3 - A Theory of Elite Influence and Popular Legitimacy

Summary

This chapter presents the books theory of elite communication effects on public opinion in global governance. It begins by defining legitimacy beliefs and introducing its favored empirical measure of such beliefs. In developing the book’s theory, the chapter starts from what is distinctive about the global setting in which elites and their messages may impact public opinion. The core of the argument is presented in two steps. The first step explains why elite communication can be expected to shape citizens’ legitimacy beliefs, given that citizens rely on cognitive heuristics when forming opinions. The second step specifies when elite communication can be expected to be particularly influential, identifying conditions associated with all three components of the communicative setting – elites, messages, and citizens.

What determines the legitimacy beliefs of citizens toward IOs? In this chapter, we present our theory of elite communication as a source of popular legitimacy in global governance. While other accounts emphasize individual characteristics, institutional qualities, and societal structures as sources of legitimacy (Tallberg et al. 2018), we argue that citizens’ legitimacy beliefs are profoundly shaped by the ways in which elites speak about IOs. Since citizens have limited independent information about global governance, they rely on elite communication as a shortcut to opinions, which in turn opens up space for elites to influence how citizens perceive the legitimacy of IOs. We outline this argument in four steps.

First, we specify our conceptualization of citizen legitimacy beliefs – our dependent variable. We explain that our conceptualization rests on a sociological approach, which conceives of an organization’s legitimacy as derived from the beliefs and perceptions of a given audience. We then discuss four possible ways to operationalize legitimacy beliefs for empirical research and explain why we settle on the measure of confidence, which privileges individuals’ deep-seated approval of an organization.

Second, we introduce the key components of elite communication in the context of global governance – the messengers and the message. We suggest that the relevant messengers comprise both globally active elites, such as member governments, nonstate actors, and IOs themselves, and domestically oriented elites, such as political parties. In contemporary politics, IOs are not the exclusive concern of globally oriented actors, as sometimes assumed in international relations, nor are political parties exclusively communicating about domestic political issues, as sometimes assumed in comparative politics. We also suggest that messages about IOs typically invoke the institutional qualities of these organizations: an IO’s social purpose, the authority it has been granted, the procedures it uses to make decisions, and the performance it achieves in terms of outcomes. Elites communicate about these qualities because they are central to IOs as political systems and because they expect them to matter for people’s attitudes.

Third, we explain why we expect elite communication to shape citizens’ legitimacy beliefs toward IOs. We derive this general expectation from three core assumptions, which are anchored in theories of heuristic opinion formation and empirically applicable to opinion formation in the context of global governance: (1) citizens’ political awareness tends to be low, (2) citizens rely on heuristics to form political opinions, and (3) reliance on heuristics makes citizens susceptible to elite influence. The general takeaway is that citizens rely on elite communication as a heuristic when forming opinions toward IOs, just as they rely on heuristics when developing attitudes toward domestic political issues or making other decisions in their daily lives. If anything, elite influence should be even more pronounced in the context of global governance, which citizens tend to know relatively less well.

Finally, we explain when we expect elite communication to be particularly powerful in shaping citizens’ legitimacy beliefs. Consistent with our overall approach, we theorize conditions associated with the three core features of the communicative situation: the elite, the message, and the citizen. Specifically, we identify six moderating factors: elite credibility, elite polarization, tone of the message, object of the message, citizens’ political awareness, and citizens’ political beliefs. The central point is that citizens are varyingly susceptible to elite influence depending on a set of identifiable conditions in the communicative situation. These conditions shape the extent to which elite information about the institutional qualities of IOs influences citizens’ legitimacy beliefs.

Our theory is inspired by, and contributes to, three larger movements in political science research. The first is the turn toward an empirical study of legitimacy based on a sociological understanding of this concept (Caldeira and Gibson 1995; Gilley 2006; Esaiasson et al. 2012; Bernauer and Gampfer 2013; Tallberg et al. 2018; Dellmuth and Schlipphak 2020; Tallberg and Zürn 2019; Bexell et al. 2022; Dellmuth et al. 2022b; Sommerer et al. 2022). This move involves conceptualizing legitimacy in empirically tractable ways and developing theories of the determinants, processes, and consequences of legitimacy. The second is the ambition to learn from cognitive psychology when developing theories of political opinion formation. This motivation underpins the rich literature on cueing, framing, and heuristics in American and comparative politics (Sniderman et al. 1991; Mondak 1993; Lau and Redlawsk 2001; Chong and Druckman 2007a; Bullock 2011; Druckman et al. 2013; Carmines and d’Amico 2015), as well as the recent boom in international relations research on individual attitudes and beliefs (e.g., Hainmueller and Hiscox 2007; Tomz 2007; Rathbun 2012; Guisinger and Saunders 2017; Hafner-Burton et al. 2017; Lenz and Viola 2017). The third is the effort to theorize political processes that are generic in nature and defy conventional distinctions between domestic and international politics. We count legitimacy beliefs, elite influence, and heuristic opinion formation to such topics which benefit from a closer marriage between comparative politics and international relations (Milner 1998).

Conceptualizing and Operationalizing Legitimacy Beliefs

Our analytical focus are citizens’ legitimacy beliefs toward IOs. In this section, we specify this dependent variable in two steps. To begin with, we describe the sociological approach to legitimacy and explain how this understanding allows us to study legitimacy empirically, as a property of IOs derived from the beliefs of citizens. We then discuss alternative ways of operationalizing legitimacy beliefs and explain why we have chosen to rely on people’s confidence in an organization.

A Sociological Approach to Legitimacy

Legitimacy has two main alternative conceptual meanings: normative and sociological. Normative legitimacy refers to a governor’s right to rule based on its conformity to certain philosophically formulated values and principles, such as democracy, justice, and fairness. This is the notion of legitimacy that is studied in normative political theory (e.g., Buchanan and Keohane 2006; Christiano 2010). In contrast, sociological legitimacy refers to the beliefs or perceptions within a given audience that a governor’s exercise of authority is appropriate (e.g., Weber 1922/1978; Zürn 2018). Normative and sociological inquiries of legitimacy are thus guided by different questions. Normative studies typically ask: “By what ethical standards should we evaluate a governor’s right to rule, and how do particular governance arrangements measure up against these standards?” Meanwhile sociological inquiries ask: “To what extent, on what grounds, through what processes, and with what consequences is a governor perceived to be legitimate by a given audience?”

Considerable research on legitimacy in global governance has been normative. One strand of scholars has explored normative values associated with the input side of global governance, such as participation, transparency, and accountability (e.g., Held 1995; Dahl 1999; Christiano 2010; Scholte 2011; Archibugi et al. 2012). Another strand of scholars has emphasized values associated with the output side of global governance, assessing the extent to which IOs produce outcomes that contribute to welfare, justice, and fairness (e.g., Pogge 2002; Buchanan and Keohane 2006; Westergren 2016).

