The Political Asymmetries of Euroskeptic Parties ...

The Political Asymmetries of Euroskeptic Parties and Coalition Governments: The Cases of Austria and Slovakia

Robert Pahre Department of Political Science University of Illinois at Urbana-Champaign 382 Lincoln Hall, 702 South Wright Street Urbana, IL 61801 [email protected]

ABSTRACT Euroskeptic parties exhibit four asymmetries in their behavior. They are less likely to join governing coalitions than are parties that support the European Union, and there are no countries in which they are more likely to form a government. When they are internally divided, the masses will be more Euroskeptic than the leadership, and the reverse pattern is not found. If these change their platforms, they will be more Euroskeptic when in opposition than in government. Finally, Euroskeptic parties are more likely to be extreme on issues other than the EU. This paper explains these asymmetries through a spatial model of policy. All four stem from the added utility available to Europhile parties through their ability to negotiate policy concessions from other members of the EU.

Euroskepticism has become an increasingly visible part of the European political landscape. Though found in countries such as Denmark and the United Kingdom since their accession to the Union, public concerns about the direction of the EU have gained force throughout Europe since the deep integration of the 1986 Single European Act and the 1991 Maastricht Treaty on European Union. Outright Europhobia may also be found in many countries. Both Euroskepticism and Europhobia also play an important role in some parties in many new accession countries. Euroskeptic and Europhobic parties tend to share several features in coalitional polities, that is, those countries regularly governed by parliamentary coalitions. First, Euroskeptic parties are much less likely to participate in government than are other parties, and Europhobic parties do not participate in government at all. In contrast, Europhile parties are never ostracized in this way. Second, many political parties have Europhile elites and Euroskeptic masses. However, there are no parties in coalitional polities that have a Euroskeptic leadership and a Europhile mass membership. This pattern is especially remarkable when we consider that public attitudes vary widely, from the Europhile Benelux countries to Euroskeptic Scandinavia. This variation would seem to provide a political basis for the alternative kind of mass/elite cleavage at least somewhere. Third, Euroskeptic parties that participate in government tend to become reconciled to the European Union, while parties in opposition (even those with a history of Europhilia) tend toward Euroskepticism. Once again, we do not find the reverse pattern of a Euroskeptic government and Europhile opposition. Fourth, parties that are extreme on non-EU issues are more likely to be Euroskeptic or Europhobic. Denmark’s Radikalt Venstre (“Radical Left”) seems to be the only centrist party in a European coalitional polity that has a strong tradition of Euroskepticism, though some others have 1

Euroskeptic factions. Extremist parties on both Left and Right tend to be Euroskeptic or Europhobic, though this tendency is somewhat weaker than the first three of my generalizations. In short, we can identify four asymmetries of Euroskepticism: in government coalitions, in elite/mass cleavages, in government/opposition dynamics, and extremism. All four are familiar to observers of at least some countries, though scholars seem to be more hesitant about drawing the cross-national generalizations that I have presented. A few comparative studies have identified some of these patterns inductively, but these methods are better at finding patterns than providing causal explanations of them. The deductive approach here provides a unified causal explanation consistent with all these generalizations, and points at new directions for future research. This paper shows that these asymmetries reflect the same political logic and stem from the needs of government coalition formation. I use a spatial model of policy and parliamentary coalition formation to show how coalition negotiations that include EU issues differ from negotiations over purely domestic policy dimensions. EU issues bring with them the policy externalities of European integration: any EU member government that cooperates with other EU members to address policies of joint concern brings home policy gains from abroad. For example, a German government that works with France can change not only German policy in a way beneficial to German voters but can also change French policy. A Polish government willing to join Europe may obtain policy concessions in the medium to long term, and will obtain cash for Polish agriculture and regional development in the short run. These benefits from cooperation produce political gains for Europhile parties that Euroskeptic and Europhobic parties cannot match, making these anti-EU parties more expensive coalition partners. This logic produces a pattern in which government/opposition dynamics will be associated with Europhilism/Euroskepticism. Elite/mass cleavages within some parties reflect the same logic. Elites who value 2

officeholding will be more likely to make their positions congenial to European integration so as to be acceptable coalition partners. The masses, who value policy but not officeholding, will not take this consideration into account. The result is a possible elite-mass cleavage, always in the direction of a Europhile elite and Euroskeptic mass. The tendency toward Europhilia in office and Euroskepticism in opposition reflects the same logic. Finally, extremist parties are more difficult coalition partners in any system simply because of their extremism. Since governing parties will be overwhelmingly Europhile, extremist parties can pick up Euroskepticism or Europhobism as a costless oppositional position when they are out of power. This strategy also helps opposition parties pick up votes from Euroskeptic factions of Europhile governing parties. However, this Euroskepticism makes these extremist parties even less attractive as coalition partners, which reduces the pressure on them to become less extremist. All four of these explanations stem from the fact that Europhile parties are cheaper coalition partners because they can deliver policy externalities from abroad. This means that countries with majoritarian governments—such as the United Kingdom, Malta, Greece, Spain or (increasingly) France—will not be subject to the same pressures that I identify. After all, the costs of various coalition partners are irrelevant for a single-party government of any political persuasion. As a result, these countries should not exhibit these asymmetries. The British Conservatives under Margaret Thatcher, like Malta’s Labour party today, had a Euroskeptic leadership and somewhat more Europhile mass membership. Both cases indirectly confirm the causal logic of the theory here because such parties are only found in majoritarian countries.

Europhilia, Euroskepticism and Europhobia Euroskepticism, like the European Union itself, has only become visible enough to affect 3

national politics in the last few decades. The EU first shaped domestic politics significantly in Denmark, immediately after it became a member in 1973 (Pahre 1997; Thomsen 1993). It became more salient in other countries in the 1980s, especially with Margaret Thatcher in office. Research on the subject of Euroskepticism naturally began inductively, describing the phenomenon, working up analytical concepts and categories, and beginning to make generalizations about Euroskepticism and its importance for politics. Scholars have sought to document the extent of Euroskepticism among voters, or to examine the relationship between mass Euroskepticism and the Euroskepticism of political parties. For example, some studies have argued for a distinction between “hard” and “soft” Euroskeptics (i.e., Taggart 1998; Szczerbiak 2001; Taggart and Szczerbiak 2002). Kopecký and Mudde (2002) argued that this hard/soft distinction did not capture the differences between parties who objected to the EU in principle and those who favored European integration but disagreed with the practice of the EU. Alongside such typological work, others have begun to examine the correlates of Euroskepticism. Research has traced Euroskeptic beliefs to a voter’s material interests, the national economic interests of a country; or to a particular ideology (i.e., Anderson and Kaltenthaler 1996; Eichenberg and Dalton; 1993; Gabel 1998; Hooghe 2003 inter alia). In addition to simple material interests, some have connected Euroskepticism to political attitudes on other political economy matters (i.e., Cichowski 2000; Henderson 2001). Others find it associated with a person’s location on the Left-Right ideological spectrum (i.e., Aspinwall 2000; Marks et al. 2002; Noury 2002). Many strong nationalists tend to oppose the EU for ideological reasons as violations of sovereignty or being overly internationalist or “cosmopolitan” (which sometimes means “Jewish”). These kinds of Euroskeptics include, for example, the Slovak National Party (SNS) and their splinter party, the “Real” Slovak National Party (PSNS), as well as much of the right-wing in 4

Poland (Strobel 2001). Euroskepticism may also be connected to a citizen’s or party’s general view of foreign affairs, or even to its position on NATO or relations with the United States. The association between Euroskepticism and anti-Americanism has been sufficiently strong that it was newsworthy when the previously-Euroskeptic leader of Denmark’s Left Socialists (SF), Holger K. Nielsen, came out in favor of stronger European cooperation in foreign and defense policy because Denmark and the EU should not have a foreign policy “pawned” (pantsat) in Washington (“EU skal være garanten for demokrati,” 12 May 2003). Many scholars have noticed a divided between elite and mass views on the European Union, and divisions between national and local elites (Hughes et al. 2002 inter alia). Instead of seeing Euroskepticism as a feature of public opinion, some scholars focus on the parties, often viewing it as a strategic choice of opposition parties (i.e., Taggart 1998; Sitter 2001). We can think of the literature as having focused on Euroskepticism at three levels: public opinion, parties, and governments. Important as this approach has been for establishing the foundations for studying the subject, it has been limited by its methods. The limits of an inductive approach are most evident when we consider how these different levels affect each other. Focusing on public opinion ignores the role of government and political parties in shaping public opinion, though this opinion leadership has played a very important role in referenda campaigns in particular. Examining party platforms downplays the strategic calculations of party elites, and internal divisions within parties. Some parties with Europhile platforms may nonetheless attract support from Euroskeptic voters for reasons that have nothing to do with the EU. A party might also take Europhile positions but blur these enough to attract moderate Euroskeptics. Finally, looking at behavior only when a party is in government misses the possibility that a party may profess Euroskeptic views but not act on them in office. It may even take 5

