736808 EEPXXX10.1177/0888325417736808East European Politics and SocietiesCasal Bértoa and Enyedi / Post-communist Party Systems research-article2017 Institutionalization and De-institutionalization in Post-communist Party Systems East European Politics and Societies and Cultures Volume 32 Number 3 August 2018 422–450 © 2018 SAGE Publications 10.1177/0888325417736808 journals.sagepub.com/home/eep hosted at http://online.sagepub.com https://doi.org/ Zsolt Enyedi Central European University, Budapest, Hungary Fernando Casal Bértoa University of Nottingham, Nottingham, United Kingdom In an article written in 1995 titled “What Is Different about Postcommunist Party Systems?” Peter Mair applied the method that he called “ex adverso extrapolation.” He matched his knowledge of the process of consolidation of party systems in the West with what was known at that time about Eastern European history, society, and the emerging post-communist party politics. Considering factors such as the existence of fluid social structures, the weakness of civil society, or the destabilizing impact of the so-called triple transition, his article predicted long-term instability for the region. In the present article, we evaluate the validity of Mair’s predictions, thereby also contributing to a lively debate in the current literature about the scale and nature of East–West differences and about the trajectories of the two regions. Going beyond the identification of cross-regional similarities and differences, we also differentiate between individual party systems, establish subgroups, and describe changes across time. Using four major dimensions (i.e., party system closure, party-level stability, electoral volatility, and fragmentation), the article finds that Mair’s predictions were largely, though not in every detail, right. Ironically, however, we also find that changes in the West tend to match over time the trajectory of the East. Keywords: party system institutionalization, closure, fragmentation, volatility, Eastern Europe, Western Europe Introduction In many spheres of life, well-structured, predictable patterns of behavior develop because nonstandard actions entail high risks. In an ideal-typical market situation, for example, offering products well above or well below the established price can turn out to be equally suicidal. In such situations, the negative feedback mechanisms encourage the convergence of the individual strategies, and the micro-actions add up to a system-level equilibrium. Both authors have contributed equally. This is one of several joint works and the ordering of names simply reflects a principle of rotation. Casal Bértoa and Enyedi / Post-communist Party Systems 423 In the more formalized environments, like in the world of political institutions, the actors are more obviously constrained by rules, procedures, and codified and noncodified norms. The negative feedback mechanisms are often replaced by positive reinforcements, and the hand that creates order is typically more visible than in the case of markets. But notwithstanding these differences, factors leading to order and stability dominate both types of situations.1 Party systems are neither the products of millions of autonomous decisions that cancel out each other, nor outcomes of written rules and crystallized norms. They are, of course, under the influence of the electoral choices of a large number of voters and are constrained by procedural rules such as the electoral system, but they are ultimately shaped by the behavior of a limited set of elite actors who base their decisions on a mixture of rational and normative considerations. Under such circumstances, it is a particularly intriguing question how much order will characterize the relationships within the system. Whether one considers market or state analogies or whether one compares party systems to games consisting of repeated interactions between a small number of players or rather to biological organisms, the passing of time is typically predicted to stabilize party strategies and interaction patterns. Party systems are expected to “mature” with time, and parties are anticipated to develop routine reactions toward each other. The vast literature on institutionalization of party systems (primarily Mainwaring and his coauthors, see below) warns us, however, that the evolutionary path toward stabilization cannot be taken for granted. Individual countries, regions, and time periods differ substantially in the degree of the stability and predictability of party systems. Often it is even doubtful whether the parties of a particular country constitute a “system” in the sense of operating an enduring configuration that goes beyond the coexistence of a set of parties. Post-communist Europe is a particularly interesting region from the point of view of institutionalization as it is under the influence of a set of countervailing factors. Its proximity to Western Europe, its relatively high level of social modernization, the membership of most of its countries in the European Union, and its party-centered model of political representation are supposed to foster institutionalization. The communist past, the weakness of mass political organizations, the volatile post-communist economic history, and the leader- and media-centered political culture, however, are likely to hinder the very same processes. The probably most influential early study on the region predicted long-term instability. In an article written in 1995 titled “What Is Different about Post-communist Party Systems?”2 Peter Mair applied the method that he called “ex adverso extrapolation.” He matched his knowledge of the process of consolidation of party systems in the West with what was known at that time about Eastern European history, society, and the emerging post-Communist party politics. Considering factors such as the existence of fluid social structures, the weakness of civil society, or the destabilizing 424 East European Politics and Societies and Cultures impact of the so-called triple transition, his article predicted continuous instability for the region. His approach contradicted other scholars who foresaw the emergence of more or less structured patterns of inter-party competition around certain (e.g., economic, ethnic), somehow strong, cleavages in a reduced number of countries (e.g., the Czech Republic, Slovakia).3 The peculiar sequence of political modernization in Eastern Europe provided Mair with additional reasons for skepticism concerning stabilization. While by the 1990s, all the elements of political modernity, such as incorporation, mobilization, activation, politicization, and contestation, were in place in Eastern Europe, partybased competition appeared as the last stage of the development, unlike in Western Europe, where it preceded democratization. The difference in the sequence of these steps of modernization was expected to have substantial consequences for the shape of party-politics. In the technologically advanced social environment of the 1990s, old-style mass party strategies were no longer available. As a result, the organizational networks that had been so crucial for anchoring voters in the West were substituted, or at least dominated, by top–down structured political parties. In Mair’s opinion, the fragility of collective identities, the general alienation of voters from party politics, and the overlap between social movements, parties, and interest groups would further hinder the prospects of the institutionalization of party systems. In the present article, we evaluate the validity of Mair’s predictions, thereby also contributing to a lively debate in the current literature about the scale and nature of East–West differences and about the trajectories of the two regions. Going beyond