Populism and foreign policy: charting the terrain

During the last few years, interest in populist foreign policy has increased (for reviews see Destradi et al., Reference Destradi, Cadier and Plagemann2021; Löfflmann, Reference Löfflmann2022; Chryssogelos, Reference Chryssogelos2021; Verbeek and Zaslove, Reference Verbeek, Zaslove, Kaltwasser, Taggart, Espejo and Ostiguy2017). Most of the attention, however, has been devoted to the policies and politics of populist elite actors (Cadier, Reference Cadier2024; Kesgin, Reference Kesgin2020; Mead, Reference Mead2002, Reference Mead2011; Wehner, Reference Wehner2023). This research has found that while it is possible to identify common general foreign policy themes among certain populist actors (such as anti-globalism or anti-Americanism), the effects of populism on foreign policy as a whole are quite diverse (Taş, Reference Taş2022: 564). While populist parties and governments all “cast their positions in terms of the pure people versus the corrupt elite” (Verbeek and Zaslove, Reference Verbeek, Zaslove, Kaltwasser, Taggart, Espejo and Ostiguy2017: 386), they support a wide array of foreign policies, depending on the thick-ideology populism attaches itself to, on the interests and the personality of leaders (Thiers and Wehner, Reference Thiers and Wehner2022), and the roles these actors aim at playing in the international arena (Destradi and Plagesmann, 2019). For example, in comparing Brazil’s Bolsonaro and Mexico’s Lopez Obrador, Wajner and Wehner (Reference Wajner and Wehner2023) have shown that although populist leaders typically veer isolationist in order to substantiate their anti-elitist and anti-globalization agenda, at times they embrace internationalism by pointing to transnational solidarities between peoples. Incorporating various dimensions of internationalism into existing populist rationale depends on whether a particular position brings the leader an advantage in terms of legitimacy, political support or differentiation from previous leadership (Wajner and Wehner, Reference Wajner and Wehner2023: 10). Consequently, engagement with the international arena is often driven by domestic political calculus, not based on populist ideology per se.

Other scholars also stress the diversity of populist foreign policy but zero in on the fact that populism is usually combined with other traditional (thick) ideologies. Accordingly, the latter serve as host ideologies by specifying the opposition between “the people” and “the elite” that is at the heart of populist ideology (Giurlando, Reference Giurlando2021): right-wing populist parties and movements use a nativist definition of the people, who are seen as struggling against cosmopolitan elites. Left-wing populists see disadvantaged socioeconomic groups suppressed by the rich, capitalist elite. Other blends of populism define the “pure” people as historically belonging to a smaller territorial unit who oppose territorial integration into larger entities (e.g., Verbeek and Zaslove, Reference Verbeek, Zaslove, Kaltwasser, Taggart, Espejo and Ostiguy2017: 392). Against this backdrop, some scholars have concluded that there is no single populist foreign policy; rather, populist foreign policy varies with the host ideology (e.g., Verbeek and Zaslove, Reference Verbeek, Zaslove, Kaltwasser, Taggart, Espejo and Ostiguy2017; Destradi et al., Reference Destradi, Cadier and Plagemann2021).Footnote ¹

So, while many studies have focussed on the elite level and show diversity rather than uniformity of populist foreign policy, we know almost nothing about populists at the population level (but see Rathbun, Reference Rathbun2013). Do populist citizens also show diversity in opinion or do they have common views on basic foreign policy principles that are distinct from non-populists? In other words, are there general foreign policy preferences among populist citizens, irrespective of context or (thick) ideology? The following theoretical analysis first clarifies our conceptual approach to populism at the population level. Then we discuss reasons why there might be (no) connections between populism and foreign policy preferences.

