II.— SELECTIVE BIBLIOGRAPHY 

A. BASIC SOURCES 

National Resources Committee: 

Consumer Incomes in the United States, August 1938. 
Consumer Expenditures in the United States, 1939. 
The Structure- of the American Economy, June 1939. 
Department of Commerce, Bureau of Foreign and Domestic Commerce: 
State Income Payments, 1929-37, May 1939. 
Income in the United States, 1929-37, November 1938. 
Survey of Current Business, June 1939. 
Treasury Department: 

Annual Report of the Secretary of the Treasury, 1939. 

"Federal and estimated, Bulletin, August, 1933, State and local revenue, 1938." 
Bureau of Internal Revenue: 

Internal Revenue Collections from Specific Sources, 1918-38. 
Statistics of Income for 1937: 

Part I — Individual Income Tax Returns. 

Part 11 — -Preliminary Reports of Corporation Income Tax Returns. 
Department of Commerce, Bureau of the Census: 

Statistical Abstract of the United States, 1938, 1939. 

Financial Statistics of States, 1937. ("Summary Bulletin" 1939.) 

State and Local Government, Special Sttidy No. 7. revised, "State Tax 

Collections for the Fiscal Year 1938-39." 
Financial Statistics of Cities over 100,000, 1936. 
Social Security Board: 

Social Security Bulletin, esp. April, 1939. Max J. Wasserman and John R. 
Arnold, "Old-Age Insurance: Covered Workers and Average and Median 
Taxable Wages in 1937." 
Bureau Memorandum, No. 29, 1938. "Fiscal Capacity of the States." 
Council of State Governments: The Book of the States, vol. Ill, 1939, Chicago. 
Labor Department, Bureau of Labor Statistics: Study of Consumer Purchases, 
Urban Series, 1939. 

B. SUPPLEMENTARY GENERAL MATERIAL 

National Bureau of Economic Research: Income and Wealth, vol. Ill, New York' 

1939. 
Tax Research Foundation: Tax Systems of the World, Chicago, seventh edition, 

1938. 
Twentieth Century Fund: 

Facing the Tax Problem, New York, 1937. 

Studies in Current Tax Problems, New York, 1937, especially: 

Newcomer: "Estimates of the Tax Burden on Different Income Classes." 
Shotip, Shimberg, Vickrey: "Comparison of Aggregate Burden of 

Federal and State Income Tax in Eleven Selected States." 
Reynolds: "State Property Tax Rates." 

Burr and Vickrey: "Estimating Income and Estate Tax Yields." 
Trull, Edna: Resources and Debts of the Forty-eight States, Dun & Bradstreet, 

New York, 1937. 
Dun & Bradstreet: A Survey of Taxes Paid by Business in 1938, New York, 

December 1939. 
American Economic Review: 

Vol. 27, 1937; M. Slade Kendrick— "The Incidence and Effects of Taxation: 

Some Theoretical Aspects," pp. 725-734. 
Vol. 28, 1938; Carl A. Wehrwein— "Taxes and the Consumer," pp. 92-99. 
Academy of Political Science, New York: Proceedings, vol. XI, 1924-26. W. T. 
Foster, "Some Possible Effects of Taxation on Business," pp. 39-51. 

36 



CONCENTRATION OF ECONOMIC POWER 37 

National Industrial Conference. Board Record (formerly Bulletin and Survey) : 
I, August 1939 — "Consumption Expenditures in the United States, 1909-37." 

pp. 61 flf. 
I, November 3, 1939 — "Burden of Taxation and Putolic Debt in the United 
Stfttes and other Countries 1913-38," pp. 177 ff. 
Ooeial Research Supplement No. I: G. CoUri and F. Lehmann, "Economic Conse- 
auences of Recent American Tax Policy", New York, 1938. 

C. SPECIAL TOPICS 

Academy of Political Science, New York — Proceedings, vol. XI, 1924-26, Selig- 
man — " ncome Taxes and the Price Level," pp. 3-23. 

American Economic Review: Vol. 26, September 1936. Russell Bauder — "Prob- 
able Iniiidence of Social Security Taxes," pp. 463-465. 

Automobile Manufacturers Association: Automobile Facts and Figures, 1939. 

Jacoby, Neil H.: Retail Sales Taxation, CommeTce Clearing House, Chicago, 1938. 

Jouri al of Political Economy: Vol. 47, April, June, 1939. Brown, Harry Gunni- 
s'^ A, "The Incidence of a General Output or a General Sales Tax." 

Department of Labor — Monthly Labor Review, October 1938. 

Law and Contem-porary Problems, Duke University Law School, vol. VII, No. 2, 
Spring 1940 — "Federal Income and Estate Taxation." 

Michigan — Tax Study Commission Report, "Detroit," 1939. (Similar commissions 
have functioned in many States and have issued valuable reports.) 

National Tax Association Bulletin, vol. 22, 1937. L. L. Chisholm — "Effect of 
Retail Sales Tax on the Economic Resources of the Various States." 

Taxes 17 (5) : May 1939. L. R. Bloomenthal— "Survey of Tax Exemptions." 

The Annalist, October 30, 1936. Lahee, A. W. — "An Estimate of Taxes Borne 
by Families With Incomes Under $2,000." 

Tax Magazine, October 1937. Harvey LeBun — "Consumers' Taxes v. Income 
Taxes," pp. 602 ff. August 1939. "State Tax Legislation in 1939." 

W. P. A., Division of Social Research — Urban Housing, Washington, 1938. A 
symmary of studies carried on between 1934 and 1936 in 203 localities. Data 
w^ere collected and tabulated to tehow number of dwelling units by type (e. g., 
one-family house, row house, etc.) ; the number occupied by owners, renters, or 
standing vacant; the number of owner-occupied units, by value; rental value of 
tenant units. 
. Tax Delinquency Survey, City of Hartford, Conn., 1936, Analysis of Tax Values and 
Tax Collections in Fo^ir Hamilton County, Ohio, Census Tracts, Preliminary, 1934. 
Tax-delinquency studies furnish finely detailed local break-downs of various 
types of tax-roll property, cross-classified by assessed valuation. 

The Effects of Possible Property Tax Exemption in Georgia, a Preliminary Report 
of the Survey of County and Municipal Fiscal Affairs, February 1937. Uni- 
versity of Georgia. 

Property Classification for Tax Purposes and Homestead Tax Exemption, Report of 
the North Carolina Classification Amendment Commission. 

State of Utah — Report of the Subcommittee on the Homestead Exemption Proposal 
and Taxation, October 1936. 

University of Nebraska — Studies in Business, No. 38. E. D. Schmidt — "Home- 
stead Tax Exemption." 

Arkansas — Biennial Report of the State Commissioner . of Education, 1934-36, 
pp. 76-79. These five reports include tabulations of total property in various 
low-value classes; they are suitable for correlating to income classes of occupants. 

Taxation of Real Property in Pennsylvania, 1934. E. B. Logan — Break-down on 
basis of type of property (commercial, residential, etc.) value group, average 
income, and tax as percentage of income. 

Definitions to Accompany a Quantitative Analysis of Land Use, Richmond, Va., 
Department of Public Works, Bureau of Survey and Design. 

Assessment of Montana Farm Lands. Montana Acricultural Experiment Sta- 
tion—Bulletin 348, October 1937. 

University of Denver — Bureau of Business and Social Research, Reports, vol. 9, 
No. 5, "Real Estate in Denver," June 1933. 

