A Theory of International Economic Structures
Many statesmen have been introduced to the benefits of free trade in undergraduate economics courses or, in earlier periods, through the writings of Adam Smith and David Ricardo. But over the last two centuries, the teachings of liberal trade economists have been followed only sporadically. The law of comparative advantage is one of the basic tenets of economics. Yet in perhaps no other area is the gap between economic theory and political practice so large.
As noted in the Introduction, the most widely accepted explanation of this anomaly is interest-group or, in public choice parlance, endogenous tariff theory. Government leaders, this theory maintains, are under pressure from highly organized, rent-seeking societal groups. Dependent upon such groups for support, politicians have little choice but to grant protection to claimants, thereby benefiting selected producers at the expense of consumers. This theory is attractive, and I build on it in Chapter 2.
Yet even within the advanced industrialized democracies, governmental forms vary in their degree of centralization and autonomy—factors that should allow for differential insulation from society and rent-seeking pressures.1 All of the major economic powers have maintained extensive protectionist systems at some stage in history despite clear differences in domestic structures. Great Britain in the seventeenth and eighteenth centuries, the United States in the late eighteenth and nineteenth centuries, France and Germany in the late nineteenth century, and Japan in the twentieth century have all vigorously pursued protectionist policies. These same countries, with the possible exception of Japan, have all pursued free trade at other times with almost equal vigor and often without a significant change in domestic structure.
Moreover, politicians and political economists have articulated important rationales for protection based on the “public interest.” Jean Baptiste Colbert’s mercantilism, Alexander Hamilton’s “infant industry” argument, the German historical school, and MITI technocrats in Japan all maintain that protection is in the national interest at certain stages in a country’s economic development. This economic philosophy has often dominated the political agenda in the past and may do so again in the future. At other times, the teachings of economic trade theorists have gained prominence and greater support. The current allegiance of politicians in the United States to the principle of free trade (but not necessarily the practice) provides ample evidence: even protectionist policies have to be couched in liberal rhetoric and framed in terms of reciprocity, or the opening of foreign markets to American exports. In each case, protection or free trade has been generally espoused as being in the “national interest.”
Finally, free trade and protection appear to be contagious and cyclical. The brief era of European free trade in the mid-nineteenth century rapidly gave way to near universal protectionism by the end of the century, only to be superseded by a period of free trade among the advanced industrialized democracies following World War II. Specific explanations for each country’s policy vary widely: France adopted free trade in 1860 in an attempt to secure British support for its northern Italy policy,2 the United States maintained its highly protectionist system because Republicans, who formed the dominant party after the Civil War, had declared themselves to be the “party of protection,”3 and so on. As Kenneth Waltz has persuasively argued in a slightly different context, this proliferation of explanations indicates the need for a systemic-level theory of trade policy which abstracts from national-level characteristics.4
From Adam Smith on, economists have struggled to make the real world of trade policy conform to their model, with limited success. In doing so, they have often ignored the systemic incentives for free trade and protection and the very real political issues at stake. I start here with the assumption that the political process is at least partly rational, that means relate to ends, and that the beliefs of statesmen have some basis in reality. Accordingly, I seek to explain the trade strategies adopted by nation-states. Furthermore, given the apparent presence of contagion and cyclical processes, the wide variance in unit-level explanations of national trade policies, and the public interest rationales developed by both protectionists and free traders, I attempt to build a systemic-level theory of national trade interests and strategy.
International Politics and International Trade
The international system is anarchic and based upon the principle of self-help.5 This is the basic fact of modern international relations.6 Anarchy is defined here as the absence of any centralized political authority higher than the nation-state. The absence of central authority does not necessarily indicate a lack of “order” within the international system.7 Anarchy is also compatible with strong international regimes, defined as “principles, norms, rules, and decision-making procedures around which actor expectations converge,”8 or other patterns of behavior recognized as legitimate by the community of nation-states. Regimes and order, however, are problematic under anarchy. Nationstates themselves must choose whether to participate in the global arena and what rules they will observe. In a self-help system, nation-states are responsible only to themselves. They are, in other words, the sole judge of their national interests.
Anarchy renders the construction of a stable, liberal, and open international trading order difficult. In the construction of free markets within countries, governments must provide the basic economic infrastructure, for example, property rights, transportation networks, and a stable currency, and they must coerce or induce the compliance of citizens whose interests are harmed or at least not benefited by the functioning of market processes.9 Governments succeed at home because their authoritative nature is backed by a monopoly of the legitimate use of violence.10 Within a society, for instance, thieves are imprisoned, taxes are levied for highway construction, and disadvantaged groups are compensated. In the international arena, nation-states must also provide a basic infrastructure and obtain the compliance of countries that gain less from free trade than others.11 Yet, in the absence of centralized authority, nation-states face a problem of collective action: each prefers that others bear the burden of providing the infrastructure and coercing or compensating the relatively disadvantaged. Free riding, for reasons developed below, can be overcome only by large and possibly middle-sized countries.
Anarchy and the principle of self-help also shape national interests in a second way. Self-help implies, first, that countries are concerned only with their national advantage and not with global welfare. The gains and losses to their own country capture the attention of central decision makers. Thus the reliance on self-help necessitates what Friedrich List termed a “political” rather than a “cosmopolitical” economy.12
Self-help also implies that a country will be concerned not only with its own absolute gain from trade but also with its relative gain, or the difference between its gain and the gains of others. Power, the pursuit of which follows from anarchy, is a relational concept. To the extent that one country is strong, others must be weak. Wealth, on the other hand, is an absolute trait. All countries gain from free trade, but—as John Stuart Mill first demonstrated—not necessarily to the same degree or extent. This simultaneous concern with both absolute and relative gains from trade and the contradiction between them has three implications for trade strategy. First, those countries that gain absolutely and relatively from free trade will have a stronger interest in such trade than those that gain only absolutely. Second, countries that gain absolutely and relatively will seek to construct regimes or other mechanisms of control to reinforce their advantageous positions. Third, disadvantaged countries will seek to strengthen their relative positions within the international division of labor and may be willing to bear considerable short-term absolute costs to accomplish this aim.
This chapter makes two other assumptions regarding the character of the “units” or nation-states.13 First, it is assumed that nation-states are unitary actors, that is, they act within the international system as single individuals. Thus, this chapter abstracts up from divisions within the state, between the state and society, and within society. This assumption is relaxed in Chapter 2 and the case study below. Second, it is assumed that nation-states are rational, that they make means-ends calculations and choose the policy with the greatest net return. This assumption has engendered a wave of legitimate criticism in the broader international relations and social science literatures.14 It has been proposed that nation-states, like individuals, may be satisficers or cybernetic decision makers.15 Individuals and, even more, nation-states are clearly not rational; few are able to make the means-ends calculations necessary for full or partial rationality to hold. Yet the assumption of rationality is useful and adopted here because it is the only decision principle that allows explanations or predictions to be derived from the international system without examining the internal decision processes of the individuals involved and the nation-state more generally (for example, the threshold of the satisficer and the order in which she encounters options, or the variables the cybernetic decision maker is tracking). Thus, despite its shortcomings, the assumption of rationality remains the only alternative available for the construction of systemic-level theories of international politics. Of course, to the extent that the predictions of systemic theories are incorrect, it may then be appropriate to introduce questions of misperception and cognition.16 Nonetheless, the assumption of rationality remains useful as a first approximation. When the unitary actor assumption is waived in Chapter 2, this rationality assumption is also partly relaxed.
Lessons from International Economics
Classical international trade theories demonstrate that nearly all countries are absolutely better off under free trade than in autarky or under protection.17 Adam Smith criticized the mercantilist trade restrictions predominant in his era for restricting the market, raising prices, and interfering with the division of labor, but he did not possess a fully developed theory of international trade. It was in the early nineteenth century that David Ricardo first demonstrated that the gains from trade did not depend upon absolute differences in productivity between countries, as was commonly believed. Even if one country is more productive in the manufacture of all goods (absolute advantage), Ricardo showed that as long as commodity prices differ in autarky, two countries will still gain from trade by specializing in the production of those goods in which they are relatively most productive (comparative advantage) and exchanging through international trade. The Heckscher-Ohlin theory of international trade confirms Ricardo’s insights but relates comparative costs and prices and, in turn, the pattern of trade to differences in factor (typically capital and labor) endowments. Accordingly, a country will export goods relatively intensive in the use of those factors it possesses in relatively abundant supply. Thus we can predict that a capital-rich country will specialize in capital-intensive goods and export them in exchange for relatively labor-intensive commodities, thereby increasing its welfare. Within this theoretical framework, any government intervention that inhibits the free exchange of goods across national boundaries reduces the gains from trade, introduces social deadweight losses into the economy, and decreases the utility of the country.
As all countries are believed to benefit from free trade, classical theories of international trade offer little guidance for a systemic theory that seeks to explain variations in national trade strategies across countries and over time. Moreover, although the propositions on the pattern of trade are relatively robust,18 the conclusion that all countries maximize their welfare over time under free trade is based on several restrictive assumptions. This classical model, though insightful, does not necessarily yield a complete understanding of the complex reality of international trade and trade policy making. Three more recent schools of thought relax the assumptions of classical theories, however, and provide an alternative view of the optimality of free trade.
First, the classic arguments in support of free trade assume that all countries are equally small “price takers,” or cannot affect the prices of the goods they buy and sell. In the real world, however, some countries are “price breakers” and others “price makers.” Optimal tariff theory, which has recently been subsumed under discussions of strategic trade policy,19 posits that a relatively large country can shift the terms of trade to its advantage (increase the ratio of its export to import prices) and capture a larger share of the total gains from trade by imposing a tariff. With an optimal tariff, the large country’s demand for the imported product falls, thereby lowering the world price, while the difference between the new lower world price and the now higher domestic price is retained by the government as tariff revenue. If the world price falls far enough, the large country will be better off with its optimal tariff than under free trade, despite the reduction in global welfare and the introduction of social deadweight losses into its own economy.20
Other countries may retaliate, however, and attempt to recapture the gains from trade by imposing their own optimal tariffs. The success of retaliation depends upon the relative sizes of the countries involved.21 If two countries are of equal size, both lose from the imposition of optimal tariffs after retaliation is taken into account. A large country imposing an optimal tariff against a smaller nation-state, however, may be better off even after retaliation than under free trade. Small countries can effectively retaliate against a large country only through collective action, but because retaliation also imposes costs on the small countries a free rider problem emerges. Relatively large countries, then, may possess optimal tariffs greater than zero following retaliation.
The second school of importance is the “new mercantilism,” founded on the Keynesian and neo-Marxist analyses of the balance of trade. From at least the age of mercantilism in the sixteenth century to the present, statesmen have believed in the salubrious effects of balance-of-trade surpluses (that is, exports exceed imports). Mercantilists perceived surpluses as the most effective means of augmenting the gold stock and expanding the money supply.22 The recent export-led growth strategies of the Federal Republic of Germany and others are contemporary manifestations of this old adage. Even today, trade deficits are commonly seen as a political problem which governments should rectify, whereas trade surpluses are generally desired.