We rely on the so-called ideational approach to the study of populism (Mudde, Reference Mudde, Kaltwasser, Taggart, Espejo and Ostiguy2017; Hawkins et al., Reference Hawkins, Carlin, Littvay and Kaltwasser2018). Accordingly, populism is defined as a thin ideology that “considers society to be ultimately separated into two homogeneous and antagonistic groups, ‘the pure people’ versus ‘the corrupt elite’, and which argues that politics should be an expression of the volonté générale […] of the people” (Mudde, Reference Mudde2004: 543). In other words, populism consists of a set of ideas that can be found at different levels, including the citizen level. Populism at the individual level is generally referred to as populist attitudes (e.g., Akkerman et al., Reference Akkerman, Mudde and Zaslove2014; Castanho Silva et al., Reference Castanho Silva, Andreadis, Blanuša, Corti, Delfino, Rico, Ruth, Spruyt, Steenberg, Littvay, Hawkins, Carlin, Littvay and Kaltwasser2018; Erisen et al., Reference Erisen, Guidi, Martini, Toprakkiran, Isernia and Littvay2021; Hawkins et al., Reference Hawkins, Riding and Mudde2012; Schulz et al., Reference Schulz, Müller, Schemer, Wirz, Wettstein and Wirth2018). Together, these ideas form an attitudinal syndrome linked through a non-compensatory mechanism (Wuttke et al., Reference Wuttke, Schimpf and Schoen2020). As such, populism lies at the intersection of several dimensions such as anti-elitism, support for popular sovereignty, and Manicheanism (e.g., Akkerman et al., Reference Akkerman, Mudde and Zaslove2014: 1326; Castanho Silva et al., Reference Castanho Silva, Andreadis, Blanuša, Corti, Delfino, Rico, Ruth, Spruyt, Steenberg, Littvay, Hawkins, Carlin, Littvay and Kaltwasser2018; Elchardus and Spruyt, Reference Elchardus and Spruyt2016; Spruyt et al., Reference Spruyt, Keppens and Van Droogenbroeck2016).

Building on this conceptualization, at the core of populist ideas lies the strict binary distinction between citizens and elites. This highlights two integral components that deserve attention when it comes to establishing links between the public’s populist attitudes and their views on foreign policy.

First, attitudes towards the role of elites and their (adversarial) relationship to citizens are of key importance in populist ethos; it involves opposition to the delegation of power to elites and mistrust of positions held by elites and elite-controlled organizations (e.g., Pevehouse, Reference Pevehouse2020). In the foreign policy domain, international organizations are a prime example. Populist leaders usually portray them as lacking clear accountability structures and as following their own agenda that is detached from the public interest (Bearce and Jolliff Scott, Reference Bearce and Jolliff Scott2019; Zürn, Reference Zürn2022). Since, according to populists, the people are the primary source of legitimacy, people-centrism often justifies and even calls for anti-elitism.

Defining the elite against whom populists fight is, essentially, a matter of context: while Donald Trump attacked the Washington establishment, European populist parties often discredit the EU’s bureaucracy (Destradi and Plagemann, Reference Destradi and Plagemann2019: 713–714; see also Chryssogelos, Reference Chryssogelos2021 on European parties and Euroscepticism). Likewise, populists promise a voice for the underrepresented and a come-back for the paradigm of national interest/anti-globalization. Consequently, the enemies needed for staging the fight can differ from case to case (Jakupec, Reference Jakupec2018).

Second, a Manichean worldview implies that populists perceive the world in a strictly dichotomous way that distinguishes between good and evil, which may give rise to moralistic thinking (e.g., Hawkins et al., Reference Hawkins, Carlin, Littvay and Kaltwasser2018). This rather simplistic, black-and-white view of the world seems akin to a strong need for cognitive closure (Kruglanski and Webster, Reference Kruglanski and Webster1996), which expresses individuals’ desire for clear answers – even if they are wrong – and an aversion towards ambiguity. Following this line of reasoning, high scorers on populist attitudes should prefer simple distinctions and decision rules, whereas low scorers should be more open to complexity and compromise.

