Ideological Radicalism and Anti-democratic Views: A Comparative Study of New Democracies

Transcription

1 Ideological Radicalism and Anti-democratic Views: A Comparative Study of New Democracies Willy Jou University of Tsukuba (Japan) Sejin Koo University of California, San Diego Abstract In contrast to established democracies where most citizens take their existing political system as a given, democratic principles and procedures are a subject of political debate in many countries where authoritarian rule is within recent memory. In the latter cases, ideological orientations may still be structured by a 'regime cleavage', that is, disagreements about the merits of democratic institutions or evaluations of the previous regime, rather than bread and butter issues. While a number of authors have attributed this to elite mobilization in new democracies, the present paper approaches this topic from the perspective of ordinary citizens. Against the background of a higher proportion of citizens who identify with radical ideological positions in new democracies, we analyze cross-national survey data to explore the extent to which such radical opinion is associated with anti-system attitudes and, if so, whether this applies equally to the radical right and left. Prepared for the European Consortium for Political Research General Conference, Université de Montréal, August

2 Introduction Recent decades have seen a marked increase in both academic and journalistic attention to extremist political forces in established Western democracies, in part reflecting the success of radical parties in several European countries. However, for all their rhetoric of challenging the system, few of these parties or their supporters question the institutional framework within which they operate. By contrast, the same does not necessarily hold true in new democracies. In societies where authoritarian rule is still within the living memory of large sections of the public, and where its legacies can still be contentious, extremist political views may pose a challenge to the procedures, or even principles, or democratic politics. This highlights the importance of examining radical ideological views in newly democratized countries. The process of democratic transition can be considered completed when democracy becomes the 'only game in town', referring to its acceptance in 'behavioral, attitudinal, and constitutional dimensions' (Linz and Stepan 1996). The behavioral and constitutional aspects require that actors agree to abide by certain legal procedures in the pursuing their objectives or resolving conflicts, and for the most part can be readily observed. On the other hand, the attitudinal component is less directly discernible, but may have the most enduring consequence since belief in democratic procedures (or the lack thereof) can affect actors' choices in the other two dimensions. Furthermore, this process 'may last for decades and only be complete after a change of generations' (Merkel 1998:40). Thus, the present study focuses on skeptical opinions about democratic principles, examine whether and in what form these beliefs relate to the main axis of issue competition in new democracies, and assess how this relationship changes over time. Issues comprising the main dimension of competition obviously vary from one country to another, and often a uni-dimensional axis may be inadequate to fully capture the complexities of political debate. However, for the purpose of a comparative study, it is necessary to have a measure that is applicable cross-nationally, and we utilize the widely used left-right ideological scale. In many established democracies, the left-right spectrum has long been primarily defined by economic policies, but in new democracies these issues can be superceded, at least in the period immediately following the end of autocratic rule, by questions over the legitimacy of the old and new regimes, or an 'authoritarian-democratic cleavage' (Moreno 1999). This 1

3 phenomenon has received most attention at the level of political parties, particularly those associated with the former regime (e.g. Druckman and Roberts 2007; Grzymala-Busse 2001; Klingemann 2005). Yet this cleavage can be found among the general public, and not only as a consequence of elite mobilization. This is an aspect that has been less extensively analyzed, leaving a vacuum that our study seeks to fill. In addition, we utilize longitudinal data to trace the relationship between democratic beliefs and ideological orientations over time. One may assume that experience with continuous democratic government helps to cultivate a reservoir of diffuse support for the political system (Easton 1965), or at least affirm in citizens' minds that democracy is here to stay and encourage them to seek change by working through the new rules of the game rather than attempting to overturn them. While it is unlikely that all doubts about the operation of democratic procedures can be dispelled, it augurs well for prospects of consolidation if the main axis of political competition is no longer structured by opinions toward the system itself. In view of this, our paper 1) investigates the extent to which democratic scepticism is concentrated among citizens identifying themselves with the radical right or left, or in both extremes of the spectrum, and 2) explores how the accumulation of experience with democratic rule, as well as other factors such as economic performance and the quality of governance, affect the proportion and distribution of radical ideological views. System support and polarization in new democracies While the concept of democratic consolidation has been variously defined, one important condition involves a large proportion of citizens accepting that 'the democratic regime is the most right and appropriate for their society', and preferable to any other regime type (Diamond 1999:65). In fact, democracy can be reversed in the absence of popular belief in its legitimacy (Linz and Stepan 1978). A number of historical studies have illustrated that extremist political forces pose a significant threat to the persistence of democracy, the most prominent cases being inter-war Germany, and most of these analyses focus on extremist parties. This was particularly damaging to the survival of new democracies, such as in the nascent Spanish Second Republic where 'the new system was threatened with subversion and violence by both an extreme right and an extreme left' (Payne 2006:37). Yet examples of countries where the presence of such parties 2