In classical international trade theories imports and exports are always assumed to be instantaneously balanced. Accordingly, many economists dismiss the political emphasis on trade surpluses as antiquated “mercantilist” thought. As John Maynard Keynes wrote, however, “We, the faculty of economists, prove to have been guilty of presumptuous error in treating as a puerile obsession what for centuries has been a prime object of practical statecraft.” Starting with Keynes, the new mercantilists have argued that trade surpluses and the pursuit of a competitive edge in international trade do benefit a nation-state in the long run by stimulating profits, domestic investment, technical progress, and—ultimately—growth.23
The new mercantilist literature suggests two complementary goals for nation-states. Drawing upon several centuries of experience and the beliefs of statesmen, it is recommended that countries seek a trade surplus. It is also argued that countries need to obtain a competitive edge in international trade, although this would have to extend to a substantial cross-section of industries to have a significant effect. By stimulating exports, such an edge is sufficient to set off a virtuous cycle of growth.24
Critically important to this view is the open economy multiplier, which highlights the differential effects of imports and exports. Most simply, spending on imports puts income into foreign but not domestic hands; like savings, imports create a leakage out of the domestic income cycle. Like investment, however, increased exports provide a net injection into the domestic economy, stimulating an increase in national income the magnitude of which is determined by the marginal propensities to save and import. Increased exports or a reduction in the marginal propensity to import (conditions often reflected in a trade surplus but not necessarily so by definition), as a result, create virtuous cycles of growth whereas decreased exports or a higher marginal propensity to import can initiate vicious cycles.
The third new school focuses on increasing returns to scale. The conclusions of the Ricardian and Heckscher-Ohlin theories depend upon the assumption of constant returns to scale. Yet increasing returns to scale are “an indisputable fact of life.”25 Increasing returns or economies of scale are of two general types. First, internal economies are specific to the plant or firm and are clearly present as the expanded division of labor within the modern firm suggests, but limited, as indicated by the general absence of monopoly. Second, and more important, external economies or positive externalities apply to the sector or country. Here, increased production over time improves the quality of labor, develops entrepreneurship, expands intrafirm specialization in the production of intermediate goods, increases growth of technical knowledge and research and development, and enhances the national infrastructure. Positive externalities are generalized “spin-offs” of the production process.26 The presence of positive externalities is indicated by the tendency of production to concentrate around geographic centers within countries (cities) and within the international economy more generally (core countries).27 Increasing returns (both internal and external) are also the best explanation for the recent growth of “intraindustry” trade between the advanced industrialized countries.28
Increasing returns occur naturally in some industries and are often created by technical change in others.29 They can be static or, more central to the analysis developed here, dynamic and cumulative (that is, each technological innovation that increases returns begets a second round of innovation and change). Increasing returns tend to be associated with industries that intensively use physical and human capital. Labor-intensive products are typically agricultural and primary commodities or manufactured goods in the last stages of their product cycles. They generally produce little value added, are made with existing technology, and create relatively few positive externalities. Capital-intensive goods, on the other hand, are usually high value-added manufactured items early in their product cycles. Because of their relative “infancy,” internal economies are often not completely exhausted. And through learning or by stimulating innovation, capital-intensive production often creates significant external economies. The founding of the first steel mill eases the way for a second. Likewise, the training of a semiskilled or skilled labor force benefits all potential entrepreneurs in the region. Although debate continues on the magnitude of these and other external economies, capital-intensive production clearly appears to generate greater spin-offs than labor-intensive production. The best, and perhaps simplest, indicator of dynamic economies of scale is relative labor productivity (output per worker-hour) across industries or, at the level of aggregation used here, countries at any given moment, or within industries and countries over time (see Appendix).30 High relative labor productivity reflects both the capital intensity of production and, more directly, the larger increases in output that result from increasing returns production.
Making the more realistic assumption of varying returns across industries alters the clear predictions of the Ricardian and Heckscher-Ohlin theories. With increasing returns in some sectors and constant returns in others, it is difficult to make any generalizations because much depends on the type and extensiveness of the economies of scale and whether the production possibilities frontier retains its traditional concave shape or becomes convex.31
Most important, free trade may not be the optimal long-term policy in the presence of dynamic increasing returns. Under the assumption of constant returns to scale all industries are, in a very real sense, equal. It is immaterial whether a country possesses a comparative advantage and specializes in agriculture, textiles, or steel. When some industries yield increasing returns, however, this equality diminishes. The long-term benefits to countries that specialize in industries with dynamic increasing returns will be greater than those for countries concentrating on constant returns sectors. Because both internal and external economies are greater, countries of relatively high labor productivity enjoy a disproportionate share of the long-term gains from specialization and trade. Some measure of protection designed to “capture” a larger share of the world’s increasing returns industries may be warranted.32 This point is developed in more detail below.
The effects identified by these three new schools of international economics on national trade interests and strategy can be arrayed along the two dimensions of relative size, important for optimal tariffs and—as will be seen below—the policy implications of the open economy multiplier, and relative labor productivity, an indicator of differential returns to scale.
The International Economic Structure
The central proposition of this book is that the international economic structure and the position of nation-states within it create constraints and opportunities that shape the trade strategies of countries in important and predictable ways. The international economic structure is defined here as the configuration of nation-states within the two dimensions of relative size and relative labor productivity.33 Relative size is determined by each country’s proportion of world trade (exports plus imports). Relative labor productivity is defined and measured by national output per worker-hour relative to the average national output per worker-hour in the other middle and large-sized countries (see Appendix).
Neither of these two dimensions is entirely distinct from the national trade strategies I seek to explore. A country’s proportion of world trade, for instance, is affected by prior policy decisions. A protectionist trade policy will reduce a country’s share of world trade, but so will the policies of other countries, localized depressions, changes in exchange rates, and a host of other factors. Because of the numerous factors that influence a country’s relative size and labor productivity, it is possible to assume, for reasons of simplicity, that the causal relationship is unidirectional and specifically from the international economic structure to trade strategy.
Within the two dimensions of relative size and relative labor productivity, it is possible to identify at least seven categories of nation-states.34 These seven categories are defined graphically in Figure 1.1.
Table 1.1. The international economic structure, 1870–1938: Proportion of world trade (PWT) and relative labor productivity (RLP)
The empirical demarcations between categories are guided by the theory of hegemonic stability (see Introduction) but, in the end, remain somewhat tentative. The division between small and middle-sized nation-states is placed at 5.0 percent of world trade. Over the last one hundred years, no more than five nation-states were in the middle- and large-sized categories at any one time. For these categories to be meaningful, they should be limited to countries that do or can exert a major influence over the international economy or the policies of others. Nearly all of the nation-states traditionally regarded as major or important actors within the international economy have been above this 5.0 percent level. Raising the threshold between small and middle-sized countries by 1 or 2 percent would not dramatically affect the composition of the latter category (see Table 1.1). Reducing this cutoff point by 1 or 2 percent, however, would greatly expand the number of nationstates within the middle-sized category. A division of 15.0 percent of world trade between middle- and large-sized countries places the United Kingdom from the eighteenth century until the early twentieth century and the United States from World War II until the mid-1960s in the large category. This is largely congruent with the theory of hegemonic stability and was chosen for that reason.35 Most countries have been well below the 15.0 percent level. The precise division between the large and middle-sized categories is important here only for dating the transition of the United Kingdom from the former to the latter level. Given the rapid rate of decline in the British position before World War I (see Table 1.1), shifting the cutoff point between categories slightly in either direction would not have a major impact upon the analysis.
The relative labor productivity of each country is calculated by dividing that country’s absolute labor productivity by the average absolute labor productivity in the other middle- and large-sized nation-states (see Appendix). The 1.0 level, the demarcation between high and low relative labor productivity, indicates that labor productivity in one nationstate is exactly equal to the average of the others. A level below 1.0 indicates that the nation-state’s labor productivity is less than average; a level greater than 1.0 indicates that the country’s labor productivity is higher than average. The 1.0 division between nation-states of high and low relative labor productivity is intuitively meaningful and accurately reflects the second theoretical dimension discussed below.
Although all countries within a category share a common trade interest, the two dimensions of relative size and relative labor productivity are continuous. It is possible, as a result, to have variations within categories. The further to the right a nation-state lies on the horizontal axis within any specific category, for instance, the greater its interest in free trade will be relative to other members of that same category.
It is also important not to reify the demarcations between categories: they are clearly artificial constructs and are not intended to reflect how actual statesmen understand the world around them. Nor should they be taken too literally. The United States, for instance, accounted for 13.9 percent of world trade in 1929 and 15.3 percent in 1960, and even though the difference is only 1.4 percent I classify the United States as an opportunist in the first period and a hegemonic leader in the second (see Tables 1.1 and 1.2). More important for understanding America’s trade interests, however, is the configuration of nation-states within the international economic structure as a whole. In 1929 the United States shared its dominant position with Great Britain, which possessed 13.3 percent of world trade. In 1960, the United States’s share was nearly twice that of Great Britain, still its nearest competitor. The categorizations should be treated skeptically; although they are more formalized than other definitions of the international economic structure,36 assigning cutoff points between categories does not wholly substitute for judgment and intuition.
Table 1.2. The international economic structure, 1950–1977: Proportion of world trade (PWT) and relative labor productivity (RLP)
The relative size of a country is important for three unrelated reasons. Because the international infrastructure necessary to regulate foreign commerce and stabilize the international economy approximates a collective good, relatively large countries bear a disproportionate share of the costs of providing the infrastructure. In addition, relatively large countries possess international market power and may also enjoy an optimal tariff greater than zero. Finally, the ability of governments to manipulate the open economy multiplier is determined by their relative size within the international economy.
As in domestic political economies, an international infrastructure must be established to facilitate trade. The international infrastructure is composed of two separate dimensions: a set of regimes governing international transactions and international economic stability. For substantial trade to arise between countries, three tasks must be fulfilled, and they can be most easily accomplished through international regimes. First, a monetary system must be created specifying reserve assets and exchange rates. For the same reasons that money within a country is created as a store of value and a medium of exchange, an international reserve asset is necessary. International monetary relations, however, are complicated by the need to exchange one national currency for another, requiring rules governing such transactions. The content of this international monetary regime is less important: reserve assets have changed over time from gold, gold and silver, gold and United States dollars, to dollars and Special Drawing Rights (an international monetary unit created by the International Monetary Fund [IMF]); likewise, exchange rates have been fixed, or officially tied to the reserve asset, or floating, when prices are set by international market forces.37 It is important to have some common rules. In the absence of an international monetary regime, countries will be dependent upon barter—an inefficient form of exchange.
The second task is to ensure freedom of transit for trade, just as property rights must be enforced within domestic political economies. Both buyers and sellers must be confident that the goods they contract for will reach their destination with minimal interference. Traditionally, freedom of transit has been equated with “freedom of the seas,” a legal construct first developed under Dutch hegemony in the seventeenth century and continued by British and American leadership thereafter.38 Today, the expanded use of trucks and railroads for intracontinental transport and airplanes for rapid delivery has greatly expanded the subjects covered by the international transit regime.
Finally, an international financial regime must be constructed to facilitate capital flows between countries.39 Foreign finance accelerates the long-term economic expansion of less-developed markets. Some investment in local production or warehousing and distribution facilities may also be required to sustain a high volume of trade in goods. Most important, short-term credit is necessary to facilitate current trade in goods and to ease temporary balance-of-payments constraints. Without the necessary short-term credits, or unless markets clear instantaneously, trade is inhibited.
The specific content is often less important than the existence of some rules governing international monetary, transportation, and financial issues. Strong regimes will also, as Robert Keohane has argued, reduce transaction costs, provide information, and lower uncertainty,40 not only allowing international trade to occur but facilitating its expansion.
International economic stability, the second dimension of the international infrastructure, is also a necessary requirement for an open, liberal trading order. Societies seek homeostasis.41 When the international economy is unstable, or characterized by widely fluctuating exchange rates, prices, and, in turn, patterns of trade, countries are more likely to insulate themselves from this potentially disruptive force.42 Historically, protection has been a key instrument of insulation, but protection creates social deadweight losses. Only when the benefits of insulation exceed the costs of protection will countries withdraw from the international economy.
