OUP user menu

Wedges and Widgets: Liberalism, Libertarianism, and the Trade Attitudes of the American Mass Public and Elites

Brian Rathbun
DOI: http://dx.doi.org/10.1111/fpa.12037 85-108 First published online: 2 February 2016


What are the ideological sources of free trade attitudes? Free trade plays a crucial role in classical liberal theory as a way of increasing the prospects of peace between states. Are liberal individuals more supportive of free trade? The literature on foreign policy beliefs largely neglects the question of trade, and those exceptions that find support for the liberal hypothesis generally rely on faulty conceptualization. Using surveys of the American mass public and American elites, this article finds that the combination of views that marks classical liberalism does not in fact predict support for free trade at either the mass or the elite level. Support for free trade at the mass level has libertarian, not liberal, foundations, predicted by a combination of social and economic libertarianism. At the mass level, the combination of cosmopolitanism and dovishness that constitutes foreign policy liberalism has no effect on trade attitudes. At the elite level, cosmopolitanism is actually generally negatively associated with support for free trade. Free trade is a wedge issue that creates strange alliances at the elite level between cosmopolitans and isolationists generally hostile to one another on foreign policy and at the mass level between social and economic libertarians typically antagonistic to each other's domestic agenda.

Free trade is not only an interest. It is one of the most prominent ideas in classical economic theory. Basic classes in microeconomics almost always begin with the proverbial manufacturers of widgets who gain from exchanging the item in which they enjoy comparative advantage in production for other needs and wants. And what is true within a country is true between them. The notion that there is mutual benefit in specializing in producing certain goods or services serves as the foundation of any introduction to economics. Free trade, however, is also a contested idea. There has always been considerable debate about whether the free exchange of goods and services across borders is a social good. Therefore, like all policy positions drive by broader principles about what is good for society as a whole, free trade is likely ideological. As such, it should be tied into broader organizing frameworks of political ideas.

Free trade has historically occupied a crucial place in international relations thinking as a means for improving relations between states and increasing the prospects for peaceful conflict resolution. By generating interdependence, trade raises the costs of war. Free trade also encourages cosmopolitanism by encouraging individualistic rather than group-centered thinking and facilitating contacts with others abroad. This historical association is still echoed in contemporary liberal international relations scholarship, in which the commercial interdependence fostered by free trade serves as one of the legs of the “Kantian tripod” in the promotion of peace, alongside multilateral organizations and the promotion of democracy (Wilson 1995; Doyle 1997; O'Neal and Russett 1997; Doyle 1997; Russett, O'Neal, and Davis 1998; Osiander 1998; Rathbun 2010).

Is the same combination of attitudes present at the individual level? Are beliefs about trade predicted by commitments to classical liberal principles of foreign affairs such as multilateralism? Scholars have devoted considerable attention in testing empirically whether trade actually encourages peace, but are those individuals most interested in peace and cooperation more ideologically favorable to free trade? Do we, for instance, see a kind of “folk liberalism,” just as we see a “folk realism” (Kertzer and McGraw 2012)? It is a well-established finding in the literature on foreign policy beliefs that attitudes on international relations demonstrate a particular structure that does not vary by level of sophistication, applying to masses and elites both (Holsti and Rosenau 1988, 1990; Wittkopf 1990; Chittick, Billingsley, and Travis 1995; Rathbun 2007). However, this literature largely neglects the influence of these broader constructs on issues of international economy, focusing generally on international security issues.

Most research on individual attitudes toward trade in recent years has centered on whether we can predict positions based on the personal, pocket-book effects of free exchange across borders. Scholars have devoted most of their attention to the individual material factors that might explain attitudes, such as whether one stands to personally benefit based on his or her position in the global political economy. It is thought that those who have higher skill levels generally profit from a more open international political economy.

Only a few studies have explored the ideational foundations of support, and not in a systematic way (Scheve and Slaughter 2001; Mayda and Rodrik 2005; Hainmueller and Hiscox 2006). Isolationism and hawkishness are found to be negatively associated with support for free trade, leading scholars to conclude that support for free trade is indeed promoted by more inclusive, cosmopolitan ideological viewpoints (Herrmann, Tetlock, and Diascro 2001; Mansfield and Mutz 2009). Yet there are methodological reasons to be cautious of drawing conclusions based on these results. In terms of the former, neither isolationism nor hawkishness are the conceptual opposite of cosmopolitanism, so there has been no thorough test of the association of the latter's influence on free trade.

Studies of trade attitudes also continually and curiously neglect perhaps the most obvious source of ideological support for free trade, a belief in the benefits of free markets and liberty in general. They overlook the “free” in free trade. Free trade, like democracy and capitalism, is about choice. Constructivist scholars of international relations have found that countries often export their domestic economic thinking abroad. For instance, social democrats favor greater foreign aid, based on a general embrace of humanitarianism and concern for others’ welfare that links both (Lumsdaine 1993). The same might be true at the individual level of free trade.

This paper makes a more systematic effort at understanding the ideology of free trade, drawing on work that is generally not given much attention in the international relations literature — research in political philosophy and liberal theory, particularly that of Dworkin (1977, 1985). I ask whether attitudes on free trade, at both the mass and elite levels, are predicted by support for classical liberalism or libertarianism. Liberalism is the ideological commitment to guaranteeing equal concern and respect for all individuals. It is expressed most clearly in efforts by the state to ensure that no one's view of the good life is privileged over others. This takes the form of both limiting any governmental interference that privileges some over others, through, for instance, guaranteeing freedom of speech and assembly. But it also involves governmental intervention, such as through social insurance, in instances in which the exercise of untrammeled freedom generates profound inequalities. In terms of foreign policy, liberalism is evident in a cosmopolitan identification with the welfare and rights of those beyond one's shores, dovish support for peaceful conflict resolution and a belief that multilateral cooperation promotes joint gains. Libertarianism, by contrast, is the view that individuals should be free to pursue the path that they choose without interference, particularly by government. Its most obvious manifestation is untrammeled support for free market mechanisms. Crudely, libertarians feel that they do not owe others anything and vice versa. Liberalism and libertarianism are related historically and their advocates agree on many policies, such as support for civil liberties at home. However, they are conceptually distinct and often at loggerheads in practice.

Utilizing surveys of both the American mass public and elites, I find that contemporary support for free trade does not have liberal foundations. Foreign policy liberalism, indicated by support for peaceful conflict resolution abroad and cosmopolitan cooperation, does not predict the belief that free trade presents an opportunity for people in rich and poor countries at the mass level. Among elites, an analysis of the Foreign Policy Leadership Project survey reveals that while doves are somewhat more predisposed toward free trade than hawks, cosmopolitanism does not induce these individuals to support free trade, except when it is framed in terms of joining multilateral trading arrangement such as the WTO. In fact, cosmopolitanism is negatively associated with support for free trade generically. Isolationism is strongly associated with opposition to free trade in all forms among both the masses and elites.

Libertarianism is associated with pro-free trade attitudes at the mass level, but not at the elite level. Economic libertarianism is found to be a very strong and positive predictor of the notion that free trade presents opportunities for the poor to thrive. More striking is the finding that social libertarianism also consistently predicts trade attitudes. This indicates that support for free trade is about more than just free market economics, at least at the level of the mass public. It is about a broader ideological commitment to freedom in general. At the elite level, however, only social conservatism has an effect.

These findings reveal that free trade is a wedge issue. At the elite level, it creates Baptist–bootlegger coalitions between isolationists and cosmopolitans who are on the opposite side of many, perhaps most, other foreign policy issues. At the mass level, it generates curious alliances between economic conservatives (also known as economic libertarians) and social libertarians, both of whom support greater individual choice in politics, even as the two groups are generally found on opposite sides of the domestic political spectrum.

