The Policy Studies Journal, Vol. 31, No. 4, 2003
Welfare and the Discriminating Public: Evaluating Entitlement Attitudes in Post-Communist Europe
Christine S. Lipsmeyer
Although post-communist Europe retains elements of its socialist past, public opinion shows discernment in its welfare preferences. This analysis of post-communist social welfare attitudes finds that post-communist societies are selective in their support for social policies. First, welfare preferences center on two underlying realms: government responsibility and government spending; and second, welfare opinions and beliefs are not uniform across several social policies. Although many of the conclusions highlight the selective nature of these preferences, the example of the support for unemployment benefits points to how the transition from communist-era welfare states to capitalist-led assistance continues to influence public opinion in these countries.
The dual transition altered not just the economies and politics of post-communist Europe but also the societies. With the advent of democratic governance, what these societies endure, believe, and desire can influence the governing and policy processes through representation and voting. Therefore, as these governments begin to change their cradle-to-grave welfare systems, public opinion may become a mitigating factor in the subsequent outcomes. But the underlying structure of post-communist public opinion remains underexplored. Although analyses of attitudes within individual countries offer insights into specific cases, we are left without a comparative view of mass welfare preferences during the transition.
The transitional changes to democracy and markets call into question both the rationale for welfare policies and the breadth of the support system in society. Under communism, there existed a wide range of programs that buffered citizens from economic currents; however, the dual transition introduces changes to the social welfare system more in line with the new marketization systems. Along with the creation of new social policies (i.e., unemployment benefits) and the tinkering with old ones (e.g., pensions, health care, and housing) comes the added burden of new democratic publics. Electoral forces emerge as pressures on
545
0162-895X © 2003 The Policy Studies Journal Published by Blackwell Publishing. Inc., 350 Main Street, Maiden, MA 02148, USA, and 9600 Garsington Road, Oxford, OX4 2DQ, UK.
546
Policy Studies Journal, 31:4
post-communist governments, creating situations in which public preferences and attitudes on public policy—more specifically, welfare policy—matter. In this contemporary post-communist situation, how the societies view the role of government in welfare policy and their opinions on spending levels are more than abstract questions, for the governments are now accountable to the democratic electorates. Therefore, which policies receive mass support and which ones relatively less may matter for various policy and governmental reasons.
The question remains: Did the vast communist welfare systems create parallel public opinion that embraces the notions of government social action? Or is public opinion receptive to movements away from cradle-to-grave social assistance? In this article, I undertake an analysis of welfare preferences in post-communist countries with an aim to investigating these questions. Beginning with the selectivity of responses across various social programs—health care, pensions, education, housing, and unemployment benefits—provides a way of evaluating welfare preferences that is inclusive of a range of welfare assistance. Then, I analyze the independence of welfare attitudes by separating preferences between government responsibility and government spending, finding through factor analysis that respondents distinguish between the government's role in specific social welfare programs and its budgetary authority for social policies. Because my findings show that respondents view unemployment benefits separately from other welfare policies, I more thoroughly investigate the lack of support for these benefits and locate an overall negative attitude toward unemployment assistance. This lack of backing for unemployment benefits may correspond with the newness of this assistance in these countries, as well as governments' attempts after the policy's initial creation to restrain unexpectedly high spending on these benefits.
Public Opinion Research and Welfare Attitudes in Post-Communist Societies
The dual transition in post-communist Europe has been contentious for the last decade, and public opinion polls remain an important means of measuring citizens' perceptions of political and economic change. Surveys have been widely used, for example, to examine political and ideological cleavages in these societies (Evans & Whitefield, 1993, 1996; Kitschelt, 1995; Kitschelt, Mansfeldova, Markowski, & Toka, 1999; Miller, Erb, Reisinger, & Hesli, 2000); to analyze political legitimacy and social justice (Mason & Kluegel, 2000); to explain voting and protest patterns (Bahry & Lipsmeyer, 2001; Colton, 1996; Whitefield & Evans, 1994; Wyman, White, Miller, & Hey wood, 1995); and to investigate the connection between economic and political change (Duch, 1995; Gibson, 1998; Miller, Hesli, & Reisinger, 1997; Rose & Haerpfer, 1994).1 As a whole, these works have centered on explaining the characteristics of voters, highlighting
Lipsmeyer: Welfare and the Discriminating Public
547
attitudes toward the transformation, or comparing social attitudes across post-communist countries.