the identification of cross-regional similarities and differences, we will contrast individual party systems with each other, establish subgroups within the region, and describe change across time. Party system institutionalization became a prominent topic at the turn of the century, primarily thanks to the work of Scott Mainwaring,4 either alone (1998) or with Scully or Torcal or Zoco.5 As a result of these studies, the level of institutionalization is often regarded today as one of the fundamental dimensions of party systems. While the importance of institutionalization is rarely questioned, there is considerable diversity in the literature about how we should conceptualize and measure this aspect of party politics. Our approach, based on the works of Giovanni Sartori and Peter Mair, is rather straightforward: because the essence of party systems is defined by the relations among parties, the institutionalization of the system should also be applied primarily to these relationships.6 Party systems are institutionalized if the relations among parties are stable. This stability and predictability is well captured by another concept of Peter Mair’s, “party system closure,” to be discussed in detail below.7 The stability of party relations is profoundly shaped by three further major phenomena of party politics. The first is the strength and the stability of the parties themselves. This aspect will be gauged in the present study at three levels: party continuity, the strength of party organizations in terms of party membership, and the Casal Bértoa and Enyedi / Post-communist Party Systems 425 assessment of the overlap between parties and interest groups. The second factor is the stability of voting behavior, represented here by net electoral volatility. Finally, the third aspect is the fragmentation of the party system, measured by the index of effective parties. Countries and time periods that are characterized by large electoral volatility, low organizational stability of parties, and high fragmentation are expected to be, both for simple mechanical and for more substantive causal reasons, low on party system closure. By considering all these factors, we hope to provide a rather comprehensive picture of the cross-regional and cross-country differences in Europe three decades after the collapse of Communism. Next to our fundamentally descriptive goals, we also intend to evaluate the claim that post-communist party systems, because of the specific conditions under which they emerged, are bound to remain fundamentally more open, volatile, and unstructured than the Western European party systems. In line with the argument that the stabilization of party relations requires a period of predemocratic organization building, we expect party system closure to be substantially lower in Eastern Europe than in Western Europe. Furthermore, while acknowledging that the pre-communist history of the countries of the region differed substantially, we expect no major differences between them given their long communist common history. To the extent that the cases are clustered into distinct groups, we expect the group boundaries to either follow historical trajectories, meaning primarily the differences classified as national-accommodative (e.g., Hungary, Poland), patrimonial (e.g., Romania), and bureaucratic-authoritarian (e.g., Czech Republic) by Kitschelt and his colleagues8 or to reflect on the country’s relationship to the European Union.9 The countries with the longest EU membership and with the most modern and bureaucratic background are expected to be the most institutionalized. The study covers fourteen post-communist democracies (Bulgaria, Croatia, the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Montenegro, Poland, Romania, Serbia, Slovakia, Slovenia, and Ukraine) and contrasts them to Western European averages.10 We will only consider those periods when the country obtained a score of 2.5 or less in the Freedom House Index. For some of the analyses, only eleven of the listed cases will be presented, because of data limitations. Party System Closure: Conceptualization and Operationalization The central concept of this section, party system closure, taps the degree of stability of party relations from the point of view of coalition patterns. Coalitions reflect the cooperative and competitive relations among parties. Mair identified three major components of party system closure: the form of government alternation, the degree of the access to office, and the innovative nature of the coalition formulae. Together, these factors show whether a party system is closed, predictable or open, inchoate.11 426 East European Politics and Societies and Cultures The first dimension is alternation in government. As far as this dimension is concerned, three patterns exist: wholesale, partial, and nonalternation. In the first case, the incumbent government leaves the office in its entirety and is replaced by a completely different group of parties (or a single party). The second option materializes when the new cabinet is a mixture of parties from both inside and outside of the previous government. The third possibility is marked by a complete absence of alternation, as the same party or parties remain in exclusive control of government over an extended period of time. The second major component of the model, governing formulae, shows whether the partisan composition of the governments is innovative or familiar. Familiarity prevails if there are stable groups of parties that tend to govern together. If there is a tendency toward previously unseen party compositions of cabinets, the system is considered to be open, innovative. The final component is access to government. The relevant difference on this aspect is between, on one hand, systems in which all parties have the opportunity to participate in government and, on the other, polities in which some parties are permanently excluded. Closed governments consist exclusively of parties that governed in the past; the open ones include, or are dominated by, novices. By assessing these three factors, one can determine whether a party system is closed or open. Party systems are considered to be open when (1) the alternations of governments tend to be partial, (2) no stable configuration of governing alternatives exists, and (3) the access to government is granted to all relevant parties, including the newly established ones. It is closed when (1) alternations of governments are wholesale or none, (2) the governing alternatives are stable over a long period of time, and (3) governments are confined to a narrow circle of parties. Following Casal Bértoa and Enyedi’s12 operationalization of Mair’s framework, this article will measure governmental alternation by adapting Pedersen’s (1979) index of electoral volatility to the measurement of ministerial volatility, computed by adding the net change in percentage of “ministers” (including the PM) gained and lost by each party in the cabinet from one government to the next, and then dividing it by two. The degree of familiarity of the governing formulae is captured by the percentage of ministries belonging to familiar combinations of parties. Thirdly, the percentage of ministers belonging to parties that governed in the past is employed to display the degree to which access to government is open or closed. These figures are calculated for each year. If in a year there was no change (i.e., no new party joined the government, no new coalition formula was introduced and there was no alternation in office), then to that particular year a score of 100 is assigned, meaning complete closure. The composite and cumulative closure is calculated by averaging all years and all the three components. Party System Closure: East versus West