Domestic Production and Foreign Trade; The American Capital Position Re-Examined

Author(s): Wassily Leontief

Source: Proceedings of the American Philosophical Society, Vol. 97, No. 4 (Sep. 28, 1953), pp. 332
-349
Published by: American Philosophical Society
Stable URL: http://www.jstor.org/stable/3149288
Accessed: 11-03-2016 15:11 UTC
REFERENCES
Linked references are available on JSTOR for this article:
http://www.jstor.org/stable/3149288?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact [email protected].

American Philosophical Society is collaborating with JSTOR to digitize, preserve and extend access to Proceedings of the
American Philosophical Society.

http://www.jstor.org

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

DOMESTIC PRODUCTION AND FOREIGN TRADE; THE AMERICAN

CAPITAL POSITION RE-EXAMINED

WASSILY LEONTIEF

Professor of Economics, Director of the Harvard Economic Research Project, Harvard University,

(Read April 24, 1953)

"A fin d'etre absolument clair dans nos recettes, nous n avons pas craint de repeter

plusieurs fois un meme mot dans une phrase. Nos lecteur nous seuront gre d'avoir

evite les recherches de style dans un ouvrage qui n'en comporte pas."

French Cookbook by Tante Marie

I. THE STRUCTURAL BASIS OF

in each of the trading countries, was originally

INTERNATIONAL TRADE
developed in the writings of David Ricardo and

other so-called classical economists of the late

COUNTRIES trade with each other because this

eighteenth and the early nineteenth centuries. It

enables them to participate in and to profit from

still constitutes the basis of the modern theory of

the international division of labor. Not unlike

international trade. The theory of comparative

businesses and individuals, each area specializes in

costs-as many other economic theories-reigns,

those lines of economic activity to which it hap-

however, in the pages of college text books without

pens to be best suited and then trades some of its

actually governing the practice of emipirical ecoown outputs for commodities and services in the

nomic analysis.
production of which other countries have a com-

Until recently, we had so little systematic

parative advantage. The word comparative is in

knowledge of the productive structure of our own

this connection of particular significance.

or of any other national economy that the appliThe United States, for example, exports auto-

cation of such general theoretical principles to the

mobiles and imports newsprint. It does so be-

analysis and explanation of actual foreign trade

cause the quantity of Canadian paper which we

relationships has been practically out of the quescan obtain in exchange for, say, a million dollars'

tion. M-ost of what has been said on that subject

worth of American cars is larger than the addi-

consisted of reasonable common sense conjectures

tional amount of newsprint which we would be

or of plausible examples which-like the automoable to produce at home if we withdrew the capital,

bile and newsprint reference used above-serves

labor, and other resources now absorbed in the

well enough to illustrate the logic of the theoretical

manufacture of one million dollars' worth of auto-

argument, but had hardly any specific base in

mobiles and used it instead to increase the output

detailed facts and figures.2

of our domestic paper industry. Canada, for

A widely shared view on the nature of the trade

analogous but in a sense opposite reasons, finds it

between the United States and the rest of the

advantageous to obtain its automobiles from the

world is derived from what appears to be a comUnited States in exchange for newsprint rather

mon sense assumption that this country has a comthan to divert resources from their present em-

parative advantage in the production of commodi-

ployment in its paper industry into an increased

domestic production of cars.

2 As an example of the recent empirical studies in that

This explanation of the international exchange

field, see Macdougall, G. D. A., British and American

of goods and services in terms of the comparative

export: a study suggested by the theory of comparative

costs, Econ. Jour. 61 (1) : 697-724, 1951; also, Mac-

advantage of the alternative allocation of resources

dougall, G. D. A., British and American exports: a study

1 The study described in this paper constitutes a part

suggested by the theory of comparative costs, Econ. Jour.

62 (2) : 487-522, 1952. A succinct discussion of the theo-

of the systematic analysis of The Structure of the Ameri-

retical problems involved can be found in Samuelson,

can Economy conducted by the Harvard Economic Re-

P. A. International trade and the equalization of factor

search Project. Miss Sue Smulekoff, assisted by Mrs.

prices, Econ. Jour. 58: 163-184, 1948; and International

Nancy Bromberger, has prepared the statistical tables

factor price equalization once again, Econt. Jour. 59: 180-

presented in this paper and performed the numerical com-

197, 1949.

putations underlying these tables.

PROCEEDINGS OF THE AMERICAN PHILOSOPHICAL SOCIETY, VOL. 97, NO. 4, SEPTEMBER, 1953

332

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

VOL. 97, NO. 4, 1953] DOMESTIC PRODUCTION AND FOREIGN TRADE 333

ties which require for their manufacture large

culture, each kind of transportation and distribu-

quantities of capital and relatively small amounts

tion-in short each sector of the American econ-

of labor. Our economic relationships with other

omy-depends upon every other sector. A single

countries are supposed to be based mainly on the

column of an input-output table, shows, for ex-

export of such "capital intensive" goods in ex-

ample, how many steel sheets, steel bars, and other

change for foreign products which-if we were to

steel products automobile manufacturers buy from

make them at home-would require little capital

the steel industry for every million dollars' worth

but large quantities of American labor. Since the

of cars they produce; it also shows how many

United States possesses a relatively large amount

yards (or dollars' worth) they need of upholstery

of capital-so goes this oft repeated argument-

material, how much paint from the chemical indus-

and a comparatively small amount of labor, direct

try and so on. Similarly, the "steel industry col-

domestic production of such "labor intensive"

umn" of the same table describes the various kinds

products would be uneconomical; we can much

of inputs, such as, coal, ore, etc., which the steel

more advantageously obtain them from abroad in

industry must obtain from the other sectors of the

exchange for our capital intensive products.

economy in order to produce an additional million

Recent progress in the collection and systematic

dollars' worth of its own output which, of course,

organization of detailed quantitative information

consists of various steel products. The table con-

on the structure of all the various branches of the

tains as many columns as there are separate indus-

American economy, accompanied by a parallel

tries so that it presents each link connecting any

advance in the technique of large scale numerical

two sections of the economy.

computation, now enables us to narrow the frus-

On the basis of the statistical information con-

trating gap between theory and observation.3

tained in an input-output table one can determiine

This is the first preliminary progress report on

the effect of any given increase or decrease in the

a study designed to analyze the structural basis of

level of output in any one sector of the economy

trade relationships between the United States and

upon the rate of production in all the other sectors.

the rest of the world.

Using the 1947 input-output structure of the

American economy as the basis of such computa-

II. DIRECT AND INDIRECT INPUT

tions, one finds that to produce an additional mil-

REQUIREMENTS

lion dollars' worth of automobiles the output of

None of the basic factual information used here

had to be collected especially for this particular

steel would have to increase by 235 thousand dol-

lars, the output of chemicals by 58 thousand dol-

inquiry. Both the statistical data and the analyti-

lars, while raising the production of non-ferrous

cal procedure employed constitute an integral part

metals by 79, of textiles by 39 thousand dollars

of the so-called input-output or inter-industry re-

and so on. Even the communication services-

search program jointly conducted by various agen-

telephone and telegraph-would have to contribute

cies of the government and private institutions, of

indirectly to the production of a miillion dollars'

which the Harvard Economic Research Proj ect

worth of additional automobiles.

is one.

Column 2 in our table 1 shows the result of this

The factual information referred to above com-

particular computation. Without entering into

prises many sets of figures of which the largest and

the discussion of technical details it may be suf-

in a sense the most important is organized in terms

ficient to observe that the miagnitude of every one

of a so-called input-output table.4 This table de-

of the entries depends upon all the input-output

scribes the actual flow of commodities and services

among all the different parts of the American

economy. Specifically, it shows how each one of

relationships among all the sectors of the economy,

and that the computation of each one of these fig-

ures is equivalent to the solution of a system of as

our manufacturing industries, each branch of agrimany simultaneous equations as there are dis-

tinct sectors in the economy.

3For description of the so-called input-output approach

The more minute the breakdown of industries

to structural economic analysis, see Leontief, Wassily, and

members of the Harvard Economic Research Proj ect,

in the basic input-output table, the more detailed

Studies in the structure of the American Economy, N. Y.,

the final results will be. The following analysis

Oxford Univ. Press, 1952.

is based on a 200 industry breakdown consolidated

4Evans, W. Duane, and Marvin Hoffenberg, The inter-

in some of its stages-for purposes of computation

industry relations study for 1947, Rev. of Economics and

Statistics 34: 97-142, 1952.

and simplified presentation-into fifty sectors (38

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

334 WASSILY LEONTIEF [PROC. AMER. PHIL. SOC.

ITABLE 1

CAPITAL AND LABOR REQUIREMENTS FOR THE FINAL OUTPUT OF ONE MILLION DOLLARS

WORTH OF MOTOR VEHICLES

Requirements per million dollars Requirements per mil!ion dollars

Industrya Output of output of industry of final output of

Industrya requirementsb listed on left motor vehicles

Capital Labor Capital Labor

23456

(Thousands (Thousands (Man years) (Thousands (Man years)

of dollars) of dollars) of dollars)