When considering the possible connection between populist attitudes and foreign policy attitudes, we focus on foreign policy postures, that is, attitudes towards general principles to which the foreign policy of the individual’s country could conform (Hurwitz and Peffley, Reference Hurwitz and Peffley1987: 1104). They provide benchmarks for evaluating specific policies and therefore bring coherence, consistency (Hurwitz and Peffley, Reference Hurwitz and Peffley1987; Kertzer et al., Reference Kertzer, Powers, Rathbun and Ravi2014; Rathbun et al., Reference Rathbun, Kertzer, Reifler, Goren and Scotto2016) and – because they are relatively context-independent (e.g., Mader and Schoen, Reference Mader and Schoen2023) – stability to citizens’ attitudes. According to Gravelle et al. (Reference Gravelle, Reifler and Scotto2017), four dimensions underlie citizens’ beliefs vis-à-vis foreign policy in the Western hemisphere. Isolationism refers to the question whether international involvement should be sought at all. Second, militant internationalism revolves around projecting military power abroad. Third, cooperative internationalism covers features such as working through international organizations and cooperation. Finally, global justice refers to the redistribution of wealth across borders (Gravelle et al., Reference Gravelle, Reifler and Scotto2017). When considering potential links between populist attitudes and foreign policy attitudes, we rely on these four dimensions.

Theorizing the role of populist attitudes for foreign policy postures

Links between citizens’ populist attitudes and foreign policy postures may result from multiple mechanisms. According to one line of reasoning, people could reflect on policy themselves by evaluating them in terms of whether or not they are compatible with general principles that are important to them (e.g., Hurwitz and Peffley, Reference Hurwitz and Peffley1987; Peffley and Hurwitz, Reference Peffley and Hurwitz1985). From this perspective, populism might represent such principles against which to judge foreign policies, just like others have argued that (thick) ideologies (Hurwitz and Peffley, Reference Hurwitz and Peffley1987; Nincic and Ramos, Reference Nincic and Ramos2010), value priorities (Goren et al., Reference Goren, Schoen, Reifler, Scotto and Chittick2016; Rathbun et al., Reference Rathbun, Kertzer, Reifler, Goren and Scotto2016) or moral principles (Kertzer et al., Reference Kertzer, Powers, Rathbun and Ravi2014) serve such a function. Going back even further in the funnel of causality, research suggesting that personality influences foreign policy preferences also implicitly assumes some type of natural-affinity mechanism based on abstraction and reasoning (Schoen, Reference Schoen2007; Wajner and Wehner, Reference Wajner and Wehner2023).Footnote ²

Such cognitive processes may imply some level of conceptual sophistication, as applying general principles to specific policy problems can be cognitively challenging (Goren, Reference Goren2001). This requirement is mitigated, however, when considering another possible mechanism linking populist attitudes and foreign policy postures. Accordingly, citizens often rely on cues from trusted, more sophisticated sources to do the conceptual heavy lifting for them (e.g., Zaller, Reference Zaller1992; Kertzer and Zeitzoff, Reference Kertzer and Zeitzoff2017). By receiving such cues over extended periods of time, citizens may learn “what goes with what” (Converse, Reference Converse and Apter1964: 212) and how to link populist ideas and foreign policy postures – or at least appear to do so.Footnote ³

In sum, prior research on foreign policy attitudes suggests that we may expect links between populist attitudes and foreign policy principles where ideational affinities between populist principles and foreign policy positions exist and repeated elite cueing is likely to have established a quasi-logical connection of “what goes with what.” We hence proceed to discuss such potential links between populism and each of the four foreign policy principles in order to arrive at expectations about the possible effects of the former on the latter at the population level, irrespective of political context and host ideology. In a second step, we turn to the more complex notion that host ideologies must be accounted for when thinking about links between populism and foreign policy.