“Liberty! Equality! International Fraternity!”?: Liberalism and Free Trade

There has been a surge of interest recently among scholars of international political economy in explaining individual-level support of trade policy. They have applied models of the international economy such as Hecksher–Ohlin, Stolper–Samuelson, and Ricardo–Viner that were heretofore used to generate expectations about the domestic political cleavages on foreign economic issues such as trade or exchange rates among broader collective groupings such as parties, industries or factors of production (Rogowski 1989; Frieden 1991; Frieden and Rogowski 1996; Hiscox 2001, 2002; Lake 2010). The core finding of this literature was that those industries or factors that are more competitive in the international economy benefit from a more open international trade regime and a commitment to free trade. Applying the Stolper–Samuelson model at the individual level, Scheve and Slaughter (2001) hypothesize that high-skilled individuals fare better than low-skilled workers in an integrated global marketplace and will support free trade. Using education as a proxy, they find confirmation for their hypothesis, although subsequent research argues that education captures much more than simply skill levels (Hainmueller and Hiscox 2006).

Political economy models are undoubtedly important in explaining those aspects of trade attitudes that are based on personal, pocketbook concerns. However, this hardly exhausts the reasons why one might personally support or reject free trade. For instance, support for free trade might be a part of a general endorsement of the merits of competition and the importance of market mechanisms for optimally distributing resources. However, with a few exceptions, such ideological variables have generally been lacking in studies to date.

Historically, free trade plays a prominent role in classical liberal theorizing and advocacy in international affairs (Zacher and Matthew 1995:113). Long explains, “Liberal internationalism had its roots in an extension to international relations of classical political economy of Adam Smith and in David Ricardo's theory of international trade, which emphasized the interests of each nation in concentrating on its comparative advantage in the worldwide division of labor. Richard Cobden extended the economic principle of free trade into a political doctrine for peace and prosperity through free trade and non-intervention” (1995:313). Bentham argued that utilitarian considerations would lead those who traded toward peace (Zacher and Matthew 1995:117). Scholars now refer to “commercial liberalism,” the broad notion that exchange between states reduces the chances of armed conflict by creating interdependence (Keohane 1990; Zacher and Matthew 1995; Doyle 1997). Why this association? What is it about the philosophical foundations of liberal thinking that leads liberals to stress the merits of free trade? Answering this question requires us to engage the literature on the ideological foundations of liberalism.

The core of liberal thinking is that all individuals must be treated with “equal concern and respect” (Dworkin 1977, 1985). Individuals should not be treated as objects or means but rather as ethical subjects (Doyle 1997:207). Every individual has value. In the liberal mind, freedom and liberty are derivative of a commitment to equal concern and respect. Equality begets liberty and restraints on authority. To restrict these freedoms is to deny equality. Howard and Donnelly explain, “Personal liberty, especially the liberty to choose and lead one's own life, clearly is entailed by the principle of equal respect: for the state to interfere in matters of personal morality would be to treat the life plans and values of some as superior to others” (1986:803).

This “liberal view of man” (Howard and Donnelly 1986:803) has major implications for politics, leading naturally to democratic governance. All citizens should have the equal right to participation in politics. Otherwise, public life will be hierarchical and discriminatory. All have the right to organize politically and voice their concerns. Individuals also must make their own choices about faith without interference by the state or the dominant religious group in a community. Liberals do not believe that there is a harmony of interests in political and social life. Liberalism is not synonymous with idealism (Keohane 1990; Zacher and Matthew 1995:109). However, they do contend that by restraining impulses toward imposing one's own views and interests on others, all will benefit more. As Owen writes: “Liberals have transformed, rather than transcended, selfishness” (1997:35).

Liberalism must be distinguished from libertarianism argues Dworkin (1977). He distinguishes between liberalism's “liberty as independence,” which stresses that individuals should be autonomous because they are entitled to equal concern and respect, and libertarianism's “liberty as license,” in which individuals are given the maximum possible free reign from social and legal constraints. “In this neutral, all embracing sense of liberty as license, liberty, and equality are plainly in competition. Laws are needed to protect equality, and laws are inevitably compromises of liberty” (Dworkin 1977:267). For this reason, libertarians are most insistent that governmental regulation, of all behavior, should be kept to an absolute minimum.

While libertarianism is sometimes treated as simply a laissez-faire variant of liberalism (Doyle 1997:206), Dworkin (1977, 1985), Howard and Donnelly (1986) argue that it is qualitatively different. As Doyle writes, “The Liberals were not the first philosophers to conceive of the value of human liberty or equality. What makes a Liberal a Liberal is making equal, nondiscriminatory liberty the center of one's political philosophy” (1997:306; emphasis added). Libertarianism is generally how economists conceive of the classical liberal position. And in the international political economy literature, the conceptualization of market activity, in which political actors each rationally seek their individual, material gain, what others have called the “open economy” model, is often described as liberalism, as Lake (2010) notes.

Equality and liberty might come into tension for liberals as well. However, equal concern and respect is considered more fundamental. Howard and Donnelly explain the essence of liberalism: “When taken too far, liberty as license is threatening to equal concern and respect. Liberty alone, however, cannot serve as the overriding value of social life, as the end to be maximized by political association. Liberty readily degenerates into license and social atomization unless checked by a fairly expansion, positive conception of the persons in relation to whom it is exercised. If liberty is to foster dignity, it must be exercised within the constraints of the principle of equal concern and respect. In fact, autonomy and equality are less a pair of guiding principles than different manifestations of the central liberal commitment to the equal worth and dignity of each and every person” (1986:803). Constraints on liberty as license are therefore sometimes necessary to ensure equal concern and respect. Consider, for instance, governmental efforts to ensure nondiscrimination on the basis of sex and race. Libertarians object to the restrictions on freedom that this entails. Libertarians and liberals are individualist, but in different ways. Liberals stress the value of the individual, which requires restrictions on freedom. Libertarians value individuals’ freedom to do as they please.

The tension, and its resolution in different directions, becomes most clear in contestation over the role to be played by the market in economic life. Market mechanisms are an important part of liberal thinking. “A certain amount of economic liberty is also required, at least to the extent that decisions concerning consumption, investment, and risk reflect free decisions based on personal values that arise from autonomously chosen conceptions of the good life,” write Howard and Donnelly (1986:805). Liberals prefer market mechanisms over statist forms of economic governance that privilege and favor the most powerful interests in society and preserve hierarchy. Historically, liberals drove the creation of market capitalism, opposed by mercantilistic and aristocratic interests.

However, when it generates inequalities that threaten equal concern and respect, liberals favor some degree of remedial action. “[M]arket distribution of resources can have grossly unequal outcomes. Inequality per se is not objectionable to the liberal, but the principle of equal concern and respect does imply a floor of basic economic welfare; degrading inequalities cannot be permitted” (Howard and Donnelly 1986:805). This does not make liberals into radical socialist communitarians. Howard and Donnelly (1986) explain how communist societies are not founded on the notion of individual rights, but rather communitarian duties. Liberals favor “pragmatic and selective intervention over a dramatic change from free enterprise to wholly collective decisions about investment, production, prices and wages,” writes Dworkin (1985:187). They are middle of the road on economic issues.

The core credos of equal concern and respect leads naturally to three liberal principles of international relations. First, as part of a resistance to coercion, liberals prefer peaceful conflict resolution between states. This is not always possible, but it is always desirable. Liberals are relatively more dovish in international affairs than their ideological opponents — conservatives and realists. Violence is the authoritarian imposition of one's views and interests. Conflicts are best resolved through reasoned dialogue. Of course, when they confront others who do not share their values, the lowest common denominator might prevail. Liberal states do fight illiberal states, for instance. The common denominator of all the seemingly disparate strands of liberal theorizing in international relations is the supposition that greater peace is possible (Rathbun 2012). International relations is “potentially progressive” (Keohane 1990:10).

