Forms of government in post-communist countries

Forms of government in post-communist countries Klaus Armingeon University of Berne Switzerland April 2005 Paper prepared for the ECPR Joint Workshops, Workshop Dominant Parties (M. Bogaards) Granada, Spain, April 2005 Abstract: This paper deals with the forms of government in 28 post-communist nations, 1989-2004. Applying the findings on the forms of government from Western European countries to the post-communist nations, one expects three different groups: In democratic countries with many societal cleavages there will be a large number of political parties that are forced into a politics of compromise. One indicator of this politics of compromise are oversized coalitions. In democratic countries with few cleavages there will be a small number of political parties. These parties can compete according to a majoritarian model. Finally, if countries are not democratic to a large extent, there will be great likelihood of governments by dominant parties. All these expectations are not borne out by the data. There is neither a sizeable group of consensual governments nor of dominant parties. This is due to a lack of a dominant position of political parties in the political process or to a lack of strong and durable linkages between citizens and parties. Since the premises of the application of these models to the postcommunist world are not met, it is little wonder that these models do not reflect political realities in these political systems 1

I Introduction My theme are forms of governments and their underlying modes of party competition in post-communist countries. Do political parties pursue a politics of accommodation and build grand coalitions in fragmented societies? Or do they opt for a competitive style of politics and for alternating single party government (or minimal winning coalitions), such as known from the Anglo-Saxon democracies? Or do some of them succeed in dominating the party system and therewith government? These are the guiding questions of this paper. The underlying problematique concerns the interaction of the party system with citizens and socio-cultural groups to which they belong. In Continental-European countries societal fragmentation frequently has led to a situation in which no single party has had the chance to get an majority in parliament. Even if one of them would receive a little bit more than 50% of seats in parliament it would not be wise to govern by using this majority. Powerful minorities that have never a chance to become a majority would feel suppressed. Based on historical experience politicians would have reason to worry about secession or internal war. An example for the first situation is the Netherlands in the early 20 th century. An illustration for the latter case is Austria after the Second World War, when politicians realized that both Lager the catholic-bourgeois and the secular-left groups of society were of similar size. None of them would be able to dominate the other as long as elites were not willing to repeat the experience of the civil war between the Lager in the inter-war period. As a result for a long time majority rule has not become the dominant decision technique in the political systems of countries such as Austria, Netherlands or Switzerland. Rather parties try to reach agreement by compromise. This has become know as Konkordanzdemokratie (Switzerland), Proporzdemokratie (Austria) or pillarization (Netherlands). In these countries parties joined in grand oversized coalitions. 2

Such government by negotiation between parties can be avoided, as the British case showed, if there a few societal cleavages and a two party system with each party having a fair chance of alternating in power. This frequent alternance reduced the danger of long term suppression of one major party and its constituency. This mode of party competition a small number of parties compete for office and there is alternance for parties in single-party governments or governments of minimal winning coalitions -- has been labeled Westminster democracy. A crucial precondition for both modes of party competition is a democratic order allowing for free party competition. If democracy is defect there is some likelihood of dominant parties. These parties have a majority in parliament over a long period, giving competitors no chance of coming to power. This is the case in many African nations. Dominant parties may emerge in democratic nations such as Japan too. In these cases, competition takes place within the dominant party. However, the very idea of a dominant party does not correspond to the notion of democracy based on effectively competing political parties. Generalizing these observations to the post-communist countries, three predictions follow: In democratic nations with many cleavages, many parties will emerge. They have strong incentives to work together in grand or oversized coalitions. In democratic nations with little cleavages the party system will developed towards a two or three parties systems. Single party governments or minimal winning coalitions will result. If the political system is not fully democratic though, effective party competition may be stalled and one party has the chance to become dominant. Hence in defect democracies or autocratic regimes the likelihood of dominant parties will be great. In this paper I take issue with this plausible application of Western experiences to the post-communist countries. I will show that cleavages are not related to the mode of 3