Europhile positions in office but depend on Euroskeptics at election time. By emphasizing belief systems, this research has been insensitive to the strategic environment in which Euroskeptic parties find themselves. For example, everyone seems agreed that the Czech party ODS is a Europhile party, though it makes use of Euroskeptic rhetoric. Conversely, Slovakia’s HZDS is generally classified as Euroskeptic despite its official position in favor of the European Union. As I will argue throughout this paper, these cases become less puzzling if we explicitly take account of the strategic environment. In some environments, a Euroskeptic HZDS may find it politically expedient to profess support for the European Union, while rhetorical Euroskepticism may help ODS distinguish itself from other Europhile parties. In short, there can be important disjunctures between the different factions of a political party, its political tactics, and role as government or opposition that shape its Euroskepticism. This paper develops a deductive theory that makes causal connections between these different levels of Euroskepticism. I focus on the strategic calculations of party elites, where the assumptions of rational choice fit much more easily than in the study of public attitudes. This strategic perspective sometimes identifies a different causal logic than does the literature, and sometimes makes different predictions about behavior.

Coalition formation in a spatial model without the European Union The basic processes of coalition formation are easily understood in a spatial model. As its name implies, the spatial model of policy represents policy as points in space. It is easiest to think of this model in terms of a one-dimensional issue space, with political goals as points on a line. For example, we can think of monetary policy as a line segment whose endpoints are labeled “tight” and “loose.” It is easy to think of many policies as falling on such a line. 6

Just as any policy can be placed on this line, so too can people’s views of that policy. The spatial model of policy assumes that every political actor has some ideal policy, which is his or her ideal point. The model assumes further that each actor judges policies by how far they are from that actor’s ideal point; the closer a policy, the better. Each individual's goal is to obtain a policy that is as close as possible to his or her own ideal point. The farther away a policy is in either direction from the individual's ideal point, the less the individual likes the policy. In a coalitional polity, parties form governing coalitions because none controls enough seats to govern alone. The costs and benefits of including each party in the coalition lie at the heart of my analysis. Because I do not model the process formally, this section does not draw explicitly from the growing literature on coalition formation in parliamentary democracies, but is certainly informed by it (i.e., Laver and Shepsle 1990; Martin and Stevenson 2001; Pahre 1997; Strom 1990). Consider three parties, A, B, and C, located

Figure 1 Coalition Formation without the EU

in two-dimensional space (see Figure 1). If A and B form a government, they will choose a policy on the line AB between them. All points on this line are efficient, since any point off this line is dominated by some point on AB that 7

would make both A and B better off. Similarly, a coalition of B and C would choose a point on the BC line, while a coalition of A and C would choose a point on the AC line. Finally, a grand coalition of A, B, and C might choose any policy in the triangle ABC, shaded in Figure 1. Inside this triangle, a move of policy in any direction toward one (A) or two (A and B) ideal points must also move policy away from the third ideal point (C). Such moves would be rejected by the grand coalition, so any point in the triangle is efficient. Not all points on these lines (or in this triangle) are attractive alternatives to the status quo, however. When A and B form a governing coalition, each party will only favor a coalition policy that it prefers to the status quo. I have illustrated three possible status quos in Figure 1, labeled Q1, Q2, and Q3. The set of points that B prefers to Q3 is inside the circle with center B through Q3 (its acceptance set). This means that a coalition of A and B will only choose a policy inside both of their acceptance sets and on the line segment A and B. I have drawn these policies with a thick black line in the figure. This framework can predict the policies chosen by any coalition, given any SQ. For example, A and B (or B and C) would choose from a much smaller set of point with the status quo at Q1, limited to the small, dotted circle. However, the framework does not predict which coalition will form, a question on which there is an extensive literature (i.e., Hardin 1976). The next section will show that coalition formation is different when we consider the question of European integration. In this case, some policy dimensions are outside the control of domestic parties alone, even when they act together. Denmark may care about Germany’s asylum policy, but it can only change that policy by negotiating with the German government. As in the preceding section, parties not only have different policy preferences over domestic policy but also 8

over international cooperation. As a result, parties do not form coalitions in order to choose policy directly but choose coalitions representing some win-set. In other words, parties choose policy only indirectly, since the outcome of negotiations with the EU cannot be known in advance, but the general outlines of such intergovernmental negotiations (the possible win-set) is known.1

The EU, coalition formation, and a strategic definition of Euroskepticism This section will examine the politics of government coalition formation in an environment dominated by cooperation with other member states of the European Union. It applies best to that majority of European countries with parliamentary forms of government and without single-party majority governments. The analysis here should not apply to politics in Great Britain or other countries with majoritarian politics. One feature of intergovernmental bargaining is that each negotiator controls some policy that other negotiators do not. Each negotiator offers concessions on its policy in exchange for movement on other policies that it cares about but does not control. To capture this simply, I will use a two-dimensional model in which the home country controls the x-axis, while the EU controls the y-axis.

1

An alternative assumption is that each minister has dictatorial control over the policy dimensions for which his ministry is responsible (see Laver and Shepsle 1990). Because this minister would have to negotiate with foreign counterparts, government formation and institutional choice would still depend on the anticipated win-set in the bargaining game, which drives the results here. 9

Figure 2 shows a game with this bargaining

Figure 2 Negotiations with the European Union

problem. If they did not negotiate with foreigners, a coalition of parties A and B would choose an x-axis policy in the direction shown by the arrow in the figure.2 However, both A and B would like reductions in the y-axis policy, which they can obtain in negotiations with EU. The possible set of agreements between A, B, and EU is shaded in Figure 2, the intersection of (a) all three players’ acceptance sets; and (b) the set of efficient policies, characterized by the triangle connecting their ideal points. None of these points are acceptable to the third party, C, because they lie outside C’s acceptance set (the circle with center C through Q). It is easy to verify that a minority government of A or B would also negotiate agreements that are unacceptable to C. These observations lead to our first definition: Remark 1. It is possible that a party will reject any agreement that any party or parties, not including itself, would sign with the European Union. Definition 1. A “Europhobe” party will reject any agreement that any party or parties, not

2

I do not consider how the status quo ended up at Q, a question that can affect analysis considerably in such games (see Pahre 2002). Clearly, the x-component of Q is a possible outcome of a previous BC or ABC coalition in Figure 2. 10

including itself, would sign with the European Union. The strategic context is essential to this definition. Note that a C minority government could negotiate agreements with the EU. The win-set for C and the EU consists of those points on the dashed line C-EU and inside the small lens to the south-southwest of Q.3 In other words, looking solely at whether a party would reject integration with the EU is misleading, since even a Europhobe government can reach some agreement with the EU.4 In this view, Europhobic views can only be defined relative to the views of other parties. Returning to Figure 2, we see that all of the possible agreements negotiated by a B minority government would be acceptable to A, though only some of those agreements negotiated by an A minority government would be acceptable to B. Both parties would accept any agreement negotiated by a C minority government. These illustrations exhaust the logical possibilities and provide our next two definitions: Remark 2. A party might accept some but not all of the agreements that any other party or parties, not including itself, would negotiate with the EU; a party might also accept all such agreements. Definition 2. A “Euroskeptic” party will accept some but not all agreements that any party or parties, not including itself, would sign with the European Union. Definition 3. A “Europhile” party will accept all agreements that any party or parties, not including itself, would sign with the European Union. While the public opinion literature tries to define Euroskepticism in terms of preferences in the abstract, I define all three concepts strategically. This means that each position is defined in

3

Similarly, C would accept a small set of possible agreements between an AC or BC coalition government and the EU. 4

In a spatial model, an agreement will always be possible unless the status quo already lies on the contract curve. Spatial theory cannot capture the situation when a party refuses to negotiate with the EU as a matter of principle. 11

terms of the agreements that a party will or will not accept in its own country’s political context. These definitions resemble, but are not the same as, some definitions of Euroskepticism in the literature. For example, my Europhobes are essentially Taggart’s hard Euroskeptics, while my Euroskeptics and his soft Euroskeptics overlap to a considerable degree. My definitions resonate particularly well with those Euroskeptic parties that accept European integration in principle but dislike the current form of integration (see Kopecký and Mudde 2002). Such parties would accept some other forms of integration, and often accept most of the EU status quo, even as they reject current proposals for deeper integration. This strategic definition has the advantage of relying on observables. It is notoriously difficult to describe a party’s position in the abstract. It is also very difficult to classify positions so defined as being Euroskeptic or not. In contrast, it is easy to determine whether a party accepts or rejects another party’s policy toward the EU. Such questions are major issues in party platforms, coalition negotiations, and election campaigning.

