publisher colophon

Chapter Two

Structure, the State, and Trade Strategy

The systemic theory set forth in Chapter 1 identifies the constraints and opportunities of the international economic structure. It argues that this structure creates a systemically “best” trade strategy, referred to as the national trade interest, for all countries. Given the need for national survival in an anarchic and competitive international environment, the conclusion posits a primacy for systemic concerns and argues that the trade strategies of nation-states will generally conform to the constraints and opportunities of the international economic structure.

This theory posits that countries have a strong stimulus to follow the incentives of the international system, but it is mute on the domestic process or mechanism by which these constraints and opportunities are communicated and translated into specific policies. The theory of international economic structure, like the Realist and neo-Realist traditions from which it is drawn, treats domestic politics as a “black box” in which a rational and unitary state exists. This chapter attempts to elucidate the insides of the black box to identify the agent and process by which the constraints and opportunities of the international economic structure are recognized and acted upon within the domestic political sphere so as to result in observable trade strategies.

In this chapter, I present a simplified conception of the interaction of the domestic and international political systems. I posit a state inclusive of the central government, functionally differentiated in its parts, and differing in the degree of autonomy possessed by its constituent parts. Following from this conception, I isolate two categories of actors within the state : the representative element, which serves as the basic link of the state to society and, as a result, primarily reflects the interests of society, and the foreign policy executive, which seeks to promote the power and wealth of the nation-state within an anarchic international system. Because of its concern with national power and wealth, the foreign policy executive, I argue, is particularly sensitive to the constraints and opportunities of the international economic structure and acts as a conduit through which these systemic incentives pass into the sphere of domestic politics. The foreign policy executive cannot act unilaterally, however, and must bargain with the politically mobilized groups within society and the representative element of the state. This bargaining process, in turn, is influenced by the structure of the state, which creates a particular pattern of political action which endures over time, and by executive or presidential leadership. The second half of this chapter provides an overview of the domestic political process in the United States in the period 1887–1939.

The International System, the State, and Domestic Politics

Following Max Weber, the state is defined as a “compulsory organization with a territorial basis” which monopolizes the legitimate use of force.1 As such, it is “a relation of men dominating men”2 and the sole source of authoritative decisions within a society. In short, the state is a governing apparatus which is distinct from and superordinate to society.

The state-society distinction embedded in this definition is useful because it emphasizes the hierarchical nature of government. Hierarchy exists within the state and between the state and society. Unlike pluralist conceptions of politics, which find the origins of all government actions in societal demands,3 this approach allows for a more independent role for the state in the formulation of public policy. If societal approaches explain policy from the demand side, employing the state-society distinction allows for an investigation of the supply side.4

Yet the state itself does not exist in actuality. Like the very concept of a “system” central to the argument of Chapter 1, the state is merely an analytic notion, a useful fiction. Depending upon the larger theoretical context in which it is embedded, the concept of the state can vary along three interrelated and often confused dimensions. The state can be limited to central decision makers or nearly synonymous with government. It can be an instrument of society, relatively autonomous from society, or even wholly autonomous. And the state can be understood as a unified actor with a single set of interests, or an actor composed of hierarchically arranged but functionally differentiated parts possessing individual interests. The state, in this book, is assumed to be inclusive of the central government, functionally differentiated in its parts, and differing in the degree of autonomy obtained across its constituent parts.

The state has been conceived as both narrow, limited to only the central decision makers of the government who are relatively well insulated from domestic political pressures (the president and the secretary of state in the United States),5 and extensive, as broad as the government.6 These two perspectives appear to derive from different interpretations of the authoritative nature of the state. The first conception defines authoritative as “imposed,” or not merely reflective of the interests of any group in society. Here, the composition of the state is problematic: only those decision makers who are or can be autonomous are included. Thus the state can vary in size and membership over time and across issue areas.

In this book I adopt an extensive definition of the state in which authoritative is defined as “binding”; decisions are legitimately enforced by the coercive power of the state whether or not they reflect the interests of some group(s) in society. In this conception, all parts of the governing apparatus—including the legislature and executive agencies highly permeated by societal groups—are seen as participating in the making of authoritative or legally binding political decisions. Here, the autonomy of the state is problematic: rather than equating autonomous decision makers and the state, this conception recognizes that states may differ in their degree of autonomy.

Although scholars have not commonly done so, this conception of the state also allows for it to be disaggregated into its various but nonetheless hierarchically arranged parts.7 The foreign policy bureaucracy, for instance, may possess a different mandate and purpose than the agricultural bureaucracy, but both are responsible to the chief executive. Similarly, the legislative and executive functions are conceptually distinct in nearly all countries, although the organizational relationship between them may differ cross-nationally. By disaggregating the state, this conception also allows for differences in autonomy across the various parts of the state, depending upon the function each performs. In nearly all advanced industrialized democracies, for example, the legislature is typically the least and the central bank the most independent component.

In the analysis of trade policy, the state can be simplified into two basic categories: the representative and the foreign policy elements.8 The representative element includes the legislature, which serves as the principal link of the state to society, and the “constituent” agencies, such as the Departments of Agriculture, Commerce, and Labor in the United States. Although in some countries the legislature serves only to ratify and legitimize decisions made in other parts of the state, its more important function is to aggregate and channel societal interests into the political decision-making process. It is assumed here that legislators are primarily motivated by the desire for reelection and are therefore responsive to societal demands.9 Thus individual members of the legislature, in one form or another, represent constituencies organized on a geographic basis. If the legislature is organized into substantive committees, these subunits will also represent functional societal interests.