Involvement in international affairs, which for a long time were considered a domaine réservé, implies delegation of authority to elites and potentially the need for compromise in the international arena. Thus, the will of the people can hardly serve as the sole guide for policy-making; rather, elites have some leeway in making decisions, and these decisions may deviate from the “popular will.” Such deviations may be instrumentalized by populist actors in order to cultivate electorally profitable anti-elitist sentiment among the public. Consequently, populist leaders may embrace and promote isolationism as a measure of “protecting” the people from either (foreign) ideas, enemies or alleged economic hardship (Wajner and Wehner, Reference Wajner and Wehner2023). Hence, we anticipate that high scores on populist attitudes go hand-in-hand with increased support for isolationism.

Turning to cooperative internationalism, we may expect populist citizens to be sceptical of it as well. International organizations and negotiations usually mean that elites play a significant role and make decisions that may not be in line with the popular will. In a similar vein, elite cooperation at the international level usually leads to compromise, which populists dislike. Cooperative internationalism may also imply high costs, economic and beyond, and populist leaders may communicate these costs to the people, thus leading to less support. In sum, populists’ lack of support for cooperative internationalism may be anchored both in an anti-elitist stance and also a “people’s first” attitude. Accordingly, high levels of populist attitudes may be expected to be linked to opposition to cooperative internationalism.

For sharing wealth across borders, a similar reasoning yields the expectation that high scores on populist attitudes might correlate with a rejection of global justice. In particular, the provision of foreign aid implies a long chain of delegation from the people to the elites, often involving international organizations. Hence, populists may reject such policies due to their anti-elite stance (Heinrich et al., Reference Heinrich, Kobayashi and Lawson2021). Furthermore, concern for the well-being of others in far-removed corners of the world may be used in populist rhetoric to imply disregarding the interests of the people at home (Bayram and Thomson, Reference Bayram and Thomson2022; Suzuki, Reference Suzuki2023).

Finally, militarist internationalism revolves around the projection of military power abroad. To be sure, the military is a hierarchical organization in which elites play a huge role. At the same time, the use of force may be considered a simple method of conflict resolution that helps avoid complicated and complex bargaining at the elite level. Consequently, support for militarist internationalism may resonate well with populists’ elite scepticism and taste for clear-cut distinctions and decisions; support for populist ideas may thus go hand-in-hand with support for projecting military power abroad. As shown by Rathbun (Reference Rathbun2013: 30) when discussing the role Jacksonian tendencies play in explaining militant internationalism among Tea Party voters, isolationist tendencies can go hand-in-hand with militant internationalism, with the association rooted in the military and not the internationalist aspect of militant internationalism. The importance of the military for isolationists is in line with the high value placed on national sovereignty, typical of populism in Europe and elsewhere.

Thus far our theoretical analysis has ignored that populism is often combined with other traditional (thick) ideologies. Do we have to account for this fact if we want to understand the link between populism and foreign policy? Which implications does this blending of populism and host ideologies have for how citizens think about foreign policy? Theoretically, there are two possibilities. First, both populism and host ideology might provide different considerations about foreign policy, which in turn shape citizens’ postures independently and additively. This is what we have implicitly assumed above, when we derived expectations about the link between populism and postures without regard to host ideologies. In this scenario, host ideologies are only relevant as potential confounders of the association between our quantities of interest.Footnote ⁴ To understand the link between populism and postures, we would merely have to account for them in the statistical analysis and should expect the association between populism and foreign policy postures to be the same among citizens on the left and the right.

The second possibility is that populism and host ideology interact in shaping foreign policy postures, that is, that one set of ideas dampens or amplifies the effects of the other. Operationally, this would mean that the associations between populism and foreign policy postures differ between left-wing and right-wing citizens or, viewing the interaction “from the other side,” that the associations between host ideology and foreign policy postures differ between populists and non-populists. Several mechanisms might create such interactions. Building on the idea that populism makes people less likely to accept compromises, for example, we might expect that left- and right-wing populists differ more strongly in their foreign policy postures than left- and right-wing non-populists. However, our goal here is not to speculate on potential mechanisms or forward specific hypotheses on interactions. We take a more exploratory route and merely perform the diagnostic task of documenting whether or not there is empirical evidence of such interactions.

