Although it is widely acknowledged that economic interests influence the politics of trade policy, most research on international relations treats security issues differently. The realist paradigm prevailing in the analysis of security issues subordinates economic and other sources of domestic political conflict to the interests of the state as a unitary actor in the international system. Do conflicting economic interests shape foreign policy debate even when security issues are highly salient? I argue here that they do. The early Cold War era was a time of intense bipolar rivalry when security questions might have been expected to override all other concerns. I will present evidence that, even during this period, major political actors' positions on many foreign policy issues relate to patterns of conflict among economic interests.
"National Security" and Economic Interests
In most international relations theory, national security concerns involve the protection of unitary state interests, such as political sovereignty or territorial integrity, from international threats. These interests arise from objective international conditions and the state's position within the international system. This view of national security as the survival interests of the state drives Kenneth Waltz's structural realist theory and other influential accounts of the international system.(1) Incentives for political action on security questions differ from those existing in other issue areas, such as trade policy, because they involve the common interests of the society, removing the basis for distributive conflict. Opinions may vary on classic national security issues because of individual tastes and idiosyncrasies, but there is no reason to expect political conflict over it to be related to more enduring social cleavages.
Although this understanding of national security may be a reasonable description of security concerns in some cases, it is not universally applicable. For many great powers, security concerns have ranged far beyond the survival interests of the state. Powerful states with imperial ambitions have historically treated the preservation of access to particular sites for trade and investment as vital national interests, securing them with military means. Although this expansive "national security" policy certainly benefits some segments of the society, others may be unable to take advantage of the investment opportunities it protects. Some industries may even have to compete with goods imported from these protected areas. Despite potential domestic disagreement over their desirability, the preservation of these economic interests has sometimes been a critical goal of national security policy. Indeed, the overextension of military power to protect imperial possessions, even when doing so has made defense of the homeland more difficult, is a recurring theme in world history.(2)
In extending their foreign policy goals beyond securing the political sovereignty and territorial integrity of the state, U.S. policymakers have followed the example of many previous great powers. After World War II, American policymakers sought to establish an international system open to U.S. trade and investment, one where American power would predominate. The major policy statements of the early Cold War era acknowledge that these ambitious goals went well beyond merely coping with the Soviet military threat to the United States. One of the most important policy statements, NSC 68, noted that establishing a "healthy international community" was "a policy we would have to pursue even if there were no Soviet threat."(3) Paul Nitze, one of the principal authors of NSC 68, later recalled that when work on the report began, "our prime concern had remained with the economic situation in Europe."(4)
In principle, policies aimed at protecting the basic political sovereignty and territorial integrity of the state are distinguishable from those directed at potentially divisive economic goals. Different political processes might exist side by side in these two areas. In practice, however, policymakers often use the same programs to pursue many goals. Economic and security concerns are often closely related. States give economic aid to their allies for security reasons, and industries deploy arguments about national security to gain trade protection. The recent work of Joanne Gowa, presenting evidence that states take the security externalities of trade into account when selecting their trading partners, underlines the difficulty of distinguishing between these two issue areas.(5) Conversely, a particular military posture may be necessary to establish and maintain the conditions necessary for certain kinds of investments.(6)
The analysis of conflict among firms and industries with competing interests offers one way of explaining political conflict over national security policy when its objectives are not consensual. Endogenous tariff theory links the position of various industries in the international economy to their demands for government policy. Assuming that factors are relatively immobile between industries, Timothy McKeown, Robert Baldwin, and others argue that conflict between export-oriented and import-competing sectors shapes the politics of trade.(7) Of course, as Douglas Nelson points out, not all endogenous tariff arguments assume factor immobility.(8) Some, such as Stephen Magee, William Brock, and Leslie Young treat conflict between the owners of different factors as the basis for divergent preferences on trade protection.(9) However, given the evidence that political cleavages form along sectoral lines, at least in the short run, a sector-based theoretical approach provides a reasonable starting point for empirical research.(10)
Although sectoral differences in the benefits of tariffs are well known, related conflicts extend to issues other than trade protection.(11) To the extent that the benefits of a foreign policy do not extend to everyone, those who do not expect to benefit should oppose committing national resources to these goals. Not everyone shared the Truman administration's broad foreign policy goals or favored the range of related economic and security policies these goals required. Patterns of political conflict reflected the differences in the costs and benefits of these policies to different segments of American society. Two such differences are particularly important. First, some firms and industries stood to benefit more than others from access to international markets and sites for investment. Although efforts to secure the interests of internationally oriented firms may not have directly harmed the interests of domestically oriented firms, economic interests that gained no clear benefits from these efforts nevertheless had to bear their considerable costs. Second, conflict occurred over the regional emphasis of the administration's foreign policy. Faced with limited resources, administration foreign-policymakers emphasized economic and security concerns in developed areas, especially in Europe. Not everyone shared these priorities.
The early Cold War era offers a strong test of the argument that variation in the benefits of security policies in powerful states may generate the same distributive conflicts found in foreign economic policy. Of course, economic and security concerns were closely related in Cold War foreign policy. Its framers intended the Marshall Plan to serve the security goal of rebuilding allies as well as the economic purpose of reconstructing trading partners. Similarly, the rearmament program associated with NSC 68 was intended to provide a source of dollars to cover the European balance-of-payments deficit as well as to arm American allies against the Soviet military threat. Despite the close relationship between these two policy areas, security concerns have never been more prominent for the United States in the postwar era than they were during the early Cold War period. If political actors' national security concerns override their ties to conflicting economic interests, it should be evident during this time. In the remainder of this article, I will develop and test hypotheses about the role of divergent interests in international trade and investment in shaping political conflict over early Cold War foreign policy.
Economic Interests and Foreign Policy Preferences During the Early Cold War Era
Many crucial beliefs about the appropriate world role of the United States underlying Cold War foreign policy have become common assumptions during the last fifty years. Given the prevalence of these beliefs, it is easy to forget that many basic elements of Cold War foreign policy were controversial when first introduced. The Cold War entailed much greater global activism and a correspondingly larger commitment of resources to foreign policy than the United States had previously undertaken during peacetime. As the title of Dean Acheson's memoirs, Present at the Creation, suggests, the framers of this new policy were well aware that they were breaking with previous practice. Although few influential American political figures argued for friendly relations with the Soviet Union, there was nevertheless intense disagreement over the scope of postwar American commitments and the allocation of the resources required to meet them. The viciousness of this debate reflected the importance of the issues at stake. Those opposed to the Truman administration's foreign policy characterized both minor and important administration officials as fundamentally out of touch with the American people, or even as traitors.(12)
Despite the intensity of the political struggle over foreign policy during the early Cold War era, many historians pay relatively little attention to it. Some historical works, including those of Melvyn Leffler and John Gaddis, assume a realist perspective, downplaying domestic political conflict and focusing instead on international threats to perceived security interests.(13) These interpretations attribute opposition to the administration's foreign policy to the misunderstanding of international conditions or partisan opportunism. On the other hand, other historians, such as Michael Hogan and Bruce Cumings, relate conflict over Cold War foreign policy to deeper cleavages within American society.(14) These accounts reflect theoretical arguments about conflict among firms and industries with different foreign policy interests.(15)
The conflict over early Cold War foreign policy resembles the commonly discussed antagonism between "internationalist" and "isolationist" perspectives on American foreign policy. Because these categories are so well known, it is tempting to label the Truman administration's opponents "isolationists." This temptation should be resisted for at least two reasons. First, some of the administration's opponents favored just as much international activism but placed greater stress on securing access to less developed areas, especially in Asia. Second, because internationalists ultimately prevailed, the term "isolationist" carries a normative stigma. When the winners write history, their first concern is not with the selection of terms facilitating even-handed analysis. The term "nationalist" is probably more descriptive and is certainly less pejorative.(16)
Hogan, Cumings, and others point out the prevalence within the Truman administration of policymakers linked to internationally oriented banks and multinational enterprises located mainly in the Northeast. These policymakers sought to establish and maintain an international economic order open to U.S. trade and investment, especially in the industrialized countries of Western Europe and Japan. Preserving access to Europe and Japan entailed not only a military commitment to prevent possible conquest or intimidation by the Soviet Union but also an effort to reconstruct their devastated economies and cover their enormous balance-of-payments deficits. U.S. policymakers feared this "dollar gap" would prompt European states either to seek greater economic ties with eastern Europe and the Soviet Union or to construct quasi-autarkic arrangements with their colonial empires, sharply limiting U.S. trade and investment. After 1950, when economic aid to Europe was no longer politically viable at home, American policymakers turned to military aid to accomplish the same purposes. Concern about the reconstruction of the European and Japanese economies also drew the United States into defending their access to Third Word markets and sources of raw materials.(17)
While internationalists dominated policy making in the Truman administration, many leading Republicans, such as Robert Taft, reflected the interests of other segments of the American political economy. Domestically oriented businesses were not in a position to reap the benefits of international trade and investment and so had little incentive to support the administration's risky and expensive plans to ensure the security of Western Europe and Japan. Lynn Eden and others have pointed out that political leaders associated with these nationalist interests preferred less costly military strategies based on the use of air power for continental defense.(18) They opposed the Marshall Plan and other foreign aid initiatives, as well as the stationing of American ground forces in Europe.