Although scholars have started to use surveys to explore preferences for social welfare policies in this region, some of the research relies upon questions of redistribution and inequality (Evans, 1996, 1998) or government intervention more generally (AndreB & Heien, 2001) rather than using indicators on specific welfare items. Evans (1998), for example, locates a cluster of Central and Eastern Europe (CEE) countries in an interventionist sphere; however, his topic remains one of government intervention rather than a comparison of particular social programs. Although research that touches on distinctions in post-communist welfare beliefs suggests that preferences vary across countries and across general public policy areas within each country, Rose & Makkai (1995) combine government spending responses on education, health, and pensions into one measure rather than disaggregating and distinguishing between these programs. If citizens are in fact selective in their opinions concerning welfare policies, then a single measure of "welfare orientation" will not capture the distinctions between various kinds of programs. Nevertheless, the findings from these studies highlight an Eastern Europe with stronger ties to government intervention than in the West.
In studies that use strictly "welfare" attitude measures at the individual level, the findings are somewhat inconclusive for the post-communist region. Whereas some find a typical socialist response—an aversion to decreases in social spending—others find elements of liberal preferences to social benefits in different countries. On one hand, a comparison of East and West German attitudes notes that the low support in East Germany for social spending cutbacks blames socialization under communism, in addition to the economic hardship (Roller, 1999). In the Czech Republic, Rabusic & Sirovatka (1999) notice a shift towards a desire for more government provision from 1991 to 1998, which coincides with a majority (59%) of the Czech public in 1998 being dissatisfied with the welfare system (252). On the other hand, in a recent study, Stephen Whitefield (2002) evaluates perceptions of welfare in the Ukraine and finds modest support for the targeting of benefits. And Csontos, Kornai, & Gyorgy (1998) add to the puzzle of post-communist attitudes in their survey of Hungarians in 1996 concerning the link between taxes and welfare policies. They find that when individuals are presented with different institutional arrangements, the proportions of those who support pure "state" or laissez faire options are slight (9.8 and 0.9, respectively) (1998, p. 291). Nevertheless, the research up to this point highlights post-communist welfare attitudes using indicators of redistribution or analyzing preferences within one country, neither of which offers a comparative picture of how post-communist societies in general view welfare policy issues. There appears to be a potential contradiction with the old image of broad public support for cradle-to-grave welfare, which may indicate a consensus with West European studies that show a selectivity in preferences. For instance, more respondents support the govern-
548
Policy Studies Journal, 31:4
ment provision of health care and pensions than unemployment benefits (Haller, Hollinger, & Raubal, 1990; Shapiro & Young, 1989).
Research Design
The goal of this work is to explore the following inquiries in the post-communist context:
First, do public preferences on welfare vary across individual policy areas? Are they selective in their support for government responsibility and for spending on social policies?
Second, are there different patterns of public opinion on welfare across the countries with some adopting a less-government-centered approach, showing some countries to be closer to a "Western" model of welfare preferences than a "Communist" pattern?
Third, and finally, do respondents view government responsibility and government spending as independent realms of welfare policy?
In order to be systematic and comparative, I use the International Social Survey Program (ISSP): Role of Government III survey from 1996. This survey provides answers across a range of welfare questions in seven post-communist countries: Bulgaria, the Czech Republic, Hungary, Latvia, Poland, Russia, and Slovenia. These cases range from advanced reformers in the west to slower reformers in the east, but they all had undergone some welfare changes from their communist systems by the time of this survey in 1996. The ISSP data provide individual-level evidence of the publics' attitudes toward the role of government in providing social assistance, as well as more specific answers on individual policies—unemployment benefits, old-age pensions, health care, education, and housing.
Indicators and Statistical Methods
The indicators of welfare responsibility and spending are questions measuring attitudes on specific social policies from the ISSP survey. Because my main concern is social welfare preferences, I use responses to "welfare"-oriented queries: health care, pensions, unemployment, housing, and education. These particular areas include policies that correspond with particular types of welfare programs: insurance-based, in-kind, and market-oriented policies. Welfare analysts at times will distinguish among the social insurance sources of health care and pensions, the in-kind benefits of education and housing, and market-based unemployment assistance (Barr, 1994; Andrews & Ringold, 1999). They argue that not only the funding sources but also the type of benefit program may influence the public and partisan support for welfare assistance. While some benefits, such as pensions, are employment based, others, such as education, are universal. At the beginning of the transition, the differences between benefit types were minimal,
Lipsmeyer: Welfare and the Discriminating Public
549
but by 1996 most welfare systems had developed into complex programs incorporating an assortment of policy types.
