The Social Origins of Silent Majorities

Michael Zürn

THE SOCIAL ORIGINS OF SILENT MAJORITIES

Paper prepared for the Workshop “Beyond Representative Democracy? Design and Legitimacy of Non- Majoritarian Institutions”, Berlin, 5–7 October 2017 “And so tonight – to you, the great silent majority of my fellow Americans – I ask for your support.”

(from President Richard Nixon’s Address to the Nation on the War on Vietnam on 3rd November 1969)

A spectre is haunting Europe – the spectre of authoritarian populism. 1 When Karl Marx coined this phrase referring to communism, he believed that the spectre will eliminate the old order. Indeed, many contemporary authoritarian populists target some of the existential pre-requisites of liberal democracy. They are illiberal and anti-pluralist.

The ideology behind authoritarian populism is illiberal in the sense that it questions individual and especially minority rights. It emphasizes the “rights” of the majority culture and often views individual and especially minority rights as instruments depriving the people of their control over themselves. Authoritarian populism also is anti-pluralist in the sense that it usually contains a de-proceduralized notion of majority. There is most often a reference to the silent majority, to those who – according to Richard Nixon – did not join in the large demonstrations against the Vietnam War at the time, who did not join in the counterculture, and who did not participate in public discourse. The will of the people, of the silent majority, is known in advance. There is no need for debate, no need for institutional rules, no need for checks and balances, since “HE wants, what WE want” (“ ER will, was WIR wollen ”) 2 – as an election poster for Heinz-Christian Strache, the leader of the Freedom Party of Austria ( Frei- heitliche Partei Österreichs , FPÖ) has stated. As Jan-Werner Müller (2016, 42) puts it in his essay on right-wing populism, “[p]opulism, so I argue, is a very particular political concept, according to which morally pure, and homogeneous people always stand in opposition to immoral, corrupt, and parasitic elites, although these type of elites do not actually belong to the people.” Alexander Gauland, the chairman of the German authoritarian populist party, Alternative for Germany (Alternative für Deutschland , AfD), put the message immediately after receiving the news of his election success on 24 th September 2017 in quite confused words: “We’ll get our country and our people back”.3

Indeed, when these authoritarian populists come to power, they aim at transforming the political system. They try to weaken the institutionalization of individual rights and minority rights, they disempower the third branch of government, and they try to control the media.

1 I prefer the term authoritarian populism over right-wing populism. The term posits this political orientation deliberately on an axis that is orthogonal to the old right–left axis. Indeed, many of the right-wing populist parties have – in economic terms – by now established an agenda that is protectionist and state- interventionist, which puts them rather on the leftist side of the class cleavage. Moreover, the term authoritarian populism does capture both parties like the Nationalist Front ( Front National , FN) in Western Europe as well as some authoritarian leaders in power such as Orban and Erdoğan. 2 All translations by the author, unless otherwise noted. 3 Michelle Martin, “Germany’s jubilant far-right has Merkel in its sights,” Reuters , 24 September 2017, online at , last accessed 02 October 2017. 2

At the same time, they keep a minimal level of freedom of opinion and allow for some elec- toral competition. This transformation is usually accompanied by fueling internal or external conflicts, which allows playing the national card. In these respects, Putin, Erdoğan, Orban, and Krasnodębski show remarkable similarities. While Trump has – given the resilience of American institutions – so far not succeeded, it seems that authoritarian populism in power is a fundamental danger to democracy.

Yet, as Christoph Möllers (2017, 8) has riposted in a recent commentary, “[t]he election of authoritarian figures is not per se undemocratic”. The appeal of many populists is to bring the normal people and silent majorities back into power and challenge those liberal elites who do not care about the “ordinary (white) man on the street”. Such an ambition does not sound undemocratic. Moreover, it is not only the emphasis on the majority, but very much the juxtaposition between elites and the broader population that seems to be at stake. Wolfgang Merkel (2017) likes to speak of a “representation gap”, pointing to the fact that many supporters of the authoritarian populists feel excluded from the political process. From this perspective, it appears that the populist challenge revives democracies; voter turnout increases and the political confrontation is back. So, is the contemporary spectre haunting Europe part of the democratic process, or does it endanger democracies in Europe and beyond?

In this paper, I would like to argue that the appeal of the spectre to “the majority” has indeed a substantial base. It is hard to deny that the link between cosmopolitan decision makers and the local people has weakened in the last decades. At the same, I do consider the authoritarian populists as a danger for democracy. It is therefore of utmost importance to understand the political dynamics that has brought many people to believe that the will of the majority does not count in Western democracies any longer. This paper therefore aims at providing an explanation of the social and institutional origins of silent majorities. It does so by arguing that the civilization of the class cleavage that Marx wrote about has given way for the rise of a new cleavage: one between cosmopolitans and communitarians. The latter feel dominated by the former, since cosmopolitans have established non-majoritarian institutions in order to shield and protect their policy preferences. My argument is quite in line with Barrington Moore’s (1966) in his groundbreaking book on the social origins of dictatorship and democracy, in that it is a historical institutionalist explanation demonstrating how the development of a new social cleavage has been influenced by the handling of a former one. To put it bluntly, the taming of the old cleavage between capital and labor that instigated the new cleavage – one between frequent travelers and friends of the homeland.

To develop this argument, I start out in the first section by setting the rise of right-wing populist parties in the context of cleavage theory. By referring to research carried out with colleagues at the WZB Berlin Social Center, it is argued that, by now, a new cleavage between cosmopolitans and communitarians co-structures the political landscape in Western democracies and beyond. According to cleavage theory, cleavages are triggered by social revolutions that create socio-structural divisions. In the case of the new cleavage, the under- 3 lying social revolution is globalization, and one therefore can expect that the winners of globalization are pitted against the losers. In section two, this socio-structural explanation however is complemented by a historical-institutionalist explanation, which is needed to understand the substantial positions taken by the losers of globalization. A purely social– structural perspective would suggest that the losers of globalization join leftist movements to fight globalization and neoliberalism. While the message obviously went out, it arrived once more at the wrong address (Ernest Gellner): at the address of those who emphasize silent majorities. In aiming to explain this, I provide in section three a thick description how the big deal between capital and labor got institutionalized in all Western democracies after the World Wars, and how this has nurtured the rise of catch-all parties. These catch-all parties (in German, tellingly, “people’s parties”) ironically got out of touch with the people. As a result, the trust in the core majoritarian institutions of democracies got lost and non- majoritarian institutions gained power. Non-majoritarian institutions, however, display a cosmopolitan bias. Accordingly, the new cleavage triggered by globalization was translated in most Western democracies into one between cosmopolitans, emphasizing the importance of non-majoritarian institutions, and nationalist communitarians, emphasizing the need to protect popular sovereignty and collective identities. In section three, I want to formalize this thick description and provide a testable causal mechanism consisting of a sequence of institutional developments that were triggered by the big deal and has led to the rise of populist parties emphasizing the will of silent majorities. Finally, I want to provide some pre- liminary evidence along three steps of the causal sequence. I conclude with suggestions for how to put the explanation to further tests, followed by the discussion of some political implications of it.

I The New Cleavage Seymour M. Lipset and Stein Rokkan (1967) have seminally described the historical sequence and development of four classic cleavages in Western Europe and how these developments are reflected in Western European party systems (see also Flora 2000). Peter Mair (2006) identified three dimensions of the cleavage concept: a structural element that manifests itself primarily in interests derived from the position in social structure; an ideational element, that is, a fairly coherent set of values and convictions that provide for a narrative; and an organizational element, that is, institutions and organizations such as parties, interest groups, and states that mobilize the cleavage.

Against this background, there are good reasons to see the current juxtaposition of two groups in Western societies as a cleavage between cosmopolitans and communitarians produced by the social revolution of globalization (Zürn and De Wilde 2016). In ideational terms, cosmopolitanism stands for a political ideology that advocates open borders and a transfer of public authority to the global level to combat, for instance, climate change or uncon- trolled financial markets effectively. In this view, individual and minority rights are seen as prior to the decisions taken by local majorities. Communitarians, on the other hand, emphasize the constitutive role of communities and identities for the development of social atti- 4 tudes. In their view, both distributional justice and democracy depend on social context that most often are territorially delimited. Given these background beliefs, the political positions differ from cosmopolitan demands. Borders are seen as meaningful and even constitutive for justice. They are therefore given some normative dignity. At the same time, any transfer of authority to supranational institutions is seen with skepticism. Communitarians emphasize the costs for democratic self-determination at lower levels and therefore are much less in favor of international institutions and regional integration processes than cosmopolitans. Finally, communitarians do reject the notion of universal values and therefore tend to sub- sume individual rights and sometimes even minority rights under the majority culture.