Constituent agencies serve a similar function as the legislature. Possessing narrow institutional mandates, these agencies are easily “captured” by the interests they are designed to serve.10 Capture can occur directly, through the appointment of interested personnel, or indirectly, as decision makers come to identify their own career interests and success with the well-being of their constituents. As the principal link between state and society, the representative elements are the least autonomous parts of the state. Indeed, they can be understood as merely reflecting the interests of society.

It is assumed here that individuals, who ultimately constitute society, pursue their material interests, defined as the maximization of their economic well-being. As these interests are pursued, society and the representative state elements are dominated by the politically mobilized groups within society. Latent or unformed groups can influence the political process only with great difficulty. Their one point of access to the political arena is the direct election of legislators, but it is nearly impossible to signal support for specific policy positions through them except perhaps in economically homogeneous electoral districts (such as rural agricultural areas). As Mancur Olson and others have demonstrated, groups have differing incentives to form and mobilize for political action. Small, homogeneous, and geographically concentrated groups, for instance, are more likely to form than large, variegated, and geographically dispersed groups. As a result, individuals and the groups they form possess differential access to the policy-making arena.11 The representative elements of the state, in other words, are not truly “representative” of all societal interests. Rather, they principally reflect the interests of only those manifest groups that have successfully overcome problems inherent in taking collective action.

The foreign policy executive constitutes a second component of the state. Defined as executive officials, who typically face a national electorate, and high-ranking bureaucrats charged with the overall conduct of defense and foreign affairs, the foreign policy executive sits at the intersection of the domestic and international political systems and regulates the interaction between the two.12 Most important, the foreign policy executive is the sole authoritative maker of foreign policy and the only national actor mandated to preserve and enhance the position of the nation-state within the anarchic and competitive international system. It is charged, in other words, with husbanding the nation-state’s wealth and power given the interests and actions of other countries.13 It is this unique position of the foreign policy executive which renders it particularly sensitive to the national trade interest and, in turn, to the international economic structure that shapes this interest.

The societal pressure brought to bear on the policy-making process through the representative element of the state must differ from the national trade interest of the foreign policy executive for three reasons. First, societal interests cannot aggregate into the national trade interest as defined here. A country’s position within the international economic structure is determined by economic and political aggregates for the nation-state as a whole. All industries and producers are reflected in the measure of relative labor productivity and all importers and exporters are summed into the measure of relative size (see Chapter 1). Thus if problems inherent in collective action exist, as they surely do, and only some groups mobilize or become manifest, there is no reason to assume that the “bottom-up” interests of society will be identical to the “top-down” national trade interest of the foreign policy executive. Indeed, the greater the problems of collective action within society, the more these two interests must diverge.

Second, producer groups possess relatively narrow interests. They support protection when facing competition from imports in their own markets and oppose it only if threatened with retaliation against their products abroad. Groups have little incentive to oppose protection on their own products if another industry is likely to bear the costs of foreign retaliation. In the pursuit of national wealth and power, and in responding to its national rather than regional electorate, the foreign policy executive must take these trade-offs into account and make judgments about what is good for society as a whole. It must also make appropriate choices on the means to obtain these goals in the face of opposition or resistance from foreign national and state actors.

Third, to the extent that the executive’s national trade interests are shaped by considerations of relative advantage over other countries, as might be expected within an anarchic and competitive international environment, the interests of the representative and foreign policy elements of the state must also diverge. In the pursuit of material interests, no group in society—even encompassing coalitions—has any incentive to maximize the relative resources or power of the nation-state. Group interests may, at times, complement this power interest, but they possess very different roots.

This is not to argue that the executive is entirely free from societal constraints. Presidents and prime ministers must periodically stand for election. Yet whereas the representative element of the state can be best understood as acting in the interests of society, to use Vilfredo Pareto’s famous distinction, the executive acts in the interests for society.14 The executive is responsible to all of society, charged with responsibility for foreign affairs, and, as a result, specifically concerned with the constraints and opportunities of the international economic structure.

Because competing trade interests exist within the domestic political arena, the foreign policy executive will rarely be able to translate its systemically derived policy preferences directly and unilaterally into trade strategy. In few countries is trade policy entirely within the purview of the foreign policy executive. Trade strategy affects society, and the representative element of the state can be expected to block or at least partially undermine foreign policy initiatives. Consequently, foreign policy leaders are dependent upon the support or at least the acquiescence of society and the representative element of the state. Support from society, and particularly from politically mobilized groups within society, is needed even in the most totalitarian countries. Coercion can substitute for societal consent, but it becomes extremely costly and decreasingly effective at high levels of state-society divergence. Such conflict will also be reflected in the state, as the representative and foreign policy elements split along functional lines, thereby rendering the effective use of coercion by the state problematic.

For the constraints and opportunities of the international economic structure to be transformed into public policies, they must pass through a bargaining process between the foreign policy executive and the politically mobilized groups in society as manifested in the representative element of the state. Resolution of the conflict between these sets of interests—whether in favor of one, the other, or both—is ultimately determined by many contextual factors. Two intervening variables, however, are important in the case study discussed in Part II.