4 did not bring about the collapse of democracy (see Capoccia 2001) remind us that the political environment, including not only the strength of extreme groupings but also the preferences of ordinary voters, played a key role in determining the survival or demise of democratic government. In other words, the existence of sceptical views toward democracy, especially if persisting over time, can have grave implications for the viability of democratic regimes. Whereas most citizens in countries with a long history of uninterrupted democratic government likely take their system as a given, and engage in political competition over various issues within the framework of existing institutions, this does not necessarily hold true in societies where democracy has not been introduced until relatively recently. In the latter case, democracy is seen as only one among several possible regime types (Rose et al. 1998), and a non-democratic system is not an abstract and distant concept, but part of real-life experience for a considerable proportion of the citizenry. Indeed, it is likely that in the aftermath of transition, segments of the population who supported the former regime, whether due to political convictions or material benefits, would regret its passing and perhaps even yearn for its restoration. Furthermore, in countries where authoritarian rule induced (or at least coincided with) high rates of economic growth and thus retained some credibility, an 'authoritarian nostalgia' may persist long after transition (Chang et al. 2007). The legitimacy of the old and/or new regimes frequently constituted the most salient axis of competition in the first, 'founding' election after the end of authoritarian rule, which in turn shaped subsequent patterns of party competition (Bermeo 1987; Reich 2001). Concentrating on the fundamental distinction of democratic versus authoritarian values, instead of bread-andbutter issues that characterize most election campaigns in established democracies, can create a polarizing political environment in which differences of opinion are conceived in terms of choices over regime principles. A number of studies highlight the salience of a democraticauthoritarian axis of competition in post-communist new democracies (Evans and Whitefield 1998; Rohrschneider and Whitefield 2009). Compared with policy decisions such as taxes and spending, the regime divide would generate and sustain a higher level of polarization because the issues involved are intrinsically less liable to compromise. This implies that some groups of 3

5 citizens may continue to take the extreme position of rejecting principles that underpin a democratic system. Comparing established and new democracies, survey-based analyses indeed affirm the presence of greater polarization in the latter countries. For example, using the World Values Survey that includes cases from Africa, Asia, Eastern Europe and Latin America as well as North America and Western Europe, Dalton observes that 'polarization along the left/right dimension is substantially greater in the less affluent and less democracies than in advanced industrial democracies' (2006:20). Another study reports that twice as many respondents in Eastern and Central European new democracies placed themselves towards each pole on the left-right spectrum as in Western Europe (McAllister and White 2007). In addition, Moreno concludes from his study of Latin American cases that 'the underlying meaning of left and right party polarization was linked with democratic issues' (1999:123). Thus, there is both theoretical and empirical basis for the proposition of ideological polarization over a regime divide in new democracies. It follows that a decline in the level of disagreement between supporters and sceptics of democratic government could serve as an indication of democratic consolidation, as new rules of the game come to find more widespread acceptance. De-polarization 'constitutes a major achievement of the new political system' in Spain (McDonough et al. 1986:752), and can be used as a standard by which to measure the success of other new democracies. While one cause of depolarization probably lies in generational placement, another can be found in greater acceptance of democracy among older cohorts through processes of political re-socialization and learning (Bermeo 1992; Montero and Torcal 1990). One can thus interpret a decrease in the proportion of self-identified radicals in new democracies as a positive development for the likelihood of democratic persistence. Another possible indicator of consolidation relates to the content rather than level of polarization. With the accumulation of democratic experience, the main axis of political competition may no longer revolve around the question of regime preferences or legitimacy, but instead shift from the 'opposing moral absolutes' of democracy versus authoritarianism to more 'modest differences in political means and ends' characteristic of established democracies (Dalton 4