International infrastructure approximates a public good.43 Regimes can be exclusive (for example, the dominant naval power can ignore privateers who prey only on ships flying the flag of a particular country) or partial (for example, the Soviet bloc has generally been excluded from the IMF and General Agreement on Tariffs and Trade). Yet, in practice, excluding specific countries from international economic regimes is difficult and self-defeating because it undermines the very purpose of the regimes. Excluding countries from the benefits of international economic stability is even more difficult. Thus, although the infrastructure of international trade is not a “pure” public good, countries do possess an incentive to free ride on the efforts of others, and a collective action dilemma emerges.
Insofar as the international infrastructure approximates a collective good, the insights of Mancur Olson and Richard Zeckhauser, first developed to explain the unequal sharing of burdens in alliances, become helpful. In the provision of any collective good in a political economy in which members differ in size, these authors argue, there is a strong tendency for the largest members of the group to bear a disproportionate share of the burden of providing the good. The larger countries, in Olson’s earlier terminology, form a “privileged” group who value the collective good enough to provide it for themselves, even though they cannot prevent others from enjoying it as well.44 Following this logic, larger countries can and will bear a greater share of the costs of providing the international infrastructure.
Charles Kindleberger argues that because international economic cooperation is unstable, only a single hegemonic nation-state can lead an international economy. “With a duumvirate, a troika, or slightly wider forms of collective responsibility,” he states, “the buck has no place to stop.”45 There is no doubt that cooperation is more difficult to achieve in nonhegemonic systems. At the very least, the gains from cooperation must be larger to offset the additional costs of negotiating and enforcing agreements among the parties, and the results have been unstable historically.46 Yet Kindleberger’s conclusion has no grounding in collective goods theory; privileged groups need not be limited to one actor, although such a group is unlikely to exist in a system in which there are many equally small actors. Given the wide disparity in size in the international system even without hegemony, there is no a priori reason to conclude that international cooperation under a nonhegemonic system is impossible, but cooperation in the absence of hegemony will be most likely to occur between the middle-sized nation-states.47
The willingness and ability of nation-states to stabilize the international economy is also influenced by the resources available to them for regulation relative to the size of the disturbance(s) they must control. The present theory treats instability as an external shock to the international economic structure. The type and magnitude of economic disturbances may be related to the international economic structure, but several important types of disturbances, such as the accelerating pace of technological advance which led to the rapid decline in prices during the late nineteenth century or the condition of global agricultural surplus in the late 1920s and 1930s, span several different international economic structures and, as a result, can be safely treated as exogenous to this structure.
The ability of nation-states to regulate disturbances successfully is determined by two factors: the absolute level of resources available to them for regulation and the efficiency with which they use these resources.48 Each of these factors may compensate for the other. A single country with large resources, such as the United States at its hegemonic zenith, may be able to regulate a disturbance successfully even though the application of its resources is inefficient. Conversely, a nation-state with fewer resources may successfully regulate a similar disturbance if it uses these resources more efficiently. Two or more nation-states, particularly if they are middle-sized, are likely to possess greater collective resources than a single hegemonic leader. The combined shares of world trade of the United States and the United Kingdom, two middle-sized nation-states after 1912, were 25.2 percent in 1913 and 27.2 in 1929, compared with the United Kingdom’s 24.0 percent at the peak of its hegemony in 1870. Likewise, the United States, Federal Republic of Germany, and France accounted for 33.4 percent of world trade in 1977, whereas the United States alone held only 18.4 percent in 1950. Two or more nation-states, however, will be unlikely to use their combined resources as efficiently as a single hegemonic leader. At the very least, there will be costs involved in organizing the joint intervention necessary for the successful regulation of disturbances in the international economy.
Because disturbances can differ in magnitude, and assuming that nation-states differ in their resources and ability to manage them, it is not axiomatic that a hegemonic leader will be successful in regulating or stabilizing the international economy. This may have been the case during the Great Depression of 1873–96, when the United Kingdom’s international leadership faltered and the international economy moved toward greater closure.49 If hegemony does not guarantee stability, it is equally true that nonhegemonic nation-states may be able to cooperate and successfully stabilize the international economy under appropriate circumstances.
To summarize, large countries, and to a lesser extent middle-sized nation-states as well, possess incentives voluntarily to provide the infrastructure necessary for a liberal international economy. As a result, they will bear a disproportionate share of the burden of providing this quasi-collective good.
Relative size also affects the trade interests of individual countries in a second manner, discussed above in reviewing the literature on optimal tariffs. Trade barriers always reduce global welfare, yet large countries have an incentive to impose optimal tariffs because they increase national welfare. By adopting a tariff, the large country reduces its demand for the imported product, lowers the world price of the good (if the elasticity of supply is great enough), and—in this static model—turns the terms of trade to its advantage. This relationship is continuous: the larger the country, the greater are its incentives to impose optimal tariffs.
John Conybeare has argued that optimal tariff theory militates against the theory of hegemonic stability because large countries benefit not from free trade, as the latter theory posits, but from optimal tariffs.50 Yet it is difficult to draw any specific generalizations on this point. Whether or not a country’s postretaliation optimal tariff is greater than zero depends not only on its own relative size but also on the sizes of its trading partners. Even a hegemon is unlikely to benefit from protection if its trading partners are also substantial—or at least middle-sized in the terminology adopted above—and retaliatory. The relative size of a country by itself indicates little about the incentives for imposing optimal tariffs. More important is the relative size of the country’s trading partners, which requires examining the international economic structure as a whole.
For at least the last century, the international economic structure has always consisted of several middle-sized nation-states and, in two relatively brief periods, a single hegemonic leader (see Tables 1.1 and 1.2). All of these countries are likely to have possessed some market power, although this depends more precisely on the composition of trade and the various elasticities of supply. The circumstances under which optimal tariffs would be highest, or when a single hegemonic leader confronted a large number of poorly organized small countries, have been historically rare. Although large countries and, to a lesser extent, middle-sized countries may still possess postretaliation optimal tariffs greater than zero, these duties are not likely to be as great as sometimes estimated.51 Consequently, the static welfare gain from optimal tariffs may be offset over time by the third effect of relative size.
Relative size also influences the ability of governments to manipulate the open economy multiplier for political ends. The new mercantilists do not dispute the classical contention that global welfare is maximized by free trade. They do argue, however, that individual countries gain more from exports than from imports because of the multiplier effect. As J. B. Burbridge summarizes, “From the point of view of one nation, exports tend to increase employment, output, and profits; imports tend to reduce them.” Moreover, welfare gains from imports are static, but gains from exports are dynamic and cumulative. “Higher levels of profit, in turn, provide the finance to expand capacity and install still more efficient equipment, which tends to make their competitive advantage all the stronger.”52 Thus virtuous and vicious cycles of growth are set in motion in which export “success leads to success and failure engenders failure.”53
The presence of virtuous and vicious cycles opens the possibility of national gains through political manipulation of exports and the marginal propensity to import. All countries enjoy the multiplier effect, but their ability to influence it varies with their size. The demand for a small country’s exports is exogenously determined and, for practical purposes, independent of its trade policies.54 Although a country might gain a temporary advantage from export subsidies, such payments are likely to bankrupt its national treasury or be countered by similar measures abroad. Import protection, on the other hand, can reduce the marginal propensity to import and, as a result, the drain on the national income cycle without damaging the small country’s exogenously set exports. At any given level of exports, the open economy multiplier will then be larger. And by shifting consumption from imported to domestic goods, protection will also stimulate investment and, consequently, growth. This is not necessarily an argument in favor of domestic protection for small countries. Protection reduces the gains from trade and, by raising prices for goods, creates social deadweight losses. It does demonstrate, however, that protection is not an unmitigated evil for small countries.
The efficacy of export subsidies and import protection is reversed in large countries. In large countries, exports are not determined exogenously, but rather are partly a function of levels of imports and government export policies. Large countries are able to influence world prices of goods and may stimulate exports through subsidies. Any significant reduction in a large country’s imports, however, reduces the ability of other countries to purchase its exports. As a result, protection in large countries reduces both imports and exports. The loss of the export stimulus may be offset by increased domestic investment, but the net result is indeterminate. Moreover, the dynamic gains from the open economy multiplier may vitiate the static gains reaped by large countries through optimal tariffs; the terms of trade effect may increase welfare in the short run, but reduced exports may decrease welfare by an even larger amount in the long run. Whatever the precise effect, however, large countries are less effective than small countries in using protection to reduce the marginal propensity to import and thereby stimulate growth. This is an important disparity between small and large countries, with middle-sized countries occupying an intermediate position, which becomes more important in the discussion of relative labor productivity.
Relative Labor Productivity
Statesmen have traditionally believed that countries gain from industrialization and the expansion of manufacturing plant and that in the long run protection is an effective instrument in pursuit of this goal. Even today, industrialization and not specialization in agricultural, primary, or craft production is the goal of many statesmen in the Third World. And the economic success of Japan and the newly industrializing countries of East Asia has reinforced this belief. The problems arising from specialization in traditional industries and the dynamic advantages from industrialization are highlighted by Joan Robinson in her reexamination of Ricardo’s classic example of free trade between England and Portugal. “In reality,” she writes, “the imposition of free trade in Portugal killed off a promising textile industry and left her with a slow-growing export market for wine, while for England, exports of cotton cloth led to accumulation, mechanization and the whole spiralling growth of the industrial revolution.”55
Yet, according to classical trade theories, countries are best off specializing in goods that intensively use their most abundant factor. It is immaterial, in the terms of these theories, whether a country specializes in wine or textiles, raw material extraction or high-tech industry—a conclusion again at odds with practical statecraft. Two characteristics of classical trade theories need to be reassessed to reconcile them with political practice.
First, classical theories are static and based upon the existing productivity of current resources. Recent theoretical work has argued and historical experience has demonstrated, however, that comparative advantage is not fixed or immutable, but rather open to political manipulation. By intervention in the market, and particularly through protection of infant industries, industrial targeting, selective credit policies, and state-sponsored research and development, governments can reshape their economies creating comparative advantage, or what John Zysman and Laura Tyson refer to as “competitive advantage,” in new areas.56
Second, both Ricardo and Heckscher-Ohlin assume that internal and external returns to scale are constant, and the conclusion that the area of specialization is irrelevant follows from this simplification. In actuality, some industries are characterized by diminishing returns, others by constant returns, and still others by increasing returns. The latter industries are relatively intensive in the use of human and nonhuman capital and, accordingly, are marked by relatively high labor productivity. Consequently, capital-rich countries will tend to specialize in industries with increasing returns, as reflected in their high relative labor productivity, whereas land- or labor-abundant countries will specialize in industries with constant or diminishing returns to scale, as indicated by their low relative labor productivity.