governance and I will show that in the defect democracies of the post-communist world there is greater likelihood for multi-party systems than for dominant parties. My argument is that one crucial link is missing: The basic assumption underlying all three predictions is that political parties are dominant in organizing political conflict and they do so by representing homogeneous societal groups. If this is not the case there is no reason to assume that the Western model will work in the post communist nations. I will present evidence that to a large extent this linkage between parties and citizens is still missing in the post-communist nations. There is no evidence for an ubiquitous increase in the strength of formerly weak linkages in this part of the world. 2 Hypotheses The theory linking societal cleavages and the use of majority rule as dominant decision technique has been developed by Gerhard Lehmbruch and Arend Lijphart (Lijphart, 1968). Their conceptual work is based on an analysis of the political systems of Austria, Switzerland, and the Netherlands. These democracies have been labeled consociational. The theory has been further developed and modified taking into account obvious difficulties to operationalize the concept so that it can be used in systematic comparative studies. In particular Arend Lijphart transformed the theory from the politics of accommodation to the theory of consensus versus majoritarian democracy (Lijphart 1984; Lijphart 1999). This has met many criticisms (e.g. Bogaards 1998; Bogaards 2000). Critizing Lijphart s analysis Ganghof makes the point that institutional aspects (e.g. electoral laws) and behavioral aspects (e.g. the building of coalitions) are mixed up and that the causal sequences are disregarded (Ganghof 2004). I do not have to enter that debate on the development of the theory here. Rather for my research question I start from the original argument made most clearly by Lehmbruch (Lehmbruch 1967): In fragmented societies that are politically organized by 4

political parties and interest groups majoritarian techniques are either not feasible or highly risky. Therefore in a process of learning and supported by the development of corresponding institutions, elites in these societies tend to opt for decision-making by compromise (Lehmbruch 1993; Lehmbruch 1996). An operationalization of these politics of compromise are grand or oversized coalitions. 1 The opposite ideal type of democracies are competitive or Westminster democracies. In the ideal type construction, these societies do not have more than one major cleavages. If there is any cleavage as in the UK -- it is organized by two competing parties, alternating in government. In other nations there are recurrent social conflicts but they are not or only weakly organized by major political parties being spokesmen of different societal sections and having a clear programmatic profile corresponding to social groups. This is the case of the USA. Post-communist nations are frequently fragmented along lines of class, ethnicity, religion and language (see Alesina, Devleeschauwer et al. 2003; Fearon 2003). Shortly after the transition to democracy, in these countries political parties developed programmatic profiles similar to those in Western Europe (Budge, Newton et al. 1997: 217-220; Batory, 2004; Bielasiak, 1997; Kitschelt, 1999), or they have developed particular configurations of programatic elements (Marks, Hooghe et al. 2004). Applying the arguments by Lehmbruch and Lijphart to Eastern European countries, one would expect that in countries with many cleavages the likelihood for consociational modes of government is high. In contrast in countries with only one or two cleavages majoritarian governments are to be expected. Recent analyses by Jessica Fortin (Fortin 2004) and Andrew Roberts (Roberts 2004) use the concept of consenus democracy and its operationalization by Lijphart. They show that there is no consensus (and hence in all likelihood no consociational democracy) in the post- 1 Lijphart (Lijphart, 1999) argues that minority governments are similar to oversized coalitions since in both systems pliticians are forced to negotiate for reaching a majority. This may be true. But minority governments may also survive because they can create alternating majorities with ad hoc compromises and, in addition, they may be successful in reaching plurality simply because the opposition is strongly fragmented. Hence, in this paper I consider only oversized coalitions as an indicator of durable politics of compromise. 5

communist world. This comes as a surprise only if one expects that cleavages are immediately processed to electoral laws, interest groups systems, party systems, balanced distribution of power between parliament and executive, and particular coalition behavior in forming the government. This is far from plausible, though. The historical analyses of consociational democracies convincingly showed that this particular elite behavior is learnt during a longer period only a few years after the abandonment of the communist system. Hence there is no reason to expect on average a consociational mode of government in postcommunist countries. Rather, a sensible test of the Lehmbruch-Lijphart thesis has to rest on a diachronic analyses. According to these authors we do not expect that post-communist political systems in societies with many cleavages are fully developed as consociational systems only a few years after the breakdown of communism. Rather we would expect that in post-communist countries in the early 2000s there are more elements of consociational politics than had been shortly after the break-down of the communist system. Hence we expect a development towards consociationalism, but no fully developed consociational system. A crucial precondition for these developments towards majoritarian or consociational government is party competition. If competition is limited neither majoritarian nor consociational patterns of government will emerge. The most likely result in these cases will be dominant parties that govern the country without alteration in power and without effective challenger parties. Dominant parties have been defined as parties that has won a parliamentary majority plus the presidential elections, where present, in three consecutive multi-party elections (Bogaards 2004: 175). Dominant parties need not be non-democratic. For example established democracies such as Japan have had dominant parties. But there is a great likelihood that dominant parties are at the same time autocratic parties. Either they use their power to keep down competitors or they have come into being because challengers have been suppressed (cf. Bogaards, 2004: 192). 6