6 2006:8). While ideological semantics remain the same, the substantive meanings associated with such vocabulary would change, so that 'left' and 'right' come to connote stances over issues typically debated in established democracies such as taxes, pensions or foreign policy, rather than preferences toward different regime types. Under this scenario, regardless of fluctuations in the proportion of self-identified radicals, the success of consolidation can be found is a weakening relationship between radical ideological orientations and attitudes toward democracy. Ideological extremists are certainly not absent in established democracies, but for the most part they aim at challenging certain policies within the confines of existing institutional rules rather than seek to overhaul those rules. Thus, even high levels of ideological polarization would not undermine the democratic foundations of the polity. Ideological radicalism Putting the theoretical discussion above to an empirical examination requires one to operationalize the concept of ideological radicalism. The issues which dominate political debate in different countries, and even at different times in the same country, can vary considerably. In new democracies, even the generic rubric of a democratic-authoritarian dimension involves disparate sets of questions, due for example to the nature of the preceding autocratic regime and the process of democratic transition. To address the need for a means to measure and compare ideology across countries with distinct political, economic and social trajectories, we follow many scholars in choosing the left-right schema. Given the complexity of political issues in many countries, the assumption of a uni-dimensional axis of competition is probably simplistic, but as Inglehart and Klingemann state in a seminar study, the left-right spectrum has the advantages of capturing and summarizing 'the most important issues of a given era' (1976:244). Furthermore, these semantics serve as 'a universal solvent [that] tend to absorb whatever major conflicts are present in the political system' (Inglehart and Rabier 1986:470), and are understandable and translatable in different cultural settings (Laponce 1981:24-27). Is the left-right schema applicable in the context of new democracies? A few expert questionnaires have been conducted in several new democracies and show meaningful party placements along this spatial spectrum (Benoit and Laver 2006; Huber and Inglehart 1995). More importantly for the purpose of the present study, scholars have examined the relevance and 5

7 meaning of the left-right spectrum at the mass level in new democracies Central and Eastern Europe (e.g. Markowski 1997; Tworzecki 2003), Latin America (Coppedge 1998; Kitschelt et al. 2010; Mainwaring and Scully 1995), and East Asia (e.g. Dalton and Tanaka 2007). In a comparative study of post-communist new democracies, McAllister and White point out that voters in the emerging democracies appear to understand the concept and to align themselves on it' (2007:204). A particularly dramatic example comes from a study in South Korea which reports that, whereas nearly half of South Korean respondents could not provide an answer when asked to place themselves on the ideological scale in a survey conducted under authoritarian rule, this proportional fell to just 1 percent when the same question was asked a decade after democratic transition (Shin and Jhee 2005:387). What set many new democracies apart from their more established counterparts with respect to ideology may be the absence of social cleavages politicized by competing elites. In contrast to the 'frozen' party systems that characterized many Western democracies (Lipset and Rokkan 1976; see also Knutsen 1989), similar sociological differences are often less politically salient in many new democracies (e.g. McAllister 2007), and suppression of competition under authoritarian rule meant that, with the possible exception of the former ruling party, there is no generational transmission of partisan identification. Indeed, given the often shallow roots of parties, in some southern European new democracies ideological orientations 'rushed in to fill the gap left by the collapsing social-structural supports for partisanship' (Gunther and Montero 2001:127). Given that parties in new democracies often lack mass memberships, and have been liable to undergo more frequent mergers and splits, the close association between ideological orientations and partisanship identified in several seminal studies of Western democracies is likely much weaker in new democracies. Finally, one should note that polarization is not a function of party system institutionalization (Mainwaring and Torcal 2006), meaning that even when the party system becomes more stable in a new democracy, this does not necessarily lead to an increase in moderate views. Data and variables Data for this study are taken from the World Values Survey (WVS) and the European Values Study (EVS), which contain the same wordings and response categories for questions 6

8 pertaining to regime preferences and ideological orientations. Specifically, the empirical analysis below is based on the third, fourth, and fifth waves of the WVS and the third and four waves of the EVS. While the WVS began in the 1980s, questions on regime types - assessment of rule by strongmen, technocracy, military rule, and democracy - are only available since the mid-1990s, thus restricting the time span of our inquiry. Measurement of individual respondents' ideological orientation is based on their self-placement on a 10-point scale, with higher values denoting positions toward the right. Our sample includes 24 'Third Wave' democracies in southern Europe, Latin America, East Asia, and Central and Eastern Europe. Selection is based on each country having data available for at least two points in time, with a total of 69 cases. Details are provided in Table 1. One key variable in the present study, length of democratic experience, is simply calculated by the number of years between the survey date and the year of democratic transition in each country. The latter is determined by when a country attained the status of 'free' according to Freedom House ratings, provided that it does not relapse into 'partly free' or 'not free' status later. Ideological radicals are operationalized as respondents who put themselves in the two leftmost and right-most response categories on the left-right self-placement question, namely those with scores of 1, 2, 9 and 10. The mean proportion of radicals in the sample is 19.7%, comprising of 11.1% on the right and 8.6% on the left. For the purpose of comparison, averages for established democracies in Western Europe, as measured by the third and fourth waves of the EVS, are 12.4% and 14.0%, respectively, clearly lower than in new democracies. It is also important to note that these averages belie wide variations. The proportion of radicals exceeds 40% in both Mexican cases (with extreme rightists outnumbering extreme leftists by a large margin) and was just slightly under 40% in Brazil in By contrast, less than one in ten respondents identified with a radical position in Latvia in 1996, Slovenia in 1995, and Estonia in both 1996 and Support for democratic principles is a composite index constructed by adding respondents' ratings of strongman rule, technocracy (rule by experts), military rule, and democracy. The more they disapproved of the first three regime types and approved the last one, the higher the score. This reflects the understanding that approval of a democratic form of 7