Most fundamentally, the bargaining process is influenced by the distribution of authority within the state as codified into existing laws and institutions and referred to here as state structure. This structure forms a continuum from near anarchy, in which competing centers of political authority vie for leadership as in present-day Lebanon, to totalitarianism, in which almost all forms of political control and influence are centralized in the highest reaches of the state. Although this continuum is not open to precise gradations, differences across countries can nonetheless be found. Among the advanced industrialized democracies, Japan and France possess relatively centralized states and the United States a relatively decentralized state.15 State structure may also vary by issue area. Because trade affects groups differentially and has direct implications for national power, many societal and state actors will be involved, and authority is likely to be torn between the representative and foreign policy elements of the state. Monetary policy, on the other hand, has relatively symmetrical effects and does not mobilize social groups into the political process to the same extent as does trade. Thus the overall autonomy of the state will be higher and authority will be more concentrated.16

The structure of the state does not necessarily determine the outcome of the bargaining process between the representative and foreign policy elements. But by specifying which elements of the state possess authority over an issue and which actors can legitimately be involved in the political process, the structure of the state does create a set of constraints within which the bargaining process occurs. As argued below, the decentralized structure of the American state and the constitutional delegation of authority over international commercial policy to Congress magnified the importance of society and the representative element of the state in the United States during the nineteenth and early twentieth centuries. The principal task for the foreign policy executive was to gain legitimacy in and access to the trade policy-making arena. The state’s structure also conditions the bargaining strategies and resources open to the representative and foreign policy elements. The foreign policy executive, for instance, can use its position at the intersection of the domestic and international political systems to redefine issues and build transnational coalitions to bind the nation-state. It can also mobilize new or existing societal groups with complementary interests into the political system so as to gain access to the representative element of the state.

At a more proximate level, presidential or executive leadership is also important in explaining the outcome of the bargaining process between the representative and foreign policy elements. Executives bring to office differing conceptions of appropriate executive-legislative roles and varying degrees of political acumen. Although it is difficult to generalize about this idiosyncratic factor, a president who has a strong view of his policy-making role or highly developed political skills is clearly more likely to obtain his wishes when faced with legislative opposition.

Domestic Politics and American Trade Strategy, 1887–1939


For an investigation of trade strategy, society can be conceptually divided into four broad and not necessarily homogeneous or exclusive groups: manufacturers (capitalists and workers), financiers, farmers, and consumers.17 Measuring the protectionist or free-trade inclination of groups is always difficult. Nonetheless, it would appear that the domestic political position and interests of each group changed in important ways over the period 1887–1939.

The political importance of industry grew dramatically over the era. In 1890, 65 percent of the United States population resided in rural areas where little manufacturing activity occurred. By 1930, rural dwellers had declined to 44 percent of the population. Similarly, manufacturing’s share of national income increased from 18.2 percent between 1889 and 1899 to a high of 21.9 percent between 1919 and 1929. Although not direct measures, these figures do indicate the growing political strength of industrial interests in the United States.

Export dependence provides a useful proxy for measuring the trade interests of industry.18 Any industry with substantial markets abroad is likely to favor free trade for two reasons: in the absence of extraordinary export subsidies it is at least competitive with foreign rivals, and it may fear retaliation abroad for protection at home. Levels of export dependence for all of American manufacturing, disaggregated into fifteen sectors, are presented in Tables 2.1 and 2.2.

In 1889, the first census year in the period studied here, only the chemicals sector, which made up 4.5 percent of American manufacturing by value, exported more than 10 percent of its production (see Table 2.2).19 The largest manufacturing sector of the economy, food and beverages, was moderately dependent, exporting 9.7 percent of its output. All other sectors exported less than 5 percent of their products. By 1899, two years after the passage of the protectionist Dingley Act, moderately export-dependent sectors more than doubled in value to 52.9 percent of American manufacturing. Although chemicals remained the only highly export-dependent industry, food and beverages now exported 9.3 percent of its output, nonferrous metals 9.2 percent, iron and steel 6.8 percent, and miscellaneous industries—covering everything from agricultural implements to dental goods, to glue, musical instruments, and toys—5.6 percent. In 1909, no sector exported more than 10 percent of its output. Chemical exports as a proportion of total output declined, and chemicals joined the four sectors just noted and leather in the moderately export-dependent category. These industries now constituted approximately 63.6 percent of American manufacturing. Thus over the period 1889 to 1909, American manufacturers became relatively more export-dependent and, by implication, liberal, although the movement in this direction after the turn of the century was not strong.

As a result of increased foreign demand in wartime, the export dependence of American manufacturers expanded. In 1919, 25.2 percent of United States industry exported between 5 and 10 percent of its output and 36.8 percent exported more than 10 percent of its production. Because the ratio of low-export-dependent industries remained relatively constant, it is clear that the dramatic increase in highly export-dependent industries came largely at the expense of the middle category. Industries that did not export significantly before the war were similarly disadvantaged during and after the hostilities. By 1929, after a decade of relative domestic prosperity, levels of export dependence had fallen to more traditional ranges and had receded below their prewar marks. Only petroleum, which until 1919 was not important enough to warrant a separate census category, exported more than 10 percent of its output, and vehicles, machinery, rubber, nonferrous metals, and forest products (largely manufactures of wood) exported more than 5 percent of their production.

Table 2.2. Export dependence (XD) of American manufacturing, by sector and value of manufacturing (VM)

This trend toward declining export dependence and, in turn, liberalness, was partially offset by rising foreign direct investment by American industry, which took hold in the late nineteenth century and accelerated rapidly after World War I. In 1929, the leading foreign investors (foreign direct investment divided by book value of fixed capital, by sector) were in machinery and equipment (23.3 percent), mining and petroleum (17.7), motor vehicles (14.9), and rubber products (13.8).
20 Although the categories differ slightly, the similarity between this list and that of the export-dependent industries in Table 2.2 is striking. The most important foreign investors were also the most export-dependent industries. As a result, it is possible to infer that increased foreign direct investment did not create new interests in trade liberalization, but most likely served to reinforce the liberal tendencies of the already export-dependent sectors. Thus the essential political cleavage between sectors was not affected, but the commitment to a liberal trade policy was probably strengthened for sectors that both invested in and exported to foreign markets.