We join others in adopting a sociological understanding of the legitimacy of IOs (e.g., Hurd 2007; Reus-Smit 2007; Tallberg et al. 2018; Zürn 2018; Bexell et al. 2022; Dellmuth et al. 2022b; Sommerer et al. 2022). Central to this approach is a focus on legitimacy as an attribute of an IO, based on audience beliefs, and (de)legitimation as a process of justification and critique of an IO’s exercise of authority, with the aim of affecting audiences’ legitimacy beliefs (Tallberg and Zürn 2019). An audience comprises the group of actors whose beliefs are considered when evaluating the legitimacy of an organization (Bexell and Jönsson 2018). Audiences of IOs may include both state and societal actors, ranging from government elites to NGOs, business actors, and ordinary citizens. In this book, we focus exclusively on citizens as the relevant audience when examining the effects of elite communication on popular legitimacy perceptions.

Grounding the legitimacy of an IO in the beliefs of citizens has several implications. It means that the legitimacy of an IO ultimately is a subjective matter rooted in individual perceptions (Easton 1975; Tyler 2006). It also means that the legitimacy of a political institution is an aggregate construct, which may vary across various subsets of an audience, such as citizens in different countries or societal groups. Finally, it enables us to study drivers of legitimacy at the individual level, by examining what factors contribute to such beliefs among citizens.

When citizens develop legitimacy beliefs, they do so in a context of social norms about the appropriate exercise of authority. The social embeddedness of legitimacy beliefs suggests that sociological and normative conceptions of legitimacy may be empirically related while still analytically distinct (Beetham 1991; Keohane 2006; Bernstein 2011). The normative may shape the sociological inasmuch as audiences are influenced by philosophical standards of appropriateness when assessing an institution’s legitimacy. Conversely, the sociological may shape the normative inasmuch as standards developed in political theory reflect the time and place of their formulation. Showing such an interplay, both normative and sociological theories tend to emphasize an institution’s conformance to standards of democracy, effectiveness, and justice as a source of legitimacy (Scholte and Tallberg 2018).

The social embeddedness of legitimacy opens up possibilities for actors to affect popular legitimacy beliefs by invoking norms and values broadly associated with appropriate governance. In line with a growing literature, we refer to such justification and criticism aimed at affecting legitimacy beliefs as legitimation and delegitimation (Tallberg and Zürn 2019; Bexell et al. 2022). Thus, supporters of IOs may engage in legitimation practices that seek to cultivate beliefs in the legitimacy of these organizations (Symons 2011; Zaum 2013; Gronau and Schmidtke 2016; Dingwerth et al. 2019). Conversely, opponents of IOs may engage in delegitimation practices that aim to undermine beliefs in the legitimacy of these organizations (O’Brien et al. 2000; Gregoratti and Uhlin 2018; Zürn 2018). Elite communication, as theorized later in this chapter, represents a form of discursive legitimation and delegitimation of IOs.

Operationalizing Legitimacy Beliefs

Legitimacy beliefs may be operationalized in multiple ways for empirical research. All alternatives have their strengths and weaknesses, and all involve an element of simplification, as we move from theoretical conceptualization to empirical measurement. “However defined, ‘legitimacy’ belongs to the family of abstract concepts that are hard to measure directly” (Esaiasson et al. 2012, 790). Yet, as Gilley puts it: “[T]he complexity of a concept is neither a valid objection nor an insuperable obstacle to its measurement” (2006, 500). In the following, we distinguish between four alternative ways of operationalizing legitimacy beliefs, building on existing work in comparative politics and international relations. While all four approaches take individuals and their attitudes as the starting point, they are tied to different conceptualizations of legitimacy and lead to different suggestions for measures to use in empirical research.¹ We conclude by explaining why we have chosen to opt for an individual’s confidence in an institution as our preferred measurement of legitimacy beliefs.

The justification approach measures legitimacy beliefs as individuals’ perceptions that an institution conforms to normative criteria motivating its right to rule. This approach is tied to a conceptualization of legitimacy that combines normative and sociological elements, by restricting legitimacy to support for an institution that arises from certain morally appropriate justifications of its authority (Beetham 1991; Agné 2018). What justifications count as appropriate is established externally by researchers, drawing on normative political theory. In empirical research, this approach is expressed through attitudinal measures that seek to capture individual beliefs in an institution’s conformance to these justifications, such as whether an institution makes decisions democratically or produces fair outcomes. In this vein, Esaiasson et al. (2012) measure legitimacy beliefs as perceptions that an institution’s procedures are fair, while Bernauer and Gampfer (2013) rely on assessments of transparency, representation, skill, and expertise in an institution, and Agné et al. (2015) on perceptions of an institution’s conformance to standards of representation, deliberation, and accountability.

The approval approach measures legitimacy beliefs as individuals’ deep-seated approval of an institution. This approach reflects Easton’s (1965, 1975) conceptualization of legitimacy beliefs in terms of diffuse support, as distinguished from specific support. Diffuse support refers to “a reservoir of favorable attitudes or good will that helps members to accept or tolerate outputs to which they are opposed” (Easton 1965, 273). Such foundational approval is distinct from specific support for an institution that rests exclusively on particular policies or actions. From an approval perspective, legitimacy beliefs involve more durable and fundamental backing of an institution. The grounds for this approval lie with individuals’ subjective attitudes, and it is not for researchers to specify what counts as the “right” reasons for approval, as in the justification approach. Empirical research following the approval approach has usually operationalized legitimacy in terms of confidence or trust. While asking individuals about their support for an institution presents an additional alternative, it is usually discarded in research on legitimacy, as support can involve short-term specific support rather than deeper attachment (Dellmuth and Schlipphak 2020; Tallberg and Zürn 2019).

The acceptance approach measures legitimacy beliefs as individuals’ willingness to defer to an institution. This approach is linked to a conceptualization of legitimacy that emphasizes consent, acceptance, deference, obedience, and compliance (e.g., Bodansky 1999; Tyler 2006). In this vein, Levi et al. conceive of legitimacy as “a sense of obligation or willingness to obey authorities (value-based legitimacy) that then translates into actual compliance with governmental regulations and laws (behavioral legitimacy)” (2009, 354). Building on this conceptualization, they then measure legitimacy beliefs as the extent to which people are willing to defer to government authorities. In a similar way, Anderson et al. (2005) examine the legitimacy of political systems based on the degree to which losers in elections accept outcomes and extend their support to the new regime.

Finally, the multidimensional approach measures legitimacy beliefs by way of a combination of the other three approaches. This approach builds on the idea that legitimacy as a concept incorporates multiple dimensions, including justifications, approval, and acceptance, and therefore is too rich to capture through measures focused on a single dimension. This approach was particularly influential in comparative politics in the 1990s and 2000s, when several scholars sought to explain a decline in the legitimacy of political systems around the world (Weatherford 1992; Norris 1999; Gilley 2006; Booth and Seligson 2009). For instance, Norris (1999) marshalled indicators of five components of political legitimacy: attitudes toward political community, regime principles, regime performance, regime institutions, and political actors. In the study of global governance, Anderson et al. (2019) follow this approach when measuring legitimacy using five items that capture substantive support, principled approval, and deference in the area of climate change.

