Why Hasn't Trade Grown Faster Than Income? Inter-Industry Trade Over the Past Century

Board of Governors of the Federal Reserve System International Finance Discussion Papers Number 371

January 1990

WHY HASN’T TRADE GROWN FASTER THAN INCOME?

INTER-INDUSTRY TRADE OVER THE PAST CENTURY

Joseph E. Gagnon and Andrew K. Rose

NOTE: International Finance Discussion Papers are preliminary materials circulated to stimulate discussion and critical comment. References in publications to International Finance Discussion Papers (other than an

acknowledgement that the writer has had access to unpublished material) should be cleared with the author.

ABSTRACT

Over the past century, the ratio of international trade to GDP has not grown substantially for most major OECD economies. We conjecture that growth in intra-industry trade has been offset by a decline in interindustry trade. Inter-industry trade may have declined either because of biased growth in factor inputs so that factor proportions have become more similar, or because preferences have become more similar with rising per

capita income.

WHY HASN'T TRADE GROWN FASTER THAN INCOME? INTER-INDUSTRY TRADE OVER THE PAST CENTURY

Joseph E. Gagnon and Andrew K. Rose?

I: Introcuction

It is commonly asserted that the world is highly integrated and rapidly becoming more so. Indeed, it is well known that trade has grown faster than income since the second world war (e.g., Krugman and Obstfeld (1988), pp 1-2). In this paper, we discuss long-run trends in international trade flows. Although the ratio of exports plus imports to GDP (hereafter, the "trade ratio") has grown for many economies during the post-war period, we note that the trade ratio does not typically display a strong trend over the past century. This fact (previously noted by Grassman (1980) among others) motivates our theoretical analysis. We point out that a model which incorporates both inter-industry trade and intra-industry trade need not display any long-run trend in the trade ratio.

In our analysis, intra-industry trade refers to trade in similar goods that require similar proportions of factor inputs. Inter-industry trade refers to trade motivated by differences in factor proportions or preferences across countries, as popularized by Eli Heckscher and Bertil Ohlin. We argue that inter-industry trade may have declined in importance for one or both of two reasons: a) factor proportions have grown more similar across countries as

physical and human capital have increased relative to land and labor; and b)

‘Board of Governors of the Federal Reserve System, and University of California at Berkeley, respectively. This paper represents the views of the authors and should not be interpreted as reflecting the views of the Board of Governors of the Federal Reserve System or other members of its staff.

We thank: Gwyn Adams and Russell Tarver for assistance with the data; Brian Cody, Neil Ericsson, Bill Helkie, Dale Henderson, David Howard, Linda Hunter, and Janet Yellen for comments. Our data and programs are available upon request.

2 non-homothetic preferences lead to a growing similarity in preferences across countries as per capita income increases. Both mechanisms can lead to a decline in the trade ratio unless they are offset by an increase in intraindustry trade.

In addition to resolving the apparently mysterious long-run behavior of the trade ratio, we also believe that a better appreciation of the characteristics of international trade during the late nineteenth century is important in understanding the intellectual development of the inter-industry theory of trade advocated by Heckscher and Ohlin. It is commonly argued that Heckscher-Ohlin trade theory cannot explain key features of internatioral trade, such as the large and quickly growing trade between similar countries in similar products (e.g., Helpman and Krugman (1985)). However, the Heckscher-Ohlin theory was of greater relevance in the past, when more tracle was motivated by differences in factor proportions or preferences.

Empirical evidence on long-run trends in the trade ratio is contained in the section which follows. Section III presents a theoretical discussi.on of potential sources of the decline in inter-industry trade. Our conclusions are

summarized in a brief final section.

II: Empirical Evidence on Trends in International Trade

Despite the recent increasing importance of international trade lows in many countries, there is little evidence that international trade has grown faster than output over the last century. Figures 1 through 14 contain timeseries plots of the trade ratio for fourteen different OECD countries, where the trade ratio is defined as the ratio of nominal exports and imports to GDP.