Therefore we would expect that in non-democratic countries the likelihood of dominant parties is high. Non-democratic countries lack civil liberties (such as the right to form political parties) and political rights (such as free and fair elections). In sum we expect the following pattern of forms of government or modes of party competition by level of democracy and size of cleavages. See Table 1. Table 1: Hypothetical distribution of types of government/mode of party competition Many cleavages Few/no cleavages Democratic System Consociational government Majoritarian government Non-democratic System Government by dominant party Government by dominant party All these arguments make an -- as I will show unwarranted assumption: That parties organize social conflicts and that they are the dominant actors in the political process. At least one of both assumption is met or has been met in all established democracies. Western European party systems have evolved out of societal cleavages (Lipset and Rokkan 1967; Bartolini/Mair 1990). Even if they currently have lost much of their closeness to societal segments (Franklin, Mackie et al. 1992; Dalton et al. 1985; Dalton, 2002), they still bear the imprints of this former links between homogeneous socio-cultural groups and political parties. In other words there have been (or still are) stable coalitions between parties and citizens. 7

Parties that have not been representing homogeneous societal groups such as in the USA are at least the major organizing factor of the political process: they nearly monopolize the access to parliamentary seats and cabinet posts. In addition they have a monopoly for political recruitment and political socialization. They can do so because they are considered to be legitimate spokesmen of their constituencies in the electorate. I argue that this is not true for post-communist countries. Although being programmatic parties, parties have failed to build up stable coalitions with groups of the electorate. That applies to programmatic as well as to clientelistic coalitions (Kitschelt, 2000). If linkages exist this is more often the case on an individual basis be it charismatic or clientelistic between a particular politician and citizens. In that regard a crucial link is lacking in the post-communist countries. This explains why cleavages do not lead to consociational forms of governance and why a lack of democracy does not lead to dominant parties. The idea of a missing link is not new. This point has been made most clearly and forcefully by Ingrid van Biezen in a recent contribution (Biezen, 2005 see also Olson 1998; Mair 1998). She argues that the logics of party politics converge in the West and the East. In the West, previous coalitions between parties and citizens deteriorate; in the East they never have existed. However, Biezen bases her analyses on a small empirical basis (Hungary and the Czech Republic for the post-communist countries). It is far from clear whether this applies to other post-communist countries, though. I will show that her findings are confirmed if all post-communist nations are taken into account. 3 Data and research design The research question implies a comparative research design covering all the postcommunist nations. Considering the idea of successive learning of elites, I need data over 8

time too. Therefore my research design requires a matrix of data across nations and across time. This applies to the dependent variables, i.e. indicators of consociational / majoritarian governments and indicators of dominant parties. For other independent variables that do not vary strongly in the short term (e.g. ethnic fractionalization or the general level of political rights and civilian liberties) averages for the whole period under consideration or measurements at one point in time (e.g. ethnic fractionalization) will do. For my argument, I would need data on the evolution of coalitions between electorates and parties, though. Unfortunately, these time series are not available in many cases. However, the size of respective indicators measured at the mid/end 1990s can be gauged by data for established Western countries. In the 1990s these countries have suffered a significant reduction of formerly strong ties between parties and electorates. Hence if the magnitude of indicators of coalitions is as low or lower than in the West this serves as empirical evidence for a lack of such coalitions in the East. The dependent variables are forms of governments: consociational government (includes all significant societal segments and their respective political parties), majoritarian government (government is formed by the largest party or a minimal winning coalition), and government by a dominant party. The indicators for majoritarian or consociational governments are the size of the governing coalition. If more parties are in government than are needed for a minimal winning coalition this can be considered a consociational ( grand coalition, oversized majority) government. On the contrary, minimal winning coalitions, single party governments, minority governments and presidential cabinets are coded as non consociational governments. They do not intend to represent all the significant social sections even if these are not momentarily need for reaching a plurality of seats in parliament. With regard to dominant parties in post-communist nations it does not make much sense to apply the criterion of a majority of seats in three consecutive elections considering the fact that first elections took place only in the early to mid 1990s. Therefore I relaxed the 9