Figure 3 Euroskepticism in Strategic Terms

These definitions have the interesting feature that two parties with identical preferences could be classified differently in two different countries. Figure 3 illustrates a slightly different constellation of party preferences, with B’s 12

ideal point moved significantly to the right. Consider party C in Figures 2 and 3. In Figure 2, C is Europhobic, since it rejects all of the agreements that A and B would negotiate (individually or jointly) with the EU. In Figure 3, C is Euroskeptic. It would accept any agreements negotiated by an AB coalition (or a B government), though it would reject agreements negotiated by an A government. In Figure 3, then, our classification of C depends on the location of B. Of course, B’s interests have also changed. B would reject any agreement between A and EU, though it would accept all agreements between C and EU and some agreements between an AC coalition and the EU. In both Figure 2 and Figure 3, defining A as a Europhile party certainly captures the spirit of the qualitative literature. Party A has preferences somewhat close to EU preferences, and will accept any agreement negotiated by any combination of other parties. Party B is similarly Europhile in Figure 2 but not in Figure 3. Party C rejects all agreements reached by others in Figure 2, but will accept some but not all agreements in Figure 3 even though its own preferences have not changed. In effect, my definitions reflect the notion that “extremism” is context-dependent. A party whose views lie well outside one country’s mainstream may not lie outside the mainstream if it were moved to another country. The same is true of Europhobia and Euroskepticism. Those who seek to improve our cross-national understanding of Euroskeptic public opinion will probably find this feature of my definitions very troublesome. In my definition, beliefs that would be Europhobic in Luxemburg could easily be Euroskeptic or perhaps even Europhile in Denmark. Because my focus does not lie in public opinion, I will simply note these problems without solving them. For legislative behavior, this feature of my definitions is not a problem, since coalition 13

formation in a legislature is inherently a strategic problem. Successful professional politicians with good information about one another and excellent negotiation skills care more about what agreements they can or cannot reach with other parties and with the EU. A strategic definition of Euroskepticism better captures the problems that these elites face. The main challenge for with these definitions is that too many parties might be labeled “Euroskeptic” in particular countries. Accepting only one possible agreement will move a Europhobic party into the Euroskeptic classification, just as the theoretical rejection of any agreement will make a Europhile party Euroskeptic. Impressionistically, this problem is most likely found among social-democratic parties. Even when in opposition, center and center-right parties in most countries will accept likely any deeper integration that the government negotiates. Many social-democratic parties, in contrast, would reject a range of integration negotiated by center-right parties, even if they generally favor the European Union. Sometimes this opposition reflects purely tactical considerations, while in other cases it stems from the existence of a sizeable Euroskeptic faction within each social-democratic party—an issue to which I return below. The fact that the definitions and typology of Euroskepticism are not settled in the literature certainly leaves some room for ad hoc definitions. Because the literature has been entirely empirical and inductive, existing typologies are largely unconnected to any broader theoretical perspective. This too means that these definitions may be inappropriate for particular empirical or theoretical problems. The strategic approach here sets out an alternative foundation for categorizing Euroskepticism that may (or may not) prove to be useful as research progresses.

Euroskepticism and parliamentary behavior The different types of parties discussed in the previous section affect both negotiations over 14

new levels of integration and parties’ attitudes toward existing levels of integration. The effects are particularly noticeable when we consider existing levels of integration negotiated by Europhile parties and a Euroskeptic party that seeks to join the governing coalition.

Consider three parties with different

Figure 4 Existing Integration and New Coalition Negotiations

evaluations of the EU. Suppose that two nonEurophobe parties form a coalition, which then negotiates deeper integration with the EU. Figure 3 shows such a case, with parties A and B having forming a coalition government that negotiated integration with the EU at the point labelled I. Now consider negotiations over a new government, after elections for example. If A and B form a coalition, European integration does not present any issue, since both prefer the point I to Q. However, C prefers Q to I. This poses a problem if it wants to negotiate a possible new government with either A or B. The main challenge is that the Euroskeptic party C cannot accept the current level of integration, and will need to be compensated on other issues. As a result, 15

Remark 3. Satisfying a Europhobic or Euroskeptic party in a new coalition entails a loss in utility for the Europhile party or parties. This means, of course, that the Europhiles will avoid coalition with the Euroskeptics/Europhobes so as not to pay those costs. Phrased differently, the Euroskeptics and Europhobes will have to bring something new to the table to overcome this inherent unattractiveness. This tendency is strongest for Europhobes, weaker for Euroskeptics who may resemble either Europhiles or Europhobes. In short: Remark 4. Non-Europhobe parties are better off reaching coalition agreements with other non-Europhobes. Remark 5. Europhobe parties that enter a government will make extraordinary concessions to enter. A necessary condition for Europhobe participation in government, then, is the existence of some benefits available to coalitions outside this figure, such as other issues or the benefits of office-holding. If C wants a coalition to reject current integration (I) in favor of the unilateral policy Q, it will have to offer either A or B some additional gains in these other issues, since A and B can always form a coalition with one another in which they would not have to forego the utility gains of current integration. By raising the cost of a coalition with C, this logic makes C's entry into the government less likely than it otherwise would be. This means that Proposition 1. Euroskeptic parties are less likely to join governing coalitions than Eurofriendly parties. European integration offers utility gains to political actors above and beyond those available from domestic politics, since European integration also affects the policy of other member states. Parties that forgo these utility gains are more expensive coalition partners, and are therefore less likely to be invited to join a government. Instead of looking at the implications of Euroskepticism for coalition formation we might 16

also consider how coalition formation process shaped a party’s own preferences. To continue with the illustration above, C could acquiesce to the policy I when negotiating with A or B. It might demand additional compensation elsewhere, but A and B may have little incentive to provide this compensation since they could form a coalition with one another in which they would not need to offer this additional compensation. When C joins a government, then, it is may accept existing integration as a fait accompli as part of the coalition formation process. In this way, it absorbs the utility costs of European integration: Corrolary 1A. Euroskeptic parties that participate in government become reconciled to existing integration the European Union, even if they oppose future deepening. The higher utility cost of reversing integration means that Euroskeptics in government will be reconciled to existing integration for two reasons. First, the utility gains of integration means that some Euroskeptics will accept integrative policies at a Union level that they would reject unilaterally, especially if the party’s elite receives personal utility from participating in government. Second, there is a selection bias in that Euroskeptics unwilling to compromise this particular principle will not be found in government. As a result, the Euroskeptics we find in government will all be reconciled to existing levels of integration. These dynamics are especially evident when a new country joins the European Union. Parties that opposed membership generally shift from opposing the EU to supporting existing levels of integration but opposing new integration. One early example is the RV in Denmark. Though originally opposed to membership, it accepted membership as a fait accompli and became a pivotal player in EEC affairs for most of the next two decades (see Pahre 1997). The strategic definition of Euroskepticism and Europhobia above is clearly relevant to this problem. A Europhobe party is so hostile to European integration that they will require breaking

17

the treaty commitment at I. Some Euroskeptics will actually accept this level of integration; those that do not will only have to pay relatively low costs to join a government. As a result, Corrolary 1B. Euroskeptic parties will become reconciled to issues where integration occurs, and thus reconciled to the European Union itself; Europhobic parties will remain locked into permanent opposition. One only sees this distinction in issue areas where integration has already occurred. Since the definitions and operationalizations depend in part on behavior, this claim is difficult to falsify. However, it does imply an existence claim for which one may find confirming evidence, namely, that there exist parties whose preferences are such that they oppose integration before it happens but who will not unmake integration once it has occurred. My explanation of why anti-EU parties tend to be out of government differs significantly from others, which treat the relationship as a spurious result of centrist politics. For example, Aspinwall (2002: 106) argues that pro-integration centrists are elected more often, and are more likely to form governments because consensus (PR) systems are biased toward centrist parties.5 This alternative claim would be unable to explain why a centrist Euroskeptic party, such as Denmark’s “Radical Left” (Radikalt Venstre, RV), would typically find itself outside government in the 1970s and 1980s (Pahre 1997). In sum, this section has argued that Euroskeptic parties find themselves disadvantaged in the coalition formation process of a parliamentary democracy. They are less likely to join governments and typically compromise some of their Euroskeptic principles to get there. The next

5

My explanation resonates with Martin and Stevenson’s (2001: 36) hypothesis that coalitions containing anti-systemic parties are less likely to form, if we suppose that many antisystemic parties are also Euroskeptic, or if we simply define anti-EU parties as anti-systemic. However, the existence of pro-systemic anti-EU parties such as Denmark’s RV, the British Conservatives, or the Polish Peasants’ Party suggests that no such definition would be appropriate. 18

section argues that joining the government may also produce internal cleavages in these parties if we make some simple assumptions about preferences.