(Our data, documentation, and programs are available upon request; mos: of the

Bl 7@ 80 50 be 18 28 8 SO kt 7 86 50 00 10 28 36 4 650 66) Ote

Figure 1: American Trade Ratio Figure 2: British Trade Ratio

a8

8.7

g.6

a5

a4

a2

8.2

ary) 99 60 «18 (

70 30 0 hea Figure 3: Canadian Trade Ratio Figure 4: Danish Trade Ratio

re

2 70 0 em 18 20 30 8 OOS

Figure 5: Dutch Trade Ratio Figure 6: Finnish Trade Ratio

fe 73 86 «58 00 616 628 30 8

63788

Figure 7: French Trade Ratio Figure 8: German Trade Ratio

0.45 a6 4.48 a.5 0.35 0.38 aa 0.25 4.3 8.28 ag 0.15

at 8.18 8.05 2.8

“70 88 #96 B@ 10 20 30 48 SG O68 7” $8 98 8 10 20 36 OH O68

Figure 9: Italian Trade Ratio Figure 10: Japanese Trade Ratio

ph

98 20 16

7 838 «98 8 10 20 30 0 Hw GP “1 80 70 6300 OGD $2

a 7 #86 96 B@ 16 2 39 @ 3G 6B

Figure 13: Swedish Trade Ratio Figure 14: Swiss Trade Ratio

data are taken from Mitchell (1980).) A striking feature of the plots is the apparent lack of any long-run trend. We examine the ratio of nominal merchandise exports and imports to

nominal GDP for two reasons.” It seems appropriate to use economic value

?It would be desirable to examine total trade in goods and services rather than just trade in merchandise. Unfortunately, there is little data on trade in services before World War II. In 1986 merchandise trade accounted for almost 90 percent of total trade in goods and services for the United States, Germany, and Japan, according to OECD National Accounts.

5

rather than physical volume data, since the economic question of interest to us concerns the relationship between the value of goods and services actually traded and total economy-wide value added (i.e., GDP). Also, it is difficult to obtain reliable price indices for exports, imports and GDP for the sample in quesition.? We note in passing that the trade ratio clearly is not an ideal measure of the openness of an economy; the focus of this paper is the magnitude of international trade flows, rather than openness. *

Table I contains estimates of the average growth rates of the trade ratio and the average propensity to import. These estimates have been

multiplied by 100 for ease of interpretation as percentages; t-statistics are

included in parentheses. Also contained in Table I are regression-based

°Taere has been no discernible trend in the ratio of the wholesale price index to the consumer price index in the United States over the past 100 years. Since wholesale goods are mainly tradable, while many consumer goods are not, there is no prima facie evidence of a long-run trend in the tradable/non-tradable price ratio. Beenstock and Warburton (1983) provide a more detailed discussion of this point.

“Grassman (1980) argues that the trade ratio also may be a biased measured of openness because government spending (as a fraction of GDP) has increased over the past century. Grassman suggests adjusting the denominator of the trade ratio to exclude government spending. It is not clear that this is helpful; such a trade ratio of an open economy would change if the government decided to nationalize an industry, even if this did not affect either trade flows or GDP. Moreover, governments clearly influence international trade flows, either directly (e.g., current American military expenditures in Europe) or indirectly (e.g., by domestic purchases of exportable goods such as defense and agriculture).

In any case, the trade ratio measures actual rather than potential trade; it may be low despite the fact that the elasticity of trade flows with respect to small perturbations in the terms of trade is large. Consider two countries with identical tastes and identical Ricardian production functions. There will be no reason for such countries to engage in international trade, even if there are no barriers to trade (such as tariffs or transportation costs) so that the economies are perfectly open. However, an exogenous shock

to the production function of one of the economies would lead to substantial — trade flows.

