ECPR Workshop about European Aspects of Post-Communist Party Development (Mannheim, March 26-31, 1999)

Transcription

1 ECPR Workshop about European Aspects of Post-Communist Party Development (Mannheim, March 26-31, 1999) The Soviet Legacy in Voting Behaviour?: Support for Left Parties in Russia, Ukraine, Bulgaria, Slovakia and the Czech Republic (Version 1.2) Sarah Oates, William L. Miller and Åse Grødeland Address for correspondence: Sarah Oates Lecturer, Politics Department Adam Smith Building University of Glasgow Glasgow G12 8RT United Kingdom Phone: (0141) ; Fax: (0141) While Western analysts have touted the arrival of pro-market parties and Keynesian economics in the former Soviet bloc, parties of the Left have continued to attract votes in elections throughout the former Soviet Union and Eastern Europe. Although support for communists and socialists has varied a great deal, voters elected parliaments in which roughly one-quarter of the deputies were communist in Russia in 1995 and Ukraine in In turn, Bulgarian voters have shown strong support for the communist successor party, the Bulgarian Socialist Party, and the Czech Republic has a small, but well-supported communist party. It is important to consider why voters have continued to support communist parties long after they have rejected the Communist system as well as the differences between communist and socialist party supporters. This paper will use election results and public opinion surveys to define and discuss Left electoral support in Russia, Ukraine, Bulgaria, Slovakia and the Czech Republic. This paper will first examine the ability of Left voters to identify and support various parties, as well as their commitment to voting itself. The paper also will search for cross-national trends among both communist and socialist constituencies in these five countries to suggest reasons for the enduring presence of the Left parties. What emerges from this research is that Left voters, in general, are better able to identify parties to support and are more committed to voting itself. In addition, communist voters in Ukraine and Russia share similar socio-economic traits. These constituency characteristics correspond to a degree to those of socialist voters in Bulgaria, the Czech Republic and Slovakia, countries in which communist parties currently are less successful. Results on attitudes among Left voters, aside from a strong dislike for a market economy, are more mixed. Still, it would appear that Left voters in these five post-communist countries share important socio-economic characteristics, some attitudes and a heightened ability to translate their ideological preferences into actual votes. This could be quite useful in understanding the persistence of a Left vote throughout the post-communist sphere. This paper draws on some classic and some newer literature. In particular, it draws on the conclusions on Lipset and Rokkan about the development of parties along cleavage lines in society (1967). Their theories suggest that parties traditionally enjoy support from various groups in society based on cleavages formed during upheavals what Lipset and Rokkan term revolutions in the history of each 1

2 country. Not only is this type of study important for the foundations of party strategies, as party leaders try to play to their strengths among the electorate to gain office, but they are critical for plotting the general political course of a country as well. For example, if an ageing population supports a certain party, opponents may have to wait for a population shift to the younger cohort or find some way to appeal to that older constituency to win votes. If economic problems reduce the real income of a large segment of voters, parties that cater to the lower end of the economic spectrum may find themselves with a greater base of support. Much of the debate about the former communist countries has focused not on support for particular party families, be they communist, socialist, nationalist, etc., but on support for the concept of democratisation and Westernisation. Indirectly, this can elucidate the support for certain parties, but it is not a direct link and much like examining the support for lower taxes to suggest voting patterns for the Republicans in the United States. If one switches the focus to a study of voting behaviour and party organisation, the information generated can be much more precise and useful. In this manner, this paper borrows from U.S. literature on voting behaviour, particularly on party affiliation and identification on the part of individuals. Scholars of U.S. voting behaviour have found that U.S. voters have fairly fixed party identification (Campbell et al. 1960, Keith et al. 1992) and that voters in the U.S. tend to use this internal political compass, generally fixed since childhood, to make decisions at each election. Other work, especially Nie et al. (1979) suggest that the strength of party identification is eroding over time and that American voters are more sensitive to issues, candidates, the media and other short-term variables during elections. Those who study the often puzzling world of post-soviet politics will be relieved to know that there are fierce debates within the field of U.S. voting behaviour over the validity and reliability of the U.S. elections studies upon which much of research is based (in particular see Keith et al. 1992). The study of post-communist voting behaviour is much more exciting and I would venture to say more difficult than the exploration of the relatively static ties between voters and party in established democracies such as the United States and Great Britain. While the same concepts certainly could apply, they are thrown into the much more extreme conditions in fragile democracies, a sort of Big Bang of political effects. As Peter Mair suggests (1997), we should see post-communist Europe as not a terminus but rather a departure point (p. 180). Thus, if we look to the roots of cleavage formation in Lipset and Rokkan or the use the ideas of the formation of party identification from Campbell et al., Nie et al. and others. we have a useful lens through which to view voter identification and party formation in new democracies. Meanwhile, a rapid sequence of elections, pressure to integrate with a world capitalist economy and quick formation of a new state and society force change at breakneck pace. As such, voter preferences and allegiances may be formed within years rather over the slower pace of decades explained in other societies. As a result, these formative years can become immensely important, solidifying cleavages, consolidating party gains and fixing party identification within the minds of voters. Evidence in this paper suggests that more so than other party groupings communists have been successful in consolidating their support within a certain group in society. In addition, it appears that socialists appear to enjoy much of the same advantages. While many scholars have looked at the fluidity and somewhat chaotic nature of voting behaviour and party organisation in post-communist countries, it is equally important to consider where cleavages appear to have frozen and party affiliation has been locked into certain individuals. 2