In this book, we have chosen to rely on the approval approach in general and the confidence measure in particular. We use individuals’ confidence in IOs as our preferred measurement of legitimacy beliefs in all empirical analyses in Chapters 4–7, recognizing that confidence and trust are substitutable measures of approval. Our choice of measure does not reflect a principled rejection of alternative strategies, as much as a recognition of the comparative advantages of this operationalization for our purposes. More specifically, this choice is based on three considerations: conceptual fit, research purpose, and scientific cumulativity.

First, the confidence measure aligns well with our conceptualization of legitimacy as the belief that an institution exercises its authority appropriately. By capturing individuals’ general faith in an institution, it taps into that reservoir of long-term support that is indicative of legitimacy. Confidence picks up on a sense of institutional attachment and on a willingness to put one’s judgement in the hands of that institution. This link is borne out empirically in Gibson et al. (2003, 361), which shows how confidence in an institution reflects a combination of diffuse support measures and is not driven by specific support for certain policies.

Second, the confidence measure has advantages when studying sources or effects of legitimacy. Different from the other three approaches, the approval approach does not integrate into the privileged measure either potential sources of legitimacy (such as fairness or effectiveness) or potential consequences of legitimacy (such as acceptance or compliance). Relying on confidence thereby avoids a conflation between the operationalization of legitimacy and potential sources or effects of legitimacy (Mishler and Rose 2001, 40–41; Booth and Seligson 2009, 12). This is a central concern for us, since we are interested in establishing the sources of legitimacy beliefs, including the role of elite communication about institutional features of IOs, such as democracy, effectiveness, and fairness.

Third, the confidence measure facilitates scientific progress by allowing us to relate the findings of this book to the large literature in comparative politics and international relations that already employs this indicator of legitimacy (e.g., Caldeira 1986; Newton and Norris 2000; Inglehart and Welzel 2005; Norris 2009; Bühlmann and Kunz 2011; Voeten 2013; Dellmuth and Tallberg 2015, 2020; Dellmuth et al. 2022b). Next to scattered observational data, the absence of consistent measures for legitimacy is a prominent reason why scientific cumulativity has been slow in this field (Dellmuth 2018).

Having conceptualized and operationalized legitimacy beliefs, we now turn to our theory of why and when elites shape such beliefs through their communication. We outline this argument in three steps. First, we identify the key players and narratives in elite communication about global governance. We suggest that the central communicators include both globally and domestically oriented elites and that the key focus of communication are the institutional qualities of IOs. In a second step, we explain why elite communication should matter for citizens’ legitimacy beliefs, grounding the argument in theories of heuristic opinion formation and contextual conditions in global governance. Finally, we theorize the conditions under which elite communication should be more or less important in shaping citizens’ legitimacy beliefs in the global realm. Consistent with our overall theory, we identify conditions associated with the elite, the message, and the citizen.

Elites and Messages in Global Governance

Who are the elites engaged in communication in global governance and what do they communicate about? While research on elite communication in the domestic setting conventionally focuses on political parties communicating about issues and candidates, global governance presents a different context. In the following, we describe what we consider to be the main messengers and messages in elite communication about IOs, thus identifying the central contextual components and boundaries of our topic.

The first important component is the messenger – the elite conveying information about an IO through a communicative act. One of the hallmarks of global governance is the multitude of elite actors and organizations that aspire to influence its outcomes. We regard both global and domestic elites as relevant communicators with a potential to shape how citizens think about the legitimacy of IOs. While this distinction between global and domestic elites is a simplification, since most elites these days are transnational, it is helpful for organizing our analyzes.

At the global level, elites that frequently communicate about IOs include states, nonstate actors, and IOs themselves. Member governments often criticize IOs in order to force policy change, deflect blame or mobilize domestic supporters, but may also endorse them to rally support for ambitious goals, lock in policy preferences and protect multilateral arenas (Morse and Keohane 2014; Zürn 2018; Tallberg and Verhaegen 2020; Kentikelenis and Voeten 2021). Nonstate actors, such as NGOs, business associations, and philanthropic foundations, frequently challenge IOs for insufficiently ambitious policies and undemocratic decision-making procedures, but may also praise them for their policy achievements and efforts to consult with stakeholders (O’Brien et al. 2000; Gregoratti and Uhlin 2018). Finally, IOs themselves not only increasingly engage in self-legitimation, trumpeting their democratic credentials, expertise-based policies and critical achievements, but also occasionally admit to mistakes when seeking to minimize political damage or generate support for reorientation (Zaum 2013; Gronau and Schmidtke 2016; Dingwerth et al. 2019; Von Billerbeck 2020).

At the domestic level, political parties and candidates frequently communicate about IOs when debating political issues on which these organizations have prominent roles, such as climate change, pandemics, and international trade. While research in international relations, with some notable exceptions (e.g., Gabel and Scheve 2007; Guisinger and Saunders 2017; Dür and Schlipphak 2021), has been slow to acknowledge the role of political parties as communicators, we consider partisan politics an important arena of contestation over IOs. Political parties not only communicate about domestic concerns but also often take positions on international issues involving IOs as well. Parties in the political mainstream tend to be relatively supportive of international cooperation aimed at solving cross-border problems. In contrast, parties on the far right tend to contest IOs because of their constraints on national sovereignty, while parties on the far left tend to challenge IOs because of insufficient representation and redistribution (de Vries and Edwards 2009; Hooghe et al. 2019). Given the central role that parties occupy in structuring the choices that citizens confront (Druckman et al. 2013; Leeper and Slothuus 2014), we expect their influence over citizens to extend to international issues as well.

The elites that engage in communication about IOs are the focus of two empirical chapters in the book. In Chapter 4, we examine the effects of communication by globally active elites, in the shape of member governments, NGOs, and IOs themselves. In Chapter 5, we turn to communication by domestically oriented elites, in the shape of political parties, examining their influence on citizens’ legitimacy beliefs vis-à-vis IOs. These two chapters are complementary, by exploring communication by elites that primarily are active at two different levels of governance and usually the topic of separate literatures in international relations and comparative politics.

The other important contextual component is the message – the information about an IO that is conveyed to citizens. By definition, all communication about IOs refers to the organization in some capacity. We assume that elite communication typically attempts to affect individual attitudes by invoking the institutional qualities of IOs. We conceive of an IO’s institutional qualities in broad terms, distinguishing between four main features: the organization’s purpose, the authority it has been granted, the procedures it uses to make decisions, and the performance it achieves in terms of outcomes. Elites communicate about the institutional qualities of IOs because they expect these features to matter for people’s attitudes toward these organizations.

The identification of procedure and performance as two discrete institutional qualities draws on Scharpf’s (1999) influential distinction between the processes by which an IO makes decisions (“input”) and the consequences of this decision-making (“output”). The dichotomy of procedure versus performances structures most research on how institutional qualities of IOs may matter for people’s legitimacy beliefs toward these organizations (e.g., Hurd 2007; Bernauer and Gampfer 2013; Binder and Heupel 2015; Scholte and Tallberg 2018; Strebel et al. 2019; Tallberg and Zürn 2019; Bernauer et al. 2020). Procedure encompasses qualities such as participation, accountability, deliberation, transparency, efficiency, legality, impartiality, proportionality, and expertise. Performance includes qualities such as problem-solving, collective gains, rights protection, and distributive justice.