On this basis, a group of researchers at the WZB Berlin Social Science Center has studied to which extent such a new cleavage between cosmopolitanism and communitarianism takes hold of established and defect democracies. 4 We have chosen the terms cosmopolitanism and communitarianism – both of which have their roots in recognized and respected political–philosophical traditions – in order to emphasize that current conflicts may not be temporary ones between modern and atavistic factions in society, but a much more perma- nent cleavage inspired by two opposing political ideologies. One of the goals is to reveal and understand the normativity behind both of these positions and, where possible, to bind them back to structurally induced interests on both sides. 5

Both cosmopolitans and communitarians can come from different sides of the traditional class cleavage. “Right-wing” cosmopolitans emphasize above all open economic borders, while “left-wing” cosmopolitans urge for political re-regulation on the level beyond the nation-state including global redistribution. Left-wing communitarians usually emphasize the dangers of globalization for equality within states; right-wing communitarians and authoritarian populists highlight the dangers of globalization for national cohesion. However, on other globalization-related issues, left and right communitarians often have more in com- mon with each other than with the cosmopolitan camp. Euroscepticism, for instance, has the shape of an “inverted U-curve” with peaks on the two ends of the traditional one- dimensional political spectrum (Hooghe et al. 2002).

Among others, we asked two questions in this project: Do the labels “cosmopolitanism” and “communitarianism” grasp the contentious issues between the two camps? Which economic, social, and political actors are found on which side of these conflicts? We answer these research questions by analyzing political conflict formations in five countries – the United States (US), Germany, Poland, Turkey, and Mexico – and two international organizations (IOs) – the European Union (EU) and the United Nations (UN). Previous research on the political impacts of globalization had a strong focus on Western Europe, with the work of

4 The Bridging Project “The Political Sociology of Cosmopolitanism and Communitarianism” started at the WZB Berlin Social Science Center in 2011. It is based on a collaboration of three research units, involving Pieter de Wilde, Ruud Koopmans, Wolfgang Merkel, Oliver Strijbs, Céline Teney, Bernhard Wessels and myself, among others. See de Wilde et al. (2018). 5 Introduction to the project found in Koopmans and Zürn (2018, in preparation). 5

Hanspeter Kriesi, Edgar Grande, and their colleagues as the most influential example (Kriesi et al. 2008; Kriesi et al. 2012). However, if globalization is really the driving force behind these shifts in the political landscape, we should also observe its transformative effects outside of Western Europe, and we should inquire to what extent other regions show similar or different patterns.

In order to study our questions, it was necessary to focus on issue areas that are dominated by issues of societal or political denationalization in the first place. First, the regulation of international trade has long stood at the forefront of globalization studies (Jackson 2006; Zangl 2006). Issues of tariffs and non-tariff trading barriers, of state procurement, of intellec- tual property, and the public regulation of products are at the core of this issue area. Migra- tion is a second major issue area dominantly understood to be a feature of globalization (e.g., Koopmans et al. 2005; Kymlicka 2010). This issue area includes questions such as rights to employment and residence permits, roads to citizenship, and asylum policy, as well as brain drain and the societal integration of immigrants and diasporas. The third issue area of our choice is climate change. Questions included are, among others, whether a country should subscribe to obligations as currently laid down in the Paris Convention or earlier in the Kyoto Protocol to limit its emissions of greenhouse gasses; to what extent combating global climate change should be voluntary; and who should pay for mitigating the effects of climate change (Biermann 2014). Fourth, we address the issue area of human rights. Here, the diffusion and global application of norms and values about the relationship among states, other powerful actors, and individuals is at stake. The build-up of a global human rights regime under the auspices of the UN is one of the cornerstones of arguments in favor of a shared global responsibility among humanity (e.g., Pogge 2002; Risse et al. 2013). Fifth and final, we look at regional integration by focusing on debates around the EU and the North American Free Trade Agreement (NAFTA). This includes questions such as whether countries should be a member of a regional organization, how much power regional institutions should have over the member states, on which policy issues these institutions should be allowed to create collectively binding decisions, and how regional decision-making processes should be organized – including such questions as voting weights and unanimity versus majority voting (Börzel 2005; Hooghe et al. 2016). In each of these issue areas, we look at whether actors take positions for or against open borders and for or against political authority beyond the nation state. For each of these countries, we follow a multi-method approach that is based on three components: mass population surveys, an elite survey, and political claims data drawn from print media.

Is the juxtaposition between cosmopolitan and communitarian ideas useful to better grasp societal conflicts? Our analyses indeed support the notion that a new conflict line that pits cosmopolitan positions against communitarian ones. 6 This holds for all the five countries under question. In all of them, the debate between those for and against open borders and between those for and against the transfer of public authority (co-)structures the political

6 See Merkel and Zürn (2018, in preparation) for a summary of the findings and their normative implications. 6 landscape (de Wilde 2018, in preparation). While radical and merely profit-seeking neoliber- als are part of the cosmopolitan coalition, communitarianism is most often dominated by authoritarian populists. The struggle over borders has drastically and one-sidedly reduced the rich philosophical tradition of communitarianism during the last decades within the political realm. “Community” has become largely associated with the nation and exclusive defi- nitions of citizenship within it. Authoritarian populists have increasingly identified with the nation-state in opposition to further supranational integration. In this process, they began to reimagine the nation as an ethnically and religiously homogenous body. In their view, the nation and its silent majority has to be protected against the supranational usurpations, mi- grants, and self-seeking elites. It is this one-sided and narrow interpretation of communitarianism that has become an attractive political narrative for the various losers of globalization.

In organizational terms, there is a strong liberal internationalist actor coalition in the media of our five countries that advocates open borders on all policy issues and is willing to transfer some authority beyond the nation state. The main actors of the cosmopolitan coalition include mainstream political parties, state agents in the government, judiciary, and the liberal media. However, there is also a visible left–right divide among cosmopolitans. The leftist faction favors supranational regulation and is critical about the further opening of economic borders; the rightist faction emphasizes open economic borders and resists supranational authority in the regulation of markets.

The communitarian camp is dominated by authoritarian populists. These parties have occupied much of the political space to the right of the traditional conservative parties in Western Europe. The Danish People’s Party (Dansk Folkeparti , DF), the FPÖ in Austria, the National Front ( Front National , FN) in France, and Geert Wilders’ Party for Freedom (Partij voor de Vrijheid , PVV) stand out, but they are by no means the only examples. In Switzer- land, the Swiss People’s Party ( Schweizerische Volkspartei , SVP) is integrated in a cross-party ruling coalition and, in Germany, the AfD just entered the parliament with almost 13 percent of the vote. In parts of Eastern Europe, especially in Hungary and Poland, authoritarian populists are the dominant parties of government. These parties are above all right-wing in cultural matters, such as immigration policy and anti-Islam attitudes, in addition to being against regional integration. Together with their new trend towards national protectionism on economic issues, they have become the true anti-globalization parties at least in Europe.

In Western Europe, the cosmopolitan–communitarian divide additionally cuts right through the programs, policies, memberships, and electorates of catch-all parties. It is especially the center-right (Christian Democrats) and the center-left parties (Social Democrats) that needs to handle this tension. The social-democratic catch-all parties (center-left) seem to be affected more substantially than the center-right catch-all parties because their programs traditionally have both strongly cosmopolitan (internationalist) and communitarian (“people’s home”) roots.

Which economic, social, and political actors are found on which side of these conflicts? Cosmopolitans, in socio-structural terms, have above-average levels of education, above- average incomes, and high levels of cultural and human capital; they prefer open cultures and are highly mobile both spatially and occupationally. In short, they tend to be the winners of globalization (Koopmans 2018, in preparation). Communitarians are expected to defend the political achievements of post-World War II welfare states. They can be conceptualized ideal-typically in terms of the opposite attributes to those of cosmopolitans. They tend to have below-average levels of education, below-average incomes, and limited cultural and human capital; they reject multiculturalism and are not very mobile, spatially or profession- ally, beyond the nation-state. Communitarians tend to be the losers of globalization and to have a particular economic and cultural interest in the preservation of tightly controlled nation-state borders (Strijbs and Wessels 2018, in preparation).