Table I: Descriptive Statistics for Trend Growth in Trade Ratios

Average Growth Average Growth Rate Estimated Growth Rate of Ave. Prop. to Imp. Rate (excl. wars)

Canada . 2% 1% -.5% (t-statistic) (.2) (.1) (.5)

Denmark 4 .3 1.

Finland .2 «4 3,

France 7 ./7

Germany -.1 -.5

Italy wl -.4

Japan .7 -.1

Netherlands -1.1 -1.3

Norway 4 .3

Spain -2.1 -2.0

Sweden .5 .5

Switzerland

7 estimates of the average growth rate of the trade ratio. In column 3, the OLS estimate of the constant term (multiplied by 100) and the Newey-West t- Statistic are reported from a regression of the change of the log of the trade ratio on a constant and dummies for each year of both world wars.

The statistics in Table I effectively corroborate the impression conveyed by the plots, namely that there is little evidence of long-run growth in the trace ratio. Judged at conventional levels of Statistical significance, none of the trade ratios has exhibited trend growth. As more sophisticated time-series techniques (e.g., non-linear state-space methods) do not lead to different conclusions, it seems reasonable to conclude that there have not been strong trends in the trade ratios of these fourteen countries over the past century.

The lack of a strong drift in the trade ratio is all the more surprising in light of the numerous factors that might have led to substantial increases in the importance of trade. Worthy of particular mention are: 1) declines in protectionism (due in part to the establishment of GATT); 2) declines in transportation costs; 3) advances in communications technology; 4) increased liquidity of the international financial system (and the establishment of

central banks); and 5) increased barriers to labor mobility.°

°On the other hand, some researchers have proposed that there are factors that tend to reduce the importance of trade. Indeed, only a generation ago it was widely believed that international trade tended to stagnate or decline in importance secularly. Hicks (1953) argues that exogenous technological change was biased towards trade creation in the nineteenth century, but has been biased against trade in the twentieth century. Kuznets (1959) uses a century of data for ten countries to argue that there is little evidence of a long-run trend in the trade ratio; he asserts that a variety of factors affect the trade ratio, including technological developments that affect internal and external transportation costs at different rates and the changing sizes of various countries. Haberler (1964) characterizes the 1950s as "the first time in almost a hundred years [that] world trade has grown faster than world production" and argues that this was a result of two factors: rapid output

III: Two Theoretical Explanations

While inter-industry trade is motivated by differences in factors of production or preferences across countries, intra-industry trade is motivated by economies of scale in the production of differentiated products. Hirschman (1945) presents evidence that intra-industry trade was much less important than inter-industry trade during the three decades prior to World War Il. Since World War II, many researchers have documented a secular rise in intraindustry trade; Deardorff (1984) provides a survey. The conjecture of this paper is that there are forces leading to a secular decline in inter-industry trade. This section presents two non-mutually-exclusive hypotheses that may account for this phenomenon. Given the offsetting movements in its components, the aggregate trade ratio may rise, fall, or stagnate over the long run.

Inter-industry trade shrinks if countries tend to grow more similar in their effective factor proportions over time. There are several reasons to expect growing similarity in factor proportions. If natural resources were once a major source of differences in factor proportions, technology that

conserves resources also reduces the size of trade based on resource differen-

ces.® More generally, in a world where some factors are exogenously deter-

mined (e.g. land and labor) and other factors can be endogenously augmented

(e.g. physical and human capital), growth of the augmentable factors relative

growth and the removal of trade barriers; see also Cooper (1964). Finally, it has been argued that the loss of forced trade with colonies after liberation tends to reduce the trade ratio, especially because former colonies sometimes choose development strategies involving import substitution. A similar argument may apply to countries that become communist.

°Maddison (1987) documents a declining ratio of energy inputs to GDP for a group of OECD countries.