Authority and purpose have so far received less attention in empirical studies of IO legitimacy. Authority refers to an IO’s formal right to make decisions in a particular area. More specifically, the authority of an IO reflects the degree to which member states have conferred policy-making competences, delegated independent authority to supranational bodies, and pooled authority in collective decision-making (Zürn et al. 2012; Hooghe et al. 2017; Zürn 2018). Purpose refers to the organizational goals of IOs – their stated normative objective (Dowling and Pfeffer 1975; Scott 1991; Lenz and Viola 2017). For instance, the purpose of an IO may be to protect human rights, fight pandemics, reduce poverty, promote free trade, ensure peace and security, or combat climate change. Some IOs, such as the UN and the EU, combine multiple goals in general-purpose designs, while other IOs, such as the WHO and the WTO, focus on particular purposes in task-specific designs (Lenz et al. 2015). Purpose is distinct from performance: While purpose indicates the normative goal of the IO, performance refers to its success in achieving this goal.

Communication about the institutional qualities of IOs is at the center of two empirical chapters in the book. In Chapter 6, we examine how information about procedure- and performance-related qualities of IOs affects people’s legitimacy beliefs. In Chapter 7, we extend the analysis to also consider the impact of information on the authority and purpose of IOs. Our argument gives us not reason to expect a priori that information about one particular type of institutional quality would be more influential than information about some other institutional quality. We therefore focus on developing the logic for why information about each quality should matter for legitimacy beliefs and on assessing the conditions under which such communication effects are stronger or weaker.

Why Elite Communication Matters

Why would elite communication about the institutional qualities of IOs affect the way citizens think about the legitimacy of these organizations? Our core argument is that citizens care about the qualities of IOs but lack sufficient information to form independent opinions, leading them to rely on communication by elites as a cognitive shortcut. While elite communication thus offers an efficient solution to an information deficit experienced by citizens, it simultaneously opens up possibilities for elites to shape the opinions of citizens.

This logic rests on three key assumptions about heuristic opinion formation. These assumptions are anchored in cognitive psychology, inform research on domestic opinion formation, and apply to contextual conditions present in global governance. Together, these assumptions suggest why citizens make use of heuristics when forming political opinions, why elite communication presents a form of heuristic, and why such communication affects citizens’ legitimacy beliefs toward IOs.

Citizens’ Political Awareness Tends to Be Low

The starting assumption for understanding heuristic opinion formation pertains to individuals’ political awareness. According to a long line of studies, awareness tends to be quite low (e.g., Converse 1964; Delli Carpini and Keeter 1996; Lupia 2016). While there is variation across individuals, people on average have a relatively weak grasp of politics. As Zaller (1992, 18) puts it: “The two main points about political awareness […] are (1) that people vary greatly in their general attentiveness to politics, regardless of particular issues; and (2) that average overall levels of information are quite low. More succinctly, there is high variance in political awareness around a generally low mean.”

This finding has been attributed to a number of conditions. Classic theories in cognitive psychology and behavioral economics focus on individuals’ cognitive limitations in information processing. Simon (1957, 1982) famously argued that people are boundedly rational agents who experience limits in thinking capacity, information, and time, making them satisfice rather than optimize their choices when making decisions. Building on Simon’s insights, Tversky and Kahneman (1974) took individuals’ cognitive limitations and bounded rationality as the starting point for an influential research program on how people make decisions under uncertainty. From this perspective, citizens’ low awareness about politics is nothing particular for this domain, but the expression of a general limitation in humans’ ability to process information and a “rational” adaptation to this condition. The complexity of the social world in combination with limitations in time and cognition means that citizens are selective in their learning.

In political science, one strand of research links citizens’ limited political awareness to inconsistency in individual belief systems. According to Converse’s (1964) seminal contribution, citizens’ belief systems suffer from weaknesses in three connections: horizontally, between opinions on different issues; vertically, between abstract concepts and positions on specific issues; and temporally, between positions on issues taken at different times. Research after Converse concludes that citizens’ opinions tend to be minimally consistent, minimally stable, and rest on minimal levels of comprehension of political abstractions (Sniderman et al. 1991, 2). While this picture has not stood unchallenged given the crude measures and pessimistic assumptions it is based on (cf. Sniderman et al. 1991; Dolan 2011; Cohen and Luttig 2020), it has proven highly influential in the field (Chong and Druckman 2007a, 103; Levendusky 2010, 112).

Other research in political science attributes citizens’ low awareness about politics to structural conditions, such as economic inequalities and flailing educational systems. In this vein, Solt (2008, 48) argues that economic inequalities in a society put a damper on all but the most affluent citizens’ incentives to develop political knowledge and become politically engaged, concluding that “higher levels of income inequality powerfully depress political interest, the frequency of political discussion, and participation in elections.” Delli Carpini and Keeter (1996, 188), for their part, highlight limitations in education as the foremost source of low political awareness, establishing that education is “the strongest single predictor of political knowledge.”

The take-home message of these accounts is the same: People tend to possess limited information about politics. Documenting these limitations has been a favorite pastime among public opinion researchers, usually with reference to the US context. Yet people in comparable countries appear to be only slightly better informed than Americans (Delli Carpini and Keeter 1996, 89–91). Delli Carpini and Keeter (1996, 101–102) offer the most comprehensive overview and find the overall picture disturbing from a democratic perspective:

Many of the facts known by relatively small percentages of the public seem critical to understanding – let alone effectively acting in – the political world: fundamental rules of the game; classic civil liberties; key concepts of political economy; the names of key political representatives; many important policy positions of presidential candidates or the political parties; basic social indicators and significant public policies.

We have good reasons to believe that this pattern extends to people’s awareness of global governance. In fact, citizens are probably even less informed about international cooperation than about domestic politics. Global governance is more remote and complex than national politics, IOs are less frequently covered in national media than domestic institutions, and the transnational public sphere is less developed than its domestic counterparts. Citizens therefore rarely have access to rich and varied information about IOs. Even if citizens have usually heard of the most prominent IOs (Gallup International Association 2005, 2011), they likely have limited knowledge about their political mandates, decision-making procedures, and policy impacts (Dellmuth 2016). This assumption is substantiated by recent data on people’s awareness of global governance based on three knowledge items included in the WVS7, fielded during the time period 2017–2020 (see Chapter 2).

This picture of citizens’ awareness of global governance is consistent with findings on people’s knowledge regarding foreign affairs more broadly. Delli Carpini and Keeter (1996, 82–86) find that only 14 percent of more than 500 questions regarding world politics were answered correctly by at least three quarters of US survey respondents. Figures such as these have contributed to a predominant view among students of foreign policy and public opinion that citizens tend to possess relatively little information about foreign affairs, except when such issues become highly politically salient or touch directly on people’s everyday lives (e.g., Aldrich et al. 1989; Berinsky 2009; for a discussion, see Kertzer and Zeitzoff 2017, 544–545).