With differentiated returns, a country’s area of specialization is no longer unimportant. Countries that specialize in increasing returns industries enjoy greater spin-offs or positive externalities. And—in conjunction with the effects of the open economy multiplier—a consistent bias is introduced over time in the distribution of the gains from trade in favor of relatively productive countries. “In this … case,” Nicholas Kaldor writes, “it is not the differences in the relative prices of factors that are important, but relative differences in labor productivities (measured in terms of a common currency) which, even in the absence of any differences in capital productivities (in the amount of capital employed per unit of output) will make the country with the higher productivity a favorable one for exports and a relatively unfavorable one for imports. It will therefore tend to have [ceteris paribus] a surplus of exports over imports. In the other … [countries] with which it trades, the opposite will take place.” Goods produced by increasing returns industries, in other words, tend to become relatively cheaper as production expands whereas the products of industries with constant or diminishing returns become comparatively more expensive. As exports increase, a country with relatively high labor productivity will grow more rapidly over time than others. “Its productivity growth will be accelerated and unless its domestic absorption (meaning its domestic consumption and investment) keeps pace with its faster productivity growth, its export surplus will reappear, giving rise to another push, making for faster growth rates for itself and slower growth rates for the others.” Thus increasing returns and expanding exports combine to create a doubly virtuous cycle of accelerating relative growth. “This,” Kaldor concludes, “is the principle of cumulative causation whereby some regions gain at the expense of others, leading to increasing inequalities” within the international economy. This is also the process referred to as “backwash” by Gunnar Myrdal, “polarization” by Albert Hirschman, and—in a slightly different context—“unequal exchange” by Arghiri Emmanuel.57
All countries continue to gain from trade in a world of nonconstant returns, but over time and in the presence of increasing returns countries with relatively high labor productivity gain disproportionately as a result of the doubly virtuous cycle. Because of these unequal gains, free trade is no longer the optimal policy. A country may be able to improve its condition by adopting protection and shifting domestic production toward the increasing returns good, thereby moving down its cost curve and, perhaps, manifesting an underlying comparative advantage it could not otherwise have obtained because of the entrenched position of existing competitors. This is the classic “infant industry” argument first put forth by Alexander Hamilton. In increasing returns production, however, external economies are created only over the long run; thus the period of “infancy” may be extended. In addition, if the positive externalities are large enough, it may still be beneficial to protect a specific industry even if it never outgrows its infancy. Thus “infant industry protection” is a misnomer and the term “increasing returns protection” is used here instead.
The increasing returns argument for protection applies equally to countries of both high and low relative labor productivity, although the height and extensiveness of the trade barriers should differ considerably. Countries with high relative labor productivity already possess a comparative advantage in increasing returns industries and gain both absolutely and relatively from trade. As a result, they possess a strong interest in free trade abroad. This allows the highly productive country to expand its exports and thereby to enjoy the fruits of the doubly virtuous cycle, and limits the ability of others to create a comparative advantage in competitive increasing returns industries. Free trade abroad, in other words, expands the virtuous cycle and reinforces the favored position of countries with high relative labor productivity at the top of the international division of labor.
Even though countries with higher than average labor productivity gain from free trade abroad, they may still use protection to compete with one another for larger shares of the world’s increasing returns industries.58 But because the doubly virtuous cycle depends upon exports leading imports, relative size also becomes important here. As noted above, in small and, to a lesser extent, middle-sized countries exports are set exogenously and domestic protection will have little or no effect on foreign demand. In large countries, however, exports are a function, at least in part, of imports. Consequently, domestic protection—of whatever form, including increasing returns protection—creates an impediment to exports. Large countries with relatively high labor productivity, as a result, will be less likely to adopt increasing returns protection than their small or middle-sized counterparts.
Countries with low relative labor productivity gain from international trade but typically reap fewer benefits than highly productive nationstates. Less productive countries, therefore, have a strong incentive created by their concern over the relative gains from trade to adopt an extensive policy of protection designed to stimulate increasing returns industries, disrupt the existing pattern of comparative advantage, and raise their position in the international division of labor over time, while continuing to export in areas of traditional or previously created comparative advantage. The ability of less productive nation-states to adopt a protectionist increasing returns strategy, however, is largely conditioned by the size of the domestic market, an admittedly nonsystemic but nonetheless indispensable factor.59 Countries with small domestic markets cannot realize economies of scale in production in the absence of international trade. Although they fail to gain to the same extent as more productive countries, less productive nation-states with small domestic markets may not be able to improve upon their free-trade utility. Protection will not significantly expand production or stimulate growth but will only introduce inefficiencies into the economy, and subsidizing production may be prohibitively expensive. Less productive countries with large domestic markets, on the other hand, can gain in the long run from stimulating increasing returns industries within their borders. Rather than specializing in the constant returns commodity, as a less productive economy would typically do under free trade, the country diversifies its economy, realizes economies of scale, and—if successful—creates a new comparative advantage in the increasing returns industry raising its level of relative labor productivity. The social deadweight loss from protection is offset, in this view, by the increased growth stimulated by the combination of export growth and increasing returns.
Structure and Interests
The effects of relative size and relative labor productivity on national trade strategies can be summarized by the following propositions: (1) Relatively large countries bear a disproportionate share of the burden of creating and maintaining the international infrastructure. This role may be provided by a single hegemonic leader, although it is not certain that even here such a leader will possess the necessary ability. Even though negotiating and enforcement costs will be higher, middle-sized countries may also be able to provide the international infrastructure through cooperative leadership. (2) Relatively large countries may possess optimal tariffs greater than zero, but this depends on the size and retaliatory proclivities of their trading partners and the dynamic effects of the open economy multiplier. (3) Countries with high relative labor productivity possess a strong interest in free trade abroad, although they may still use domestic protection to compete for greater shares of the world’s increasing returns production. Countries with low relative labor productivity and small domestic markets can do little to improve upon their free-trade utility. While also desiring to export goods in their areas of traditional or created comparative advantage, countries with relatively low labor productivity and large domestic markets can gain by stimulating increasing returns production. Consequently, they possess a strong interest in domestic protection.
These three propositions are interactive. Drawing upon each proposition, I derive trade policy preference orderings for the seven categories of nation-states identified above (see Figure 1.1). Ambiguities and plausible alternative preference orderings are noted where appropriate. Even though protection and free trade are continuous concepts, the choices available to a nation-state are simplified for purposes of analysis to greater free trade (FT) or greater protection (P) for itself (first term) and all others (second term). In this notation, for example, P/FT represents a preference for protection at home and free trade abroad. It should be emphasized that this notation does not necessarily imply policies of complete free trade or protection. A free-trade strategy can countenance some measure of protection for selected industries, and a protectionist policy may still leave some sectors open to international competition. The two choices merely refer to the general thrust or central tendency of policy.
As defined above, there are three categories of nation-states with high relative labor productivity. In order of increasing size, they are liberal free riders (LFRs), opportunists (OPs),60 and hegemonic leaders (HLs). Countries within all three categories share a strong interest in free trade abroad as a result of their high relative labor productivity. Free trade in other countries allows liberal free riders, opportunists, and hegemonic leaders to export and, by inhibiting the development of comparative advantage in competing increasing returns industries, preserve their favored positions within the international economy. Their incentives for optimal tariffs and increasing returns protection, which differ according to relative size, however, are contradictory. The larger the country, the greater is the likelihood that its postretaliation optimal tariff will be greater than zero. Conversely, the smaller the country, the more efficacious increasing returns protection will be over time.
Hegemonic leaders have a strong preference for universal free trade (FT/FT) when they confront at least middle-sized and retaliatory partners. In this case, their optimal tariffs will be low, if still greater than zero, and increasing returns protection will inhibit exports, thereby reducing the open economy multiplier. Even in the absence of free trade abroad, hegemonic leaders still possess few incentives to adopt domestic protection, which will only further restrict their exports. This suggests a dominant hegemonic strategy of free trade at home regardless of the policies of others. If hegemonic leaders do adopt protection, they are likely to prefer that others remain open for their exports. This yields a trade policy preference ordering of FT/FT > FT/P > P/FT > P/P. If, however, all of a hegemonic leader’s trading partners are small and its optimal tariff with retaliation is significantly greater than zero, a second preference ordering is suggested: P/FT > FT/FT > P/P > FT/P. The conditions underlying the first preference ordering are more typical, and that ranking is used in the analysis which follows. Interested readers can easily substitute the second ranking where appropriate. The United Kingdom in the mid-nineteenth century and the United States in the mid-twentieth century are the only two post—Industrial Revolution examples of hegemonic leaders.61
Opportunists, or middle-sized, relatively productive nation-states, possess less market power than large countries and, typically, face retaliatory trading partners. As a result, their optimal tariff is normally close to zero. Given their limited market power, however, exports are determined exogenously and their incentives for increasing returns protection are substantial. Although the protection/free trade dichotomy overstates their interests in domestic protection, the preferences of opportunists can be ranked as P/FT > FT/FT > P/P > FT/P.
Because opportunists possess only moderate influence within the international economy, their mixed interests create the tendency and their middle size the ability to free ride when a hegemonic leader is present to maintain a liberal international economy. The hegemon’s dominant strategy of free trade at home not only creates incentives for an opportunist to adopt increasing returns protection confident that its exports will remain robust, but may actually encourage the latter to pursue its albeit relatively low optimal tariff by removing the fear of retaliation. During periods of hegemonic decline, as we shall see, opportunists become critical in determining the openness or closure of the international economy.
The United States was an opportunist from at least the early nineteenth century until World War II. After a brief two decades as a hegemonic leader, the United States returned to this category in the mid-1960s. The Federal Republic of Germany and France joined the United States within this category in approximately 1965 and 1975 respectively (see Table 1.2). If past trends continue, Japan will also become an opportunist within the next decade.
Liberal free riders, although smaller, possess interests similar to those of opportunists. Their optimal tariffs are very low but, like those of their middle-sized counterparts, are offset by the desire to stimulate increasing returns production through protection. Thus the preferences of liberal free riders are ordered as P/FT > FT/FT > P/P > FT/P. Belgium and Sweden today would be classified as liberal free riders.
Corresponding to the categories of liberal free riders, opportunists, and hegemonic leaders are several categories of nation-states with low relative labor productivity. Like countries of high relative labor productivity, these nation-states all favor free trade abroad. International openness facilitates exports, hinders other less productive countries from fostering increasing returns industries, and prevents highly productive nation-states from sheltering their increasing returns industries.
Assuming that they possess large domestic markets, as might be expected, imperial leaders (ILs)—relatively large countries with less than average labor productivity—can increase their long-run relative gains from trade by stimulating increasing returns industries through domestic protection. This interest in protection derived from relative labor productivity reinforces tendencies in imperial leaders created by their large size to impose optimal tariffs. As a result, the question of retaliatory trade partners affects only the cardinal and not the ordinal preference rankings. The preferences of imperial leaders can be ordered as P/FT > P/P > FT/FT > FT/P. Imperial leaders, as a result, will pursue protection at home regardless of the policies of other nation-states unless induced or coerced to do otherwise.
The international trading system would be quite different under imperial than hegemonic leadership. Imperial leaders might still bear a disproportionate share of the costs of providing an international economic infrastructure, but the component regimes would be substantially weaker and most likely based on some form of administered trade. Since the Industrial Revolution, no imperial leaders have existed within the international economy.62
Even though small and middle-sized countries with low relative labor productivity have optimal tariffs close to zero they may still be extremely protectionist. Domestic market size will vary, however, and exert important effects.
As noted above, countries with low relative productivity and small domestic markets, referred to here collectively as free trade free riders (FTFRs), can do no better than to adopt free trade at home. Their preferences, as a result, are FT/FT > FT/P > P/FT > P/P. At first glance, this ranking appears somewhat paradoxical, because free trade free riders have stronger preferences for free trade than the more productive opportunists or liberal free riders. Free trade free riders, however, have little chance to expand their share of the world’s increasing returns industries, whereas the latter already have a comparative advantage in this area and compete for even larger shares. Free trade free riders, as a result, have few options and little opportunity to improve their relative condition. Although the concept of domestic market size is difficult to operationalize,63 many Third World nation-states and several of the less productive European countries today would be considered free trade free riders. Historically, countries with less than average labor productivity and small domestic markets have also been relatively small within the international economy. As a result, although their interests coincide with those of hegemonic leaders, they have generally played an inconsequential role in creating or maintaining an open international economy.
Small and middle-sized countries with less than average labor productivity and large domestic markets, referred to here as protectionist free riders (PFRs) and spoilers (SPs) respectively, have ordinally ranked preferences identical to those of imperial leaders: P/FT > P/P > FT/FT > FT/P. Although their optimal tariffs are considerably lower than those of imperial leaders, they have the same potential for stimulating increasing returns production through trade protection. Because of their middle size and protectionist preferences, spoilers can play a critically disruptive role within the international economy. France and Germany in the late nineteenth century and France and Japan throughout most of the post-World War II era are the principal examples of this category.