From these trends in the industrial structure of the United States, it is possible to infer that the underlying need and demand for trade protection by American manufacturers gradually eased over the period examined here. Export dependence rose until 1919, and its later decline was partially offset by increased foreign direct investment. As more industries exported and later invested abroad, they reflected the new underlying competitive strength of the American economy and developed important cross-cutting cleavages in trade policy.

The dominant political pressures on Congress nonetheless remained protectionist. The statistics on export-dependence just cited do not capture collective action problems. Both Joan Hoff Wilson and William H. Becker, in their widely cited studies of the role of business in American politics during this period, confirm the already well-accepted observation that protectionist producers organize more readily than free-trade producers or consumers. Both Wilson and Becker find that smaller and less internationally competitive industries were more politically active on the tariff than were the larger, more internationally competitive sectors, which tended to become involved in tariff making only when their interests were directly threatened.21 Although the underlying demand for protection within the United States may have eased slightly over the period, the protectionists still sang louder than others in the chorus serenading Congress.

Before World War I, the interests of American finance were closely linked to those of industry and, by implication, were largely protectionist. Finance began to expand its international lending decades before, but these foreign activities remained relatively small compared to domestic operations. During and after World War I, however, American finance greatly expanded its role in the international economy and developed interests separate from those of American industry as a whole. As a net creditor after the war, the United States played a major role in international financial markets. American banks now possessed an interest in Europe’s ability to pay, which necessitated lower tariffs in the United States and a reversal in America’s traditional balance-of-trade surplus. In advocating a more liberal trade strategy, American finance allied itself not with industry as a whole but only with the largest and most internationally oriented segments of the industrial community.22

As American industry and finance became more internationalist over time, agriculture drifted toward protectionism.23 Southern farmers, the world’s premier growers of cotton and tobacco, feared few competitors and generally supported free trade throughout the late nineteenth and early twentieth centuries. As high-cost producers of wool, Northeastern and Midwestern sheep growers benefited from protection and supported the policy (see Chapter 3). The interests of most farmers, however, were more complex and mixed.

In the early phases of the period studied here, farmers sold most of their products in the home market and were largely indifferent or slightly favorable to protection, believing—perhaps not without merit—that high tariffs provided a spur to domestic economic growth. After 1902, however, the ranks of farmers began to split as the “Iowa idea” gained widespread attention. Because farmers sold their marginal products in the unprotected international market, which thereby determined prices in the domestic market as well, and bought their goods in the protected home market, the Iowa idea convinced many farmers that they were actually suffering rather than benefiting from high tariff policies—leading some Republican legislators from the Midwest to vote against the party’s Payne-Aldrich Act of 1909.

By 1921, farmers had reversed their position and become generally supportive of protection. After World War I, agricultural prices fell and stocks rose dramatically. The new position of the United States as a net creditor and the confusion over war debts and reparations limited the ability of Europeans to import American agricultural goods. And as European farmers returned to production, facilitated by renewed protection, prices fell (see Chapter 5). As Murray R. Benedict notes, “By the spring of 1921, American agriculture found itself in a more unfavorable position than it had experienced at any time in the memory of men then living, or possibly at any time since the nation’s beginnings.”24 Yet conditions grew worse. This prolonged crisis altered the historic position of American agriculture in the nation’s trade balances. A major contributor to the national trade surplus before 1919, farm imports exceeded farm exports on a regular basis after the mid-1920s.25

Protection now offered a real advantage to agriculture, and farmers supported it wholeheartedly, demanding and receiving increased protection for their products in 1921, 1922, and 1930. As the logrolling found by E. E. Schattschneider in the making of the Smoot-Hawley Tariff of 1930 demonstrates, farmers were increasingly willing to offer their support to manufacturers in return for increased tariffs on farm commodities.26

Nearly all Americans were both producers and consumers in the period 1887–1939, although few identified themselves with the latter. Protectionists consistently emphasized the benefits of higher profits and wages for producers. Tariff reformers noted that the tariff led to higher prices and, despite higher wages, lower real incomes. Even though the second position is paramount today, the protectionists were clearly winning the rhetorical war before World War II.

When Louis D. Brandeis appeared before the House Ways and Means Committee in 1897 as a “representative of the consumers,” he was jeered by the audience. Yet the emergent middle class, led by Brandeis and other progressive reformers, soon mobilized—albeit weakly—consumer interests into the tariff-making process. As Walter Lippmann wrote in 1914, “We hear a great deal about the class-consciousness of labor… . My own observation is that in America today consumers’-consciousness is growing very much faster.” This consumer consciousness was driven by the rapidly expanding “new middle class,” composed of technicians and salaried professionals, clerical workers, salespeople, and public service personnel. Whereas the population as a whole increased by 230 percent between 1870 and 1910, this new middle class grew by almost 800 percent, from 756,000 to 5,609,000 people.27

Despite their consumer consciousness and in part because of their rapidly growing numbers, these new white-collar workers were difficult to mobilize, and they exerted little direct impact on the producer-dominated and relatively closed congressional tariff-making process. The one exception occurred in 1913, when Woodrow Wilson made a strong appeal to the public in general and his progressive supporters in particular to offset the influence of the tariff lobby (see Chapter 5).