The elite-mass cleavage in Euroskeptic parties The politics of coalition formation may stretch beyond the interparty negotiation sketched out in the previous section. The analysis above assumes some benefits outside the integration policy dimensions from joining a coalition government. These benefits could lie in other policy areas, or they could be the benefits of office-holding to the elite of a party. This section will examine how the benefits of office-holding combine with the political economy of integration in a spatial model to produce elite-mass cleavages of a certain type. I will argue that we will observe parties with a pro-integration leadership and anti-integration mass membership, but will not observe the reverse type of party, with anti-integration leaders of pro-integration masses. This argument rests on the assumption that the leadership and the masses have different preferences over government membership even if they have the same preferences over policy. The rank-and-file in a party will receive any policy benefits that the elite obtains in a coalition treaty, but only the elite receives the benefits of office-holding.6 Presumably the party leadership considers both policy and officeholding gains when it decides to enter a coalition. This helps explain an otherwise-puzzling regularity in elite attitudes across Europe. It is often noticed that national elites have different views than local elites (Hughes et al. 2002) as well as the public (for a nuanced view, see Hooghe 2003). Yet, if only policy mattered, both ordinary

6

This elite would include cabinet and sub-cabinet level people as well as anyone who benefits from their patronage at lower levels of government (or business). Obviously, patronage systems such as the Austrian Proporz give the rank-and-file some personal benefits from government membership. 19

members and the leadership of a party should have a similar distribution of preferences on policy matters. Disagreements over policy will naturally occur within the leadership as in the membership at large, but there is no a priori reason to expect different ranges or distributions of policy preferences. However, only the leadership can enjoy the non-policy perks of office, so it will value joining the governing coalition more than the membership does. If faced with a choice between joining a pro-integration government and remaining hostile to the EU in opposition, it is entirely possible that the leadership will find the utility of coalition greater than that of opposition, while the rank-and-file prefers opposition, even if they have the same preferences over policy. In other words, Proposition 2. We may observe parties in government with a pro-European leadership and Euroskeptic membership, but will not find parties with a Euroskeptic leadership and proEurope mass membership. This claim depends on the argument in the previous section that Euroskeptic parties are less likely to join government coalitions, combined with the assumption that leaderships are more likely to value joining the government for personal, officeholding reasons. We find this pattern most frequently among the Social Democratic parties in Europe. In Sweden, for example, forty to fifty percent of the Swedish Workers’ Party (Svenska Arbeiderpartiet, SAP) are Euroskeptic despite the pro-European orientation of the leadership. About two-fifths of the party electorate defects during elections to the European Parliament, voting mostly for the Euroskeptic Left Socialists instead. Germany’s, Austria’s and Denmark’s Social Democratic Parties, among others, exhibit a similar, if smaller, cleavage. Defection from the party position is also common in European referenda for these parties (Svensson 2002). Various other non-Socialist parties also have such cleavages, including Finland’s Center, Germany’s FPD, and Denmark’s Liberals (Venstre). 20

The reverse cleavage, with a Euroskeptic leadership and pro-integration mass membership, is not found in coalitional polities. In majoritarian Britain, on the other hand, the two largest parties are both split on the EU issue, and both Euroskeptic and Europhile leaders have appeared at the head of each party since membership. Because majoritarian systems do not fit the assumptions of the theory here, the fact that this different elite-mass cleavage is found there provides circumstantial evidence in favor of the coalitional logic identified here. Additional circumstantial evidence comes from studies of European parties and European integration. Elites are consistently more integrationist than the masses in their views of the EU as a whole, though this general attitude is not found in every single issue area (Hooghe 2003). However, this is not a settled question empiricially, and it is sensitive to differences between center and peripheral elites, elites and masses (see Aspinwall 2002; Hughes et al. 2002; Marks et al. 2002). My analysis suggests that different national strategic contexts matter, and they probably confound existing results. In a liberal democracy, elites who consistently pursue policies contrary to the wishes of a large share of their electorate risk the wrath of that electorate. This seems to be the case for those parties with the elite/mass cleavage, especially as European integration becomes more important to the public. As a result, we should expect Corrolary 2A. Governing parties with a split between leadership and masses on the EU the EU will lose votes to anti-EU parties who are excluded from the government. When voters have alternatives, especially opposition parties that have chosen not to compromise their Europhobic principles in order to join the government, they will eventually chose those alternatives. I have already mentioned the case of Swedish Social Democrats defecting to the Left Socialists in elections to the European Parliament.

21

The European Union therefore provides a particularly salient issue for opposition parties. It allows them to exploit a natural division between elites and masses, in hopes of attracting disaffected members away from the pro-integration policies of their leadership. Factionalism is rarely an advantage, so the advantageous Euroskepticism of the opposition differs from the disadvantageous factionalism of governing parties (Taggart 1998: 372-373). The success of Euroskepticism as a tactic of opposition does not stem from any inherent popularity of Euroskeptic positions. In fact, Euroskepticism tends to be a second-order issue that cuts across the usual party divide. The reasons for Euroskepticism vary enormously, from pragmatic material interests (farmers in many countries), religious nationalism (Poland), secular nationalism (France), xenophobia and/or anti-semitism (Austria, Belgium, France), and secular socialism (Scandinavia). Reflecting this heterogeneity of motivations, Euroskepticism is an uncertain vote getter in itself, and its popularity makes sense most as part of a strategy of opposition (Sitter 2001) Because of party factions, many parties will change their position on various issues over time. Attitudes toward the EU should be no different, and need not be constant. However, shifts in positions will be patterned. Because party elites benefit from holding office, we should see a relationship between Europhilia and office-holding in those parties that change positions: Proposition 3. A party whose position on the EU changes depending on whether it is in government or opposition will be Europhile in government, Euroskeptic in opposition. There are two possible reasons for this. The first is a kind of selection effect. Membership in government may reflect the outcome of internal struggles between factions. When the Europhiles are ascendant, for whatever reason, the party is more likely to join a governing coalition. Second, office-holding benefits may have a direct effect. Party elites wavering between Europhilia and

22

Euroskepticism may decide to support the EU more strongly in order to join a governing coalition and reap the rewards of office. Taken together, these propositions suggest that the “Franklin thesis” debated in the literature may reflect a spurious relationship between public support for a governing party and support for the EU, especially on referenda (see inter alia Franklin et al. 1994; Ray 2003; Svensson 2002). If EU parties are more likely to join governments, if elite-mass cleavages are structured in a particular way, and if even divided parties are more likely to be Europhilia when in office, then we should see Europhile voters being more likely to vote for governing parties because they reflect their ideology (regardless of governing status). Similarly, they should be more likely to vote against the government when they are already Euroskeptic or Europhobe. In the discussion so far, I have regularly noticed a tendency for centrists to be Europhiles (and vice versa). This stems naturally from the logic of a coalitional system, where governing coalitions are usually built in the center of the political system, and where governing parties are more likely to be Europhile. Parties with little coalitional possibilities, especially those on the extremes of the political system, will not be pressured toward Europhilia. As a result, we should find that Proposition 4. Otherwise extreme parties are more likely to be Euroskeptic. This does not reflect any necessary extremism in the ideology of Euroskepticism. In majoritarian countries such as the UK, it is part of the mainstream, an issue on which reasonable people differ. Euroskepticism is not a mark of extremism in such countries, and is also a mainstream position in Malta (especially among the Labour Party). In countries with governing coalitions, however, the pressures of office-holding tend to drive Euroskepticism out of the political center. The reasons for this follow from the preceding 23

propositions. It is easier to build coalitions in the political center, a tendency that appears most clearly in the highly fractured systems of Denmark, Finland, and the Low Countries. Office-holding elites and parties will tend to shed their Euroskepticism. To win votes, parties outside the center with less chance of joining a governing coalition may emphasize Euroskeptic positions. By a kind of natural selection, then, we find Euroskeptics on either the Left or Right but not in the center, center-left, or center-right (see also Taggart 1998). Aspinwall notes that It is possible that governments become centrist (and thus pro-European) when they take office, but it is also possible that centrists are predictably elected more often. Proposition 4 is consistent with many observations in the literature, though the causal reasoning may vary. For example, it is consistent with Taggart’s (1998) claim that Euroskepticism found in parties on the periphery of their party systems and are therefore used as an issue to differentiate them from more established parties, which only express Euroskepticism through factions (see also Sitter 2001; Taggart and Szczerbiak 2001). Because centrism and Europhilia do not necessarily go together in countries lacking parliamentary coalition dynamics, such as the United Kingdom (and perhaps France). These countries fall outside the theory here, but would be anomalous for theories linking the left-right spectrum to Euroskepticism. As far as I know, no one has yet identified the presence or absence of parliamentary coalitions as a variable affecting Eurosceptic party behavior. If I am right, controlling for majoritarian parliamentary systems in studies of Euroskepticism should affect quantitative results. Taken as a whole, then, my analysis implies that Europhile governments will tend to stand against Euroskeptic and Europhobe parties, and will also be threatened by losses from their own members. Daniela Raschhofer (2001: 28) finds a similar cleavage across Europe and in the 24