First, let us compare the countries of the region as well as the entire region with the old Western European consolidated democracies. We do the comparison by Casal Bértoa and Enyedi / Post-communist Party Systems 427 Figure 1 Party system closure in fourteen post-communist European democracies (1990–2016) Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. taking the latest pieces of information on closure, the ones that refer to 2016. Since we use a cumulative closure index, the scores summarize both the characteristics of the latest year and the characteristics of all the previous years. In other words, the figures do not only capture the state of affairs in 2016 but also summarize the entire post-1990 period. As Figure 1 indicates, a substantial, 6.5 points-large difference exists between the two regions, as the Western Europe average is close to 93 percent, while the Eastern European one is only slightly above 86 percent.13 These figures confirm the fundamentally more structured nature of Western European politics. This conclusion has to be qualified, however, in two ways. First, there has been a move toward higher level closure across post-communist Europe: in 2000 the average closure score was 84.1; in 2010, 84.5; and in 2016, 86.3.14 The progress toward stabilization was larger during the last six years than during the entire previous decade. But because in the meantime closure has increased in the West too (it was 90.5 at the beginning of the century, 92.7 ten years later, and 92.8 at the end of 2016), the gap remained as large as it was one-and-a-half decade earlier. Secondly, Eastern Central Europe appears to be much more differentiated than expected. The most closed systems are to be found in Montenegro, Hungary, Croatia, and Romania; the most open ones in Latvia, Serbia, Estonia, Lithuania and Ukraine. The difference between the most and least closed systems in the region is 18.5,15 almost three times as much as the difference between Eastern and Western Europe. While the closed systems tend to come from the Western rim of Eastern Central Europe, and the open ones from the Eastern part of the region, the grouping of countries appears to have little to do with the differences in pre-communist pluralism, level of modernization or the degree of suppression during communism, that is, factors emphasized by Kitschelt.16 A country can develop a relatively closed system emerging 428 East European Politics and Societies and Cultures from both national-accommodative or patrimonial regimes, and the same applies to open systems. Poor countries can have both predictable (Montenegro) and unpredictable (Ukraine) party relations. EU membership is not a major factor either: Old members can have very open party systems and new members as well as excluded countries can develop considerably closed systems. The only noticeable structural pattern is the overrepresentation of the post-Soviet countries on the open end of the continuum. Table 1 provides information on temporal changes, excluding the two cases that do not meet currently the Freedom House threshold, namely, Ukraine and Montenegro. The table reveals that the overall ranking of the party systems within the region has not changed much during the years. But contrary to the modest increase of the regional averages, there was a decline of structure in Bulgaria and in the Czech Republic, while in Lithuania and Hungary the party systems increased their level of institutionalization rather rapidly. The Estonian party system underwent a period of destabilization but it has subsequently stabilized, while Slovenia first followed the standard route toward stabilization but then, during the 2010s, it deinstitutionalized. The modest shift toward closure in Eastern Europe was due to the growing stability of governmental formulae and to the decreasing likelihood of new parties entering governments. While for the first part of the period Eastern European party systems were particularly innovative in trying various coalition patterns, lately the composition of the governments became somewhat more predictable. Probing deeper into the issue of intraregional tendencies, Figures 2 and 3 present the trajectory of party systems in the region,17 disregarding the first five years of their democratic history. The figures indicate a tendency toward homogenization: the party systems that have a high level of closure tend to change little, while the most inchoate systems tend to improve their closure scores. On the other hand, the graphs show the clustering of the cases into two separate groups: the closed systems virtually never fall below 80 percent, while the open ones virtually never rise above 85 percent. The trajectories demonstrate the steady institutionalization of the Estonian, Latvian, and Lithuanian party systems since the early 2000s and they highlight the continuously high level of structuration of the Hungarian party system. The figures also reveal that some of the most inchoate systems have low scores because their institutionalization has suffered major setbacks either during the early 2000s (Bulgaria, Lithuania, and Estonia) or in the middle of that decade. The early frontrunners, the Czech Republic and Romania, underwent similar periods of de-alignment, but in their case the years of instability were not followed by a similarly steep recovery. Here again we see that Slovenia, a country that appeared to be on a steady road of institutionalization of party relations, became significantly less predictable in the 2010s. Party (In)stability In order to place the trends in closure into the larger context of party politics characteristics, we turn now to the analysis of more conventional dimensions. First Table 1 Closure of Party Systems in Eastern Europe, a Contrast of Three Decades Decade 1990s 2000s 2010s Bulgaria Croatia Czech R Estonia Hungary Latvia Lithuania Poland Romania Serbia Slovakia Slovenia Average 88.2 85.8 84.7 84.5 88.1 91.7 87 87.9 84.7 82.6 82.7 86.1 89.3 91.8 72 77.8 79.8 79.2 80.4 83 83.7 83.2 85.7 87.1 88.6 77.8 80 86.6 86.1 86.6 79.4 88.7 86.9 83.5 84.2 85.5 Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. 429 430 East European Politics and Societies and Cultures Figure 2 The degree of closure in the party systems with currently above-average level of closure Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. Figure 3 The degree of closure in the party systems with currently below-average level of closure Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. we consider the stability at the level of parties. While the characteristics of individual parties cannot be directly translated to the systemic level, the degree of stability at party level is a major cause of the institutionalization of party systems. Some of the relationships between the two levels are mechanical: if individual parties disappear then the relations among these parties cannot survive either. Others are more indirect: The fluctuation in the list of actors may have a negative impact on the Casal Bértoa and Enyedi / Post-communist Party Systems 431 Figure 4 Average number of new parties in fourteen post-communist European democracies (1990–2016) Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. individual party strategies in the sense that the frequent disappearance of partners may push parties toward prioritizing short-term considerations. We examine party-level continuity by comparing the number of new parties above 0.5 percent of support, the average age of large parties, and the number of splits, mergers, and new party foundations. The difference between the two regions in terms of the stability of party rosters is striking. As Figure 4 indicates, the number of new parties with more than 0.5 percent of the votes is more than twice as large in the East (3.4) as in the West (1.4).18 With the sole exception of Montenegro, all post-communist party systems experience the entry of more than two parties per election, while in the West only Greece and, more recently, Cypriots and Icelanders can be accustomed to a similar level of renewal (not shown). Furthermore, in the East many of these new parties, some of them formed just few months ahead of the elections, have a realistic chance of gaining governmental power. The Action of Dissatisfied Citizens (ANO) in the Czech Republic, the Bridge of Independent Lists (Most) in Croatia, the Serbian Progressive Party (SNS) in Serbia, the Liberal Reformist Party (PLR) in Romania, the Network (Siet’) in Slovakia, Modern (.N) and Kukiz 2015 (Kukiz’15) in Poland, and the Modern Centre Party (SMC) in Slovenia constitute the most recent examples of new parties performing well at elections. In this context, the Slovenian case is particularly dramatic, with SMC, a party that was created just forty-two days before the elections, wining 35 percent of the votes and 40 percent of the seats. The variation on party continuity is largely in line with the variation of party system closure, although Estonian and Ukrainian party systems achieved relatively high 432 East European Politics and Societies and Cultures Figure 5 Average age of parties in fourteen post-communist European democracies (1990–2016) Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. level of openness with relatively few new parties, while in Croatia surprisingly many new parties were established given the relative stability of the relations. A contrast between the average age of parties confirms the gap between the two parts of the continent. Figure 5 ranks party systems according to the age of the parties that received at least 10 percent of the vote during the last parliamentary elections.19 Obviously, given the shortage of the democratic time span, Eastern Europe has an in-built disadvantage in this aspect. Out of the parties that received more than 10 percent of the vote in the most current legislative elections, only Fidesz has a longer past than the post-communist regimes. But the difference between the regions is not due exclusively to the newness of democracies. In fact almost half (twenty of forty-one) of the currently competing Eastern European political parties were created after the beginning of the twenty-first century, while in the West, if Italy is excluded, only nine such parties are relevant. Furthermore, of the twenty cases of recent party formations in the East, five were created in the 2010s. Only in a few countries, such as Croatia and Hungary, are the party systems dominated by parties that are more than twenty-five years old. In contrast, and in spite of the recent spectacular successes of new political initiatives, Western citizens keep on voting for parties that were established well before the citizens themselves were born. This is particularly so in countries such as Portugal, Cyprus, Malta, Luxembourg, Ireland, Sweden, United Kingdom, and until very recently, also Greece and Spain. The only two significant exceptions to this general rule are Italy and France. In the former case, the “implosion” of the old party system in 1994 brought about new political parties that continue to struggle organizationally, while the French parties, mainly on the right of the political spectrum, Casal Bértoa and Enyedi / Post-communist Party Systems 433 underwent a considerable degree of organizational restructuring after 2002.20 If we exclude the extreme Italian case, then we find that out of the sixty-two relevant West European political parties, forty-eight have been active prior to 1989. This means a survival rate of 77.4 percent. In Eastern Europe, only eighteen of the forty-one major parties were already active in 1994: This implies a survival rate of 43.9 percent. The average age of parties typically changes little from year to year and, again typically, the change is in the positive direction even in Eastern Europe. There are some notable exceptions, however: 2015 Croatia, 2010 and 2013 Czech Republic, 2010 Hungary, 2011 Latvia, 2012 Montenegro, 2008 Romania, 2012 Serbia, 2010 Slovakia, and 2011 Slovenia. In these countries and years from one year to another, the party system rejuvenated, because of the disappearance of established parties and the breakthrough of novices. Notwithstanding the higher renewal rate in Eastern Europe, the temporal trends concerning average party age appear as rather similar across the two regions. From 1994 to 2016 the Eastern European average age changed from four to seventeen, the Western European one from 57.2 to 66.4. Finally, considering splits and the formation of new parties as indicators of centrifugal tendencies, the East again shows consistently and significantly more instability. The difference is not simply due to the turbulence characterizing the formative years of the post-communist party systems. In the post-2000 period, as many splits occurred in the East as in the West, but the ratio of new foundations, 6 versus 4, shows a larger Eastern “advantage.” The difference in the number of mergers is even higher than the difference in number of new foundations: nine mergers in the East and only four in the West—though mergers could be interpreted, in principle, not only as forms of discontinuity but also as intermediary steps toward consolidation. Party Membership The fluctuation of the roster of parties is closely linked to the general weakness of the organizational networks surrounding parties. Concerning party organizations, a number of scholars have confirmed the elitist nature of parties and the weakness of organizational encapsulation in Eastern Europe. According to the data of van Biezen and her colleagues,21 the total party membership as percentage of the electorate (i.e., M/E) is lower in the East than in the West, notwithstanding a sharp decrease in the last decades in the West (Figure 6). In particular, while in Western Europe only five countries fail to display an M/E ratio higher than 3 percent, in the East only Bulgaria, Estonia, Romania, and Slovenia reach this figure. Surely, organization is more than membership, and the significance of organizational characteristics cannot be established simply by raw organizational data. Tavits, for example, has shown that better organization provides a significant advantage in the electoral competition in Eastern Europe.22 And one must also acknowledge that 434 East European Politics and Societies and Cultures Figure 6 Party membership in post-communist Europe (2010/2011) * Does not include Croatia, Montenegro, Serbia, and Ukraine. ** Does not include Iceland, Luxembourg, and Malta. Source: I. van Biezen, T. Poguntke, and P. Mair, “Going, Going…Gone?”