3 Richard Rose and William Mishler (1998) have examined conceptions of party identification though cross-national surveys in former communist countries. Survey data from 1995 in Hungary, Poland, Romania and Slovenia found that 77 percent of the respondents had negative identification with particular parties while only 30 percent had positive identification. This study legitimised the conception of party identification in post-soviet states, although the authors posited a greater movement toward scepticism than positive party identification. Rose (1996) examined the opinions of ex-communist Party members on a range of issues and suggested that post-soviet communists must adjust to the times: they must produce platforms and promises that are in demand in the competitive marketplace dominated by voters. To survive in the new political system, ex-communist politicians must do more than convince voters that they will defend social welfare programmes; they must also make clear that they have abandoned authoritarianism and all it conveys, such as censorship, armed border guards and the secret police (p. 24). While this is true, communist parties also enjoy association with what many citizens would perceive as positive aspects of the former Soviet rule, such as political and economic stability, guaranteed minimum standards of living, free health care and global influence. Communist Parties in Post-Communist Times Russia has seen the most steady growth of support for communists since the collapse of the Soviet Union, both through two parliamentary elections (1993 and 1995) as well in the 1996 presidential elections (see Chart 1: Left Support and Successes in Five Post-Communist Countries). The communist party also has enjoyed significant popularity in Ukraine in post-soviet elections. The number of seats the Communist Party of the Ukraine holds in the Ukrainian parliament has increased from 91 out of 450 from the 1994 elections to 122 in the 1998 elections, rising from 20 percent to 27 percent. 1 In Bulgaria, the former communists were able to transform themselves into a successful socialist electoral movement (the Bulgarian Socialist Party) even before the collapse of the Soviet Union; in the 1990 parliamentary elections the Bulgarian Socialist Party, won a majority of seats in the National Assembly. They lost their majority, however, in new elections in The Bulgarian Socialist Party won an even greater number of seats 125 out of 240 in parliament in the 1994 elections and has hung on to win 22 percent of the vote in the April 1997 parliamentary elections. There are several communist movements, most notably the Bulgarian Communist Party. 2 In the 1996 parliamentary elections in the Czech Republic, left-wing parties gained ground, with the Social Democrats increasing their support four-fold and the Communist also gaining seats (see Chart 1). In Slovakia, the Party of the Democratic Left (formerly the Communist Party of Slovakia Party of the Democratic Left) ran second in the 1992 general election, but lost ground in the general election of fall After falling to 13 seats in the 150-member National Council of the Slovak Republic in 1994, the Party of the Democratic Left won 23 seats and about 15 percent of the vote in the September 1998 parliamentary elections. 3 There also is a Communist Party of Slovakia, but it received just 2.7 percent of the vote in (Chart 1 about here.) 1 The 1994 figures are from The Statesman s Yearbook, (135 th Edition), Barry Turner (ed.), London: Macmillan, p The 1998 figures are from Sarah Birch and Andrew Wilson, The Ukrainian Parliamentary Elections of 1998, Electoral Studies (forthcoming). 2 Political Handbook of the World 1997, Arthur S. Banks, Alan J. Day and Thomas C. Muller (eds.), Binghamton, N.Y.: CSA Publications, pp RFE/RL Newsline, 28 September