While some opposed the administration's internationalist foreign policy because of its costs and risks, others opposed it because it did not seek more extensive political control of less developed areas, particularly in Asia. Cumings presents evidence that some "expansionists" with ties to extractive mineral industries preferred a strategy based on direct control of territory in less developed areas.(19) This preference makes sense in terms of their interests. As Jeffry Frieden points out, extractive industries have a greater incentive than other foreign direct investors to seek colonial control of areas where they have extensive investments.(20) After the decision not to intervene in the Chinese civil war, it appeared that the administration's foreign policy would not secure these interests.
The debate over China policy had clear implications for investors in other less developed areas. Conflict over U.S. policy toward these areas focused mainly on Asia because access to this region was much more seriously threatened than access to Africa or Latin America. The administration's decision to avoid intervention in China raised questions about its plans for other parts of the less developed world. Although Latin America was a relatively safe location for U.S. investors, the political future of the European colonies in Africa and Asia was in doubt by 1949. Investors in these areas could not have been comfortable with the idea that the U.S. government would do nothing to prevent the emergence of regimes that might confiscate their local assets. Among the most important effects of the political outcry over China policy was to make subsequent U.S. policymakers much more reluctant to permit communist gains in any less developed area, a tendency Fred Block aptly labels the "loss-of-China complex."(21)
If the economic stakes were really so important for early Cold War politics, the views of political leaders about both the overall benefits of internationalism and the regional emphasis of U.S. foreign policy should reflect their ties to different interests in international trade and investment. Comparing congressional voting patterns and the available data on the economic structure of the United States in the late 1940s is a good way to evaluate the influence of economic interests on foreign policy preferences. The implications of the Truman administration's foreign policy for American firms and industries involved in international trade or foreign investment suggest some testable hypotheses.
Because the Truman administration intended to establish and maintain a relatively open international trading system, commercial policy interests should influence attitudes toward its foreign policy. The benefits of this new international order differed for exporting and import-competing industries. Political leaders from states with a large import-competing sector should be less likely to support the Truman administration's foreign policy. Those from states with more export-oriented industries should be more likely to favor it. The regional emphasis of the administration's policy also had implications for international traders. Firms that exported primarily to developed countries, or to European markets, should influence their representatives in favor of the administration's policy. Those oriented toward Asian markets or less developed countries should take the opposite position.
The administration also sought a favorable climate for American investment, concentrating its efforts on the developed states of Europe and Japan. Concern about the future of U.S. investments in Europe not only fueled fears about potential Soviet domination of the continent but also caused concern about the continuing European balance-of-payments deficit with the United States. As Block points out, the NSC 68 rearmament program was intended to deal with both of these problems.(22) Because they may behave differently, I consider bank lending and direct investment separately in the empirical analysis.
Differences over the general benefits of internationalism and the regional orientation of the Truman administration's foreign policy affect the foreign policy interests of U.S. banks. Political leaders from states where banks were especially active in international lending, particularly to Europe, should be more likely to support the administration's foreign policy. On the other hand, the administration was frequently criticized for failing to show as much concern about Asia, especially after the collapse of the Nationalist regime in China. Political leaders from states where banks lent more heavily to Asia should oppose the administration's policy.
Of course, lenders and direct investors might simply respond to changing opportunities in international markets rather than seeking to influence these opportunities by political means. If so, there is no reason to expect any pattern of political influence over foreign policy by banks and corporations with foreign investments. On the other hand, not all banks are equally well positioned to take advantage of international lending opportunities. Those with greater experience in international markets, existing business relationships overseas, and other advantages might seek government action to secure international lending opportunities. The geographic concentration of international financial activity in the United States during this period - evident in Table 4, discussed later - supports the argument that some banks were simply more internationally oriented than others. Similarly, not all areas of the world are equally suited for direct investments by particular firms. The size of the local market, the local endowment of natural resources, and other considerations may considerably narrow the range of international opportunities available to some industries. Access to regions where opportunities exist might be important enough to prompt political action. This is particularly true for direct investors in less developed areas, who were concentrated more heavily in extractive industries depending on local mineral resources or agricultural conditions. In any event, whether or not lenders or direct investors engaged in political action is an empirical question. If these firms are not politically active, there should be no relationship between their geographic distribution and congressional voting patterns.
Finally, the Truman administration's effort to establish an international system open to U.S. trade and investment benefited firms with direct foreign investments outside the Western Hemisphere. Although U.S. access to the Western Hemisphere was not in danger during the early Cold War era, conditions in other areas of the world were less certain. Political leaders from states home to many corporations active outside the Western Hemisphere ought to support the administration's foreign policy more readily than those from states where foreign investment was less important. The high costs and risks of the administration's foreign policy probably boded larger for domestically oriented firms that did not expect to benefit from its preservation of access to potential investment sites. The regional orientation of the administration's foreign policy is also relevant to the interests of direct foreign investors. Because the administration's commitment to preserving access to less developed areas was in doubt after the fall of China, corporations with direct investments in extractive industries in these areas should also oppose the administration's foreign policy and pass these concerns along to their political representatives.
Overall, Cold War foreign policy had different implications for exporters, import-competing industries, banks with different levels of international lending, and firms involved in different levels of direct foreign investment. Table 1 summarizes the pattern of conflicting interests in trade and investment discussed here. Those with a stake in the establishment and preservation of an international political and economic order open to U.S. trade and investment, especially in Europe, should support the Truman administration's foreign policy more than groups without such a stake.
Two caveats apply to this argument about economic interests and foreign policy preferences. First, the influence of economic interests on politicians' policy preferences need not be attributed exclusively to campaign contributions from interested groups and individuals. Although campaign finance is an important mechanism for transmitting interests, the economic structure of their home states may influence politicians in more subtle ways. For example, their state's economy may decisively shape the attitudes of those with whom politicians discuss the issues of the day and form their ideas about "the national interest." These ideas are likely to vary among groups of people dependent on different types of economic activity. Elite conversations about foreign policy in Omaha probably sounded quite different from those in New York. For present purposes, it is not necessary to specify all the ways economic interests may shape politicians' preferences, only to establish that this pattern of influence existed.
Second, the empirical analysis presented here can consider only a few of the avenues through which economic interests might influence foreign policy preferences. Powerful interest groups can of course influence members of Congress outside their home states. Furthermore, firms in a particular industry may not all have the same interests. The firm-based data I use for the foreign investment variables may capture these intra-industry differences, but the industry-based data I use for the trade variables will not. In industries with few international interests, firms planning future international activity might favor internationalism more than their existing investments suggest. The hypotheses developed here also do not consider the possible influence of defense industries, which have a different stake in an internationalist foreign policy. Although there are many other ways economic interests might influence policy preferences, they do not invalidate the empirical analysis presented here. These complicating factors should bias the empirical tests presented here against finding any significant influence by economic interests. The models presented here offer only a rough test of the role of economic interests, but, nevertheless, it is a very demanding test.