Dun & Bradstreet— Frederick L. Bird, The Trend of Tax Delinquency, 1930-39 
(Cities over 50,000 population). New York, 1940. 

National Municipal Review— XXYIU, No. 12, December 1939. Mohaupt, 
Rosina— "Comparative Tax Rates of 287 Cities, 1939." 

Tax Policy, V. 7, June 1938 — "Tax Delinquency and Premature Land Use Tran- 
sition." Excellent stuay as base for more precise break-downs. 



Appendix UL— TAX CAPITALIZATION 

The possibility of tax ca,pitalization was not considered in the com- 
putations but may notably affect tax and saving ratios. The property 
taxes imposed on land and buildings may be passed neither forward 
nor backward but result in a decrease in capital value obtainable by 
the owner when and if he sells such real property. Such capitaliza- 
tion is predicated upon the availability of other types of investments 
which become more attractive. The future expected taxes, thus 
capitalized, would depress real property values, being absorbed by 
wealth rather than income. When taxes are capitalized, their impact, 
by lowering the value of property, may lead to the need for reassess- 
ment and the raising of tax rates, or discovery of new sources of 
revenue — or else to increased tax delinquency and the com.ing into the 
hands of the tax authority and therefore on the m.arket of large amounts 
of property, thus further depressing real-property values. Conversely, 
lowered tax rates may be capitalized into higher capital values, leading 
to windfall profits for sellers of such property. 

The emphasis on real property in American property taxation fosters 
tax capitalization, since discrimination in favor of other types of 
property provides attractive alternative investment opportunities. 
The extent of such capitalization is perhaps evidenced by the 14 per- 
cent decline in assessed property valuations between 1932 and 1937, 
although assessment practices and lags in reassessment may explain 
it in part. However, this theoretical and actual effect assumes, first, 
the prospective transfer of property after a change in tax rates; secdnd, 
the impossibility of shifting the tax to tenants; third, the absence of 
any compensating factors which over a period of time might restore 
the old relationsliip. 

In other words, assuming that the property tax on a residence 
rented out is lowered, it is not inevitable that the rent will be likewise 
lowered, proportionately or at all. Instead, the land value might 
rise because of the lessened carrying charge. In this case, the tax 
reduction would m.ean a capital gain for the former owner but no 
advantage for the tenant. Conversely, a tax increase need not neces- 
sarily m.ean a compensatory rise in rents. 

Of course, these considerations apply mainly to the value of old 
property or only the land where new construction takes place.' 
■ The possibility of tax capitalization has been discarded in the 
preceding study, which represents simultaneous income and tax outgo, 
while tax capitalization is effective only over time and sporadically 
as individuals dispose of and acquire property. (A rapid ownership 
turn-over ^ in Aro.erica contributes to quick capitalization and the 
need of annual reassessment.) 

But the inclusion of ro.ortgage extinction and real-estate acquisition 
as increases in assets in the consum.er studies m.eans that realized 
capital gains affect the final aggregate saving figures. Whatever role 
taxes play in affecting capital values is reflected in the saving and 
dissaving shown. 



' Old property value is established mainly by the maintenance costs of new houses, including interest, 
taxes, insurance, and repairs. Interest and taxes, in torn, are determined by land values and construction 
costs. As regards land value, tax amortization may play a role, so that its part in establishing or changing 
rents (or imputed rent value of tenant-owned homes) is very minor. 

2 See U. S. Department of Agriculture Circular No. 548, M. M. Regan, "The Farm Real Estate Situa- 
tion, 1936-37, 1937-38, and 1938-39," October 1939. On p. 34, all farms changing hands by whatever method 
are estimated at from 6.2 to 7.4 percent in each of the years 1935 through 1939. 

38 



Appendix IV.— DISTRIBUTION OF TAXES ON REAL ESTATE 
BY TYPE OF OWNERS ' 

As a part of the inquiry into the economic effects of taxation, a 
questionnaire on general property taxes was prepared by the Depart- 
ment of Commerce, and the questionnaire was sent by the Urban 
Land Institute to approximately 450 local boards that are members of 
the National Association of Real Estate Boards. A copy of the ques- 
tionnaire is presented herewith. Among the inquiries was a request 
for the distribution on a percentage basis of the total taxes on real 
property by four classes of owners: (a) Businesses, including industries; 
(6) farms; (c) residential property — (1) owned homes and (2) homes 
rented out; and (d) vacant lots. 

Of the replies received, 53 supplied percentage data on this inquiry; 
the remainder either made no reply or stated that the percentage 
distribution by classes of owners was not available. The distribution 
by class of owner for the 53 cities or counties for which the information 
was tabulated is presented in the accompanying table. 

URBAN LAND INSTITUTE AND THE DEPARTMENT OF COMMERCE 

QUESTIONNAIRE 

1. What was the approximate amount of general property taxes levied or collected 

(specif}' which is being reported) in your city or county (state which unit is 
being reported) by whatever jurisdiction levied, for the fiscal year ended in 
1938? 

2. (a) Is tangible personal property taxed in your city or count}'? 

(b) Is intangible personal property taxed in your city or county? 

3. Of the total tax levy, or collection, reported in No. 1, state the approximate 

amount levied on, or collected from, real property, $ 

4. Of the total taxes on real property reported in No. 3, approximately what 

percent was levied on, or collected from, the following classes of owners: 

(a) Businesses, including industries percent 

lb) Farms percent 

(c) Residential property percent 

1. Owned homes percent 

2. Homes rented out 1 percent 

id) Vacant lots ^ percent 

Total 100 percent 

5. Are there differential tax rates on any of the classes of real property listed in 

No. 4? 

If so, please specify 

6. Please state, for the data reported in No. 3, your judgment as to what was the 

estimated ratio of assessed valuation to true valuation of real property, by 
classes: 

(a) Business and industries percent 

(b) Farms . percent 

(c) Residences percent 

1. Owned percent 

2. Rented out percent 

id) Vacant lots percent 

' Prepared by C E. Rightor, Chief Statistician, Division of State and Local Govcrnmenty Bureau of tiie 
Census, Department of Commerce. 



40 



CONCENTRATION OF ECONOMIC POWER 



URBAN LAND INSTITUTE AND THE DEPARTMENT OF COMMERCE— 

Continued 

QUESTIONNAIRE— Continued 

7. What types of privately owned real property are exempt from general property 

taxes in your city or county? (Charitable, religious, educational, industrial, 
homestead, other) 

8. Have you observed any influence of changes in tax rates or assessed valuations 

on the prices of property and the turn-over in property of various types? 

9. Any additional comments? : 



(Name of Board). 

(Address) 

(City) 



. (State) . 



Distribution by Class of Owner of Total Taxes on Real Property for 53 Cities or 
Counties, by Population Groups, 1938 



City or county and population group 



Population, 
1930 



Percentage of real-property taxes paid by class 
of owner 



Busi- 
ness 



Farms 



Vacant 
lots 



Homes, 
total 



Homes Homes 



Group I.— HAVING A POPULATION OF'500,000 AND OVER 



City: 

Milwaukee, Wis 

County: 

Los Aneeles, Calif. (Los Angeles) 

Cook, m. (Chicago) _'. 