4 These election outcomes, as well as results from public opinion surveys show consistent support the Left (communists and socialists) ranging from a low of 16 percent in Bulgaria up to more than 25 percent in the Czech Republic in these five countries. This suggests that the Left is well-entrenched in these five countries. While the success of communist parties or socialists might once have been dismissed as the result of naïve voting or confusion over the benefits of the market system, these election and public opinion results show that the Left has a well-defined, permanent base of support in post-communist Europe. The task of this paper is to examine the nature of that support, using public opinion surveys. Is the electoral behaviour of Left supporters different, i.e. do they vote more often or appear to be more loyal to their ideology than other voters? Are Left supporters essentially confined to a narrow band of the socio-economic scale? Many analysts have noted a connection between older age and poverty with support for the Left. Are there critical issues such as advocating market reform or interest in authoritarian leaders -- on which Left supporters stand out from their fellow citizens? Are there important differences between communist and socialist voters on any or all of these points? Research Design Most of the data used for this paper comes from a set of surveys conducted in the Czech Republic, Slovakia, Bulgaria and Ukraine. In each country, a representative sample of the adult population was surveyed during the winter of , with sample sizes ranging from 1,003 in the Czech Republic to 1,519 in Bulgaria. 4 The Russian data are from a survey carried out by the All-Russian Centre for the Study of Public Opinion in Moscow on a nation-wide sample of 1,568 respondents from December 20 to December (carried out immediately after the December 1995 parliamentary elections). The question set is different for the Russian survey. For example, while the four-country questionnaire included specific questions on party family preference, the Russian survey merely asks people to name the party for which they voted. However, with some limitations, it is possible to make comparisons between the Russians and the citizens from the other countries using the two databases. Ideological Preferences in the Five Counties When viewed together, the four countries excluding Russia don t find parties of the Left the most popular option (Chart 2: Support for Party Types Across Four Post-Communist Countries). In fact, the 4,778 respondents from Ukraine, Bulgaria, Slovakia and the Czech Republic chose market economy parties most frequently, with 21.8 percent selecting this orientation. Socialist was the second most popular choice, picked by 13.3 percent of the respondents, followed by Communists at 7 percent, nationalists at 5.4 percent, Greens at 4.5 percent and minority nationalist at 2.7 percent. 5 A large proportion of the respondents didn t express a preference for any party type 19 percent of the respondents claimed they supported another party type from those on the list or none at all, 7.6 percent claimed it depends and 18.6 percent responded that they didn t know. This means that almost half of the 4 The research was funded by the British ODA/DFID (formerly Overseas Development Administration, now Department for International Development) under grant R6445 to Prof. Miller and Olga Koshechkina and by the UK Economic and Social Research Council under grant R to Miller and Dr. Grødeland. Translation and fieldwork were carried out by OPW of Prague, MVK of Bratislava, the Centre for the Study of Democracy in Sofia and Ukrainian Surveys and Market Research of Kyiv. 5 Minority nationalist referred to parties that supported a subnational group, such as Hungarians in Bulgaria or Crimeans in Ukraine. A special selection for "Russian nationalists" was included in the Ukraine survey. The six responses for "Russian nationalist" from Ukrainian citizens was included in the "minority nationalist" category. 4

5 respondents couldn t make up their minds to say they supported any party family at all. In the Russian data on vote choice in the 1995 Duma elections, there is very strong support for communists at 20 percent and virtually no socialist vote, as there was no strong socialist option (see Chart 3: Support for Parties in Russia). (Charts 2 and 3 about here.) While the overall findings put the market economy parties first, crosstabulations from the four countries find significant differences among them (see Chart 2). The Czech Republic and Bulgaria both had strong support for market-economy parties, at 29.6 percent and 33.6 percent respectively. However, while socialists enjoyed relative popularity in both countries as well (17.1 percent in the Czech Republic and 14.3 percent in Bulgaria), there also was a relatively robust support (8.5 percent) for the communists in the Czech Republic. Communists, however, had almost no support in Bulgaria. In Slovakia, nationalist parties and the socialists had almost equal strength, with just 4.7 percent support for the communists. In Ukraine, the communists were the most popular party grouping at 14.3 percent. The findings are affected to a fairly significant degree by the varying degrees of commitment to any party type. For example, although the communists were the most popular party type of choice in the Ukraine, an unusually large number of respondents declined to pick any party type of all. In fact, 61.9 percent of the respondents in Ukraine were unable to pick a specific party type to support. In contrast, only 43.2 percent of the respondents were unable to select a party type in Bulgaria 6 and just over a third of the respondents failed to give a specific party grouping in the Czech Republic or Slovakia. The findings from the Russian poll were different, which is not strange in that the Russian survey was started within days after a national parliamentary election. In addition, the questions were different, albeit the concepts measured are comparable to a degree. As explained above, the Russian survey team did not ask about general party preference, but about specific vote choice. The communists were the most popular choice in Russia among those who voted (only those who claimed to have voted were asked a party choice question). The mainstream Communist Party of the Russian Federation and the more hard-line Working Russia for a Soviet Union together garnered 20.1 percent of the support from the 1,161 respondents who claimed to have voted. Nationalist parties (Derzhava, Congress of Russian Communities and the Liberal-Democratic Party of Russia) won a total of 13.2 percent of the respondents support, while pro-government Our Home is Russia received 8.4 percent and promarket Yabloko 8.3 percent. 7 As Our Home is Russia explicitly supported the continuing course of a market-based economy, it is considered pro-market as well as pro-government. Perhaps most significantly, Left voters claim to translate their ideological preferences into voting. First, those who identify as either socialist or communist voters were far more likely to vote in general than either those with other party affiliation or the population over all (See Chart 4: Party Identification and Vote Intention). Among the communist identifiers in the four countries (333 out of 4,778 or seven percent) only 2.1 percent planned not to vote and a mere 1.8 percent couldn t pick a specific party to support, compared with 17.5 percent of the respondents overall who planned not to vote and 18.5 percent who couldn t pick a specific party. 6 The survey was held soon after elections in Bulgaria, which would have made people more aware of party choices. 7 The definitions of party families comes from my own research on the 1995 Russian Duma elections, which including collecting and coding party platforms on significant issue areas. 5