26. Motor vehicles (145) 1,457.450 565.8 60.340 824.6 87.942

15. Iron and steel 235.14 1,026.3 77.777 241.3 18.288

19. Other fabricated metal products 118.25 713.5 95.335 84.5 11.273

16. Nonferrous metals 78.69 1,001.6 55.715 78.8 4.384

25. Other electrical machinery 75.50 551.1 102.638 41.6 7.749

22. Other non-electric machinery 1 60.70 t 775.7 96.579 47.1 5.862

10. Chemicals s57.95 592.7 49.779 34.3 2.885

12. Rubber products 56.19 493.1 90.172 27.7 5.067

31. Railroad transportation 50.18 3,343.3 153.640 167.8 7.710

11. Products of petroleum and coal 46.85 1,397.2 29.843 65.5 1.398

4. Textile mill products 39.29 493.6 110.563 19.4 4.344

14. Stone, clav and glass products 33.64 1,026.3 128.539 34.5 ! 4.324

8. Paper and allied products 31.95 564.1 64.805 18.0 2.071

34. Trade 31.82 984.9 165.876 31.3 5.278

30. Coal, gas and electric power 29.50 2,222.6 99.318 65.6 2.930

1. Agriculture and fisheries 27.53 2,524.4 82.025 69.5 2.258

21. Metalworking machinery 27.48 1,246.9 130.705 34.3 3.592

33. Other transportation 23.88 928.3 121.576 22.2 I 2.903

9. Printing and publishing 19.72 436.0 114.038 8.6 2.249

38. Business services 18.44 144.5 97.543 2.7 1.799

39. Personal and repair services 18.10 i 681.8 183.503 12.3 3.321

6. Lumber and wood products 15.98 537.9 1 141.540 8.6 2.262

5. Apparel 13.74 262.2 108.795 3.6 1.495

29. Miscellaneous manufacturing 11.26 439.4 1 100.364 4.9 1.130

37. Rental 10.68 8,156.5 16.324 87.1 .174

28. Professional and scientific equipment 10.35 841.8 133.129 8.7 1.378

2. Food and kindred products 9.98 361.9 43.143 3.6 .431

36. Finance and insurance 9.83 28.2 92.242 .3 .907

35. Communications 6.21 4,645.4 1 163.097 28.8 1.013

44. Eating and drinking places 6.02 688.0 I 125.365 4.1 .755

27. Other transportation equipment 5.11 759.0 122.419 3.9 .626

13. Leather and leather products 5.06 264.0 109.629 1.3 .555

23. Motors and generators 4.99 404.3 117.771 2.0 .588

24. Radios 4.65 I 449.0 | 124.097 2.1 .577

7. Furniture and fixtures 4.28 485.1 116.923 2.1 .500

18. Fabricated structural metal products 3.79 441.9 1 83.300 1.7 .316

20. Agriculture, mining and construction machinery! 3.65 838.6 87.794 3.1 .320

17. Plumbing and heating supplies 2.67 509.9 99.388 1.4 .265

40. Medical, educational and non-profit org's. 2.05 2,689.5 253.044 5.5 .519

3. Tobacco manufactures .53 557.6 40.539 .3 .021

41. Amusements .10 1,082.9 166.899 .1 .017

Total requirements in all industries per million dollars of final output of motor vehicles 2,104.8 201.476

a See footnote b for table 2.

b The output required from each industry in order to produce one million dollars' worth of motor vehicles for export

or domestic consumption. See Evans and Hoffenberg, "The Interindustry Relations Study for 1947," The Review of

Economics and Statistics 34: Table 6, 1952.

c This figure includes the "back feed" within this industry, i.e., the automotive industry's purchases from itself,

as well as the million dollars' worth of motor vehicles going to final consumers and the amounts needed by the various

other industries to meet their output requirements. For detailed explanation of the technical point involved, see,

Evans, W. Duane and Marvin Hoffenberg, loc. cit., 137 and 140.

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

VOL. 97, NO. 4, 1953] DOMESTIC PRODUCTION AND FOREIGN TRADE 335

of which trade their products directly on the inter-

tors of the economy. Similar computations have

national market).

been performed for each category of commodities

and services which we export or import (in com-

III. CAPITAL AND LABOR INPUTS

petition with domestic output).

Labor is the other primary factor, the availability

The second and the third sets of our statistical

of which must obviously have a decisive role in

data (columns 3 and 4, table 1) show the direct

establishing the pattern of specialization which decapital and labor requirements of each industry.

termines the composition of our foreign trade.

These figures are based on detailed information

Not unlike capital, the man years which go into

which tells us, for example, that to produce an

the production of, say, one million dollars' worth

additional million dollars' worth of finished cars,

of automobiles are partly absorbed by the automoour automobile industry would have to invest in

bile industry itself but are partly employed also by

175 thousand dollars' worth of new buildings, 266

thousand of additional machinery and many other

fixed items. It also would have to increase its

inventories of raw materials and "goods in proc-

all the other sectors of the economy. The compti-

tation of such direct and indirect labor require-

ments is quite analogous to the computation of the

ess" by 124 thousand dollars. All together this

direct and indirect demand for capital (see col-

adds up to 566 thousand dollars which represent

umns 3 and 5, table 1 ).

the total additional capital (in 1947 prices) which

would have to be invested in the American auto-

mobile industry if its capacity were raised so as

The summary of total quantities of capital and

labor required for domestic production of each

of the many types of commodities exported and

to enable us to produce an additional million

imported by the United States is entered in col-

dollars' worth of cars per year.

umns 2 and 3, table 2. In this table most of the

But this is only one part of the total additional

38 large industry and commodity groups are

capital which would have to be invested in the

broken down into their components, described in

American economy in order to enable it to pro-

terms of the more detailed 200 industry input-

duce-say, for export purposes-these additional

output classification.

automobiles. As we saw before, the input of steel

The figures entered in columns 2 and 3 were

into the automobile industry will have to increase

actually arrived at in two steps. First the indirect

by 235 thousand dollars and the input of textile by

capital and labor requirements generated by one

39 thousand. This, of course, means additional

million dollars' worth of demand for the product

investment in both the steel and textile industries.

of each of the composite 38 sectors were computed.

The magnitude of each of these capital require-

This computation (essentially a solution of corre-

ments can be computed. To do so one must sim-

ply multiply the amount of capital which each of

these two industries requires per million dollars

of its capacity by the additional demand for its

sponding systems of linear equations) was per-

formed in terms of the consolidated 50 industry

input-output table. Next, the total capital and

labor requirements respectively of each particular

product indirectly generated by the million dollar

commodity type within the sector were obtained

rise in automobile output. The amounts of addi-

by adding its specific direct requirements to the

tional capital which each one of the various sectors

previously computed (in a sense average) indirect

of the economy would need in order to enable the

requirements of the consolidated sector as a whole.

United States to increase its automobile export by

Thus, the differences between the total capital and

one million dollars are listed in column 5 of

labor requirements of the industrial products be-

table 1. These add up to 2,105 thousand dollars

longing to the same consolidated sector are due

which is the total amount of capital which the

entirely to the difference in their direct require-

United States economy of 1947 had to invest for

ments, since their indirect requirements are as-

every million dollars' worth of cars produced for

export or final domestic use.

Like the top of an iceberg, visible above the

surface of the water, the part invested in the auto-

mobile industry itself constitutes only a small por-

tion of the total-26 per cent to be exact; the rest

is distributed among the other 42 productive sec-

sumed to be the same.

The main reason for such a two-stage procedure

is economic. If based throughout on the 200 x

200 input-output table the computation of direct

and indirect requirements would cost a thousand

dollars more. The errors caused by the short cut

are not likely to be of decisive importance since

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

336 WASSILY LEONTIEF I PROC. AMiER. P'HIL_ SOC.

TABLE 2

CAPITAL AND LABOR REQUIREMENTS PER MILLION DoLLARS OF V' S. ExPORTS AND IMPORT REPLACEMENTS a
1947

Direct and indirect Requirements per million dollars of

requirements per exports and import replacements Comparison

fiina doutputo Exports Imports of average (1947) composition of export

fina ouput per per and______ ___ r imprtmillion million rnenuire
Industryb ~~~~~~~~ ~~~dollars dollars Capital Labor

of total of total

Capital~ Labo exports~ importsf ___ ____

Exports Import EprsImport.Cp Lab.

replace. Ep rts lacap

1 ~ ~ ~~ ~ ~ ~ ~~2 3 4 5 6 7 8 9 lo 11

(Millions (Man- (Man (Mani

of dollars) years) (Dollars) (Dollars) (Dollars) (Dollars) years) years)

All industries 1,000,000 1,000,000 2,550,780 3,091,339 182.313 170.004 < >

1. Agriculture and fisheries (1-10a) 4.7 120 158.7 10 100,987 257,526 475,851 1,213,463 16.028 40.872

2. Food and kindred products 105,701 98,045 3119,593h 3,349,589 159.847 183.508 < <
Meat packing and poultry (2 1) 3.0158 149.032 17,568 7,189 52,982 21,681 2.618 1.071
Processed dairy products (22) 3.1334 165,081 15,217 2,429 47,681 7,611 2.512 .401
Canning, preserving and freezing (23) 3.2287 206.505 11,446 48,043 36,956 155,116 2.364 9.92 1
Grain mill products (24) 3.0375 146.37 1 45,928 1,522 139,506 4,623 6.723 .223

Bakery products (25) 3.2447 221.331 468 32 1,519 104 .104 .007
Miscellaneous food products (26) 3.2610 175.271 10,553 8,825 34,413 28,778 1.850 1.547
Sugar (2 7) 4.1953 148.850 1,997 12,970 8,378 54,413 .297 1.931

Alcoholic beverages (28) 3.2923 169.712 2,524 17,035 8,310 56,084 .428 2.891

3. Tobacco manufactures (29) 3.2887 173.472 13,245 21,439 43,559 70,506 2.298 3.719

4. Textile mill products 56,810 23,657 2,308,032 2,327,539 213.202 206.662 < >
Spinning, weaving and dyeing (30) 2.3114 215.250 53,758 9,796 124,256 22,643 11.571 2.109
Special textile products (31) 2.3420 201.558 684 8,922 1,602 20,895 .138 1.798
jute, linen, cordage and twine (32) 2.3412 200.639 815 4,728 1,908 11,069 .164 .949