European Parliament. She argues that The great majority of Conservatives, Social Democrats, Liberals and Greens stand opposed to a minority of Euroskeptic movements that are becoming ever stronger on the level of national domestic politics. This correspondingly manifests the discrepancy between Europe’s political elite and its citizens.7 We need hardly note that the first group of parties are the governing parties, with the partial exception of the Greens, while the movements Raschhofer has in mind (and of which she is a member) have been largely out of power. Thinking about Euroskepticism in strategic terms may also help explain existing puzzles. For example, the literature seems agreed that party family (or ideology) does not predict Euroskepticism. For example, the Italian Communists support the EU, while French Communists are vehemently against it; Norway’s Progress Party favors EU membership while Denmark’s party is Euroskeptic. Across Europe, Greens are split, with German and Austrian Greens favoring the EU while Scandinavians oppose it. If Euroskepticism simply reflected either ideology or political values, we would expect there to be a strong association between party stream and Euroskepticism. That we do not see this relationship suggests that national differences in the strategic environment may help explain Euroskepticism. The arguments in this section follow from the fact that the EU is different from other issues. Elites might ignore mass preferences on any issue in order to join the government and obtain the benefits of officeholding. Leaders compromise beliefs all the time as part of the coalition formation process.

7

“Der großen Mehrheit von Konservativen, Sozialdemokraten, Liberalen und Grünen steht eine Minderheit von europakritischen Bewegungen gegenüber, die auf nationalstaatlicher Ebene immer stärker werden. Zugleich zeigt sich die Diskrepanz zwischen den politischen Eliten Europas und den Bürgern.” 25

What makes the politics of the EU different is the systematic bias in favor of pro-integration parties joining the government. This stems from the fact that a pro-European position gives a domestic party some leverage not only over domestic policy but also over the policies chosen by other European countries. Party leaderships who wish to join the government therefore face systematic pressure to adopt a certain kind of policy (pro-European) and systematic pressure against other kinds of (Euroskeptic) policies. This means that mass/elite cleavages do not appear haphazardly, as they do on the EU issue in Britain, for example. Instead, mass/elite cleavages on the EU issue in parliamentary democracies will always have leaderships more integrationist than their followers.

Evidence from Austria and Slovakia In this section, I examine Austria and Slovakia in depth. My goals are, first, to test the above propositions when this is possible. Second, I seek to uncover the causal logic behind these relationships and to see how closely it matches the reasoning of the theory here. These cases were chosen to provide two multi-party parliamentary democracies with significant Euroskepticism among at least some parties. They are neighbors, one an accession candidate and the other a latecoming EU member. Slovakia is also interesting because it can boast the only Europhobe government among the EU-15 or accession candidates, namely, the HZDSSNS coalition of 1994-1998 (Kopecký and Mudde 2002: 317). They are also relatively understudied, though each has attracted significant attention in recent years for the demagogic politics of charismatic politicians, Jörg Haider and Vladimír Me…iar. The way in which these leaders and their parties have tried to reconcile their underlying hostility to the EU with the needs of coalition politics will provide a central focus for this section. 26

The EU only became an issue in Austria after membership in 1995. Joining the EU itself had enjoyed wide support after the end of the Cold War removed Russian objections to Austrian membership as a violation of neutrality enshrined in the State Treaty of 1955 (for a helpful overview of postwar Austrian politics, see Rautscher 2000). The two largest parties, the ÖVP (Österreichische Volkspartei) and SPÖ (Sozialistische, later Sozialdemokratische Partei Österreichs), both supported EU membership. They constituted one of their recurring grand coalition governments at the time of accession and pushed membership forward. The ÖVP’s membership tends to be pro-European, like the leadership. Farmers and rural voters who make up a significant part of the ÖVP’s constituency are culturally skeptic of European ideals but have been bought off by the EU’s Common Agricultural Policy (CAP). The leadership tends not to share the cultural reservations, and successfully pushed one of its own, Franz Fischler, as Agricultural Commissioner to help service this part of their political base. In contrast, the SPÖ includes a sizeable wing that suspects that labor mobility in the EU, especially after enlargement, will threaten Austrian jobs. Not surprisingly, this wing is often associated with the trade unions and with the chambers of labor. The leadership has been consistently pro-European. This division is consistent with Proposition 2's claims about the nature of elite/mass cleavages on European integration. Austria’s most Euroskeptic party is the Freedom Party, the FPÖ (Freiheitliche Partei Österreichs). It has traditionally consisted of two wings, a centrist, market-oriented wing not unlike the German FPD, and a xenophobic right wing that traces its intellectual heritage to the Nazis and to the German nationalist parties before them (Bailer et al 2000; Betz 2002; Ötsch 2000: Chapter 3; Schorske 1961; Wodak 2000). The right wing has dominated the party since Jörg 27

Haider took over the leadership of the party in 1986, in a strong rejection of Norbert Steger’s policy of working with the SPÖ in government (see Plasser and Ulam 2000; Zöchling 1999). His politics of resentment served the party well as Austria underwent the changes of the post-Cold War era, with increased immigration from its eastern neighbors and greater exposure to the economic uncertainty of globalization. Haider also successfully centralized the party, eliminating rivals to his leadership and purging many members of the centrist wing who ultimately formed a new Liberal Forum (LiF) as an (unsuccessful) alternative to FPÖ. Despite the party’s pro-European traditions, he also moved to a position against the European Union in 1993, resonating with traditional Austrian concerns about economic security over growth (Rautscher 2000). In the late 1990s, workers who opposed the EU had to turn to one of two non-governing parties, the Greens or the FPÖ. The SPÖ lost votes to the FPÖ in particular, who could successfully exploit themes of opposition, renewal, reform and “Wende” (a change of direction) in Austria (Betz 2002). It succeeded far better than the party could have hoped for. Daniela Raschhofer (2001: 25), an MEP for the FPÖ, argues that “In the national election of October 1999, the voters carried out something that amounts to the essence of democracy: a change [of government] that had previously been impossible in Austria.”8 The change was possible because, when out of government, the FPÖ could criticize a series of policies associated with the EU, including immigration. The Austrian case is particularly instructive since any interpretation of politics on a purely Left-Right dimension would expect SPÖ dissidents to register their protests by voting for the Greens instead of leaping rightward over the ÖVP and landing on the FPÖ. However, the Austrian

8

“Mit der Nationalsratswahl im Oktober 1999 setzten die Wähler das durch, was das Wesen einer Demokratie ausmacht: den Wechsel, der in Österreich bislang unmöglich schien.” 28

Greens generally supports European integration. In the analysis here, this made the Greens unattractive for Euroskeptic masses in the SPÖ electorate, who voted for the FPÖ instead.9 In this respect, the Austrian Social Democrats face a challenge very different than that found in Denmark or Sweden. In 1999, the FPÖ managed to earn (barely) more votes than the ÖVP and temporarily became the second-largest party in the country. Because the SPÖ rejected any coalition with the FPÖ, while both domestic and international actors were opposed to a government with Haider at its head, the ÖVP and FPÖ eventually agreed on a government in which the ÖVP would serve as the senior member though it was marginally the smaller party. Wolfgang Schüssel of the ÖVP became Chancellor, and Haider remained formally outside the cabinet though the FPÖ’s members tended to be close associates of his (for the time being).10 Given widespread concern over this government, Austrian President Thomas Klestil—a member of the ÖVP with a long-standing dislike of both Schüssel and Haider—required that the coalition treaty include a preamble reaffirming core principles of Austria’s political order. Alongside democracy, human rights, and nondiscrimination, the preamble committed both parties to continued membership of the European Union and support for enlargement. Including a

9

However, many professionals and middle-class SPÖ loyalists have moved to the Greens, in part out of frustration with the SPÖ’s moves toward Haider’s anti-immigration position. In 1999, 58% of Green voters said that they had chosen this party “because it is engaged against hostility to immigrants and against right radicalism” [weil sie engagiert gegen Ausländerfeindlichkeit und gegen Rechtsradikalismus eintrete] (Pelinka 2000: 64). Green leader Alexander van der Bellen is also highly regarded by many Austrians across the political spectrum, and is often identified as the most respected politician in Austria. 10