To provide a more complete picture of government responsibility and budgetary affairs, I also include questions concerning provision of jobs; redistribution of income; industry growth; responsibility for prices and wages; spending on law enforcement and defense; and, finally, spending on environment and culture/arts programs. These responses to general public policy issues may provide a more discriminating picture of how preferences on welfare policies fit into the wider question of post-communist public policy. If welfare policies do not adhere to a single underlying factor in the case of government spending or responsibility, then the questions become: How do they group? Do respondents think of welfare policies as a cohesive group of programs or do they disaggregate them into separate policies that correspond with other economic or redistributive beliefs? Are preferences ordered by attitudes concerning individual policies or by concerns of government responsibility and spending?
The specific wording of the questions includes a typical five-point scale measuring a respondent's opinion from (1) "Agree strongly" to (5) "Disagree strongly," and a four-point scale, which asks for an opinion on the government's role in a specific area from (1) "Definitely should be" to (4) "Definitely should not be." The lower numbers indicate more government responsibility or spending in an area, whereas higher numbers denote less government intervention and funding (See Appendix for the specific survey questions).
Methodologically, this article relies upon simple crosstabs/figures and oneway ANOVAs to compare preference levels across both policy areas and countries. But when investigating the underlying beliefs of respondents, I employ factor analysis because as Kim & Mueller explain, "[it] is based on the fundamental assumption that some underlying factors ... are responsible for the covariation among the observed variables" (1978b, p. 12). Because I predict that government responsibility and spending originate from different underlying preferences, the method used is confirmatory factor analysis, or principal axis factoring (PAF) in SPSS,2 which allows me to test hypotheses for diverse beliefs that undercut welfare preferences. Theoretically, welfare preferences on responsibility and spending are not completely separate attitudes, causing the need for an oblimin rotation that allows factors to be correlated. In the presentation section, the corresponding factor loadings extend from +0.000 to +1.000 (ranging from no relation between factor and variable to an almost perfect relationship between the two). Two general rules of thumb with factor analysis are (1) that factor loadings below +0.400 are only weakly correlated with the underlying attitude and therefore are not used in the explanation, and (2) that the number of factors is determined by the Kaiser criterion where those components with eigenvalues of less than 1.0 are not used in the analysis. Based on the previous discussion, I hypothesize that the spending and responsibility preferences will load on separate factors with the more "economically" situated questions—wages and prices—loading on
550
Policy Studies Journal, 31:4
a third component. To test whether more factors would lead to more policy-oriented underlying beliefs, I show the results of PAF using a four-factor scheme.
Welfare Attitudes in Post-Communist Societies
Communist welfare systems provided universal coverage ranging from old-age pensions, health care, and education to childcare, housing, and family benefits. Under these regimes, states supplied their citizens with the necessities in return for furnishing low wages and securing acquiescence. For decades, people had turned to the state for assistance, not only during times of need but also on a daily basis for housing and childcare. The omnipresence of a state-financed, universal benefits system may have created a society that views all welfare programs as an entitlement and is loath to support changing the system to a more individualistic version.
As previous research shows, beginning in the late 1980s into the early 1990s parliaments began the process of altering welfare systems (Cook & Orenstein, 1999; Lipsmeyer, 2000, 2002). The welfare adjustments that correspond with the transformation to markets tie many of the benefits to employee or employer contributions, enforce shorter durations, or even impose means-testing requirements in order to receive the assistance (Social Security Administration [SSA], 1991, 1993, 1995, 1997). But these adjustments have not been equally distributed across all of the welfare policies. Only a few countries, for example, have implemented higher retirement rates, whereas most have cut education funding and/or devolved it to the regional level. Parliaments and governments clearly make distinctions between programs for the purpose of budgets and political support, and I find in this article that the publics do as well.
Table 1 shows the level of support for government responsibility across social programs, job responsibility, and income redistribution in the post-communist countries. Two policies have undeniable support (from 95 to 99%) across all countries—health care and pensions. Most other policies receive a clear majority, with the Czech Republic typically posting the lowest percentages. The assistance program with the lowest support is unemployment benefits; less than a majority (39.9%) in the Czech Republic agrees, and the highest percentage (86%) comes from Bulgaria. The percentage of respondents who said, "Definitely should" to unemployment, however, remains under a majority for all countries and is the only area to earn that distinction. Income redistribution still retains majority support in all countries, but there are differences in the levels of support, from 83% in Slovenia to 58% in the Czech Republic. Analyzing across the countries, I note the results of ANOVA tests within each policy, which test the statistical significance of the differences in the country means and locate variation in all categories. Therefore, these European countries differ on the role of government in supplying social services and do not cohere to a simple "post-communist" pattern of government intervention.