Second, the belief in the inherent worth of all individuals generates a broader, cosmopolitan sense of identity. Liberalism is universalist. Locke writes of the individual, “He and the rest of all mankind are one Community, make up one society distinct from all other creatures. And were it not for the corruption and vitiousness of degenerate Men, there would be…no necessity that Men should separate from this great and natural Community, and by positive agreements combine into smaller and divided associations” (Doyle 1997:306–307). Moral psychologists stress how Enlightenment morality is based on an “ethics of autonomy” dedicated to protecting and caring for individuals rather than an “ethics of community” that values in-group solidarity and loyalty (Haidt and Joseph 2004; Graham, Haidt, and Nosek 2009; Haidt, Graham, and Joseph 2009; Graham, Nosek, Haidt, Iyer, Iyer, Koleva, and Ditto 2011). Howard and Donnelly write: “For the liberal, the individual is not merely separable from the community and social roles, but specially valued precisely as a distinctive, discrete individual, which is why each person must be treated with equal concern and respect” (1986:803). Liberals have a greater sense of global solidarity as a consequence. With the advent of liberal thinking, they argue, “No longer could persons be reduced to their roles, to parts of the community. With separate individuals, possessing special worth and dignity precisely as individuals, the basis for human rights was established” (1986:804). This does not mean that liberals cannot be intensely patriotic (Herrmann, Isernia, and Segatti 2009). However, liberalism is not consistent with a nationalistic sense of superiority, although it might generate an ideological self-righteousness, something that has fueled liberal imperialism historically (Doyle 1997).

Third, liberals stress the mutual gains from cooperation (Zacher and Matthew 1995:110, 117). By adopting general principles of behavior that respect the rights of others in international affairs, all might benefit more. Liberals are believers in reciprocity (Haidt and Joseph 2004; Graham et al. 2009, 2011; Haidt et al. 2009). In international affairs, this is most evident in liberals’ historic endorsement of multilateralism, what is sometimes called “regulative” or “institutional” liberalism (Keohane 1990; Zacher and Matthew 1995). Multilateralism involves commitment to generalized principles of conduct that prescribe certain rules to be taken regardless of the circumstances of the individual case (Ruggie 1992). For instance, collective security is premised on the idea that an attack on one is an attack on all. Although fulfilling that guarantee in practice in a particular instance might not be in a state's immediate short-term interest, it does benefit the collective welfare, and therefore each individual member's interest in the long term. It is the rising tide that lifts all boats.

Dovishness, cosmopolitanism, and multilateralism might each lead liberal individuals toward support for free trade. Committed to peaceful conflict resolution, liberals could embrace free trade as a means of creating interdependence that makes the use of force more costly. Consistent with their broader sense of identity, liberals might stress that one's pursuit of individual welfare in overseas exchange undermines nationalistic and statist attachments. The great liberal theorist Ricardo wrote that free trade “binds together, by one common ties of interest and intercourse, the universal society of nations throughout the civilized world” (quoted in Zacher and Matthew 1995:114). Smith denounced mercantilism for making “commerce, which ought naturally to be among nations, as among individuals, a bond of union and friendship…the most fertile source of discord and animosity” (Doyle 1997:234). And free trade is a multilateral commitment in that it is based on reciprocal obligations to lower trade barriers. It benefits both sides.

Previous studies on trade attitudes have alluded to these connections. On cosmopolitanism, Mansfied and Mutz write, “Activist foreign policy attitudes, a positive attitude toward out-groups, and a preference for open trade…all reflect a sense of cosmopolitanism and inclusion (Mansfield and Mutz 2009:451).On multilateralism, Herrmann, Tetlock and Diascro write, “It… is not surprising that free-traders tend to be cooperative internationalists given how closely ideas about global interdependence and gain through cooperation are connected to trade” (2001:196).

However, there are also reasons to think that classical liberalism might not lead to support for free trade. Indeed, the opposite might be true. Liberals, as cosmopolitan citizens of the world, might be resistant to free trade if it is thought to generate great inequalities between North and South and to undermine equity at home through the effect of global competition on wages. Certainly, the type of anti-globalization protesters we see at WTO, G-8, and IMF meetings embrace tolerance of international diversity and cosmopolitanism. The advocate journalist Naomi Klein, whose writings are the closest thing to an anti-free trade manifesto for the movement, explicitly denies the very label of “anti-globalization” as a consequence: “The irony of the media-imposed label ‘anti-globalization’ is that we in this movement have been turning globalization into a lived reality, perhaps more so than even the most multinational of corporate executives or the most restless of jet-setters…It is an intricate process of thousands of people tying their destinies together simply by sharing ideas and telling stories about how abstract economic theories affect their daily lives” (Klein 2002:8) Klein's cosmopolitanism is evident in her argument that it is the proponents of international free markets who want to construct the “fences” of private property across the world and that her cosmopolitan allies are the true ambassadors of globalization. In describing the protests, she writes:“Overnight, the sight is transformed into a kind of alternative global city where urgency replaces resignation, corporate logos need armed guards, people usurp cars, art is everywhere, [and] strangers talk to each other” (2002:9). While these attitudes are most likely not mainstream, it is not inconceivable that others will have similar, if less intense, reservations about free trade.

It might be that the association between free trade is a historical artifact. When it first arose, the liberal view of the merits of free trade was juxtaposed to nationalistic and mercantilistic defense of protectionism that privileged the state over society and the in-group over the individual and global welfare (Doyle 1997:214). To the extent that such economic statism is less of a force in international politics today, the contours of the debate might have shifted. The driving motivation of anti-globalization movements is an opposition to international free market mechanisms that generate inequality. As Klein writes, “The crisis respected no national boundaries. A booming global economy focused on the quest for short-term profits was proving itself incapable of responding to increasingly urgent ecological and human crises.” She goes on to assert that “the economic process that goes by the benign euphemism ‘globalization’ now reaches into every aspect of life, transforming every activity and natural resource into a measured and owned commodity” (2002:12).

Support for free trade might have libertarian foundations instead. Just as libertarians stress that the government intervention should be reduced to an absolute minimum at home, the role of the ‘night watchman,’ they might believe the same about the state's role in foreign economic policy. And lacking the same commitment to equal concern and respect, libertarians would be relatively untroubled by any adverse distributional consequences of free trade, for instance, in favor of rich individuals or rich countries. Just as libertarians bridle at restrictions on their economic and social choices at home, they might oppose governmental efforts to restrict their commercial intercourse across borders on ideological principle.

Data Analysis


To properly test whether free trade is predicted by classical liberalism, we need to call on the literature on foreign policy belief systems. Despite early studies that suggested skepticism (reviewed in Hurwitz and Peffley 1987), it is now generally believed that foreign policy beliefs are coherently structured, even at the mass level. Specific attitudes (in this case, about free trade) are a natural outgrowth of a number of more abstract postures toward foreign policy. Even if they are not up to date on all of the goings-on in world politics, individuals have “general stances,” “postures,” “dispositions” or “orientations” that allow us to predict their attitudes toward more discrete issues in international relations (Hurwitz and Peffley 1987; Hurwitz and Peffley 1990; Herrmann, Tetlock and Visser 1999). Foreign policy attitudes are marked by ideological coherence, in that the mass public possesses belief systems held together by certain core principles or values. These foreign policy attitudes may or may not be related to typical measures of political ideology, such as placement on a liberal-conservative scale, but they are ideological none the less in that, to use Converse's own words, “a few crowning postures…serve as a sort of glue to bind together many more specific attitudes and beliefs, and these postures are of prime centrality in the belief system as a whole” (1964:211).