578, 249 



2, 208, 492 
3,982,123 



Group II.— HAVING A POPULATION OF 100,000 to 500,000 



City: 

South Bend, Ind 

Des Moines, Iowa 

Grand Rapids, Mich 

Yonkers. N. Y 

Portland, Oreg 

Dallas, Tex 

Richmond, Va 

Honolulu, Hawaii 

County: 

Alameda. Calif. (Oakland) 

San Diego, Calif. (San Diego). 

Winnebaeo, 111. (Rockford) 

Genesee, Mich. (Flint) 

Tulsa, Okla. (Tulsa) 

Multnomah, Oreg. (Portland) 



104, 193 
142. 559 
168, 592 
134, 646 
301,815 
260, 475 
182, 929 
202,923 

474, 883 
209, 659 
117,373 
211,641 
187, 574 
338, 241 



Group III.— HAVING A POPULATION OF 50,000 to 100,000 



City: 

Pasadena, Cajif 

Charlotte. N. C ^ 

Altoona, Pa 

Harrisburg, Pa 

Charleston, W. Va 

Madison, Wis 

County: 

Pima, Ariz. (Tucson) 

Stanislaus, Calif. (Modesto) _ 

Burlington, N. J. (Morristown) 

Ashtabula, Ohio (Ashtabula).. 

Richland, Ohio (Mansfield) 

Grays Harbor, Wash. (Hoquiam) 

Wood, W. Va. (Parkersburg). 



76, 086 


15 




5 


80 


60 


82. 675 


34 




9 


57 


38 


82, 054 


25 




10 


65 


33 


80, 339 


60 




2 


38 


20 


60,403 


49 




15 


36 


16 


57, 899 


29 
44 






71 
44 




55, 676 


4 


8 


40 


56, 641 


25 


50 


5 


20 


15 


93, 541 


25 


15 


10 


50 


35 


68. 361 


55 


20 


5 


20 


10 


65.902 


,50 


20 


10 


20 


11 


59, 982 


60 


14 


3 


23 


18 


56, 521 


25 


5 




70 


44 



CONCENTRATION OF ECONOMIC POWER 



41 



Distribution by Class of Owner of Total Taxes on Real Property for 53 Cities or 
Counties, by Population Groups, 1938 — Continued 



City or county and population proup 



Population, 
1930 



Percentage of real-property taxes paid by class 
of owner 



Busi- 
ness 



Farms 



Vacant 
lots 



Homes, 
total 



Homes 
owned 



Homes 

rented 

out 



Group IV.— HAVING A POPULATION OF 25,000 TO 50,000 



City: 

New London, Conn 

Moline Township, 111. (Moline). 

Council Bluffs, Iowa. 

White Plains, N. Y.. 

Lima, Ohio. 

Massillon, Ohio 

Newport News, Va.. _.. 

County: 

Me.sa, Colo. (Grand Junction). _. 

Floyd, Ind. (New Albany) 

Bernalillo, N. M. (Albuquerque) 

Arlington, Va 

Whatcom^ Wash. (BeUingham) . . 



29,640 
30, 607 
42, 048 
35, 830 
42,287 
26,400 
34,417 

25,908 
34, 655 
45, 430 
26,615 
59,128 



14 
27 
25 
9 
15 



Group v.— HAVING A POPULATION OF LESS THAN 26,000 



City: 

Old Lyme, Conn 

Davtona Beach, Fla 
Ocala, Fla 

Waterville, Maine . _ 

Monroe, Mich 

Westfield, N. J 

Bron.\ville, N. Y 

Scarsdale, N. Y 

Salisbury, N. C 

Shawnee, Okla 

Waukesha, Wis 



1,313 
16. 598 

7,281 
15, 454 
18, 110 
15, 801 

6,387 

9,690 
16, 951 
23,283 
17, 176 



40 


5 


5 


5 




5 


10 


30 




12 




10 


13 


2 


45 


1 




16 
15 
20 

1 
13 

9 
14 
32 



254815— 40— No. 3- 



Appendix V.— THE METHODS OF STUDY 

The original National Resources Planning Board figures were ac- 
cepted as basic data for 1935-36, although methodological changes in 
the allocation of taxes required minor additions, described below. 

As has been pointed out in the text, it was believed preferable to 
employ crude and simple methods of allocation where data did not 
warrant refinement, and to describe the existence and extent of possible 
error, rather than mislead the reader with apparent statistical refine- 
ments no more trustworthy than the original source. 

The 1935-36. per unit income in each income class was multiplied 
by the number of consumer units in that class to get its "basic in- 
come." This amount was then raised in each case to make the total 
equal the 1938-39 consumer income computed by using 1938-39 
figures on Commerce Department incomxC payments. 

The 1938 figure of $66,271,000,000 for income payments was sub- 
tracted from the 1939 estimates of $69,683,000,000, and the difference 
split and added to the lower figure to get the 1938-39 income pay- 
ments corresponding most closely to the 1935-36 consumer income 
figure developed by the sample method of the National Resources 
Planning Board. The figure derived approximately fits fiscal year 
1939, tax receipts of which are used. 



Table C. — Basic Data on Consumer Incomes 



Income classes 


Number of 
consumer 
units ' (in 
thousands) 


Consumer 
income ' 
(millions 

of dollars) 


Income 
per con- 
sumer unit 


Increase 

over 
1935-36 per 
consumer 

unit 


Income 

including 

death 

taxes ^ 

(millions 

of dollars) 


Same 
income 
including 
corporate 
income 
and ijrofit 
taxes 3 (mil- 
lions of 
dollars) 


I. Under $500 


6,828 

•11, 852 

8,887 

5,277 

4,512 

1,850 

606 

150 

69 

112 


2,363 
10, 034 
12, 273 
10, 179 
12, 131 
7,623 
4,691 
2,008 
1,344 
5,331 


$346 

847 

1,381 

1,929 

2,689 

4,121 

7,741 

12,872 

19, 477 

47, 600 


$39 

96 

156 

218 

304 

466 

875 

1,454 

2,198 

5,373 


2,363 
10, 034 
12, 273 
10, 179 
12, 131 
7,623 
4,696 
2,028 
1.403 
5,741 


2,363 
10 038 


II. $500 to $1,000 


III. $1,000 to $1,500 


12 280 


IV. $1,500 to $2,000 ,_- 


10,210 


V. $2,000 to $3,000 


12 194 


VI. $3,000 to $5,000 


7,743 


VII. $5,000 to $10,000 _ 

VIII. $10,000 to $15,000 

IX. $15,000 to $20,000 

X. $20,000 and over 


4,861 
2,238 
1,601 
6,333 


Total 


40, 149 


67, 977 


1,693 


191 


68.471 69 Sfil 









• Extrapolated, on the basis of census estimates, from figures in National Resources Planning Board— 
Consumer Incomes in the United States, 1938. 

2 Extrapolated, on the basis of Department of Commerce estimates, from figures ibid. 

3 Taxes from Annual Report of the Secretary of the Treasury, 1939, and State and Local Government, Special 
Study No. 7, revised "State Tax Collections." For allocation, see pp. 45-8. 

A "basic expenditures" figure was derived in a similar manner. 

The difference between the original 1935-36 and the new 1938-39 

per unit incomes was broken down to show per unit increases in 

savings and in expenditures. The ratio of these items was obtained 

42 



CONCENTRATION OF ECONOMIC POWER 



43 



by interpolating the percentages of each $10,000,000,000 additional 
income saved and spent by the receivers of the three-thirds of con- 
sumer income. (See tables in National Resources Planning Board 
Consumer Expenditures in the United States, pp. 174 and 184, and 
Structure of the American Economy, p. 91.) 