criterion for a dominant party to six consecutive years having more than 50% of seats in parliament. The two major independent variables are the strength of societal fractionalization and the level of democracy. The former is measured based on a data set by Alesina et al (Alesina, Devleeschauwer et al. 2003). I created a summary indicator being the mean of the fractionalization scores for ethnic groups and for religion. These data have been collected for the period between 1989 and the early 2000s. The indicator varies between 1 and 0. 1 denotes a situation in which each individual belongs to a separate ethnic and religion group; 0 characterizes a situation in which all individuals belong to the same ethnic and religious group. The calculation follows the rational of Rae s index of party fragmentation (Rae 1968). My operational definition of democracy is based on the rankings of political rights and civil liberties by Freedom House (www.freedomhouse.org). I calculate an indicator of the level of democracy. It is the combined score of the Freedom House scales for political rights and civil liberties. Both scales vary between 1 (free, democratic) and 7 (non free, not democratic). I added both scales, subtracted two and divided by two. Now the combined scale runs from 0 (free and democratic) to 6 (not free and not democratic). For the missing link is used various indicators: (a) Share of independent candidates in parliament. The larger the share of independent candidates the less the central role of political parties in the political process. (b) Net volatility. This is sum of the amount of the difference between the votes for party i at election j and election j-1 divided by two. The major drawback of this measure is that it does not measure inter-block volatility but only inter-party volatility (Bartolini, 1990; Kitschelt et al., 1999: 400). However, a large or increasing net inter-party volatility is hardly compatible with a hardening coalition between segments of society and parties. As in case of the share of independent members of parliament this 10

indicator has the major advantage to be available for the whole period under consideration and for all countries. [c] Party attachment: This is the share of respondents in surveys that feel close to a particular political party. The larger the share of citizens with party attachment, the larger and the more stable the programmatic coalition between electorates and political parties. (d) Active party membership. In nations with strong links between societal groups and political parties we expect a larger share of citizens being active members of a party than in countries where this link is missing. (e) Trust in parties. The voters lack of trust in parties denotes a lack of coalition between voters and parties. The major disadvantage of this and the former two indicators (party attachment and active party membership) is the lack of reliable time series and an incomplete coverage of all post-communist countries. Hence as an auxiliary measure I calculated the comparable figures for established democracies in order to see if trust, membership and closeness is more marked in post-communist countries than in the modernized and individualized Western nation. If that is not the case I can conclude that the post-communist political system have not yet reached a level of alignment that even Western countries have after a prolonged period of deterioration of former links. The universe of my comparative analysis are the 28 post-communist political systems in the period between the first election and the last election or government formation for which data are available. Most countries hold their first (founding) election between 1990 and 1992. Turkmenistan elected its first post-communist government only in 1999. Data for democracy, parties, coalitions and elections come from Armingeon, Careja et al. 2004. This data base extends to 2004, but the last election for which results are entered into the data base happened in 2003. 11

Data for closeness to political parties are calculated from two surveys: The European Social Survey (ESS) of 2004 and the survey Citizens, Involvement, Democracy (CID) conducted by Jan van Deth and collaborators in the early 2000s. The share of active party members and the extent of confidence in political parties are calculated on the basis of the third and fourth wave of the World Values Surveys, conducted in the mid 1990s and in the early 2000s. Univariate survey statistics have been calculated from weighted data. 4 The missing correlation: cleavages, democracy and mode of party competition Crosstabulating the level of democracy and the extent of societal fragmentation we get the following distribution (Graph 1): Graph 1 Democracy and Fractionalization COMPUTE Comb_Frac = MEAN (frac_ethnic,frac_religion) (COMPUTE) 0.70 0.60 0.50 0.40 0.30 0.20 0.10 Bosnia-Herzego Georgia Kazakhstan Latvia Kyrgyz Republi Yugoslavia Estonia Moldova Ukraine Czech Republic Bulgaria Belarus Tajikistan Slovakia Croatia Lithuania Russia Hungary Albania Azerbaijan Uzbekistan Armenia Romania Turkmenistan Slovenia Mongolia Poland 0.00 1.00 2.00 3.00 4.00 5.00 6.00 COMPUTE Democracy = ((fh_pr + fh_cl)-2)/2 (COMPUTE) 12