Floor coverings (35a) 2.1591 154.206 1,553 1 211 3,353 456 .239 .033

5. Apparel ~~~~~~~~~~~~~~~~~21,129 36,029 1,661,527 2,213,875 233.802 207.139 <

Canvas products (33) 1.6106 237.848 174 0 280 0 .041 0

Apparel except furs (34) 1.6050 250.169 15,493 12,630 24,866 20,2 71 3.876 3.160
House furnishings, etc. (35b) 1.6492 188.151 4,479 1,814 7,387 2,992 .843 I .341

Furs (hunting and trapping) (lOb) 2.6176 183.571 983 21,585 2,5 73 56,501 .180 3.962

6. Lumber and wood products 10,223 31,787 1,560,785 1,617,"910 242.003 231.636 < >
Logging (36) 1.6383 188.365 378 9,149 619 14,989 .071 1.723
Sawmills, planing and veneer mills (37) 1.6383 251.604 7,153 20,435 11,719 33,479 1.800 5.142

Plywood (38) 1.3366 209.125 863 761 1,154 1.017 . 180 .159
Fabricated wood products (39) 1.3465 226.188 1,217 632 1,639 851 . 2 75 .143

Wood containers and cooperage (40) 1.3491 242.168 612 810 826 1,093 .148 1 .16

7. Furniture and fixtures (41-43) 1.682 1 233.687 2,075 471 3,490 735 .485 1 .102

aAll figures refer to 1947. Wassily, and members of the Harvard Economic Research Project,
b The thirty-eight composite industries are found in Evans, W. Duane, Studies in the Structure o.f the American Economy, Chapter 6, New York,
and Hloffenberg, Marvin, 'The Interindustry Relations Study for Oxford University Press, 1952.

1947," The Review of Economics and Statistics 34: 97-142, 1952. The d See text, page 335, for the derivation of these figures. The direct
component industries are based on Bureau of Labor Statistics, Division labor requirements (labor coefficients) wvere computed by the Harvard
of Interindustry Economics, Interindustry Relations Study, 1947 Economic Research Project from B. L. S. and census data.

Emergency model classification; 1-25, 1952. In column 1, the numbers e Export figures are based on Bureau of Labor Statistics, Division of
in parentheses correspond to this latter classification. Interindustry Economics, Table I-Interindustry flow of goods and
Some of the 200-order industries were split in the process of aggre- services by industry of origin and destination, section 6, October 1952.

gating them into the 50-order classification. These industries are Exports are valued at producers' value: transportation, insurance and
indicated by a or b following the 200-order indust~ry number. Their trade margins are charged separately as export items. The total value
composition in terms of the Standard Industrial Classification is as of exports in 1947 was $16,678.4 million; the actual value of the exports
follows: of each industry can be obtained by multiplying each item in column 4
200-order industry SIC No. by $16,678.4.

10a Fisheries 091 f ~~~~~~Import figures are based on Bureau of Labor Statistics, op.cit.

lOa Funisheries apin 0941 All import figures refer to competitive imports only. Imports are
lOb Hunting and trapping 0741 ~valued at domestic port value, i.e., foreign port value plus transpor-

35a Floor coverings 2274, 2295 tation, insurance, etc., plus duties. The total value of competitive

35b House furnishings, etc. 2391-2399 imports in 1947 was $6,175.7 million; column S times $6,175.7 gives the
lOOb FBriaed ho pipect 3592 actual value of each type of competitive import.

100b Fabricated pipe 3592 g ~~The sign " > indicates that the export requirement exceeds the

1 12a Tractors 3521 corresponding requirements for import replacement; " <' shows the

112b Industrial trucks 3565 opposite. The signs "?> and "< " mark differences amounting to less
135a Electrical appliances 3621 than 2 per cent of the larger of t-he two italicized figures.

135b Heating appliances 3581, 3583, li For the meaning of the italicized figures, see page 347 in text.
3584, 3589 i These two industries are numbered 38 and 41, respectively, in

186a Radio broadcasting 771 Table 1. They are numbered consecutively here because the inter186b Advertising 731 vening industries do not directly participate in international trade.

cThe derivation of these figures is given in the text (see page 335). i Both the capital and labor coefficients for "Other nonferrous min-

The basic data on the direct capital requirements (capital coefficients) ing" (15) must be considered unreliable (too high) since they were
of individual industries were computed by the Harvard Economic Re- based on output statistics which probably did not include operations
search Project. For a general description of methods, see Leontief, performed under the authority of the Atomic Energy, Commissio:i.

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

VOL. 97, NO. 4, 19531 DOMESTIC PRODUJCTION AND FOREIGN TRADE 337

TABLE 2-Continued

Direct and indirect Requirements per million dollars of

requirements per exports and import replacements Comparison

filiona doutputo Exports Imports of average (1947) composition of export

finaloutput per per ~~~~~~~~and imports

IndustrVb million million require

dollars dollarsmng

of total of total Capital Labor

Capitale Labord exportse importsf

Eprs Import Fxport import. a.Lb

replace. replace. P.Lb

1 2 3 4 5 6 7 8 9 10 I1I

(Millions (Man (Man (Man

of dollars) years) (Dollars) (Dollars) (Dollars) (Dollars) years) years)

8. Paper and allied products 9,743 103,616 1,726,891 1,859,722 165.764 161.346 < >

Pulp mills (44) 1.8611 152.803 1,337 42,732 2,488 79,529 .204 6.530

Paper and paper board mills (45) 1.8611 167.325 4,401 60,447 8,191 112,498 .736 10.114

Converted paper products (46) 1.5346 169.389 4,005 437 6,146 671 .678 .074

9. Printing and publishing (47) 1.32 16 196.597 4,329 1,425 5,721 1,883 .851 .280

10. Chemicals 49,153 105,398 2,337,851 2,390,120 167.681 147.602 < >

Industrial inorganic chemicals (48) 2.2968 171.293 7,693 9,748 17,669 22,389 1.318 1.670

Industrial organic chemicals (49) 2.8055 161.081 7,303 4,340 20,489 12,176 1.176 .699

Plastic materials (50) 2.5614 159.740 3,082 97 7,894 249 .492 .015

Rubber (5 1) 2.5208 141.238 342 55,751 862 140,537 .048 7.874

Synthetic fiber (52) 2.9200 212.841 1,739 2,720 5,078 7,942 .370 .579

Explosi'ves (53) 2.2814 197.963 342 0 780 0 .068 0

Drugs and medicines (54) 2.1666 184.150 9,329 1,457 20,212 3,157 1.7 18 .268

Soap and related products (55) 2.1417 146.365 2,524 405 5,406 867 .369 .059

Paints and allied products (56) 2.0430 152.411 3,663 340 7,484 695 .558 .052

Gum and wood chemicals (57) 2.4267 184.907 2,140 3,854 5,193 9,353 .396 .713

Fertilizers (58) 2.3700 180.631 450 356 1,067 844 .081 .064

Vegetable oils (59) 2.007 1 128.889 2,734 20,063 5,487 40,268 .352 2.586

Animal oils (60) 2.0062 136.738 1,079 2,672 2,165 5,361 .148 .365

Miscellaneous chemical industries (61) 2.2467 170.497 6,733 3,595 15,127 8,077 1.148 .613

lla. Crude petroleum and natural gas (17) 3.2118 108.844 6,248 37,372 20,067 120,031 .680 4.068

1lib. Products of petroleum and coal 34,566 21,730 2,600,946 2,674,929 94.110 93.465 < >

Petroleum products (62) 2.5514 94.011 32,881 19,658 83,893 50,155 3.091 1.848

Coke and products (63) 3.8708 87.760 1,355 2,040 5,245 7,896 .119 .179

Paving and roofing materials (64) 2.3237 13 1.557 330 32 767 74 .043 .004

12. Rubber products 10,199 389 1,817,051 1,801,799 194.823 205.656 > <

Tires and inner tubes (65) 1.8305 185.087 6,044 49 11,064 90 1.119 .009

Miscellaneous rubber products (66) 1.7975 208.989 4,155 340 7,469 611 .868 .071

13. Leather and leather products 5,054 5,974 1,667,016 1,668,681 233.874 227.151 < >

Tanning and finishing (67) 1.6900 183.095 1,901 2,817 3,213 4,761 .348 .516

Other leather products (68) 1.6395 271.302 749 1,360 1,228 2,230 .203 .369

Nonrubber footwear (69) 1.6574 262.612 2,404 1,797 3,984 2,978 .631 .472

14. Stone, clay and glass products 12,788 27,560 1,961,425 2,345,091 192.211 177.794 < >

Stone, sand, clay and abrasives (18) 2.582 1 226.822 330 3,854 852 9,951 .075 .874

Sulphur (19) 2.5821 139.703 1,385 0 3,576 0 .193 0

Other nonmetallic minerals (20) 2.582 1 154.790 881 17,456 2,275 45,073 .136 2.702

Glass (70) 1.9293 199.932 4,419 1,295 8,526 2,498 .883 .259

Cement (7 1) 2.4944 167.940 1,043 0 2,602 0 .175 0

Structural clay products (72) 1.7718 271.334 959 49 1,699 87 .260 .013

Pottery and related products (73) 1.3682 261.934 929 2,477 1,271 3,389 .243 .649

Concrete and plaster products (74) 1.6727 205.466 246 65 412 109 .051 .013

Abrasive products (75) 1.4890 159.882 1,127 1,765 1,678 2,628 .180 .282

Asbestos products (76) 1.4890 176.167 600 32 893 48 .106 .006

Other miscellaneous nonmetallic minerals (77) 1.4948 179.324 869 567 1,299 848 .156 .102

15a. Iron ore mining (1 1) 3.1683 2 12.434 552 7,675 1,749 24,317 .117 1.630

15b. Iron and steel 37,732 4,695 2,724,880- 2,655,654 181.305 151.438 > >

Blast furnaces (78) 2.6394 142.525 396 3,676 1,045 9,702 .056 .524

Steel works and rolling mills (79) 2.7599 180.703 35,585 955 98,211 2,636 6.430 .173