These figures do not correspond to the arithmetical progression 
indicated by the independent Securities and Exchange Commission 
estimates of savings of $17,500,000,000 in 1936, with a national 
income of $65,226,000,000, and $18,500,000,000 in 1937, when income 
rose to $71,853,000,000. Smce $6,000,000,000 was believed to be 
consumer savings in 1936 (which agrees very well with the National 
Resources Planning Board computation of $5,978,000,000), the same 
percentage (34.3) applied to 1937 would have given only $6,120,000,000 
consumer savings. On the other hand, the 1935-36 National Re- 
sources Planning Board estimates show consumer saving increasing 
more rapidly than consumer income. The same proportion of savings 
as 10 percent of income would have given $6,735,000,000 as the con- 
sumer savings for 1937-38, and $6,847,000,000 in 1938-39. But if 
savings do increase at a faster rate than does income, as the increase 
in all bank-deposit figures and the decrease in Government housing 
and agricultural loans would suggest, then the savings estimate 
developed in the study is fully justified. 

The aggregates for basic and for additional expenditures were 
added to get a preliminary expenditure figure for each income class. 
To get the final expenditure figures in line with the Board's definitions 
for the purpose of comparability, it was necessary to deduct personal 
taxes and gifts (1935-36 figures raised in ratio to the income increase) 
from either expenditures or savings. Deductions were made from 
expenditures for the income groups up to $2,000 (see text, p. 11), and 
from savings for other brackets (with one minor exception) . 



Table D. — Gifts, Savings, and Expenditures 



Income classes 


Gifts (in 

millions of 

dollars) 


Total sav- 
in?.'! (in 

millions of 
dollars) 


Tax deduc- 
tions for 

calculating 
expendi- 
tures fin 

rrillions of 
dollars) 


Net ex- 
penditures 
(in millions 
r,( doUars) 


I. Under .'!>500 - 


39 
277 
418 
382 
490 
352 
231 
99 
56 

n6 


-575 

-197 

642 

,193 

1.167 

1,299 

1,380 

72C 

517 

2,425 


31.7 
85.1 
113.5 
117.2 
124.7 
141.5 
146.6 
140.8 
125.5 
1,060.3 


2,868 


II. $50'no.$f.000_. . 


9. 869 


in. $i.oocto$i..'iOo... 


11,100 


IV. $1,500 to $2,000 


9,086 


V. .'i;2,(HX) to .$3,000 ..- -- 


10, 350 


Vr. .$3,000 to .fi.5,000. 


5,831 


VII. .$5,000 to $10.000 


2,934 


VIII. $10,000 to $15,000 


1,052 


IX. $15,000 to .$20,000 


645 


X. $20,000 and over... . 


1,730 






Total'- 


2; 459 


7,977 


2,p78 


55, 465 







' Totals arc given to the closest million. 

Sources: Basic data from National Resources Planning Board Constimer Eipendituns in the United •'>'iatfs, 
extrapolated. For tax sources, see table C, note 3. 

The e.xpenditure and savings figures thus derived were final; but 
income figures, although used as the basis for distributing dependent 
data, were further adjusted before their use for developing group 
percentages by the addition of inheritance, estate, and gift taxes to 



44 



CONCENTRATION OF ECONOMIC POWER 



the proper income brackets; likewise corporate and profits taxes, 
after distribution to the appropriate income brackets, were added to 
income, since the imputation of cofasumer payment of these taxes 
implies precedent receipt of at least that amount of profits. (See 
table C.) 

Taxes were attributed to particular income brackets by methods 
appropriate, as far as possible, to the tax involved. Statistics of 
Income ratios of taxes paid in various income brackets were used for 
allocating individual income taxes. 



Table E. — Tax Receipts, Fiscal Year 1939 

[In millions of doUars] 



Item 


Federal J 


State 2 


Local' 


Customs revenues. - - 


318.8 
1, 028. 8 
1, 122. 5 

127.2 
27.1 

360.7 






Individual-income taxes __ _ _ 

Corporate-income taxes. I 


186.9 
* 142. 7 








Excess-profits taxes - -.- --. 






Estate, inheritance, gift taxes 


133.0 

5.8 

362.3 

800.9 

59.5 

' 218, 5 




Poll taxes .- 


(') 






'30 




207.0 
580.2 

.587.8 
190.0 
41.1 

8 62.7 




Tobacco taxes 


20 


Alcoholic-be veraee taxes ... . . .:. .. 


20 


Manufacturers excises 




Stamp and documentary taxes ..1. 


27.6 

9 206. 8 

44.4 

11 161.6 

442.3 

241.2 

799.0 




Taxes on amusements, utilities, and insurance companies 










10 40. 6 


330 






General property taxes . ... ... 




4,300 




740.4 
65.4 




Sugar Act of 1937 










Total 


5, 500. 3 


3, 832. 


4,700 







• Federal tax receipts from Annual Report of the Secretary of the Treasury, 1939. 

2 State tax collections from Bureau of the Census, State and local government, Special Study No. 7, Re- 
vised— "State Tax Collections, Fiscal Year 1938-39." 

3 Local taxes estimated roughly on basis of 1938 figure cited in Bulletin of the Treasury Department, August 
1939. See text, table V, note 4, regarding later estimates by the Tax Policy League. 

< "Undistributed" State income taxes of $29,700,000 are included here, but not in the distribution of 
corporate income taxes to income brackets. 

5 The estimated receipts from local poll taxes in earlier years has been around $10,000,000. The trend 
away from this tax precluded any rational estimate. 

« Including gasoline taxes. 

' Includes selective sales tax and licenses. 

8 Includes taxes on communication and transportation facilities, leases of safe deposit boxes, checks, etc., 
aimissions and club dues. 

' Gross receipts taxes. 

1" Includes taxes on oleomargarine, narcotics, processed oils, crude petroleum, bituminous coal, unjust 
enrichment. , 

'1 Includes selective sales tax on commodities other than gasoline, tobacco and alcoholic bever^es and 
license taxes other than on alcoholic beverages. 

The inconsistencies in the concepts of consum.er incoin.e as found in 
the National Resources Planning Board study and net income as 
defined for individual income tax purposes might well affect the dis- 
tribution of income taxes aro.ong consumer incom.e levels. No adjust- 
m.ent was sought for the diflPerences between consum.er unit and 
individual, between the inclusiveness of deductions and incom.e under 
the two respectivfi definitions, of the Board and the Treasury for the 
two following reasons: In the first place, the adjustm.ents for income 
had already been m.ade in the Board study, and particularly affected 
the totals of the uppcr-incoro.e brackets; in the second place, what 
was sought was the distribution of actual taxes paid in relation to 
income received by consumers. So aggregate actual receipts of 
individual incomes taxes were distributed among the income groups 



CONCENTRATION OF ECONOMIC POWER 



45 



on the basis of the percentage of such receipts paid by the correspond- 
ing individual income group (Statistics of Income, pt I, 1937; table 3, 
p. 121). As has been pointed out, this may lead to an overestimate 
of the taxes paid by lower-incoro.e families; but the concentration of 
income taxes paid in the thi-ee highest brackets indicates that any 
error for the groups below would be extrem.ely sm.all in terms of per- 
centage of income. A decrease of Federal individual income taxes 
unputed to the $5,000-$10,000 group by even $15,000,000 would 
lower their total tax burden by 0.3 percent of their income. Since 
adjustments would be hypothetical (see National Bureau of Eco- 
nomic Research, Income and Wealth, vol. Ill, Enid Baird and 
Selma Fine, "The Use of Income Tax Data in the National Resources 
Committee Estim.ate of the Distribution of Incom.e by Size," pp. 149- 
203, and discussion by A. J. Golden thai, pp. 204-215), and some 
degrees removed from even 1935-36 accuracy by the extrapolation 
of income data to 1938-39 and the use of incomplete income tax 
data not corresponding fully to the income year, the making of ad- 
justments was again felt to be misleading, since they would imply a 
greater precision and give an appearance of real accuracy to merely 
plausible figures. Furthermore, as has also been indicated, the range 
of each of the higher income brackets eliminates error to some extent. 
An increase of even $77,400,000 (assuming a 10-percent error) in 
the top bracket's share would raise their tax burden by 1.2 percent 
of income. 