Moreover, the communitarian-cosmopolitan conflict is reflected in an elite–masses divide. Belonging to the elites has a significant and strong positive impact on cosmopolitan positions in all five countries and on most of the five issues (Strijbs et al. 2018, in preparation). Moreover, the two fora beyond the nation-state under question, the General Assem- bly and the European Parliament, are dominated by cosmopolitans (de Wilde et al. 2018, in preparation). The cosmopolitan–communitarian cleavage is thus, to some extent, a cleavage between elites and significant parts of the wider public. While the elites are predominantly cosmopolitan in outlook, the general public seems to be more divided. It remains true, however, that the higher the level of education, the more we can expect cosmopolitan attitudes among the wider public (Strijbs et al. 2018, in preparation).

Altogether, the hypothesis that globalization has produced a new cleavage that pits the winners of globalization with a cosmopolitan ideology against the losers with a communitarian ideology seems to hold. To be sure, the new cleavage did not replace the old divide between the Left and the Right but complemented it. It is also important to note that there are significant differences between countries depending very much on the question of whether it is immigration or emigration country, and also depending on the degree of integration in a regional integration process The cleavage is most clearly accentuated in countries that are strongly integrated in regional and global institutions, and are a target of migration.

Many of the findings seem to be in line with a socio-structural reading of cleavages, according to which the social revolution of globalization has produced losers and winners that are now pitted against each other. However, two questions remain: First, why do the oppo- nents to open borders and international institutions do not coalesce with old and new leftist parties? Why do they not aim at democratization and at strengthening the capacity of liberal democracies to change policies? Why do they appeal to silent majorities instead? If neoliberalism is the major enemy of communitarians, why vote for projects like Brexit or candi- dates like Donald Trump, both of whom are quite obviously increasing material inequalities within the societies? Why are authoritarian populists spearheading the communitarian side in the new cleavage? 8

A second question pops up when the differences between the views of the broad population and the view of the elites are analyzed more closely. For this purpose, we developed an elite survey that contains some of the questions used in the World Value Survey (Strijbs et al. 2018, in preparation). In general, elites are significantly more cosmopolitan than masses. This is true for preferences regarding the permeability of borders (immigration, trade, fighting climate change) as well as for the allocation of authority and for all five countries under question (Germany, Mexico, Poland, Turkey, and the US). The gap between masses and elites is particularly strong for all cases when it comes to fighting climate change. It is also strong for Germany, Mexico, and to a lesser extent the US in the case of immigration. The mass–elite gap is somewhat weaker for international trade and even minimal in the US- American case. With the exception of Mexico, the gap between masses and elites is even more drastic when it comes to the allocation of authority in the form of delegation of public authority to the regional or international level (see Figure 1). While the elite take uniformly pro-integration positions, the distribution of preferences on the side of the masses is much more diverse.

Gap between masses and elites: Allocation of authority in the form of delegation of public authority to the regional or international level

Figure 1: Histograms on pro supranational integration preferences (Source: Strijbs et al. 2018, in preparation)

The finding that there is little variance in mass–elite gaps across countries and that these gaps are weaker for trade than for immigration and climate changes speaks also against the explanation of cosmopolitan attitudes as motivated by structurally determined socio- economic differences. In particular, the fact that the cosmopolitan–communitarian divide is 9 weaker in the highly developed US and Germany than in less developed Poland and Turkey speaks against such a socio-structural explanation. In these terms, one would expect the strongest difference in countries that have a high wage level for labor and high levels of social protection. From a perspective of material distribution, one would rather expect a reviv- al of the class cleavage in the most developed countries and moderation with a broad majority of globalization winners in rising economies (Rogowski 1989; Milanović 2016).

Why do the losers of globalization support authoritarian populists instead of leftist parties? And why do we observe in societies outside of Europe with less developed, but currently very dynamic economies similar cleavage constellations, in spite of the fact that globalization produces mostly winners in these countries? In order to answer these questions, I want to move beyond a purely socio-structural explanation and understand the political and institutional dynamics in Western societies during the last decades. We need to complement the socio-structural explanation with one that points to the path-dependent effects of certain institutional decisions taken at historical junctures.

II An Institutionalist Explanation: Embedded Liberalism, the Big Deal, and Non- Majoritarian Institutions The core argument elaborated in the remainder of this contribution is simple. The emphasis on silent majorities by the communitarian side is directed against the cosmopolitan bias of non-majoritarian institutions. With the rise in importance of non-majoritarian institutions that govern the contested issues of the new cleavage in Western democracies and beyond, communitarians lose political influence and control. Those who are on the communitarian side in the current cleavage and who advocate closed borders, national sovereignty, and the validity of the majority culture consider themselves as a silent majority that is excluded by the cosmopolitan elites who dominate the non-majoritarian institutions. They feel that they have no say in the political deliberations and in the public discourse. They consider themselves as suppressed, or at least forgotten, by cosmopolitan experts controlling non- majoritarian institutions. It is therefore that they posit the notion of silent majority against these non-majoritarian institutions and that we observe – against the economic hypothesis – similar cleavages in societies with both established and rising economies.

This perception – in spite of being instrumentalized for anti-democratic purposes – is based on real institutional changes in Western democracies. Since the 19 th century, democratic political systems are usually considered as those in which parliaments – in connection with parties and the government – play the decisive role (see Dahl 1989). Since parliaments in democratic political systems consist of elected representatives, who in turn are organized in political parties, they are at the same time the most typical cases of majoritarian institutions. Parliaments decide via majority, and the members of the parliament are selected on the basis of elections of parties. These majoritarian institutions embody the idea of popular sovereignty. Non-majoritarian institutions like courts and central banks have always played an important role in democratic political systems as well. Non-majoritarian institutions can 10 be defined as governance entities “that (a) possess and exercise some grant of specialized public authority, separate from that of other institutions, but (b) are neither directly elected by the people, nor directly managed by elected officials” (Thatcher and Sweet 2002, 2). In democratic theory, their major task is to control and to limit the public powers so that they do not violate individual as well as minority rights and thus do not undermine the democratic process. In addition, they need to implement the norms set by the legislative. 7 One may also see international institutions as non-majoritarian institutions, since they intrude into domestic affairs based on similar, mostly technocratic justifications than central banks, for instance. 8 In this conception of democratic rule, parliaments are the norm setters and, together with the executive, stand for the foundational component of constitutionalism, while non-majoritarian institutions play the limitational role. 9 They are seen as prior to non- majoritarian institutions, which are – in this model – parasitic on majority institutions. 10

It is this relationship between majoritarian and non-majoritarian institutions that has changed significantly. In the last decades – so my argument goes – specialized authorities have risen in quantity and quality so that they nowadays do play not only a limitational, but also a foundational role – that is, a norm-setting one. As a result, non-majoritarian institutions control decision making in some crucial issue areas leading to a perception of a growing share of the people that their interests are not represented. I see the rise of non- majoritarian institutions as a function of both the decline of the class conflict and the growing complexity of politics in a globalized world. The central political decisions that triggered these developments were made after World War II when embedded liberalism (Ruggie 1983) was established. This created a political situation in which catch-all parties rose and non-majoritarian institutions controlled by cosmopolitan elites gained a significant share of power over time. In other words, the cosmopolitan bias of non-majoritarian institutions nur- tures the idea that the silent majority needs to take over again. This demand however comes in form of an illiberal and anti-pluralist democracy, understood as a combination of plebisci- tary elections and de-proceduralized authoritarian leadership. In the remainder of this section, I want to develop this explanation by pointing to a casual mechanism that consists of a sequence of institutional developments and connects social responses to the old class cleavage with the rise of the new cleavage.

In the beginning, there was the industrial revolution and the rise of the class cleavage . In interaction with older conflict lines in European societies, it produced the turmoil and the twenty year’ crisis that in turn led to World War II as the greatest disaster in global history. In a second step, and as response to the disaster, “embedded liberalism” was established in the years after the end of World War II. The idea of coordinating trade and currency policies

7 See the contributions in Preuß (1994), especially by Stephen Holmes and Jon Elster. See also Ackermann (2000) und Möllers (2013). 8 See, e.g., Keohane et al. (2009) for a characterization of international institutions as democracy-enhancing, which is very much in line with the normative reasoning in favor of majoritarian institutions. 9 See Krisch (2010) for an excellent discussion of the foundational and and limitational role of constitutions. 10 Habermas (1996), however, has put the non-majoritarian institutions on equal footing with the concept of co-originality of rule of law and democracy. 11 in order to reduce tariffs and to avoid devaluation races – which were seen as decisive for the depth of the crisis of 1929 (Kindleberger 1986) – led to the setting-up of the trinity of the General Agreement on Tariffs and Trade (GATT), the International Monetary Fund (IMF), and the World Bank. Harry Dexter White (for the US) and John Maynard Keynes (for the United Kingdom) were the architects of this setting. Based on the learnings from the interwar period, it needed American leadership plus an increased willingness of the European states to follow the lead in order to establish these institutions that reduced transaction costs and thus fostered international cooperation (Keohane 1984).