. the general erosion of organizational strength is not universal. Indeed, the party that Mair cited as the best example of the conscious rejection of grassroots involvement, the Hungarian Fidesz, has developed, since then, one of the most formidable organizational structures in the region. The relevance of party organizations needs to be considered also in the context of the relationship of parties and the state, and of parties and other civic society actors. Eastern European parties are often more powerful in these relations than Western European ones, even leading to claims that both spheres are colonized by parties.23 Boundaries The organizational strength can be undermined not only by a lack of membership and lack of bureaucratic apparatuses but also by the blurred boundaries between parties and various other social actors such as movements, churches, or trade unions. Mair’s assessment was that these boundaries are particularly porous in Eastern Europe. Unfortunately, there is no comprehensive data set measuring this aspect, but in our view this evaluation is no longer correct. Today parties have fairly clear identities. Movements and interest groups tend to be relegated to the margins of the political systems.24 In some sense, the opposite of Mair’s assertion seems to be the case: In the East the emotional and organizational distance between political parties and interest groups, social movements, trade unions, and other civic organizations is larger than in the West. The relatively high autonomy of parties, however, does not seem to foster the stabilization of the party systems. The gap between parties and governments, on the one hand, and the civil society, on the other, provides rather the (often populist) newcomers with ammunition to pose a continuous threat to the established parties.25 Casal Bértoa and Enyedi / Post-communist Party Systems 435 Electoral Volatility The institutionalization of party systems is typically measured via electoral volatility indices. While in our view the party systems should be understood primarily in terms of relations among parties, the fluctuation in the support of parties has a direct influence on the institutionalization of party relations. In order to calculate volatility scores, we use Pedersen’s index, which measures “the net change within the electoral party system resulting from individual vote transfers.”26 In terms of coding rules, we adopt Bartolini and Mair’s rules:27 1. Volatility figures are always rounded up from the second decimal point. . . . 2. When two or more parties merge to form a new party, or when one or more parties merge with an existing party, the relevant electoral volatility is computed by subtracting the combined vote of the merging parties in the election immediately preceding the merger. . . . 3. When a party splits into two or more parties, the relevant electoral volatility is computed by subtracting the combined vote of the new parties from that of the original party in the election immediately preceding the split. . . . 5. When the name of a party changes, volatility is computed as if it were the same party.28 Following Sikk,29 and taking into consideration the specific organizational instability of political parties in Eastern Europe,30 we have considered as splits all those cases in which a minority of the leaders or a relevant member of the leadership established a new party. In line with previous studies, following somewhat different rules and principles, we detect a dramatic difference between East (24.1) and West (12.2), (significance .000). The only systems in the West approaching—but not reaching—the Eastern European mean are Iceland (17.9 percent), France (16.4 percent), Italy (19.7 percent), and the Netherlands (20.7 percent). Of 140 pairs of elections held in Western Europe, barely four-fifths (i.e., 73.6 percent) remained below the 15 percent ceiling. Higher than 20 percent volatility figures are common in fact in only one Western European country, the Netherlands, where four such elections occurred recently (and all elections since 1990 were above the 15 percent threshold). Among the remaining 133 Western European election-pairs, only ten had a higher than 20 percent score, namely, Italy in 1994 (36.7 percent) and 2013 (35.7 percent), Greece in May 2012 (34.4 percent), Ireland in 2011 (29.7 percent) and 2016 (36 percent), Iceland in 2013 (28.5 percent) and 2016 (21.8 percent), France in 2012 (24.1 percent), Portugal in 1995 (20.7 percent), and Spain in 2015 (35.1 percent). In the East, on the other hand, the average election would be considered by Pedersen’s standards to be “earthquake” election (i.e., with no less than 15 percent of 436 East European Politics and Societies and Cultures Figure 7 Average electoral volatility in fourteen post-communist European democracies (1990–2016) Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. electoral volatility) in all countries. Eleven party systems display scores well above 20 percent. Of the seventy-seven pairs of elections considered, only fourteen31 remained below 15 percent: 7.7 percent in Poland (2011), 8.4 percent and 12.7 percent in Hungary (2006 and 2014), 8.5 percent and 13.2 percent in Serbia (2008 and 2016), 9.4 percent in Bulgaria (2014), 12.2 percent in Ukraine (2007), 13 percent in Estonia (2011), 13.7 percent in Slovakia (1994), 14.7 percent and 13.8 percent in Romania (2004 and 2016), and 14.9 percent, 13.1 percent, and 7.8 percent in Croatia (2007, 2011, and 2016). Figure 7 displays the average volatility scores between 1990 and 2016 for all the fourteen post-communist countries studied here, and contrasts Western and Eastern Europe. Two patterns stand out clearly. First, post-communist electorates differ from each other substantially. The difference between the most and the least volatile systems is higher than 20 percentage points. Second, West European voters are far more loyal than their Eastern European counterparts, in line with Mair’s predictions.32 Even though the Western European average has increased in the recent years, it is well below—actually it is about half of—the Eastern European levels. The difference between the two regions, considering almost three decades of democratic politics, is 11.9 percentage points.33 As far as temporal trends34 are concerned, three relevant findings need to be mentioned. First, while during the first decade of the twenty-first century electoral volatility decreased in both East and West, in the last six years Western European party systems have experienced a reversal of this trend, resulting in a convergence between the two parts of the European continent.35 In fact, the East–West difference decreased Casal Bértoa and Enyedi / Post-communist Party Systems 437 decade by decade: it was 15.5 in the 1990s, 14.3 in the 2000s, and 7 in the 2010s. Second, in spite of the general increase in volatility in Western Europe, most of the examples of decline during the most recent democratic legislative elections (eight of fourteen) still belong to the West. Third, while most Eastern European countries (eight of twelve)36 have undergone electoral earthquakes (see above) at the time of the most recent elections, only seven of nineteen (France, Germany, Iceland, Ireland, Italy, the Netherlands, and the United Kingdom) have experienced a similar fate in the West.37 While in Eastern Europe there was a small decrease in electoral volatility since the 1990s, the trends are not uniform across all countries (Table 2). The solid drop in countries such as Estonia, Latvia, and Poland contrasts sharply with the general tendency of increase in the Czech Republic, Slovakia, and Slovenia. Out of the recently more tranquil systems, only Croatia had no major rupture in the development of its electoral landscape. At the same time, Figures 8 and 9 show that the countries with currently high levels of volatility more often had above-average scores than the currently more stable systems, suggesting a clustering of cases into two broad groups, with only a few cases shifting from one cluster to another. Number of Parties in the Electoral and Parliamentary Arenas Finally, we need to consider the factor that is perhaps most often discussed as a cause of electoral volatility, and therefore of party system institutionalization: the number of parties. While the association between the degree of party system fragmentation and inherent party system instability is disputable, extreme fragmentation typically makes elections less consequential and increases uncertainty in government formation. Below we compare countries and the two regions by contrasting the effective number of electoral and parliamentary parties (ENEP/ENPP).38 Figures 10 and 11 reveal that fragmented party systems constitute the norm in the East. The average Eastern system can be