Table F. — Personal Taxes by Income Brackets 
[All money figures in millions of dollars] 



Income classes 



Individual 
income 
taxes, 
Federal 

and State 



Social se- 
curity 

employee 
taxes 



Inheritance, 

estate, gift 

taxes. 

Federal 

and State 



Poll taxes, 
State 



Personal 
property 

taxes— 
two-thirds 

of total 



I. Under $500 

II. $500 to $1,000. 

III. $1,000 to $1,5001 - 

IV. $1,500 to $2,000-. 
V. $2,000 to $3,000.. 

VI. .$3,000 to $5,000-- 

VII. $5,000 to $10,000. 

VIII. $10,000 to $15,000. 

IX. $15,000 to $20,000 

X. $20,000 and over- 

TotaP- 



0.0 
.1 

6.3 
12.2 
16.7 
41.4 
89.0 
73.9 
61.3 
914.7 



26.2 
65.2 
77.4 
63.4 
54.7 
33.3 



5.0 
20.0 
59.0 
410.0 



0.0 
2.6 
1.4 
■ .8 
.7 
.3 
.1 
.1 



5.6 
17.4 
28.4 
40.8 
52.6 
66.5 
67.6 
66.9 
64.2 
145.6 



1, 216. 



320.0 



494.0 



546. 



1 Totals are given to the closest million. 



Sources: See table E; for allocations of taxes to income brackets, see text, p. 17, on income taxes; p. 17, 
on Social security taxes; pp. 17, 18, on "death" taxes, pp. 17, 18, on poll and personal property taxes. 

The allocation of State individual income taxes according to Federal 
tax ratios was made necessary by the absence of adequate State data 
on income distribution by levels corresponding to the consumer income 
study. The small amount involved ($187,000,000) in comparison to 
total income made this procedure practicable, particularly since dis- 
tortion would be in the opposite direction from that caused by the 
nonadjustment of Federal individual income taxes, for State taxes 
tend to have a lower exemption than do Federal. 



46 CONCENTRATION OF ECONOMIC POWER 

Social Security employee old-age contributions were obtained from 
the Annual Report of the Secretary of the Treasury, 1940, page 1568, 
as follows: 

A. Total Federal employment taxes $631, 000, 000 



1. Under title VIII 529,800,000 

2. Under title IX 101, 200, 000 



B. Railroad carriers' taxes .. 109, 400, 000 



Paid bv employees: 

One-half of 1 264,900,000 

One-half of B ■ 54,700,000 



Total 319,600,000 

State and Local Government, Special Study No. 7, revised, Bureau 
of Census — "State Tax Collections", page 9: 

C. Unemployment compensation taxes, (all except railroad) $799,000,000 

Percent 

Employer taxes $1, 165, 000, 000 54. 7 

Employee contributions 265, 000, 000 54. 7 

They were allocated on the basis of the percentage of total taxable 
wages received by each income bracket, stopping at an open-end class 
of "over $3,000," as given in the April 1939 number of the Social 
Security Board Bulletin (p. 4); this class may have compensated for 
the somewhat arbitrary assignment of death taxes at a steeply pro- 
gressive rate to the four top income brackets. Again, no adjustments 
for family income have been made, nor could they change the shape 
of the curve, since despite the concentration of these taxes in the six 
lowest income classes, their percentual relationship to income is in the 
vicinity of 1 percent or less. Consumption taxes and shifted taxes are 
of primar}^ importance in the tax burden of these classes. 

The estate taxes (Federal plus State) were split as between the two 
top brackets by arbitrarily assigning taxes on net estates over $50,000 
(actually $100,000, in view of the specific exemption) to the income 
classes above $15,000, and splitting taxes on smaller estates in propor- 
tion to aggregate income received by that class and by the next two 
preceding ($5,000 to $15,000). Gift taxes were all attributed to the 
two top brackets, as it is doubtful that traceable gifts over the exemp- 
tions would be made, even over time, by receivers of income less than 
$15,000. 

The proportions to be allocated to each income class were derived 
as follows: 

The total tax liability of half of the net estates smaller than $50,000 
was figured in proportion to the liability of total net estates. This 
ratio was applied to inheritance and estate taxes to be attributed to 
the income classes between $5,000 and $15,000. The rest, plus gift 
taxes, was attributed to the classes over $15,000. Each total was 
divided between its respective two income classes in proportion to total 
personal taxes paid by the income classes. 

Poll taxes were treated as proportionate to consumer unit number. 
Personal-property taxes were the most difficult to allocate, if only 
because the very data had to be estimated. The personal-property 
tax revenue, aside from the special property taxes on utilities, was 
divided into three parts — one third was attributed to business, the 



CONCENTRATION OF ECONOMIC POWER 



47" 



other two parts to consumers. It was assumed that one of these 
parts was paid on intangibles and one on tangible property. The 
latter was therefore allocated in proportion to 1935-36 expenditures 
on furnishings, automobiles, and, in the, four high brackets, on part 
of clothing. The tax on intangible personal property was allocated 
by developing and extrapolating the estimates on the distribution of 
stockholdings given by Colm and Lehmann in Economic Conse- 
quences of Recent American Tax Policy (p. 46) (based on Twentieth 
Century Fund figures). The dividends were also assumed to be 
proportionate, and therefore corporate income taxes, in assuming 
forward shifting of all taxes, were distributed on the same basis. 
The original percentages used for the extrapolation were: 34.3 percent 
of dividends go to brackets which are nontaxable or under $6,000; 
23.1 percent go to people in the $6,000 to $20,000 classes, and 42.6 
percent to the rest. The results of the interpolation and adjustment 
of these figures are shown in table J. 