Formation of this government led to “sanctions” against Austria freezing bilateral political ties between the other fourteen member states and the Austrian government, but not affecting the participation of Austria or Austrians in EU organs. These sanctions provide a notable subject in themselves but lie outside the focus here. 29

previously Euroskeptic party in the government raised new challenges for Austria and forced that party to compromise its policies on that issue.11 The central role of the EU in debates over the FPÖ is consistent with the logic of Propositions 5-7. The FPÖ faces the kind of divisions between its electorate and some of its leadership as the theory would expect. These divisions have racked the party internally because Haider remained outside the government and (formally) renounced the leadership of the party, while more pragmatic elites took over important roles in the government. Ironically, these divisions mirror those in the SPÖ, from which the FPÖ seems to have taken many of its more fickle voters in the 1999 election. Amusingly enough, the FPÖ’s cabinet members seem to have “gone native” while in government, moving toward many of the establishment policies that their party had long decried out of office. From his position as prime minister of the state of Carinthia (Kärnten), Haider was free to criticize his own ministers as sell-outs on many issues, including European questions. For Haider, the EU became an arch-enemy that worked closely with domestic symphathizers demonstrating against the government every Thursday (Ötsch 2000: Chapter 5). If we assume, quite reasonably, that these cabinet members valued officeholding to some degree, it is not surprising that a cleavage would emerge between them and non-office-holders such as Haider and much of the mass electorate (see Proposition 8). These internal FPÖ divisions eventually drove the government apart in 2002, as Schüssel decided it would be propitious to humble the FPÖ in new elections.

11

Wags might add that the coalition agreement also forced Haider’s FPÖ to compromise its principles on democracy, human rights, and nondiscrimination, but we need draw no such conclusion here. 30

Events justified Schüssel’s tactics. After fighting a mistake-prone campaign12 in which it emphasized its hostility to foreigners, immigrants, and the Czech Republic, the FPÖ lost well over half of its support. Its protest voters left the FPÖ when it joined government, and as Schüssel expected, those FPÖ voters embarrassed by Haider’s regular verbal slipups on Nazism (NaziAusrutscher) well outnumber those Austrians appealed by them (Rauscher 2000: 23-24). At the same time, the ÖVP leaped past both of its major rivals to become the largest party in Austria for the first time in decades. Having compromised its principles on Europe (and other issues), the FPÖ could no longer appeal to SPÖ working-class voters on that issue. Out of government, the SPÖ also gained some flexibility on the EU issue that helped paper over its internal divisions. As a result, FPÖ support moved most of the way back to its base level of under ten percent. The new government is a significant humbling of the FPÖ, with an “unacceptable” turncoat as Finance Minister, Karl-Heinz Grasser, and FPÖ ministers relegated almost entirely to minor ministries such as culture or sport. The Economist (1 March 2003, p. 48) describes FPÖ participation in terms that treat office-holding and the EU issue as two different issues: with his [Haider’s] influence down, the main aim of his disgruntled colleagues is, it seems to cling to whatever jobs in government they may be offered. Moreover, though many Freedom Party voters remain xenophobic and hostile to the EU, the party itself no longer threatens to veto the EU’s enlargement to the east, nor could it do so if it tried. Of course, the theory here links the office-holding motive to the leadership’s changed position on the EU, at least in part.

12

The party had trouble finding someone willing to lead the party in the campaign, for it had expelled four of its promising cabinet members and many other promising candidates hesitated to lead the party into an electoral debacle. The first peak candidate eventually had a nervous breakdown during the campaign and had to withdraw in favor of Haupt, an unimpressive functionary. 31

These events illustrate much of the logic behind the theory here, though the political color and tactical details obviously fall outside any theory. The EU is associated with internal party cleavages in the direction we expect, and we do not observe the anomalous pattern of a Euroskeptic leadership and pro-integration membership. A Euroskeptic party faced significant hurdles before joining the government, which required caving in on the EU issue and ultimately produced devastating divisions within the party.

Slovakia has a much more complex party system than Austria.13 The government of 19982002, essentially reelected in 2002, is a broad coalition of many pro-democracy parties from across the political spectrum. These governing parties can be classified easily on the conventional leftright dimension found in Europe and elsewhere, and these parties are generally members of the corresponding international party federations such as the Socialist or Liberal International. They were joined in 2002 by ANO (Alliance of the New Citizen), a new and somewhat unconventional media-based party of the center-right.14 The opposition consists of right-wing parties HZDS (Movement for a Democratic Slovakia), led by Vladimír Me…iar, prime minister in 1994-1998, and the SNS (Slovak Nationalist Party). Another unconventional party, Smer (“Direction), is led by Robert Fico. It is often labeled rightist but includes elements from the old democratic left (SD´, Party of the Democratic Left).

13

Much of this information comes from interviews in Bratislava in June 2001 and June 2002; see Mesežnikov 2002 for an overview of Slovak politics. 14

The array of governing parties can be bewildering. In 1998-2002, the largest party, SDK, was in fact a coalition of four parties (the KDH, DS, LDÚ, and SZS) who gradually left the party coalition (but not the government), leaving a rump SDKÚ in the place of the former SDK. It was joined in government by the SD´, SMK, and SOP. For full names and descriptions, see Mesežnikov 2002. 32

These unconventional parties use rhetoric contemptuous of “politics as usual” and often use demagogic techniques of political mobilization that rely heavily on nationalistic appeals.15 All parties except SNS officially support Slovak membership in both the EU and NATO, and all give much stronger support for the EU than for NATO. However, it is not clear how sincere this official position is in the case of HZDS, and its behavior in office was very different than the policies of the Dzurinda coalitions (Bil…ík 2001). I would classify SNS as Europhobic, HZDS as highly Euroskeptic, SD´ and Smer as Euroskeptics. As in other countries, mass support for the EU is not nearly as strong as a survey of party positions would expect. In a young and unstable party system such as Slovakia, this means that dissidents regularly defect from parties to form new parties. Each of these parties faces some version of the elite/mass cleavage identified in Proposition 3. When the EU is an issue, it always has effects in the direction that we would expect. As the HZDS leadership tried to convince Slovaks and outsiders of the party’s support for the EU in the lead-up to the 2002 elections, it stirred up domestic Euroskepticism and Europhobia. Party leaders such as Me…iar have been willing to take a more European line since it has become clear that support for the EU is a precondition for joining the government. The mass membership of the party has been much less likely to change its views in response to such considerations. HZDS’s efforts to become acceptable for participation in government therefore came at heavy internal cost, ultimately producing a split in the party during the lead-up to elections. Though acceptable as a coalition partner, Smer’s opposition role and “alternative” rhetoric

15

Smer fits uneasily here, much less nationalistic than the other two, and much more willing to take concrete and even difficult policy stances (see http://www.strana-smer.sk/program). As will become clear in the rest of this section, Smer is also much more acceptable to the conventional parties than are the HZDS and SNS. 33

allow it to appeal to Euroskeptics.16 Like many opposition politicians in other accession countries (see Szczerbiak 2001 for Poland), Fico argued for a harder bargaining stance, with a list of “untouchable national interests.” He argued for example that “the strategy of accession to these institutions, as pursued by Dzurinda’s cabinet, is not self-confident enough. And it looks as though we are losing our dignity” (quoted in Mesežnikov 2002: 88). The Christian Democratic Movement was originally a center-right reformist Christian Democratic party. Its leader, „arnogurský, was too unpopular to lead the Slovak Democratic Coalition, so the party’s vice-chair Dzurinda took the premiership. This led to a split in the Movement, with Dzurinda remaining in the Slovak Democratic Coalition and then joing the coalition’s second-largest party to form the Slovak Democratic and Christian Union. The CDM continued to exist with a new leader and is veering toward Catholic fundamentalism and Euroskepticism. As these examples suggest, parties consistently tend to lose factions in a Euroskeptic direction. There do not seem to be contradictory examples of any importance, in which Europhiles left a party because of its Euroskeptic or Europhobic drift, though one could imagine that the HZDS, in particular, might have suffered that fate. Negotiations with the EU stagnated under the HZDS in 1994-1998, when Me…iar was increasingly accused of non-democratic behavior such as firing protected civil servants, violating some civic rights, working closely with organized crime, and even kidnaping the president’s son. Quite aside from not having met the Copenhagen criteria that candidate countries have established stable democracies with market economies, HZDS’s strongly Euroskeptic preferences account for

16

Fico’s former party, the SD´, also appeals to a substantial number of Euroskeptic voters, though it too supports EU membership (Henderson 2001: 21). 34