Additional Influences on Foreign Policy Preferences
Even on clearly economic issues, foreign policy preferences are not simply a function of economic interests. Noneconomic factors also influence political actors' foreign policy outlooks. A substantial body of research contends that ideology influences congressional voting patterns on foreign policy and many other issues.(23) During the early Cold War era, liberals tended to favor the Truman administration's internationalist foreign policy, whereas conservatives tended to oppose it.(24) Although ideology reflects the influence of economic interests, sincerely held ideological beliefs probably diverge from the interests that back them on some issues. For example, some firms and industries that supported conservative positions on domestic issues might have differed from most conservatives regarding foreign policy during the early Cold War era. Including ideological variables in the empirical analysis represents a cautious theoretical approach and a tougher empirical test for the influence of economic interests. If ideology independently influences foreign policy preferences, it should be included as a control variable. On the other hand, if ideology is essentially a rationalization for economic interests, its inclusion will bias against finding any influence by economic interests.
Caution also warrants a separate consideration of the role of region in shaping foreign policy preferences. Most of the literature on foreign policy treats region as a proxy for economic structure.(25) This approach makes sense because regions differ in economically important ways, such as climate, soil quality, prevailing wages and labor practices, and the presence or absence of barriers to international markets. Although variations in economic structure account for many regional differences, region may still independently influence political outlooks. American institutions define congressional constituencies in regional terms, making them more politically salient than other societal cleavages. Since regional issues unite their constituencies, members of Congress may find them particularly attractive. Furthermore, regional political cultures change more slowly than the regional economy. When economic change occurs rapidly, as it did during the immediate postwar era, this lag effect may be even more important.
During the early Cold War era, region influenced foreign policy preferences in several ways. First, there was more opposition to the administration's foreign policy in the Midwest and Mountain West than in other parts of the country. As many accounts of American isolationism have noted, the political culture of these states, as well as their relative geographic isolation from international markets, encouraged a skeptical view of foreign entanglements.(26) Second, southern members of Congress - all of whom were Democrats in the late 1940s and early 1950s - supported the administration's foreign policy less than other Democrats did. This tendency represents a division within the Democratic party rather than simply a regional effect. Southern Democrats were more conservative than their Northern copartisans on most issues.
Whereas ideology and region represent competing influences on foreign policy preferences, political party complements economic interests. The literature stressing the role of economic interests generally recognizes the mediating influence of political parties in organizing these interests and linking them to concrete political action. James Kurth notes that competition between coalitions of different industries will determine foreign economic policy when a single industry does not dominate the national economy.(27) Thomas Ferguson and Jeffry Frieden both stress the role of political parties in organizing these competing blocks of industries.(28) Viewing party as a competing explanation for foreign policy preferences is a mistake. It is a crucial part of the way economic interests influence foreign policy preferences.
Treating parties as coalitions of economic interests has practical implications for the empirical analysis of these interests. Particular firms and industries may strongly influence politicians in one party but have much less influence outside that party. Politicians' responsiveness to particular interests may depend as much on their party affiliation as on the economic structure of their home states. Variations in the influence of particular firms and industries in different parties help explain the empirical evidence that political candidates do not always converge toward the position of the median voter.(29) If a state contains sharply conflicting interests seeking incompatible policies, politicians would have difficulty securing the support of all major interests in any event. They may simply adopt a position favoring interests associated with their own party instead of moving toward the middle.
The requirements of maintaining party unity may also lead members to positions antagonistic to major interests in their own states on some issues. Parties encompass interests on many different foreign and domestic policy issues, and the positions of all its members are unlikely to coincide on all policy questions. Compromises are necessary to preserve the coalition. Richard Bensel, for example, offers evidence that southern support for the liberal programs enacted by the Democratic party during the New Deal era depended on the toleration of racist southern institutions by the northern wing of the party.(30)
Overall, any model of the economic sources of foreign policy preferences should consider the influence of such factors as ideology, region, and, above all, party. Even if ideology and region can largely be reduced to economic structure, they may still have independent effects on foreign policy preferences. Including them in the model is the most cautious approach. Political parties, on the other hand, mediate the influence of economic interests, orienting their members toward the influence of some firms and interests and away from others. Their effects complement the influence of economic interests.
Modeling Foreign Policy Preferences
The influence of economic interests and other variables on foreign policy preferences during the early Cold War era will be tested in the model that follows. The operationalization of the major variables requires some explanation. Table 2 presents descriptive statistics on the variables used here.
V = f([E.sub.1], [E.sub.2], [E.sub.3], P, I, R)
V is the proportion of votes each senator cast in favor of the administration's foreign policy.
[E.sub.1] is the commercial policy orientation of the senator's home state, indicated by export orientation and import sensitivity.
[E.sub.2] is the international orientation of the financial sector in the senator's home state, indicated by the size of portfolio investor interests outside the United States and in different regions of the world.
[E.sub.3] is the strength and regional orientation of foreign direct investors based in the senator's home state, indicated by the interests of the major corporations headquartered in the state.
P is the senator's party affiliation.
I is the senator's ideology.
R is the senator's home region.
The Dependent Variable: Senate Voting on Foreign Policy During the 81st Congress
The lack of congressional district-level economic data makes the analysis of voting patterns in the Senate easier than in the House of Representatives. However, there are also substantive arguments for concentrating on the Senate. Because of the Senate's constitutional foreign policy prerogatives and because administration opponents were overrepresented there, the Truman administration focused its legislative concerns about foreign policy on the Senate. The administration's legislative strategy centered on securing Senate support with the assistance of Senator Arthur Vandenberg of Michigan and other internationalist Republicans. Paul Nitze later recalled that "what we decided to do was to build up Senator Vandenberg, as opposed to Senator Taft, and create a split within the Republican Party, and to drive our policy in between these two poles in the Republican Party, and it worked."(31)
The 81st Congress, which met in 1949-50, provides an especially demanding test of the influence of economic interests on the debate over U.S. foreign policy. This Congress was the last one in which the major foundational issues of Cold War foreign policy, including the nature of the U.S. commitment to Europe, and the size of the military budget, were subject to serious debate. By 1949, the antagonism between the United States and the Soviet Union was obvious to members of Congress. Furthermore, the foreign policy issues this Congress faced, including the North Atlantic Treaty, the Military Defense Assistance Program, and the massive military budget increases that followed the Korean War, were more clearly "security" initiatives than were earlier administration proposals such as the Marshall Plan. If security concerns overrode economic interests in shaping preferences, it should be evident in the 81st Congress.
I used all the roll-call votes on foreign policy taken during the 81st Congress to develop an index of each senator's support for the administration's foreign policy. I omitted only votes on measures directed exclusively at a particular industry or special interest group, because they do not reflect general positions on foreign policy but mainly attitudes toward that special interest. Of the remaining ninety-two votes, I coded those supporting the administration's position "1," and votes against such measures, or in favor of initiatives designed to interfere with the administration's program "-1." I then calculated mean scores between 1 and -1 for each senator across all ninety-two votes. Appendix A contains a complete list of these votes and the way they were coded. As I will explain in detail when discussing the empirical results, I also calculated several alternative voting scores to test the robustness of my findings.
Specifying the Influence of Economic Structure
The model estimates the effects of economic interests in three types of international economic activity: trade, portfolio investment, and direct investment. Appendix B explains in greater detail how I collected the economic data and calculated each variable. Although the available data are limited in some respects, these limitations should, if anything, make it more difficult to find statistically significant relationships.
To test the influence of commercial policy interests, I used variables indicating the size of the export sector and the import-sensitivity of each state. I calculated these based on national aggregate data on trade and state-level data on manufacturing, mining, and agriculture. Unfortunately, data on the regional orientation of U.S. exports during the early Cold War era were unavailable. Although hypotheses about interests in European and Asian trade follow from my argument and are included in Table 1, I cannot test them here.