Table 1. Attitudes Toward the Role of Government in Specific Welfare Programs—Post-Communist Europe, 1996
% of respondents who answered "Definitely should" or "Probably should" when asked if the policy should be the government's responsibility
Bulgaria        Czech        Hungary        Latvia        Poland        Russia         Slovenia ANOVAs
a job for everyone
Definitely should	49.1	48.1	46.0	66.0	55.4	69.8	57.3	F = 48.3
Probably should	30.1	26.8	39.4	21.3	31.0	21.8	30.6	
Total	79.2	74.9	85.4	87.3	86.4	91.6	87.9	
health care for the sick								
Definitely should	68.4	71.5	69.4	82.3	67.5	80.8	79.2	F= 16.9
Probably should	27.4	24.2	28.3	15.0	27.7	16.5	16.8	
Total	95.8	95.7	97.7	97.3	95.2	97.3	96.0	
decent standard of living for the old								
Definitely should	63.7	62.9	62.2	78.9	66.2	85.8	73.6	F = 42.7
Probably should	32.6	32.7	34.9	19.2	29.9	12.2	21.7	
Total	96.3	95.6	97.1	98.1	96.1	98.0	95.3	
decent standard of living for unemployed								
Definitely should	46.3	9.4	15.3	37.3	26.9	36.5	41.7	F= 165.1
Probably should	39.6	30.5	45.1	40.1	44.7	39.4	42.3	
Total	85.9	39.9	60.4	77.4	71.6	75.9	84.0	
reduce income differences								
Definitely should	41.2	27.2	40.0	35.3	44.4	45.8	55.8	F = 51.7
Probably should	31.6	30.7	35.9	31.6	33.3	29.5	27.3	
Total	72.8	57.9	75.9	66.9	77.7	75.3	83.1	
decent housing								
Definitely should	35.3	27.3	21.1	37.0	35.8	49.6	50.2	F = 57.3
Probably should	39.9	47.0	52.1	43.4	47.3	39.4	37.9	
Total	75.2	74.3	73.2	80.4	83.1	89.0	88.1	
N	997	1,089	1,482	1,484	1,154	1,675	993	8,995
Source: ISSP, Role of Government, 1996.
552
Policy Studies Journal, 31:4
Comparing policies within countries, the distinctions are clear: respondents vary their level of support depending on the individual policy. Although a majority of countries retain high levels of backing for government action in all welfare policies, responses vary more widely in the Czech Republic and Hungary depending on the specific policy. But here again, there is no obvious pattern of post-communist beliefs. The Central European averages do not differ widely from those of the Former Soviet Union (FSU) countries, and the advanced transition states of the Czech Republic, Hungary, Poland, and Slovenia also lack consistent similarities on government responsibility. Regardless, across all policies and countries, respondents make distinctions between policies. There is little evidence of the "Communist" ideal of unanimous consent for government-controlled social policies. Even in the countries where support is generally high for government responsibility, unemployment receives one of the lowest ratings and pensions one of the highest, showing gradations in opinion depending on individual policies.
Turning to government spending levels on social policies, the pattern of program distinctions continues. In Table 2,1 again find that health care and pensions are more likely to find large levels of support, but additionally, education surfaces with majority backing. Russia and Latvia appear to be the only countries where a majority of their respondents desired "Spending much more" on most policies (except unemployment benefits), highlighting a possible FSU/Central and Eastern Europe pattern of more intense responses versus moderate support. The only policy of the four included in the survey in which citizens respond overwhelmingly against spending much more is unemployment assistance—with positive responses ranging from just 4% in the Czech Republic to 25% in Russia. As in the previous table, the ANOVAs show statistical differences across the countries on these policies. When comparing policy percentages within countries, I find that responses vary more depending on the individual policy than for the previous responsibility measure; no country retains the 80-90% responses witnessed in the previous table. Across the policies and countries, the support for government spending never reaches the same levels of backing as those for government responsibility, showing a willingness to accept government intervention but not increased budgetary allotments.3 But this does not explain how these areas correspond at a more structural level, for that factor analysis is necessary.