9 government by itself is insufficient for the attitudinal dimension of consolidation. Rather, it needs to be complemented and bolstered by rejection of non-democratic alternatives. As explained below, one of the primary research questions of the present study is to explore whether and how this democratic support index relates to both ideological orientations and radicalism. Empirical analysis The passage of time under a democratic form of government may not by itself be sufficient to reduce sceptism about the new rules of the game on either or both sides of the ideological spectrum, not least in view of high expectations of democratic government that transition from a long period of authoritarian rule often encourages. Hofferbert and Klingemann (1999) categorize the action of new democracies that may affect satisfaction with democratic performance into 'virtues of commission' and 'virtues of omission'; the former refer to measures to facilitate individual (especially economic) well-being while the latter denote actions practised under autocratic rule that the new regime has ceased to engage in. Building on this logic, we also include a series of economic and governance indicators that may exert an impact on both the proportion of radicals and the relationship between ideological radicalism and democratic support. These variables, taken from the World Bank, include: rule of law, voice and accountability, control of corruption, rate of GDP growth and unemployment rate. The first three measures, as part of the widely cited Governance Indicators, each range from 2.5 to -2.5, with higher values denoting better performance. Between these are reported every two years. Where the indicator for the survey year is not available, the figure for the subsequent year is used. To start with simple bivariate relationships, the correlation between the length of democratic government and the proportional of all radicals is While the negative sign points in the expected direction, meaning that the longer the democratic experience, the less prevalent extreme opinions are, this is only significant at the 90% confidence interval, indicating a rather weak linkage. Interestingly, when treating radical rightists and leftists separately, it turns out that there is a high correlation of for the former (significant at the 99% confidence interval), but an insignificant and surprisingly positive for the latter. In other words, to the extent that accumulation of democratic experience assuages extremist views at the aggregate 8

10 level, this effect is only found on the right side of the ideological spectrum. Using the log term of length of democracy, to ensure that countries with longer experience of democracy in our sample (mainly the southern European cases of Greece, Portugal and Spain) do not exert a disproportionate influence on the results, points to a similar conclusion. How does the impact of democratic experience on the prevalence of extreme ideological orientations compare with those exerted by economic and governance indicators? A multivariate regression test, with the proportion of radicals as the dependent variable, yields results shown in Table 2. Length of democratic rule is insignificant, as are unemployment and rate of GDP growth. Only control of corruption and rule of law emerge as highly significant and substantive factors. In other words, radical views are moderated not simply by having a democratic government in place, but rather by what the government does - or more precisely, what it refrains from doing. Instead of 'positive' actions such as enhancing economic productivity, creating jobs, and giving citizens a greater say in the policy-making process, Table 2 suggests that the key to reducing extremism lies in reining in illicit and unfair practices which were probably common under the previous authoritarian regime. Note that when dummy variables are inserted for each region (Central and Eastern Europe, southern Europe, Latin America, East Asia), none of these approach statistical significance and the results are unchanged. Once again, when one inserts the proportion of extreme rightists and leftists as dependent variables, different results emerge. Tables 3 and 4 show that findings from the entire sample are largely replicated when one examines the radical left, namely that corruption and the rule of law contribute to diminishing radical views. But on the right side of the ideological spectrum it is the voice and accountability variable that turns out to be significant instead. This implies that in many new democracies, extreme right orientations are associated with a sense of exclusion from the decision-making process, or lack of external efficacy, so adopting measures to enhance freedom of expression and opening up more channels of participation could reduce alienation on the extreme right. The finding that causes of left- and right-wing radicalism do not overlap is noteworthy, and gives pause to treating the two groups as a single unit. Another approach to examining the effect of the independent variables on the proportion of citizens with radical ideological positions is to separate the sample by the orientation of the 9