To test hypotheses about the influence of portfolio investor interests, I calculated three variables on international lending. Using data on international lending reported to the Federal Reserve by member banks, I developed indicators of the level of international lending by banks in each state and two variables on their regional orientation. Because a variable on the proportion of bank lending to domestic borrowers would be perfectly collinear with the variable indicating the proportion of lending to foreign borrowers, I used just one variable to test the influence of banking interests. Because the variable for total foreign lending was highly collinear with the regional lending variables, I have estimated their effects in separate models.
Using data on the international activities of the two hundred largest U.S. manufacturing, transportation, merchandising, and utility firms, I created aggregate indicators of the international interests of the corporations headquartered in each state. To test my hypotheses about the effects of these direct investor interests, I calculated a variable representing the proportion of the major corporations in each state with subsidiaries outside the Western Hemisphere. A variable indicating the proportion of these large firms without such investments would be essentially identical, so it is not included.
Based on the arguments of Frieden and others about the special interests of extractive investors,(32) as well as Cumings's evidence of the importance of extractive mineral interests in the debate over Asia policy,(33) I have modified my emphasis on regional interests in the case of direct investors. Rather than simply examining their regional orientation, I also coded them according to the type of investments they held. The variable used here indicates the proportion of the major corporations headquartered in each state with extractive subsidiaries in less developed areas. Since investment in less developed areas was concentrated in extractive industries during the late 1940s and early 1950s, the empirical difference between this variable and one simply examining holdings in less developed areas is not very great.
Party, Region, and Ideology
In contrast to the data on the structure of the U.S. economy during the early Cold War era, data on the home states and party affiliations of individual senators are unambiguous and easily available. I have used a dichotomous variable coded "1" for Democrats and "0" for Republicans. Because party mediates the influence of economic interests, Democrats and Republicans may pay closer attention to some economic interests than to others. To evaluate these effects, I estimated interactions between the partisanship variable and the economic structural variables. To test the regional hypotheses, I use two dichotomous variables. One is coded "1" for senators from the Midwest and Mountain West and "0" for senators from all other states. The other is coded "1" for senators from the South, and "0" for senators from all other states.(34)
Although region and party affiliation are straightforward, discerning an individual senator's ideology poses special problems. The use of voting scores to test the influence of ideology on other voting scores is problematic. Indicators of ideology based on voting records, such as the scores given by the Americans for Democratic Action (ADA), reflect the influence of other variables that affect voting, including the other variables in these models. Even though ADA scores are strongly related to the voting scores used here, John Jackson and John Kingdon point out that this relationship may indicate only that most votes load onto a single dimension, a widely recognized feature of congressional voting.(35) Appendix C discusses in greater detail the means I chose to handle ideology. Briefly stated, I identified those senators who were more ideologically consistent than their party affiliation, region, or the economic structure of their home states would lead one to expect. The differences between these groups of liberals and conservatives, and the rest of the Senate, offer a reasonable approximation of the role of ideology in shaping political conflict on foreign policy.
Empirical Results
Table 3 contains the results of several regression models estimating the influence of the variables discussed earlier on the voting score of each senator. I will discuss the results concerning trade, portfolio investment, and direct investment in turn.
The Influence of Export Orientation and Import Sensitivity
In every specification of the model, the effect of import sensitivity is significant and has the predicted sign. Senators from states with large import-competing sectors tended to oppose the administration's foreign policy. In principle, it is possible to calculate the precise impact of a change in the import sensitivity or export orientation of a state on the voting score of its senators using the coefficients estimated in the model. However, the imperfections of the indicators for economic structure used here make exact statements about the influence of each variable less than convincing. As I noted earlier, because the limitations of the data bias against finding statistically significant influences by economic interests, they do not invalidate the evidence that these relationships exist. In order to avoid false precision, though, I limit my interpretation of the variables to the direction of the impact of the independent variables rather than the magnitude of their impact.
Export orientation had no statistically significant relationship to the foreign policy voting score. Although the theoretical argument presented here suggests that exporters should influence senators from their states in favor of the Truman administration's foreign policy, the apparent political weakness of this sector is not surprising in view of previous empirical research. Baldwin also found that districts dominated by import-sensitive industries elected more protectionist legislators, but that an export-orientation variable was statistically insignificant.(36) He points out that it is probably more difficult for exporters to overcome the collective action problems of organizing for political lobbying. Unlike import-competing firms, for which the costs and benefits of trade protection are immediately evident, exporters are not immediately affected by domestic trade protection. Although foreign retaliation against them is possible, it is only likely if the countries importing their products are also the countries against which tariffs are directed. Because of the different incentives they face, import-competing interests are more likely to be politically organized than exporters.
The Influence of International Lenders
Banks in their home state also influenced senators' preferences. Whereas the total amount of international lending, included in model 1, had no statistically discernible impact on Senate voting, models 2, 4, and 5 indicate that lending to particular parts of the world had significant political effects. As expected, senators from states where banks lent heavily to Asia tended to oppose the administration's foreign policy, whereas senators from states where banks lent heavily to Europe tended to favor it. Although European lending was greater than Asian lending in all Federal Reserve districts where foreign lending occurred, Asian lending was most important in the San Francisco and Cleveland districts. European lending was relatively most important in the New York and Boston districts.
Interests in the overall costs and benefits of internationalism are theoretically distinguishable from interests in the regional orientation of the Truman administration's foreign policy. However, it is possible that these interests have complementary effects in practice. States where banks conducted little international business and states where lending to Asia was an important concern might have elected senators with similar foreign policy positions. In both cases, bankers were likely to be critical of the administration's foreign policy, albeit for different reasons. Conversely, European lending and overall international activity might have had complementary influences in favor of the administration's internationalist foreign policy. Unfortunately, multicollinearity prevents the inclusion of both overall international activity and regional orientation variables in the same model. As an alternative way of testing the possible complementary effects of these interests, I used a variable adding the proportion of lending to Asia to the proportion of lending to domestic borrowers. Although this hypothesis is plausible, the results of model 3 do not support it. The regional orientation of U.S. banks, rather than their overall volume of international business, appears to have shaped their political influence during the early Cold War era.
The influence of regionally oriented banking interests is particularly striking in view of the relatively small amounts of money involved. As Table 4 indicates, even European lending was a tiny fraction of total bank activity. The volume of Asian lending was extremely small. International lenders apparently had disproportionate political influence relative to their role in the economy. Political action by banks on foreign policy issues might reflect their future ambitions for Europe or Asia rather than their actual lending in 1950. Despite the small value of loans to these areas, the lenders may have expected greater business following postwar reconstruction, provided they had continued access to these regions.
Not all regional concerns were politically important. For example, when used in a model like this one, a variable representing lending to Latin America is not significant. This result is easy to explain in terms of the international realities of the time. Although U.S. portfolio investment in Latin America was almost as great as in Europe, and was much greater than in Asia, policy toward this region was not as controversial as policy toward Europe or Asia. Latin America did not face the same threat of communist domination or incorporation into autarkic European imperial trading areas as other areas of the world did in the late 1940s. Because the region was not seriously threatened, there is no reason to expect interests in it to affect the broader debate about the outlines of U.S. foreign policy.
Historical evidence suggests that the international financial sector was also influential in the administration. Paul Nitze, Robert Lovett, Averell Harriman, James Forrestal, and John McCloy had been investment bankers before entering government service. Dean Acheson, Thomas Finletter, and John Foster Dulles, among others, were lawyers for large firms in New York and Washington. These leading officials drew many of their subordinates from the same circles. Many were Republicans, but all shared a commitment to the idea that the United States should play an active role in promoting a more liberal international economic order to preserve American access to trade and investment opportunities, especially in Europe. These same financial interests also played a direct role in implementing the aid program. Through 10 August 1950, the Economic Cooperation Administration had issued letters of commitment to fifty-nine U.S. banks for Marshall Plan aid funds, at the request of participating countries. The top thirteen banks received more than 90 percent of the $4 billion these letters of commitment represented. All but one of these banks were based in New York. The group included Brown Brothers Harriman, in which Averell Hardman and Robert Lovett were partners.(37)
Other research suggests that the financial sector has significant influence over political action by business groups in general. Ferguson notes that large financial institutions often play central roles in business organizations.(38) Similarly, McKeown presents statistical evidence that financial institutions have played a significant role in political action committee formation during the 1975-84 period.(39) The evidence presented here supports these arguments about the political influence of banks.