Slovakia’s failure to join the first wave of candidates. Me…iar did yield power democratically, however, and the next government pursued EU membership with gusto. In the summer of 2001, it had caught up with its former compatriots in the Czech Republic in terms of the number of EU negotiating chapters closed. As the theory here would expect, the result was to put pressure on Euroskeptic opposition parties to accept existing integration. Interestingly, events next door in Vienna heightened the importance of positions in favor of the European Union and against racism and demagogy. Candidate members had previously thought of the Copenhagen requirements of democratic governance in terms of various basic laws but now saw them in larger terms as a matter of democratic values. One member of the Dzurinda government said that “We are grateful to the 14 states for this decision”17 (cited in Martin 2000: 19). Not surprisingly, then, much of the politics around the election of 2002 concerned which parties were acceptable to the governing parties and to the EU. During the campaign, Fico categorically excluded a coalition with Me…iar, saying that the HZDS with Me…iar has “no coalition potential.” Interestingly, he admitted some of the officeholding motives behind his reasoning. Entering a coalition with a man who is “absolutely unacceptable” to Europe would, Fico said, end his own political career at the young age of 37 (Haughton 2002: 5). Fico also emphasized this position in visits with American congressmen, and with Slovak President Rudolf Schuster. Thanks to such statements, Smer was probably acceptable as a coalition partner in 2002. However, its weaker-than-expected electoral performance made it a less attractive partner than

17

“Wir sind den 14 Staaten für diesen Beschluss dankbar.” 35

many had anticipated. HZDS’s support of the EU was believed to be less credible, and neither it nor SNS were acceptable partners for the existing government. Yet HZDS’s efforts to be accepted came at heavy internal cost, ultimately producing a split in the party during the lead-up to elections. After the doubts of the mid-1990s, reelection of the Dzurinda coalition government in 2002 confirmed Slovakia’s position in the mainstream of post-communist democracies. The division between coalitionable, pro-EU parties seems likely to persist. Smer will probably remain in an ambiguous position, theoretically acceptable as a potential coalition partner at the elite level, despite its Euroskepticism, but also a magnet for opposition to the EU at the mass level. More overtly nationalist opponents of the EU, especially the SNS and (ambiguously) the HZDS, will be condemned to opposition roles. These patterns also appeared in the referendum campaign of 2002. Though Slovakia achieved the highest “yes” vote of any country (over 90%), there was a widespread lack of enthusiasm for EU entry. In this environment, the central challenge was reaching 50% turnout, as required for a referendum to be valid. Dzurinda argued that the official unity across the political spectrum even encouraged Euroskeptic factions to try to depress turnout in a “sneaky” (rafinovane) way (“Dzurinda vyhlásil”).

Conclusions Both Austria and Slovakia illustrate the domestic political dynamics highlighted by the theory here. Euroskeptic parties find it more difficult to join government, and Europhobic parties find it impossible. Euroskeptic parties that do join the government face internal splits and loss of support to parties outside government. Parties that have been hostile to the EU find it necessary to establish a willingness to work within EU institutions and policies if they wish to join the 36

government. I have argued that that elite-mass differences of opinion are contingent on the political system, and should not be found in majoritarian systems. Controlling for this feature of the political system in future analyses of Euroskepticism would be a natural direction for quantitative research. Several of my propositions show how European integration changes domestic politics. This poses a rebuttal to, among others, Mair’s (2000) claim that Europe has a limited effect on national party systems. The bulk of my propositions focus on the growing divide between prointegration governing elites and those dissidents both in the governing parties and in the opposition. The result of the processes that I identify is a narrowing of political perspectives within government and within the spectrum of political parties in opposition. This narrowing highlights differences with the broader range of views among the public and doubtless contributes to public alienation from the political establishment. One could criticize this effect of the European Union as undemocratic in shutting out certain mainstream political views—namely, skepticism about the European project—from participating in government. Indeed, one sees the EU’s institutional hostility to Euroskeptic parties in the case of the sanctions against Austria. As one might expect, the Austrian government responded in terms reasserting the rules of democratic politics and the need to accept the will of the people. The government even proposed a “referendum” against the sanctions that sought to teach the EU-14 something about popular sovereignty (cited in Fleischhacker 2001: 32-33): In the course of the coming reforms of the EU treaty, should the federal government secure, with all available means, that . . . the European Union will guarantee the fundamental right of every country to choose its own government in free elections, and protect the rights of all democratic parties as well as the instruments of direct

37

democracy . . . 18 Viewed differently, the familiar “democratic deficit” of the European Union appears not only in Brussels but in stifling certain aspects of national-level politics. These dynamics doubtless contribute to the extra-establishment opposition to the pro-EU mainstream that we have seen across Europe in recent years, including the growth of ultranationalist or anti-immigration parties in Belgium and Denmark, the rise of Pim Fortuyn in the Netherlands, and Le Pen’s first-round success in France’s presidential race. Moreover, this theory predicts that these cleavages are endemic to parliamentary government formation inside the European Union, for as long as the EU itself remains a political issue. Only if (con)federation becomes as accepted as in Switzerland, Canada, Australia or the United States will this cleavage likely wither away. In this case, one might imagine the anti-foreign views currently being directed at Brussels might be directed elsewhere, probably at the United States. Denmark’s Left Socialists already seem to have begun to move in this direction, but they remain exceptional. As long as government formation is organized at the national level and not the supranational level, the patterns of Euroskeptic parties identified here should remain a feature of European politics.

18

“Soll die Bundesregierung im Zuge der bevorstehenden Reform des EU-Vertrages mit allen geeigneten Mitteln sicherstellen, dass . . . . C die Europäische Union das Grundrecht jedes Landes auf freie demokratische Wahl seiner Regierung garantiert und den freien Wettbewerb und die Rechte aller demokratischen Parteien sowie Einrichtungen der direkten Demokratie achtet . . . .” 38

REFERENCES Anderson, Christopher J. and Karl C. Kaltenthaler. 1996. “The Dynamics of Public Opinion toward European Integration, 1973-93.” European Journal of International Relations 2(2): 175-199 (June). Aspinwall, Mark. 2002. “Preferring Europe: Ideology and National Preferences on European Integration.” European Union Politics 3(1): 81-111 (March). Bailer, Brigitte, Wolfgang Neugebaur, and Heribert Schiedel. 2000. “Die FPÖ auf dem Weg zur Regierungspartei. Zur Erfolgsgeschichte einer rechtsextremen Partei.” in Haider: Österreich und die rechte Versuchung, edited by Hans-Henning Scharsach. Reinbek bei Hamburg: Rowolt Taschenbuch Verlag, pp. 105-127. Betz, Hans-Georg. 2002. “Haider’s Revolution or The Future Has Just Begun.” in Austria in the European Union, Günter Bischof, Anton Pelinka, and Michael Gehler, eds. (Contemporary Austrian Studies Volume 10.) New Brunswick (US) and London: Transaction Publishers, pp. 118-143. Bil…ík, Vladimír. 2001. “Slovenská cesta do Európskej únie—vývin po voµbach 1998.” Integrace, http://integrace.cz/integrace. Crombez, Christophe. 2002. “Information, Lobbying and the Legislative Process in the European Union.” European Union Politics 3(1): 7-32 (March). Dogan, Mattei and Robert Pahre. 1990. Creative Marginality: Innovation at the Intersections of Social Sciences. Boulder, Colorado: Westview Press. “Dzurinda vyhlásil, že predreferendovú kampaÁ vyhodnotí chladnokrvne.” Sme online, 18 May 2003, http://www.sme.sk/clanok.asp?cl=982481. Eichenberg, Richard C. and Russell J. Dalton. 1993. "Europeans and the European Community: The Dynamics of Public Support for European Integration." International Organization 47(4): 507-534 (Autumn). “EU skal være garanten for demokrati.” LO Ugebrev, 12 May 2003, http://www.ugebreveta4.dk. Franklin, Mark, Michael Marsh, and Lauren McLaren. 1994. “Uncorking the Bottle: Popular Opposition to European Unification in the Wake of Maastricht.” Journal of Common Market Studies 32(4): 455-472. Gabel, Matthew J. 1998. Interests and Integration. Market Liberalization, Public Opinion, and European Union. Ann Arbor: The University of Michigan Press. Haider, Jörg. 1993. Die Freiheit, die ich meine. Das Ende des Proporzstaates. Plädoyer für die Dritte Republik. Frankfurt am Main. Hardin, Russell. 1976. “Hollow Victory: The Minimum Winning Coalition.” American Political Science Review 70(4) 1202-1214 (December). Haughton, Tim. 2002. “Slovakia’s Robert Fico: A Man to be Trusted or Feared?” RFE/RL East European Perspectives 4(1), 29 May 2002, http://wwww.rfel.org. Hinich, Melvin J. and Michael C. Munger. 1994. Ideology and the Theory of Political Choice. Ann Arbor: University of Michigan Press. Hooghe, Liesbet. 2003. “Europe Divided? Elites vs. Public Opinion on European Integration.” European Union Politics 4(3): 281-304 (September). Hosli, Madeleine O. 1993. “Admission of European Free Trade Association States to the European Community: Effects on Voting Power in the European Community Council of Ministers.” International Organization 47(4): 629-643 (Autumn). Hosli, Madeleine O. 1995. “The Balance Between Small and Large: Effects of a Double-Majority 39