Iron foundries (80) 2.0344 232.540 672 32 1,367 65 .156 .007

Steel foundries (81) 2.0349 236.564 90 16 183 33 .021 .004

Iron and steel forgings (92) 2.03 11 179.672 989 16 2,009 33 .178 .003

16a. Nonferrous metal mining 468 47,154 4,402,991 4,372.254 286.325 281.885 > >

Copper mining (12) 3.2280 197.862 0 5,263 0 16,989 0 1.041

Lead and zinc mining (13) 2.6210 230.618 12 5,360 32 14,049 .003 1.236

Bauxite mining (14) 2.6948 22 1.395 114 3,757 307 10,124 .025 .832

Other nonferrous mining (15)i 5.0347 310.689 342 32,774 1,722 165,007 .106 10.183

16b. Processing nonferrous metals 9,516 57,759 2,402,427 2,445,386 149.222 127.461 < >

Primary copper (82) 2.4334 12 1.184 2,788 22,216 6,784 54,060 .338 2.692

Copper rolling and drawing (83) 2.4348 155.831 1,565 49 3,811 119 .244 .008

Primary lead (84) 2.4340 120.806 30 6,720 73 16,357 .004 .812

Primary zinc (85) 2.4350 166.224 1,379 2,672 3,358 6,506 .229 .444

Primary metals, n.e.c. (86) 2.4348 131.553 396 18,913 964 46,049 .052 2.488

Nonferrous metal rolling, n.e.c. (87) 2.4349 148.977 983 16 2,394 39 .146 .002

Primary aluminum (88) 3.2849 144.156 204 761 670 2,500 .029 .110

Aluminum rolling qnd drawing (89) 2.1816 177.628 1,769 0 3,859 0 .314 0

Secondary nonferrous metals (90) 2.4355 125.398 282 6,396 687 15,578 .035 .802

Nonferrous foundries (91) 2.182 1 -244.406 120 16 262 35S .029 .004

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

338 WASSILY LEONTIEF IPROC. AMER. PHIL. SOC.

TABLE 2-Continued

Direct and indirectReurmnsprilondlasf

requirements per Rexportseand iport repinolaceens of Comparisoni

million dollars ofexotanimotrpamns ofxot

final output Exports Imports of average (1947) composition ofd exports

per per __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ require-

Industryb million million mentsg

dollars dollars CailLbo

of total of total CailLbo

Capital Labor'exportse importsf __

______________________ ____ I ~~~~~~~~~ ~~~Exports Exports~ Cap. Lab.

1 2 3 4 5 6 7 8 0 10 1t

(Millions (Man (Man (Man

of dollars) years) (Dollars) (Dollars) (Dollars) (Dollars) years) years)

17. Plumbing and heating supplies 3,202 49 2,048,157 2,046,700 211.118 204.647 > >
Metal Plumbing and vitreous fixtures (97) 2.0510 223.913I 1,085 0 2,225 0 .243 0

Heating equipment (98) 2.0467 204.647 ,11 49 4,333 100 .433 .010

18. Fabricated structural metal products 4,053 179 1,748,187 1,796,648 182.087 178.771 < >
Structural metal products (99) 1.6954 183.767 2,518 49 4,269 83 .463 .009
Boiler shop products (100a) 1.8348 178.945 1,535 130 2,816 239 .275 .023

19. Other fabricated metal products ~~~~~16,531 1,262 2,011,342 1,971,712 203.738 207.607 > _

Tin cans and other tinware (93) 2.1458 174.998 791 32 1,697 69 .138 .006
Cutlery (94) 2.0414 241.579 1,229 178 2,509 363 .297 .043
Tools and general hardware (95) 2.042 1 227.946 3,130 259 6,392 529 .713 .059
Hardware, n.e.c. (96) 2.0459 228.406 1,811 16 3,705 33 .414 .004

Metal stampings (101) 1.8530 202.075 2,075 453 3,845 839 .419 .092
Metal coating and engraving (102) 2.0457 264.165 0 0 0 0 0 0
Lighting fixtures (103) 2.0419 195.244 2,140 16 4,370 33 .418 .003
Fabricated wire products (104) 2.0401 169.167 3,286 49 6,704 100 .556 .008
Metal barrels, drums, etc. (105) 2.0397 164.918 486 130 991 265 .080 .021
Tubes and foils (106) 2.0399 206.580 282 32 575 65 .058 .007
Miscellaneous fabricated metal products

(107) (0)2.0406 190.366 258 65 527 133 .049 .012

Steel springs (0)2.0397 172.761 0 0 0 0 0 0

Nuts, bolts and screw machiine products (109) 1.8550 2 16.333 1,043 32 1,935 59 .226 .007

20. Agriculture, mining and construction machinery 34,518 5,667 2,083,252 2,115,952 193.059 202.400 < <
Tractors (I112a) 2.1098 185.783 11,722 1,457 24,731 3,074 2.178 .271
Farm equipment (113) 2.1183 208.218 5,504 4,194 11,659 8,884 1.146 .873

Construction and mining machinery (1 14) 2.0541 188.271 12,081 1 6 24,816 33 2.2 75 .003
Oil field machinery and tools (1 15) 2.0541 204.419 5,211 0 10,704 0 1.065 0

2 1. Metal working machinery (1 16-117) 2.1793 212.211 12,633 227 27,531 495 2.681 .048

22. Other non-electric machinery I58,836 3,238 1,901,679 1,978,413 195.442 192.7120 < >
Fabricated pipe (100b) 1.6724 176.071 0 0 0 0 0 0

Steam engines and turbines (1 10) 1.6334 234.085 1,409 1 6 2,302 26 .330 .004
Internal combustion engines (111) 1.6334 183.850 6,212 389 10,147 635 1.142 .072
Industrial trucks (I112b) 1.8509 175.047 851 0 1,575 0 .149 0
Special industrial machinery (1 18) 2.1146 202.576 19,684 1,943 41,624 4,109 3.988 .394
Pumps and compressors (1 19) 1.8797 179.349 4,335 0 8,149 0 .7 77 I 0
Elevators and conveyors (120) 1.8754 181.040 1 2,452 0 4.599 0 .444 0
Blowers and fans (1 21) 1.8744 182.857 396 01 742 0 .072 0
Power transmission equipment (122) 1.8749 204.820 162 0 304 0 .033 0
Industrial machinery, n.e.c. (123) 1.8748 170.428 2,494 648 4.676 1,215 .425 .110
Commercial machines and equipment, n.e.c.

(124) 1.8185 224.616 7,051 32 12,822 58 1.584 .007

Refrigeration equipment (125) 1.6074 169.170 6,697 0 10,765 0 1.133 0
Valves and fittings (126) 2.2257 2 11.626 2,782 0 6,192 0 .589
Ball and roller bearings (12 7) 2.2110 233.258 1,457 32 3,22 1 71' .340 1 .0

Machine shops (128) 1 2.2 131 2 12.2 77 156 () 345 0 .0331 ol

Electrical appliances (135a) 1.6404 170.386 2,698 178 4,426 292 .460 .030

23. Motors and generators (131) 1 1.3747 '202.568 4,383 97 6,025 133 .888 .020

24. Radios and related products (139) 1.5768 1249.783 6,763 130 10,664 205 1.689 .032

25. Other electrical machinery I15,794 193 1,767,716 1,771,503 1218.121 202.073 <K
Wiring devices and graphite products (129) 11.7708 200.531 1 1,745 16 3,090 28 .350 .003
Measuring instruments (130) 1.7748 224.339 714 i 16 1,267 28 .160 .0
Transformers (132) 1.7713 204.589 971 0i 1,720 0 .199 0
Control apparatus (133) 1.7731 2 14.419 1,679 0 2,977 0 .360 0
Welding apparatus (134) 1.77 17 1183.887 1,289 49 2.284 87 I .237 .009o

Heating appliances (135b) 1.7181 179.511 1,163 0 1,998 0 .209 0

Insulated wire and cable (136) I1.7663 172.350 1,457 16 2,574 I 28 .251 .0

Engine electrical equipment (137) 1.7690 297.422 971 0I 1,718 0 .289 0

Electric lamps (138) 1.7678 226.812 726 32 1,283 571 .165 .007
Tubes (140) 1.7763 297.568 947 0 1,682 0 I .282 0
Communication equipment (141) 1.7744 231.621 2,147 32 3,810 57 1 .497 i .007
Storage batteries (142) 1.7695 154.3 18 576 16 1.019 28' .089 .002

Primary batteries (143) 1.7697 209.119 486 0 860 0 .102 0

X-ray apparatus (144) 1.7742 276.505 923 16~ 1,6381 28 .255 .0041

26. Motor vehicles 61,151 1,085 2,104,799 2,104,799 201.779 201.476>

Motor vehicles (145) 2.1048 201.476 59,892 1,085 126,061 2,284 12.067 1.219
Truck trailers (146) 2.1048 216.227 1,259 0 I 2,650 0 .2 72 0

Automobile trailers (147) 2.1048 210.641 0 0 0 0 0 0

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

VOL. 97, NO. 4, 1953] DOMESTIC PRODUCTION AND FOREIGN TRADE 339

TABLE 2-Continued

Direct and indirect Requirements per million dollars of

requirements per exports and import replacements Comparison

million dollars ofofaeae197copstn ofxor

final output Exports Imports of average (1947) composition of imports

per per anduimprts

Industryb million million require

dollars dollars Capital Labor mentsg

of total of total CailLbo

Capitalc Labord exportse importsf |

Exports Import Exports Import. Cap. Lab.

replace. replace.