These institutions were rooted in a social purpose: the global extrapolation of the democratic welfare state that required a certain international environment to flourish (Katzen- stein 1985). The social purpose on which these international institutions were based is aptly expressed by the term “embedded liberalism”. Embedded liberalism is understood as a basic focus on free trade and open borders, which, however, is concretely embedded in national political systems that can cushion the shocks and inequalities triggered by the global market (see Ruggie 1983). The international institutions created on the basis of the principle of embedded liberalism facilitated relatively unlimited trade among all industrialized states, simul- taneously allowing for significant differences in the political and societal developments of the nation-states (see Hall and Soskice 2001). The concept of embedded liberalism thus refers to a specific combination of liberalized international markets and nation-state market interventions.

Against the background of these features of the international order, it was possible to in- stitutionalize the big deal between capital and labor in almost all Western democracies. Cap- ital and labor developed a historical compromise. Unions accepted open borders and the dangers for employment it implied, while the export-oriented business associations sup- ported the building up of the welfare state. This turned out to be a self-reinforcing process: The export capital gained more and more power relative to producers focusing on the domestic market only (e.g., Milner 1988), and the unions together with the social-democratic parties moved into many power positions in most Western democracies. As Peter Katzen- stein convincingly showed in his seminal work on small states in world markets (Katzenstein 1985), the big deal succeeded. Indeed, the Katzenstein law – the more a national economy was trade-dependent, the stronger the development of the welfare (see also Rieger and Leibfried 1997) – is one of the core features of the period between 1945 and 1989.

The big deal , in a third step, changed the handling of the class cleavage in Western democracies dramatically: it led to the rise of political systems dominated by catch-all parties and moderation of the most important actors. The big deal changed the party system in a significant way from the 1960s on in almost all Western states: the abandonment of ideological positions and class orientations on the side of the most important parties; the strengthening of party leadership; and very close association between ruling parties and the interest groups of capital and labor, called “corporatism” (Lehmbruch 1977).

Two towering political science figures can be referred to in order to capture these developments. Otto Kirchheimer (1965) coined the term “catch-all party” as part of an investigation into political party transformation in Britain and Germany. A catch-all party is a political party that aims to attract people with diverse political viewpoints, appealing to a large amount of the electorate. While it is driven by the decline of class alignment and the need to win elections, it can cause problems with internal discipline, as more traditionally oriented party members and parliamentary members may rebel against some of the policy directions the party takes. As a consequence, the centralization and strengthening of the party leadership accompanied the development. In this way, the rise of catch-all parties can be seen as a major cause for the decline in traditional political participation (see also Hay 2007; Mair 2013).

Robert Dahl pointed early on to the consequences of this development for the meaning of democracy:

The politics of this new democratic Leviathan are above all the politics of compromise, adjustment, negotiation, bargaining; a politics carried on among professional and quasi- professional leaders who constitute only a small part of the total citizen body; a politics that reflects a commitment to the virtues of pragmatism, moderation and incremental change; a politics that is un-ideological and even anti-ideological … This new Leviathan [is seen by many citizens] as too remote and bureaucratized, too addicted to bargaining and compromise, [and] too much an instrument of political elites and technicians. (Dahl 1965, 21–22)

As a consequence, party leaders and parties as a whole are increasingly seen as mechanisms to gain and maintain power instead of social organizations defending a given social purpose. Thus, the alienation between parties and representatives on the one hand and people on the other was set in train. In this view, it is the taming of the old cleavage between capital and labor that instigated the new cleavage.

With the benefit of hindsight, it can be argued that, as the fourth step, growing distrust in majoritarian institutions can hardly surprise. While the democratic principle increasingly became accepted worldwide as a key principle of desirable political order, parties and parliaments lost trust in Western democracies. In all countries covered by the World Values Survey, the majority of respondents are in favor of democracy. No government in the world fails, at least rhetorically, to depict itself as democratically legitimated. At the same time, people seem to dislike most of the political institutions that are majoritarian and thus inher- ently connected to Western democracies: parties and parliaments. In recent decades, comparative research on democracy has diagnosed a decline in political participation, evidenced in the average Organisation for Economic Co-operation and Development (OECD) country by a steady fall in voter turnout and a marked decrease in party membership since the 1960s (Hay 2007, 21). This is only critical if developments are accompanied by low confidence in party politicians and parliaments. In fact, among the multitude of public institutions, political parties in the US and the EU consistently scored worst on confidence, worse even than big 13 business and the media. Parliaments, too, scored badly, generally ranking fourth-last of twelve public institutions.

This rejection of majoritarian institutions is accompanied by a high level of trust in constitutional courts, central banks, and other non-majoritarian institutions. They do have a very high reputation in most countries. In Germany, for example, the constitutional court out- scores consistently all other national political institutions. More generally, in all 22 countries covered by the 2008 European Social Survey, people had greater confidence in the legal system than in parliament and parties (Norwegian Social Science Data Services 2011). Interna- tional institutions are also grounded in a remarkable degree of recognition. In the consoli- dated democracies of the West, the UN enjoys greater political trust than national parties and parliaments (Hay 2007, 34).

In sum, dissatisfaction with the core institutions of parliamentary democracy such as parliaments and parties contrasts with the considerable esteem enjoyed by non-majoritarian institutions. Thus while the core institutions for majority formation, which define the realm of politics in the narrower sense, are seen with an extremely critical eye, institutions em- powered to make decisions affecting society as a whole that escape the political process – such as central banks, constitutional courts, or international institutions – have a much better standing everywhere than do the core democratic institutions.

The decline of trust in majoritarian institutions is accompanied by a growing relevance of non-majoritarian institutions in the political process , as the fifth step. Increasingly, non- majoritarian institutions do not only double-check and implement decisions of majoritarian institutions, but are also involved in policy making and norm setting.

In the domestic realm, independent central banks in the Western world have become more important. They have gradually been introduced in many countries, and their independence has been strengthened. At the same time, the importance of monetary policy tools in the general economic control toolbox has generally increased with the spread of monetarism. Central banks thus became more autonomous and more important at the same time (Rapaport et al. 2009). At least as important is the increase in the importance of constitutional courts (Hirschl 2004). In general, the increase in so-called “independent agencies” (Shapiro 1997; Shapiro 2005; Gilardi 2008) has been observed widely. On average, according to a quantitative study, “autonomous regulatory agencies” play by now a role in 73 percent of all policy areas in the countries under investigation (Jordana et al. 2011).

In addition to these developments within national political systems, a dense network of international arrangements and organizations has developed that differ in both quality and quantity from traditional international institutions (cf. Zürn 2004). The EU is only the best- known example in this respect; yet it is no longer an exception in the international institutional landscape, but part of a general trend, the effects of which are similar to the ones described in the domestic realm. The new international arrangements exercise authority over

14 their constituent members and, at the same time, intervene profoundly in the internal affairs of countries. This means that the consensus principle of international politics and hence the logic of national sovereignty is undermined. The principle that a given territory is governed exclusively by the national government does not hold any longer.11 In the last decades, especially since 1990, IOs have gained significantly in authority (see Hooghe et al. 2017; Zürn et al. 2015).

The rise of international authority also transforms the role of government along the majoritarian–non-majoritarian axis. Since governments are elected either by parliaments or directly in a presidential election, traditionally they have been seen as a case of majoritarian institutions. To the extent, however, that the more powerful Western governments are controlling international authorities, they can use international institutions to circumvent parliaments and party members. In fact, the rise of multi-level governance systems including all the new space created for blame shifting and credit claiming detaches the executive from the legislative and makes government to a significant player in the world of non-majoritarian institutions. The rise of international institutions empowers governments and weakens parliaments (Zürn 1996).

Overall, some domestic institutions within democracies, as well as international institutions, became more powerful relative to the parliaments and parties in the last three decades. These non-majoritarian institutions do not only implement and control policies – as foreseen by the notion of democratic constitutionalism – they are also strongly involved in setting norms and rules. Most of these institutions represent a similar type of public authority. Their power is limited and often refers to only one issue area (e.g., central banks or World Trade Organization) or only a certain function within a constitutionalized system of rule (e.g., constitutional court; see Zürn 2017).