The most prominent model of the structure of foreign policy beliefs by most accounts (Chittick et al. 1995:313; Murray, Cowden, and Russett 1999:458; Nincic and Ramos 2010:122) argues that foreign policy beliefs at both the mass and elite levels are structured along two related but distinct dimensions – cooperative internationalism (CI) and militant internationalism (MI). Originally articulated by Eugene Wittkopf, these Faces of Internationalism served as the subsequent basis for numerous studies of foreign policy attitudes (Holsti and Rosenau 1988, 1990; Wittkopf 1990; Murray 1996; Murray et al. 1999). Murray (1996) demonstrates that the structure of foreign policy attitudes persists even across vastly different epochs in international affairs, such as before and after the Cold War. This literature is particularly useful for our purposes as it includes all the elements of classical liberal thinking. Liberalism in foreign affairs is marked by the combination of a high degree of cooperative internationalism and a low degree of militant internationalism. This conjuncture of attitudes captures the three liberal principles of international relations – dovishness, cosmopolitanism, and multilateralism.

Wittkopf derived his dimensions inductively through the finding that particular items consistently defined the same dimensions in factor analyses of survey data. Yet recent work has more precisely defined the values and beliefs underlying these constructs. Cooperative internationalism can be defined as a sense of attachment to, concern for, and identification with those in other countries. It is cosmopolitanism. After all, a cosmopolitan is literally a “citizen of the world.” Chittick et al. (1995) have called it the “identity” dimension. They write, “What all these [CI] questions seem to have in common is a concern for the wider community. We believe that those who emphasize the importance of these goals have a more inclusive identity than those who de-emphasize these same goals” (1995:318). Nincic and Ramos (2010) write of “other-regarding” objectives. Rathbun defines this dimension as denoting a “sense of obligation to the broader international community” (2007:388). Global solidarity seems to be central to cooperative internationalism.

Yet cosmopolitanism is not self-abnegating and completely altruistic. Rather, cosmopolitans stress the joint gains to be had through cooperation. In previous studies, support for multilateral organizations and generic expressions of belief in international collaboration also load on the CI dimension (Holsti and Rosenau 1988, 1990; Wittkopf 1990; Rathbun 2007). Liberalism combines both self-regarding and other-regarding elements. CI entails both multilateralism and cosmopolitanism.

Cooperative internationalism is distinct from, although it is related to, MI. Militant internationalism, at least among Americans, is the embrace of an aggressive and hawkish approach to foreign affairs based on a vision of the world as full of threats. It consequently privileges the use of force. Those who score high on MI advocate acting quickly and decisively vis-à-vis potential dangers in the international system lest they fester and proliferate and believe that American military preponderance is the key to international peace and stability. Hegemony brings stability. These are part of a general frame of mind akin to what is known as the “deterrence model” in the international relations literature (Jervis 1976). Even those who do not embrace the CI/MI framework make use of this distinction. Hurwitz and Peffley use a “dimension of militarism,” “anchored, on the one end, by a desire that the government assume an assertive, militant foreign-policy posture through military strength and on the other by a desire for a more flexible and accommodating stance through negotiations” (1987:1107). From this point, I will refer to cosmopolitanism and hawkishness, as these terms better specify the underlying intuition behind the concepts. Cosmopolitanism entails multilateralism.

Cosmopolitanism and hawkishness are both distinct conceptually and empirically from isolationism, which is a generic belief that the United States is best served by avoiding political entanglements with other countries, whether they be cooperative or militant in nature. Research sometimes indicates a third isolationist dimension in the structure of foreign policy attitudes defined by a generic endorsement of withdrawing from international affairs(Chittick et al. 1995; Rathbun 2007). An isolationist opposes both extensive military engagements abroad as well as collaborative multilateral efforts to reach common problems.

Insights from the foreign policy belief system literature have never been properly or systematically applied to free trade attitudes. Most studies use these concepts to predict attitudes on international security issues, such as support for military interventions (Rathbun 2007). Those that have drawn on it have misapplied key constructs. Herrmann et al. (2001) infer a positive effect for cooperative internationalism on free trade through the negative effect of militant internationalism. In their study, they actually only measure militant internationalism and treat cooperative internationalism as its opposite, even though previous literature cautions against this (Murray 1996; Rathbun 2007). The two dimensions are negatively related but not reducible to a single continuum. Mansfield and Mutz (2009) find a negative effect of isolationism on free trade attitudes, indicating that out-group anxiety makes individuals hostile to free international economic exchange(Shahrzad Sabet, unpublished manuscript). However, as the literature reviewed above indicates, isolationism is not the opposite of cosmopolitanism. Out-group anxiety is not the opposite of in-group solidarity.2 We can make no indirect inferences from either study, which show us that isolationism and hawkishness are negatively related to free trade, but not that cosmopolitanism is positively related to free trade. We need measures of both constructs to fully specify a model of individual trade preferences. Cosmopolitanism is also not accounted for by simply including measures of nationalism. Previous studies find that nationalism is negatively associated with free trade (O'Rourke and Sinnott 2002; Mayda and Rodrik 2005), but others have found both positive and negative associations between national and broader international identities (Haesly 2001; Klandermans, Sabucedo, and Rodriguez 2003; Citrin and Sides 2004).

How might we know if attitudes toward free trade emerge from a more abstract and general libertarian outlook given laissez-faire's overlap with liberalism? First, free trade should be predicted by support for free market mechanisms as this is the economic expression of liberty as license. This can be captured by whether belief in free trade is positively predicted by support for economic liberty at home, whether it be opposition to redistribution and entitlement programs or support for deregulation and tax cuts. Both are about greater individual choices about consumption and the removal of government regulations over the economy, just at different levels of analysis. Despite the historical association between free market support and free trade, none of the studies on individual trade attitudes — neither those that stress individual position in the world economy nor even those that concentrate on ideological preferences — include variables measuring perhaps their most obvious ideological predictor — an individual's general economic philosophy. This is a curious omission considering the very subject of these studies — that is, free trade. Intuitively, we can hypothesize that just as individuals have overarching conceptions of foreign affairs in which attitudes toward free trade are couched, they also have attitudes toward competition and the free market. This is Hurwitz and Peffley's (1987) logic. Individuals have broader abstract postures on fundamental issues which lead then to specific policy preferences consistent with those foundations.

On the other hand, some might dislike free trade because they generally oppose the market allocation of resources both at the state and the international level. In this vein, Herrmann et al. (2001) hypothesize that individuals might adopt a more Rawlsian, as opposed to neoclassical, approach to free trade, in which they are concerned about the least advantaged.3 Constructivist scholars of international relations have found that countries often export their domestic policy agenda abroad. For instance, social democrats favor greater foreign aid, based on a general embrace of humanitarianism and welfare that links both (Lumsdaine 1993). The same might be true at the individual level. Consistent with this hypothesis, Rathbun (2007) finds that among elites, a number of measures of free market attitudes load negatively on the CI dimension.

To place more causal distance between the dependent and independent variables and to probe for a broader libertarian foundation for free trade attitudes, we also want to know the effect of other elements of libertarianism. It might also be the case that a belief in free exchange is part of an even broader ideological belief in liberty generically. Free trade is about choice, which is another way of saying it is about freedom. This would be reflected in support among free traders for individual choice on social issues in addition to economic issues, such as on matters of free speech or gay rights. Social conservatism is generally defined by a series of issue attitudes bound together by a belief in restraints on individual rights so as to protect the general social welfare, whether it be capital punishment or traditional values. Social conservatism (but not economic conservatism) is highly correlated with right-wing authoritarianism, and authoritarianism is the very opposite of libertarianism (Altemeyer 1988, 1998; Duriez and Van Hiel 2002; Feldman 2003; Jost, Glaser, Kruglanski, and Sulloway 2003).