1 2 3 4 5 6 7 8 9 10 11

(Millions (Man (Man (Man

of dollars) years) (Dollars) (Dollars) (Dollars) (Dollars) years) years)

27a. Other transportation equipment 20,236 1,247 1,678,459 1,528,148 189.761 169.206 > >

Aircraft and parts (148) 1.7328 235.024 7,525 130 13,039 225 1.769 .031

Locomotives (150) 1.6663 170.126 4,731 16 7,883 27 .805 .003

Railroad equipment (151) 1.6663 158.126 6,433 0 10,719 0 1.017 0

Motorcycles and bicycles (152) 1.5019 161.216 1,547 1,101 2,323 1,654 .249 .177

27b. Ships and boats (149) 2.1404 263.615 5,360 810 11,473 1,734 1.413 .214

28a. Professional and scientific equipment 6,566 11,529 1,844,913 1,840,559 251.904 238.442 > >

Scientific instruments (153) 1.8465 266.625 3,748 65 6,921 120 .999 .017

Medical and dental instruments and supplies

(155) 1.8437 229.939 2,039 97 3,759 179 .469 .022

Watches and clocks (156) 1.8405 238.387 779 11,367 1,434 20,921 .186 2.710

28b. Optical, ophthalmic and photo equipment

(154) 1.8465 311.213 4,707 680 8,692 1,256 1.465 .212

29. Miscellaneous manufacturing (157-163) 1.4382 186.429 10,762 23,771 15,478 34,188 2.006 4.432

30. Coal, gas and electric power 22,083 1,133 1,790,214 3,702,030 209.573 136.805 < >

Coal mining (16) 1.7821 209.883 22,011 259 39,226 462 4.620 .054

Electric light and power (167) 4.2709 115.066 72 874 308 3,733 .008 .101

Natural, manufactured and mixed gas (168) 2.2676 97.194 0 0 0 a 0 0

31. Railroad transportation (169) 3.9285 186.879 40,957 0 160,900 0 7.654 0

32. Ocean transportation (172) 2.6324 165.090 80,361 40,157 211,542 105,709 13.267 6.630

33. Other transportation 20,068 2,364 2,007,843 2,151,946 165.238 150.592 < >

Trucking (170) 1.1152 152.922 9,018 0 10,057 0 1.379 0

Warehousing and storage (171) 3.9155 376.255 1,529 0 5,987 0 .575 0

Other water transportation (173) 4.2776 119.141 3,933 696 16,824 2,977 .469 .083

Air transportation (174) 1.2650 163.866 4,976 1,668 6,295 2,110 .815 .273

Pipeline transportation (1 75) 1.8485 127.555 612 0 1,131 0 .078 0

Local and highway transportation (178) 1.0436 173.106 0 0 0 0 0 0

34. Trade 62,302 0 1,417,208 185.452

Wholesale trade (176) 1.4157 185.346 62,158 0 87,997 0 11.521 0

Retail trade (177) 2.0683 228.730 144 0 298 0 .033 0

35. Communications 2,272 0 5,097,887 246.360

Telephone and telegraph (179) 5.0979 246.360 2,272 0 11,582 0 .560 0

Radio broadcasting (186a) .8310 57.460 0 0 0 0 0 0

36. Banking, finance and insurance (181) .4699 134.774 8,106 16,516 3,809 7,761 1.092 2.226

37. Business servicesi (186b-187) 1.6345 240.990 156 0 255 0 .038 0

38. Amusementsi (190) 2.2801 237.204 7,687 0 17,527 0 1.823 0

the similarity of their structural relationship to IV. COMPUTATION OF EXPORT AND OF

the rest of the econonmy constituted the guiding IMPORT REPLACEMENT COSTS

principle in the aggregation of the individual in- Now we are ready to find out whether it is true

dustries into the larger sectors. What is even that the United States exports commodities the

more important, whatever errors do occur in these domestic production of which absorb relatively

basic computations, can have no biasing effect on large amounts of capital and little labor and im-

the final results of our numerical analysis. The ports foreign goods and services which-if we had

disregard of differences between the indirect capi- produced them at home-would employ a great

tal and labor requirements of industries belonging quantity of indigenous labor but a small amount of

to the same group has, furthermore, a theoretical domestic capital.

reason which will become clear in the course of Let us imagine a situation in which the United.

the later argument. States, for some reason wanted to reduce its de-

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

(I (IOA)

63 oO

167

27

//

/ 88 28 29 26

17 /2

/ ~~~~~~~~~~22

/ ~~~~~~~24 21/

79

t1 1 / ~~~~~~~~49

162 19 50

84 82 90 96 87 83 85 48 57
/64 61 58 3

104 107

/ 125135A 939 Sop

108IO0A I12A~

591 6011B 1 114 103

59 60 45 120 65

123 134 134 129

125 135A 135

67

(157-163

174 34 8,53 B, 124 B, 164 8) 47

Overlay Chart for figure 1, Wassily Leontief, Domestic producti

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

CAPITAL AND LABOR REQUIREMENTS PER MILLION DO

5. X

4.75

4.504

4.25 417

(1) 3.5

0 30

9 2.75-

19

LI

2.25

200

1.75

-or-7 ~ 36

1.50

1.590 100 120 130 140 150 160 170 180 190 2

LABOR (MANs

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

CAPITAL AND LABOR REQUIREMENTS PER MILLION DO

5.

4.75

Q. I.. OA)

4.4.00

63

42.75 1 X L

80 U7

3. /

0~~~

o I ~~~~~~~~~~~~~~~22

IL 2 21

o 3.0

( ) 79

10

1 ~ ~ ~~ 8 3 8 48 al5

< 61 ~~~~~~~~~~~~~~~~~~~~~~5311

!2.256468

a.. 104 07

< ~~~~~~~~~~~A9

o I~~~~~~~~~~~~~~~~~~~~08~ 1

2.00 60 45 jL~~6

(f)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I

1.75 12 125 135 12

151 ii 36

1.50 75

0(157 -16%1

I ~~~ 34 B ,5 Bs, 124 B 64 B)7

.290 100 20 130 140 Ovr5hr f'6i01 lo* 18 0 190

Over ayhart foLure 1bomestic d t

0 R ""MA~Fi

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

! 23 25

2'

52 ,

3 13

116-1 17) Q

14815 127149

4106

/ ~~~~~~~~91

go ,,,.-147 146 97 153

I."s0 121$ 124 155 9110

1221t6 124 IfS{141 156 102 72

66 09 130 14815

143 133 13814 1013

74 138(41-43) 37

110 33 34A 69

'39

73

131 38 39 40

ion and foreign trade, Proc. Amer. Philos. Soc. 97: (4): 340-341, 1953.

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

ILLARS OF U.S. EXPORTS AND IMPORT REPLACEMENTS

LEGEND

EXPLANATION OF PLOT OF

TOBACCO MANUFACTURES (29)

LOGARITHMIC SCALE

10000/ 13,245 EXPORTS

iO,OOOt --21X439 IMPORTS

173.5 LABOR (MAN YEARS) - X COORDINATE

3.29 CAPITAL (MILLIONS) - Y COORDINATE

too 210 220 230 240 250 260 270 280 300 310

1 YEARS)

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

lLLARS OF U.S. EXPORTS AND IMPORT REPLACEMENTS

LEGEND

EXPLANATION OF PLOT OF

TOBACCO MANUFACTURES (29)

LOGARITHMIC SCALE

00, OO 13,245 EXPORTS

IO,00121 439 IMPORTS

1,0001

5 }fi t - l 173.5 LABOR (MAN YEARS) - X COORDINATE

3 ~~~~~~~~3.29 CAPITAL (MILLIONS) - Y COORDINATE

31 I

16 W149

115 1 27

1153

6~~~~~~~~~ 4~5

9 1%~~~~~~~~

t ae,roc Aer i 09 1

n.ap f td,? 1 mr () 4,

ZI 0,1 2?h9 os. Soc. (4:g34o93 270 280 300 31

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

342 WASSILY LEONTIEF [PROC. AMER. PHIL. SOC.

pendence on foreign countries and, to achieve this

aggregate dollar value of all competitive imports

end, decided to decrease both its imports and ex-

gives us the amounts by which the domestic out-

ports by one million dollars each. Let us, in

puts of these goods and services would have to be

particular, examine the rather plausible case in

increased if our economy proceeded to replace

which the reduction of exports is to be achieved

by an equal proportional cut in each export com-

modity so that after the reduction the percentage

composition of exports remains unchanged. The

same procedure can be applied to so-called com-

petitive imports, i.e., imports of commodities which

can be and are, at least in part, actually produced

by domestic industries. The level of non-competi-

tive imports which, conventionally, are taken to

comprise coffee, tea, jute (but not rubber,.which

can now be commercially synthesized) and a few

other, minor items, is assumed to remain at the

same time unchanged. Such an exemption ob-

comnmodity by commodity an aggregate million

dollars' worth of (proportionally reduced) conm-

petitive imports. Column 5 in table 2 shows the

composition of an average million dollars' worth

of competitive imports. To compute the total

amount of capital which would be required to

produce domestically this particular collection of

commodities, one has only to multiply each of

these figures by the corresponding capital require-

ments listed in column 2 and then find the sum

total of the resulting products. The products-

one for each kind of the competitive imports-are

entered in column 7.

viously has a good common sense basis. More-

over, within the context of the present analysis,

it also has the closely related reason that labor

and capital requirements for the domestic produc-

tion of, say, coffee, cannot be realistically assessed.