The cosmopolitan bias of non-majoritarian institutions is the final and decisive step in the causal story. Non-majoritarian institutions are based on expertise and the capacity to make arguments referring to a complex and globalized world. This in effect leads to an exclusion of people with less education and less transnational contacts. Yet people with less education and less transnational contacts are typically associated with communitarian positions as has been shown in the above mentioned project on the cosmopolitan–communitarian divide. This translates into a cosmopolitan bias of non-majoritarian and international institutions (Zürn 2014). On the basis of claims data in the project on the cosmopolitan and communitarian cleavage, Pieter de Wilde et al. (2018, in preparation) have shown with respect to international institutions that the higher the distance of political actors to majoritarian institutions, the more accentuated the tendency to claim cosmopolitan positions. Claims in the General Assembly are more cosmopolitan than the ones on European level; within Europe,

11 Of course, as Stephen Krasner (1999) has powerfully pointed out that the principle of sovereignty was disre- garded by powerful states all the time in modern history. Yet, the current developments point to an institutionalization of interventions into domestic affairs (see Zürn and Deitelhoff 2015) for a discussion of the development of the principle of democracy). 15 claims by members of the Commission are more cosmopolitan than those of the European Parliament; in general, claims on the European level are more cosmopolitan than those on the national level; and within national political systems, members of the executive are less communitarian than members of the parliament.

While these findings are mainly about the cosmopolitan bias of international institutions, there are good arguments to believe that non-majoritarian institutions in general take individual and minority rights as well as international concerns more into account than majoritarian institutions. This development underlies the division between masses and elites on issues like borders, authority transfer, and individual rights. It is against this background that the appeal of slogans like “ die da oben, wir da unten ” (“they on the top, we on the bottom”) is very strong. The appeal to the silent majority becomes very attractive, since those who feel that they belong to this group indeed feel excluded and deprived from their voice. At the same time, already the claim to be in the majority comes across anti-pluralistic since the mass is divided on these issues. The policies of the silent majority in power, moreover, becomes necessarily illiberal, since it needs to target those non-majoritarian institutions such as higher courts, which are also there to defend liberal rights against the temper of the majoritarian will. While globalization has triggered the new cleavage, it is the specific political taming of the old cleavage that has led to dynamics that lead to the idea of silent majorities as a central component of the real-world communitarian narrative. In this sense, the oppor- tunity to abuse the notion of silent majorities on the side of authoritarian populists has been endogenously produced by the big deal that domesticated the class cleavage. In effect, the notion of silent majority is at the core of the new cleavage.

III The Causal Mechanism and Propositions The model of institutional dynamics proposed here suggests a causal mechanism leading from the handling of the old cleavage to the rise of the silent majority narrative as a constitutive part of the new cleavage. Rather than immediate bivariate relationships, mechanisms refer to “recurrent processes linking specified initial conditions to a specific outcome” (Mayntz 2004, 241). My understanding of such causal mechanisms includes both recurrent sequences of events and institutional relationships (composite mechanisms) that put relationships between social facts in a broader context, and micro-processes that describe the social choices leading from one macro phenomenon to the next (linking mechanisms) (see Bennett and Checkel 2015; Bunge 1997; Elster 1989). The silent-majority mechanism is in line with this reasoning as summarized in Figure 2. It consists of the six components (t 1-t6) discussed in the preceding section and five links (arrows a–e). The basic claim behind this causal mechanism is that the taming of the old cleavage between capital and labor instigated the new one between frequent travelers and friends of the homeland.

The posited causal mechanism builds on the idea of reactive sequences (Mahoney 2000; Hanrieder and Zürn 2017). Accordingly, decisions taken at critical junctures produce path dependent processes. Yet, path dependency does not consist only of self-reinforcing mechanism as one-sidedly emphasized by most historical institutionalists, but also of self- undermining ones. At a certain point, the losers of self-reinforcing dynamics organize resistance against the institutional set-up that leads to contestation, conflicts, and resistance. The communitarian challenge to the cosmopolitan order represents such a reactive sequence. It is the domestic analogy to the reactive sequence caused by the growing challeng- es to the international order (Zürn 2018, in preparation).

Figure 2: The Silent Majority Mechanism

For the purposes of this paper, I take the first three steps in the model as given and blind out all the micro-mechanisms. In order to present a first plausibility probe in support of this causal mechanism, I focus only on the last three steps in the sequence, which are certainly much less established and much more contested than the first three. In other words, I take as given and empirically established that the postwar international order made the big deal between capital and labor possible, and that the domestication of the class cleavage led to catch-all parties and the politics of compromise and adjustment. Therefore, I focus on the last three steps of the silent majority mechanism and provide some data in support of the three propositions that underlie the final three steps in the causal sequence.

P 1: After the rise of catch-all parties, the 1960s and 1970s saw a decline of trust in majoritarian institution in Western democracies. P 2: The 1980s and 1990s saw a parallel rise of different non-majoritarian institutions. P 3: By now, non-majoritarian institutions are controlled by actors who lean towards cosmopolitanism.

IV A First Plausibility Probe Proposition 1 requires descriptive data on the trust in majoritarian institutions like the parliament and parties. In general, a look at the World Value Survey as carried out in OECD countries in 2006 points to something that may be labelled a “democratic paradox”. In all of the included 16 countries, over 85 percent of the population wants to be governed by a “democratic political system” (WVS 2006). A closer look, however, reveals that they have little confidence in parliaments and parties. In 8 out of 16 countries, respondents rank each of the three non-majoritarian institutions higher than parties, parliaments and governments (***). In three additional countries, parties and parliaments have the lowest value of all institutions with governments moving partially into the ranks of non-majoritarian institutions (**). In three more cases, at least the average of confidence into the non-majoritarian institutions is higher than the average of confidence in majoritarian institutions (*). Only Finland and Turkey deviate – to some extent – from this pattern, both still with very low values for parties.

Percent of respondents who have a great deal or quite a lot confidence in the courts, the national government, political parties, national parliament, the EU and the UN

Justice National Political National EU UN government parties parliament Italy*** 51.6 26.4 16.5 33.1 67.2 59.0 Spain*** 55.6 44.9 28.5 50.8 61.3 59.8 USA** 58.2 38.6 15.4 20.3 33.4 Canada*** 65.6 38.4 23.1 36.5 61.4 Japan*** 82.0 31.0 18.3 23.2 64.1 Mexico** 37.7 44.7 23.9 25.5 53.0 Australia*** 53.8 40.1 14.3 35.4 45.5 Norway* 86.0 54.1 28.6 62.3 44.3 85.2 Sweden* 74.2 42.3 33.4 54.9 36.6 78.2 Finland 81.8 64.5 29.1 56.2 36.8 63.7 Poland*** 32.9 18.2 07.1 12.5 46.2 49.5 Switzerland* 76.8 69.3 27.6 57.4 43.5 52.8 Chile 30.0 48.2 19.1 26.2 45.4 Slovenia*** 32.9 24.0 09.0 16.4 35.9 32.6 Turkey 75.0 62.7 33.0 60.0 30.2 30.4 Germany*** 56.4 23.9 13.0 22.0 30.6 37.8

Table 1: Confidence in Majoritarian and Non-Majoritarian Institutions (Source: World Value Survey, Wave 5; data collected between 2005 and 2006)

Did the decline of trust into majoritarian institutions indeed start from the 1960s on as the causal mechanism suggests? Unfortunately, most of the measures that compare different institutions in different countries lack the historical depth needed to answer this question. I use therefore two sources to probe the expectation. Colin Hay (2007) has gathered and analyzed series of national survey on the issue. Data from three countries asking a similar question, “Do politicians care?”, shows a clear decline. Positive answers declined from the late 1950s on. By the mid-1970s, the share of negative answers passed the 50 percent threshold in the three countries. While this question refers to “all politicians”, there are good reasons to believe that most respondents thought about party politicians and parliamentary members when they gave their answer.

Do Politicians care? Comparative attitudinal trends, France, Sweden and the US

Figure 3: Trust in Politicians (Source: Hay 2007: 30; compiled from National Election Studies – various years)

The World Value Survey data referred to in Table 1 does not go back until the 1960s, yet for some countries, data for all the waves from the early 1980s on is available. In order to elabo- rate Proposition 1, I focus on the question about confidence in parliament and select those countries that over the whole time period were democratic OECD countries and were part of the survey for all the six waves. With the US and Mexico, at least two of these countries are also part of data gathered for the cleavage project. For reasons of simplicity, I compound the values of “A Great Deal” and “Quite a Lot”, which is in the survey contrasted with “not very much” and “not at all”, as well as “don’t know” (usually clearly below 10 percent).