The Influence of Foreign Direct Investors
Direct investment also helped shape support and opposition to Cold War foreign policy. In models 1, 2, and 3, the coefficient for the proportion of the largest firms headquartered in the state with direct investments outside the Western Hemisphere is positive and significant. Senators from states with a higher proportion of these internationally oriented firms tended to favor the administration's foreign policy. Senators from states where fewer large firms were internationally active were more likely to oppose Cold War foreign policy.
As with the portfolio investment variables, those representing foreign direct investment were collinear and do not produce reliable estimates when all are included in the same model. I included only one in each regression. Model 4 in Table 3 tests the hypothesis that the regional orientation of foreign direct investors from their home state influenced senators' foreign policy outlooks. Those with extractive interests in less developed areas should influence senators against the administration's foreign policy. In this case, however, the empirical evidence does not support the regional-orientation hypothesis.
As with the international lending variables, the effects of the regional orientation and overall international interests of direct investors may have been complementary. Domestically oriented firms and those with extractive investments in less developed areas should both oppose the administration's foreign policy. To test this hypothesis I added the proportion of large firms with no foreign subsidiaries and the proportion of firms with extractive investments in less developed areas. As model 5 indicates, this variable was significant. Domestically oriented corporations and those with extractive investments in the less developed areas had complementary political effects. The negative coefficient on this variable indicates that having a high proportion of these large firms in their home state made senators more likely to oppose the Truman administration's foreign policy.
The Effects of Party, Ideology, and Region
In all specifications included in Table 3 the variables indicating party, region, and ideology performed as expected. In each case Republicans, senators from the Midwest and South, and conservatives were more likely to oppose the administration's foreign policy than were other senators, holding all other variables constant. Liberals and Democrats were more likely to favor administration policy. As I noted earlier, because all of the southern senators were Democrats, the results for this variable indicate intraparty differences as much as the effects of region. As Democrats, southerners were still more likely than Republicans to favor the administration's foreign policy.
Alternative Voting Scores
In order to check the robustness of these relationships, I used three alternative voting scores in place of the dependent variable used in Table 3. Table 5 presents these results. The patterns evident in foreign policy voting do not vary greatly if the voting score is calculated differently or if a restricted set of issues is examined.
In principle, a list of Senate roll-call votes may overrepresent some issues. There were many more votes on foreign aid than on military spending, trade, or the NATO alliance. To test the possibility that the overall voting score reflected cleavages unique to one issue area, I calculated an alternative voting score for each senator. Dividing the ninety-two votes into the twelve issue areas listed in Appendix A, I calculated a mean score weighting each issue area, rather than each individual vote, equally. The results of the second model in Table 5, which uses this alternative dependent variable, do not differ substantially from those in the first model using the original independent variable. Although some of the coefficients changed slightly, all of the same variables were significant, and all of the signs were in the same direction.
I have argued that the fundamental stakes in the debate over early Cold War foreign policy were economic, even on issues with important military implications. It is difficult to distinguish security issues from economic issues, because both served the same broad internationalist goals, and many programs were intended to further both the economic and the politico-military aspects of the overall foreign policy. Others might expect different voting patterns on economic and politico-military issues. To test this argument, I constructed two more voting scores, one including only votes on programs where the security rationale was clearest, the other including only votes on trade and economic aid. The security votes included those on NATO, military spending, and military aid but excluded those pertaining to trade, administrative issues, and economic aid. The results of the last two models in Table 5 indicate that the same patterns evident in the broader voting scores were also evident in votes on security and economic issues considered separately. The differences are minor. The coefficients vary slightly for each set of votes, and the variable indicating foreign direct investment narrowly misses statistical significance in the model of voting on security issues. However, all the other variables are significant, and all their signs are the same.
Interaction Effects: The Relationship Between Party and Economic Structure
I argued earlier that party mediated the influence of economic structure. The parties organize societal interests and link them to actual state policymakers. Republicans or Democrats should be more attuned to the interests most closely associated with their party. To test this argument, I examined interactions between party and all the economic variables. Table 6 presents the results. For brevity it includes only the significant effects. The party interaction effects of the two trade variables and two of the three direct investment variables were not statistically significant.
The first two models in Table 6 estimate the mediating influence of party on the effect of the international lending variables. Although the overall level of foreign lending had no significant effect when used alone in Table 3, model 1 in Table 6 indicates that it did have a significant effect when considered together with partisan differences in its influence on foreign policy preferences. Because the coefficients for the main variable and its interaction effect with party have opposite signs and similar coefficients, a change in the proportion of international lending by banks in their home state would have little influence on Democrats. Because the interaction effect is always zero for Republicans, however, an increase in foreign lending by banks in their home states would make Republican senators more likely to support the administration's foreign policy. Because this variable was not significant in the model presented in Table 3, it is a good example of why party and economic structure are complementary rather than competing influences. It would be incorrect to conclude that the overall level of international lending by banks in their home state had no effect on senators' foreign policy views. However, its effects differed greatly depending on the senator's party affiliation. These partisan differences obscure the role of international lending when this interaction effect is not considered.
Model 2 indicates that the effects of banks' regional orientation also varied by party. As in the models presented in Table 3, a greater proportion of Asian lending by banks in their home state made senators more skeptical of the administration's foreign policy, whereas a greater proportion of European lending increased support for this policy. As was the case with international activity as a proportion of total lending, however, these effects varied by party. Banks' regional orientation had more influence on Republicans than on Democrats. Both the Asian and European lending variables have the opposite sign from their interaction effects with party. A change in the regional orientation of banks in their home state would have produced a larger effect on Republicans, on whom the effect of the interaction term is always zero, than it would on Democrats.
This evidence that lending activity in their home state influenced Republican senators' foreign policy attitudes supports the general observation that Republicans from the northeastern states, where banks lent most heavily to foreign borrowers, tended to be more internationalist than other Republicans. Of the sixteen Republicans with positive voting scores, indicating more frequent support than opposition to the administration's foreign policy, ten were from the Northeast. They included such well known internationalists as John Foster Dulles and Henry Cabot Lodge.
Although the evidence is somewhat weaker, party also appears to mediate the influence of direct investors. Of the three direct investment variables used in Table 3, only the variable indicating the proportion of firms with either no foreign subsidiaries or with subsidiaries involved in extractive activities in less developed areas had a significant interaction effect with party. Model 3 in Table 6 estimates this effect. As was the case with the lending variables, the opposite signs of the main effect and the interaction term indicate that these direct investor interests had more influence on Republicans than on Democrats. Firms with no direct investments outside the Western Hemisphere, or investments in extractive industries, strongly influenced Republicans against the administration's policy but had no significant effects on Democrats.
Conclusion
Conflicting economic interests, mediated by political party, help shape foreign policy preferences. Economic interests matter even when security issues seem paramount, as they did in the United States during 1949 and 1950. Not all U.S. firms and industries stood to benefit equally from the Truman administration's costly and ambitious foreign policy. Senators linked to those benefiting less tended to oppose it. In particular, the administration's opponents tended to come from states with relatively large import-competing sectors, more Asian-oriented banks, more large corporations with few international investments, and more investors in Third World extractive industries. Some of these interests were more strongly associated with the Republican party and had little influence on the preferences of Democrats. On the other hand, administration supporters tended to come from states with more European-oriented banks and internationally oriented corporations. The Cold War obviously involved important security issues. However, these security concerns did not produce a unitary set of national preferences. The pattern of support and opposition to U.S. foreign policy simply cannot be explained in terms of "core values" or security imperatives.