Welfare Belief Structures in Post-Communist Societies
General Findings
The figure and crosstab tables identify differences in the responses between government responsibility and government spending, but it is unclear whether or not these two aspects of welfare policy are grounded in the same policy preferences or if they are separate entities based within different beliefs. Factor analysis highlights the underlying correlations among these variables and situates them
Table 2. Attitudes toward Government Spending Levels for Specific Welfare Programs—Post-Communist Europe, 1996
% of respondents who answered "Spend much more" and "Spend more" when asked if the spending should be increased or decreased for a specific policy
Bulgaria Czech Hungary Latvia Poland Russia Slovenia ANOVAs
Health care
Spend much more Spend more Total
Retirement
Spend much more Spend more Total
Unemployment benefit Spend much more Spend more Total
Education
Spend much more Spend more Total
N
43.8 48.4
92.2
30.8
46.3 77.1
21.0
43.4 64.4
27.3 51.7 79.0
981
32.5
48.4 80.9
18.1
46.5
64.6
3.7 14.8 18.5
18.3
47.2 65.5
1,074
42.7 48.9
91.6
31.9
51.0 82.9
8.4
24.7
33.1
28.2 51.7 79.9
1,474
51.2 38.0 89.2
51.7
38.8
90.5
19.2 38.2
57.4
47.6
36.9
84.5
1,452
41.8 48.8 90.6
35.6 43.4 79.0
13.8 27.2 41.0
30.8 48.2 79.0
1,142
62.3 31.9 94.2
57.2 32.8
90.0
25.4
34.1
59.5
52.1
35.4
87.5
1,657
38.1
40.3
78.4
22.5
34.6
57.1
16.0
31.2
47.2
39.4
43.3
82.7
976
F = 68.4
F= 179.7
F= 167 i
F = 87.1
8,995
Source: ISSP, Role of Government, 1996.
554
Policy Studies Journal, 31:4
Table 3. Factor Analysis of Welfare Responsibilities and Spending—Post-Communist
Countries, 1996
Factors:
	1	2	3	4
Govt, redistribute wealth	0.252	0.020	0.516	0.193
Govt, control of wages	0.236	0.153	0.650	0.249
Govt, control of prices	0.249	0.156	0.740	0.143
Govt, create new jobs	0.251	0.233	0.181	0.158
Spend on environment	0.094	0.508	0.056	0.048
Spend on health care	0.259	0.638	0.174	0.110
Spend on law enforcement	0.016	0.378	0.012	0.107
Spend on education	0.253	0.661	0.099	0.152
Spend on defense	0.099	0.406	0.203	0.274
Spend on pensions	0.297	0.486	0.204	0.238
Spend on unemployment benefits	0.187	0.299	0.202	0.620
Spend on culture arts	0.153	0.545	0.053	0.203
Govt, responsibility to provide jobs	0.551	0.124	0.505	0.336
Govt, responsibility to keep prices under control	0.444	0.108	0.691	0.223
Govt, responsibility: health care	0.717	0.196	0.335	0.152
Govt, responsibility: pensions	0.735	0.219	0.326	0.232
Govt, responsibility: assist industry growth	0.488	0.210	0.303	0.359
Govt, responsibility: unemployed	0.406	0.128	0.341	0.805
Govt, responsibility: reduce income differences	0.429	0.062	0.595	0.396
Govt, responsibility: financial help for students	0.569	0.228	0.267	0.324
Govt, responsibility: housing	0.527	0.186	0.352	0.477
Variance explained:	210.6	8.0	4.5	3.8
Extraction Method: Principal Axis Factoring (PAF). Rotation Method: Oblimin with Kaiser Normalization. KMO Test = 0.846. Bartlett's Test of Spericity = 0.000.
into manageable, interpretable components. Table 3 displays the results of the principal axis factor analysis for the post-communist countries using variables that measure government responsibility and government spending in an assortment of policy areas—welfare, economy, defense, environment, and so on.
In general, the factors correspond with the expected vision between government responsibility and state welfare spending. Government responsibility for general welfare policies comprises the first factor, with all individual policies loading strongly, highlighting the belief that the government's role continues to be one of vast welfare responsibility, from cradle to grave. The second factor taps into government spending on welfare with two of the more "non-welfare" policies—law enforcement and defense—loading the weakest. Responses on welfare questions tend on one level to cohere to a responsibility versus spending divide while at a secondary level they point to some selectivity in responses. Unemployment appears to be a separate issue from economic or welfare concerns,
Lipsmeyer: Welfare and the Discriminating Public
555
because both responsibility and spending on unemployment assistance load onto a fourth factor. The breakout pattern of government responsibility and spending on unemployment benefits creating its own category indicates that respondents view welfare policies from three dimensions: government responsibility, state spending, and specific policies. Unlike the other factors, the third factor remains separate from the "welfare"-oriented components with questions of redistribution and economics loading onto it, which shows that respondents make a distinction between purely welfare responsibilities and the government's economic role.