Table G.- 



-Specific Consumption Taxes and All Consumption Taxes, by Income 
Brackets 



Income classes 


Taxes on specific 
consumption 
(in millions of 
dollars) 


Taxes shifted to 
consumption 
(in millions of 
dollars) 


Total taxes on 
consumption 
(in millions of 
dollars) 


Percent- 
age of 
consump- 
tion taxes 
paid by 

each 
bracket 




Federal 


State and 
local 


Federal 


Stat • and 
local ' 


Federal 


State and 
local 


I. Under $500 . . . 


71 
287 
345 
286 
327 
181 
87 
31 
19 
48 


103 

356 

445 

397 

484 

282 

147 

55 

32 

92 


90 
309 
348 
284 
324 
183 
92 
33 
21 
54 


222 
764 
861 
705 
804 
453 
,228 
81 
52 
134 


161 
596 
693 
570 
651 
364 
170 
64 
40 
102 


325 

1.120 

1,306 

1,102 

1,288 

735 

375 

136 

84 

226 


3.7 


II. $500 to $1,000 - 


12.8 


III. $1,000 to $1,200 


15.0 


IV. $1,200 to $2,000 

V. $2,000 to $3,000 


12.9 
15.1 


VI. $3,000 to $5,000 


9.7 


VII. $5,000 to $10,000- 


6.2 


VIII. $10,000 to $15,000 


4.0 


IX. $15,000 to $20,000 


3.6 


X. $20,000 and over 


17.0 


Totals 


1,682 


2,393 


1,738 


4,304 


3,420 


6,697 


100.0 

■ 



■ Includes general sales taxes. 

2 Totals are given to the closest million. 

Sources: See table E, and text, pp. 12, 13 ff, 18 ff. 

Consumer taxes were distributed on the basis of the National 
Resources Planning Board Consumer Expenditures study. Gasoline 
taxes and the nonbusiness part of' manufacturers' automobile excises 
were distributed in proportion to spending on automobiles. As has 
been noted in the text, this procedure probably moderated the 
regressiveness of gasoline taxes, since with decrease in size of income 
a progressively larger portion of automobile and gosoline expenditures 
goes for the fuel, while in the upper-income brackets a much larger 
part of automobile expenditures in vcJ.'v^ represents the annual or 
semiannual purchase of a new car, and the car is generally higher 
priced. Since manufacturers' excises on automobiles are very small 
in comparison to levies on gasoline, the procedure overrepresents, 
maj^be substantially the share of consumer taxes paid by the upper 
brackets. 

The only uncertainty with regard to tobacco taxes is the relative 
consumption by various classes of particular types of tobacco. It is 



48 CONCENTRATION OF ECONOMIC POWER 

doubtful, however, that any appreciable difference would result from 
a distribution of total tobacco taxes among the income classes in 
proportion to their expenditures on tobacco, since such spending, 
however tremendous as a sum, makes a very small percentage dent 
in consumer expenditures. 

Taxes on foods are laid by the Federal Government on oleomar- 
garine, mixed flour, filled cheese, processed butter, and various oils 
which are substitutes for butter. These taxes amount to little and 
are probably paid by the lowest-income groups who tend to use the 
cheaper substitute foods, but since the actual proportions are not 
available and the amounts involved are very small, they were attributed 
in proportion to food expenditures without regard for actual inci- 
dence. 

The allocation of other excises, especially on liquor and gasoline, 
has been discussed in detail in the text. 

BUSINESS TAXES 

There is a sizable discrepancy between figures on actual receipts 
and taxes payable by corporations (such differences exist for other 
taxes but are of far smaller amounts). There would be reason to 
take either figure, since the economic effects of taxes result both from 
the fact of their being due and from their actual payment. However, 
in view of the fact of the receipts basis of the personal income and tax 
tables, it was thought advisable to use the lower figures on taxes 
actually paid. 

On the basis of Statistics of Income figures, ratios of taxes to pro- 
duction in various industrial branches vary widely and disparately, 
at least for corporate enterprise. Thus, normal income and surtaxes 
and excess-profits taxes varied in 1937 from 1.3 percent of value 
added in services, to 6.1 percent for mining (although the ratio is 
even higher in particular manufacturing industries, as for motor 
vehicles and tobacco). But for all taxes (i. e., those specified for 
deduction, not including those included in costs of operation, etc.) 
in relation to value added, the spread was from 6.6 percent of trade 
to 22.9 percent in transportation and utilities. The average for all 
corporations was 3.2 percent and 12.3 percent for the respective 
ratios. The same taxes as a ratio of gross receipts of all corporations 
were 3.5 percent, 1.2 percent for trade and 8.6 percent for transpor- 
tation and utilities. 

This method of utiUzing Statistics of Income data on corporations 
provides an interesting check on the direct allocation of taxes to 
sectors of industry. Of course, the reliability of the check itself 
depends on the extent of corporate enterprise in the particular sector. 

Since unincorporated professional people report earnings subject to 
individual income tax and are not liable to business taxes as such, 
there appears to be no need to include them in the business structure, 
although incorporated professional services are so included. It is 
always possible that the professional man makes allowance, in setting 
his fees, for his taxes; but the approximation is so vague and the 
eventual incidence so nebulous (and infinitesimal in the sum total of 
taxes) that it seemed advisable to regard his earnings and income tax 
with the personal income tax of other strata; i. e., as nonshif table. 

With regard to trade and construction, the solution was more prob- 
lematical. Construction is still largely in an- imorganized condition. 



CONCENTRATION OF ECONOMIC POWER 49 

with many independent unincorporated contractors who pay general 
business and personal income but not corporation taxes. This dis- 
tinction may well be ignored in view of the depression of the building 
industry and the very small taxes paid by it. (It paid $15,000,000 
Federal corporate income and surtax and excess-profits tax in 1937.) 

This is likewise true of the numerous though perhaps not very 
profitable unincorporated retailers.^ The small margin of net income 
on a rather tremendous gross done by trade and the large proportion 
of annual failures, suggests that small traders make small profits not 
subject to profit taxes. 

In analyzing th6se data, it is important to include the gross income 
of concerns without net income, as well as those making a profit. In 
•evaluating taxes due, various deductions are allowed; except for cost 
of operations, cost of goods sold, rent on business property, and depre- 
ciation and depletion none would diminish value added. The deduc- 
tions should therefore be computed on this limited base in arriving at 
taxes paid by various industries. Almost the same results are achieved 
by subtracting these items from total gross receipts, except for finan- 
cial institutions, where the difference is considerable, apparently due 
to the inclusion of nonfinancial business conducted by such institutions. 

The allocation of State and local taxes was made for aggregates. 
However, their actual incidence may depend very largely on the extent 
and localization of the market of each industry. The National Re- 
sources Planning Board report on the Structure of the American 
Economy, presents the following list of industries with local markets: 

Nonalcoholic beverages. Stereotyping, electrqtyping. 

Bread and bakery products. Compressed and liquefied gases. 

Ice cream. Manufactured gas. 

Manufactured ice. Unclassified sheet-metal work. 

Engraving (lithographing, photo- Foundries. 

engraving, etc.). Machine shops. 
Newspaper and periodical printing and Signs, advertising novelties. 

publishing. 

It will be noted that some of these products (viz, ice cream, news- 
papers), despite their- local market, have prices set by custom or 
Nation-wide corporations, so that special taxes in difl"erent localities 
must be absorbed by the enterprise itself in whatever manner is 
possible. The same holds true jfor some regional industries, which 
follow : 

Fertilizers. Wrought pipe. 

Feed's for animals and fowls. Structural and ornamental metal work. 

Liquors, malt. Lighting equipment. 

Wooden boxes, cooperage. Coke-oven products, fuel briquets. 

Planing-mill products. Cement, concrete products. 

Iron and steel forgings. Lime, sand-lime brick. 

Steel barrels, kegs, drums. Marble, granite, and other stone. 
Steam and hot-water apparatus and Carriages, wagons, sleighs, sleds, 

fittings. Paving materials, blocks, mixtures. 

The large number of construction items with a regional market and 
the known divergent prices in that field suggest that different local 
costs, including taxes, may well be apportioned very diversely. 