Towards the upper left corner, countries are expected to come close to the consociational type of party competition; towards the lower left corner countries should increasingly show signs of majoritarian modes of government. In the right part of the figure countries with dominant parties are expected. None of these expectations is borne out by the data. Considering the trends in government formation, these results can be found 2 : Ten countries are governed by a presidential system in which the president nominates his cabinet without being effectively hindered by a parliamentary opposition (Armenia, Azerbaijan, Belarus, Georgia, Kazakhstan, Kyrgyz Republic, Russia, Tajikistan, Ukraine, Uzbekistan). In five parliamentary democracies there was no single instance of a oversized coalition (Croatia, Czech Republic, Estonia, Macedonia, Mongolia). Eight parliamentary democracies have been governed at least one year by a grand coalition. In three of them, these experiences have been made at the beginning of the period under consideration (Bulgaria, 1991, Romania 1992, Poland 1994). In another four cases, grand coalitions happened towards the end of the period under consideration. But these cases have been limited in duration (Latvia 1996, 2000; Lithuania 1997-2000; Slovakia 1999-2002; Slovenia 1993, 2001-2002). Only in one country there was a longer period of grand coalition but that ended in 1998 (Hungary 1990-1998). Therefore we can confirm the findings by Fortin and Roberts, that there is no sizeable consociational government in postcommunist nations and we add that more importantly there is no tendency towards consociationalism. This applies even for countries with strong societal fractionalization and a large number of effective parties (the Baltic countries, Romania, Slovenia, Slovakia and the Czech Republic). Given the fact that in all but one post-communist countries a majoritarian style of government be it majoritarian-parliamentary or presidential dominates and taking into account that in many of these countries democracy is defect, we would not be surprised to 2 Cases have been coded for the dominant type of government during a year and the discussion refers to these coded information.data for Moldova and for Bosnia-Herzegowina are missing. 13

find many dominant parties. But this is not the case neither. It is only in five nations that one party has had more than 50% of seats in parliament for six or more consecutive years: Albania 1997- (Socialist Party); Azerbaijan 1995- (New Azerbaijan Party); Croatia 1990-99 (Croatian Democratic Union); Mongolia 1990-95 (the communist Mongolian People s Revolutionary Party) and Turkmenistan 1999- (the communist Democratic Party of Turkmenistan) 3. Although many post-communist nations have politically organized societal cleavages they do not develop consociational governments and although many of these nations have serious deficits with regard to the democratic quality of their political systems, they hardly have dominant parties. Why do explanations of forms of government fail in post-communist nations whilst they have been plausible in other parts of the world? 5 The missing link The answer to this question can be found in premises that are frequently not met in these countries: that parties are the dominant actors organizing the political process and that they build up stable coalitions between citizens and their own programmatic appeal. In 11 out of 28 post-communist countries more than 10 % of all parliamentary seats are held by independent politicians, i.e. members of parlament not affiliated to a particular political party. Taking the Kyrgyz Republic as an example, these are often successful businessman to whom voters have a clientelistic relationship. Calculated over the whole sample of post-communist nations, the share of seats of independents receded from about 18 % in the mid 1990s to about 14 % in the early 2000.. 3 Turkmenistan held its first election only in 1999. There are no other political parties scoring votes than the communist Democratic Party of Turkmenistan. 14