It is virtually impossible to aggregate the political influence of these four conceptual groups within society and reach a definitive statement on the balance between the pressures supporting free trade or protection exerted on the tariff-making process and manifested in Congress, the principal representative agency of the state. Nonetheless, several tentative conclusions may be suggested. Industry and finance gradually became more internationalized and liberal over time, although protectionists continued to be politically dominant in the former. Although its interests were mixed beforehand, agriculture clearly moved in a protectionist direction after 1919. Consumers emerged as a political force after the turn of the century and supported a policy of freer trade, but their influence in the halls of Congress was relatively minor. In sum, protectionist pressures, at the very least, were mixed and most likely did not expand in importance during the period 1887 to 1939. Although societal interests may have become more liberal over time, protectionist interests continued to exert a disproportionate influence because of the inherent difficulties of mobilizing latent free-trade-oriented manufacturers and consumers. Thus, although the predictions of the interest-group model are not particularly clear, there seems to be little reason to expect significant shifts in trade strategy during this period. Specific anomalies for an explanation of trade strategy based on interest groups are discussed in Part II.

Societal factors are most important in this analysis as a counterweight to the national trade interest derived from the international economic structure. As discussed in Chapter 1, the United States remained an opportunist throughout the period examined here. The national trade interest is consequently reflected in the preference ordering P/FT > FT/ FT > P/P > FT/P. The preference ordering of American society during this same period appears to have been P/FT > P/P > FT/FT > FT/P. The first preference, P/FT, satisfied both protectionists and export-oriented sectors. Assuming that protectionists dominated Congress, P/P would then be preferred over FT/FT. The last preference, of course, was FT/P, because both protectionists and export-oriented sectors would be dissatisfied. Although the absolute differences between the national trade interest and society’s trade interest may have varied over the period, the central conflict between society and the representative element of the state, on one hand, and the foreign policy executive, on the other, clearly revolved around the second and third options. In the first and second phases of American trade strategy (discussed in Chapters 3 and 4, respectively), when the United States could free ride on Britain’s hegemonic leadership and obtain its first choice of P/FT, little conflict existed between society and the foreign policy executive. Both the national trade interest and societal demands could be easily reconciled. Only when Britain declined and the United States could no longer free ride would conflict between the foreign policy and representative elements of the state become acute.

The State and State Structure

In the United States, Congress is the principal representative element of the state. During the period studied here, three constituent agencies were formed: the Departments of Agriculture, Commerce, and Labor. Despite their role as representatives of specific functional interests, these agencies played relatively minor roles in the debates over trade policy.28 Congress was central. In this body geographic constituents are represented by individual legislators, and functional interests, with a degree of overlap, are represented by substantive committees. Within Congress, trade policy is formulated primarily by the House Ways and Means Committee and the Senate Finance Committee. The foreign policy executive is principally composed of the president, in his role as commander in chief of the armed forces and head of government, and the State Department.29

The American state is relatively decentralized but not anarchic. Perhaps more than any other, the American state was designed to be fragmented. Fearful of centralized political power, the founding fathers purposely dispersed authority within the governing apparatus. They also created numerous checks and balances by giving many if not all parts of the state a role in each and every policy decision. This system has led, as Samuel Huntington notes, to a “fusion of functions and division of power.”30 The decentralized structure of the American state is highlighted by international comparisons. As Peter J. Katzenstein cogently summarizes,

The decentralized structure of the state clearly constrains the policymaking process. As David B. Truman notes, the “diffusion of leadership and disintegration of policy are not hallucinations.”32 In seeking to create a political system responsive to the varied needs of society, the designers of the Constitution ensured an ongoing struggle among the elements of the state, that policy would ultimately result from the striking of pragmatic bargains, and that overall coherency in policy would be low.

In the Constitution, Congress was given specific authority to raise revenue and set tariff rates. Depositing that authority with Congress ensured that trade policy would be responsive to the needs of society, or at least the needs of politically mobilized groups within society. The tariff is an infinitely divisible and symmetrical political good. Congress can easily divide categories of goods and “tailor-make” rates for individual producers. During the period 1887–1934, societal conflict on the tariff was resolved through logrolling or the creation of encompassing coalitions.33 As Senator Knute Nelson of Minnesota sarcastically noted during the Payne-Aldrich debate in 1909, the tariff was based on the principle of “You tickle me and I tickle you. You give us what we on the Pacific Coast want for our lead ore and for our citrus fruit, and we will tickle you people of New England and give you what you want on your cotton goods.”34 Even with the passage of the Reciprocal Trade Agreements Act in 1934, an event often interpreted as signaling an enhancement of the executive’s role in the formulation of trade policy, Congress still maintained ultimate control by ceding authority to the president for only limited periods of time. In the era studied here, Congress reigned supreme and served the interests of the many producers who clamored for aid.

For the same reason that Congress was dominant in issues regarding trade, the president and the State Department were relatively isolated from the policy-making process. This is not to argue that the foreign policy executive lacked influence. The distribution of authority did, however, create a specific pattern of intrastate and state-society relations which endured over the entire period.

The principal political task for the foreign policy executive was to penetrate and gain access to the otherwise closed and jealously guarded congressional tariff-making process. The president, of course, possesses veto power over any tariff bill passed by Congress. But because of the months of work and careful bargaining that Congress normally devoted to omnibus tariff legislation and the identification of the tariff as a “party issue,” the president was often reluctant to exercise his veto power; in the period studied, the president explicitly threatened such action only in 1894, 1909, and 1930, and these threats were not taken seriously. More often, the president would intervene informally through the definition of the party platform and consultations with congressional leaders to shape the final legislation. Nevertheless, the informal power and personal influence of the president ultimately rested in the legislature’s need to obtain executive approval of the tariff bill. The State Department, and more precisely, the secretary of state—often the individual most clearly cognizant of the constraints and opportunities of the international economic structure—could also shape legislation through consultation and the exercise of personal influence. Though lacking the veto power of the president, the secretary often exercised considerable influence because the post was normally awarded to an “elder statesman” of the party.