Citizens Rely on Heuristics to Form Political Opinions

Our second core assumption is about the role of heuristics in helping citizens to form political opinions. When individuals confront limitations in processing relevant information, they rely on cognitive shortcuts – heuristics – to make decisions. This fundamental insight in cognitive psychology has been highly influential in the study of public opinion, where it offers the favored answer to the puzzle of how citizens make up their minds on political issues under conditions of uncertainty. It suggests that individuals are “cognitive misers” (Fiske and Taylor 2017), who have become quite skilled at applying various forms of heuristics to compensate for their own limitations. Part of why it has been so influential is because it portrays political opinion formation as an extension of generic cognitive processes. If people use cognitive shortcuts to make up their minds about all kinds of things, why not politics as well?

Heuristics are problem-solving strategies, employed automatically or unconsciously, that serve to keep information-processing demands within bounds (Lau and Redlawsk 2001, 952; see also Sniderman et al. 1991, Ch. 2; Shah and Oppenheimer 2008; Gigerenzer and Gaissmaier 2011). Simon’s (1957) strategy of satisficing is an early example of a heuristic used to deal with the impossible information demands involved in optimizing choices. Kahneman and Tversky deepened and broadened the study of heuristics (Tversky and Kahneman 1974; Kahneman et al. 1982). On the one hand, they identified several prominent heuristics that people use when making judgments under conditions of uncertainty. On the other hand, they linked people’s use of heuristics to systematic biases in judgment. While presenting an efficient strategy for dealing with cognitive limitations, heuristics simultaneously cause systematic errors in judgment, according to this argument. Cognitive psychology suggests that all humans rely on cognitive shortcuts. While not everyone relies on the same heuristics to the same extent, all people simplify choices in life by way of cognitive short cuts. Heuristics are associated with System I of the human intellect, which is fast, automatic, and instinctive, and thus different from System II, which is slower, more deliberative, and involves more effort (Stanovich and West 2000; Kahneman 2011).

For political scientists, heuristics offer a compelling answer to the question of how people make political choices in the face of limited information. If individuals generally use heuristics to deal with uncertainty in their everyday lives, would they not use similar strategies when making political choices? The consensus position is affirmative and can be summarized in two points (Lau and Redlawsk 2001, 952; see also, e.g., Hamill et al. 1985; Sniderman et al. 1991; Lupia 1994; Lau and Redlawsk 2001; Carmines and d’Amico 2015). First, everyone uses cognitive shortcuts in thinking about politics. It is not limited to a particular group, such as more uninformed citizens or political experts. Second, the use of heuristics partially compensates for limited political awareness, so that individuals who are less informed nonetheless can make reasonably accurate political judgments.

Research in comparative politics highlights how political parties, political ideology, and political endorsements present people with useful heuristics (e.g., Lodge and Hamill 1986; Sniderman et al. 1991; Lau and Redlawsk 2001; Bullock 2011; Torcal et al. 2018). According to this literature, citizens typically look to political parties for cues. As long as citizens know what parties they like or dislike, they can use basic information about a party’s position to figure out their own opinion. Political ideology works in a similar way: provided that citizens identify more or less with specific ideologies, information on whether an argument or a candidate is advanced by the left or the right offers a shortcut to opinions. Finally, political endorsements can help citizens to arrive at opinions on complex matters. Rather than gathering information and carefully thinking through the options, people choose to listen to political elites they trust. These political heuristics are not always neatly distinguishable in practice. Yet their impact is the same: “Citizens can overcome informational shortfalls about policies, not because they (mysteriously) can simplify public choices effectively, but because their choices are systematically simplified for them” (Sniderman 2000, 81).

There is much to suggest that citizens rely on heuristics also when forming opinions toward global governance (for a theoretical argument, see Lenz and Viola 2017). To begin with, research in the European context shows that political parties and political ideology serve the same function in structuring attitudes toward the EU as they do in shaping domestic public opinion (Hooghe and Marks 2005; Hobolt 2007; Hellström 2008; Torcal et al. 2018). These studies extend the logic of partisan politics at the domestic level to governance at the regional level. The central argument is that “European integration is too complex and remote from the daily lives of most citizens for them to have sufficient interest, awareness, or emotional attachment to base their attitudes on an evaluation of the implications of the integration process. Instead, citizens rely on proxies or cues to overcome their information shortfalls” (Hobolt and de Vries 2016, 421–422).

In addition, a number of studies suggest that citizens use their attitudes toward domestic political institutions they know better as a heuristic when forming attitudes toward IOs they know less well (Harteveld et al. 2013; Voeten 2013; Armingeon and Ceka 2014; Dellmuth and Tallberg 2015; Schlipphak 2015). The evidence for this heuristic is a stable and strong correlation between people’s attitudes toward domestic political institutions and IOs, consistent across institutions, time, measures, and surveys (for a discussion, see Dellmuth and Tallberg 2021). In this vein, Armingeon and Ceka (2014, 82) conclude that “support for the EU is derived from evaluations of national politics and policy, which Europeans know far better than the remote political system of the EU.” Similarly, Schlipphak (2015, 367) suggests that this link in attitudes reflects “general satisfaction with or broad trust in domestic political actors that is extrapolated to a more diffuse trust of political actors operating at other levels beyond the national level.”

Finally, some accounts suggest that citizens are rationally ignorant about world politics, as these issues are far removed from their daily lives and therefore rely on a variety of heuristics to form opinions on these matters. Berinsky (2009) argues that citizens consider the positions of party elites when making up their minds about foreign policy issues. Kertzer and Zeitzoff (2017) submit that citizens rely on informational cues from peers in society, next to information from political elites, when developing opinions about foreign policy. Nielson et al. (2019) show that NGO representatives tend to use heuristic shortcuts when taking a stand on the legitimacy of election observer organizations. Guisinger and Saunders (2017, 425) summarize: “Foreign affairs are distant from most voters’ everyday concerns and thus are especially ripe for cue-giving by elite actors.”

Heuristics Make Citizens Susceptible to Elite Influence

The third core assumption focuses on the implications of people relying on heuristics. It suggests that the use of cognitive shortcuts comes with a consequence, namely, that people develop opinions and make choices influenced by these heuristics. In cognitive psychology, this possibility has spurred a debate on whether heuristics lead to biased or accurate judgments. In political science, it has invited analyses of how elites may influence public opinion through cueing and framing.

The implications of people relying on heuristics are front and center in cognitive psychology. Two alternative accounts dominate the debate. The standard assumption has been that heuristics, while helping people to make choices, simultaneously produce systematic errors in judgment. Indeed, Kahneman and Tversky’s original research program is generally known as “heuristics and biases,” and an important part of their classic 1974 contribution was to link the use of specific heuristics to concrete biases in judgment. This insight has informed dual-processing accounts in which biases typically are associated with the fast, automatic, and instinctive System I, rather than the slow, deliberative, and thoughtful System II (Evans 2008; Kahneman 2011).