11 former regime, in other words, whether it was a rightist or leftist dictatorship. Categorizing authoritarian regimes by their ideological inclination carries inherent risks due to the often marked gaps between their theoretical proclamations and grim reality. With this in mind, we divided our sample into two groups: half of the cases (34) are comprised of ex-communist countries (including ones which attained independence after the collapse of communist rule), and the other half (35) had an assortment of right-wing, anti-communist dictatorships. Regression results show that in the latter set of cases, only improvements in the rule law exerts a significant impact on lowering the proportional of radicals, while the length of democratic experience has a negative albeit insignificant coefficient. In the ex-communist cases both control of corruption and the number of years under democratic government are significant, but the latter has an unexpectedly positive sign. This means that with longer experience of democracy, a higher percentage of citizens move to an ideologically extreme position. While one can speculate that democratic competition creates space for the cultivation and articulation of more diverse opinions, including radical ones, we hesitate to offer any elucidation in the absence of more detailed analysis, and only note that the magnitude of this effect is not substantial. In addition to how democratic experience, governance and economic indicators relate to the strength of ideologically radical views, it is also pertinent to look at how these factors may affect the contents of extremist views. Even if the ranks of self-professed extremists expand, this may not endanger prospects of consolidation if democracy itself no longer constitutes part of the main axis of political contention. Theoretically this could mean that democracy has been accepted by citizens across the left-right spectrum - or equally detested by both sides. Empirically, the latter scenario is not found anywhere. With this in mind, we can look at whether the relationship between radical orientations on one hand, and democratic support on the other, diminishes with longer experience of democracy. In other words, is support for a democratic form of government significantly greater on one end than the other, or greater in the centre than on the extremes, of the ideological spectrum, and if so, does this relationship change with the passage of time? Correlations between aggregate democratic support and ideological self-placements in each country are shown in Table 5. The third column tests the extent of a linear relationship, with 10

12 positive correlations denoting that support for democratic government and rejection of other regime types are higher on the right. The fourth column tests whether there is a curvilinear relationship, with negative coefficients meaning that democratic support is higher in the ideological centre (on a 1-10 scale, positions of 1, 2, 9 and 10 are coded as 1, the remainder coded as 0). With respect to the latter, in only 16 out of 68 cases are radicals significantly less supportive of democracy, and the strongest correlations only in Chile 2000 and in Greece In seven cases, including two from Slovenia, the correlation is actually positively significant, meaning that radicals are more in favour of democracy than moderates. In the majority of cases, however, there was simply no discernible linkage between extreme orientations and democratic support. The same certainly cannot be said of a linear relationship between ideological orientations and democratic support. In 27 out of 68 cases, support increases significantly as one moves to the left (negative correlations), while in 14 other cases this is true as one moves to the right (positive correlations). The implication is that democracy has indeed been a polarizing issue, with differences of opinion much more likely between leftists and rightists rather than between ideological moderates and radicals. Furthermore, many correlations between left-right selfplacement are attitudes toward democracy are quite strong and persistent, such as in Chile, Spain and Uruguay. Interestingly, these are all examples where leftists are more inclined to uphold the banner of democracy. The only country with similarly robust and consistent association between rightist positions and democratic support is the Czech Republic. The signs of the correlation coefficients are largely in the expected direction: leftists in former right-wing dictatorships and rightists in ex-communist countries - those who were most likely to have held dissenting views under autocratic rule - are more inclined to embrace democracy and reject alternative regime types. Using the correlations in column 3 of Table 5 as the dependent variable in a regression analysis yields results shown in Table 6. The length of democratic experience, control of corruption, and voice and accountability are significant at the 99% confidence interval, and the two economic indicators also turn out to exert a significant impact. At the country level, living under a democracy by itself does not reduce the proportion of radicals, but it does have an effect 11

13 in diminishing the linear relationship between ideological orientations and attitudes toward democracy. Note that the coefficient for this variable is negative, meaning that the longer a country has had a democratic government, the more leftists come to accept democratic principles. Theoretically, this could also be interpreted that longer democratic governance leads rightists to spurn the new regime, but this proposition has little empirical support. Substantively, the two governance indicators of corruption control and voice and accountability exert a larger influence, in the direction of reducing leftists' scepticism toward democracy. This finding raises intriguing questions about the how citizens who identify with the radical right and radical left in new democracies differ. The sample sizes of extremists in most countries are not sufficiently large to be divided into demographic sub-samples for further statistical analysis, but one may speculate about the reason behind the negative correlation reported above for the relationship between ideological self-placement and democratic on one hand, and the length of democratic rule on the other. There is some evidence suggesting that, in several Central and Eastern European new democracies, older adherents of former leftist dictatorships have either exited the electorate or become more reconciled to the new regime. But ironically, segments in the younger generation, who grew up since democratization and never experienced the restrictions imposed by authoritarian rule, are turning to far-right parties out of disillusionment with established parties on both the mainstream right and left. This may be part of the explanation why, a quarter-century after the collapse of the Soviet bloc, the once strong linear association between democratic support and ideological orientations has weakened. This is a topic that warrants further research. Summary and conclusion The present study has explored the relationship between radical ideological views and political system support in new democracies at the aggregate level. Support for the system is usually less deeply rooted in countries that have only recently undergone transition from authoritarian rule, but it is precisely in these countries where such support may be most crucial for the survival of democracy. Challenges to the new democratic regime are more prone to emerge from segments of the population with extremist views, as the radical right and left are more likely to comprise of either adherents of the former authoritarian regime, or its fiercest 12