6 Socialist supporters also were making relatively concrete plans to pick a party and vote only 2.5 percent said they didn t plan to vote and just 3.1 percent claimed they couldn t identify a party of choice. One might argue that any party identification, ranging from pro-market to communist, would make a respondent more likely to pick a specific party or vote. However, 12.9 percent of pro-market identifiers across the four countries said they either wouldn t vote or couldn t pick a specific party. As is shown in the section below on the results from individual countries, pro-market and nationalist party identifiers seem to have a harder time making up their minds not only on one specific party, but on deciding whether to vote at all. Party identification of any type does have an impact on vote intention, for it is those that can t pick a party type that are least likely to vote 53.6 percent compared with 17.5 percent overall. (Chart 4 about here.) In Ukraine, 94.2 percent of those identifying themselves as communist supporters listed the Communist Party of Ukraine as their favourite party and 96.7 percent of those calling themselves socialists picked the Socialist Party. Interestingly, there was no overlap in Ukraine, i.e. no communist identifiers picked the Socialist Party and no socialist identifiers picked the Communist Party. In contrast, other party identifiers were far more lax in picking out a single party to reflect their ideological views. For example, those identifying themselves as most closely aligned with market economy parties chose 13 different parties for which to vote. Perhaps most ominously for the pro-market cause, 26 percent of the pro-market voters said they didn t plan to vote and an additional 18.2 percent said they either didn t have specific party preference or planned to vote for an independent candidate. 8 In addition, nationalist supporters were spread across eight parties, with 6.4 percent claiming they couldn t pick a specific party or didn t plan to vote. Meanwhile, the well-disciplined communist and socialist party identifiers in Ukraine show little plans of ignoring their social responsibility to vote only 4.6 percent of the communist identifiers and a tiny 1.7 percent of the socialist identifiers either wouldn t pick a favourite party or said they wouldn t vote in Ukraine. Does this pattern hold in other countries, particularly as one moves from the former Soviet heartland of Ukraine into Eastern Europe? Bulgaria had the smallest amount of communist identifiers (only 26 out of the 1,519 respondents), but 14.3 percent of the respondents (217 respondents) were socialist supporters. Of the handful of communist identifiers, 57.2 percent did pick the Bulgarian Communist Party. The pattern is far stronger, however, among the socialist identifiers. Among this group, 91.2 percent picked the Bulgarian Socialist Party. This is interesting on two counts. Not only does it show that the socialist identifiers were picking a single party, but it also suggests the Bulgarian Socialist Party has made a successful transition from its communist past into being accepted as a real socialist alternative. 9 Pro-market supporters in Bulgaria, however, were more consolidated on party choice than promarket supporters in Ukraine: 78.1 percent of them chose the UDF and only 11.5 percent claimed that they either would prefer an independent candidate, didn t plan to 8 It should be noted that Ukraine changed its electoral law between the 1994 and 1998 elections so that half of the parliamentary candidates are elected through parliamentary lists and half are elected through single-member districts. So technically voters either chose a party or don t vote on the party-list half of the ballot, so this survey question is a bit odd. 9 Also interesting is the idea that if the Bulgarian Socialist Party still had "communist" in its name, would these voters consider themselves communist rather than socialist? This is were the nature of party competition and the choices offered to voters becomes important in discussing ideological meaning and cleavages. 6