This perspective is challenged by Gigerenzer and colleagues, who suggest that heuristics in fact may produce more accurate judgments than more complicated cognitive strategies (Gigerenzer and Brighton 2009; Gigerenzer and Gaissmaier 2011; Gigerenzer 2015). Using simple decision rules that ignore part of the information can yield greater predictive accuracy. Heuristics appear to involve less-is-more effects, stemming from “an inverse-U-shaped relation between the level of accuracy and the amount of information, computation, or time” (Gigerenzer and Gaissmaier 2011, 453).

This debate has carried over to political science. Mirroring the discussion in cognitive psychology, political scientists have debated the effects of heuristics on decision efficacy, citizen competence, and democracy at large. On the one hand, party cues, ideological prompts, and elite endorsements may lead citizens to disregard other information, shy away from independent assessments, and arrive at positions that do not fully reflect their preferences (Kuklinski and Quirk 2000). In a larger perspective, substituting one’s own judgment with a mechanical adoption of partisan positions may even hurt democracy itself (Achen and Bartels 2016). On the other hand, political heuristics may make citizens more efficacious by guiding them to the “correct” political choices in view of their preferences (Lupia and McCubbins 1998; Lau and Redlawsk 2001). Partisan cues thus lead to higher – not lower – quality decisions. In the larger perspective, it is the influence of parties, ideologies, and elites on citizens that should be credited for the consistency in public opinion that is so essential to democracy (Sniderman and Bullock 2004; Leeper and Slothuus 2014).

Yet, most important for our purposes, the use of heuristics in politics has spawned a rich research agenda on elite influence. The starting point is the argument that citizens turn to elites for information that can simplify their political choices.² Downs (1957, 233) offers an early (and somewhat antiquated) statement of this idea: “[A citizen] cannot be expert in all fields of policy that are relevant to his decision. Therefore he will seek assistance from men who are experts in those fields, have the same political goals he does, and have good judgment.” Zaller (1992, 6), in another classic contribution, explains how ordinary citizens, by necessity, rely on elites to make sense of the political world: “To an extent that few like but none can avoid, citizens in large societies are dependent on […] others for most of their information about the larger world in which they live. […] The ‘others’ on whom we depend, directly or indirectly, for information about the world are […] political elites.”

Building on this insight, later work focuses on the communicative processes through which elites influence public opinion, distinguishing between cueing and framing. A cue refers broadly to “a piece of information that allows individuals to make inferences without drawing on more detailed knowledge” (Druckman et al. 2010, 137; see also Bullock 2011, 497). Cueing effects, then, arise whenever people’s opinions or choices are influenced by this limited piece of information. Such effects are at the core of the rich literature on party cues, which examines how simple information about partisanship shapes people’s opinions on issues and candidates (e.g., Lupia and McCubbins 1998; Hobolt 2007; Levendusky 2010; Bullock 2011; Leeper and Slothuus 2014; Torcal et al. 2018).

A frame, on the other hand, refers to a particular definition or interpretation of a political issue (Chong and Druckman 2007a). Framing effects arise whenever individuals reorient their thinking about an issue because of how it is presented. A classic example is that of attitudes toward a hate group rally, which become more positive when this rally is discussed in terms of free speech than when it is discussed in terms of public safety (Druckman et al. 2010). Framing effects are ubiquitous in politics and have given rise to an impressive literature on the conditions under which such influence is likely (e.g., Iyengar 1991; Druckman 2004; Chong and Druckman 2007a; Busby et al. 2018).

There are indications that cueing and framing by elites may influence the opinions of citizens on international issues as well. First, studies of public opinion in Europe show that citizens’ reliance on party cues affects their support for the EU. For instance, Gabel and Scheve (2007) find that more negative elite messages about European integration result in decreased public support for the EU. Maier et al. (2012) establish that party cues about economics and identities shape citizen support for the EU, especially among people who identify with the respective parties. Hobolt (2007) and Torcal et al. (2018) examine how party cues help to increase voter competence by giving citizens the information they require to make choices in line with their political preferences.

Second, studies of US public opinion on foreign policy demonstrate that party cues shape people’s positions toward international affairs. Berinsky (2009) examines public opinion in times of war and finds that patterns of consensus and dissensus among US political elites influence people’s attitudes. Guisinger and Saunders (2017) show how party cueing results in differential effects across international issues depending on the baseline distribution of mass opinion on these issues. Kerzer and Zeitzoff (2017) confirm that cues from party elites shape American’s opinions on foreign policy, while also demonstrating that cues from social peers matter. Zvobgo (2019) examines how competing frames about human rights and national interests affect US public opinion on the issue of whether or not to join the ICC. Taken together, these studies from the European and US contexts suggest that citizens’ reliance on heuristics make them susceptible to elite influence on international issues as well.

When Elite Communication Matters

Our general expectation is that elite communication about IOs affects citizens’ legitimacy beliefs toward these organizations. Yet elites are unlikely to be equally influential under all circumstances. While some conditions favor greater effects of elite communication on people’s legitimacy beliefs, other conditions restrict the scope for elites to shape citizens’ attitudes toward IOs. In this section, we therefore consider the conditionality of elite communication. Under what circumstances should we expect elite communication to be more or less influential in relation to citizen legitimacy beliefs?

We theorize conditional effects arising from all three constitutive components of the process of elite communication: the elite, the message, and the citizen. The moderating factors we consider are all consistent with the underlying logic of heuristic opinion formation, as developed in cognitive psychology and applied in comparative politics.³ First, the ability of elites to shape popular legitimacy beliefs is likely conditioned by credibility of these elites in the eyes of citizens, as well as the extent to which elites are polarized over IOs. Second, elite influence likely varies with tone of the messages, as well as the extent to which messages target IOs that have been subject to more or less prior contestation. Third, elite communication likely affects individuals differently, depending on their degree of political awareness and their political beliefs. In the following, we outline the general logic of each expectation. In the empirical chapters that follow, we formulate specific hypotheses to test these propositions.

Elite Credibility

Elite credibility refers to the belief that a speaker has relevant knowledge and can be trusted to reveal that information accurately (Lupia 2000). The basic expectation is that elites who are perceived as more credible are more likely to sway the opinions of citizens. According to this view, citizens are not mindless followers of elites who might want to manipulate them, but instead listen specifically to the elites they perceive to be credible on a particular issue. There are reasons to expect that elites exhibit varying degrees of credibility when communicating about IOs, which would have implications for the effects of this communication on citizens’ legitimacy perceptions. Based on this general expectation, we theorize the drivers of elite credibility in the domestic and international contexts, respectively.