14 opponents who are disenchanted by what they perceive as lack of sufficient political progress under the new system. Would the passage of time, i.e. increasing familiarity with how democracy functions in practice, induce more widespread acceptance of the new 'rules of the game'? And what other factors can play a role in facilitating the attitudinal aspect of democratic consolidation? And insofar as there is a link between ideological radicalism and support for democracy, is the relationship more likely to be linear or curvilinear? These are questions that we have attempted to address. Analysis of data from 24 new democracies, with at least two points in time from each countries, yields the expected negative relationship between length of democracy and proportion of self-identified radicals, albeit not a strong one. When extremist rightist and leftist respondents are treated separately, a significant relationship is found only for the former group (i.e. the percentage of radical right identifiers decreased over time). However, length of democratic rule does not emerge as a significant influence on the prevalence of radical opinions in a multivariate regression, and neither do economic indicators such as rates of GDP growth and unemployment. Instead, control of corruption and the rule of law exert are the independent variables that exert the most substantive impact on the proportion of self-identified radicals. Once again, however, results differ somewhat when the sample is divided into extremist right and left. For example, giving citizens a greater say helps to reduce the ranks of the radical right, but has little effect on the radical left. Where an ideological divide exists concerning attitudes toward democracy, it mostly lies between left and right, rather than between moderates and extremists on both ends of the scale. Not surprisingly, leftists in former communist regimes express greater scepticism toward democratic principles, and the same is mirrored on the other side of the spectrum in countries previously ruled by right-wing dictatorships. Also, the correlation between ideological selfplacement and attitudes toward democracy is negative, meaning that overall leftists are more supportive. Furthermore, the length of democratic experience exerts a statistically significant impact in assuaging radical leftists' doubts toward the new regime's underlying principles, though substantively several measures of good governance (control of corruption, voice and accountability) exert a greater influence. 13

15 Finally, several notes of caution are needed when interpreting results reported above. First, it is important to stress that all data are measured at the aggregate level, and one should thus avoid the risk of ecological fallacy by applying the findings to individuals. Second, we have left out two independent variables that may have considerable effect if included in the regression equations: the poverty gap and electoral systems. One would expect that a higher Gini index or institutional rules that render election outcomes less proportional (i.e. more difficult for radical parties to gain parliamentary representation) would augment the ranks of the discontented, and thus increase levels of ideological extremism. Unfortunately these two measures are not available for a large minority of cases discussed here, so we have not been able to test the existence and extent of the impact exerted by these two factors. Lastly, more detailed analysis is called for regarding the different incentives behind radical left and radical right doubters of democratic principles, and by implication the possible policy prescriptions that can be adopted to alleviate their scepticism toward the political system. 14

16 References Benoit, Kenneth and Michael Laver Party Policy in Modern Democracies. London: Routledge. Bermeo, Namcy Democracy and the Lessons of Dictatorship. Comparative Politics 24(3): Bogdanor, Vernon 'Founding Elections and Regime Change'. Electoral Studies 9(4), Capoccia, Giovanni Defending Democracy: Reactions to Political Extremism in Inter-war Europe. European Journal of Political Research 39, Chang, Yu-tzung, Yun-han Chu, and Chong-Min Park Authoritarian nostalgia in Asia. Journal of Democracy 18(3): Coppedge, Michael The Dynamic Diversity of Latin American Party Systems. Party Politics 4(4): Dalton, Russell J Social Modernization and the End of Ideology Debate: Patterns of Ideological Polarization. Japanese Journal of Political Science 7(1): 1-22 Dalton, Russell J. and Aiji Tanaka The Patterns of Party Polarization in East Asia'. Journal of East Asian Studies 7(2): Diamond, Larry Developing Democracy: Toward Consolidation. Baltimore: Johns Hopkins University Press. Druckman, James N. and Andrew Roberts Communist Successor Parties and Coalition Formation In Eastern Europe. Legislative Studies Quarterly 32(1), 5-31 Easton, David A Framework for Political Analysis. Eaglewood Cliffs: Prentice-Hall. Evans, Geoffrey and Stephen Whitefield The Structuring of Political Cleavages in Post- Communist Societies: the Case of the Czech Republic and Slovakia. Political Studies 46(1): Grzymala-Busse, Anna Coalition Formation and the Regime Divide in New Democracies: East Central Europe. Comparative Politics 34(1), Gunther, Richard and Jose R. Montero The Anchors of Partisanship: A Comparative Analysis of Voting Behavior in Four Southern European Democracies. In P. Nikiforos Diamandouros and Richard Gunther (eds.), Parties, Politics, and Democracy in the New Southern Europe. Baltimore: Johns Hopkins University Press. 15