Confidence in p arliament: A Great Deal + Quite a Lot

Wave 1981 –1984 1990 –1994 1995 –1998 1999 –2004 2005 –2009 2010 –2014 Countries Japan 27.8% 28.3% 24.2% 19.7% 21.4% 19.8% Mexico 27.6 % 34.3% 43.2% 20.8% 25.0% 25.0% South Korea 43.1% 33.9% 31.0% 10.2% 26.2% 25.5% US 42.1% - 28.2% 37.1% 19.5% 20.2%

Table 2: Confidence in Parliaments (Source: Compiled from World Value Surveys)

The overall trend is in line with the expectation. Given the relatively low levels of support already in the early 1980s, it seems plausible to assume that the decline in confidence started earlier. While we see some outliers (especially in the 1999–2004 wave), probably due to specific political developments within these countries, the values for all five countries in the most recent two waves are clearly below the values for the first two waves. For Poland, a country that became democratic only in the early 1990s, the values went down from 55.2 percent in wave two, via 30.1 percent in wave 3, to about 11 percent in waves 5 and 6. It seems that this new democracy has replicated the path in only twenty years, for which established democracies took four decades.

Proposition 2 requires data about the rise of non-majoritarian institutions. Jordana et al. (2011) have documented the rise of executive agencies used especially in the US to circumvent the congress. The authors point to a movement that mainly took place in the 1980s and 1990s. While this data is already impressive, it might be seen as too broad and encompass- ing. Many of these agencies may not move beyond a limitational function. Instead, I want to focus on those non-majoritarian institutions that clearly play a rule-setting role from time to time: supreme courts, central banks, and international institutions that can act without the consent of the member states.

According to Ran Hirschl (2004), more than 80 legislatures in the world have decided to give supreme courts more rights and competences. Most of these changes took place in the 1980s and 1990s, as well. Ran Hirschl (2004, 1) sums up this development in the opening sentence of his critical study Towards Juristocracy : “Over the past few years the world has witnessed an astonishingly rapid transition to what may be called juristocracy . Around the globe, in more than eighty countries and in several transnational entities constitutional re- form has transferred an unprecedented amount of power from representative institutions to judiciaries.” He bolsters this finding by providing four case studies about Canada, New Zea- land, Israel, and South Africa. In all of these cases, he observes a judicial empowerment and a shift away from traditional principles of parliamentary sovereignty. In all the cases, the development speeded up especially in the 1990s.

Moreover, according to Rapaport et al. (2009) , no fewer than 84 countries passed legisla- tion to enhance the formal autonomy of central banks between 1990 and 2008 . Cukierman et al. (1992) has provided a somewhat more complex picture by his sophisticated measure- ments of legal independence of Central Banks since World War II. While this data shows a first – although much less systematic – rise of legal independence of central banks in some countries from the 1970s on, the speed increases in the 1990s. Data provided by Jácome and Vázquez (2008) build on the Cukierman index for central bank independence and show a remarkable growth especially in areas without a tradition of central bank independence (see also Cukierman 2008). The data for Latin America is most telling in this respect.

The Average Aggregate Index of Legal Independence of Central Banks in Latin American Countries

Figure 4: Central Bank Independence (Source: Cukierman 2008)

For measuring the authority of IOs, I use the data of the International Authority Database group at the WZB Berlin Social Science Center. This database uses the following operational definition of international authority: International and/or transnational institutions have authority when the direct and indirect addressees recognize, in principal or in practice, that an institution can make competent judgments and decisions that come with the ambition to be binding at least for some members of the global governance system (see Zürn et al. 2012, 70). The data is retrieved from a careful analysis of all the treaties, their amendments, and procedural protocols, starting with 1948. Competence is measured as the product of autonomy of the institution times the scope, whereby scope refers to the number of issues regu- lated by the IO and autonomy points to the sum of delegation and pooling. Recognition is measured in terms of the level of obligation and bindingness for the members.

We collected data on the political and epistemic authority of a sample of 34 IOs (and more than 230 IO bodies) across seven policy functions derived from the policy cycle. 12 These are (1) agenda setting, (2) rule making, (3) monitoring, (4) norm interpretation, (5) enforcement of decisions, (6) evaluation, and (7) knowledge generation. 13 These functions focus on different activities of IOs reflecting the notion that authority is exercised when making decisions or interpretations. We used founding treaties and all the changes made to them over time to assess formal authority. In doing so, we have developed a coding instrument with which we collect – for each of our seven policy functions – information according to a three-step logic. First, does the IO have the right to authorize any of these functions? Second, if so, who carries out the function (IO body, member state, other actor, including non-state actors)? Third, how is the function carried out and how “authoritative” is it? The possible authority values range from 0 to the (theoretically possible) value of 10.25. 14

Regarding the historical development of international authority, we can state that the overall growth of authority by international institutions is indeed an ongoing process in which it is possible to identify two growth phases: a growth period running from the end of World War II to 1970, and a second – an even steeper one – from 1990 until today.

The Rise of International Authority

Figure 5: IOs: Variation in authority over time (Source: Zürn et al. 2015)

12 We draw on similar selection decisions as Tallberg et al. (2013) and Hooghe et al. (2016), and arrive at a joint sample, which allows the participating scholars to pool their respective analyses and thereby produce a more complete understanding of international authority. 13 These functions draw from the public administration literature (see, e.g., Anderson 1975; Jenkins 1978; May and Wildavsky 1978). Several studies have adapted similar functions for the study of international politics and institutions, including, for instance, Bradley and Kelley (2008); Abbott and Snidal (2009); Avant et al. (2010). 14 This value is possible when applying the following formula with the theoretically possible maximum values: = Σ , where i is IO, j policy function, and t year. The resulting maximum total authority score is: √5 21 10.246 . 22

These three data sources together provide remarkable support for the proposition that the political relevance of non-majoritarian institutions grew especially in the 1990s. These three measures could be combined in a next step into one global index of non-majoritarian institutions with norm-setting character. With all likelihood, one would see two things that are both fully in line with the causal mechanism as explicated above: First, there is an enormous growth in non-majoritarian institutions, and second, this growth mainly took place in the 1980s and 1990s.

Proposition 3 finally contains the bivariate statement that non-majoritarian institutions hold positions that are leaning more to the cosmopolitan pole than majoritarian positions. In order to establish this hypothesis, the claims-making data – as described in section one – is used. This data offers the possibility to investigate the interrelatedness of issue positions directly on the actor level. We especially consider two variables used to characterize actors. First, actors were distinguished according to the function they play in the different political systems: the executive (e.g., the UN Secretary General, national governments, the EU Coun- cil of Ministers); the legislative (e.g., the European Parliament, national parliaments, city councils); the judiciary (e.g., the European Court of Human Rights, the German Constitution- al Court); and experts. Second, regarding the the polity level of the claim maker, we distin- guish among global (e.g., the UN High Commissioner for Refugees), regional (e.g., the Euro- pean Commission), or national (e.g., the German Chancellor). To investigate actors’ positions on the three dimensions of denationalization, “issue position” is taken as the dependent variable. The issue position denotes an actor’s general position towards open borders (“integration”) or closed borders (“demarcation”), as well as demands for further transfer of authority (“integrate”) or demands for political nationalization (“demarcate”). Issue position then can take four values: “integrate” (1), “keep integrated” (2), “demarcate” (3), and “keep demarcated” (4) (see De Wilde, Koopmans, and Zürn 2014, 34–35 for the codebook).

The results are clear-cut. Regarding the function of actors, legislatives – as the archetype of a majoritarian actor – are close to the middle position with an average of around 2.1. To the contrary, all the claims made by representatives of non-majoritarian institutions much more clearly lean towards the cosmopolitan pole. In line with the theoretical reasoning, the positions for governments are closer to non-majoritarian actors than to the legislatures. More concretely, representatives of IOs are most cosmopolitan with an average value of 1.5 (note that 1.0 is the cosmopolitan pole), followed by executives, experts, and the judiciary, all of them with an average value between 1.6 and 1.7. All the differences to the legislative baseline are significant (see Figure 6).

Means of Political Positions of Actors

Figure 6: Means of Political Positions of Actors. Legend: 0=Legislative; 1=Judiciary; 2=Experts; 3=Government/Executive; 4= IOs. Political Position ranges from 1 (Cosmopolitan) to 4 (Communitarian). 95% Confidence Intervals included.

Moving to the level of the institutions for which actors make their claims, the result again is clearly in line with the expectation. While national actors (including the executive and all parties) lean already towards cosmopolitanism with an average of slightly above 1.7, the cosmopolitan leaning of IO representatives and also of representatives of regional institutions is more accentuated. Again, actors on the regional and international levels (which are dominantly non-majoritarian) lean significantly more towards the cosmopolitan position than national actors, which are a mix of majoritarian and non-majoritarian actors.