Structure and Strategy
International economic structures are distinguished by the number of middle- and large-sized nation-states present in the international economy and the categories into which they are classified. Changes in the international economic structure are of two kinds: changes within a structure and changes of a structure.64 Changes within a structure occur within specific categories. A hegemonic leader’s willingness and ability to stabilize the international economy, for example, is affected as its relative size increases or decreases, even though it may remain in a position of hegemony. More important, changes of structure occur when any middle- or large-sized nation-state changes category. The decline of a hegemonic leader into an opportunist, the rise of a protectionist free rider into a spoiler, or the transformation of a spoiler into a opportunist would each constitute a change of structure. Since the midnineteenth century there have been six distinct international economic structures: two of hegemony (United Kingdom, until 1912, and United States, 1945–65), two of bilateral opportunism (United States and United Kingdom, 1912–32, and United States and Federal Republic of Germany, 1965–75), one of multilateral opportunism (United States, Federal Republic of Germany, and France, 1975- ), and one of unilateral opportunism (United States, 1932–45).65 Each structure has its own processes for resolving the conflicting preferences of nation-states.
A hegemonic international economic structure is depicted in Figure 1.2. As in all game-theoretic matrices, the row player’s (in this case, the OPs, LFRs, SPs, PFRs, and FTFRs) payoffs are given first and the column player’s (HL) payoffs second. The payoffs are represented on an ordinal scale of one to four, with four being the most preferred outcome, three the second most preferred outcome, and so on. The Nash equilibrium cell is the one in which each player receives its highest payoff obtainable given the actions of the other. A dominant strategy exists when any actor would adopt the same policy regardless of what the other does.
As can be seen in Figure 1.2, a hegemonic structure will not axiomat-ically lead to a liberal international economy: universal free trade (FT/FT) is the Nash equilibrium only for hegemonic leaders and free trade free riders. In every interaction between a hegemonic leader and opportunists, liberal free riders, spoilers, or protectionist free riders, the equilibrium is FT/P (or P/FT from the perspective of the small and middle-sized countries; this lies in the southwest cell of matrices a and b). In each of these cases, the hegemonic leader must impose (offer) greater or lesser positive or negative sanctions (side payments) to obtain the nation-states’ compliance with universal free trade.66 In other words, the hegemonic leader must directly alter the costs and benefits of free trade that spoilers, protectionist free riders, opportunists, and liberal free riders face as a result of their positions within the international economic structure.
The sanction imposed (offered) by the hegemonic leader for compliance with a liberal international economy must be at least equal to if not greater than the difference between the best payoff obtainable by nation-states in the absence of compliance and the free trade payoff. In the cases of protectionist free riders, spoilers, liberal free riders, and opportunists the size of the sanction (or the price of compliance) must be equal to or greater than P/FT—FT/FT. The price of compliance is likely to be high for spoilers and protectionist free riders (difference between first and third choices) and moderate for opportunists and liberal free riders (difference between first and second choices).
The sanctions imposed (offered) by the hegemonic leader can take the form of threatening to close its market to the goods of the noncomplying country, absorbing the costs of adjustment, or accepting a measure of protection in some areas to secure cooperation in others. Historically, hegemonic leaders have also drawn upon their political-military power to ensure compliance by others to a free-trade regime. The United Kingdom’s role as the “balancer” of the nineteenth-century European state system and Napoleon Ill’s need for British support for his Italian policy, for example, may have induced France to adopt a free-trade policy in the Cobden-Chevalier Treaty of 1860.67 Likewise, America’s military presence in Europe has been a critical influence on the Federal Republic of Germany’s liberal orientation toward the international economy during the postwar era.68
In addition to paying the price of compliance, a hegemonic leader must also provide the necessary international infrastructure. This requires that relatively strong monetary, transit, and financial regimes be created or maintained and that the international economy be stabilized. The cost of providing these necessary conditions of international openness is referred to as the price of infrastructure (PI) and can vary over time.
Establishment or maintenance of a liberal international economy depends on the price of infrastructure and on the resources of the hegemonic leader available for influencing other nations relative to the sum of the individual prices of compliance. The total resources available to the hegemonic leader are limited to the difference between the outcome derived from the independent decisions of opportunists, liberal free riders, spoilers, and protectionist free riders (FT/P) and the free trade outcome (FT/FT). These resources, less the price of infrastructure, must be equal to or greater than the sum of the individual prices of compliance, or
(FT/FT - FT/P)HL - PI ≥ Σ (P/FT - FT/FT)OP, LFR, SP & PFR
for universal free trade to be established. It is not axiomatic that hegemony lead to a liberal international economy. In an international economic structure composed only of a hegemonic leader and spoilers, for example, the price of compliance would most likely exceed the benefits of universal free trade received by the hegemonic leader. Similarly, even if the hegemonic leader can pay the price of compliance, it may fail to regulate international instability successfully. Even in the absence of hegemony, some international openness will exist as free trade free riders pursue their dominant strategies of free trade at home.
The constraints and opportunities of a hegemonic international economic structure are relatively unambiguous; each actor possesses clear interests within the structure. The hegemonic leader will pursue free trade at home and abroad until the costs of doing so exceed the benefits. Free trade free riders will also adopt a liberal trade policy, although their actions will have little effect on the international economy as a whole. Opportunists, liberal free riders, spoilers, and protectionist free riders will protect their domestic economies while taking advantage of the openness provided by the hegemonic leader. They will deviate from this course only when coerced or induced to do so by the hegemonic leader.
An international economic structure of bilateral opportunism is shown in Figure 1.3.69 When two or more opportunists exist in an international economy, they confront a classic prisoner’s dilemma. If each opportunist attempts to maximize its individual gains (P/FT), the net outcome will be suboptimal (P/P). Opportunists have a high incentive to cooperate, however. Opportunists and liberal free riders possess strong interests in free trade abroad, although the latter will only marginally influence support for free trade within the international economy as a whole. Consequently, the cost of not cooperating (P/P – FT/ FT) is likely to be significantly greater than the gain that would be obtained by protecting selected industries (P/FT—FT/FT).
Moreover, drawing upon the recent literature on prisoner’s dilemma, there are at least two reasons for expecting the cooperative or universal free trade (FT/FT) outcome to occur. First, trade relations between nation-states approximate an iterated game. Trade policy making is a continuous process and tariff systems are periodically revised. Consequently, any defection by one party can be easily punished by others. When two actors participate in an iterated prisoner’s dilemma and, in Robert Axelrod’s terminology, the “shadow of the future” is large “tit-for-tat” or conditional cooperation is the maximizing strategy.70 In other words, when actors value future returns highly enough it is rational and possible for them to cooperate and accept their second best outcomes rather than attempt to achieve their most preferred outcomes.
It is also possible to identify the conditions under which cooperation will fail to arise in an iterated prisoner’s dilemma. Tit-for-tat is stable only if the shadow of the future is sufficiently large. International economic instability, and particularly fluctuations in the exchange rates and prices that determine the pattern of international trade, will increase the discount rate. In other words, instability changes a nation-state’s evaluation of the future trading system, leading it to place greater weight on present returns and devaluing possible but uncertain future gains. Thus opportunists are less likely to cooperate (adopt FT/FT) and more likely to pursue narrow short-term interests (P/FT) during periods of international economic instability.
Tit-for-tat is also unstable when an end point for the game exists or is perceived to exist. Defection (or protection in this case) is rational on the last play of the game. Knowing this, each player has an incentive to defect on the next to last move, and so on. Once an end point is perceived, it is extremely difficult to maintain cooperation. If an opportunist is believed to be changing or about to be changing categories within the international economic structure, an end point is effectively created. The shadow of the future diminishes and the structure of cooperation will rapidly break down. Thus if the international economic structure is perceived to be changing, each opportunist can be expected preemptively to defect from free trade and adopt P/FT.
Second, cooperation can emerge within prisoner’s dilemma by formal or informal agreements or “contracts” between the parties.71 Arthur Stein has identified prisoner’s dilemma as a subset of “dilemmas of common interest.” Such dilemmas, Stein argues, can be and have been resolved historically through “collaborative regimes” which constrain the self-seeking behavior of nation-states and allow for cooperation through highly formalized sets of rules which “specify what constitutes cooperation and what constitutes cheating” and procedures by which each party can be “assured of its own ability to spot others’ cheating immediately.” Likewise, Robert Keohane has argued that once created, regimes will be respected because they serve the useful purposes of reducing uncertainty and providing information. Cooperation is thereby reinforced by the presence of a regime. As they provide a “good” in and of themselves, regimes will tend to persist even after the underlying national interests that brought them into being have changed.72
Thus, cooperation can be expected to occur in prisoner’s dilemma when multiple plays of the “game” are possible and the shadow of the future is large, and/or when the parties are able to structure a regime to constrain their self-seeking behavior. Interestingly, the two arguments supporting the possibility of cooperation under prisoner’s dilemma reach different conclusions about the likelihood of cooperation persisting when underlying interests change. For those who focus on the iterative nature of the relationship, changing interests are likely to prompt preemptive defection. For at least some regime theorists, however, cooperation can be expected to persist for at least a relatively short period of time despite such change. The case study developed in Part II supports the former conclusion.
If two or more opportunists agree to cooperate and provide free trade among themselves, they still face the same costs as does a single hegemonic leader if they desire to expand or maintain free trade in the international economy as a whole. Figure 1.3c indicates that opportunists will exploit free trade free riders by adopting domestic protection. In this case, the crude dichotomization between freer trade and greater protection used in the game matrices proves misleading. Free trade free riders typically specialize in constant returns industries in areas of traditional comparative advantage, so their exports will not impinge upon the increasing returns industries protected in opportunists. The more likely equilibrium in these dyadic relationships, as under hegemony, is universal free trade (FT/FT). In their relations with spoilers and protectionist free riders, on the other hand, opportunists must be willing and able to pay the price of compliance demanded by others and the price of infrastructure. In addition, the opportunists will incur negotiating costs (NC) in orchestrating the joint interventions necessary to bring about a liberal international economy. These costs may be considerable, precisely because the negotiations aim to constrain or alter national behavior from what it would be in the absence of cooperation. These negotiating costs are likely to escalate as the number of opportunists increases. Although the aggregate gains from trade will increase as well, reaching a mutually acceptable agreement among a large number of actors will be difficult. As the number becomes large, and perhaps even as it approaches five or six, the problem of free riding among the opportunists may become insurmountable.
A liberal international economy will arise in bilateral opportunism only when the net benefits of free trade for the opportunists, less the costs of negotiation, and less the price of infrastructure, exceeds the price of compliance of the other nation-states. In other words, universal free trade will occur in a structure of bilateral opportunism only if
Σ(FT/FT – P/P)OP – NC - PI ≥ Σ(P/P - FT/FT)SP, PFR & LFR.
Thus the absence of hegemony does not necessarily mean the absence of leadership; under these conditions, effective collective leadership of a liberal international economy is possible and, indeed, probable.
The national trade interests revealed by a structure of bilateral opportunism are more ambiguous than those of a hegemonic structure. As under hegemony, free trade free riders will adopt liberal trade policies, while spoilers, protectionist free riders, and liberal free riders pursue protection at home and free trade abroad unless coerced or induced to do otherwise by powers with stronger interests in universal free trade. It is the mixed interests of the opportunists themselves which introduces the ambiguity into the system. The policies pursued by the opportunists and particularly their willingness to cooperate will be strongly influenced by the degree of instability within the international economy—making their efforts to control this instability even more problematic—and their perceptions on the likelihood of the present structure enduring into the future.