Available data give total State and local property taxes (including 
personalty which is estimated unofficially at 19 percent), farm prop- 

I (See National Resources Plannintr Board, Structure of the American Economy, pp. 102-104.) 30 psr- 
oent of retail sales of 1935 were made by independent stores with annual sales under $30,000 and employing 
one or two persons. One-third of the construction by private firms was done by firms doing less than 
$50,000 work a year each. 



50 



CONCENTRATION OF ECONOMIC POWER 



erty value, at 5-year intervals, with subdivisions for operators' 
dwellings, assessed valuations of all taxable property mortgage debt, 
and value of all homes. Since most of the figures are estimates, they 
are usable only with constant corrections, which reduce the value of 
final results even more. The originally computed agricultural real- 
property figure was adopted because it was best borne out by several 
spot studies. Thus, in Utah in 1935, two-fifths of the privately 
owned land in the State Avas owned by farmers and stockmen (cf. 
State of Utah, Report of the Subcommittee on the Homestead Exemp- 
tion Proposal and Taxation, October 1936). (See also Montana 
Agricultural Experiment Station Bulletin No. 348, October 1937, 
"Assessment of Montana Farm Lands", p. 49, citing that farm land 
and buildings are one-third the property tax base; and A Preliminary 
Report of the Survey of County and Municipal Fiscal Affairs Relating 
to the Effect of Possible Property Tax Exemption in Georgia, February 
1937, University of Georgia, showing that farm taxes are about half 
the total homestead taxes.) Making allowance for the predominantly 
agricultural character of some of the "check" States (but also for the 
poverty and low values of their farms, in comparison to the more pros- 
perous dairy and grain States) and homestead exemption, the selected 
figure of $1,203,000,000 as nonresidential farm real property taxes ap- 
pears as adequatel}^ reasonable, taking the total as $3,630,000,000. 

On a different basis, using Statistical Abstract figures (1938, pp. 
568-604) for farm property and taxes, it mu^t be admitted that the 
agricultural property figure may be rather heavily overestimated. 
If so, the amount of difference should be attributed to homes, for which 
the minimum figure was taken. In the final analysis, i. e.. the pro- 
cedures used to shift all business taxes to the incomes, the precision 
of intermediate allocation is of less importance than is the inclusiv€- 
ness of the figure. 

The remainder of $1,186,000,000 property taxes was allocated on a 
1935-36 wealth basis, except for utilities, which are specially taxed. 

The unemployment compensation tax and one-half the old-age 
insurance tax were each independently allocated on the basis of per- 
centages paid by each industry. 



Table H. — Taxes on Business Shifted to Consumers and Shareholders 

[In millions of dollars] 





Corporate taxes shifted 


Taxes 
shifted to 
consump- 
tion I 


Total taxes shifted 


income classes 


Federal 


State 


Federal 


State and 
local 


I. XTnder.$500 l ".. 


0.0 
3.7 

6.4 
28.5 
.57.9 
110.2 
151. 6 
192.9 
181.9 
543.9 


0.0 

.3 

.6 

2.5 

5.1 

9.8 

13.4 

17.1 

16.1 

48.1 


312.1 

1, 073. 

1, 209. 2 

989.0 

1, 128. 1 

636.4 

319.6 

114.1 

73.2 

187.5 


90 
313 
3.54 
313 
382 
293 
244 
226 
203 
598 


222 


n. $500 to $1,000 ■ 


764 


in. $1,000 to $1,500 


862 


IV. $1 ,500 to $2,000 

V. .$2,000 to .$3,000 


708 
809 


VI. $3,000 to .$5,000 


403 


VII. $5,000 to $10,000 


241 


VIII. $10,000 to $15,000.--. 


98 


IX. $15,000 to $20.000 


68 


X. $20,000 and over 


182 






Total 2 


1, 277. 


113.0 


6, 012. 2 


3, Bio 


4,417 



' This is the sum of the fourth and fifth columns in table G. 
2 Totals are given to the closest million. 

Sources: See table E, and text, pp. 15 f, 23 fl; appendix V, p. 48 ff. 



CONCENTRATION OF ECONOMIC POWER 



51 



Table J. — All Personal Taxes, Including Taxes Shifted to Shareholders 

[AH money figures, except income classes, in millions of dollars] 





All direct personal 
taxes 


Corporate taxes 
shifted 


All per- 
sonal 
taxes, 
direct 
and 
shifted 


Percent 

of all 
personal 


Income classes 


Federal 


State 


Federal 


State 


taxes paid 
by each 
income 
bracket 


I. Under $500- 

II. $500 to $1,000 . 


26.2 
65.2 
82.7 
73.7 
68.8 
68.3 
79.0 
77.1 
94.9 
1, 073. 7 


5.5 
19.9 
30.8 
43.5 
,55. 9 
73.2 
72.6 
83.7 
89.6 
396.6 


0.0 

3.7 

6.4 

28.5 

57.9 

110.2 

151.6 

192.9 

181.9 

543.9 


0.0 

.3 

.6 

2.5 

5.1 

9.8 

13.4 

17.1 

16.1 

48.1 


31.7 
.89.1 
120.6 
148.2 
187.7 
261.5 
316.6 
370.8 
382 5 


0.8 
2 3 


III. $1,000 to $1,500 - 


3 1 


IV. $1,500 to $2,000 


3 7 


V. $2,000 to $3,000 


4 7 


Vr. $3,000 to $5,000 


6 6 


VII. $5,000 to $10,000 


8 


VTII. $10,000 to $15,000 _. 

IX. $15,000 to $20,000 


9.3 
e, 


X. $20,000 and over 


2. 062. 3 51. 9 


Total > 


1-, 710. 


871.0 


1, 277. 


113.0 


3,971.0 


100 







' Totals are given to the closest million. 

Sources: See table E, and text pp. 17, 23; 27, appendix V, pp. 45, 48. 

The shifting of corporate income and profits taxes has already been 
adequately covered. Most other business taxes, with the exception 
of sales taxes but including employers' Social Security taxes, were 
shifted in proportion to total expenditures. Adequate data exist for 
the allocation of- sales taxes to the products on wliich these taxes are 
paid. The percentages in several States making surveys of their in- 
cidence have been surprisingly constant, with major differences due 
to the liability or immunity to such taxes of utilities and housmg. 
The percentages shown by four ^ such State surveys were averaged, and 
total sales-tax revenues distributed proportionately. Then the por- 
tion falling on each commodity was further allocated in proportion to 
the expenditures of each income group on that com.m.oditj'". 

It would, of course, be desirable to have a similar segregation for 
each business tax paid. 

2 Only four were used, because the sales tax had been in force in eacb for 2 or more years, so that the initial 
difRculties of enforcement in certain fields had been ironed out and a relative stability of yield achieved. 
Beatable VI. 



I K D E X 



Page 
BARTON, BRUCE: Questionnaire sent to families in $6,000 to $40,000 
income classes, 1939: Where would they cut expenditures in case of 10- 

percent-income-tax increase; comment .. 11 

BUSINESS TAX: Business taxes as percent of income produced, 1938-39; 

comment and table 1 22 

Data available; comment on 48 

Taxes on business shifted to consumers and shareholders; dollar 

amounts, by income classes; comment and table 50 

CAPITALIZATION OF TAX 38 

COLM, GERHARD. Consultant in preparation of monograph vii 

CONSUMER INCOME. See Income, Consumer. 