What Is It About Unemployment Benefits?
Figure 1 presents welfare preferences across the various policies in multiple post-communist countries to illustrate why respondents view unemployment benefits more negatively than other policies. At a basic level of analysis, citizens appear to distinguish between government responsibility and government spending on social welfare. However, unemployment benefits appear to be widely unsupported. Whereas responsibility for pensions and health care remain the most popular responses in every country, unemployment tends to be the least favored of policies, for both the responsibility and spending realms.
During the transition, these countries moved from communist-era full employment to unemployment caused by marketization, resulting in the creation of unemployment benefit schemes. Unlike assistance policies that had existed under communism, these benefits were a result of the transition and its new capitalist ideology. Figure 1 also includes an indicator of the unemployment rate in
556
Policy Studies Journal, 31:4
each country from 1995. One could expect that the economic and social hardships of the transition, shown through high unemployment, might influence society's preferences for buffering the out-of-work, but the picture here is one of mixed signals. A low unemployment rate does correlate with a negative attitude toward unemployment benefits in the Czech Republic, which corresponds with other's findings in that country from 1998 (Rabusic & Sirovatka, 1999). But no other country shows such a strong relationship.
If we examine aggregate support for unemployment, then the negative response occurs across most age and income groups—and for both government responsibility and spending. As Figure 2 illustrates, the average responses for unemployment benefits are the most negative across the age cohorts for all welfare policies. When asked about government responsibility, there is little change between policies for the age groups; that is, health care remains the most supported of the policies and unemployment the least. For government funding, the youngest cohort is slightly more likely to favor unemployment spending, but overall, the negative trend continues. Note that the other policies correspond with their expected cohort groups. The youngest cohort favors education spending, which changes by the oldest cohort to the second-least supported policy. But this pattern is the opposite for pensions; the oldest cohort backs pension funding, whereas the youngest group ranks it second to unemployment as a least favorite. These figures show that age groups distinguish between policies, yet they all place unemployment benefits in the lowest of categories for government responsibility and funding.
Because unemployment benefits are income replacement, Figure 3 shows preferences based on income level. The pattern across both responsibility and spending attitudes remains a negative one; unemployment continues to be the least supported policy of the group. Unlike the previous graphs, though, the relative means correspond somewhat with what one would expect—the groups most likely to require unemployment benefits, the lowest income quintiles, are the least negative. The middle-income group is the least supportive: the group on the cusp of getting by without support and stigmatizing those in the new system that require assistance. These figures point to a complex relationship of unemployment benefits with marketization and welfare, for the newness of the policy combined with its linkage to the transition presents a problem for countries accustomed to full employment and going through a difficult economic and social hardship.
Conclusion
The newly democratic publics of post-communist Europe are astute political players, for not only do they distinguish between the governments' role in structuring social policies and their spending on them, but they also select individual policies to support. Although these societies retain elements of their socialist pasts, citizens' attitudes show discernment in their welfare preferences and attitudes.
Lipsmeyer: Welfare and the Discriminating Public 557
Responsibility Attitudes for Individual Welfare
3.0 i-1
2.5
I 2.0
1.5
1.0
\
Policy
I Health Care
23 Old-Age Pensions Unemployment
CSM Education I Housing
15-30 31-45 46-60
Age Cohort
61-
Spending Attitudes for Individual Welfare Policies
cz cd
cd
Policy
Health Care
Education
Old-Age Pensions
Unemployment
15-30
31-45 46-60
Age Cohort
Figure 2. Welfare Preferences for Welfare Benefits by Age Cohort, Post-Communist
Countries, 1996.
558
Policy Studies Journal, 31:4
Responsibility Attitudes for Individual Policies
3.0
Policy
Health Care Old-Age Pensions Unemployment Education Housing
Missing 2 4
1--Highest        3~Middle 5~Lowest Income
Spending Attitudes for Individual Policies
3.5>
c cg
cd
3.0
2.5
2.0
1.5
1.0
Policy Health Care Education Old-Age Pensions
^^Unemployment
Missing 2 4
1-Highest        3~Middle 5-Lowest Income
Figure 3. Welfare Preferences for Welfare Benefits by Income Group, Post-Communist
Countries, 1996.
Lipsmeyer: Welfare and the Discriminating Public
559
Respondents may be accustomed to universal welfare assistance, but in accordance with preferences in established democracies, not all welfare policies receive the same public endorsement and support.