An international economic structure of unilateral opportunism is depicted in Figure 1.4. The most important difference between bilateral and unilateral opportunism is that the constraints on domestic protection imposed upon an opportunist by the need for cooperation are absent. The relationship between an opportunist and liberal free riders is similar to that in bilateral opportunism. Because of their small size, however, it is unlikely that liberal free riders can resist the temptations of free riding or supply the necessary constraints upon the opportunist to prevent it from adopting protection. Likewise, the dominant strategy of protection at home possessed by protectionist free riders and spoilers will fail to restrain protectionism in the opportunist. Only in its relations with free trade free riders does a single opportunist face incentives for free trade.
A single opportunist with a strong interest in free trade abroad may be able to exert a moderating influence upon protectionism in the international economy. The ability and willingness of a single opportunist to construct or maintain a liberal international economy will be determined by the benefits of universal free trade less the price of infrastructure, relative to the sum of the individual prices of compliance, or
(FT/FT – P/P)OP – PI ≥ Σ(P/P - FT/FT)SP, PFR, & LFR.
Given the magnitude of the task and the limited resources of the opportunist, it is not likely that a liberal international economy can be either constructed or maintained, although it remains a possibility. To the extent that the opportunist can induce or coerce others to adopt a measure of free trade, it may be able to obtain a partial success and create or maintain a modicum of openness within the international economy.
A paradox emerges in unilateral opportunism, however. Under conditions of international economic openness, which may exist temporarily as the legacy of a previous international economic structure, and high international economic instability, which will increase the desires for protection in all countries, the single opportunist may actually undermine the liberal order instead of moderating protectionism in the international economy and attempting to create at least a measure of free trade abroad. A single opportunist can achieve its highest payoff (P/FT) only by preempting the protectionist policies of spoilers and protectionist free riders, thereby creating a higher relative level of protection approximating its first choice. By preemptively adopting a higher level of protection, the opportunist channels imports formerly absorbed by its market toward other countries (which are now relatively more open) while, in the short run and to the extent that these now diverted goods are noncompetitive with the exports of the opportunist, not significantly reducing its foreign markets. The opportunist, as a result, obtains greater protection at home than it would have otherwise achieved and maintains its exports. This strategy, however, will quickly lead to a tariff war in which, through a pattern of action and reaction, the opportunist and spoilers build higher and higher tariff walls around their economies. As a result, this strategy will create benefits for the opportunist only if implemented preemptively. If the opportunist waits until other nation-states have increased their levels of protection, it will lose its export markets and fail to gain additional protection for its own economy. This strategy is logical only if greater protection abroad is imminent and is likely to occur in the transition from some other structure to one of unilateral opportunism. Because the opportunist gains from openness in the international economy and its preemptive protection will clearly act as a catalyst for closure, acting too soon or before protection abroad is imminent will cause an undue loss of exports for the opportunist. Achieving maximum benefits, therefore, requires precise timing. Even with such timing, the strategy will provide benefits to the opportunist only in the short run, defined as the time it takes other nation-states to retaliate.
Once tariff levels reach heights that prohibit international trade and nation-states become locked into the extreme P/P outcome, the interests of the opportunist will be best served by a return to greater free trade abroad. Any reduction in foreign tariffs then benefits the opportunist. Although it most likely cannot lead the international economy effectively, the opportunist will benefit from even limited bargaining with other countries over tariff reductions. Because the influence of a single opportunist will be limited, the bargaining will most likely center on the exchange of tangible concessions between countries, and the actual reductions are likely to be moderate.
Like bilateral opportunism, the constraints and opportunities of unilateral opportunism are more ambiguous than under hegemony. Whereas under normal conditions a single opportunist will seek free trade abroad, it is influenced by the preexisting international economic structure and the level of stability. During the transition from an international economic structure of hegemony or bilateral opportunism, and particularly if it coincides with a period of high international economic instability, the single opportunist is likely to seek preemptive protectionism, a paradoxical strategy. If it does so, however, the opportunist will soon return to a strategy of seeking to build free trade abroad.
In summary, hegemony is neither a necessary or sufficient condition for the creation or maintenance of a liberal international economy. If the price of compliance and/or the price of infrastructure exceeds the benefits of universal free trade, the hegemonic leader will be both unwilling and unable to lead the international economy toward greater openness. Conversely, two or more opportunists, and in some cases even a single opportunist, may be willing and able to construct or maintain a liberal international economy if the conditions specified above are met. Thus it is important to distinguish between different nonhegemonic international economic structures. Each possesses distinctive politics and processes important for understanding the international economy.
In conclusion, it should be emphasized that the theory of international economic structures developed here does not contain within it a theory of change. There is no mechanism endogenous to the international economic structure which drives change at the systems level from one structure to the next or at the country level from one category to another. On the basis of the theory sketched above, however, it is possible to suggest a tentative model of change. Robert Gilpin and others see hegemony as fragile, ultimately insecure, and doomed to decay. “From a political perspective,” Gilpin writes, “the inherent contradiction of capitalism is that it develops rather than that it exploits the world. A capitalist international economy plants the seeds of its own destruction in that it diffuses economic growth, industry, and technology, and thereby undermines the distribution of power upon which that liberal, interdependent economy has rested.”73 In practice, this is correct. In the theory developed above, however, successful hegemony, bilateral opportunism, or (rarely) unilateral opportunism is self-reinforcing. To the extent that a hegemonic leader, or one or several opportunists, is successful in creating a wholly open international economy, their positions will be strengthened and others weakened by limitations placed on the latters’ ability to create a comparative advantage in the increasing returns production upon which dominance rests. The “tragedy” of hegemony or opportunism is that it is never entirely successful. With the onset of the Great Depression of 1873–96, Great Britain failed to take any active measure to preserve openness in Europe or the United States and, instead, took the easier route of turning inward upon its empire.74 Similarly, the United States, preoccupied with Cold War concerns, compromised on its newly found free-trade principles during the 1950s so as to rebuild Western Europe and Japan. In both cases, protectionist competitors did create comparative advantage in increasing returns industries, ultimately undermining the economic base of the hegemon.
The International Economic Structure and American Trade Strategy
The constraints and opportunities of the international economic structure are manifested at two levels. Depending upon the long-term benefits of free trade for various countries, price of infrastructure, and negotiating costs (for two or more opportunists), greater free trade or protection will result in the international economy as a whole. This chapter has focused on these necessary and sufficient conditions for openness or closure in the international economic system.
Individual nation-states, however, also face constraints and opportunities created by the international economic structure. Two or more opportunists, for instance, cannot simultaneously obtain protection at home and free trade abroad. Conversely, an opportunist can free ride on the international economic openness provided by hegemonic leadership. Stated more formally, the constraints and opportunities of the international economic structure create opportunity costs, defined as the relative costs and benefits of alternative trade strategies, for individual countries. The international economic structure, in other words, creates differing rewards and punishments for alternative policies. Any nation-state that acts against or ignores the constraints and opportunities of the international economic structure will receive less than the maximum reward and will be less well off than it otherwise could have been. The policy alternative with the highest reward (lowest punishment) can be thought of as the “national trade interest.”75
This book examines the systemic-level theory of international economic structures in the case of American trade strategy between 1887 and 1939. Rather than attempting to analyze the constraints and opportunities of the international economic structure at the level of the international economy as a whole, they are examined at the level of an individual nation-state. Problems of operationalization and data availability would make the former approach difficult, although perhaps not impossible. By focusing on a specific country, the behavior of other nation-states can be used as proxies for the costs and benefits of free trade for the hegemonic leader, the price of infrastructure, and the price of compliance. Thus, much of what would be problematic if the theory were to be tested at the level of the international economy can be treated descriptively in a case study of trade strategy in a single nation-state. I am not attempting to explain the trade strategies of countries other than the United States in the case study. When I refer to Great Britain as a hegemonic leader, for instance, I am not purporting to explain British policy as a function of its structural position. I am merely using the label as a shorthand for the syndrome of policies associated with hegemony and pursued by Britain at this time.
The case of American trade strategy between 1887 and 1939 was chosen for three reasons. Four distinct international economic structures existed during the period (see Figure 1.5). An international economic structure of British hegemony existed from the late eighteenth century until approximately 1897. At that date, America’s relative labor productivity finally exceeded Britain’s and a change within the structure occurred. Although Britain’s position within the international economy had been gradually declining since approximately 1870, the period after 1897 is referred to as one of declining hegemony because Britain was no longer the largest and most productive country. In 1912, a structure of bilateral opportunism emerged, with the United States and the United Kingdom as the two opportunists. This structure lasted until approximately 1932, when Britain declined into a spoiler, creating a structure of unilateral opportunism. Thus there is significant variation in the international economic structure during the fifty-two years covered. Each of these four structures is associated with a major change in American trade strategy. Specific propositions and expectations are outlined in the four chapters in Part II.
The United States was chosen also because it approximates a least likely critical case study for the systemic theory developed here. Its large domestic market, low level of international economic dependence, domestically oriented ideology, and strong social groups should have enabled it to ignore the constraints and opportunities of the international economic structure if any nation-state could. To the extent that American trade strategy reflects the constraints and opportunities of the international economic structure, this constitutes strong support for the theory.
Finally, the case is important for the insights it can yield into the present era of declining hegemony. A simplistic analogy is often drawn between the decline of the Pax Britannica and the decline of the Pax Americana. By differentiating between nonhegemonic international economic structures, the theory outlined in this chapter invalidates this analogy. During the late nineteenth and early twentieth centuries the international economic structure evolved from hegemony, into bilateral opportunism, and finally into unilateral opportunism. The international economy experienced its greatest strains during the transition between the second and third structures, ultimately breaking down into a series of regional trade and currency blocs in the early 1930s. The Pax Americana, on the other hand, appears to be evolving from hegemony, into bilateral opportunism, and into multilateral opportunism—where it is likely to remain for the foreseeable future. This is a much more stable evolutionary path, and multilateral opportunism allows for considerable international economic cooperation and openness (discussed in more detail in the Conclusion). If the theory developed above is correct, the best historical analogy for the present is the period from immediately before World War I until the late 1920s, the only other era of either bi- or multilateral opportunism. By studying the problems and potentials of trade policy making in this earlier period, perhaps some pitfalls can be avoided within the present international economic structure.
Figure 1.5. The international economic structure, 1870–1938Methodological Considerations
One strategy for assessing the utility of the theory of international economic structures would be to model empirically America’s national trade interest and the relevant counterfactuals deduced from the theory. Although this approach would surely be revealing, it would require an extremely sophisticated model of the United States economy and an even more rigorously specified and operationalized theory than presented here. Because of the inherent difficulties of such an exercise, particularly at the early stage of theoretical development reached in this book, and my own inclinations and skills, this approach is not pursued. Rather, the costs and benefits of free trade or protection are treated by assertion as if the theory developed above has indeed identified the national trade interest. Thus the focus of the analysis is not on whether the United States maximized its long-term national income, but whether national policy makers chose trade strategies and conducted affairs in pursuit of the national trade interest identified by the theory.76 From the viewpoint of a political scientist, at least, the latter is the more interesting question.
Even when focusing on policy choice, however, substantiating a structural argument like the one developed here is difficult. Any nation-state can choose to contravene its systemically derived national trade interest. Indeed, countries are subject to multiple domestic and international pressures. Although the nation-state as a whole may lose by acting against the constraints and opportunities of the international economic structure, groups and interests within the country may benefit. These groups may seek to pressure the government into adopting policies that fulfill their narrow interests at the expense of the national trade interest (this possibility is developed further in Chapter 2). Likewise, a country’s dependence on the international economy, a nonsystemic attribute, influences the choice of trade strategy. In short, the constraints and opportunities of the international economic structure may simply be overwhelmed by other pressures within the policy-making process. To determine when the constraints and opportunities of the international economic structure will be followed and when they will be overwhelmed would require a theory of political economy which integrated all of these sometimes complementary and sometimes competing pressures. A first step is taken in this direction in Chapter 2, but the field of international political economy does not possess the necessary theory at the present time.