The foreign policy executive possesses two strategies for increasing its leverage over tariff policy. Despite its relative isolation from the tariffmaking process, the foreign policy executive can mobilize societal groups with complementary interests into the policy-making process. Numerous domestic coalitions potentially exist. During the period examined here, American society was far from united on questions of protection or free trade and activism or passivism. And as Kenneth Arrow’s paradox and coalition theory indicate, even under weak and plausible assumptions, majorities and coalitions are likely to be unstable.35 The structure of interests facing the government is not rigid or predetermined. Instead, it resembles a clay which the relatively autonomous elements of the state can—within limits—mold and shape in ways they desire. To the extent that society is open to manipulation, the role of the foreign policy executive is magnified in importance. Somewhat paradoxically, one implication of a decentralized state structure is that it creates many actors and points of entry into the government.

By appealing to particular groups whose interests overlapped with its own, the foreign policy executive during the period examined here was able to circumvent the usual protectionist lobbies and place new offsetting pressures on Congress. As discussed in more detail below, Secretary of State James G. Blaine adopted this strategy in 1890 when he rallied midwestern farmers, who had not previously been active on the tariff issue, behind his proposal for bilateral reciprocity agreements with Latin America. President Woodrow Wilson followed a similar course in 1913, when he denounced the tariff lobby and thereby activated his progressive supporters. By mobilizing various groups, the foreign policy executive was able to turn the decentralized structure of the state into a source of strength.

The foreign policy executive’s second strategy is to use its unique position at the intersection of the domestic and international political systems to generate several entries into the otherwise closed congressional tariff-making process. The foreign policy executive can redefine domestic political issues as foreign policy issues, thereby gaining a legitimate policy-making role it otherwise would not have and strengthening its influence relative to its own society. Accordingly, President Grover Cleveland initiated the process of redefining the tariff in 1887, when he emphasized the export advantages that would derive from his duty-free raw materials program. From then on, nearly every president emphasized the foreign policy implications of the tariff. The foreign policy executive can also use its authority to enter into legitimate transnational coalitions and thereby change the options facing Congress. The final recommendations of the International American Conference of 1890 proposed—partly at Secretary of State Blaine’s urging—that bilateral reciprocity agreements be used to expand trade within the hemisphere. By agreeing to the recommendations of the conference as an official representative of the United States, Blaine effectively increased his bargaining leverage relative to the protectionists in Congress. If the legislature now failed to adopt reciprocity, it would risk disappointing the same Latin American countries the United States had so recently attempted to court. Secretary of State Cordell Hull attempted a similar but ultimately unsuccessful strategy in 1933, when he set off for the London Economic Conference with hopes of obtaining the consent of other countries to the opening of negotiations for global tariff reductions.

A focus on the structure of the state reveals which elements of the state will be the primary locus of decision making in a particular issue area and the domestic political strategies available to the relatively isolated foreign policy executive. Although the structure of the state does not determine outcomes, it does create a pattern of politics which endures over time. The legislature’s constitutionally delegated authority over trade policy forced the foreign policy executive to influence Congress indirectly by mobilizing social groups. In doing so, it transformed its isolation into an important source of influence. The foreign policy executive was also able to use its role as the sole authoritative maker of foreign policy to redefine the political debate and alter congressional options.


To the extent that Realists and neo-Realists have looked inside the black box of domestic politics, they have seen only a rational and unitary state acting in the often vaguely defined national interest. In this chapter, I started from this familiar ground but took a different path. By disaggregating the state, domestic politics once again becomes important.

The foreign policy executive is the critical link in the process by which the constraints and opportunities of the international economic structure are transformed into trade strategy. Concerned with national power and welfare, the foreign policy executive seeks to adopt strategies consistent with the national trade interest. In doing so, the foreign policy executive acts as a conduit, channeling the constraints and opportunities of the international economic structure into the domestic political process.

Yet despite the efforts of the foreign policy executive, the nation-state may choose not to follow its national trade interests. Competing demands exist. Society pursues its material interests through the representative elements of the state. Because of the relative nature of power and problems of collective action, the politically mobilized societal interests must differ from the national trade interest. Out of the resulting bargaining process, trade policy emerges, but not necessarily in a form consistent with the desires of the foreign policy executive.

Despite the disproportionate authority granted to Congress, American trade strategy in the period 1887–1939 did reflect the constraints and opportunities of the international economic structure. In the early phases of the period, societal interests and the national trade interest, though not identical, were not incompatible. Each set of interests could be satisfied at little or no cost to the other. Later, confronting a national trade interest which increasingly pointed toward the need for freer trade, the foreign policy executive was able to force through a fundamental reform of the tariff immediately before World War I. Finally, as the national trade interest continued to grow out of step with societal interests, the foreign policy executive was able to gain primary control over the tariff in the 1920s and 1930s. This historical progression of policy and the policy-making process highlights the success of the domestic political strategies adopted by the foreign policy executive and suggests the importance and high political salience of the constraints and opportunities of the international economic structure.

1Max Weber, Economy and Society, 2 vols., Guenther Roth and Claus Wittich, eds. (Berkeley: University of California Press, 1978), 1:56.

2Alfred Stepan, The State and Society: Peru in Comparative Perspective (Princeton: Princeton University Press, 1978), p. xii.

3The most developed “societal” explanations of tariff policy are found in the public choice literature; see note 1, in the Introduction.