Graph 2 shows the development of the share of independents over time in those countries that have about more than a quarter of representatives from non-party list: Graph 2: Share of members of parliament not affiliated to a political party Mittelwert Share of seats of independents 100.00 80.00 60.00 40.00 20.00 country Armenia Azerbaijan Belarus Kazakhstan Kyrgyz Republi Russia Tajikistan Ukraine Uzbekistan 0.00 1990 1991 1992 1993 1994 1995 1996 1997 year 1998 1999 2000 2001 2002 2003 2004 In three of these countries this share increases over time; in the remaining 6 countries the data can be interpreted as a decline of the share of independents. But in all of the nations except Russia the share remains above a fifth of all members of parliaments. Hence there is little likelihood of dominant parties since in many post-communist nations parties still have difficulties to be the dominant players in the political process. Crucial for applying the concept of consociational democracies to post-communist nations is a second assumption: there are stable linkages between parties and citizens. These 15

linkages take time to develop, in particular if they are programmatic or clientelistic. The chances of political elites to represent the particular interests of their constituencies depend crucially on the linkage of these societal segments to the parties. In the post-communist world these linkages did not yet develop to a substantial extent, and there are at best weak hints that they will develop in the near future. This conclusion is based on the inspection of a number of indicators: The first indicator is net electoral volatility. Graph 3 shows electoral volatility in post communist countries after 1989 in comparison to established democracies in the OECD country group, 1989-2002. Graph 3: Electoral volatitlity Volatility 63.7 66.1 61.0 70.1 72.474.0 74.474.8 35.2 37.4 31.231.7 52.5 53.253.454.5 49.0 43.8 46.0 24.3 13.414.716.016.717.718.4 19.6 12.4 8.8 9.8 9.810.010.410.8 5.8 6.3 6.4 7.3 7.5 7.5 2.4 USA UK Switzerland Spain Australia Germany Portugal Ireland Belgium Denmark Greece Luxembourg Austria Finland Sweden Norway France Iceland Netherlands Canada New Zealand Japan Italy Czech Rep. Albania Hungary Georgia Romania Bosn.Herzeg. Bulgaria Slovenia Croatia Russia Ukraine Estonia Poland Latvia Slovakia Armenia Moldavia Lithuania It is no surprise that in new democracies volatility is larger than in established democracies. In the post communist nations citizens simply did not have the chance to build up strong ties to political parties over a long period of time. But the difference between both 16

country groups are very strong, indicating that in post communist nations the party system is far from being established to a similar extent as in the old democracies. In addition there are no sights of receding volatility. It is only in Estonia, Moldavia and Slovakia that volatility decreases over at least three elections. In all other post-communist countries for which data are available there is no decline at all. Hence the indicator of net volatility does not give any evidence that stable and sizeable links between electorates and political parties are growing. The great advantage of the indicator of volatility is its availability across all nations and across time. Unfortunately other indicators for linkages are not available as time series. But we can use the data from established democracies as bench-marks. From these democracies particularly in Western Europe -- we have evidence that until the 1950s or 1960s such coalitions have existed. These coalitions have been forming the party system and its mode of party competition. Since the 1960s or 1970s there are hints of a decline of these coalitions: Party attachment, party membership, and confidence in parties are eroding in the established democracies (Dalton, Flanagan et al. 1985; Dalton 2002; Franklin, Mackie et al. 1992). Therefore if links between citizens and parties are emerging in post-communist countries the respective indicators are expected to be not much lower than the values for the Western countries where the links are in a process of decline. 17

Graph 4: Closeness to political parties (party attachment), early 2000 s Closeness to political parties 69.1 66.5 67.4 54.1 56.4 56.5 57.2 59.1 59.2 59.2 59.8 50.7 47.8 48.5 45.5 45.8 43.1 41.8 38.6 39 35.2 24.7 26.2 26.8 28.1 30.2 31.1 Germany East (1) Slovenia (1) Russia (1) Poland (2) Spain (1) Romania (1) Slovenia (2) Portugal (1) Moldova (1) Switzerland (1) Germany West (1) Ireland (2) Czech Republic (2) UK (2) Spain (2) Switzerland (2) Finland (2) Greece (2) Israel (2) Netherlands (2) Denmark (1) Hungary (2) Norway (2) (1) CID (2) ESS Respondents (%) saying that they feel close to a political party Graph 5: Active membership in political parties, 1995-2001 Netherlands (1) Norway (1) Sweden (2) Portugal (2) Active party membership 6.8 1.2 1.8 2 2.1 2.6 3.2 Armenia Azerbaijan Georgia New Zealand Australia Norway Switzerland Respondents (%) saying that they are active members of a political party 18