7 vote or couldn t pick a specific party. This suggests that a clear party alternative can make a great difference in consolidating pro-market supporters. Of course, given that the survey took place soon after parliamentary elections in Bulgaria, the relatively consolidated support of the pro-market supporters could be a reflection of greater party awareness due to the recent campaign and vote. Nationalist supporters were spread over five parties, with 23.8 percent of them unable to make a specific choice. In the Czech Republic, there were more socialist identifiers (17.1 percent) than communist identifiers (8.5 percent), but the latter group were more consolidated in their support for a single party. According to the survey, 96.5 percent of the communist identifiers picked the Communist Party to support. In contrast, socialist identifiers were split among 11 parties, although 68.6 percent of them picked the SDP. While all of the communist identifiers were able to pick a party to support one picked the SDP the socialist identifiers were less sure-footed in their party choice. Of the socialist identifiers, 8.1 percent either couldn t pick a party, would choose an independent candidate or didn't plan to vote. As in Ukraine, pro-market supporters were distributed over a relatively wide number of parties (10), with about a third of them picking ODS. As in other countries, though, the pro-market supporters were less inclined to either pick a specific party or vote 18.6 percent responded that they either couldn t pick a specific party or would choose an independent candidate and 4.7 percent didn t plan to vote at all. There were so few nationalist party identifiers in the sample (14 or 1.4 percent) that it is impossible to draw conclusions from their party preferences. Left respondents in Slovakia had the weakest relationship between party identification and a choice of a specific party to reflect their ideology. The 50 communist identifiers (4.7 percent of the sample of 1,056 people) mostly picked the Communist Party (58 percent), but 26 percent chose the Lavice Party, suggesting that there is not a clear choice for communist identifiers in Slovakia as there is in Russia, Ukraine, Bulgaria and the Czech Republic. The socialist identifiers were even less consolidated in their party choice: They were split among 15 parties, including 26.6 percent for Lavice, 13.8 percent for the SDP and 12.2 percent for MDS. Other party identifiers in Slovakia showed the same type of schizophrenia, including a spread about 14 parties for the pro-market identifiers and 12 parties for nationalist identifiers. The only other pattern of party consolidation of ideological interests aside from the communists can be found in the minority nationalist identifiers with 50.7 percent choosing the Hungarian CD and among the Greens, with 44.4 percent picking Pod, Ziv a Rol (the Green party). Once again, there was a pattern of a greater ability on the part of communist identifiers in picking a party and planning to vote: Only 4 percent of the communist identifiers in Slovakia couldn t pick a party and only an additional 4 percent didn t plan to vote. In contrast, 8.5 percent of the socialist identifiers couldn t pick a party and 2.1 percent didn t plan to vote. Even before considering socio-economic and attitude data, it is clear from the questions about party identification and vote choice that communist and socialist voters are different political animals. Most critically, they seem better able to translate their party identification into vote choice. The reasons for this remain slightly unclear and need to be examined at different levels. First, is there an individual-level attachment to parties of the Left, particularly the communist party with its tangible presence for decades in each country, that makes voters more loyal? In other words, is there a type of Left party identification in these five post-communist countries akin to that of the union member for the Labour Party in Britain or of the feminist activist to the Democrats in the United States? If so, is this individual affiliation also linked to 7

8 cleavages among the population, i.e. do older people feel more attachment to the Left, be it the communists or the socialists? The individual-level attachment, however, can only be part of the picture. It is clear from the investigation of the data in the five countries that both structures of the party system and proximity to elections seem to make a difference in both party identification and the ability to translate that affiliation into a vote for a specific party. The data from these five countries suggests that communist identifiers are particularly good at identifying specific parties to support. But in places in which there aren t viable communist alternatives, such as in Bulgaria, a similar pattern emerges with the support of socialist identifiers for the dominant Bulgarian Socialist Party (the successor to the Soviet-era communists). Although it is beyond the scope of this paper, it should be noted that there must be a link between clear choices in the electoral arena and the support from ideologues. For example, if there are several parties claiming to represent the socialist cause, voting may be scattered. It is up to the parties to create and transmit political platforms that make clear their underlying ideology in the confusing and transitional society in post-communist Europe. The data suggest that the communist and, to a slightly lesser degree in most countries, the socialists are doing a better job of providing ideological alternatives than the nationalist and pro-market parties. As a result, the Left vote appears to be more consolidated, while the nationalist and pro-market vote is split. This means that Left support is amplified at elections because of greater party consolidation, party identification and voting patterns. As a result, the Left enjoys a greater degree of electoral success than its underlying support base in the population. What about the lingering question of possible transnational characteristics of the Left electorate? In cross-tabulations measuring age, income, gender, urbanicity and education, the Left voters were most distinct as being older (See Chart 5: Socio- Economic Characteristics of Left Voters in 4 Countries). While roughly 25 percent of the respondents overall were under the age of 32, only 13.1 percent of the Left supporters were that young. Conversely, 36.6 percent of the Left supporters were 58 or older, compared with about a quarter of the respondents overall. In other categories, the differences were not so marked: Left supporters were slightly more likely to say they have not enough or just enough income and they were less likely to be from the capital city in the four countries (5.7 percent of the Left supporters compared with 9 percent of the respondents overall.) The differences in levels of education did not appear to be significant in these cross-tabulations. (Chart 5 about here.) But when the group of Left supporters were split into their communist and socialist components other differences emerged (See Chart 6: Socio-Economic Characteristics of Communist and Socialist Voters in 4 Post-Communist Countries). Communist supporters were even more likely to be older; with only 10.2 percent of them in the age range of to 32 and 40.2 percent aged 58 or over. In addition, they felt the pinch of poverty more keenly than the socialist supporters overall: 84.3 percent of them reported that their income is not enough or just enough compared with 76 percent of the respondents overall. In addition, very few were from the capital cities a mere 1.8 percent and they were markedly less likely to have completed higher education (2.4 percent of the communist identifiers compared with 5.7 percent overall among the respondents). On the other hand, the socialist identifiers more closely mirrored the respondents in general, albeit still being older in general, with one interesting exception socialist identifiers were very slightly more likely to have a university degree (6.8 percent of socialist identifiers compared with 5.7 percent of the 8