labeled as a case of “extreme pluralism” (ENEP = 5.3), while its Western counterpart rather belongs to the category of “limited pluralism” (ENEP = 4.8). But the difference between the regions is in fact much smaller than what is implied by these qualitative labels. The gap is only 0.5 points, a difference that is statistically not significant, and one can find highly fractionalized systems in Western Europe (Switzerland, Finland, the Netherlands, Italy, and Belgium [last one being by far the most fragmented system of Europe]), and highly concentrated systems in the East (especially Hungary and, Montenegro). Moreover, the decline in fragmentation in the East is striking. Taking the 1990s as a starting point, the number of electoral parties decreased from 6.5 to 4.9 in the 2010s, the number of parliamentary parties from 4.5 to 3.8. Between 1992 and 2016, the decline was continuous, and the correlation between year and average fragmentation is very high (–.74 at the parliamentary and –.82 at the electoral level). Since the 438 Table 2 Electoral volatility in Eastern Europe, a contrast of three decades Decade Bulgaria Croatia Czech R Estonia Hungary Latvia Lithuania Montenegro Poland Romania Serbia Slovakia Slovenia Ukraine Average 1990s 2000s 2010s 21.3 45.2 14.8 16.1 14.5 22.1 17.4 34.6 23.5 25.4 15.1 29.5 11.5 23.2 40.4 27.6 29 36.5 36.7 33.1 17.2 24.1 31 19.8 21.9 17 20.4 19.5 Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. 17.5 20 24 25.1 24 42.3 23.4 26.7 24.7 23.4 Casal Bértoa and Enyedi / Post-communist Party Systems 439 Figure 8 Degree of volatility in the currently more volatile electorates Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. Figure 9 Degree of volatility in the currently less volatile electorates Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. fragmentation of West European party systems had increased in the meantime (from 4.6 in the 1990s to 5.3 in the 2010s at the electoral level and from 3.9 in the 1990s to 4.3 in the 2010s at the parliamentary level), the two halves of Europe, in fact, switched positions.39 In the West, the fragmentation of sociopolitical structures (especially of the working class and of religious blocs), and in the East the impact of the restrictive electoral rules, contributed primarily to this convergence.40 The last point is underscored by the fact that Gallagher’s “least-squares index,”41 measuring the level of distortion of electoral systems, has a score of 6.8 for the EE14, 440 East European Politics and Societies and Cultures Figure 10 The average effective number of electoral (ENEP) parties in fourteen postcommunist European democracies (1990–2016) Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. Figure 11 The average effective number of parliamentary (ENPP) parties in fourteen post-communist European democracies (1990–2016) Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. as opposed to only 5 for the WE20. As a result of the disproportional electoral systems42 adopted in the post-communist region, the average level of fragmentation at the parliamentary level is exactly the same in the two regions (ENPP = 4). As Tables 3 and 4 show, the number of parties is far from being a fixed trait of Eastern Central European party systems. Hungary is the only country that had a Table 3 The Effective Number of Electoral Parties in Eastern Europe, a Contrast of Three Decades Decade Bulgaria Croatia Czech R Estonia Hungary Latvia Lithuania Montenegro Poland Romania Serbia Slovakia Slovenia Ukraine Average 1990s 2000s 2010s 3.7 4.7 5.6 5.2 4.1 5.8 4.4 7.2 7.2 5.2 5 6 2.9 3 7.6 7.2 5.1 6.1 7.3 7.2 3.2 3.4 9.4 4.6 4.1 6.1 4.4 3.2 4.5 4.6 Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. 5.5 7.5 5.7 7.4 5.4 5.5 5.4 6.5 5.1 4.9 441 442 Table 4 The Effective Number of Parliamentary Parties in Eastern Europe, a Contrast of Three Decades Decade Bulgaria 1990s 2000s 2010s 2.5 3.7 4.2 Croatia Czech R Estonia Hungary Latvia Lithuania Montenegro Poland Romania Serbia Slovakia Slovenia Ukraine Average 3.6 2.9 4.2 3.4 5.3 5.2 4.6 4.3 3.4 2.3 2 6.1 5.5 4.5 3.2 5.2 4.9 2.5 3 5.9 3.6 2.9 4.3 3.5 2.9 Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. 3.7 3.5 4.1 5.5 4.2 6.1 4.7 4.4 3.8 4.5 4.0 3.8 Casal Bértoa and Enyedi / Post-communist Party Systems 443 Figure 12 The effective number of electoral parties in the currently more fragmented systems Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. consistently below-average level of fragmentation (though in terms of the number of electoral parties it was originally close to average), and Latvia, Slovakia, and Slovenia are the only continuously highly fragmented systems. The other party systems have oscillated around the regional average. While originally, Slovenia, Estonia, Latvia, and Poland had the largest number of parties, in the 2010s these countries have been overtaken by the Czech Republic and Lithuania, party systems that were originally rather concentrated. The most concentrated, and in this sense most “consolidated,” party systems are to be found today in Romania, Poland, Hungary, Croatia, and Montenegro. As indicated by the data, Latvia, Slovakia, and Lithuania took turns as the most fragmented systems of the region, never falling below the threshold of four parties. Interestingly, the destructuration of the Slovenian electorate noticed above is not reflected by a growing fragmentation of the party system. Actually the number of Slovenian parties slightly declined recently. On the other hand, the increase of electoral volatility in the Czech Republic is rather well mirrored by the increase in the number of parties. The most dramatic decline in the number of parties happened in Poland, turning the most fragmented system into one of the most concentrated ones (Figures 12 and 13).43 While in most cases the number of parliamentary parties is significantly lower than the number of electoral parties, showing the power of electoral rules, in the Czech Republic the distortion is minimal, and consequently the effective number of parties in the Parliament remained recently above six: an unusually high value. Interestingly, the countries that democratized belatedly (and show up in our database only for the second part of the timecontinuum), Serbia, Montenegro, Croatia, and Ukraine, had, for most of the time of their democratic career, a relatively low number of parties (Figures 14 and 15). 444 East European Politics and Societies and Cultures Figure 13 The effective number of electoral parties in the currently less fragmented systems Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. Figure 14 The effective number of parliamentary parties in the currently more fragmented systems Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. Conclusions Finally, we revisit the most important changes that took place in post-communist party politics along the classical dimensions with the help of Figure 16 and provide a comparative perspective by showing the respective figures for Western Europe in Figure 17. Casal Bértoa and Enyedi / Post-communist Party Systems 445 Figures 15 The effective number of parliamentary parties in the currently less fragmented systems Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. Figure 16 The evolution of post-communist party systems over time (1990–2016) * For presentation purposes, closure scores have been divided by 10. Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. The contrast between the two regions is the largest in terms of electoral volatility, but the difference is significantly large in terms of closure and party-level stability too. But on most dimensions, with the possible exception of party instability and party system closure, there is a convergence going on, helped by changes in both 446 East European Politics and Societies and Cultures Figure 17 The evolution of Western European party systems over time (1990–2016) * For presentation purposes, party instability and closure scores have been divided by 10. Source: F. Casal Bértoa, “Database on WHO GOVERNS in Europe and Beyond,” PSGo (2017), whogoverns.eu. Eastern and, especially, Western Europe.44 The process reached furthest concerning party system fragmentation, eliminating all differences between the regions and opening the prospect for a more concentrated post-communist Europe. The lack of convergence on closure is surprising because the de-institutionalization of Western party politics is often taken for granted. We have found some indications indeed of a structural dealignment, but this process does not (yet?) manifest itself at the level of party relations in the governmental arena.45 The various analyzed dimensions do not always show the same dynamic, but together they help us differentiate between few clusters of party systems. Hungary, Croatia, and Romania (and among the newly democratized countries, Montenegro) are party systems that have been, throughout their career, relatively institutionalized, at least compared to the regional averages. The Baltic countries, Poland, and Slovakia increased their level of structure from a very low one, while the Czech Republic and Slovenia have gradually lost much of their stability and predictability. The emerged grouping is not in line with theoretical models based on a pre-communist or communist past, on socioeconomic development, EU membership, or on distance from the West. The existence of a transparent conflict structure (see the examples of Montenegro, Croatia, and Hungary) appears as a more relevant factor, but bipolarity does not guarantee stability and predictability, as Lithuania and the Czech Republic of the 1990s have demonstrated. Mair’s cited article suggested long-term instability. And yet the author still expected that the passing of time would make a difference. He emphasized that one Casal Bértoa and Enyedi / Post-communist Party Systems 447 needs time for coalitional norms to develop and for collective identities to crystallize. From today’s vantage point, he appears as even too cautious. He wrote, for example, [m]uch like Germany and Italy in the postwar decades, and Portugal and Spain in the last twenty years, the post-communist democracies will also eventually assimilate to the pattern set by the established democracies.46 As the figures cited above indicate, the expectations toward maturation and stabilization are not (yet) met. The consolidation of party politics in the region remained partial. Fragmentation declined, the closure of the governmental arena slightly increased, and there are fewer new parties in the parliaments. But electoral volatility and closure changed only marginally, and occasionally new parties are still able to capture governmental power. This means that some of our theoretical expectations have been entirely met, others less so. The difference between East and West has been found to continue to be substantive on all analyzed aspects, except fragmentation. The low survival rate of parties in the East, or the current de-institutionalization of the Czech or the Slovenian systems contributed to the survival of the East–West difference. But the convergence is apparent on many dimensions. The growing fragmentation, volatility, and rejuvenation of the Western systems are as responsible for the trend of convergence between the regions as the “maturation” of the East.47 The differences within Eastern Europe are often striking. While the current article was not aimed at uncovering causal patterns behind the changes in party system institutionalization, the presented differences cast doubt on the systemic role of fragmentation: Eastern party systems continued to stay volatile in spite of declining fragmentation. To conclude, Peter Mair was largely, though not in every detail, right. The twenty years’ time span of convergence observed in the West–South relation may need to be multiplied when applied to the East–West relation, but the direction of convergence is undisputable. Notes 1. P. Pierson, “Increasing Returns, Path Dependence, and the Study of Politics,” American Political Science Review 2 (2000):671–87. 2. Included later on as a chapter in Party System Change: Approaches and Interpretations (Oxford: Oxford University Press, 1997), 175–98. 3. H. Kitschelt, “The Formation of Party Systems in East Central Europe,” Politics and Society 20 (1992): 7–50; and “Formation of Party Cleavages in Post-communist Democracies,” Party Politics 1 (1995): 447–72; also G. A. Evans and S. Whitefield, “Identifying the Bases of Party Competition in Eastern Europe,” British Journal of Political Science 23 (1993): 521–48. For a critical view of the above-mentioned studies, see F. Casal Bértoa, “Party Systems and Cleavage Structures Revisited: a 448 East European Politics and Societies and Cultures Sociological Explanation of Party System Institutionalization in East Central Europe,” Party Politics 20 (2014): 16–36. 4. Mainwaring originally proposed four criteria for capturing party system institutionalisation: stability in the patterns of party competition, stable roots in society, legitimacy of parties and elections, and strength of party organisations. 5. S. Mainwaring, “Party Systems in the Third Wave,” Journal of Democracy 9 (1998); S. Mainwaring and T. Scully, Building Democratic Institutions: Party Systems in Latin America (Stanford, CA: Stanford University Press, 1995); S. Mainwaring and M. Torcal, “Party System Institutionalization and Party System Theory after the Third Wave of Democratization,” in Handbook of Party Politics, ed. Richard S. Katz and William J. Crotty (London: Sage, 2006); S. Mainwaring, and E. Zoco, “Political Sequences and the Stabilization of Inter-party Competition,” Party Politics 13 (2007): 155–78. For a comprehensive review of the studies on the topic, see F. Casal Bértoa, “The Three Waves of Party System Institutionalization Studies,” Political Studies Review, published online 27 December 2016. 6. Mair, “Party System Change.” See also P. Mair, “The Freezing Hypothesis: And Evaluation,” in Party Systems and Voter Alignments Revisited, ed. L. Karvonen and S. Kuhnle (London: Routledge, 2001). 7. In a comprehensive overview of the party systems literature, Steven Wolinetz discusses Mair’s concept of closed competition as one of the most significant post-Sartorian models, but he also remarks that “its full potential has not yet been explored.” We agree on both counts. See S. Wolinetz, “Party System Institutionalization: Bringing the System Back In,” in Katz and Crotty, Handbook of Party Politics, 58. 8. H. Kitschelt, Z. Mansfeldová, R. Markowski, and G. Tóka, Post-communist Party Systems: Competition, Representation and Inter-Party Cooperation (Cambridge: Cambridge University Press, 1999). 9. M. A. Vachudová, “Tempered by the EU? Political Parties and Party Systems before and after Accession,” Journal of European Public Policy 15 (2008): 861–79. 10. By Western Europe, we mean the following countries in the text: Austria, Belgium, Cyprus, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Luxembourg, Malta, the Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, and the United Kingdom. 11. Mair, “Party System Change.” See also F. Casal Bértoa and P. Mair, “Party System Institutionalization across Time in Post-communist Europe,” in Party Government in the New Europe, ed. H. Keman and F. Müller-Rommel (London: Routledge, 2012). 12. F. Casal Bértoa and Z. Enyedi, “Party System Closure and Openness: Conceptualization, Operationalization and Validation,” Party Politics 22 (2016). 13. The difference is significant at .000. 14. If we exclude the short-lived cases, Montenegro, Serbia, and Ukraine, then 84.1, 84.8, and 86. 