Graph 6: Confidence in political parties, 1995-2000 Confidence in political parties 53.5 27.6 27.6 29.4 25.9 22.9 23.2 24.0 32.8 33.7 9.5 6.0 14.6 14.6 15.7 15.8 18.2 New Zealand Macedonia Bosnia Herzegovina Serbia Australia Armenia Japan USA Canada Moldova Montenegro Spain Switzerland Albania Norway Georgia Azerbaijan Respondents saying that they have a great deal or quite a lot of confidence in political parties Graphs 4 to 6 contain the respective information for party attachment, active party membership and confidence in parties for post-communist countries and established democracies. For the indicators attachment to parties and active party membership the general result is that in post-communist nations citizens have not yet build up strong and durable ties to political parties as this has been the case in Western Europe in the past. In the postcommunist world electors are at least as dealigned to political parties as electors are now after a long process of dealignment in the West. Both in East and West trust in political parties is low, but with regard to this indicator the post communist countries do not have generally lower values than the established democracies. 19

5 Conclusions Applying the findings on the forms of government from Western European countries to the post-communist nations, one expects three different groups: In democratic countries with many societal cleavages there will be a large number of political parties that are forced into a politics of compromise. One indicator of this politics of compromise are oversized coalitions. In democratic countries with few cleavages there will be a small number of political parties. These parties can compete according to a majoritarian model. Finally, if countries are not democratic to a large extent, there will be great likelihood governments by dominant parties. All these expectations are not borne out by the data. There is neither a sizeable group of consensual governments nor of dominant parties. This is due to a lack of a dominant position of political parties in the political process or to a lack of strong and durable linkages between citizens and parties. Since the premises of the application of these models to the postcommunist world are not met, it is little wonder that these models do not reflect political realities in these political systems In that sense, political parties in post-communist nations are different from political parties and their competition in Western Europe (Mair 1998). However this does not mean that political parties in post-communist nations are totally different. Rather, they develop a programmatic appeal (Kitschelt, Mansfeldova et al. 1999). This can be observed the better the more we move to the West in the post-communist region. But this programmatic appeal does not include durable links between electors and parties in the sense that parties represent socio-cultural group as this has been the case in Western Europe. Rather parties offer a program to citizens and these are making a choice that is highly volatile in time. Hence the 20

programmatic structuring of the party system resembles the Western European pattern while the mode of party competition is similar to that in the USA (Olson 1998). Therefore the political process even in the new EU countries may be more alike to the logic of politics in the USA than to the past logic of politics in many of the established Western European countries. 21