9 respondents overall). Could this suggest that socialism is the intellectual choice for Leftists in post-communist societies? (Chart 6 about here.) Regression analysis can be used to tease out the relationship among Left support and age, income, gender, urbanicity and education (See Chart 7: Regression Analysis for Left Support in Four Post-Communist Countries). In the regression model, the only significant independent variable in predicting Left support was age: Older age was highly correlated with support for the Left in the Ukraine, Bulgaria, Slovakia and the Czech Republic in these surveys. When either support for communist parties or support for socialist parties was used as the dependent variable (see Charts 8 and 9), there was the same result (although age was not quite as strong a predictor in the statistical model). It should be noted, though, that the r-square test for significance was quite weak in these regression models, suggesting that the model explained relatively little of the variance in Left support. However, the evidence from the cross-tabulations described above is convincing enough to suggest that the relationship between certain socio-economic characteristics and Left support is valid. This reinforces the idea that the theories of Lipset and Rokkan can be imported into post-communist systems: Voting groups do tend to align along certain socioeconomic divisions. (Charts 7, 8 and 9 about here.) If age was the only reliable predictor of Left voting across the four countries, were there distinct differences in communist and socialist supporters among the countries? For Ukraine, the patterns of Left supporters were similar to the fourcountry pattern, although Left voters were less likely to be from rural areas (see Chart 10: Socio-Economic Attitudes of Left Voters by Four Post-Communist Countries). In addition, socialists were slightly more likely to have higher education than the average respondent. Also, socialist voters in Ukraine were not likely to be older, although it is difficult to generalise because there were only 60 socialist identifiers (or 5 percent) among the 1,200 respondents. In a regression analysis using Left supporter as the dependent variable and age, income, gender, urbanicity and education as independent variables, both age and surprisingly urbanicity emerged as positively correlated with Left voting (see Chart 11: Significant Socio-Economic Attributes in Predicting Communist and Socialist Support in Five Post-Communist Countries). Thus, Left supporter would be considered a more urban phenomenon in Ukraine, but not in the four-country sample overall. The same result was obtained when communist support was used as the dependent variable. (Charts 10 and 11 about here.) On the other hand, the Left was quite old in Bulgaria 58 percent of the Left supporters are 58 or older, compared with 32.1 percent of the respondents overall in the Bulgarian sample (see Chart 10). Only 16.5 percent of the Left supporters were under the age of 45, compared with 45.1 percent of the respondents in general. They also were more bitter about their economic circumstances: 55.6 percent claimed that they had inadequate incomes, compared with 44.1 percent in the sample. Their supporters were slightly more likely to be female and more likely to live in villages (45.3 percent of the Left supporters as opposed to 32.9 percent of the Bulgarian sample). They also were more likely to have a minimum amount of education. There were so few communist voters in the sample (26 out of 1,519 or 1.7 percent) that it is not useful to look at their distribution in cross-tabulations and there are far too few for regression analysis. However, in regression analysis for the Left voters in Bulgaria (using the same model as above), it appears that only age was positively correlated 9

10 and urbanicity was negatively correlated with Left support (See Chart 11). 10 Thus, communism might be more urban phenomenon in Ukraine, but the Left in the form of socialism appears to be a rural phenomenon in Bulgaria. In the Czech Republic, Left voters were once again older and slightly more likely to be male (see Chart 10). They were far less likely to live in Prague (2.7 percent of Left supporters compared with 7.3 percent of the sample overall). However, Czech communist supporters were far more distinctive in socio-economic terms, albeit there were only 85 of them in the sample of 1,003 respondents. The communist supporters were generally much older than the rest of the respondents 45.9 percent were 58 or older while only 19.7 percent of the sample fell into this age category. In addition, they felt themselves to be poorer, as 74.1 percent of them felt that had either not enough or just enough income to survive compared with 53.9 percent of the Czech Republic respondents. They were slightly more likely to be male and far less likely to come from Prague (only one communist supporter in the survey came from the Czech capital). In addition, they were far less likely to have a university or technical higher education. In contrast, socialist voters closely paralleled the general attributes of the population, although they also were less likely to come from the capital (3.5 percent compared with 7.3 percent overall) and slightly more likely to be male. It is striking that socialist supporters had very similar attributes in terms of age, income and education as the general pool of respondents. This suggests that the communist and socialist voters in the Czech Republic were quite distinct groups, which can be traced through differences in age, income and education between the two groups within the Left camp. However, a regression analysis using the model outlined did show that older age and being male were significant in predicting socialist support in the Czech Republic (see Chart 11). There were too few communist supporters to run a regression analysis. In Slovakia, communist supporters were only slightly less likely to be older, but more likely to consider their incomes either not enough or just enough. Surprisingly, there appeared to be a few wealthy communists in the mix, as 10 percent felt their income was good, almost twice as many as the average in the sample. The communist supporters were more likely to be male than the population. However, this was perhaps an artefact of the fact that woman were less likely to have picked a party affiliation. 11 No communist supporters came from Bratislava, although 8 percent of the respondents overall lived in the Slovak capital. In addition, they were far less likely to have a high level of education. There were relatively few communist supporters in Slovakia, only 50 out of 1,056 respondents (4.7 percent). There was little difference in the ages of the socialist supporters compared with the general pool of respondents in Slovakia, although they did rate themselves as slightly poorer than average (70.3 percent said they had not enough or just enough income compared with 57.4 percent of the respondents overall). The socialists, however, were not confined to the small villages in fact, they were more likely to live in Bratislava than the average respondent. In addition, their educational attainments were quite close to those of the typical respondent. In a regression analysis using the model above, support for socialists was linked with low income and living in a large city. Interestingly, urbanicity was positively correlated with support 10 The result was the same if the 26 communist identifiers are taken out of the model and the regression is run using socialist voters as the dependent variable. 11 In the overall four-country survey, 39.3 percent of the women failed to pick a party type or didn t plan to vote compared with 34.1 percent of the men. 10