4The distinction between demand- and supply-side explanations is made in Timothy McKeown, “Firms and Tariff Regime Change: Explaining the Demand for Protection,” World Politics 36 (January 1984): 216.

5This perspective is taken most clearly in Stephen D. Krasner, Defending the National Interest: Raw Materials Investments and U.S. Foreign Policy (Princeton: Princeton University Press, 1978).

6For examples from a variety of theoretical perspectives, see Eric A. Nordlinger, On the Autonomy of the Democratic State (Cambridge: Harvard University Press, 1981); Theda Skocpol, States and Social Revolutions (New York: Cambridge University Press, 1979); and Martin Carnoy, The State and Political Theory (Princeton: Princeton University Press, 1984).

7The conception of the state developed here reintroduces a degree of bureaucratic and “intrabranch” politics into the study of the state. On the former, see Graham T. Allison, Essence of Decision (Boston: Little, Brown, 1971); and Morton H. Halperin, Bureaucratic Politics and Foreign Policy (Washington, D.C.: Brookings, 1974). For the latter, see Robert A. Pastor, Congress and the Politics of U.S. Foreign Economic Policy, 1929–1976 (Berkeley: University of California Press, 1980).

8Another important element of the state may be the economic agencies, such as the Treasury Department and the Federal Reserve Bank in the United States. As compared to the constituent agencies, the economic agencies possess broad, societywide institutional mandates. Whether focusing on the macroeconomy or economic development, these broader mandates allow the economic agencies to avoid capture by particularistic interests, rendering the agencies at least relatively autonomous. Their concern for the economic performance of the nation-state makes the economic agencies sensitive to the international economy but not necessarily to the constraints and opportunities of the international economic structure. Although they are also unconcerned with the power of the nation-state, economic agencies charged with overseeing long-term economic development are an exception to this rule and may also demonstrate a regard for the relative gains from trade. The underlying motivation, however, remains domestic economic performance and not international power. In the period studied here, the economic agencies did not play an important role in the tariff-making process. As a result, they are excluded from the analysis. Today, their role is much greater and an examination of their interests and actions would be necessary.

9Central to Anthony Downs, An Economic Theory of Democracy (New York: Harper, 1957), this is now a widely accepted assumption in rational choice models of politics. See also David Mayhew, Congress: The Electoral Connection (New Haven: Yale University Press, 1975).

10On bureaucratic capture, see Marver H. Bernstein, Regulating Business by Independent Commission (Princeton: Princeton University Press, 1955); and Grant McConnell, Private Power and American Democracy (New York: Knopf, 1967), pp. 246–97.

11Mancur Olson, The Logic of Collective Action (Cambridge: Harvard University Press, 1971).

12This point was first made by Otto Hintze in “Military Organization and the Organization of the State,” in Felix Gilbert, ed., The Historical Essays of Otto Hintze (New York: Oxford University Press, 1975).

13Although the individuals who constitute the foreign policy executive may be motivated by personalistic (career enhancement) and bureaucratic (budget maximization) concerns, it is assumed that their ability to fulfill these desires depends at least in part upon the development and implementation of “successful” foreign policies, with success defined as satisfying their institutional mandate or preserving and enhancing national wealth and power.

14Pareto is cited in Krasner, Defending the National Interest, p. 12.

15Peter J. Katzenstein, “Conclusion: Domestic Structures and Strategies of Foreign Economic Policy,” in Katzenstein, ed. Between Power and Plenty: Foreign Economic Policies of Advanced Industrial States (Madison: University of Wisconsin Press, 1978), p. 324.

16See Stephen D. Krasner, “United States Commercial and Monetary Policy: Unravelling the Paradox of External Strength and Internal Weakness,” in Katzenstein, ed., Between Power and Plenty, pp. 51–87; Theodore J. Lowi, “American Business, Public Policy, Case-Studies, and Political Theory,” World Politics 16 (December 1964): 677–715; Joanne Gowa, “Public Goods and Political Institutions: Trade and Monetary Policy Processes in the United States,” International Organization 42 (Winter 1988); and Stephen G. Walker and Pat McGowan, “U.S. Foreign Economic Policy Formation: Neo-Marxist and Neo-pluralist Perspectives,” in William P. Avery and David P. Rapkin, eds., America in a Changing World Political Economy (New York: Longman, 1982), pp. 207–24.

17Despite the attention placed on interest groups in past studies of the tariff, surprisingly few studies have rigorously measured the size and material interests of various segments of society. Neither Becker nor Wilson, for instance, presents any empirical evidence on the influence or interests they ascribe to the manufacturing groups they examine. Rather, both rely solely on statements made by the leadership of selected organizations. See William H. Becker, The Dynamics of Business-Government Relations: Industry and Exports, 1893–1921 (Chicago: University of Chicago Press, 1982); and Joan Hoff Wilson, American Business and Foreign Policy, 1920–1933 (Boston: Beacon, 1971).

Bennett D. Baack and Edward John Ray test a more disaggregated interest-group explanation of American tariffs for 1870, 1910, and 1914. Although derived from within the tradition of endogenous tariff theory, several of their findings parallel the arguments developed here. They find no support for partisan political considerations as an explanation of tariff policy; a positive relationship between basic industries—which might be expected to possess considerable positive externalities—and tariffs; and a positive (but insignificant) relationship between tariffs and skill intensity of production and capi-tal/labor ratios (“The Political Economy of Tariff Policy: A Case Study of the United States,” Explorations in Economic History 20 (January 1983]: 73–93).