At a structural level, divisions between FSU and Central/Eastern Europe apparently do not explain the variation between post-communist countries on welfare responsibility. For example, respondents in the Czech Republic— arguably, one of the most economically stable Central European countries—have on average the lowest levels of support for government responsibility of welfare policies. However, Hungarian and Polish responses, from two of the other stable Central European countries, do not correspond with the low Czech level. The most obvious example of how economic growth does not correlate with welfare responsibility preferences comes from Slovenia, where the averages rival the Russian respondents. When evaluating unemployment attitudes, overall I find that the unemployment rate does not correspond with levels of support for unemployment assistance; instead, responses are generally negative toward the new welfare assistance.
But the most intriguing finding is preferences for unemployment benefits. In the factor analysis, the responses on the responsibility for and spending on unemployment produce a separate attitude from either the economic or welfare dimensions. The first possible explanation derives from the severity of unemployment, which may cause citizens to view it as a unique problem that transcends the economic and welfare lines; however, as we saw, this attitudinal distinction exists for a variety of countries regardless of the seriousness of the unemployment rate. The second alternative diminishes the emphasis on the level of unemployment and instead notes the newness of the programs. Under communism, these countries officially had full employment and no need for benefits, so with the arrival of market forces, all countries implemented some version of unemployment assistance (SSA, 1991). Lacking the understanding of how unemployment benefits provide a buffer in a market economy, the support for unemployment benefits in these societies may require time. The benefits possibly will fit less into the traditional universal system of welfare and more into a hybrid category that straddles economic and welfare attitudes. The lack of support across age and income groups emphasizes its shallow support and may suggest a reason why governments were able to weaken these benefits during the mid-1990s by cutting their durations and replacement rates (Lipsmeyer, 2000). An interesting note, however, remains the high negative average for the middle-income groups, pointing to a potential connection with West European attitudes: "Unemployment as a social policy drives a wedge between the majority of the working class in work . . . and the minority of the unemployed" (Taylor-Gooby, 1985, p. 78).
In post-communist countries, citizens retain their dependence on the state for structuring and budgeting their social policies, but this comes at a time when the states are reevaluating their roles in social policy and making changes that do not correspond with this picture of the public's preferences. Over 90% of respondents
560
Policy Studies Journal, 31:4
across the countries, for example, prefer that the government be responsible for old-age pensions, when the reality is that countries are attempting to find ways to privatize at least part of their pension systems through multitiered schemes (e.g., Batty, Stumpa, & Kovari, 1994). A large majority of respondents also prefer state responsibility and increased spending for health care, which does not correspond with some government actions to push the costs of the policy onto salary-based funds of employers and employees (Lipsmeyer, 2000).4 The social policy in which the public's attitudes and government actions are the most consistent is unemployment benefits. Across post-communist countries, we find that the least amount of support is for increasing funding for unemployment policies, and because of the increasing number of unemployed during the transition, many governments find it difficult to continue the initial generous funding levels for those out of work.5
For future research, this picture of welfare attitudes points to clues of how governments change post-communist welfare systems. Examining welfare from both government responsibility and spending dimensions, using multiple data points during the transition, we may find that citizens' attitudes on one dimension may not preclude adjustment in the other sphere. When asked about increasing spending in specific areas, although the support varies, it remains strong for most policies. These findings might help explain previous conclusions that governments are less likely to alter spending levels directly and instead change the rules surrounding welfare benefits to make them more restrictive and less expensive. Although support for government intervention is high and support for spending remains strong for most programs, individual programs with relatively less support or those that do not tap into attitudes for particularly popular policies may be more susceptible to alteration. As a result, the cuts in funding of certain policies, the devolution of programs to regional governments, and the changes in the rules and regulations for receiving benefits (i.e., means-testing of benefits) may not spur an electoral backlash. This prompts the question: How are governments able to make such cuts without losing substantial public support and legitimacy? One answer may be that the transformation of the welfare states in post-communist countries appears loosely to correspond with society's attitudes and preferences.
Christine S. Lipsmeyer is an assistant professor of political science at the University of Missouri, Columbia. Using political economy and survey methodologies, her research investigates the relationships between welfare policy, politics, and economic influences in post-communist Europe. She has published articles in a number of journals, including the British Journal of Political Science, Electoral Studies, and the Journal of European Public Policy.
Lipsmeyer: Welfare and the Discriminating Public
561
Appendix
Wording of the Survey Questions:
Q.8   What is your opinion of the following statement:
It is the responsibility of the government to reduce the differences in income between people with high incomes and those with low incomes.