Data and national cases

To test our hypotheses, we draw on data from online surveys conducted in France, Germany, Italy, and the United Kingdom (henceforth UK) in the summer of 2019. In total, we surveyed 7,788 respondents (FRA: 1,866; DEU: 1,899; UK: 1,976; ITA: 2,047). The French, German, and British surveys were fielded by YouGov in August 2019 using nation-specific online open-access panels. Respondents were recruited based on quotas on gender, age, and education. In Italy, the survey was fielded by GfK in June–July 2019. The sample was selected from a probability panel held and managed by the same company. As the countries in our sample differ considerably in their political systems, strategic cultures, and political cultures, we analyse them separately.

The four countries were selected for two reasons. First, they are the most powerful European countries and hence of substantive importance. This importance is increased by the fact that in all four countries, populist parties have had significant electoral success and, consequently, gained political influence. It is hence important to understand the foreign policy preferences of the populists, including those at the citizen level. Second, there are important differences between the four countries – both in terms of their foreign policy cultures and populist movements. At the moment of fielding the survey, in Germany and France, for example, populists were in parliament but not in government, while in Italy and, arguably, the UK, they had executive power. In the summer of 2019, the UK was in the middle of a succession of Conservative-dominated/influenced governments, with many debates revolving around the 2016 Brexit Referendum and its implementation in 2020 (Hobolt, Reference Hobolt2016). While a natural association between populism and conservatism seems somewhat improbable (not in the least because of the former’s critical stance against elites, and the latter’s support of elites), in the context of post-Brexit debates on deciding how and when to leave the EU, populism gained traction in the UK. As Alexandre-Collier (Reference Alexandre-Collier2022) points out, the Conservative Party included either procedural or substantive populist elements, around the time of Brexit and afterwards, especially in terms of Johnson’s anti-immigration discourse, and hard stance on Brexit. It is beyond the scope of this article to provide analyses of the boundary conditions of each case. However, the contextual diversity does allow us to diagnose whether there is variation in the associations between populist attitudes and foreign policy postures in different contexts. Finding consistent patterns of associations across contexts would strengthen – but not conclusively prove – the view that there may be one distinct populist foreign policy after all.Footnote ⁵

Measures

Foreign policy attitudes were operationalized using the conceptualization suggested in Gravelle et al. (Reference Gravelle, Reifler and Scotto2017). These authors suggest a four-dimensional configuration of foreign policy beliefs in which the three traditional dimensions found in the US, namely internationalism, militant and cooperative, and isolationism, have to be integrated with a fourth one tapping public support for redistributive and humanitarian policies, such as foreign aid. According to this line of thinking, the best way of describing the belief structure of the European public is made on the basis of four main postures, called militant internationalism, cooperative internationalism, isolationism, and global justice.

To build our four measures of the foreign policy postures we relied upon items suggested by Gravelle et al. (Reference Gravelle, Reifler and Scotto2017; Reference Gravelle, Reifler and Scotto2020). Militant internationalism was operationalized using two items, one related to the role of the military and another to support for the use of force in international relations. Cooperative internationalism was measured by three items on whether the respondent’s country should work more through international organizations (e.g., the UN), whether the main goal of a world leader is to build international consensus or to be committed to diplomacy. We assess the isolationist dimension with three items stressing the importance of focusing on the need to stay aloof from international relations. The last dimension elicits attitudes towards global justice by means of three items asking whether the respondent’s country should share its wealth with other nations, spend significantly more on foreign aid, or whether, instead, the country really does enough to help the world‘s poor. Cross-country dimensionality was tested using aggregate and country-specific exploratory factor analysis and simple additive indexes for these four dimensions were computed and rescaled to range from 0 to 1.Footnote ⁶