Economic interests alone do not determine foreign policy preferences but interact with political party in important ways. Simply examining the economic structure of a senator's home state does not necessarily reveal his or her preferences on foreign policy issues. Because some firms and industries are especially influential with senators from one party, party acts as an organizing influence on economic structure. International lenders and large firms whose investments were not protected by the administration's foreign policy were especially influential among Republicans. The statistical relationships between foreign policy preferences and economic structure are not always clear until the mediating effect of party is included in the model. Economic interests and party do not represent incompatible sources of explanation for foreign policy preferences. They are closely related and should be understood together.
These findings suggest that the role of international threats and interests in foreign policy making is poorly understood in much of the existing literature on national security policy. International threats, including military ones, may have a variable impact on different parts of society. The interests driving a state's foreign policy are not necessarily "national" interests, even on security questions. Perhaps because of the neorealist emphasis on the survival of unitary actors in the international system, threats and interests that do not affect the entire society have been neglected in theory development. The protection of values vital to some elements of society but less important to others may still require means that fall under the misleading rubric of "national security policy." Administration policymakers believed that insuring the openness of Europe to U.S. trade and investment required a credible military deterrent as well as a means of coping with the dollar gap. Political actors linked to interests holding a smaller stake in Europe were willing to risk this outcome rather than pay the enormous costs of the administration's strategy to prevent it.
Understanding international threats in terms of their impact on an assumed unitary state seriously limits the existing literature on domestic and international influences on foreign policy. The central issue is not whether domestic or international factors are more important - both obviously matter - but the appropriate unit of analysis for understanding foreign policy. Focusing on the interaction among domestic actors rather than assuming a unitary state has some important advantages. In fact, domestic actors have different interests and face different sets of threats in the international system. Where one perceives dire threats and crucial interests, others may see nothing more than a dangerous and costly foreign entanglement. The influence of international conditions on policy depends on the interests comprising the dominant coalition. A purely systemic analysis of foreign policy, which leaves out the character of this dominant coalition, is informative only when international conditions affect all domestic actors in the same way. These cases are rare, and the early Cold War era in the United States is not among them. The practice of treating a particular policy position as if it represented the "perceived national interest" only begs critical questions about the reasons for this perception and how those holding it came to determine policy. Political conflict between domestic actors determines the goals of foreign policy, and economic interests play an important role in shaping this conflict.
The relationship between the underlying determinants of policy outcomes and the motives of individual policymakers is not necessarily a simple one. Factors that decision makers never consciously consider may nevertheless shape political debate and decisively influence policy. If those deciding how to respond to international conditions share basic assumptions about policy goals, they may never even discuss their masons for holding these goals. In all probability, both the supporters and opponents of the Truman administration's foreign policy simply believed they were pursuing the "national interest" with the support of like-minded fellow citizens. However their members may have understood them, though, these political coalitions reflected the interests of particular segments of the domestic political economy. Even if political actors never explicitly consider the socioeconomic origins of their policy views, these socioeconomic factors may be just as important in shaping policy outcomes as are many considerations that they do discuss. This point is crucial for historical research on the development of policy, because a theoretically uninformed reading of the archival record will not reveal the importance of these undiscussed socioeconomic factors.
Appendix A: Roll-call Votes on Foreign Policy During the 81st Congress (1949-50)
A total of 92 roll-call votes on foreign policy issues, shown in Table A. 1, took place during the 81st Congress. Affirmative votes are coded "1" if the vote supported the administration's internationalist foreign policy, and "0" if it was an effort to oppose or limit the administration's foreign policy.
Appendix B: Economic Data on the Early Postwar Era
Economic data from the 1940s raise some important issues that require explanation. Although the limitations of these data are important, they do not diminish the statistical relationships presented here. If anything, these limitations should make it more difficult to find statistically significant relationships.
Data on International Trade
Reliable figures on exports and imports for each state are not available. To estimate the size of the export and import-competing sectors in each state, I first converted the statistics on exports and imports into Standard Industrial Classification (SIC) sectors. During the 1940s, imports were recorded under a classification system known as Schedule A. Exports were recorded under a different system, Schedule B. These systems organize exports and imports by commodity, rather than by the industry that produced them, as in the SIC system. The differences between these classification systems make some error unavoidable in converting trade data into the SIC system. Schedule A and Schedule B do not always specify commodities in ways that facilitate this conversion. In some cases, a single commodity classification includes items produced by more than one SIC sector. This problem exists even when converting trade data into the very general two-digit SIC sectors and becomes more serious as one moves to the narrower classifications at the three- and four-digit levels. In the 1960s, the adoption of a more easily convertible system for recording data on trade resolved some of these difficulties. Unfortunately, data in this form are not available for earlier historical periods.
Using the converted trade and industrial census data, I calculated the value of the goods exported by each three-digit SIC sector as a proportion of the total value added in that sector in the United States. Similarly, I calculated the value of imported goods as a proportion of the total value added by the industrial sector producing those goods domestically. Assuming that each sector's export orientation and import sensitivity were constant across states, I then estimated the state's overall export orientation and import sensitivity based on the size of each sector as a proportion of total value added in the state. The indicator of the size of the export sector, then, is the sum, over all sectors in each state's economy, of the proportion of the sector's products exported multiplied by the size of the sector as a proportion of all value added in the state. Similarly, import sensitivity is the sum, over all sectors in a state's economy, of imports as a proportion of total value added in the sector multiplied by the size of the sector as a proportion of all value added in the state.
Data on International Lending
The Treasury Department released annual data on foreign lending in each Federal Reserve district under the Freedom of Information Act. The Federal Reserve requires member banks to submit monthly information on their "claims on foreigners." The forms they released contain figures on both "short-term claims," with a maturity less than one year, and "long-term claims," with a maturity of more than one year, for each applicable country and region. Each Federal Reserve Bank aggregates these data and sends them to the Treasury Department. The Treasury Department released to me the year-end figures for each Federal Reserve district.
In order to approximate each state's share of the total claims on foreigners in their Federal Reserve district, I multiplied the district figure by each state's share of total banking activity in the Federal Reserve district. I also assumed that the noncoastal states included in the San Francisco Federal Reserve district did not participate in international lending activity.(40) In order to control for the total volume of lending activity in each state, I divided total short- and long-term claims on foreigners by the total loans and discounts for the states in the Federal Reserve district. The resulting variables are estimated figures for claims on foreigners, on Asians, or on Europeans, by banks in the state as a proportion of their total lending activity.
Data on Direct Foreign Investment
To develop an indicator of the level of foreign direct investment by corporations headquartered in each state, I relied mostly on the data provided by the annual Moody's Industrial Manual and the comparable volumes on utilities, merchandising, and railroads. I examined the two hundred firms with the greatest assets according to the earliest Fortune listing, 1954. I categorized each based on their ownership of subsidiaries outside the Western Hemisphere or in extractive industries in less developed areas.
Appendix C: Ideology and Economic Structure
The inclusion of ideology poses some thorny theoretical and methodological questions. Because politicians frame nearly all their policy positions in terms of shared ideas and beliefs, it is difficult to tell genuine ideological stands from rationalizations for constituent pressure or coalitional dynamics. Whatever the reasons for particular members' ideological convictions, the ideological makeup of Congress reflects in part the influence of interests and political coalition building. If only politicians with particular views can gain the party and interest group backing needed to succeed in politics, this selection process may determine the ideological composition of the Congress even without directly influencing any individual. In addition to these conceptual issues, indicators of ideology derived from interest group voting scores also pose serious problems.(41)
To cope with these problems I first regressed the ADA scores on the other independent variables I used. Table C.1 contains the results of this regression. The residual values of the ADA scores represent the proportion of this particular voting score not explained by these other variables. Of course, this model does not exhaust the range of exogenous influences on either ideology or ADA scores. Since ideology, like party, links positions on a wide range of domestic and foreign policy issues, the influence of groups interested in issues other than foreign policy may be paramount. For example, model 2 indicates that a strong labor movement in their home states made senators more liberal. The point here is not to develop a complete model of ideology, but only to discount the voting scores used to indicate it for the influence of the other variables used here.