17 Hofferbert, Richard I. and Hans-Dieter Klingemann Remembering the Bad Old Days: Human Rights, Economic Conditions, and Democratic Performance in Transitional Regimes. European Journal of Political Research 36(2): Huber, John and Ronald Inglehart Expert Interpretations of Party Space and Party Locations in 42 Societies. Party Politics 1(1): Inglehart, Ronald and Hans Klingemann Party Identification, Ideological Preference and the Left-Right Dimension among Western Mass Publics. In Ian Budge, Ivor Crewe, Dennis Farlie (eds.), Party Identification and Beyond. London: Wiley. Inglehart, Ronald and Jacques-René Rabier Political Realignment in Advanced Industrial Society: From Class-Based Politics to Quality-of-Life Politics. Government and Opposition 21(4): Kitschelt, Herbert, Kirk A. Hawkins, Juan Pablo Luna, Guillermo Rosas, Elizabeth K. Zechmeister Latin American Party Systems. Cambridge: Cambridge University Press. Klingemann, Hans-Dieter 'Post-autocratic Party Systems and the Regime Cleavage in New Democracies'. In Ursula Van Beek (ed.), Democracy Under Construction: Patterns from Four Continents. Bloomfield Hills: B.Budrich. Knutsen, Oddbjørn Cleavage Dimensions in Ten West European Countries: A Comparative Empirical Analysis. Comparative Political Studies 21(4): Laponce, J.A Left and Right: The Topography of Political Perceptions. Toronto: University of Toronto Press. Lipset, Seymour M. and Stein Rokkan Cleavage Structures, Party Systems, and Voter Alignments: An Introduction. In Seymour M. Lipset and Stein Rokkan (eds.), Party Systems and Voter Alignments: Cross-National Perspectives. New York: Free Press. Linz, Juan and Alfred Stepan The Breakdown of Democratic Regimes. Baltimore: Johns Hopkins University Press. Linz, Juan and Alfred Stepan Problems of Democratic Transition and Consolidation: Southern Europe, South America, and Post-Communist Europe. Baltimore: Johns Hopkins University Press. Mainwaring, Scott and Timothy Scully Building Democratic Institutions: Party Systems in Latin America. Stanford: Stanford University Press. 16

18 Markowski, Radoslaw Political Parties and Ideological Spaces in East Central Europe. Communist and Post-Communist Studies 30(3): McAllister, Ian Social Structure and Party Support in the East Asian Democracies. Journal of East Asian Studies 7: McAllister, Ian and Stephen White Political Parties and Democratic Consolidation in Post-Communist Societies. Party Politics 13(2): McDonough, Peter, Samuel H. Barnes, Antonio López Pina The Growth of Democratic Legitimacy in Spain. American Political Science Review 80(3): Merkel, Wolfgang The Consolidation of Post-Autocratic Democracies: A Multi-level Model. Democratization 5(3), Montero, Jose R. and Mariano Torcal Voters and Citizens in a New Democracy: Some Trend Data on Political Attitudes in Spain. International Journal of Public Opinion Research 2(2): Moreno, Alejandro Political Cleavages: Issues, Parties, and the Consolidation of Democracy. Boulder: Westview Press. Payne, Stanley G The Collapse of the Spanish Republic, : Origins of the Civil War. New Haven: Yale University Press. Reich, Gary M Coordinating Party Choice in Founding Elections: Why Timing Matters. Comparative Political Studies 34(10): Rohrschneider, Robert and Stephen Whitefield Understanding Cleavages in Party Systems: Issue Position and Issue Salience in 13 Post-Communist Democracies. Comparative Political Studies 42(2): Rose, Richard, William Mishler, Christian Haerpfer Democracy and its Alternatives: Understanding Post-Communist Societies. Cambridge: Polity Press. Shin, Doh Chull and Byong-Kuen Jhee How Does Democratic Regime Change Affect Mass Political Ideology: A Case Study of South Korea in Comparative Perspective. International Political Science Review 26(4): Tworzecki, Hubert Learning to Choose: Electoral Politics in East-Central Europe. Stanford: Stanford University Press. 17