This discussion above suggests the following hypotheses about the ideological sources of support for free trade. If free trade has classical liberal foundations, we would expect a package of attitudes. Support for free trade will be predicted by a combination of cosmopolitanism and multilateralism (or cooperative internationalism) and dovishness(a lack of militant internationalism). These are the foreign policy manifestations of classical liberalism. As liberals are middle of the road on economic issues, given their commitment to equal concern and respect, we have no expectations in regard to the effect of economic attitudes toward free trade, although the effect of social libertarianism should be positive as this set of liberties is part of a commitment to equal concern and respect.

If free trade has libertarian foundations, we would expect that support for free trade is predicted by a combination of social libertarianism, economic libertarianism, and a lack of cosmopolitanism. Rather than emerging out of a sense of concern for equal concern and respect for those abroad, support for free trade will find its firmest basis on the part of those who think of politics as the pure individualist pursuit of his or her own interests. Under classical liberal thinking, social libertarianism would also be thought to predict support for free trade, but not in combination with economic libertarianism. And the effect of cosmopolitanism is crucial for differentiating between the two sets of expectations.

YouGov/Polimetrix Survey

The surveys utilized in previous studies of trade attitudes at the mass level do not contain the variables necessary to test our hypotheses. I use a new original data set of 1,200 Americans collected by YouGov/Polimetrix in January 2011 that includes questions measuring attitudes toward trade, foreign policy orientation and domestic policy positions, a combination missing in previous surveys at the mass level. YouGov uses sample-matching techniques to draw “representative” samples from non-randomly selected pools of respondents in online access panels, which consist of Internet users who were recruited via banner ads, purchased email lists, and other devices. The sample-matching technique begins by drawing a stratified national sample from a target population (in this case, the 2006 American Community Survey). Rather than contacting these individuals directly, which would be prohibitively expensive, YouGov utilizes matching techniques to construct a comparable sample from its existing Internet panel. Members of the matched sample are then contacted and invited to participate in the survey. After selection, the sample is weighted to match the target population on a series of demographic factors. Although the matched sample has been drawn from a non-randomly selected pool of opt-in respondents, it can, in some respects, be treated as if it was a random sample (Vavreck and Rivers 2008). These matched samples resemble the broader public on a number of socio-demographic variables; however, as respondents self-select into the original panel, they may differ from the broader public on unmeasured variables like political interest and awareness. In general, we should expect that the sample is somewhat more educated than a truly random sample. However, as discussed below, education is factored into the analysis as a variable.

For the dependent variable, respondents were given a forced-choice question on free trade measured on a 6-point scale asking them whether “free trade is a threat to people in both rich and poor countries” or whether “free trade is an opportunity for people in both rich and poor countries.” This differs from previous conceptualizations of free trade support, which generally simply ask respondents whether they support or oppose the reduction of trade barriers. Ideally, the survey would have contained multiple measures. However, trade was just a small portion of the survey. The instrument's strength, that is, its breadth in terms of capturing general foreign policy postures and domestic ideological positions, is also its weakness.

However, I would argue that the dependent variable as constructed is superior for the purposes of this particular paper for two reasons. First, the aim is to gauge how ideological factors impact beliefs about free trade independently of pocketbook factors. We want to know whether respondents believe that free trade is in general a good or bad thing, not whether one believes it is good or bad for him personally. The more general phrasing of the question is more likely to be contaminated by these individual material impacts (Fordham and Kleinberg 2012).

Nevertheless, we must guard against the possibility that responses to this question might in any case be post hoc rationalizations of more egoistic concerns. For this reason, I include dummy variables for education level in the analysis, as these are thought to capture the effect of one's individual position in the global economy and individual material concerns on free trade attitudes. While education likely captures much more than this, as Hainmueller and Hiscox (2006) pointed out, it is at least a crude measure and the best that is available within the survey.

Table 1 presents the frequency of responses among the mass public. Those who believe free trade is an opportunity outnumber those who view it as a threat by approximately 2 to 1. This is a different ratio than typically found by those who measure trade attitudes in terms of pro and con, something we can likely attribute to the different way of capturing opinions and the particular features of the sample, which as mentioned, likely include a higher number of politically aware and educated individuals.

View this table:
Table 1.

Frequency of Responses among the Mass Public

Percentage of RespondentsCumulative
Free trade is a Threat
Free trade is an opportunity
N = 1,169

I generate scores on cosmopolitanism, hawkishness, and isolationism using factor analysis of 12 foreign policy attitude questions taken from the survey, many of which have been used in previous studies to generate indices for these latent variables (Holsti and Rosenau 1988; Wittkopf 1990; Chittick et al. 1995; Murray 1996). Regression scoring is used to create continuous variables for each of the three concepts, which are subsequently used as independent variables. I do the same with economic and social conservatism, extracting the single dominant factor for each variable individually. A complete list of variables and their scaling and wordings can be found in Appendix 1.

Appendix 2 shows the factor loadings for the foreign policy items, which are consistent with previous research on the structure of foreign policy attitudes. The militant internationalism dimension shows strong loadings for striking at the heart of opponents, defense spending, the necessity of the use of force, preventing aggressive expansion by other countries, and the importance of demonstrating resolve. The cooperative internationalism dimension is defined by concern for others by providing international aid, promoting human rights and protecting the global environment as well as multilateralist inclinations in the form of increasing the power of the United Nations. Minding one's own business, focusing on problems at home, and closing overseas military bases are the main items loading on the third, isolationist dimension.


Table 2 shows the substantive effect of our ideological independent variables on belief in free trade as an opportunity. Model 1 includes only the foreign policy variables. As would be expected, isolationism is negatively associated with belief in free trade as an opportunity. The cosmopolitan measure is also negatively associated with the belief that free trade presents an economic opportunity for all, not positively, with a very high level of statistical significance. Hawkishness, on the other hand, is positively associated with support for free trade. These three factors together account for 5% of the variance on free trade attitudes.

View this table:
Table 2.

American Mass Public Attitudes on Free Trade

Model 1Model 2Model 3
Cosmopolitanism (CI)−0.24 (0.05)***−0.11 (0.06)#−0.09 (0.06)
Hawkishness (MI)0.14 (0.05)**0.13 (0.05)*0.09 (0.06)
Isolationism−0.16 (0.04)***−0.16 (0.04)***−0.16 (0.05)***
Economic Conservatism0.33 (0.06)***0.36 (0.06)***
Social Conservativism−0.16 (0.06)**−0.16 (0.06)**
Nationalism0.11 (0.07)
Party affiliation
 Democrat0.02 (0.11)
 Republican−0.10 (0.12)
Age0.01 (0.003)#
Sex−0.27 (0.09)**
 Some College0.21 (0.11)*
 College Graduate0.22 (0.13)#
 Graduate School0.26 (0.16)#
Household Income
 Prefer not to Say−0.15 (0.15)
 Middle0.10 (0.11)
 High0.13 (0.13)
  • (Notes. Table entries are OLS coefficients with standard errors in parentheses.

  • ***p < .001; **p < .01; *p < .05; #p < .10.

  • Independent is excluded category for partisan identification. Gender is dummy variable with male coded as ‘1’. High school diploma and no high school are the excluded categories for education. Household income of $0 to $30,000 is the excluded category for income. Middle income is $30,000 to $80,000; high income is above $80,000. Dependent variable is scaled from 1 to 6.)