CONSUMPTION TAX. Influence on consumption 30 

New York City cigarette tax, evasion in 1939 31 

Specific consumption taxes and all consumption taxes, dollar amounts, 

by income brackets ; comment and table 47 

Specific consumption taxes, 1938-39; comment and table 19 

Reference: Lough, W. H. High-level consumption, 1935; cited 18 

CUSTOMS DUTIES. Distribution of customs duties, by type of im- 
port, 1935-36, 1938-39; comment and table 14 

DEFENSE TAX BILL. Analysis of bill 34-35 

EXPENDITURE. Tax deductions and net; dollar amounts, comment 

and table 43 

GIFT. Savings, expenditures and gifts, dollar amounts, by income 

classes ; comment and table 43 

HOMES. Percentage distribution of tax on real property on owned and 

rented homes in each of 53 cities or counties, 1^38; table 40 

INCOME: References. Baird, Enid, and Selma Fine: Use of income tax 
data in the National Resources Committee estimate of distribution of 

income by size; cited :__: , 45 

Department of Commerce: Income in the United States, 1929-37; 

1 938 ; cited . . 28 

Hicks, Ursula: Effects of financial policy on the distribution of income 

in Great Britain since the war ; cited 27 

National Resources and Planning Board: Consumer expenditures, 

1935-36, 1939 

Consumer incomes, 1935-36, 1939 3, 6 

INCOME, CONSUMER. Basic date on consumer incomes, 1938-39: 
Number of consumer units, dollar amount of consumer incomes, income 
per consumer unit, increase over 1935-36 per consumer unit, incomes 
including death and corporate income and profit taxes; comment and 

table ■- - 42 

Business taxes as percent of income produced, 1938-39; comment and 

table - ---- 22 

Consumer savings by income brackets, related to consumer incomes 

in 1935-36 and 1938-39; comment and table 33 

Consumer units and income by income brackets; table and chart 7 

Defense taxes as percent of consumer income; comment and table 35 

Distribution and disposition of consumer income, United States, 1938- 

39 ; charts 7 

Measurement of 3 

National Resources and Planning Board, consumer income, 1935-36, 

used as basis for measuring income 3 

Original personal, and specific consumption taxes plus business taxes 

shifted forward as percentages of consumer income, 1938-39; chart. 5 

53 



54 INDEX 

INCOME CONSUMER— Continued. Page 
Personal and specific consumption taxes in percent of consumer in- 
come, 1938-39; comment and table 12 

Personal and specific consumption taxes as percentages of consumer 

income, 1938-39; chart 4 

Percentage of defense taxes paid by each income bracket, 1938-39 

taxes, taxes under H. R. 9966 and H. R. 10039 34 

Savings and taxes. Federal, State, and local, as percent of consumer 

income, 1938-39, comment, table 1, and chart ^^ 6, 7 

Tax and income percerltages for 1938-39, cumulated from lowest 

brackets; comment and table 28 

Taxes, all taxes, in percent of consumer income, 1938-39; comment 

and table 12 

Taxes and income, 1932, comment and table 28 

Taxes and savings as percent of consumer income per consumer unit, 

1938-39 30 

Taxes in relation to consumer income, 1938-39, if 1932 tax rates had 

been in force in 1938-39; comment and table 29 

Total taxes as percent of consumer income; if business taxes were 

shifted to wages, 1938-39; comment and table 26 

INCOME, NATIONAL. Department of Commerce estimates, limita- 
tions 9 

Taxes as percentage of national income in United States and Great 

Britain, 1935-36, 1938-39; comment and table 27 

NATIONAL RESOURCES PLANNING BOARD. Consumer expendi- 
tures, 1935-36; and consumer incomes, 1935-36, used as basis for measur- 
ing income in preparation of monograph 3 

Structure of American economy, cited 11 

PERSONAL TAXES. All personal taxes, including taxes shifted to 

shareholders; dollar amounts, by income classes; comment and table 51 

Income, social security, inheritance, estate and gift, poll taxes, per- 
sonal property; by income brackets; dollar amounts; comment and 

table 45 

REAL ESTATE. Distribution by class of owner of total taxe? on real 

property for 53 cities or counties by population groups, 1938; table 40 

Distribution of taxes by type of owner 40 

RIGHT or C. E. Distribution of taxes on real estate by type of owners.. 39 

SALES TAXES. Estimates of State sales taxes paid by commodity 

groups, 1937-39; comment and table 25 

SAVINGS. All taxes and savings as percent of consumer income, 1938-39; 

, comment, table 1 , and charts 6, 7 

Consumer savings by income brackets, related to consumer incomes in 

1935-36 and 1938-39; comment and table 33 

Gifts, savings and expenditures, dollar amounts, by income classes; 

comment and table 43 

Influence of taxes on 32 

Taxes and savings as percent of consumer income per consumer unit, 

1938—39 . . . .. 30 

TARASOV, HELENV " Who'Pays "the Taxes?" " FrepaTedhy'/////////." VII 

TARIFF: References, Ellis, Lippert S. Tariff" on sugar. 1933; cited 14 

Mohat, Haldore R. Tariff on wool, 1935; cited 14 

TAXES. Customs duties. See Customs Duties. 

Deductions for calculating expenditures; comment and table 43 

Federal, State, and local, and savings as percent of consumer in- 
come, 1938-39; comment, table 1 and chart 6,7 

Federal, State, and local tax receipts, fiscal year 1939, by sources; 

comment and table 44 

Income and taxes, 1932; comment and table 28 

Influence of, on business 32 

Influence on consumption 30 

Influence of, on savings 32 

Original, personal, and specific consumption taxes plus business 
taxes shifted forward as percentages of consumer income, 1938-39; 

chart 5 

Percentage of national income in United States and Great Britain, 

1935-36, 1938-39; comment and table 27 



INDEX 



55 



TAXES— Continued. Page 

Personal and specific consumption taxes as percentages of consumer 

income, 1938-39; chart 4 

Relation to consumer income, 1938-39, if 1932 tax rates had been in 

force ; comment and table 29 

Savings and taxes as percent of consumer income, per consumer unit, 

1938-39 30 

Taxes paid by sectors of industry, Federal, State, and local; comment 

and table 22 

TAXES. References: Colm and Lehman, Economic Consequences of 

Recent American Tax Policy. 1938; cited 23 

Dun & Bradstreet. Survey of Taxes paid by Business in 1938; 

cited 21 

Great Britain. Committee on National Debt and Taxation (Colwyn 

Report) 1927 1 

Hynning, Clifford. Taxation of Corporate Enterprise; cited 12 

Internal Revenue Bureau. Internal revenue collections from specific 

sources, 1918-38; cited 28 

Jacoby, Neil. Retail Sales Taxation, 1937; cited 24 

Selective bibliography 36* , 

Twentieth Century Fund. Studies in current tax problems, 1937 1,29 

Warhurst, H. P. Effect of the general sales tax levies on retail sales, 

1933-35; 1937; cited 25 

URBAN LAND INSTITUTE. Questionnaire on distribution of taxes 

on real estate 39 

VACANT LOT. Percentage distribution of tax on real property on 

vacant lots, by cities and counties, 1938; table 40 

W. P. A. Urban Housing; Division of Social Research, 1938; cited 15 

o 



.s.,,,,c.g 



JS»«™-^ 912 6