I then selected those senators who had residual ADA scores more than one standard deviation from their predicted value. This produced a set of eighteen "liberals" with unusually high residual ADA scores, and thirteen "conservatives" with unusually low ADA scores. Table C.2 lists the names, states, and party affiliations of these thirty-one senators. Examining the behavior of these sets of ideological senators, rather than simply using the residual ADA scores for the entire Senate, has some important advantages. First, it allows the effects of liberal and conservative ideology to vary independently. There is no reason to assume that liberals and conservatives will act as mirror images of one another. Also, examining groups of strong liberals and strong conservatives avoids some of the problems created by the formulation of these scores. Because the ADA treats absences in the same way as a vote against their position, the meaning of small changes in ADA score does not necessarily reflect ideology. The variables I used in the models presented here ignore these small differences.
TABLE C.2. Senate liberals and conservatives selected using residual ADA scores
"Liberals"
George Aiken (R-VT) Paul Douglas (D-IL) Frank Graham (D-NC) Lister Hill (D-AL) Hubert Humphrey (D-MN) Estes Kefauver (D-TN) Harley Kilgore (D-WV) William Langer (R-ND) Henry Cabot Lodge (R-MA) Scott Lucas (D-IL) George Malone (R-NV) Wayne Morse (R-WA) Matthew Neely (D-WV) Claude Pepper (D-FL) Margaret Chase Smith (R-ME) John Sparkman (D-AL) Glen Taylor (D-ID) Elbert Thomas (D-UT) Edward Thye (R-MN)
"Conservatives"
Harry Byrd (D-VA) Sheridan Downey (D-CA) John Foster Dulles (R-NY) James Eastland (D-MS) J. Allen Frear (D-DE) Walter George (D-GA) Clyde Hoey (D-NC) Edwin Johnson (D-CO) John McClellan (D-AR) A. Willis Robertson (D-VA) Richard Russell (D-GA) John Stennis (D-MS)
Note: Those listed had residual ADA scores more than one standard deviation from the mean.
I would like to thank Timothy McKeown, Erik Devereux, Gavan Duffy, Thomas Ferguson, David Gibbs, Eric Mlyn, Gregory Nowell, the editors of International Organization, and several anonymous reviewers for their comments and suggestions. The responsibility for any errors contained here is mine.
1. Waltz 1979.
2. See, for example, Kennedy 1987; Snyder 1991; and McKeown 1991.
3. Foreign Relations of the United States 1950, I, 252.
4. Nitze 1980, 172.
5. Gowa 1994.
6. See Frieden 1994; and Gibbs 1990.
7. See McKeown 1984; and Baldwin 1985.
8. Nelson 1988, 800-808.
9. Magee, Brock, and Young 1989.
10. Magee, Brock, and Young 1989, 101-10.
11. Sectoral conflict arguments are often used to explain foreign economic policy. Since James Kurth's seminal article on the topic, many other scholars have made related arguments about sectoral conflict; see Kurth 1979. Ferguson and Frieden link interwar U.S. foreign economic policy to competing blocs of capital-intensive, internationally oriented firms, and labor-intensive, domestically oriented industries; see Ferguson 1984; and Frieden 1988. Gourevitch relates the policy responses to economic crises in the United States and Western Europe to the coalitions among various industrial sectors; see Gourevitch 1986. Many others, including Baldwin; Cassing, McKeown, and Ochs; and Milner have addressed the influence of differently situated industries in the development of trade policy; see Baldwin 1985; Cassing, McKeown, and Ochs 1986; and Milner 1988. Whereas most recent work on sectoral conflict has focused primarily on foreign economic policy, some classic accounts of foreign policy link sectoral conflict to states' broader international orientation; see Hobson [1902] 1965, 46-63; and Kehr 1977. A few recent authors have also applied the sectoral conflict approach more broadly; see Gibbs 1990; Snyder 1991; Nowell 1994; and Cox 1994.
12. Concerning the significance of this debate and how it was resolved, see Fordham 1998. Regarding the administration's congressional opponents, see Doenecke 1979; Eden 1984, 1985; and Kepley 1988.
13. See Leffler 1992; and Gaddis 1982.
14. See Hogan 1987; and Cumings 1990.
15. Both Cumings and Hogan cite Ferguson's work; see Cumings 1990, 18-19; and Hogan 1987, 10-11.
16. Concerning the use of "nationalists" and other terms, see Eden 1985; and Cumings 1990. Alternatively, one might level the normative playing field by referring to the internationalists as "imperialists."
17. Concerning the connection between the balance-of-payments crisis and rearmament, see Block 1977, 1980. Concerning the extension of U.S. commitments to Third World economic "hinterlands" for Western Europe and Japan, see Borden 1984; Rotter 1987; and McGlothlen 1993. For more thorough reviews of this literature, see Cumings 1993; Eden 1993; Gaddis 1983; and Jones and Woods 1993.
18. Eden 1984,1985.
19. Cumings 1990, 23-24, 97-100.
20. Frieden 1994.
21. Block 1980, 54. For similar accounts of the significance of the "Who Lost China?" debate, see Cumings 1990, 97-121; and Ellsberg 1972, 82-103; among others.
22. Block 1980.
23. See, for example, Clausen 1973; Clausen and Van Horn 1977; Schneider 1979; Poole and Rosenthal 1985, 1991; McCormick and Wittkopf 1992; and Hinich and Munger 1994.
24. For a discussion of the foreign policy outlooks of liberals during the early Cold War era, see Hamby 1973; and McAuliffe 1978. For a discussion of the outlook of conservatives, see Doenecke 1979; Eden 1984, 1985; and Cumings 1990, esp. 79-121.
25. See, for example, Bensel 1984; Eden 1985; Agnew 1987; Trubowitz 1992; and Trubowitz and Roberts 1992.
26. See, for example, Trubowitz 1992; Eden 1985; and many older accounts such as Rieselbach 1966; and Smuckler 1953.
27. Kurth 1979, 33.
28. See Ferguson 1984, 1995; and Frieden 1988.
29. See Rabinowitz 1978; and Poole and Rosenthal 1984. Other theories that account for the "empty center" include Rabinowitz and MacDonald 1989; and Hinich and Munger 1994, esp. 77-79.
30. Bensel 1984, 175-76, 222-55.
31. Remarks on National Public Radio, Morning Edition, 17 February 1994. Concerning the Senate during the 81st Congress, see Kepley 1988.
32. Frieden 1994.
33. Cumings 1990.
34. Midwestern and Mountain states include Arizona, Colorado, Idaho, Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Montana, Nebraska, Nevada, New Mexico, North Dakota, Ohio, South Dakota, Utah, Wisconsin, and Wyoming. Southern states include Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, and West Virginia.
35. Jackson and Kingdon 1992.
36. Baldwin 1985, 63-67.
37. Stanley Botner to Fred Hobart, 2 October 1950; and Everett Bellows to Stanley Botner, 11 October 1950; both Harriman Papers, Box 309, Library of Congress. Concerning the social background and interest group ties of administration officials and others involved in the making of U.S. foreign policy, see Isaacson and Thomas 1986; Schulzinger 1984; Burch 1980; and Shoup and Minter 1977.
38. Ferguson 1983, 16.
39. McKeown 1994.
40. The San Francisco Federal Reserve district is extremely large, including Arizona, Idaho, Nevada, and Utah, in addition to the coastal states of California, Oregon, and Washington. The volume of financial activity in the noncoastal states of this district is extremely low and probably includes very little, if any, international lending. The Kansas City district, which includes Colorado, Nebraska, Oklahoma, and Wyoming, is probably more like these four states than is the Pacific Coast. This district did not even report on its foreign lending activity because the total volume was less than $100,000.
41. Jackson and Kingdon 1992.
References
Agnew, John. 1987. The United States in the Worm Economy. New York: Cambridge University Press.
Baldwin, Robert E. 1985. The Political Economy of U.S. Import Policy. Cambridge, Mass.: MIT Press.
Bensel, Richard F. 1984. Sectionalism and American Political Development. Madison: University of Wisconsin Press.
Block, Fred L. 1977. The Origins of International Economic Disorder. Berkeley: University of California Press.