Model 2 shows the effect of the introduction of the latent variables of social and economic conservatism into the analysis. Those who endorse economic liberty and social liberty largely endorse the opportunity-creating nature of free trade. Economic libertarianism is both strongly and positively associated with free trade, while social conservatism is strongly and negatively associated. Moving from the minimum to the maximum score on economic libertarianism found in the sample is associated with an increase of 1.4 in belief in free trade as an opportunity, about a quarter of the scale. The effect of social conservatism as one moves from one extreme to the other is almost identical.

The fact that both variables are highly statistically and substantively significant suggests a general libertarian impulse behind free trade in the American mass public. This might not be evident, however, at the surface level, given the high correlation among economic libertarianism and social conservatism (Pearson's r = .56). Despite their highly positive relationship, the two, nevertheless, lead to very different trade attitudes. Trade is a wedge issue. For instance, if we drop out these two variables and insert a simple liberal–conservative scale into the analysis done in Model 1, left-right self-placement is statistically insignificant.4

Once these variables tapping into libertarianism are included in Model 2, the belief of cosmopolitans that free trade is a threat weakens substantially in the masses (p < .10). This suggests that the opposition of cosmopolitans to free trade in the mass public emerges from a broader hostility to market mechanisms. Isolationism, however, is almost completely unchanged between the two models. Taken together, the three foreign policy variables and the two domestic ideological variables account for 8% of the variance on attitudes toward free trade.

The introduction of demographic variables in model 3 explains little. Education has a consistently positively effect, although it is weak. However, we should keep in mind what these results mean. It does not mean that those who score high on libertarianism will necessarily embrace the reduction of American trade barriers. It might be that these individuals, while generally believing that free trade presents an opportunity, nevertheless oppose free trade because they personally are negatively impacted. The data unfortunately do not allow us leverage on that question. Males are less supportive of free trade than females in the mass public. There is no real effect of age, household income, or party affiliation.

One might argue that the effect of social conservatism is actually capturing a latent nationalism, as the two have been found to be very strongly related in previous research. Social conservatives believe in-group solidarity as a necessary protection from the social problems created by excessive liberty (Altemeyer 1988, 1998; Duriez and Van Hiel 2002; Feldman 2003; Jost et al. 2003). However, in Model 3, nationalism is controlled for through its inclusion as a latent variable indicated by responses to four questions.5 Nationalism does not have an effect.

In the full model, however, neither hawkishness nor cosmopolitanism remain statistically significant, indicating that among the mass public, isolationism is the most important foreign policy orientation for explaining trade attitudes. Out-group derogration is more important for explaining trade attitudes than the in-group solidarity of cosmopolitanism. Moving from the least to the most isolationist scores is associated with a 0.86 point decrease in support for free trade on a 6-point scale. At least for this particular measure of support for free trade, foreign policy liberalism is not positively related. The effects of social and economic conservatism are robust, however, suggesting that general attitudes toward liberty, freedom, and individualism are very important in accounting for mass attitudes toward trade in the United States.

Foreign Policy Leadership Project

With the exception of Herrmann et al. (2001), none of the trade studies to date have used elite surveys, and those who have focused on the structure of foreign policy beliefs at this level have been almost completely concerned with issues of security and high politics rather than international economics. However, the Foreign Policy Leadership Project, a survey undertaken by Ole Holsti and James Rosenau and frequently used in studies of the dimensionality of foreign policy attitudes, contains a number of questions on trade and trade agreements. The FPLP respondents are American elites from both in and outside of government. The survey also contains questions used to measure cosmopolitanism, hawkishness, isolationism, and economic and social attitudes on domestic issues. I use the most recent version of the survey from 1996 (Holsti and Rosenau 1999). Revisiting this classic survey allows us a window a window into the attitudes among elites in the mid-1990s, just as the United States was considering creating both NAFTA and the WTO argument.

Attitudes toward free trade were captured using a number of items. Respondents were asked five questions relevant to the study: whether they support erecting trade barriers against foreign goods to protect American industries and jobs, whether respondents agree or disagree with joining the World Trade Organization and the North American Free Trade Agreement, and how important it is to protect the jobs of American workers and the interests of American business abroad. Table 3 shows the distribution of responses for all different measures of the dependent variable. Elites are very supportive of free trade, expressed through opposition to trade barriers and strong enthusiasm for joining the WTO and NAFTA. However, they are also largely, although not as enthusiastically, in favor of protecting American jobs and promoting American business abroad.

View this table:
Table 3.

Frequency of Responses among Elites

Against Trade BarriersPro-WTOPro-NAFTAProtect American JobsPromote American Business Abroad
Strong Opposition6.
Moderate Opposition16.58.48.923.028.2
Moderate Support28.535.829.943.747.1
Strong Support49495228.819.5
N = 1,780

The latent variables were generated in the same manner as in the previous analysis. Care was taken to find as much overlap as possible with the mass survey for items defining cosmopolitanism, hawkishness, and isolationism. Some are identical, and most have been used in previous studies (Holsti and Rosenau 1988; Wittkopf 1988; Chittick et al. 1995; Murray 1996). Appendix 3 lists the variables and the question wordings. Appendix 4 shows the factor loadings for the foreign policy items, in which we see the same three-dimensional pattern found for the masses, consistent with previous research.


Table 4 shows the results of an ordered logit analysis with these five different dependent variables. The dependent variable used in the analysis in the first column provides perhaps the most generic expression of support or opposition to free trade, whether respondents oppose “erecting trade barriers against foreign goods to protect American industries and jobs.” Those who score high on cosmopolitanism are supportive of the erection of trade barriers, not opposed, as seen in the results in the first column. Using factor scores generated from the analysis and an add-on to STATA called Clarify (Tomz, Wittenberg, and King 2003), I generated predicted probabilities for different categories of the dependent variable. As respondents move from the least to the most cosmopolitan end of the spectrum, their likelihood of disagreeing strongly with trade barriers declines 22%. The predicted probability of disagreeing strongly with the imposition of trade barriers for those with the lowest scores on cosmopolitanism is 62%. For those with the highest scores, it declines substantially to 39%.

View this table:
Table 4.

American Elite Attitudes on Trade

Against Trade BarriersPro-WTOPro-NAFTAProtect American JobsPromote American Business Abroad
Cosmopolitanism (CI)−0.18 (0.07)**0.23 (0.07)***0.01 (0.07)0.58 (0.07)***0.15 (0.06)*
Hawkishness (MI)−0.30 (0.07)***−0.01 (0.08)−0.11 (0.08)0.46 (0.07)***0.64 (0.07)***
Isolationism−0.70 (0.07)***−0.60 (0.06)***−0.70 (0.07)***0.52 (0.06)***0.07 (0.06)
Economic Conservatism0.12 (0.09)−0.02 (0.09)0.15 (0.09)#−0.06 (0.09)0.22 (0.09) *
Social Conservativism−0.41 (0.08)***−0.37 (0.08)***−0.42 (0.08)***0.39 (0.08)***0.19 (0.08)*
Party Affiliation
 Democrat−0.10 (0.16)0.09 (0.17)−0.35 (0.16)*0.29 (0.15)#−0.01 (0.15)
 Republican0.23 (0.16)0.08 (0.16)0 (0.16)−0.12 (0.15)0 (0.15)
Age−0.04 (0.06)−0.07 (0.06)0.03 (0.06)0.02 (0.06)0 (0.06)
Sex−0.59 (0.17)***−0.07 (0.18)−0.08 (0.17)0.37 (0.17)*0.13 (0.17)
 College Graduate1.45 (0.27)***1.55 (0.29)***1.82 (0.29)***−0.1.13 (0.30)***1.05 (0.29)***
 Graduate School1.43 (0.25)***1.59 (0.27)***1.76 (0.27)***−1.17 (0.28)***0.87 (0.27)***
 MBA0.50 (0.29)#0.08 (0.27)0.17 (0.29)0 (0.27)0.21 (0.27)
 Business Executive0.42* (0.17)0.06 (0.16)0.12 (0.16)−0.29 (0.15)#−0.22 (0.15)
 Labor Official0.08 (0.30)0.20 (0.31)0.17 (0.30)−0.50 (0.30)−0.74 (0.30)*
Cut 1−2.01−1.68−0.96−4.30−2.62
Cut 2−0.28−0.72−0.03−2.02−0.03
Cut 31.361.451.600.122.38
Log Likelihood−1231.6086−1127.8076−1201.8449−1309.0473−1276.6732
Chi square271.00***246.05***239.89***259.27***231.64***
Pseudo R2.
  • (Notes. Table entries are logit coefficients from with standard errors in parentheses.