9 to the exogenous factors is likely to make factor proportions more similar across countries. That is, equal rates of investment in capital-rich and capital-poor countries--over and above the (exogenous) rate of growth of the labor force--will be associated with a growing similarity of factor proportions across these countries.

It is easy to show that in a world with two countries, two factors of productior,, and two goods, a growing similarity of factor proportions leads to a lower trade ratio. The proof is constructed geometrically. In Figure 15, the production possibility set of country 1 (which is relatively well-endowed with the factor that is used intensively in the preduction of good X), is given by aa; I, and I, are representative indifference. curves. In the absence of

trade, preduction and consumption occur

Figure 15: Autarky Equilibrium

at A, where the production frontier is tangent tc the indifference map. The price ratio that supports this equilibrium is given by the slope of the tangent to the indifference curve at A.

Once country 1 is opened to trade, consumption and production need not occur at the same point. For any given international price ratio, the optimal productior, point for country 1 occurs where the production frontier has the same slope as the price ratio. A tangent drawn to the production frontier at this point. provides the budget line. The optimal consumption point is the point of tangency of the budget line with the indifference map.

Figure 16 includes the production frontiers of both country 1 and

country 2, which is relatively better endowed with the factor used intensively

10

Figure 16: Initial Trade Equilibrium Figure 17: Factor Convergence

in the production of good Y. For simplicity, both country 1 and country 2 are assumed to have identical homothetic preferences. In equilibrium, each country faces the same price ratio, and the value of exports sold by each country equals the value of imports bought by the other. Without loss of generality, it is possible to construct the production frontiers such that both countries have the same total income given by the budget line gg. Since the representative consumers have identical preferences, both countries consume at point C. Production occurs at D for country 1 and E for country 2. The ratio of total trade to total production for country 1 is given by twice the ratio of the distance from c to d to the distance from 0 to g. The trade ratio for country 2 is identical.

Now consider reallocating the factor endowments of country 1 and country 2 so as to make the countries more similar, but still of equal size, as

depicted in Figure 17.’ If the countries retain the common budget line gg,

-_

"It would be more realistic to consider expanding both countries’ production frontiers at the same time that they become more similar. However, under the assumptions of constant returns to scale in production and homothetic

preferences, it is permissible to rescale the diagram by any constant multiple.

ll they will also retain the consumption point C. Country 2's production point, E’, must now be closer to C because of the reduced slope of the new production frontier; similarly with country 1's production point, D’. Total trade, and the trade ratio, shrinks.

An alternative hypothesis that can explain the decline in inter-industry trade is the existence of non-homothetic preferences. Hunter (1988) and Hunter and Markusen (1988) show in a static framework that non-homothetic preferences are both theoretically and empirically important in North-South trade. If poor countries have faster per capita growth than rich countries, preferences become more similar across countries over time as the gap in per capita income closes. Even if poor countries are not growing faster than rich countries, the net effect of growth is likely to make the representative indifference curves of rich and poor countries more similar.®

To focus on nonhomotheticity, we consider two countries with identical production possibility sets.° Country 1 has a larger population than country 2, however, so that its per capita income is lower. Individuals in each country have identical preferences. At low levels of income, preferences are skewed toward the "staple" good X. At high levels of income, preferences are

skewed toward the "luxury" good Y. Because of the difference in per capita

®Fol lowing Hunter (1988), this result can be derived theoretically from a Cobb-Douglas utility function with a minimum subsistence level for the staple good, e.g. U[x,y] = (x-x9)*y'*, where x, is the subsistence level of good X.

*When non-homothetic preferences are combined with different production possibilty sets, the effect of growth on the trade ratio may be strengthened or weakened, depending on the nature of specialization in production. If the poorer country is almost completely specialized in the production of the staple good, then growing similarity of preferences may even increase the trade ratio. We believe that such a scenario is unrealistic, especially since

growth that yields more similar preferences is likely to yield more similar production possibilities.