-----. 1980. Economic Instability and Military Strength: The Paradoxes of the 1950 Rearmament Decision. Politics and Society 10:35-58.
Borden, William S. 1984. The Pacific Alliance. Madison: University of Wisconsin Press.
Burch, Philip H. 1980. Elites in American History. Vol. 3. New York: Holmes & Meier Publishers, Inc.
Cassing, James, Timothy J. McKeown, and Jack Ochs. 1986. The Political Economy of the Tariff Cycle. American Political Science Review 80:843-62.
Clausen, Aage R. 1973. How Congressmen Decide. New York: St. Martin's Press.
Clausen, Aage R., and Carl E. Van Horn. 1977. The Congressional Response to a Decade of Change: 1963-1972. Journal of Politics 39:624-66.
Cox, Ronald W. 1994. Power and Profits: U.S. Policy in Central America. Lexington: University of Kentucky Press.
Cumings, Bruce. 1990. The Origins of the Korean War. Vol. 2. Princeton, N.J.: Princeton University Press.
-----. 1993. "Revising Postrevisionism," or the Poverty of Theory in Diplomatic History. Diplomatic History 17:539-69.
Doenecke, Justus. 1979. Not to the Swift. Lewisburg, Pa.: Bucknell University Press.
Eden, Lynn. 1984. Capitalist Conflict and the State: The Making of United States Military Policy in 1948. In Statemaking and Social Movements, edited by Charles Bright and Susan Harding, 233-61. Ann Arbor: University of Michigan Press.
-----. 1985. The Diplomacy of Force. Ph.D. diss., University of Michigan, Ann Arbor.
-----. 1993. The End of U.S. Cold War History. International Security 18:174-207.
Ellsberg, Daniel. 1972. Papers on the War. New York: Simon & Schuster.
Ferguson, Thomas. 1983. Party Realignment and American Industrial Structure. Research in Political Economy 6:4-120.
-----. 1984. From Normalcy to New Deal: Industrial Structure, Party Competition, and American Public Policy in the Great Depression. International Organization 38:59-85.
-----. 1995. Golden Rule. Chicago: University of Chicago Press.
Fordham, Benjamin O. 1998. Building the Cold War Consensus. Ann Arbor: University of Michigan Press.
Frieden, Jeffry A. 1988. Sectoral Conflict and United States Foreign Economic Policy. International Organization 42:59-90.
-----. 1994. International Investment and Colonial Control: A New Interpretation. International Organization 48:559-94.
Gaddis, John L. 1982. Strategies of Containment. New York: Oxford University Press.
-----. 1983. The Emerging Post-Revisionist Consensus on the Origins of the Cold War. Diplomatic History 7:171-90.
Gibbs, David N. 1990. The Political Economy of Third World Intervention. Chicago: University of Chicago Press.
Gourevitch, Peter. 1986. Politics in Hard Times. Ithaca, N.Y.: Cornell University Press.
Gowa, Joanne. 1994. Allies, Adversaries, and International Trade. Princeton, N.J.: Princeton University Press.
Hamby, Alonzo L. 1973. Beyond the New Deal. New York: Columbia University Press.
Hinich, Melvin J., and Michael C. Munger. 1994. Ideology and the Theory of Political Choice. Ann Arbor: University of Michigan Press.
Hobson, J. A. [1902] 1965. Imperialism: A Study. Reprint, Ann Arbor: University of Michigan Press.
Hogan, Michael J. 1987. The Marshall Plan. New York: Cambridge University Press.
Isaacson, Walter, and Evan Thomas. 1986. The Wise Men. New York: Simon & Schuster.
Jackson, John E., and John W. Kingdon. 1992. Ideology, Interest Group Scores, and Legislative Votes. American Journal of Political Science 36:805-23.
Jones, Howard, and Randall B. Woods. 1993. Origins of the Cold War in Europe and the Near East: Recent Historiography and the National Security Imperative. Diplomatic History 17:251-76.
Kehr, Eckart. 1977. Anglophobia and Weltpolitik. In Economic Interest, Militarism, and Foreign Policy, translated by Grete Heinz and edited by Gordon Craig, 22-49. Berkeley: University of California Press.
Kennedy, Paul. 1987. The Rise and Fall of the Great Powers. New York: Random House.
Kepley, David R. 1988. The Collapse of the Middle Way. New York: Greenwood Press.
Kurth, James. 1979. The Political Consequences of the Product Cycle: Industrial History and Political Outcomes. International Organization 33:1-34.
Leffler, Melvyn P. 1992. A Preponderance of Power. Stanford, Calif.: Stanford University Press.
Magee, Stephen P., William A. Brock, and Leslie Young. 1989. Black Hole Tariffs and Endogenous Policy Theory. New York: Cambridge University Press.
McAuliffe, Mary Sperling. 1978. Crisis on the Left. Amherst: University of Massachusetts Press.
McCormick, James M., and Eugene R. Wittkopf. 1992. At the Water's Edge: The Effects of Party, Ideology, and Issues on Congressional Foreign Policy Voting, 1947-1988. American Politics Quarterly 20:26-53.
McGlothlen, Ronald L. 1993. Controlling the Waves. New York: W. W. Norton.
McKeown, Timothy J. 1984. Firms and Tariff Change: Explaining the Demand for Protection. World Politics 36:215-33.
-----. 1991. The Foreign Policy of a Declining Power. International Organization 45:257-79.
-----. 1994. The Epidemiology of Corporate PAC Formation, 1975-1984 Journal of Economic Behavior and Organization 24:153-68.
Milner, Helen V. 1988. Resisting Protectionism. Princeton, N.J.: Princeton University Press.
Nelson, Douglas. 1988. Endogenous Tariff Theory: A Critical Survey. American Journal of Political Science 32:796-837.
Nitze, Paul H. 1980. The Development of NSC 68. International Security 4:170-76.
Nowell, Gregory P. 1994. Mercantile States and the World Oil Cartel, 1900-1939. Ithaca, N.Y.: Cornell University Press.
Poole, Keith T., and Howard Rosenthal. 1984. The Polarization of American Politics. Journal of Politics 46:1061-79.
-----. 1985. A Spatial Model for Legislative Roll Call Analysis. American Journal of Political Science 29:357-84.
-----. 1991. Patterns of Congressional Voting. American Journal of Political Science 35:228-78.
Rabinowitz, George. 1978. On the Nature of Political Issues: Insights from a Spatial Analysis. American Journal of Political Science 22:793-817.
Rabinowitz, George, and Stuart MacDonald. 1989. A Directional Theory of Voting. American Political Science Review 83:93-121.
Rieselbach, Leroy N. 1966. The Roots of Isolationism. Indianapolis, Ind.: Bobbs-Merrill.
Rotter, Andrew J. 1987. The Path to Vietnam. Ithaca, N.Y.: Cornell University Press.
Schneider, Jerrold E. 1979. Ideological Coalitions in Congress. Westport, Conn.: Greenwood Press.
Schulzinger, Robert D. 1984. The Wise Men of Foreign Affairs. New York: Columbia University Press.
Shoup, Laurence H., and William Minter. 1977. Imperial Brain Trust. New York: Monthly Review Press.
Smuckler, Ralph H. 1953. The Region of Isolationism. American Political Science Review 47:386-401.
Snyder, Jack. 1991. Myths of Empire. Ithaca, N.Y.: Cornell University Press.
Trubowitz, Peter. 1992. Sectionalism and American Foreign Policy: The Political Geography of Consensus and Conflict. International Studies Quarterly 36:173-90.
Trubowitz, Peter, and Brian Roberts. 1992. Regional Interests and the Reagan Military Buildup. Regional Studies 26:555-67.
U.S. Superintendent of Documents. 1977. Foreign Relations of the United States. Washington, D.C.: Government Printing Office.
Waltz, Kenneth. 1979. A Theory of International Relations. Reading, Mass.: Addison-Wesley.
Benjamin O. Fordham is Assistant Professor of Political Science at the University at Albany, State University of New York.
COPYRIGHT 1998 World Peace Foundation; Massachusetts Institute of Technology
COPYRIGHT 1998 Information Access Company