Moving now to the independent variables, we rely on an assorted set of items of which some were adapted from survey studies of populism in other contexts (Akkerman et al., Reference Akkerman, Mudde and Zaslove2014; Oliver and Rahn, Reference Oliver and Rahn2016), while the most were of our own construction. These items mostly cover anti-expert elites’ attitudes, with two items gauging whether experts do not know much or respondents are more confident in their own opinion than other people’s facts, and a Manichean outlook when it comes to conceiving politics, via three items eliciting whether political opponents have evil intentions, whether politics is a struggle between good and evil or conceived as a conflict between “us versus them.” After testing the internal structure of the responses to these items via exploratory factor analysis, we created two subdimensions and combined them building on the idea that populism is a non-compensatory concept and the so-called “Goertz approach” (Wuttke et al., Reference Wuttke, Schimpf and Schoen2020). In this view, populism is seen as a non-compensatory multidimensional concept in which each component constitutes a necessary condition for the presence of the phenomenon. Thus, “populists” are respondents who score high in all the subdimensions. Following this approach, we use a standard operationalization strategy and create the final scores using the minimum value of the concept subdimensions. Eventually, all the scores were standardized on a 0 to 1 scale, with the mean 0 and a standard deviation of 1.Footnote ⁷

We also explored how our scale performs in comparison with other two popular batteries measuring populist attitudes – the Akkerman et al. (Reference Akkerman, Mudde and Zaslove2014) and Castanho Silva et al. (Reference Castanho Silva, Andreadis, Blanuša, Corti, Delfino, Rico, Ruth, Spruyt, Steenberg, Littvay, Hawkins, Carlin, Littvay and Kaltwasser2018) scales – using the case of Italy where all three measures were administered to the sample. Results showed them to be consistent.Footnote ⁸ To be sure, we do not claim that our measure does not come without any problems. It has, however, face validity as it captures the relevant subdimensions using an established technique and appears to perform as well as other widely used measures.

Our second main independent variable is the ideological stance on the Left-Right continuum.Footnote ⁹ Ideology is particularly relevant here because the ideological interpretation of populism suggests that populism is a “thin ideology” and the most obvious thick ideology to whom it may attach itself is represented by the left-right continuum. We measured respondents’ position by asking them to place themselves on a left-to-right continuum, with the response scale ranging from 1 (left) to 7 (right). To allow for non-linear effects, we coded respondents into six groups, one for each ideological orientation (left, centre-left, centre, centre-right, right) plus a category for those who refused to answer this question or expressly said they did not recognize themselves in this scale. Finally, we control for several sociodemographic variables, including a dummy for female respondents, age measured in years, and education measured via a categorical variable (“low-level education” being the reference category, “medium-level education,” and “high-level education”).

Statistical models

We test our hypotheses running a set of country-specific linear regression models summarized by the following general notation:

$${y_i} = \beta {X_i} + \alpha + {\varepsilon _i}$$

$${\varepsilon _i} \sim N\left( {0,{\sigma _\varepsilon }} \right)$$

where ${y_i}$ is the foreign policy posture (either militant internationalism, cooperative internationalism, isolationism or global justice posture) of respondent $i$ , ${X_i}$ is a matrix of explanatory variables with their respective coefficients $\beta $ , $\alpha $ is the country intercept, and ${\varepsilon _i}$ is the residual error term. In terms of independent variables, we include populism and left-right ideology as the main variables of interest and standard socio-demographic controls, namely age, gender and education.

In describing the results, we first assess the direct impact of populism on our four policy attitudes while controlling for host ideology. We then consider the role of host ideologies on foreign policy attitudes as dependent on populist attitudes via interaction terms. Country-specific models help us disentangle any potential heterogeneity in the considered effects at the country-level across France, Germany, Italy, and the UK. It is worth noting that our analysis cannot avoid the potential circularity of studying attitude-to-attitude relationships, so in the subsequent discussion we do not intend to imply a specific causal ordering among the considered variables as this is difficult to identify with certainty. When we use the term “effect,” this is meant in the sense of a statistical association, rather than denoting a causal effect.