Means of Political Positions of Actor’s Scope

Scope of Actor

Figure 7: Means of Political Positions of Actor´s Scope. Legend: 0=National; 1=Regional; 2=Global 4. Po- litical Position ranges from 1(Cosmopolitan) to 4(Communitarian). 95% Confidence Intervals included.

These results are corroborated when we use a logistic regression model (Table 3), which – in line with Proposition 3 – takes legislatives being most communitarian actors as baseline. Note that all the differences between the legislative – respectively national – baseline and the other actors are strongly significant, with the difference between national and regional actors as the only exception.

Cosmopolitan Positions of Institutional Actors

(1) Political Position Model 1 (1=Cosmopolitan)

Reference: Legislative

1. Judiciary 0.560** (0.268) 2. Experts 1.337*** (0.168) 3. Executive 0.913*** (0.0980) 4. All international organizations 0.861*** (0.206)

Reference: National

1. Global 1.836*** (0.383) 2. Regional -0.0437 (0.185) Constant 0.491*** (0.0867)

Observations 6,554 Standard errors in parentheses ***p<0.01, **p<0.05, *p<0.1 Table 3: Logistic Regression Model

All the propositions derived from the causal mechanism receive very strong empirical support. Moreover, the timing is clearly in line with the sequence spelled out in the silent majority mechanism. The decline in confidence in majoritarian institutions started in the 1960s after the big deal got institutionalized in Western democracies and catch-all parties took over. The rise of power of majoritarian institutions is steepest in the 1980s and 1990s. The development of a new cleavage with one side using the notion of silent majorities as a weapon against the cosmopolitan dominance in these non-majoritarian institutions is a more recent phenomenon.

V Conclusion The rise of anti-pluralist notion of silent majorities serving authoritarian populists can be seen as part of a reactive sequence that has developed in response to the class cleavage. It is the taming of the old class cleavage that has produced the new cleavage. It therefore needs

26 a historical-institutionalist complement to the socio-structural reasoning of cleavage theory. The emphasis on silent majorities by the communitarian side is directed against the cosmopolitan bias of non-majoritarian institutions and cannot be derived from the structurally pre- determined juxtaposition of winners and losers of globalization. Those who are on the communitarian side in the current cleavage and who advocate closed borders, national sovereignty, and the validity of the majority culture consider themselves as a silent majority that is excluded by the cosmopolitan elites. They feel that they have no say in the political deliberations and in the public discourse. They consider themselves as suppressed or at least forgotten by cosmopolitan experts controlling non-majoritarian institutions. It is therefore that they posit the notion of silent majority against these non-majoritarian institutions and that we observe similar cleavages in societies with both established and rising economies.

The cosmopolitan side of the new cleavage needs to overcome this sense of exclusion. Either the political institutions that run the cosmopolitan order need to be reformed, or they run the danger to perish in the turmoil produced by the new cleavage. The dangers are enormous. Authoritarian populists are anti-liberal as well as anti-pluralist thus jeopardizing open political systems as a whole. Moreover, they are nationalists, thus compromising the liberal economic order with potentially devastating costs for global welfare as a whole. Au- thoritarian populists might bring us back to the days of the twenty years’ crisis after World War I.

What is needed? Non-majoritarian institutions need to open up and re-introduce majoritarian procedures. Central banks must be moved back to the tasks they performed in the 1980s. Supreme courts may add participatory procedures to the legal reasoning. Most importantly, international institutions that exercise public authority need to be opened up for societal participation and must introduce sites for public and open-ended debates about the right policies. This is most urgent for the EU, but applies to other international institutions as well.

No doubt, there is also a need to move the testing of the correlates put forward in this paper further. There are three steps which are most important. It is first of all necessary to establish the right sequence with better data than available and along all the six steps of the causal mechanism. Second, and more importantly, it is necessary to establish causal links between the different steps in the causal sequence. For this purpose, the model needs to be complemented with micro-mechanisms that link the different steps and that allow for testing causal claims in a stricter sense. At the end of the day, we also need a test of the whole model. One way could be to test the proposition derived from the model that authoritarian populists emphasizing silent majorities are strongest in those political systems in which moderating catch-all parties were most dominant – that is, in election systems with the win- ner-takes-it-all principle and in corporatist system of interest mediation.

REFERENCES

Abbott, Kenneth W., and Duncan Snidal. 2009. “The Governance Triangle: Regulatory Stand- ards Institutions and the Shadow of the State.” [en]. In The Politics of Global Regulation [en], eds. Walter Mattli and Ngaire Woods. Princeton: Princeton University Press, 44–88. Ackerman, Bruce. 2000. “The New Separation of Powers.” [en]. Harvard Law Review 113 (3): 633. Anderson, James E. 1975. Public Policy-Making [en]. New York: Praeger. Avant, Deborah D., Martha Finnemore, and Susan K. Sell. 2010. “Who Governs the Globe?” [en]. In Who Governs the Globe? [en], eds. Deborah D. Avant, Martha Finnemore and Su- san K. Sell. New York: Cambridge University Press, 1–34. Bennett, Andrew, and Jeffrey T. Checkel, eds. 2015. Process Tracing: From Metaphor to Ana- lytic Tool [eng]. Cambridge: Cambridge University Press. Biermann, Frank. 2014. Earth System Governance: World Politics in the Anthropocene [en]. Cambridge: The MIT Press. Börzel, Tanja A. 2005. “Mind the gap! European integration between level and scope.” [en]. Journal of European Public Policy 12 (2): 217–36. Bradley, Curtis A., and Judith Kelley. 2008. “The Concept of International Delegation.” [en]. Law and Contemporary Problems 71 (1): 1–36. Bunge, Mario. 1997. “Mechanism and Explanation.” [en]. Philosophy of the Social Sciences 27 (4): 410–65. Cukierman, Alex. 2008. “Central bank independence and monetary policymaking institutions - Past, present and future.” [en]. European Journal of Political Economy 24 (4): 722–36. Cukierman, Alex, Steven Webb, and Bilin Neyapti. 1992. “Measuring the Independence of Central Banks and Its Effect on Policy Outcomes.” [en]. The World Bank Economic Review 6 (3): 353–98. Dahl, Robert A. 1965. “Reflections on Opposition in Western Democracies.” [en]. Govern- ment and Opposition 1 (1): 7–24. ———. 1989. Democracy and its Critics [en]. New Haven, CT: Yale UP. De Wilde, Pieter, Ruud Koopmans, and Michael Zürn. 2014. “The Political Sociology of Cos- mopolitanism and Communitarianism: Representative Claims Analysis.” [en]. WZB Discus- sion Paper SPIV 2014-102. Elster, Jon. 1989. Nuts and Bolts for the Social Sciences [en]. Cambridge: Cambridge Universi- ty Press. Flora, Peter. 2000. “Einführung und Interpretation.” [de]. In Staat, Nation und Demokratie in Europa. Die Theorie Stein Rokkans aus seinen gesammelten Werken rekonstruiert und ein- geleitet von Peter Flora [de], ed. Stein Rokkan. Frankfurt a.M: Suhrkamp, 14–119. Habermas, Jürgen. 1996. Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy [en]. Cambridge, MA: MIT Press.

Hall, Peter A., and David Soskice. 2001. Varieties of Capitalism. The Institutional Foundations of Comparative Advantage [en]. Oxford: Oxford University Press. Hanrieder, Tine, and Michael Zürn. 2017. “Reactive Sequences in Global Health Govern- ance.” In International Politics and Institutions in Time [en], ed. Orfeo Fioretos. Oxford, New York: Oxford University Press. Hay, Colin. 2007. Why We Hate Politics [en]. Cambridge: Polity Press. Hirschl, Ran. 2004. Towards Juristocracy: The Origins and Consequences of the New Constitu- tionalism [en]. Harvard: Harvard University Press. Hooghe, Liesbet, Gary Marks, Tobias Lenz, Jeanine Bezuijen, Besir Ceka, and Svet Derderyan. 2017. Measuring International Authority. A Postfunctionalist Theory of Governance, Vol- ume III [en]. Transformations in governance . Corby: Oxford University Press. Hooghe, Liesbet, Gary Marks, Arjan H. Schakel, Sara Niedzwiecki, Sandra C. Osterkatz, and Sarah Shair-Rosenfield. 2016. Measuring Regional Authority [en]. Vol. 1 of A postfunctionalist theory of governance . Oxford, United Kingdom: Oxford University Press. Hooghe, Liesbet, Gary Marks, and Carole J. Wilson. 2002. “Does Left/Right Structure Party Positions on European Integration?” [en]. Comparative Political Studies 35 (8): 965–89. Jackson, John H. 2006. Sovereignty, the WTO, and Changing Fundamentals of International Law [en]. Volume 18 of the Hersch Lauterpacht Memorial Lectures. Cambridge University Press. Jácome, Luis, and Francisco Vázquez. 2008. “Is there any link between legal central bank independence and inflation? Evidence from Latin America and the Caribbean.” [en]. Euro- pean Journal of Political Economy 24: 788–801. Jenkins, W. I. 1978. Policy Analysis: A Political and Organizational Perspective [en]. New York: St. Martin's Press. Jordana, Jacint, David Levi-Faur, and Xavier Fernández i Marín. 2011. “The Global Diffusion of Regulatory Agencies: Channels of Transfer and Stages of Diffusion.” [en]. Comparative Po- litical Studies: 1–27. Katzenstein, Peter J. 1985. Small States in World Markets. Industrial Policy in Europe [en]. Ithaca: Cornell University Press. Keohane, Robert O. 1984. After Hegemony. Cooperation and Discord in the World Political Economy [en]. Princeton: Princeton University Press. Keohane, Robert O., Stephen Macedo, and Andrew Moravcsik. 2009. “Democracy-Enhancing Multilateralism.” [en]. International Organization 63 (01): 1–31. Kindleberger, Charles P. 1986. The World in Depression, 1929-1939 [en]. Berkeley: University of California Press. Kirchheimer, Otto. 1965. “Der Wandel des westeuropäischen Parteiensystems.” [de]. Politische Vierteljahresschrift 6 (1): 20–41.