In light of the anarchic nature of the international system and the consequent need to ensure national survival in a competitive environment, it is hypothesized that nation-states will normally give priority to the constraints and opportunities of the international economic structure. Their trade strategies should therefore reflect, at least in part, the national trade interest as identified here. Furthermore, to the extent that any nation-state chooses to contravene its national trade interest, it is hypothesized that central decision makers will be cognizant of the trade-offs between this interest and other political pressures. The existence of competing pressures does not, in the end, pose a major problem for the theory: in the case study the constraints and opportunities of the international economic structure are reflected in trade strategy with only a few exceptions, indicating the high salience attached to systemic considerations.
Not only do competing pressures exist, but the causal linkages postulated in structural theory are often difficult to observe in specific cases. The constraints and opportunities of the international economic structure do not force a nation-state to adopt any particular policy; they only make some options more attractive and other options less so. The theory posits only constraints and not determinants of behavior. Structures are like strainers that filter out otherwise viable options. Excluded possibilities—or the counterfactuals—are not easy to define. Conversely, since favored options are seldom rationalized simply on structural grounds, heightened possibilities are difficult to discern as well.
As a first step, supporting or disconfirming evidence can be obtained by establishing the empirical relationship between the international economic structure and the substantive policy choices. Was the final policy as adopted and implemented congruent with the constraints and opportunities of the international economic structure as defined above? Consistently supporting evidence would affirm the theory.
It is also possible to make a stronger argument in favor of the theory and establish the actual presence of the constraints and opportunities of the international economic structure through two additional modes of analysis. By establishing a baseline trajectory or course of action which might have been followed in the absence of the structural constraints and opportunities it is possible to identify the effects of the international economic structure by assessing the magnitude of the policy deviations that occurred and that were predicted by the theory. Given the importance normally attached to pressures from domestic interest groups on trade strategy, the case study assumes that these pressures would have created the central thrust of policy if the constraints and opportunities of the international economic structure had not been present. It may also be possible to isolate favored and suppressed policy options through “process tracing,” or a detailed analysis of the policy-making process and the terms of political discourse. The argument set forth in Chapter 2, which states that the executive should be most responsive to the demands of the international economic structure and the legislature most responsive to domestic or societal pressures, facilitates this task by defining which domestic political actors should adopt which political viewpoint. All three forms of analysis are adopted in the case study.
1Peter J. Katzenstein, ed., Between Power and Plenty: Foreign Economic Policies of Advanced Industrialized States (Madison: University of Wisconsin Press, 1978). For approaches that examine various institutions within the United States, see Judith Goldstein, “A Re-examination of American Trade Policy: An Inquiry into the Causes of Protectionism” (Ph.D. diss., University of California, Los Angeles, 1983), and Robert E. Baldwin, The Political Economy of U.S. Import Policy (Cambridge: MIT Press, 1985).
2A. A. Iliasu, “The Cobden-Chevalier Commercial Treaty of 1860,” Historical Journal 14 (March 1971): 67–98.
3Tom E. Terrill, The Tariff, Politics, and American Foreign Policy, 1874–1901 (Westport, Conn.: Greenwood, 1973).
4Kenneth N. Waltz, Theory of International Politics (Reading, Mass.: Addison-Wesley, 1979).
6Like all “facts,” however, the permanence of anarchy is open to several interpretations. Richard Ashley has argued that Realists assume the nature of the units or states when this concept should be conceived of as a variable and made endogenous to the theory (“The Poverty of Neorealism,” International Organization 38 [Spring 1984]: 225–86). Although Ashley’s point is correct and important for “grand theory,” a middle-range theory such as that developed here can safely ignore “first principles.”
7Hedley Bull, The Anarchical Society: A Study of Order in World Politics (New York: Columbia University Press, 1977).
8Stephen D. Krasner, ed., International Regimes (Ithaca: Cornell University Press, 1983), p. 1.
9See Douglas North, Structure and Change in Economic History (New York: Norton, 1981); Karl Polanyi, The Great Transformation: The Political and Economic Origins of Our Time (Boston: Beacon, 1957).
10This is the distinguishing characteristic of the state. See Max Weber, Economy and Society, 2 vols., ed. Guenther Roth and Claus Wittich (Berkeley: University of California Press, 1978), 1:56.
11The attributes of the international infrastructure are noted in the Introduction to this book and discussed in Charles P. Kindleberger, The World in Depression, 1929–1939 (Berkeley: University of California Press, 1973), esp. p. 292, and “Dominance and Leadership in the International Economy: Exploitation, Public Goods, and Free Rides,” International Studies Quarterly 25 (June 1981): 247.
12Friedrich List, The National System of Political Economy, trans. Sampson S. Lloyd (London, 1885; rpt. New York: Kelley, 1977).
13For a general discussion of these assumptions, see Robert O. Keohane, “Theory of World Politics: Structural Realism and Beyond,” in Ada W. Finifter, ed., Political Science: The State of the Discipline (Washington, D.C.: American Political Science Association, 1983), pp. 503–40.
14For a recent and relevant version of this critique, see Timothy McKeown, “The Limitations of ‘Structural’ Theories of Commercial Policy,” International Organization 40 (Winter 1986): 43–64.
15Herbert Simon, Models of Man: Social and Rational (New York: Wiley, 1957), pp. 241–60, and John Steinbrunner, The Cybernetic Theory of Decision: New Dimensions of Political Analysis (Princeton: Princeton University Press, 1974).
16Robert Jervis, Perception and Misperception in International Relations (Princeton: Princeton University Press, 1976).
17See virtually any text on international trade theory, for instance Richard E. Caves and Ronald W. Jones, World Trade and Payments: An Introduction, 2d ed. (Boston: Little, Brown, 1977), and at a more advanced level, Jagdish N. Bhagwati and T. N. Srinivasan, Lectures on International Trade (Cambridge: MIT Press, 1983).
18“Robustness” here refers to the sensitivity of a theory’s conclusions to minor alterations in its assumptions. The predictions of the Heckscher-Ohlin theory on the pattern of trade are not very sensitive; altering the assumption of constant returns to scale to include increasing returns to scale industries does not dramatically affect the pattern of specialization. The composition and pattern of goods trade, for instance, is no longer fully determined by factor endowments, but it is possible to predict net trade in factor services as embodied in goods, which in practice is very similar to the traditional predictions of Heckscher-Ohlin. Under increasing returns, capital-rich countries specialize in and export goods that intensively use the services of capital. The converse holds for labor- and land-abundant countries. See Elhanan Helpman and Paul R. Krugman, Market Structure and Foreign Trade: Increasing Returns, Imperfect Competition, and the International Economy (Cambridge: MIT Press, 1985).
19See Avinash Dixit, “Strategic Aspects of Trade Policy,” mimeo, Princeton University, January 1986; and Gene M. Grossman and J. David Richardson, “Strategic Trade Policy: A Survey of Issues and Early Analysis,” in Robert E. Baldwin and J. David Richardson, eds., International Trade and Finance, 3rd ed. (Boston: Little, Brown, 1986), pp. 95–114.
20See Dixit, “Strategic Aspects of Trade Policy”; and Harry G. Johnson, “Optimum Tariffs and Retaliation,” Review of Economic Studies 21 (1954): 142–53.
21Johnson, “Optimum Tariffs”; Raymond Riezman, “Tariff Retaliation from a Strategic Viewpoint,” Southern Economic Journal 48 (January 1982): 583–93; and John Whalley, Trade Liberalization among Major World Trading Areas (Cambridge: MIT Press, 1985), p. 239.
22John Maynard Keynes, The General Theory of Employment, Interest and Money (New York: Harcourt, Brace and World, 1964), chap. 23.
23Ibid., p. 339; Michal Kalecki, Selected Essays on the Dynamics of the Capitalist Economy, 1933–1970 (Cambridge: Cambridge University Press, 1971), pp. 15–25; Joan Robinson, Contributions to Modem Economics (Oxford: Blackwell, 1978), pp. 201–12; J. B. Burbidge, “Post-Keynesian Theory: The International Dimension,” Challenge, November—December 1978, pp. 40–45; Hans O. Schmitt, “Mercantilism: A Modern Argument,” Manchester School of Economic and Social Sciences 47 (June 1979): 93–111; and David Vines, “Competitiveness, Technical Progress and Balance of Trade Surpluses,” Manchester School of Economics and Social Studies 48 (December 1980): 378–91.
24This debate is summarized and extended in Vines, “Competitiveness.”
25Miltiades Chacholiades, “Increasing Returns and Comparative Advantage,” Southern Economic Journal 37 (October 1970): 157. See also Nicholas Kaldor, Economics without Equilibrium (Armonk: M. E. Sharpe, 1985), p. 70.
26Johan Galtung, “A Structural Theory of Imperialism,” Journal of Peace Research 8, 2 (1971): 81–109.
27Kaldor, Economics without Equilibrium, p. 69.
28Helpman and Krugman Market Structure and Foreign Trade, esp. chap. 6. Edward E. Leamer, Sources of International Comparative Advantage: Theory and Evidence (Cambridge: MIT Press, 1984), provides an empirical discussion of the factor similarities between developed countries.
29“Industry” here primarily refers to production techniques. Automobiles, for instance, can either be made largely “by hand” or mass produced. Obviously, only the latter method is characterized by substantial increasing returns.
For a discussion of the role of positive externalities, the antebellum cotton textile industry, and trade policy see Paul A. David, “Learning by Doing and Tariff Protection: A Reconsideration of the Case of the Ante-Bellum United States Cotton Textile Industry,” Journal of Economic History 30 (September 1970): 521–601; Jeffrey G. Williamson, “Embodiment, Disembodiment, Learning by Doing, and Returns to Scale in Nineteenth-Century Cotton Textiles,” Journal of Economic History 32 (September 1972): 691–705; and Paul A. David, “The Use and Abuse of Prior Information in Econometric History: A Rejoinder to Professor Williamson on the Antebellum Cotton Textile Industry,” Journal of Economic History 32 (September 1972): 706–27.
30The argument in this chapter is developed only for cross-national comparisons at a single moment in time. The case study in Chapters 3 through 6 employs both crossnational and longitudinal comparisons.
31The literature on increasing returns to scale and trade is fairly large. See, among others, R.C.O. Matthews, “Reciprocal Demand and Increasing Returns,” Review of Economic Studies 17, 2 (1949–50): 149–56; Jan Tinbergen, International Economic Integration, 2d ed., rev. (Amsterdam: Elsevier, 1954), Appendix 2; Murray C. Kemp, The Pure Theory of International Trade (Englewood Cliffs: Prentice-Hall, 1964), chap. 8; T. Negishi, “Marshallian External Economies and Gains from Trade between Similar Countries,” Review of Economic Studies 36 (January 1969): 131–35; Raveendra Batra, “Protection and Real Wages under Conditions of Variable Returns to Scale,” Oxford Economic Papers 20 (November 1968): 353–60; and Arvind Panagariya, “Variable Returns to Scale in General Equilibrium Theory Once Again,” Journal of International Economics 10 (November 1980): 499–526.
32See Helpman and Krugman, Market Structure and Foreign Trade; and Paul R. Krug-man, ed., Strategic Trade Policy and the New International Economics (Cambridge: MIT Press, 1986).
33This theory is intentionally parsimonious. Relative size and relative labor productivity are the common dimensions underlying the several international trade theory literatures discussed above. I have purposely excluded other possible explanatory factors from the theory, including national security considerations, macroeconomic conditions, trade patterns, international monetary and financial regimes and processes, and levels of trade dependence.