  • ***p < .001; **p < .01; *p < .05; #p < .10.

  • Independent is excluded category for partisan identification. Gender is dummy variable with males coded as ‘1’. High school diploma and some college are the excluded categories for education. MBA is a separate dummy variable indicating attainment of a business degree.)

Among elites, isolationists and cosmopolitans form an unlikely coalition against free international markets, even as the two constructs are negatively correlated with one another. The covariance between the two factor score variables is −0.16. Isolationism has a much more powerful effect on trade attitudes, however. Those who score highest on isolationism have less than a 10% chance of disagreeing strongly with the imposition of trade barriers and a 25% chance of agreeing strongly. Those who score lowest on isolationism have less than a 1% chance of agreeing strongly with the imposition of trade barriers and an 87% chance of disagreeing strongly.

Situating free trade in a global multilateral organization does make cosmopolitans significantly warmer toward free trade. As scores on cosmopolitan rise, support for joining the WTO increases, but not however, support for joining NAFTA. Cosmopolitanism in foreign affairs does not lead to a generic endorsement of free trade even among elites, although it does generate support for global institutions fostering economic ties. The probability of strongly agreeing with joining the WTO for the most cosmopolitan of respondents is 60%. For the least it is only 25%. Hawkishness is negatively associated with support for free trade in terms of reducing trade barriers, although neutral in regards to the WTO and NAFTA. In general, it does not appear that the combination of high CI and low MI, the way in which we have operationalized foreign policy liberalism, leads to support for free trade among American elites.

Hawks are strongly supportive of both protecting American jobs and promoting American business abroad, indicating a nationalist approach to economic competitiveness. The same is true of cosmopolitans, although their support for promoting business is considerably weaker. Isolationists strongly favor protecting American jobs but are neutral on protecting American business interests abroad likely because this smacks of international involvement. Isolationism is defensive rather than offensive in character.

Unlike at the mass level, economic conservatism is not a major factor in predicting free trade attitudes. On only one issue, supporting American business abroad, is the variable significant and the question does not indicate support for free trade. Social conservatism, however, is a strong predictor of trade attitudes. Social conservatives favor trade barriers and oppose joining the WTO and NAFTA. They want to protect American jobs and promote American business. There does not appear to be a consistent libertarian impulse behind free trade support among American elites.

Education is highly significant for free trade support. Having a college degree and having a postgraduate degree are both associated with hostility to trade barriers and protecting American jobs and support for joining the WTO and NAFTA. This is in line with previous results highlighted in political economy models showing the importance of personal skill levels in determining trade attitudes, even as we must recognize that education might capture other variables as well. Interestingly, even amongst a population of elites, there are still substantial differences predicted by educational level. The attainment of an MBA does not seem to affect trade attitudes in any substantial way, however. This seems to indicate that it is not a more precise understanding of market mechanisms that leads educated individuals to support free trade.


The analyses indicates that trade is an ideological wedge issue at both the mass and elite levels, although in different ways. At the mass level, support for free trade is driven by support for social and economic libertarianism, two factors that are negatively correlated. Isolationism is the only foreign policy orientation that consistently predicts opposition to free trade. The combination of cooperative internationalism (cosmopolitanism) and militant internationalism (hawkishness) that marks liberal thinking in regard to foreign affairs, does not have any real effect.

However, at the elite level, cosmopolitanism and isolationism, two foreign policy orientations that are generally negatively correlated, both generate hostility to free trade. For cosmopolitans, the only exception is the global trading system of the WTO. Yet elites do not seem to be driven by their economic libertarianism toward support for reducing trade barriers. Only social conservatism is associated with free trade attitudes, negatively as is the case with the masses.

In neither case do we find that trade attitudes are a function of commitment to liberal thinking. At the mass level, trade attitudes appear to have libertarian origins. At the elite level, the dovishness associated with foreign policy liberalism shows some positive effect on free trade, but cosmopolitanism does not.

It is not clear why elites and masses show different patterns in thinking toward trade, however. One possibility might be that the masses have less fully formed attitudes toward foreign policy, and therefore views on liberty and freedom provide a more accessible heuristic for formulating opinions on trade. Elites, on the other hand, can more readily connect their foreign policy orientations to the issue of trade because they know more about international relations.

The paper makes a contribution to a number of different literatures. First, it adds to the growing number of studies interested in the role that ideational factors play in international political economy (Blyth 2002; Abdelal 2007; Abdelal, Blyth, and Parsons 2010)but which have been overwhelmingly qualitative in nature with some notable exceptions (Chwieroth 2007a,b, 2010) and which generally do not address individual attitudes and political behavior. There is every reason to think that ideational approaches have much to add to quantitative analyses and survey research.

Second, it brings political economy into studies on the structure of foreign policy attitudes which have been, for the most part, preoccupied with political and military issues and ideology into the study of trade attitudes. Unlike other analyses, the goal was to explain individuals ideological predispositions toward free trade, not the effect in the aggregate population of how the framing of the issue shifts respondents’ attitudes as a whole in one direction or another (Herrmann et al. 2001; Hiscox 2006). Issues of international security and international trade are inextricably connected in individual's minds. No wedge should be driven between them.

Appendix 1

Variables in Mass Public Survey

Foreign Policy Questions
 Agricultural SurplusCountries needing our agricultural surpluses should pay for them instead of getting something for nothing – US agricultural surpluses should be given for free to the have-nots of the world (forced choice from 1 to 6)
 Defense SpendingFederal spending on foreign aid (Decreased a lot to increased a lot, 5-pt scale)
 Foreign AidFederal spending on foreign aid (Decreased a lot to increased a lot, 5-pt scale)
 Free TradeFree trade is a threat to poor people in both rich and poor countries – Free trade is an opportunity for people in both rich and poor countries (forced choice from 1 to 6)
 Global EnvironmentProtect US economic growth even if that hurts the global environment – Protect the global environment even if that hurts US economic growth (forced choice from 1 to 6)
 Home FocusWe should not think so much in international terms but concentrate more on our own national problems and building up our strength and prosperity at home (Strongly disagree to strongly agree, 7-pt scale)
 Military Base ClosureKeep overseas US military bases to protect the US and its allies – Close US military bases overseas to avoid foreign entanglements (forced choice from 1 to 6)
 Mind Own BusinessThe US should mind its own business internationally and let other countries get along the best they can on their own (Strongly disagree to strongly agree, 7-pt scale)
 Prevent ExpansionThe US should take all steps including the use of force to prevent aggression by an expansionist power (Strongly disagree to strongly agree, 7-pt scale)
 ResolveThe US must demonstrate its resolve so that others do not take advantage of it (Strongly disagree to strongly agree, 7-pt scale)
 Striking at the HeartRather than simply countering our opponents’ thrusts, it is necessary to strike at the heart of an opponent's power (Strongly disagree to strongly agree, 7-pt scale)
 United NationsReduce the power of the United Nations so that the US might have more influence in international affairs – Expand the power for the United Nations, even if the US might have less influence in international affairs (forced choice from 1 to 6)
 War is Necessary EvilGoing to war is unfortunate but sometimes the only solution to international problems (Strongly disagree to strongly agree, 7-pt scale)
Social Conservatism
 Gay MarriageRecognize marriage only as a union between a man and a woman – Recognize marriage between gay and lesbian couples (forced choice from 1 to 6)
 LegalizationKeep marijuana illegal, as it is today – Legalize marijuana (forced choice from 1 to 6)
 New LifestylesNew lifestyles are contributing to the breakdown of our society (Strongly disagree to strongly agree, 7-pt scale)
 Deport IllegalsAllow residents who are not in the US legally to remain in the country – Deport residents who are not in the US legally (forced choice, 1 to 6)
Economic Conservatism
 Business/Individual over Gov'tBusinesses and individuals can solve problems better than government can (Strongly disagree to strongly agree, 7-pt scale)
 Government RegulationGovernment regulation unfairly hurts businesses – Government regulations protect society (Strongly disagree to strongly agree, 7-pt scale)
 Income EqualityThe government should get out of the business of trying to promote income equality – The government should do more to reduce income inequality (forced choice from 1 to 6)
 WelfareFederal spending on people on welfare (Decreased a lot to increased a lot, 5-pt scale)