12

Figure 18: Non-Homothetic Preferences Figure 19: Convergence of Preferences

income, the representative indifference curves of country 1 and country 2 are not identical. In Figure 18, the equilibrium budget line is tangent to each country’s indifference contours at consumption points that are consistent with balanced trade. Production occurs at C in both countries; consumption is at A in country 1 and at B in country 2. Since both countries have equal p-oduction and there is balanced trade, the trade ratio is equal across countries.

Now suppose that growth leads indifference curves to become more similar across countries. By reducing the skewness of country 1's preference for X and country 2's preference for Y in a symmetrical manner, Figure 19 shows that the trade ratios for both countries drop. It is also the case that the trade ratio drops when only one country’s indifference curves shift, provided that

they shift in the direction of the other country’s.

IV: Conclusion International trade is increasingly dominated by trade between similar countries in similar products; intra-industry trade seems to account for a

growing share of international trade. In this paper, we have argued trat the

13 declining importance of inter-industry trade may approximately offset the growth of intra-industry trade, so that the aggregate ratio of exports plus imports to GDP need not show any trend. We also have suggested two reasons why inter-industry may have declined in importance: a) factor proportions have grown moe similar across countries as physical and human capital have increased relative to land and labor; and b) non-homothetic preferences lead to a growing similarity in preferences across countries as per capita income increase3.

We have not presented evidence that would allow us to discriminate between our hypotheses and other explanations for the decline of interindustry trade. Indeed, we believe that such tests will prove elusive for at least three reasons. First, there is much disagreement over the exact breakdown between inter- and intra-industry trade, even though intra-industry trade is increasing in importance by most definitions. Second, it is very difficult: to determine how much of a country’s industrialization is due to government policies that discourage imports of manufactures and how much is due to fundamental changes in factor proportions and preferences brought about through population growth and investment. Third, technological progress itself may change the shape of production frontiers in a manner that is very difficult: to measure. Nevertheless, we believe that our proposed explanations for the clecline of inter-industry trade are plausible and useful additions to

the theory of international trade.

14 References Beenstock, Michael and Peter Warburton (1983) "Long-Term Trends in Economic Openness in the United Kingdom and the United States" Oxford Economic Papers

35, 132-135 (with rejoinder).

Cooper, Richard (1964) "Growth and Trade" Journal of Economic History, 609- 628.

Deardorff, Alan V. (1984) "Testing Trade Theories and Predicting Trade Flows" in Handbook of International Economics (R.W. Jones and P.B. Kenen, eds, North-

Holland, Amsterdam).

Grassman, Sven (1980) "Long-Term Trends in Openness of National Economies" Oxford Economic Papers 32, 123-133.

Haberler, Gottfried (1964) "Integration and Growth of the World Economy in Historical Perspective" American Economic Review LIV, 1-22.

Helpman, Elhanan and Paul R. Krugman (1985) Market Structure and Foreign Trade (MIT Press, Cambridge).

Hicks, John (1953) "An Inaugural Lecture" Oxford Economic Papers, 121-135.

Hirschman, Albert 0. (1945) National Power and the Structure of Foreign Trade (University of California Press, Berkeley).

Hunter, Linda (1988) "The Contribution of Nonhomothetic Preferences to Trade” unpublished paper, Federal Reserve Bank of Dallas.

Hunter, Linda and James Markusen (1988) "Per-Capita Income as a Determinant of Trade" in Empirical Methods for International Trade (R. Feenstra, ed, MIT

Press, Cambridge).

Krugman, Paul R. and Maurice Obstfeld (1988) International Economics (Scott Foresman, Glenview).

Kuznets, Simon (1959) Six Lectures on Economic Growth (Free Press, Glencoe).

Maddison, Angus (1987) "Growth and Slowdown in Advanced Capitalist Economies:

Techniques of Quantitative Assessment" Journal of Economic Literature XXV, . 649-698.