In the domestic context, we expect partisanship to be the central source of credibility for elites communicating about IOs. The general idea is that people will follow cues from parties they trust, while neglecting cues from parties they distrust, especially in polarized political environments. This association between credibility and partisanship is rooted in an understanding of domestic politics as structured in partisan terms (Sniderman 2000; Leeper and Slothuus 2014). On the one hand, parties order the political choices that citizens confront. On the other hand, citizens display partisan identification, understood as deep and emotional attachments to political parties. When citizens identify with or lean toward a specific political party, they therefore tend to interpret information from that party as more credible than alternative information from opposing parties. This process involves an element of motivated reasoning, that is, a tendency to seek out information that confirms prior beliefs (Druckman et al. 2013, 59).

We build on this general logic in Chapter 5 when we assess the conditions under which domestic party elites shape citizens’ perceptions of IO legitimacy.

In the global context, where elites are less strongly linked to specific partisan positions, and the issues often are less polarizing, we expect credibility instead to be tied to perceptions of impartiality. This expectation of a link between credibility and impartiality is inspired by research on the impact of expert endorsements, which shows that experts can affect public opinion by virtue of their perceived unbiased knowledge (Guisinger and Saunders 2017; Maliniak et al. 2021), as well as research on media priming, which shows that news sources perceived as authoritative are more likely to produce framing effects (Miller and Krosnick 2000; Druckman 2001). Analogously, we expect that citizens consider whether elites in global governance can be expected to hold and reveal accurate information about IOs. Elites who have greater incentives to convey biased information about IOs are less likely to be seen as credible sources. Conversely, elites who stand to gain less from how IOs are perceived can be expected to communicate more honestly about these organizations.

We examine this expectation in Chapter 4, when we explore the conditions under which member governments, NGOs, and IOs themselves affect citizens’ legitimacy beliefs toward IOs.

Elite Polarization

Elite polarization refers to the extent to which elites are divided on the issues they communicate about. The expectation that elite polarization matters has been developed in the context of party cueing in domestic politics, where polarization is conceptualized as having two components: the ideological distance between parties on a particular issue and the ideological homogeneity within each party on this issue (Levendusky 2010, 118). The expectation is that greater polarization leads to stronger cueing effects among people sympathizing with the communicating party, since polarization leads to clearer signals.

On this view, polarization works by strengthening or weakening the impact of people’s partisan identification on their political attitudes or voting behavior (Levendusky 2010; Druckman et al. 2013). Provided that people have partisan leanings and are prone to take cues from their favored party, polarization helps to facilitate this process. “When elites are polarized, they send voters clearer signals about where they stand on the issues of the day: the parties’ positions are distinct from one another, and each party is more internally ideologically homogeneous. In this environment, it is easier for an ordinary voter to follow his party’s cue” (Levendusky 2010, 114–115).

This expectation is consistent with Zaller’s (1992) seminal model, which suggests that consensus among political parties increases public support for government policy, while dissensus leads citizens to follow those elites who share their political beliefs. Berinsky (2009) and Gabel and Scheve (2007) have subsequently found support for this basic claim in studies of US attitudes toward war and public opinion toward the EU.

We test this expectation in Chapter 5, when we assess if communication by political parties about IOs affects citizens differently depending on the degree of elite polarization.

Tone of Message

Elite communication in global governance comprises both positive messages that endorse, praise, and defend IOs, and negative messages that challenge, criticize, and dismiss IOs. A growing literature refers to such positive and negative communication as legitimation and delegitimation of IOs (Zaum 2013; Binder and Heupel 2015; Tallberg et al. 2018; Zürn 2018). While legitimation aims to boost legitimacy beliefs through positive communication about IOs, delegitimation aims to undermine legitimacy beliefs through negative messages about the same organizations. Tone is in many ways at the core of communication as a practice in global governance. It is through the inclusion of an evaluative tone that elite messages are charged and become potentially influential communicative practices.

We expect that negative elite messages will have stronger effects on legitimacy perceptions than positive messages. We base this expectation on theories in comparative politics, economics, and psychology. While identifying slightly different mechanisms, all ground their expectations in general sociopsychological dynamics, and all suggest that negative messages should have a larger impact than positive.

First, research on voting behavior shows that people respond asymmetrically to positive and negative information about the economy (Bloom and Price 1975; Soroka 2006). Since people tend to be slightly optimistic in their basic predisposition, negative information usually diverges more from people’s reference points than positive information and therefore has a greater impact on attitudes and behavior. This dynamic has been identified in communication about political candidates and institutions as well (e.g., Lau 1985).

Second, prospect theory suggests a similar story of asymmetry, highlighting a complementary mechanism. It submits that individuals tend to be risk averse, weighing potential losses more heavily than potential gains (Kahnemann and Tversky 1979; Tversky and Kahnemann 1981). Such loss aversion leads people to react more strongly to negative information than to positive, for instance, by cutting back consumption more sharply in response to bad news than they expand consumption in response to good news.

Third, psychological research on impression formation establishes that bad emotions weigh more heavily than good emotions (Baumeister et al. 2001). Negative information tends to be processed more thoroughly, be stickier, and have greater impact than positive information.

We test this expectation in Chapter 4, when comparing the effects on legitimacy beliefs of positive and negative messages communicated by member governments, NGOs, and IOs.

Object of Message

IOs are subject to public debate to varying extents. While some IOs fly beneath the radar of public contestation, others are continuously discussed in society. When messages target IOs that already have been subject to extensive societal contestation, the likelihood of communication effects is significantly reduced. Citizens have then developed stronger priors about the IO in question, reducing the probability that additional information will shift their opinions. This is a different way of saying that prior elite communication has consequences for the effectiveness of future elite communication. As Levendusky (2010, 120) notes: “For more crystallized issues, the impact of elites has already been absorbed into citizens’ attitudes.”

Research in political psychology suggests that the strength of people’s priors has a negative conditioning effect on the influence of communicative processes, such as cueing, framing, persuasion, and priming (Chong and Druckman 2007a, 111–112). When people hold strong prior beliefs on a topic, this opinion is unlikely to change in view of new information and likely to continue dictating actions and thoughts. In contrast, when people hold weak prior beliefs on a topic, affecting this opinion is considerably easier, even if the new opinion is unlikely to stick unless it is simultaneously strengthened.

The process by which priors are strengthened is motivated political reasoning (Taber and Lodge 2006). Motivated reasoning occurs when individuals seek out information consistent with their prior opinions, while rejecting information inconsistent with those prior opinions (Kunda 1990). When people already have strong priors, they rely more extensively on motivated political reasoning. As Druckman and Leeper (2012, 877) explain: “Those with stronger attitudes are substantially more likely to engage in motivated reasoning not only because their attitudes reflect cumulative exposure over time to information, but also because they increasingly resist new information that might change those attitudes.”

Building on this logic, we expect the level of prior contestation of the IO targeted in a message to condition the effectiveness of elite communication. When people have stronger priors about an IO, they will be less susceptible to new information about this organization (see also Guisinger and Saunders 2017; Spilker et al. 2020). Their attitudes toward this IO will already be hardened from earlier discussion, and they will be reluctant to change their opinions in response to communication from elites. However, if citizens have less developed attitudes toward an IO, they will more open to new information about this organization and therefore more easily affected by elite cues and frames.