An analogous computation yields the corre-

sponding labor requirements. Column 9 shows

the number of American man years which, in

combination with the capital entered in column 7,

would have to be employed to replace the foreign

For later reference, one miight observe that hot

goods and services listed in column 3 with similar

houses and heating installations wouild in any case

g,oods produced domestically.5

require inordinately large capital investment per

- For the purposes of the present analysis, we were able

million dollars' worth of competitively produced

to utilize the previously completed computation which

Florida or California coffee.

shows the effects of any given change in "final demand"

To replace a million dollars' worth of imports

on the levels of output of all American industries. (See

we would have to raise the output of the corre-

Evans and Hoffenberg, ibid.) The results of these origi-

nal computations must, however, be subjected to a quansponding United States industries. If competitive

titatively not very significant but in principle very impor-

imports were, as has been assumed, cut propor-

tant adj ustment.

tionally all along the line, the domestic production

Common sense reasoning as well as actual experience

of the specific goods involved would have to exslhows that whenever any one of the American industries

pand by the amounts equal to the reduction in the

expands or contracts, the level of its operation tends to

increase (or to decrease) its demand for imported inputs

corresponding imports, i.e. by the same propor-

in a way analogous to the increases (or decreases) in its

tional amounts. If, for example, newsprint con-

requirements for materials and services of domestic origin.

stituted twenty per cent of all competitive imports,

An increase in the rate of our domestic outputs will,

and woolens ten, then in replacing the total of one

therefore, in general, lead to a rise in the volume of the

dependetnt imports. The usual input-output computations

million dollars' worth of competitive imports, the

thus present the United States' imports as depending on

domestic output of newsprint would have to be

the level of final demand which, in particular, itnplies

increased by two hundred thousand dollars and

that any rise in exports would necessarily require an

the production of woolens by one hundred thoti-

increase in imports.

For the purposes of the present analysis, this conclusion

sand dollars.

should certainly be retained in respect to inputs which are

Such domestic production for replacing imports

uinlikely to be replaced by a supply coming from domestic

would mean additional direct and indirect capital

sources. Coffee. jute, tin, and a number of other raw

and labor requirements. These can be determined

in the following way.

materials canl be safely included in this "non-competitive"

category. In evaluating the effect of increased exports oii

domestic capital requirements, it seems to be reasonable

The large 200 industries input-output table of

to assume that whatever additional indirect demand for

the American economy for the year 1947 shows

the above type of goods will arise, it will be satisfied by

the competitive imports for that year classified by

foreigni sources. In other words, in contemplating any

possible changes in the level and the composition of our

the commodity groups into which they would fall

exports and imports-as they would result from alterna-

if they had been produced by our domestic industive patterns of American foreign economic policy-it is

tries. Dividing each one of these figures by the

reasonable to assume that the volume of such ntoni-

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

VOL. 97, NO. 4, 1953] DOMESTIC PRODUCTION AND FOREIGN TRADE 343

The quantities of capital and of labor absorbed

These figures show that an average million dol-

by the American economy per million dollars of

lars' worth of our exports embodies considerably

its 1947 exports can be determined exactly in the

less capital and somewhat more labor 6 than would

same way. Column 4 in table 2 shows the com-

be required to replace from domestic production an

position of an average million dollars' worth of

equivalent amount of our competitive imports.

the United States' exports. The quantities of

America's participation in the international divi-

capital and labor required to produce the indicated

sion of labor is based on its specialization on labor

amount of each export-obtained by multiplying

intensive, rather than capital intensive, lines of

each figure in column 4 by the corresponding fig-

production. In other words, this country resorts

ure in columns 2 and 3-are entered in columns

to foreign trade in order to economize its capital

6 and 8, respectively.

and dispose of its surplus labor, rather than vice

versa. The widely held opinion that-as com-

V. EMPIRICAL FINDINGS AND THEIR

pared with the rest of the world-the United

INTERPRETATION

States' economy is characterized by a relative sur-

The principal findings of the quantitative factual

plus of capital and a relative shortage of labor

analysis described above are summarized in the

proves to be wrong. As a matter of fact, the

following figures:
opposite is true.

What is the explanation of this somewhat unDOMESTIC CAPITAL AND LABOR REQUIREMENTS PER

MILLION DOLLARS OF U. S. EXPORTS AND OF

expected result? The conventional view of the

COMPETITIVE IMPORT REPLACEMENTS

position which the United States occupies today

(OF AVERAGE 1947 COMPOSITION)

in the world economy is based-as has been preImport

Exports Replacements
viously explained-first, on an empirical observa-

Capital (dollars, in 1947 prices) 2,550,780 3,091,339

tion and second, on a factual assumption. The

Labor (man years) 182.313 170.004

observation is that the United States possesses

competitive imports will be in the future as in the past

more productive capital per worker than any other

directly determined by structurally conditioned domestic

country. It can hardly be disputed.

requirements.

To reach the conclusion that this means that

With the typical competitive imports-such as cars,

there exists a comparative surplus of capital and

most other highly manufactured products, and also some

raw materials such as, for example, crude oil-the situa-

6; There exists a good reason to believe that the excess

tion is entirely different. If the problem of comparative

of the labor requirements per million dollars' worth of

costs, i.e., the question of possible alternative patterns of

American exports over the labor requirements for the

trade is to have any meaning in respect to such commodi-

equivalent amount of imports replacing output is actually

ties, one must explicitly consider stepped-up domestic pro-

larger than our computations shows it to be.

duction as being an alternative to imports and vice versa.

Part of the labor input entering in both of these figures

In this context, an increase in final demand' and particu-

consists of agricultural labor. Agricultural employment

larly an increase in export demand should not be assumed

figures are well-known to be biased in the upward directo result in an automatic rise in competitive imports. On

tion partly because many persons living on the farms do

the contrary, the domestic repercussion-for example, the

niot actually work on them and partly because a very large

change in domestic capital and labor requirements-of

portion of agricultural labor input is absorbed, one could

additional exports must first of all be computed on the

nearly say wasted, in marginal subsistence farming.

assumption that whatever virtual demand for competitive

Since the agricultural employment contributes less to,

importation might arise, it will be satisfied entirely and

the labor requirement of our exports than it does to the

only through expansion of domestic output. The possi-

replacement requirements for our competitive imports, any

bility of increasing the imports of such competitive com-

downward revision in that figure would tend to increase

modities has to be considered as a separate alternative.

the difference between these two figures.

The capital saving effects of such imports are explicitly

The labor requirements shown in the summary table

taken into account when one separately postulates the

presented above are split between the agricultural and alt

expected changes in the. level of specific competitive im-

otlher labor as follows:

ports and computes the repercussion of such imports on

domestic capital requirements.

AGRICULTURAL AND NON-AGRICULTURAL LABOR

In a very open economy, such, for example, as the

REQUIREMENTS PER MILLION DOLLARS OF U. S.

British, the difference between the domestic reactions

EXPORTS AND OF COMPETITIVE IMPORT-

computed first on the assumption of an automatically

REPLACEMENT (OF AVERAGE

induced change in the level of competing imports and

1947 COMPOSITION)

then without such induced changes might be quite large;

Import

in the case of the United States-the most self-sufficient

of the modern western economies-such discrepancy will

be quite small. It was still, however, taken into account

in the present study.

Exports Replacements

Agricultural labor (man years) 22.436 40.934

Non-Agricultural (man years) 159.872 129.069

Total 182.308 170.003

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

344 WASSILY LEONTIEF [PROC. AMER. PHIL. SOC.

a scarcity of labor in this country, the conventional

rather than larger, than that of many other

argument must combine the foregoing observation

countries.

with the implicit assumption that the relative pro-

ductivity of capital and labor-if compared indus-

try by industry-is the same here and abroad.

This, I submit, is the analytical explanation of

the results of our empirical findings. In terms

of the relative production possibilities here and

Concretely, this assertion means that if in the

abroad, the United States is rich in man power

United States we can transform ten pounds of

and poor in capital. This country resorts to for-

yarn into a corresponding amount of finished cloth

by using, say, one man year and two thousand

eign trade to save its capital and to dispose of its

relative surplus labor.

dollars' worth of machinery and transform a

barrel of oil into gasoline by using one man year

Our data obviously cannot explain why Ameri-

can labor is more productive than foreign labor.

and twenty thousand dollars' worth of equipment,

The problem of productivity is so intricate and

the corresponding foreign industries can perform

has been so thoroughly discussed elsewhere that

each of these two operations either with exactly

no casual remarks can possibly advance its solu-

identical inputs of capital and labor or-if this is

tion. The following negative observation, how-

not the case-at least with inputs differing in

ever, has a direct bearing on the subject of the

both (and all the other) industries in the same

present analysis and on the possible interpretation

proportion. So, for example, if in India one

of its principal findings.

The extent to which the high relative efficiency

could weave ten pounds of yarn by using two man

years and four thousand dollars' worth of ma-

of American man power causes this country to

chinery (instead of one man year and two thou-

exchange goods which absorb relatively little capi-

sand dollars as in the United States) the cracking

tal for those which would require more capital if

of one barrel of oil could also be accomplished by

we chose to produce them at home, cannot be due

using a double quantity of both factors, i.e., two

man years and forty thousand dollars' worth of

simply to the large amount of capital which Ameri-

can industry uses per employed worker.

The fact that workers are frequently replaced

equipment.

by machines cannot be denied. But such technoO-nly on the basis of such an assumption, will

logical stubstitution, if profitable in United States,

the comparative costs argument necessarily lead to

would in general be profitable also in the correthe conclusion that a country possessing a large

sponding industries abroad. The argument that

stock of capital and a relatively small number of

the comparative shortage of capital might prevent

workers will find it advantageous to specialize in

the use of the same labor-saving technology by

industries which, in terms of its own productive

foreign countries would only hold if international

possibilities, require much capital and relatively

trade, i.e., the international division of labor, did

little labor.

not exist. Actually, it does take place and if it

Let us, however, reject the simple but tenuous

were simply the problem of substituting capital

postulate of comparative technological parity and

for labor, foreign countries could and would imi-

make the plausible alternative assumption that in

tate the American production practice industry by

any combination with a given quantity of capital,

industry. At the same time, their prodtuction

onle man year of American labor is equivalent to,

would be concentrated on those commodities which,

say, three man years of foreign labor. Then, in

comparing the relative amounts of capital and

labor possessed by the United States and the rest

of the world-a comparison used for the explana-

tion of their respective specialization in capital or

both there as well as in the United States, require

relatively little capital and large amounts of labor.