Appendix 2

Three Dimensions of Foreign Policy Attitudes among the Masses

Prevent Expansion0.840.130.06
Striking at the Heart0.64−0.090.18
Defense Spending0.51−0.25−0.15
War is Necessary Evil0.74−0.03−0.05
Agricultural Surplus0.100.73−0.09
United Nations Power−0.060.790.22
Global Environment−0.100.670.01
Foreign Aid0.070.58−0.23
Home Focus0.3100.73
Mind Own Business−
Close Military Bases−0.62−0.070.34
Proportion of Variance0.290.240.15
  • (Note. Table entries are factor loadings derived from a principal components analysis retaining all factors with an eigenvalue >1 and using oblique, promax rotation. N = 1070.)

  • (Note. MI, militant internationalism; CI, cooperative internationalism.)

  • Appendix 3

    Variables in Elite Survey

    Foreign Policy Questions
     BurdensharingOur allies are perfectly capable of defending themselves and they can afford it, thus allowing the United States to focus on internal rather than external threats to its well-being (disagree strongly to agree strongly, 4-pt scale)
     DominoThere is considerable validity in the ‘domino theory’ that when one nation falls to aggressor nations, others nearby will soon follow a similar path (disagree strongly to agree strongly, 4-pt scale)
     Defense SpendingWe are spending too little money, about the right amount, or too much (3-pt scale)
     Global EnvironmentProtecting the global environment (not at all important to very important, 4-pt scale)
     Home FocusWe should not think so much in international terms but concentrate more on our own problems (agree strongly to disagree strongly, 4-pt scale)
     Human RightsPromoting and defending human rights in other countries (not at all important to very important, 4-pt scale)
     International AidHelping to improve the standard of living in less developed countries (not at all important to very important, 4-pt scale)
     Maintain SuperiorityMaintaining superior military power worldwide (not at all important to very important, 4-pt scale)
     Mind Own BusinessAmerican interventions in conflicts that are none of our business (not at all serious to very serious issue, 4-pt scale)
     NAFTASigning the North American trade agreement (not at all important to very important, 4-pt scale)
     PreemptRather than simply countering our opponent's thrusts, it is necessary to strike at the heart of an opponent's power (disagree strongly to agree strongly, 4-pt scale)
     Prevent ExpansionThe United States should take all steps including the use of force to prevent aggression by any expansionist power (disagree to agree strongly, 4-pt scale)
     Protect BusinessProtecting the interests of American business abroad (not at all important to very important, 4-pt scale)
     Protect JobsProtecting the jobs of American workers (not at all important to very important, 4-pt scale)
     Scale BackAmerica's conception of its leadership role in the world must be scaled down (disagree strongly to agree strongly, 4-pt scale)
     Against Trade BarriersErecting trade barriers against foreign goods to protect American industries and jobs (agree strongly to disagree strongly, 4-pt scale)
     United NationsStrengthening the United Nations (not at all important to very important, 4-pt scale)
     WTOSigning the GATT trade agreement and joining the World Trade Organization (disagree strongly to agree strongly, 4-pt scale)
    Economic Conservatism
     Balanced BudgetAdding a balanced budget amendment to the Constitution (agree strongly to disagree strongly, 4-pt scale)
     EntitlementsReducing the growth of spending for Medicare and Medicaid (agree strongly to disagree strongly, 4-pt scale)
     RedistributionRedistributing income from the wealthy to the poor through taxation and subsidies (agree strongly to disagree strongly, 4-pt scale)
     TaxReducing federal budget deficits by raising taxes (agree strongly to disagree strongly, 4-pt scale)
     WelfareWe are spending too much money, too little money, or about the right amount (3pt scale)
    Social Conservatism
     DecriminalizationLegalizing drugs such as cocaine to reduce drug-related crimes (agree strongly to disagree strongly, 4-pt scale)
     PrivacyRequiring that applicants for marriage licenses, insurance policies, and some jobs be tested for AIDS (agree strongly to disagree strongly, 4-pt scale)
     AbortionLeaving abortion decisions to women and their doctors (agree strongly to disagree strongly, 4-pt scale)
     Gay RightsBarring homosexuals from teaching in public schools (agree strongly to disagree strongly, 4-pt scale)
     Death PenaltyBanning the death penalty (agree strongly to disagree strongly, 4-pt scale)

    Appendix 4

    Three Dimensions of Foreign Policy Attitudes among Elites

    Burden sharing−
    Scale Back−0.02−0.260.67
    Home Focus0.110.160.72
    Defense Spending0.290.59−0.11
    Striking at the Heart0.040.690.19
    Maintain Superiority0.090.70−0.18
    Mind own Business−
    Prevent Expansion−0.280.54−0.21
    Human Rights0.80−0.080.04
    International Aid0.76−0.010.06
    Global Environment0.740.10−0.05
    Strong UN0.710.11−0.02
    Proportion of Variance0.210.190.17
  • (Note. Table entries are factor loadings derived from a principal components analysis retaining all factors with an eigenvalue >1 and using oblique, promax rotation. N = 1645.)

  • (Note. MI, militant internationalism; CI, cooperative internationalism.)

  • Footnotes

    • 1The author thanks Joshua Kertzer, Benjamin Fordham, and Serena Joy for their help improving this paper. Kevin Petway's technical expertise was invaluable and not evident at all in the final version of this paper.

    • 2It is not clear that the authors are attempting to make such a statement; however, it is implied at times. When they write that “Activist foreign policy attitudes, a positive attitude toward out-groups, and a preference for open trade…all reflect a sense of cosmopolitanism and inclusion,” they immediately follow with a contrast to isolationism. “Isolationism, a negative attitude toward out-groups, and antipathy toward open trade all reflect a sense of insularity and separatism” (Mansfield and Mutz 2009:451). This implies that the two are opposites on a continuum.

    • 3Unfortunately, however, they do not measure economic philosophy as an individual-level independent variable in their analyses but rather prime Rawlsianism in a treatment condition in an experiment. Their goal is to establish how the framing of trade issue affects aggregate support for free trade rather than explain individuals’ prior ideological predispositions toward free market solutions.

    • 4Results not shown.

    • 5These are: “When I see the American flag flying, I feel great”; “I am proud to be an American”; “I believe in the motto: ‘My country, right or wrong’”; “The United States is generally on the side of good against the powers of evil in the world.” Cronbach's alpha is over 0.8.


    View Abstract