19 Table 1: Proportion of radicals in new democracies country survey/ wave year of transition survey year years of democracy (C-D+1) % all radicals % radical right % radical left ARGENTINA WVS ARGENTINA WVS ARGENTINA WVS BRAZIL WVS BRAZIL WVS BULGARIA WVS BULGARIA EVS BULGARIA WVS Bulgaria EVS CHILE WVS CHILE WVS CHILE WVS CZECH REPUBLIC WVS CZECH REPUBLIC EVS Czech Republic EVS ESTONIA WVS ESTONIA EVS Estonia EVS Greece EVS Greece EVS HUNGARY WVS HUNGARY EVS Hungary EVS INDONESIA WVS INDONESIA WVS LATVIA WVS LATVIA EVS Latvia EVS LITHUANIA WVS LITHUANIA EVS Lithuania EVS MEXICO WVS MEXICO WVS PERU WVS

20 PERU WVS PHILIPPINES WVS PHILIPPINES WVS POLAND WVS POLAND EVS POLAND WVS Poland EVS PORTUGAL EVS Portugal EVS ROMANIA WVS ROMANIA EVS ROMANIA WVS Romania EVS SLOVAKIA WVS SLOVAKIA EVS Slovakia EVS SLOVENIA WVS SLOVENIA EVS SLOVENIA WVS Slovenia EVS SOUTH AFRICA WVS SOUTH AFRICA WVS SOUTH AFRICA WVS SOUTH KOREA WVS SOUTH KOREA WVS SOUTH KOREA WVS SPAIN WVS Spain EVS SPAIN WVS SPAIN WVS Spain EVS TAIWAN WVS TAIWAN WVS URUGUAY WVS URUGUAY WVS Note: Self-placements of 1, 2, 9 and 10 on a 1-10 left-right scale are coded as radical. Tabulated from the World Values Survey (WVS) and European Values Study (EVS) 19

21 Table 2: Factors affecting overall proportional of radicals B standard error years of democracy control of corruption ** voice & accountability rule of law ** unemployment rate GDP change constant adjusted R square.360 *** p< 0.001; ** p< 0.01; * p<< 0.05 Table 3: Factors affecting proportional of radical rightists B standard error years of democracy control of corruption voice & accountability * rule of law unemployment rate GDP change constant adjusted R square.442 *** p< 0.001; ** p< 0.01; * p<< 0.05 Table 4: Factors affecting proportional of radical leftists B standard error years of democracy control of corruption ** voice & accountability rule of law *** unemployment rate GDP change constant

22 adjusted R square.150 *** p< 0.001; ** p< 0.01; * p<< 0.05 Table 5: Correlations with years of democracy country survey year years of democracy correlation with L- R placement correlation with extreme placement ARGENTINA * ** ARGENTINA ** ARGENTINA ** BRAZIL ** * BRAZIL BULGARIA BULGARIA ** BULGARIA Bulgaria * CHILE ** CHILE ** ** CHILE ** CZECH REPUBLIC ** CZECH REPUBLIC ** * Czech Republic ** ** ESTONIA ** ESTONIA ** Estonia Greece ** ** Greece ** HUNGARY HUNGARY ** Hungary INDONESIA ** INDONESIA ** ** LATVIA * ** LATVIA Latvia * * LITHUANIA ** LITHUANIA Lithuania * MEXICO ** ** MEXICO * ** PERU

23 PERU PHILIPPINES PHILIPPINES * POLAND POLAND Poland * PORTUGAL ** Portugal ROMANIA * * ROMANIA ROMANIA * Romania SLOVAKIA * * SLOVAKIA * ** Slovakia SLOVENIA SLOVENIA * SLOVENIA ** Slovenia * SOUTH AFRICA ** SOUTH AFRICA SOUTH AFRICA ** ** SOUTH KOREA ** SOUTH KOREA * SOUTH KOREA SPAIN ** Spain ** SPAIN ** SPAIN ** Spain ** * TAIWAN ** TAIWAN URUGUAY ** * URUGUAY ** Note: 1-10 ideological scale (0= left; 10= right); extreme placement = 1, 2, 9 and 10 on ideological scale. 22