The United States would for similar reasons con-

centrate on capital intensive indtustries and the

trade between it and the rest of the world would

consist in an exchange of American capital intenlabor intensive industries respectively-the total

sive against foreign labor intensive goods.7. Our

number of American workers must be nmltiplied

by three, which would increase our 1947 labor

7 To clarify the internal logic of the argument leading

force from 65 million to three times that number,

to this assertion, let us consider-from the point of view

i.e., 195 million of "equivalent" foreign man years.

of the world as a whole-the double problem of, first,

allocating capital and labor between the various industries

Spread trice as thinly as the unadjusted figures

and, second, of locating the various industries in specific

suggest the American capital supply per "equiva-

countries endowed with different relative amounits of

lent worker" turns out to be comparatively smaller,

capital and labor.

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

VOL. 97, NO. 4, 19531 DOMESTIC PRODUCTION AND FOREIGN TRADE 345

empirical findings indicate that in fact the opposite

often been mentioned in this connection. In ac-

cepting this most plausible explanation, we must,

is true.

Thus, without denying that capital can be sub-

stituted for labor, we must still look for some

other reason in explaining the high productivity

of labor in America as compared with the labor

employed by similar industries abroad.

Entrepreneurship and superior organization have

however, make the following comment. Both

these, as well as such other factors as education

or the general climate of our production oriented

society do certainly make the American economy

more efficient in the sense that it is able to achieve

the same output of finished commodities and serv-

ices with smaller inputs of capital and labor.

If in accordance with the conventional argument, but in

There exists a definite statistical evidence that the

contradiction to the argument presented in this paper, one

considers the technological possibilities to be the same

man hour and the capital investment both meas-

throughout the world, i.e., if one assumes that with a

ured per unit of output have been reduced in many

given amount of capital and a given number of indigenous

of our industries through better utilization of

man years, every industry in England, in India or any-

equipment and more rational use of labor.8 To

where else is able to produce an output equal to that

explain the comparative surplus of labor which

which the corresponding American industry could achieve

with the same amount of capital and an equal number of

our figures unmistakably reveal, we must, how-

(American) man years, that double task can be accomever, also infer that entrepreneurship, superior

plish2d in the following two steps.

organization and favorable environment must have

First, considering the total stock of capital and the

increased-in comparison with other countriescombined supply of labor of all countries and taking in

account the total world demand for various commodities

the productivity of American labor much more

and services, the proverbial "invisible band" of competithan they have raised the efficiency of American

tive adj ustment would determine-on the basis of the

capital.

uniform technological possibilities of the world as a whole

From the point of view of sheer arithmetic, the

-the proper amounts of capital and labor which each

industry would best use per, say, every million dollars'

worth of its respective output. Barring certain special,

American comparative capital shortage and labor

surplus-as revealed in our figures-could, of

unusual situations, this decision could and would be made

course, be equally well explained if instead of

without any regard to the actual distribution of the com-

assuming that American man years are more pro-

bined labor and capital resources of the world between the

ductive than foreign man years we took the labor

different countries. This distribution could be taken into

account separately in the next step in which all the indi-

productivity to be the same here and abroad, but

vidual industries would be actually assigned to the sepat the same time assumed the United States' capital

arate countries. In accordance with the "comparative

to be less productive than its dollar equivalent in

supply of factors" considerations described in the first

foreign countries. Such an alternative explanasection of this paper, this second step will result in placing

the industries requiring relatively large amount of capital

tion, implying an absolute inferiority of the Ameri-

into the countries comparatively well supplied with that

can productive technology, hardly would pass the

particular factor and in locating the labor intensive lines

test of empirical scrutiny; it is plainly contraof production in the areas having a comparatively larger

dicted by the fact an average American man year

supply of labor.

As a final result of such efficient "comparative costs"

receives a much higher remuneration than the man

allocation, the capital rich countries must specialize on

year of labor employed in most other countries.

the production and export of capital intensive goods, while

the labor rich areas will produce and export labor inten-

VI. EMPIRICAL ANALYSIS OF SUBSIDIARY

sive commodities, while importing goods which, when

RELATIONSHIPS
produced at home, would absorb comparatively large

amounts of capital and little labor.

Before directing your attention to the wider

It is particularly important to observe that under the

economic implications of these general concluassumption of technological parity the combination of

capital and labor used in each industry-having been

sions, it is well to examine once more their em-

decided in the first stage of the two stage allocation propirical background.

cedure described above-will necessarily be the same in

Although computed on the basis of a rather

all the countries. For example, any specific textile prod-

detailed industrial classification, the amounts of

uct requiring much capital and little labor when made in

the United States would require the same combination of

capital and labor used in the production of Ameri-

these two factors also, if it had been produced in Eng-

land, in India or in any other country. Being short of

capital, i.e., of the factor which this product uses most,

these other countries would, however, manufacture only

relatively small amounts of that particular textile or even

none at all.

8 See Leontief, Wassily, Machines and man, Sci. Amer.

187: 150-160, 1952. A different point of view is presented

in the detailed factual study by Rostas, L., Comparative

productivity in British and American industry, Cambridge

Univ. Press, 1948.

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

346 WASSILY LEONTIEF [PROC. AMER. PHIL. SOC.

can exports and those required for the replacement

of competitive imports have been compared above

4 and column 5 of table 2) into each of the resuilt-

ing five radial segments in figure 1.

One can clearly see that in the upper left hand

only in terms of the over-all averages. If the

explanation which has been given to these quanti-

part of the map, i.e., in the sectors containing

tative findings is correct, similar relationships

goods which require for their production larger

should also be discovered within separate com-

amounts of capital and comparatively small quan-

modity groups.

tities of labor, the white parts tend to be taller

A visual presentation of the quantitative rela-

than the black parts of the same blocks. As one

tionships revealed by the figures contained in the

moves towards the lower right hand corner, the

first four columns of table 2 is given in figure 1.

black strips tend to become higher than the cor-

Since we deal here with essentially foiir-dimen-

responding white strips; the tendency to export

sional phenomena, they cannot possibly be de-

goods requiring much labor and little capital for

scribed in an ordinary two-dimensional graph.

their domestic production and to import those

Each one of the black-white blocks on the graph,

which demand much capital and little labor can

figure 1, must be visualized as standing on the

in other words be as clearly discerned in this de-

flat surface of the paper not unlike a diminutive

tailed picture as it is reflected in the over-all

skyscraper rising above the base map in a three-

averages presented above.9

dimensional model of New York City. Each block

represents a separate commodity type. Its posi-

The results of this visual examination are sub-

stantiated by the following numerical compilation.

tion, or more exactly the position of its base, on

the flat surface of the map reflects the capital-labor

EXPORTS AND IMPORTS COMPARED BY SECTORS WITH

DIFFERENT CAPITAL INTENSITY

combination per million dollars of output required

for its production in the United States; the capital

Capital Per Man Trade Turnover a Percentage of Turnover

Year (In Dollars) (In Dollars) Exports Imports

requirement being measured upward along the

1234

scale marked along the left hand side of the chart,

More than 30,000 411,103 27.39 72.61

and the labor requirements-horizontally, along

30,000-17,750 394,465 47.90 52.10

the man years scale entered along the bottom

17,750-12,250 372,425 48.31 51.69

margin.
12,250- 9,700 395,028 61.76 38.24

The length of the black strip in each block (in

Less than 9,700 393,869 69.62 30.38

a truly three-dimensional figure it would be meas-

Aggregate 1,965,890 50.82 49.18

ured by its height above the capital-labor plane)

represents the level of exports and the white strip,

* Turnover within the line segments is not exactly

the imports of commodities of particular kind (see,

equal since they had to be summed for integral indus-

for example, the explanation of the plot of To-

tries. Aggregate turnover differs from two million dol-

lars due to rounding and the omission of the Other Nonbacco Manufactures (29) as given in the legend

ferrous Mineral Mining industry-cf. footnote f, table 2.

on the figure).

To facilitate the identification of all the indi-

It shows that as the capital/labor ratio goes

vidual blocks on the graph, their numbers-entered

down, exports make up an ever larger and imports

in brackets after the name of each industry in

a smaller fraction of the corresponding foreign

column 1 of table 2-are printed on a separate

trade turnovers.

transparent sheet which can be superimposed on

We have examined the over-all choice which

the graph. To make it possible to distinguish at

the American economy makes when it allocates

a glance the proportions in which capital and labor

its capital and labor to produce a million dollars'

are combined in the U. S. production of the variworth of the average combination of exportable

ous commodities, red reference lines are entered on

the same overlay sheet showing the capital/labor

9 The following ten service industries are omitted from

presentation in our figure: Railroad Transportation (31),

ratios of $30,000 per man year, $17,500 per man

Trucking (170), Warehousing and Storage (171), Pipe-

year, and so on. The capital/labor ratios, i.e., the

line Transport (175), Local and Highway Transportation

slopes of these four lines are chosen so as to in(178), Wholesale Trade (176), Retail Trade (177),

clude as nearly as possible one fifth of the total

Banking, etc. (36), Amusements (38), Communications

(35), and Other Water Transportation (173). Being

U. S. foreign trade turnover (i.e., of exports per

essentially non-transportable, the products of these indus-

million dollars of total exports plus imports per

tries cannot enter in any direct competition with imputed

million dollars of total imports, as listed in column

products of the same kind.