Markusen, James R. (1986) "Explaining the Volume of Trade" American Economic Review 76-5, 1002-1011.

Mitchell, Brian (1980) European Historical Statistics (Macmillan, New York).

IFDP NUMBER

371

370

369

368

367

366

364

363

362

361

360

359

358

357

Please address requests for co Papers, Division of International F

15

International Finance Discussion Papers

TITLES 1990

Why Hasn't Trade Grown Faster Than Income? Inter-Industry Trade Over the Past Century

Contractionary Devaluation with Black Markets for Foreign Exchange

1989

Exchange Rate Variability and the Level of International Trade

A Substitute for the Capital Stock Variable in Investment Functions

An Empirical Assessment of Non-Linearities In Models of Exchange Rate Determination

Equilibrium in a Production Economy with an Income Tax

Tariffs and the Macroeconomy: Evidence from the USA

European Integration, Exchange Rate Management, and Monetary Reform: A Review of the Major Issues

Savings Rates and Output Variability in Industrial Countries

Determinants of Japanese Direct Investment in U.S. Manufacturing Industries

The U.S. and U.K. Activities of Japanese Banks: 1980-1988

Policy Rules, Information, and Fiscal fffects in a "Ricardian" Model

4 Forward-Looking Multicountry Model: MX3 {Implications for Future U.S. Net {Investment Payments of Growing U.S

Net International Indebtedness

U.S. Policy on the Problems of International Debt

Federal Reserve System, Washington, D.C. 20551.

AUTHOR (s)

Joseph E. Gagnon Andrew K. Rose

Steven B. Kamin

Joseph E. Gagnon

Guy V.G. Stevens Richard A. Meese Andrew K. Rose

Wilbur John Coleman II Andrew K. Rose

Jonathan D. Ostry

Garry J. Schinasi

Garry J. Schinasi Joseph E. Gagnon

Catherine L. Mann Henry S. Terrell Robert S. Dohner

Barbara R. Lowrey Eric M. Leeper Joseph E. Gagnon Lois E. Stekler

William L. Helkie

Edwin M. Truman

pies to International Finance Discussion inance, Stop 24, Board of Governors of the

IFDP NUMBER 356

355

354

352

351

350

349

348

347

346

345

344

343

342

341

16

International Finance Discussion Papers TITLES

International Economic Policy: The Role of Exchange Rates

An Econometric Analysis of UK Money Demand in Monetary Trends in the United States and the United Kingdom by Milton Friedman and Anna J. Schwartz

Encompassing and Rational Expectations: How Sequential Corroboration Can Imply Refutation

The United States as a Heavily Indebted Country

External Debt and Developing Country Growth

An Algorithm to Solve Dynamic Models

Implications of the U.S. Current Account Deficit

Financial Integration in the European Community

Exact and Approximate Multi-Period Mean-Square Forecast Errors for Dynamic Econometric Models

Macroeconomic Policies, Competitiveness, and U.S. External Adjustment

Exchange Rates and U.S. External Adjustment in the Short Run and the Long Run

U.S. External Adjustment: Progress and Prospects

Domestic and Cross-Border Consequences of U.S. Macroeconomic Policies

The Profitability of U.S. Intervention Approaches to Managing External Equilibria: Where We Are, Where We Might Be Headed, and How We Might

Get There

A Note on "Transfers"

AUTHOR(s) Edwin M. Truman

David F. Hendry Neil R. Ericsson

Neil R. Ericsson David F. Hendry

David H. Howard

Steven B. Kamin Robert B. Kahn

Ross Levine

Wilbur John Coleman II

David H. Howard

Sydney J. Key

Neil R. Ericsson Jaime R. Marquez

Peter Hooper

Peter Hooper

William L. Helkie Peter Hooper

Ralph C. Bryant John Helliwell Peter Hooper

Michael P. Leahy

Edwin M. Truman

David B. Gordon Ross Levine