34The reasons for the overlap between spoilers and protectionist free riders, on one hand, and free trade free riders, on the other, are made clear below. They are distinguished by the size of the domestic market, a nonsystemic attribute which is nonetheless essential for understanding their systemic trade preferences. The categories are arrayed here, however, only along their systemic dimensions.
35Kindleberger, World in Depression, dates the period of potential American hegemony from the mid- to late-1920s, but he is primarily concerned with monetary and financial power. Any level of trade that would classify the United States as hegemonic during this period would be theoretically meaningless. See Table 1.1.
36Ibid.; Robert Gilpin, U.S. Power and the Multinational Corporation: The Political Economy of Foreign Direct Investment (New York: Basic Books, 1975); Stephen D. Krasner, “State Power and the Structure of International Trade,” World Politics 28 (April 1976): 317–47; and the other sources cited in the Introduction.
37Robert O. Keohane and Joseph S. Nye, Power and Interdependence: World Politics in Transition (Boston: Little, Brown, 1977), pp. 63–162.
39See Charles Lipson, Standing Guard (Berkeley: University of California Press, 1985).
40Robert O. Keohane, After Hegemony: Cooperation and Discord in the World Political Economy (Princeton: Princeton University Press, 1984).
41See Polanyi, Great Transformation, for a classic statement of this assumption.
42Constructing a theory of instability is beyond the scope of this study. Such a theory would need to specify the causes of instability and identify threshold effects of instability on the relevant political variables. I do neither here. In Chapter 5 the issue of instability becomes central to the analysis, and I compare price and exchange rate fluctuations in the eras before and after World War I. Prices and exchange rates are the principal determinants of international trade patterns and so are highly appropriate indicators. Yet one can only conclude that fluctuations were greater after 1919 than before.
43John A. C. Conybeare, “Public Goods, Prisoners’ Dilemmas, and the International Political Economy,” International Studies Quarterly 28 (March 1984): 8–9, argues that free trade is not a collective good. I agree. The argument put forth here focuses only on the intermediate product of international infrastructure and not the final product of free trade.
44Mancur Olson and Richard Zeckhauser, “An Economic Theory of Alliances,” Review of Economics and Statistics 58 (August 1966): 266–79; Olson, The Logic of Collective Action (Cambridge: Harvard University Press, 1971).
45Kindleberger, World in Depression, pp. 299–300.
46On the problem of enforcement costs see Beth V. Yarbrough and Robert M. Yarbrough, “Free Trade, Hegemony, and the Theory of Agency,” Kyklos 38 (1985): 348–64; and “Cooperation in the Liberalization of International Trade: After Hegemony, What?” International Organization 41 (Winter 1987): 1–26.
47Duncan Snidal, “The Limits of Hegemonic Stability Theory,” International Organization 39 (Autumn 1985): 579–614, makes a parallel argument.
48For a discussion of the concept of regulation, see Richard N. Rosecrance, Action and Reaction in World Politics: International Systems in Perspective (Westport, Conn.: Greenwood, 1977), pp. 220–27.
49Peter A. Gourevitch, “International Trade, Domestic Coalitions, and Liberty: Comparative Responses to the Crisis of 1873–1896,” Journal of Interdisciplinary History 8 (Autumn 1977): 281–313. Gourevitch dismisses international structural explanations in this article because Britain’s power position had not significantly declined. He does not develop the concept of regulation used here.
50Conybeare, “Public Goods”; and Trade Wars: The Theory and Practice of International Commercial Rivalry (New York: Columbia University Press, 1987), chap. 2.
51Whalley, Trade Liberalization, estimates that postretaliation tariffs for the United States, European Economic Community, and—to a lesser extent—Japan “may well be higher than 50 percent” and notes that tariff levels during the early 1930s might actually have been “optimal” (p. 246). Given the welfare losses associated with the Smoot-Hawley and other tariffs during this period, this estimate clearly appears too high.
52Burbidge, “Post-Keynsian Theory,” p. 41.
53Joan Robinson, quoted in ibid., p. 42.
54Technically, any reduction in imports must reciprocally reduce exports, but for small countries the effect of a reduction in imports is so diffuse that we generally consider exports to be exogenous. This is not so for large countries.
55Joan Robinson, Aspects of Development and Underdevelopment (Cambridge: Cambridge University Press, 1979), p. 103.
56John Zysman and Laura Tyson, eds., American Industry in International Competition: Government Policies and Corporate Strategies (Ithaca: Cornell University Press, 1983).
57Kaldor, Economics without Equilibrium, pp. 72–75; Gunnar Myrdal, Economic Theory and Under-developed Regions (London: Duckworth, 1957), pp. 23–38; Albert O. Hirschman, The Strategy of Economic Development (New Haven: Yale University Press, 1958), pp. 183—90; and Arghiri Emmanuel, Unequal Exchange (New York: Monthly Review Press, 1972).
58I recognize that trade restrictions are a “second best” policy for stimulating increasing returns protection. Because of the presumed positive externalities, social returns are greater than private returns and some government intervention is necessary to stimulate production. Subsidies might be the optimal intervention instrument, but tariffs or other trade restrictions have been more commonly used. Several reasons can be suggested for this paradox. Tariffs are a diffuse indirect tax on consumers and a source of revenue for the government. Subsidies are a direct payment to producers and therefore more transparent and a drain on government revenues. The diffuse and opaque nature of trade restrictions makes them less likely to generate political opposition from nonbeneficiaries. Also, because trade restrictions alter market incentives for all relevant entrepreneurs and subsidies are direct “rewards” to specific producers, the former are typically perceived as less “interventionist” and more consistent with a policy of laissez-faire than the latter. In this case, however, the reality may be quite different. Trade restrictions, at least for small and middle-sized states, expand tendencies toward trade surplus, further reinforcing the virtuous cycle discussed above. Finally, states vary in the policy instruments available to them (see Katzenstein, ed., Between Power and Plenty). By their definition (see Chapter 2), however, all states possess the ability to regulate interactions with foreigners, even though they may lack the more refined and sectorally specific instruments, such as subsidies, available to other, “stronger” states.
59See Kaldor, Economics without Equilibrium, pp. 65–67, on the importance of market size for increasing returns industries.
60In several earlier articles I referred to this category of nation-states as “supporters.” As the theory evolved, this label became something of a misnomer and a source of confusion. The term “opportunist” better reflects the exploitive behaviors expected of middle-sized, relatively productive nation-states. When a hegemonic leader is present they free ride; when two or more exist they restrain protectionism in one another; when only one exists, it may precipitate the closure of the system.
61Christopher Chase-Dunn, “International Economic Policy in a Declining Core State,” in William P. Avery and David P. Rapkin, eds., America in a Changing World Political Economy (New York: Longmans, 1982), argues that the Netherlands was a hegemonic power in the seventeenth century. Insufficient data exist to classify the Netherlands within the dimensions of the international economic structure specified here. For this reason, the examples and analysis are confined to the post-Industrial Revolution period.
62Readers of earlier drafts of this chapter have suggested that the Soviet Union might be classified as an imperial leader within the Communist bloc trading system. If the Soviet bloc is considered as an autonomous subsystem, the Soviet Union would appear to fit both the structural definition and the policy predictions for an imperial leader. The Soviet bloc is not an autonomous subsystem, however, but a grouping of countries only loosely integrated into the larger international economy. Although it may be useful to apply the theory developed here at the regional level, the theory’s ability to generate testable hypotheses collapses when it is recognized that regions are also situated within the global system and that trade policy is made with consideration to the larger economy. I prefer, therefore, to limit the applicability of the theory to the international economy as a whole. In this approach, the Soviet Union is classified as a protectionist free rider.
63Domestic market size is best measured by gross national product, population, or other such aggregate indicators. The absence of a precise definition and operationalization does not create significant problems for this analysis because the concept does not play a role in the case study of Part II.
64Robert Gilpin, War and Change in World Politics (Cambridge: Cambridge University Press, 1981), makes similar distinctions between types of international change; see pp. 39–49.
65The data series on relative labor productivity used for this book ends in 1977 (see Tables 1.1 and 1.2). Projecting from existing trends and drawing upon productivity growth rates available from other sources, it appears that the United States and the Federal Republic of Germany are securely placed as opportunists within the international economic structure. France’s position is more difficult to assess; it appears to have maintained its slightly higher than average relative labor productivity and, therefore, its position as an opportunist.
66Arthur A. Stein, “The Hegemon’s Dilemma: Great Britain, the United States, and the International Economic Order,” International Organization 38 (Spring 1984): 355–86; and Timothy J. McKeown, “Hegemonic Stability Theory and Nineteenth Century Tariff Levels in Europe,” International Organization 37 (Winter 1983): 73–91. Both argue that there is little evidence that either Britain or the United States acted in the manner predicted here. McKeown specifically notes the lack of evidence that Britain effectively manipulated other countries to secure free trade. For a counterargument, see Scott C. James and David A. Lake, “The Second Face of Hegemony: Britain and the American Walker Tariff of 1846,” paper presented to the Conference Group on Political Economy, Chicago, Illinois, September 3–6, 1987.
67Iliasu, “Cobden-Chevalier Commercial Treaty.”
68Michael Mastanduno, “Between Economics and Security: The Western Politics of East-West Trade” (Ph.D. diss., Princeton University, 1985), examines the relationship between America’s economic and political hegemony in the post-World War II era.
69Bilateral and multilateral opportunism are identical except in the following manner: two-person iterated prisoner’s dilemma can potentially lead to cooperation in the northwest cell of the matrix. N-person prisoner’s dilemma eventually breaks down with the introduction of the possibility of free riding. Although the dynamics of the game change, it may not constitute a practical problem for the circumstances under discussion here. The number of opportunists never has been and never will be very large. Some free riding may go undetected, but most will be caught and punished. Thus, some cooperation may occur, although it will be less stable than under bilateral opportunism. For a general discussion of N-person games see Morton D. Davis, Game Theory: A Nontechnical Introduction, rev. ed. (New York: Basic Books, 1983), pp. 163–228.
70Robert Axelrod, The Evolution of Cooperation (New York: Basic Books, 1984). In “tit-for-tat” a player cooperates on the first move and then does whatever his or her opponent did on the previous move. Thus it reciprocates cooperation, punishes defection, and forgives.
71Russell Hardin, Collective Action (Baltimore: Johns Hopkins University Press, 1982), pp. 125–230.
72Arthur A. Stein, “Coordination and Collaboration: Regimes in an Anarchic World,” International Organization 36 (Spring 1982): 312; Robert Keohane, “The Demand for International Regimes,” International Organization 36 (Spring 1982): 325–55; and After Hegemony. See Stephen D. Krasner, “Regimes and the Limits of Realism: Regimes as Autonomous Variables,” in Krasner, ed., International Regimes, pp. 355–68.
73Gilpin, U.S. Power, p. 260; see also Stein, “Hegemon’s dilemma.”
74E. J. Hobsbawm, Industry and Empire: The Making of Modem English Society, Vol. 2, 1750 to the Present (New York: Pantheon, 1968), pp. 110–26.
75Stephen D. Krasner, Defending the National Interest: Raw Materials Investments and U.S. Foreign Policy (Princeton: Princeton University Press, 1978), has partially resuscitated the use of the term “national interest.” This is a positive development for Realist international relations and international political economy. Yet Krasner defines the national interest inductively, thereby limiting the usefulness of his approach. Having worked backward from policy to establish interests, Krasner cannot then use the concept of the national interest to explain policies. To his credit, Krasner avoids this tautology. This book posits national interests deductively and can, as a result, use them to explain policy.
76Thus I cannot claim to have identified the national trade interest but only a systemically derived goal referred to here as the national trade interest.