15. 12 percent if we discard Montenegro. 16. Kitschelt et al., Post-communist Party Systems. 17. Systems that were considered to be democratic for too short time (Montenegro and Ukraine) were left out from these figures. 18. The difference is significant at .000. 19. This threshold is relatively high but it is line with the relevant literature: R. Dix, “Democratization and the Institutionalization of Latin American Political Parties,” Comparative Political Studies 24 (1992); S. Mainwaring and T. Scully, Building Democratic Institutions: Party Systems in Latin America (Stanford, CA: Stanford University Press, 1995); M. Tavits, “The Development of Stable Party Support: Electoral Dynamics in Post-communist Europe,” American Journal of Political Science 49 (2005). Of course, parties can have an important role (can be “relevant” in Sartori’s sense) even below 10 percent if they can contribute to government-building or if they can influence the strategy of other parties. 20. The French party system changed even more profoundly in 2017, but the entry of the La République En Marche! is not part of the analyzed data set. Casal Bértoa and Enyedi / Post-communist Party Systems 449 21. Biezen et al., “Going, Going . . . Gone?,” European Journal of Political Research (2012). Unfortunately, the accuracy of national data on such a sensitive issue as party membership cannot be guaranteed. The different sources converge on the relative position of the regions, but note that the difference between the figures reported in van Biezen, 3 and 5.6, across the seventeen cases, does not reach the conventional statistical threshold (significance = .106). 22. M. Tavits, Post-communist Democracies and Party Organization (Cambridge: Cambridge University Press, 2013). 23. P. Kopecký, “Political Parties and the State in Post-communist Europe: The Nature of Symbiosis,” Journal of Communist Studies and Transition Politics 22 (2006): 251–73; A. Ágh, “The Development of East Central European Party Systems: From ‘Movements’ to ‘Cartels,’” in The Challenge of the Europeanization in the Region: East Central Europe, ed. Máté Szabó; Magyar Politikatudományi Társaság; Politikai Tudományok Intézet (Budapest: Budapest Institute for Political Sciences, 1996), Z. Enyedi, “Party Politics in Post-communist Transition,” in Katz and Crotty, Handbook of Party Politics. 24. P. Webb and S. White, Party Politics in New Democracies (Oxford: Oxford University Press, 2007). 25. T. Haughton and K. Deegan-Krause, “Hurricane Season: Systems of Instability in Central and East European Party Politics,” East European Politics and Societies 29 (2015): 61–80. 26. M. N. Pedersen, “The Dynamics of European Party Systems: Changing Patterns of Electoral Volatility,” European Journal of Political Research 7 (1979): 1–26. 27. Often followed when calculating electoral volatility, M. Gallagher, M. Laver, and P. Mair, Representative Government in Modern Europe (London: McGraw-Hill, 2011); Z. Enyedi and F. Casal Bértoa, “Patterns of Inter-party Competition (1990-2009),” in Europeanizing Party Politics? Comparative Perspectives on Central and Eastern Europe, ed. P. G. Lewis and R. Markowski (Manchester: Manchester University Press, 2011); V. Emanuele, Dataset of Electoral Volatility and its internal components in Western Europe (1945-2015) (Rome: Italian Center for Electoral Studies, 2015), http://dx.doi. org/10.7802/1112. 28. S. Bartolini and P. Mair, Identity, Competition, and Electoral Availability: The Stabilization of European Electorates 1885–1985 (Cambridge: Cambridge University Press, 1990), appendix. 29. A. Sikk, “How Unstable? Volatility and the Genuinely New Parties in Eastern Europe,” European Journal of Political Research 44 (2005): 391–412. 30. S. Gherghina, Party Organization and Electoral Volatility in Central and Eastern Europe: Enhancing Voter Loyalty (London: Routledge, 2014). 31. In case we discount elections that took place barely a year following a previous election (Croatia 2016, Bulgaria 2014, Serbia 2008, and Ukraine 2007), then this figure is only ten. 32. See also J. Bielasiak, “Party Competition in Emerging Democracies: Representation and Effectiveness in Post-communism and Beyond,” Democratization 12 (2005): 331–56; J.-E. Lane and S. Ersson, “Party System Instability in Europe: Persistent Differences in Volatility between West and East?,” Democratization 14 (2007): 92–110; T. Haughton and K. Deegan-Krause, “Hurricane Season.” 33. The difference would be just of 6.7 if we were to consider the last elections in each country. 34. When interpreting the averages, one should keep in mind that occasional sudden changes in the levels of volatility from one election to another due to idiosyncratic reasons are common. In Bulgaria, for example, volatility increased from 25 to 48 between 1997 and 2001 or between 2005 and 2009, while in Poland it decreased from 23.7 to 7.7 between 2007 and 2011, but increased again in 2015, reaching 31.9. In Serbia, in a four-year period (2012–2016) volatility went up from 17.5 to 27.8 before decreasing to 13.2. A similar trajectory can be observed in Croatia: 13.1 in 2011, 22.6 in 2015, but just 7.8 in 2016. In Hungary, electoral fluidity decreased up to 10 points before 2006, just to reach 33.7 (an increase of 25 points) in 2010, while it decreased 20 points 4 years later (see also Enyedi and Casal Bértoa “Patterns,” 133–34). 35. F. Casal Bértoa, “Post-communist Politics: On the Divergence (and/or Convergence) of East and West,” Government and Opposition 48 (2013). 450 East European Politics and Societies and Cultures 36. Exceptions are Bulgaria, Hungary, Romania, and Serbia. The 2016 Croatian and the 2007 Ukrainian elections, which took place barely a year after the previous ones, are excluded. 37. The 2016 Spanish elections are not included for the reasons mentioned in the previous footnote. 38. M. Laakso, and R. Taagepera, “‘Effective’ Number of Parties. A Measure with Application to West Europe,” Comparative Political Studies 12 (1979). 39. Though the difference is not yet statistically significant. 40. Casal Bértoa, “Post-communist Politics.” 41. M. Gallagher, “Proportionality, Disproportionality and Electoral Systems,” Electoral Studies 10 (1991): 33–51. 42. Casal Bértoa, “Post-communist Politics,” 402–8. 43. F. Casal Bértoa and S. Guerra, “Earthquake or Hurricane? The Rise and Fall of Populist Parties in Poland,” in Absorbing the Blow. Populist Parties and Their Impact on Parties and Party Systems, ed. S. Wolinetz and A. Zaslove (Colchester: ECPR Press, 2017). 44. Characterized by increasing volatility, fragmentation, and polarization. 45. See also Z. Enyedi, “The Discrete Charm of Political Parties,” Party Politics 20 (2014): 194–204. 46. Mair, “Party System Change,” 197. 47. Z. Enyedi and K. Deegan-Krause, “Voters and Parties in Eastern Europe,” in The Routledge Handbook of East European Politics, ed. A. Fagan and P. Kopecky (London: Routledge, 2017). Zsolt Enyedi is Professor at the Political Science Department of Central European University, working on party politics, political attitudes, religion and politics and populism. He held research fellowships at the Woodrow Wilson Center, Kellogg Institute (Notre Dame University), the Netherlands Institute for Advanced Studies, the European University Institute (Florence, Italy) and Johns Hopkins University. Fernando Casal Bértoa is an Assistant Professor in the School of Politics and International Relations at the University of Nottingham (United Kingdom). He is also member of the OSCE/ODIHR “Core Group of Political Party Experts” and co-director of REPRESENT: Centre for the Study of Parties and Democracy. Currently he is writing with Zsolt Enyedi Party System Closure: Alliances and Innovations in Europe between 1848 and 2016 (OUP).