System on Voting Power in the European Union.” International Studies Quarterly 39(3):351-370 (September). Hosli, Madeleine O. 1996. “Coalitions and Power: Effects of Qualified Majority Voting in the Council of the European Union.” Journal of Common Market Studies 34(2): 255-273 (June). Hug, Simon and Thomas König. 1999. “Ratifying Amsterdam.” Paper prepared for presentation at the Annual Meeting of the American Political Science Association, Atlanta, September 2-5, 1999. Hughes, James, Gwendolyn Sasse, and Claire Gordon. 2002. “Saying ‘Maybe’ to the ‘Return to Europe.’ Elites and the Political Space for Euroscepticism in Central and Eastern Europe.” European Union Politics 3(3): 327-355 (September). König, Thomas and Simon Hug. 2000. “Ratifying Maastricht: Parliamentary Votes on International Treaties and Theoretical Solution Concepts.” European Union Politics 1(1): 93-124. König, Thomas and Mirja Pöter. 2001. “Examining the EU Legislative Process: The Relative Importance of Agenda and Veto Power.” MS: University of Konstanz. Kopecký, Petr and Cas Mudde. 2002. “The Two Sides of Euroscepticism: Party Positions on European Integration in East Central Europe.” European Union Politics 3(3): 297-326 (September). Lakatos, Imre. 1970. "The Methodology of Scientific Research Programmes," in Criticism and the Growth of Knowledge, edited by Imre Lakatos and Alan Musgrave, pp. 91-196. Cambridge: Cambridge University Press. Laver, Michael and Kenneth Shepsle. 1990. “Coalitions and Cabinet Government.” American Political Science Review 84:873-90. Mair, Peter. 2000. “The Limited Impact of Europe on National Party Systems.” West European Politics 23(4): 27-51 (October). Marks, Gary, Carole J. Wilson and Leonard Ray. 2002. “National Political Parties and European Integration.” American Journal of Political Science 46(3): 585-594 (July). Martin, Hans-Peter. 2000. “Die Schockwelle in Europa.” in Haider: Österreich und die rechte Versuchung, edited by Hans-Henning Scharsach. Reinbek bei Hamburg: Rowolt Taschenbuch Verlag, pp. 13-21. Martin, Lanny W. and Stevenson, Randolph T. (2001) ‘Government Formation in Parliamentary Democracies,’ American Journal of Political Science 45(1): 33-50. Mesežnikov, Grigorij. 2002. “Domestic Politics.” in Slovakia 2001: A Global Report on the State of Society, edited by Grigorij Mesežnikov, Miroslav Kollár, and Tom Nicolson. Bratislava: Institute for Public Affairs, pp. 21-92. Moser, Peter. 1996a. “The European Parliament as a Conditional Agenda Setter: What Are the Conditions? A Critique of Tsebelis (1994).” American Political Science Review 90(4): 834-838 (December). Moser, Peter. 1996b. “A Theory of the Conditional Influence of the European Parliament in the Cooperation Procedure.” Twente, Netherlands: Network on Enlargement and new Membership of the European Union Working Paper No. 96-1. Noury, Abdul G. 2002. “Ideology, Nationality and Euro-Parliamentarians.” European Union Politics 3(1): 33-58 (March). Ötsch, Walter. 2000. Haider Light: Handbuch für Demagogie. Wien: Czernin Verlag. 40

Pahre, Robert. 1996. "Mathematical Discourse and Crossdisciplinary Communities: The Case of Political Economy," Social Epistemology 10(1):55-73. Pahre, Robert. 1997. "Endogenous Domestic Institutions in Two-Level Games: Parliamentary Oversight in Denmark and Elsewhere." Journal of Conflict Resolution 41(1):147-174 (February). Pahre, Robert. 1998. “Reactions and Reciprocity: Tariffs and Trade Liberalization in 1815-1914.” Journal of Conflict Resolution 42(4): 467-492 (August). Pahre, Robert. 2001. “Divided Government and International Cooperation in Austria-Hungary, Sweden-Norway, and the European Union,” European Union Politics 2(2): 131-162 (Spring). Pelinka, Anton. 2000. “Die rechte Versuchung. SPÖ, ÖVP und die Folgen eines falschen Tabus.” in Haider: Österreich und die rechte Versuchung, edited by Hans-Henning Scharsach. Reinbek bei Hamburg: Rowolt Taschenbuch Verlag, pp. 46-66. Pennings, Paul. 2002. “The Dimensionality of the EU Policy Space. The European Elections of 1999.” European Union Politics 3(1): 59-80 (March). Plasser, Fritz and Peter A. Ulam. 2000. “Protest ohne Parteibindung. Die Wählerschaft der FPÖ.” in Haider: Österreich und die rechte Versuchung, edited by Hans-Henning Scharsach. Reinbek bei Hamburg: Rowolt Taschenbuch Verlag, pp. 128-143. Putnam, Robert D. 1988. "Diplomacy and Domestic Politics: The Logic of Two-Level Games." International Organization 42(3): 427-60 (Summer). Raschhofer, Daniela. 2001. “Die FPÖ in Europa. Ziele und Parteiprogram.” in Österreich in Europa, edited by Carl Paul Wieland. Wien: Amalthea, pp. 25-32. Rauscher, Hans. 2000. “Ein geschlossene Verdrängungskette. Warum Österreichs Konservative der Haider-Partei zur Regierungsmacht verhalfen.” in Haider: Österreich und die rechte Versuchung, edited by Hans-Henning Scharsach. Reinbek bei Hamburg: Rowolt Taschenbuch Verlag, pp. 22-45. Ray, Leonard. 2003. “Reconsidering the Link between Incumbent Support and Pro-EU Opinion.” European Union Politics 4(3): 259-279 (September). Rodrik, Dani. 1997. Has Globalization Gone Too Far? Washington: Institute for International Economics. “The same peculiar team?” Economist 1 March 2003, p. 48. Scharsach, Hans-Henning, ed. 2000. Haider: Österreich und die rechte Versuchung. Reinbek bei Hamburg: Rowolt Taschenbuch Verlag. Schorske, Carl E. 1961. Fin-de-siècle Vienna: Politics and Culture. New York: Vintage. Sitter, Nick. 2001. “The Politics of Opposition and European Integration in Scandinavia: Is EuroScepticism a Government-Opposition Dynamic?” West European Politics 24(4): 22-39 (October). Strom, Kaare. 1990. Minority Government and Majority Rule. Cambridge: Cambridge University Press. Smith, Alastair and David R. Hayes. 1997. “The Shadow of the Polls: Electoral Effects on International Agreements.” International Interactions 23(1): 27-108. Strobel, Georg W. 2001. “Das andere Polen. Struktur und Selbstverständnis der rechten und rechtsextremen Kräfte in der polnischen Politik.” Osteuropa 51(3): 259-280 (March). Svensson, Palle. 2002. “Five Danish Referendums on the European Community and European Union: A Critical Assessment of the Franklin Thesis.” European Journal of Political 41

Research 41(6): 733-50. Taggart, Paul. 1998. “A Touchstone of Dissent: Euroscepticism in Contemporary Western European Party Systems.” European Journal of Political Research 33(3): 363-388 (April). Thomsen, N. 1993. “EF som stridpunkt i dansk politik 1972-1979.” in The odd man out? Danmark og den europæiske integration, edited by B. N. Thomsen. Odense: Odense Universitetsforlag, pp. 153-178. Tolstoy, Leo. 2000. Anna Karenina, Constance Garnett trans. New York: The Modern Library. Tsebelis, George and Geoffrey Garrett. 2000. “Legislative Politics in the European Union.” European Union Politics 1(1): 9-36. Wodak, Ruth. 2000. “«Echt, anständig und ordentlich.» Wie Jörg Haider und die FPÖ Österreichs Vergangenheit, Gegenwart und Zukunft beurteilen.” in Haider: Österreich und die rechte Versuchung, edited by Hans-Henning Scharsach. Reinbek bei Hamburg: Rowolt Taschenbuch Verlag, pp. 180-187. Zöchling, Christa. 1999. Haider: Eine Karriere. Munich: Econ Taschenbuch Verlag.

42