11 for the Left or specifically the socialists, while the opposite was true in the regression model for the Left in Ukraine and Bulgaria (see Chart 11). There is no question in the Russian data set that asks respondents to identify themselves as a supporter of a particular party group; there is merely the question of vote choice in the recent parliamentary election. In the survey, 314 of the 1,568 respondents (20 percent) claimed to have voted for a communist party (the Communist Party of the Russian Federation or the more hard-line communist Working Russia for a Soviet Union). Cross-tabulations of the data show that the communist supporters tended to be older, more likely to live in rural areas, to fall into lower income categories and have a lower level of education than the average respondent (See Chart 12: Socio-Economic Characteristics of 1995 Communist Voters in Russia). A regression analysis shows, however, that only older age was significantly correlated with communist support when all other factors were held constant (see Chart 11). There were no truly socialist parties in the 1995 Russian Duma elections. 12 (Chart 12 about here.) There are other interesting points to note in the Russian data for communist supporters. Like their communist and socialist counterparts in Ukraine, Bulgaria, the Czech Republic and Slovakia, they appeared to be committed party supporters. When asked why they had voted for a particular party, 43.1 percent of the communist supporters gave the reason as this party expresses the interests of people such as myself, compared with 27.6 percent of the 1,115 respondents who claimed to have voted overall (see Chart 13: Reasons for Vote Choice in Russia). In addition, 8.9 percent of the communist supporters claimed that they were accustomed to voting for this party as opposed to 3.2 percent of the respondents overall. In this sense, the Communist Party of the Russian Federation benefited from being one of the few parties that survived from 1991 onwards. When asked to recall their votes in the 1993 parliamentary elections, 36.2 percent of the 1995 communist supporters could recall voting for the CPRF in This was a distinct advantage for the communists, as most of the parties in the 1995 elections had not existed at the time of the 1993 Duma elections. 13 (Chart 13 about here.) If one can identify communist and socialist supporters through socioeconomic groupings, how defined were they through their opinions on important issues in post-communist society? This paper looks at attitudes on five issues: whether to join NATO, whether to join the European Union, whether the respondent favours order over freedom, whether the respondent supports a market economy and whether the respondent believes a strong leader with a free hand would solve problems (see Chart 14: Regression Analysis for Communist Support/Attitudes in Four Post- Communist Countries). Looking at the data for Ukraine, Bulgaria, the Czech Republic and Slovakia, 14 we see that communist voters were distinct in that they didn t want to join NATO, didn t want to join the EU and -- understandably -- were against a market 12 There was a lone party, the Social Democrats, that incorporated the idea of socialism in its name, but this was basically a party formed around Russian politician Gavril Popov and received virtually no coverage in the media during the elections. White, Rose and McAllister identified several socialwelfare party programmes in their landmark study of Russian voting behaviour, but none had an articulated socialist programme (p. 208). 13 The Russian constitution stipulates that Duma elections are held every four years with the exception of the 1993 Duma, which was elected for just two years. 14 No opinion questions of this nature were asked on the Russia survey used in this paper, so the Russian voters are excluded from this part of the analysis. 11