In Chapters 4 and 6, we draw on this logic to interpret evidence that citizens respond less strongly to elite cues about IOs that have been subject to more intense public debate.

Political Awareness

Political awareness refers to the extent to which people have a developed understanding of politics, as a result of formal education, media consumption, own experiences, conversations with friends, and other way of acquiring information about politics. Political awareness has been described as “a relatively long-term and stable characteristics of individuals pertaining to the degree to which citizens pay attention to, understand, and know about the political world” (Claassen and Highton 2009, 539). Related concepts, often used as synonyms, are political sophistication (Sniderman et al. 1991), cognitive engagement (Zaller 1992), and political knowledge (Delli Carpini and Keeter 1996). In the context of global governance, politically more aware citizens have the training and resources to comprehend complex processes in world politics, consume news reporting on global events, and may even have international experiences themselves, making them more likely to know of IOs and their basic functioning (Dellmuth 2016).

We expect political awareness to have a positive conditioning effect on the influence of elite communication: The more politically aware citizens are, the more responsive they will be to elite cues about IOs. This may seem like a puzzling expectation, since a lack of information, after all, is the reason why individuals make use of heuristics in the first place. Yet integrating new information about IOs or other phenomena requires a level of political sophistication that allows the individual to sort and systematize this information. Ironically, then, those who are most well versed in politics may be those most easily affected by elites’ communication about politics, while those who need to build a better knowledge base about politics may be those most likely to resist new information.

Sniderman et al. (1991, 20) refer to this expectation as the “sophistication interaction hypothesis” – while all people are expected to take advantage of heuristics in politics, the politically more sophisticated are more likely to be guided by elite endorsements when forming opinions. Zaller (1992, 42), similarly, expects a positive conditioning effect of political awareness, stating that “the greater a person’s level of cognitive engagement with an issue, the more likely he or she is to be exposed to and comprehend – in a word, to receive – political messages concerning the issue.” Druckman and Nelson (2003, 732) advance the same argument with regard to framing effects, submitting that “elite frames will exhibit a greater impact on more knowledgeable individuals,” since those individuals are more able to connect the considerations put forward in a frame with their overall opinions.

Building on this general expectation, we anticipate that political awareness can help to account for individual-level variation in the effectiveness of elite communication about IOs. We distinguish between people’s general political awareness and their political awareness regarding global governance, expecting both to affect elite communication in the same way. When people are more politically aware, they will be more receptive to elite messages about the institutional qualities of IOs. Conversely, when they are less politically aware, they will have a harder time making sense of novel information about the workings and consequences of IOs.

We examine this expectation in all empirical chapters.

Political Beliefs

Political beliefs refer to individual opinions about a situation or object, which typically form clusters with related beliefs that together constitute more enduring attitudes (Rokeach 1968). Political beliefs tend to be influenced by political ideologies (such as liberalism or socialism), be reflected in relatively stable political attitudes (such as placement on the left or the right), and give rise to varying degrees of party identification (such as attachment to the Democratic or Republican Party). But political beliefs may also pertain to opinions on specific political topics, such as animal welfare, LGBTQ rights, immigration, and climate change. People’s prior beliefs about politics is perhaps the most well-established moderator of communication effects in earlier research (Chong and Druckman 2007a; Leeper and Slothuus 2014).

Political beliefs are commonly ordered according to specific conflict dimensions. The classic left–right dimension sorts political beliefs in terms of whether people support a more egalitarian distribution of income and more government intervention in the economy or consider inequality a natural social outcome and support more laissez-faire economic principles (Downs 1957; Bobbio 1996). This dimension effectively captures people’s opinions on matters of domestic politics (Schmitt and Thomassen 1999; Mair 2007) and, some argue, international politics (Hooghe et al. 2002; Nöel and Thérien 2008). Over time, the left–right dimension has been supplemented by another axis, often referred to as GAL–TAN, since it distinguishes between green, alternative, and liberal values, on the one hand, and traditional, authoritarian, and nationalist values, on the other hand (Hooghe et al. 2002; Kriesi et al. 2006; Hooghe and Marks 2018). This dimension captures attitudes on social and cultural issues that fit poorly on the left–right dimension, but have risen in prominence in recent years, such as immigration, gender equality, gay rights, ecology, national sovereignty, and international cooperation.

We expect people’s political beliefs to matter in two ways. First, they likely affect the extent to which people are receptive to messages from certain political elites. Specifically, people are more likely to listen to politically likeminded elites. The extensive literature on party cueing shows that such messages are most effective with citizens who have developed a deep and emotional attachment to a party – partisan identification (e.g., Sniderman 2000; Druckman et al. 2013; Leeper and Slothuus 2014). When they hold such partisan loyalties, “citizens can take advantage of parties’ endorsements of policies and candidates to form preferences without having to pay attention to substantive content” (Leeper and Slothuus 2014, 135).

We test this expectation in Chapters 4, 5, and 7. In Chapter 4, we assess if people are more likely to follow government cues if they identify with a party in office. In Chapter 5, we examine if citizens’ political ideology and partisan identification affect the influence of party cues on legitimacy beliefs. Likewise, in Chapter 7, we explore if citizens’ partisan identification shapes how they respond to information about the social purposes of IOs.

In addition, people’s political beliefs likely affect their receptiveness to elite communication with certain ideological content. The information that elites convey about IOs is rarely apolitical and frequently touches an ideological nerve in citizens. For instance, whether IOs possess more or less authority likely activates citizens’ political beliefs along the GAL–TAN dimension. Similarly, whether IOs pursue some social purposes rather than others likely triggers citizens’ political beliefs along the left–right dimension. All else equal, the information contained in a message may therefore be received differently depending on a citizen’s preexisting political beliefs.

We evaluate this expectation in Chapter 7, when we explore whether citizens’ political beliefs and internationalist attitudes, respectively, moderate the impact of information about the authority and purpose of IOs. In addition, we draw on this logic in Chapter 5, when interpreting evidence that communication effects vary with internationalist attitudes.

Conclusion

This chapter has outlined the theoretical framework of the book in four steps. First, we conceptualized legitimacy in sociological terms and explained how this approach directs our attention to the beliefs of citizens. Second, we introduced the two principal components of elite communication as this process unfolds in global governance: the elites at global and domestic levels, and the messages about institutional qualities of IOs. Third, we laid out the central assumptions of our theory, building on heuristic opinion formation, and explained why those translate into a general expectation of elite communication affecting IO legitimacy beliefs. And fourth, we identified conditions at the level of elites, messages, and citizens that likely shape the strength of such communication effects.

In the following four chapters, we proceed to test our general expectation that elite communication matters for citizens’ legitimacy beliefs toward IOs, as well as our specific expectations about heterogeneous effects across elites, messages, and citizens. Our analyses are structured in two parts. We begin by focusing specifically on elites, while concentrating less on the information they convey, in order to examine the effects of global and domestic elites on citizen legitimacy beliefs (Chapters 4 and 5). We then reverse the design, focusing on the information conveyed about IOs, while bracketing the elites themselves (Chapters 6 and 7).