18Import penetration, or imports as a percent of the value of manufacturing, might be the best measure, but it would certainly give misleading results in an economy with moderate to high levels of protection and, more important, in which levels of protection fluctuated dramatically. Using export dependence yields more accurate results with one important qualification: it magnifies the importance of the non-export-dependent industries, which are otherwise assumed to be protectionist, by lumping the non-traded-goods industries and import-penetrated industries together. This is offset, in part, by those competitive industries which choose not to export heavily, perhaps because the domestic market is not yet saturated. For a similar approach, see Glenn R. Fong, “Export Dependence versus the New Protectionism: Constraints on Trade Policy in the Industrial World” (Ph.D. diss., Cornell University, 1983); and Helen Milner, “Resisting the Protectionist Temptation: Industry and the Making of Trade Policy in France and the U.S. in the 1970s,” paper presented at the Annual Meeting of the American Political Science Association, August 28–31, 1986.

19These figures and the data displayed in Tables 2.1 and 2.2 should be treated skeptically. I consider them to be only estimates of the level of export dependence. No single source presents both manufacturing output and trade flows for this period, nor was there any standard classification for industries. As a result, the disaggregated trade data had to be combined into categories resembling the fifteen industries listed in the census. Definitions of categories were often incomplete in both sources, so there may be errors in the classifications. Redefinitions of categories over time create a second problem. Agricultural implements, for instance, were classified as miscellaneous until 1919 and as machinery thereafter. As long as the data are treated with the proper caution, however, I do not believe the errors are debilitating.

20See Jeff Frieden, “Sectoral Conflict and U.S. Foreign Economic Policy, 1914–1940,” International Organization 42 (Winter 1988).

21Becker, Dynamics of Business-Government Relations; and Wilson, American Business and Foreign Policy.

22The changing interests of finance are best described by Jeffry A. Frieden, “Studies in International Finance: Private Interest and Public Policy in the International Political Economy” (Ph.D. diss., Columbia University, 1984), pp. 29–110; see also Wilson, American Business and Foreign Policy, pp. 16 and 110.

23One of the best overviews of farm interests and policy is Murray R. Benedict, Farm Policies of the United States, 1790–1950 (New York: American Book-Stratford Press, 1953).

24Ibid., p. 172.

25See David A. Lake, “Export, Die, or Subsidize: The International Political Economy of American Agriculture, 1875–1940,” Comparative Studies in Society and History (forthcoming).

26E. E. Schattschneider, Politics, Pressures, and the Tariff (New York: Prentice-Hall, 1935).

27Richard Hofstadter, The Age of Reform: From Bryan to F.D.R. (New York: Vintage, 1955), pp. 172, 218; Lippmann quoted p. 171.

28The Department of Agriculture was formed as the Agricultural Division of the Patent Office in 1839, established as a separate agency with bureau status in 1862, and elevated to a regular department with cabinet rank in 1889. The department was the first singleinterest service agency in the government. Until the expansion of agricultural support programs in the 1930s, the Department of Agriculture focused primarily on education, seeking to improve agricultural methods and disseminate this knowledge to farmers.

The Department of Commerce and Labor was created in 1903. Its early mandate was to support smaller businesses, which needed extra assistance compared to larger firms in their quest for foreign markets. The initial mandate of the Commerce Department was reconceptualized under the leadership of William C. Redfield, appointed as secretary of commerce by Woodrow Wilson in 1913. Redfield was the first secretary actively to court support from big business for the department. Ostensibly seeking to bridge the gap between small and large business, Redfield concentrated his efforts on the latter. The department avoided the tariff issue because of its divisiveness. No clear stand could be taken on this issue without alienating some important group of business constituents. After the war, the power and role of the Commerce Department expanded, in large part because of the personal influence and prestige of Herbert Hoover, the new secretary. As secretary, Hoover sought to exert greater control over international economic relations, thereby challenging the traditional supremacy of the State Department in foreign policy. Despite strong support from the business community, Hoover was ultimately unsuccessful. The role of the Commerce Department in the tariff-making process during the 1920s was strongly colored by the moderately protectionist views of the secretary (see Chapter 6).

The Department of Labor was separated from Commerce in 1913. Although organized labor was generally but not unanimously free-trade-oriented in the 1920s and 1930s, the Labor Department took a protectionist stand. In 1927, James L. Davis, the second secretary of labor and a former union official, maintained that workers’ interests were best promoted by protection. The agency’s stand was invoked by protectionists to support their cause, but the Department of Labor itself was not a key actor in the tariff-making process, perhaps because of its split with organized labor on this issue.

29The secretary of war, later renamed defense, might also be included in the foreign policy executive, although in the period examined here he was seldom involved in trade policy.

30Samuel Huntington, Political Order in Changing Societies (New Haven: Yale University Press, 1968), p. 110.

31Peter J. Katzenstein, “International Relations and Domestic Structures: Foreign Economic Policies of Advanced Industrial States,” International Organization 30 (Winter 1976): 15–16.

32David B. Truman, The Governmental Process: Political Interests and Public Opinion, 2d ed. (New York: Knopf, 1971), p. 529.

33See Lowi, “American Business,” for the best theoretical grounding of this classic argument.

34Quoted in Richard Cleveland Baker, The Tariff under Roosevelt and Taft (Hastings, Neb.: Democrat Printing, 1941), pp. 89–90.

35Kenneth Arrow, Social Choice and Individual Values, 2d ed. (New Haven: Yale University Press, 1963); on coalition theory see Robert Abrams, Foundations of Political Analysis (New York: Columbia University Press, 1980), pp. 41–101 and 235–79.

Additional Information

Related ISBN
MARC Record
Launched on MUSE
Open Access
Creative Commons
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.