References: Alesina, Alberto; Devleeschauwer, Arnaud; Easterly, William; Kurlat, Sergio; Wacziarg, Romain (2003): Fractionalization. In: Journal of Economic Growth 8: 155-194. Armingeon, Klaus; Careja, Romana (2004): Comparative Data Set for 28 Post-Communist Countries, 1989 2004. Bern: http://www.ipw.unibe.ch/mitarbeiter/ru_armingeon/cpd_set_en.asp#cpdsii. Bartolini, Stefano; Mair, Peter (1990): Identity, Competition, and Electoral Availability. The stabilisation of European electorates 1885-1985. Cambridge: Cambridge University Press. Batory, Agnes; Sitter, Nick (2004): Cleavages, Competition and Coalition-Building: Agrarian Parties and the European Question in Western and East Central Europe. In: European Journal of Political Research 43: 523-546. Bielasiak, Jack (1997): Substance and Process in the Development of Party Systems in East Central Europe. In: Communist and Post-Communist Studies 30/1: 23-44. Biezen, Ingrid van (2005): On the Theory and Practice of Party Formation and Adaptation in New Democracies. In: European Journal of Political Research 44: 147-174. Bogaards, Matthijs (1998): The favourable factors for consociational democracy: A review. In: European Journal of Political Research 33/1: 475-496. Bogaards, Matthijs (2000): The Uneasy Relationship between Empirical and Normative Types in Consociational Theory. In: Journal of Theoretical Politics 12/4: 395-423. Bogaards, Matthijs (2004): Counting parties and identifying dominant party systems in Africa. In: European Journal of Political Research 43: 173-197. Budge, Ian; Newton, Kenneth; McKinley, R.D.; al, et (1997): The Politics of the New Europe. Atlantic to Urals. London, New York: Longman. Dalton, Russell J. (2002): The Decline of Party Identification. In: Dalton, Russell J.; Wattenberg, Martin P. (Hrsg.): Parties without Partisans. Political Change in Advanced Industrial Democracies. Oxford: Oxford University Press: 19-36. Dalton, Russel J.; Flanagan, Scott C.; Beck, Paul Allen (Hrsg.) (1985): Electoral Change in Advanced Industrial Societies: Realignment or Dealignment. Princeton: Princeton University Press. Fearon, James D. (2003): Ethnic and Cultural Diversity by Country. In: Journal of economic Growth 8: 195-222. Fortin, Jessica (2004): Patterns of Democracy? Counterevidence from Sixteen Post- Communist Countries. APSA Chicago, Chicago. Franklin, Mark N.; Mackie, Thomas T.; Valen, Henry (1992): Electoral Change. Responses to evolving social and attitudinal structures in western countries. Cambridge: Cambridge University Press. Ganghof, Steffen (2004): Institutionelle Typen parlamentarischer Demokratie und Muster der Parteieninteraktion. Köln/Lüneburg: Paper presented at the DVPW Arbeitskreis Interkulturelle Demokratie/Ad Hoc Gruppe Empirische Methoden der Politikwissenschaft, Oktober 2004. Kitschelt, Herbert (2000): Linkages Between Citizens and Politicians in Democratic Polities. In: Comparative Political Studies 33/6-7: 845-879. Kitschelt, Herbert; Mansfeldova, Zdenka; Markowski, Radoslaw; Toka, Gabor (1999): Post- Communist Party Systems. Competition, Representation, and Inter-Party Cooperation. Cambridge: Cambridge University Press. Lehmbruch, Gerhard (1967): Proporzdemokratie. Politisches System und politische Kultur in der Schweiz und Österreich. Tübingen: Mohr (Siebeck). 22

Lehmbruch, Gerhard (1974): A Non-Competitive Pattern of Conflict Management in Liberal Democracies: the Case of Switzerland, Austria and Lebanon. In: McRae, Kenneth D. (Hrsg.): Consociational Democracy: Political Accommodation in Segmented Societies. Toronto: McClelland and Stewart: 90-97. Lehmbruch, Gerhard (1993): Consociational Democracy and Corporatism in Switzerland. In: Publius: The Journal of Federalism 23/2: 43-60. Lehmbruch, Gerhard (1996): Die korporative Verhandlungsdemokratie in Westmitteleuropa. In: Armingeon, Klaus; Sciarini, Pascal (Hrsg.): Deutschland, Österreich und die Schweiz im Vergleich (Sonderheft der Revue Suisse de Science Politique). Zürich: Seismo: 19-41. Lijphart, Arend (1968): The Politics of Accommodation. Pluralism and Democracy in the Netherlands. Berkeley/Los Angeles: University of California Press. Lijphart, Arend (1984): Democracies. Patterns of Majoritarian and Consensus Government in Twenty-One Countries. New Haven/London: Yale University Press. Lijphart, Arend (1999): Patterns of Democracy: Government Form and Performance in Thirty-Six Countries. New Haven: Yale University Press. Lipset, Seymour Martin; Rokkan, Stein (1967): Cleavage Structures, Party Systems, and Voter Alignments: An Introduction. In: Lipset, Seymour Martin; Rokkan, Stein (Hrsg.): Party Systems and Voter Alignments: Cross National Perspectives. New York: The Free Press: 1-64. Mair, Peter (1998): Party System Change. Approaches and Interpretations. Oxford: Oxford University Press. Marks, Gary; Hooghe, Liesbet; Nelson, Moira; Edwards, Erica (2004): Mapping Support for European Integration in a United Europe: All in the Family? Chapel Hill: unpublished paper. Olson, David M. (1998): Party Formation and Party System Consolidation in the New Democracies of Central Europe. In: Political Studies XLVI: 432-464. Rae, Douglas (1968): A note on the fractionalization of some European party systems. In: Comparative Political Studies 1: 413-418. Roberts, Andrew (2004): What Type of Democracy is Emerging in Eastern Europe? Chicago: APSA Chicago. 23