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

VOL. 97, NO. 4, 1953] DOMESTIC PRODUCTION AND FOREIGN TRADE 347

goods instead of using them to replace an equiva-

labor required to produce a million dollars' worth

lent average combination of imports. Behind it

of exports falling within each such commodity

are subsidiary choices based on differences in the

group were computed; similar computations were

labor and capital requirements of specific export

performed for the corresponding sets of competi-

and import goods belonging to the same com-

tive-in this case directly competitive-imports.

modity group and because of that directly compet-

In each instance the average was obtained by

ing with each other. The presence of direct com-

weighting the capital and labor requirements of

petitive relationships-or at least of more direct

an individual product (as listed in columns 2 and

competitive relationships than those which exist

3 of table 2) in proportion to the value of the

among all commodities entering international trade

exports and imports of that particular product per

-is of the essence for the existence of such sepa-

million dollars of the exports and imports re-

rate subsidiary allocation problems. A proper

spectively for the group into which it belongs as

isolation and detailed quantitative description of

a whole. The results of these computations are

such "internally competitive" groups constitutes

entered (in italics) in columns 6, 7, 8, and 9

the necessary prerequisite for their empirical

opposite the names of the groups listed on the

analysis.

left in column 1.

The study of this particular aspect of our pri-

To facilitate the interpretation of these sub-

mary data has not yet been completed.10 A care-

sidiary computations, the results of the comparison

ful perusal of the composition of American ex-

of the capital and labor requirements for export

ports and imports as listed in columns 4 and 5 of

and import replacement within each of the twenty-

table 2 enables us, nevertheless, to delineate a

six distinct "internally competitive" groups are

number of commodity groups which might reason-

ably, i.e., on the basis of the general knowledge

shown in the last two columns (columns 10 and

11) of table 2. The sign " >" indicates that the

one has about them, qualify for preliminary analy-

export requirement exceeds the corresponding re-

sis. As should be expected, they correspond

quirement for import replacement, "<" shows that

rather closely to the thirty-eight consolidated in-

the import replacement requirement is the larger

dustries described above. Some of the latter,

however, had to be broken down so as to separate

important sets of obviously non-competing opera-

tions, such as, for example, the mining and final

of the two. To mark very small differences

(amounting to less than 2 per cent of the larger

of the two figures) which should perhaps be inter-

preted as equalities, we used the signs ")" and

fabrication of metals; from some others, single

non-competitive components had to be eliminated.

A large number of export and import goods (al-

The following box scores summarize the final

results showing the values of exports and competi-

tive imports which fall within each of the distinct

though all of these, of course, were included in

the computation of the over-all average capital and

Exports (Unit: one thousand dollars)

labor requirements) had to be omitted from the

following analysis because they either did not fall

Capital

into any definite competitive set, or formed small

sets containing only two or three items.

> Total

Most commodities were actually combined in

"internally competing" groups and each set was

subj ected separately to the same analysis which

< 106 45 0 150

was previously applied to all exports and all com-

petitive imports taken together. The average

amount of capital and the average quantity of

L<

a > 97 78 0 176

10 This study leads directly toward the problems in-

volved in generalized formulation of interregional input-

output theory. The distinction between typically "do-

> 145 97 58 300

mestic" and the predominately "international" commodities

(+309) (+309)

is as fundamental for such analysis as the lower order

distinction between "national" and "regional" commodities

used in the study of the regional structure of the United

States' economy. (See, Leontief, Wassily, et al., Studies

in the structure of the American economy, Chap. 4 and 5.)

Total 348 220 58

(+309)

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

348 WASSILY LEONTIEF [PROC. AMER. PHIL. SOC.

Competitive imports (Unit: one thousand dollars)

a third factor or rather as a whole additional set

of factors determining this country's productive

Capital

capacity and, in particular, its comparative advan-

tage vis-a-vis the rest of the world, are the natural

< < >; > Total

resources: agricultural land, forests, rivers, and

our rich mineral deposits. Absence of systematic

quantitative information, similar to that which has

< 98 6 0 104

been collected, organized, and used in this paper

with respect to capital and labor, prevents us as

yet from introducing this important element ex-

a _ 25 50 0 75
plicitly into this preliminary analysis.

b>

However, indirect but clear signs of the influ-

ence of natural resources can easily be traced in

r

> 308 99 6 413

the capital and labor input figures presented in

(+408) (+408)
table 2 and depicted in our graph. This influence

is revealed mostly in their deviation from the

dominant pattern reflecting the comlparative capital

Total 431 155 6

(+408) _

shortage and labor surplus of the American econ-

omy. Without emiibarking on a detailed but neces-

'comparative cost types" per million dollars of

all exports and competitive imports, respectively.

Only 63 per cent of all exports and 59 per cent

of imports fell into specific competitive groups.

The rest which did not fit into any one of them

constitutes, so to say, a separate group. On the

basis of its comparative labor and capital require-

ments for exports and import replacements, this

residual group falls in the lower left box. It is

represented by the bracketed figures.

The examination of these figures shows that the

direct competition between exports and imports

belonging to the same commodity groups is domi-

nated by our relative capital shortage and labor

surplus, as is the over-all average picture of

American foreign trade which we have considered

before. Goods of the type requiring comparatively

more American man years (but a smaller amount

of capital) on the export side have a lion's share

($145 + $309 thousand) of our exports, while our

competitive imports consist. primarily of goods

($308 + $408 thousand) which, if they were pro-

duced at home, would absorb relatively large quan-

tities of capital but smaller amounts of American

labor. Disregarding the labor requirement en-

tirely, we also see that commodities requiring for

their production relatively small amounts of capi-

tal dominate our exports ($348 +$309 thousand)

while the capital intensive commodities-irrespec-

tive of their labor intensity-are preponderant

among competitive imports ($431 + $408 thou-

sarily conjectural examination of such special cases,

let me point to only a few of them as seen in fig-

ure 1.

Near its lower right hand corner we find a few

entries in which, contrary to the general tendency

prevailing in that part of the graph, the white part

of the block is taller than its black part. Consult-

ing table 2, we find that these labor intensive and

capital extensive industries showing such unusu-

ally weak position vis-a-vis competitive foreign im-

ports comprise, Sawmill (37), Pottery (73). and

Leather Products other than Shoes (68); all of

them are based on natural materials in which the

United States is obviously short as compared with

the foreign countries. On the other side of the

cluster among the capital intensive and labor ex-

tensive commodities of which we import as a rule

more than we export, Sulphur (19), Meat Pack-

ing (21), and Grain MIill (24) products show a

considerable export surplus. The United States

is apparently comparatively well situated with re-

spect to the domestic supply of such specific min-

eral and agricultural natural resources as are re-

quired in the production of these particular goods.

Without the necessary additional information

any further pursuit of this line of reasoning is

bound to become highly speculative. Conjecture

about facts is intriguing but-at least in the field

of economics-essentially futile in the long run.

Since the facts pertaining to this particular subject

are now being collected and organized, it might be

well to refrain from further speculation, however

sand).

Invisible in all these tables but ever present as

tempting it may be.

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions

VOL. 97, NO. 4, 19531 DOMESTIC PRODUCTION AND FOREIGN TRADE 349

VII. SOME GENERAL IMPLICATIONS

derived by the United States from its foreign

trade. Only a spectacular additional increase in

This study has been designed to ascertain the

domestic capital stock could tip the balance of

structural basis of the United States' trade with

comparative advantage to the other side and thus

the rest of the world. We find that, contrary to

bring about conditions which by common assump-

widely held opinion, our exchange of domestically

tion are already supposed to exist, i.e. a situation

produced goods for competitive imports serves as

in which the United States would actually find it

a means to compensate for the comparative shortadvantageous to use its foreign trade as a means

age of our domestic capital supply and a correto save American labor and io dispose of surplus

sponding over-supply of American labor.

American capital. In view of the determined ef-

Without attempting a systematic exploration of

fort of many so-called backward countries to in-

the possible wide-reaching implications of these

empirical findings, let me merely mention here a

few questions, the answers to which might be

seriously affected by the results of this preliminary

crease their own capital stock, such tipping of the

scale will take some time. On the other hand, the

factors, whatever they may be, which are responsi-

ble for the peculiarly high relative productivity of

American labor might soon become operative in

investigation.

Foremost among them is the problem of the

changing position of the United States in the

other economies and thus accelerate the elimina-

tion of disparity between the effective comparative

world economy. A richly abundant supply of

supply of capital and labor here and in foreign

natural resources-as compared with capital and

countries. This signifies, of course, a reduced in-

labor-dominated our early development and our

centive to the continued exchange of commodities

trade relations with foreign countries up to about

and services between the United States and the

1910. From the fact that at the present time

capital appears to be comparatively more scarce

rest of the world.

Since no discussion of foreign trade is consid-

than labor, one might surmise that this scarcity has

ered to be well rounded off without some mention

dominated our entire economic development until

of free trade and protection, I conclude with an

now. This would mean that-in terms of a com-

observation on that timeless subject. An increase

parison with the rest of the world-our capital

in the United States' tariff must obviously reduce

supply, while steadily growing, has still not caught

the volume of our competitive imports below what

up with the increase in our labor force, if the

it otherwise would have been; by restricting the

peculiarly high effectiveness of that labor force is

effective foreign demand for American goods, it

taken into account. A larger supply of domestic

would bring about also a corresponding cut in our

capital, if not matched by a corresponding increase

in domestic man power, will, in any case, reduce

exports. Since the exchange of goods and services

with foreign countries serves as a means to relieve

rather than increase the comparative advantage in

the pressure of our domestic labjor surplus and our

labor supply on which our present exchange of

capital shortage, a partial closing of that valve will

goods and services with foreign countries seems to

tend to increase such pressure. In other words,

be based. In other words, a more rapid rise in

protectionist policies are bound to weaken the bar-

our average productive investment per worker

gaining position of American labor and corre-

would diminish rather than increase the advantage

spondingly strengthen that of the owners of capital.

This content downloaded from 186.137.52.48 on Fri, 11 Mar 2016 15:11:36 UTC
All use subject to JSTOR Terms and Conditions