12 economy. Their positions on strong leaders and order over freedom were not significant in the regression model. Socialist voters in the four countries, however, didn t share the same opinions (see Chart 15: Regression Analysis for Socialist Support/Attitudes in Four Post-Communist Countries). They also were against joining NATO and against a market economy, but they were not against joining the EU as well. While they may appear less isolationist, however, they did favour order over freedom: An attitude supporting order was linked with socialist support in the regression model. (Charts 14 and 15 about here.) It is possible to look at the critical attitudes in countries in which there were at least 100 socialist or communist supporters to make the regression analysis statistically valid. Thus, this paper can examine the impact of these four attitudes on communist support in Ukraine and their impact on socialist support in Bulgaria, the Czech Republic and Slovakia (see Chart 16: Significant Independent Attitude Variables in Predicting Communist and Socialist Support in Four Post-Communist Countries). The same attitudes -- anti-market, anti-eu and anti-nato were significant for communist supporters in Ukraine. The attitudes within countries varied so much for socialist supporters it calls into question whether the three groups could all be considered supporters of the same party type. In Bulgaria, a regression analysis showed that socialist supporters there shared the same three distinct attitudes as socialists in the overall sample -- anti-nato, anti-market and supportive of order over freedom. Socialists in the Czech Republic, however, were less isolationist and didn t object to joining the EU or even NATO, although they were anti-market. Also, Czech socialist supporters supported the idea of order over freedom. Slovakian socialist supporters also appeared to be indifferent about joining the EU or NATO, but their support was negatively correlated both with a market economy and a strong leader. Thus it would appear that, on a national level, the socialist supporters had quite different opinions and seem quite different from classic Socialism, particularly in supporting authoritarianism in the form of preferring order to freedom (Chart 16 about here.) Conclusions, Hypotheses and Further Questions This research suggests that Left voters are unusually important because they are relatively confident at choosing a party for which to vote and they exercise their right to vote more often than other groups in societies, even those that support other well-defined party types. As a result, their electoral preferences are amplified into disproportionate gains for the Left. Because of their superior ability to make and exercise vote choices, their wishes are better translated into seats in the election of Left parties and candidates. As a result, there is more representation for their socioeconomic group, typically the older segment of the population, in post-communist countries. In addition, the opinions of the Left, particularly their somewhat nationalistic and anti-market stance, has an unusually strong voice in elected office. There are, however, many questions raised by this research. Among them are the following queries: 1. Can the concepts of party cleavage and party identification be imported from the Western European and U.S. models? Fairly wide-ranging research in Russia (including Clem and Craumer 1995a, Clem and Craumer 1995b, Evans and Whitefield 1996, Ioffe 1995, Klyamkin 1995, Levada 1995, Lowenhardt 1998, Segal 1995, White, Rose and McAllister 1997, White, Wyman and Oates 1997, Wyman 1996, Wyman, White and Oates 1998) suggests that this is possible, but is it then fair to attempt it in other post-communist countries? Is Russia exceptional, 12

13 and therefore a poor model in studying party development other post-communist states? In particular, it should be considered whether the different experiences of Ukraine, Bulgaria, the Czech Republic and Slovakia in the 20 th century invalidate the importation of concepts that appear to be useful for explaining voter affiliation and behaviour in Russia. 2. Are the variations among countries in terms of the number and behaviour of Left voters more meaningful than any cross-national trends? For example, the percentage of communist supporters varied from about 20 percent down to 1.7 percent in the surveys. Perhaps the real story is the differences among countries rather than any attempt to discuss the Left in a cross-national context. 3. Following on to point No. 2, it is critical to consider the impact of party behaviour on Left support, hinted at in this paper but not supported by any empirical evidence. For example, are those voting for the Bulgarian Socialist Party really supporting socialism per se or rallying around a successor communist party? In order to answer this question, one needs far more information about the party and its marketing strategies. In addition, one needs to consider how the choices of voters are informed and/or constrained by their options. Can the Left thrive when there aren t viable alternatives? Why have pro-market and nationalist parties, in general, failed to capitalise on their support within the population in the same way? 4. Should one consider the presence of the Left in post-communist states more as an artefact of the former Soviet system or as a movement that is allied with the Socialist movement in Europe? Communist parties, particularly in Russia, are fond of using the positive elements of the Soviet past stable economy, more equitable distribution of wealth, military might as political marketing tools. Are those supporting the Left looking backward to the glorious Soviet-dominated past or forward to a European Socialist future? If there is a nostalgic element to the support for the Left, can that be considered a part of party identification forged under Soviet rule or domination? 13

14 14

15 Chart 2: Support for Party Types Across Four Post-Communist Countries (in percentage) Party type All 4 countries Ukraine Bulgaria Czech Republic Slovakia Communist Socialist Market Economy Nationalist Minority Nationalist* Green Other/None Depends Don t Know N 4,778 1,200 1,519 1,003 1,056 *Includes Russian minority party supporters in Ukraine. 15

16 Chart 3: Support for Parties in Russia (in percentage) Party type Parties used in definition in 1995 Duma Total elections Communist Communist Party of the Russian Federation, 20.0 Working Russia for a Soviet Union Socialist Social Democrats 0.2 Market Economy Our Home is Russia, Yabloko 16.6 Nationalist Liberal-Democratic Party of Russia, 13.2 Congress of Russian Communities, Derzhava Minority Nationalist Nur (Muslim party) 0.2 Green KEDR 1.0 Other 19.7 Hard to say 1.3 Against All 1.8 Didn t Vote 26.0 N 1,568 16