Koopmans, Ruud, Paul Stathan, Marco Giugni, and Florence Passy. 2005. Contested Citizen- ship: Immigration and Cultural Diversity in Europe [en]. Vol. 25 of Social movements, pro- test, and contention . Minneapolis, MN: University of Minnesota Press. Krasner, Stephen. 1999. Sovereignty: Organized Hypocrisy . Princeton: Princeton University Press. Kriesi, Hanspeter, Edgar Grande, Martin Dolezal, Marc Helbling, Swen Hutter, Dominic Höglinger, and Bruno Wüest, eds. 2012. Political Conflict in Western Europe [en]. Cam- bridge, MA: Cambridge University Press. Kriesi, Hanspeter, Edgar Grande, Romain Lachat, Martin Dolezal, Simon Bornschier, and Tim- otheos Frey. 2008. West European Politics in the Age of Globalization [en]. Cambridge: Cambridge University Press. Krisch, Nico. 2010. Beyond Constitutionalism: The Pluralist Structure of Postnational Law [en]. Oxford: Oxford University Press. Kymlicka, Will. 2010. “The rise and fall of multiculturalism?: New debates on inclusion and accommodation in diverse societies.” [en]. International Social Science Journal 61 (199): 97–112. Lehmbruch, Gerhard. 1977. “Liberal Corporatism and party government.” [en]. Comparative Political Studies 10 (1): 91–126. Lipset, Seymour M., and Stein Rokkan. 1967. “Cleavage Structures, Party Systems and Voter Alignments. An Introduction.” [en]. In Party Systems and Voter Alignments. Cross-national Perspectives [en], eds. Seymour M. Lipset and Stein Rokkan. New York, NY: Free Press, 1– 64. Mahoney, James. 2000. “Path Dependence in Historical Sociology.” [en]. Theory and Society 29: 507–48. Mair, Peter. 2006. “Cleavages.” [en]. In Handbook of Political Parties [en], eds. Richard S. Katz and William Crotty. London: Sage, 371–75. ———. 2013. Ruling the Void: The Hollowing of Western Democracy [en]. London: Verso. May, Judith V., and Aaron B. Wildavsky. 1978. The Policy Cycle [en]. v. 5 of Sage yearbooks in politics and public policy . Beverly Hills: Sage Publications. Mayntz, Renate. 2004. “Mechanisms in the Analysis of Social Macro-Phenomena.” [en]. Phi- losophy of the Social Sciences 34 (2): 237–59. Merkel, Wolfgang. 2017. “Kosmopolitismus versus Kommunitarismus: Ein neuer Konflikt in der Demokratie.” [en]. In Parties, Governments and Elites [de], ed. Harfst P., Kubbe I., Poguntke T. Wiesbaden. Milanović, Branko. 2016. Global Inequality: A New Approach for the Age of Globalization [eng]. Cambridge, MA: Harvard University Press. Milner, Helen V. 1988. Resisting Protectionism: Global Industries and the Politics of Interna- tional Trade [en]. Princeton: Princeton University Press.

Möllers, Christoph. 2013. The Three Branches: A Comparative Model of Separation of Powers [en]. Oxford constitutional theory . Oxford: Oxford University Press. ———. 2017. “Wir, die Bürger(lichen).” Merkur 818: 5–16. Moore, Barrington. 1966. Social Origins of Dictatorship and Democracy [en]. Boston: Beacon Press. Müller, Jan-Werner. 2016. Was ist Populismus [de]. Berlin: Suhrkamp. Pogge, Thomas. 2002. “Moral Universalism and Global Economic Justice.” [en]. Politics, Phi- losophy and Economics 1 (1): 29–58. Preuß, Ulrich K., ed. 1994. Zum Begriff der Verfassung [de]. Frankfurt a. M: Fischer Taschen- buch Verlag. Rapaport, Orit, David Levi-Faur, and Dan Miodownik. 2009. “The Puzzle of the Diffusion of Central-Bank Independence Reforms: Insights from an Agent-Based Simulation.” [en]. The Policy Studies Journal 37 (4): 695–716. Rieger, Elmar, and Stephan Leibfried. 1997. “Sozialpolitische Grenzen der Globalisierung. Wohlfahrtsstaatliche Gründe außenwirtschaftlicher Schließung und Öffnung.” [de]. Politi- sche Vierteljahresschrift 38 (4): 771–96. Risse, Thomas, Steve C. Ropp, and Kathryn Sikkink. 2013. The persistent power of human rights: From commitment to compliance [en]. Vol. 126 of Cambridge studies in international relations . Cambridge, United Kingdom: Cambridge University Press. Rogowski, Ronald. 1989. Commerce and Coalitions. How Trade Affects Domestic Political Alignments [en]. Princeton, N.J: Princeton University Press. Ruggie, John G. 1983. “International Regimes, Transactions, and Change: Embedded Liberal- ism in the Postwar Economic Order.” [en]. In International Regimes , ed. Stephen D. Kras- ner. Ithaca: Cornell University Press, 195–231. Tallberg, Jonas, Thomas Sommerer, Theresa Squatrito, and Christer Jönsson. 2013. The Opening Up of International Organizations: Transnational Access in Global Governance [en]. Cambridge: Cambridge University Press. Thatcher, Mark, and Alec S. Sweet. 2002. “Theory and Practice of Delegation to Non- Majoritarian Institutions.” [en]. West European Politics 25 (1): 1–22. Zangl, Bernhard. 2006. Die Internationalisierung der Rechtsstaatlichkeit. Streitbeilegung in GATT und WTO [de]. Frankfurt a.M: Campus. Zürn, Michael. 1996. “Über den Staat und die Demokratie im europäischen Mehrebenensys- tem.” [de]. Politische Vierteljahresschrift 37 (1): 27–55. ———. 2004. “Global Governance and Legitimacy Problems.” [en]. Government and Opposi- tion 39 (2): 260–87. ———. 2014. “The politicization of world politics and its effects: Eight propositions.” [en]. European Political Science Review 6 (01): 47–71. ———. 2017. “From Constitutional Rule to Loosely Coupled Spheres of Liquid Authority: A Reflexive Approach.” [en]. International Theory 9 (2): 261–85. 31

Zürn, Michael, Martin Binder, and Matthias Ecker-Ehrhardt. 2012. “International Authority and Its Politicization.” [en]. International Theory 4 (1): 69–106. Zürn, Michael, Martin Binder, Alexandros Tokhi, Xaver Keller, and Autumn Lockwood Payton. 2015. The International Authority Data Project, Paper presented at the International Au- thority Workshop, December 10-11, Berlin. [en]. Zürn, Michael, and Pieter De Wilde. 2016. “Debating Globalization: Cosmopolitanism and Communitarianism as Political Ideologies.” [en]. Journal of Political Ideologies 21 (3): 280– 301. Zürn, Michael, and Nicole Deitelhoff. 2015. “Internationalization and the State - Sovereignty as the External Side of Modern Statehood.” [en]. In The Oxford Handbook of Transfor- mations of the State [en], eds. Evelyne Huber, Matthew Lange, Stephan Leibfried, Jonah Levy, Frank Nullmeier and John Stephens. Oxford: Oxford University Press.