1. Agree strongly
2. Agree
3. Neither agree nor disagree
4. Disagree
5. Disagree strongly
Q.9   Here are some things the government might do for the economy.
Please show which actions you are in favour of and which you are against. Please tick one box in each line.
Q.9a Government action for economy: Control of wages by law Q.9b Government action for economy: Control of prices by law Q.9c Government action for economy: Government financing of projects to create new jobs.
1. Strongly in favour of
2. In favour of
3. Neither in favour of nor against
4. Against
5. Strongly against
Q.10 Listed below are various areas of government spending.
Please show whether you would like to see more or less government spending in each area. Remember that if you say "much more", it might require a tax increase to pay for it.
Q	10a	The environment.
Q	10b	Health.
Q	10c	The police and law enforcement.
Q	lOd	Education.
Q	lOe	The military and defense.
Q	lOf	Old age pensions.
Q	10g	Unemployment benefits.
Q	10h	Culture and the arts.
1. Spend much more
2. Spend more
3. Spend the same as now
4. Spend less
5. Spend much less
562
Policy Studies Journal, 31:4
Q.12 On the whole, do you think it should be or should not be the government's responsibility to:
Q.12a Provide a job for everyone who wants one.
Q.12b Keep prices under control.
Q.12c Provide health care for the sick.
Q.12d Provide a decent standard of living for the old.
Q.12e Provide industry with the help it needs to grow.
Q.12f Provide a decent standard of living for the unemployed.
Q.12g Reduce income differences between the rich and poor.
Q.12h Give financial help to university students from low-income families
Q.12i Provide decent housing for those who can't afford it
1. Definitely should be
2. Probably should be
3. Probably should not be
4. Definitely should not be
Notes
1. Given the vast literature on post-communist surveys, this paragraph only supplies examples of the various topics and works. Many of the pieces offer more in-depth bibliographies.
2. The most widely used method of factor analysis is principal components analysis (PCA); however, in this paper, I am interested in hypothesis testing, which uses PFA (Kim & Mueller, 1978a). Paincipal Factor Anaysis (PFA) seeks a solution using the common variance among a set of variables, whereas PCS accounts for the common and unique variance in the variables.
3. Scholars have argued that for the U.S. case, citizens tend toward "ideological schizophrenia": they argue for smaller government while at the same time demanding increased government spending (Free & Cantril, 1967; Bennett & Bennett, 1990; see Jacoby, 1994, for an alternative hypothesis). The post-communist countries tend to have the opposite problem of preferring more government intervention but less state spending.
4. Tentatively, researchers have located the cause of these government changes within the ideologies of the political parties in power (Lipsmeyer, 2000, 2002; Cook & Orenstein, 1999).
5. The World Bank has advocated this position stating: "Declining contributions and a growing number of unemployed placed more programs in financial jeopardy" (World Bank, 2000).
References
AndreG, H. J., & Heien, T. (2001). Four worlds of welfare state attitudes? A comparison of Germany, Norway, and the United States. European Sociological Review, 17, 337-356.
Andrews, E. S., & Ringold, D. (1999). Safety nets in transition economies: Toward a reform strategy. World Bank Social Protection Discussion Paper No. 9914, Washington, DC.
Bahry, D., & Lipsmeyer, C. S. (2001). Economic adversity and public mobilization in Russia. Electoral Studies, 20(3), 371-398.
Barr, N. (Ed.). (1994). Labor markets and social policy in Central and Eastern Europe: The transition and beyond. Oxford: Oxford University Press.
Batty, L, Stumpa, R., & Kovari, I. (1994). Pension reform in Eastern Europe: Hungary and Czech Republic lead the way. Benefits and Compensation International, 23, 8-14.
Lipsmeyer: Welfare and the Discriminating Public
563
Bennett, L. L., & Bennett, S. E. (1990). Living with leviathan: Americans coming to terms with big government. Lawrence: University Press of Kansas.
Central Intelligence Agency. (1996). The world factbook. Washington, DC: Office of Public Affairs.
Colton, T. (1996). Economics and voting in Russia. Post-Soviet Affairs, 12, 289-317.
Cook, L. J., & Orenstein, M. (1999). The return of the Left and its impact on the welfare state in Russia, Poland, and Hungary. In L. Cook, M. Orenstein, & M. Rueschemeyer (Eds.), Left parties and social policy in postcommunist Europe (pp. 109-143). Boulder, CO: Westview Press.
Csontos, L., Kornai, J., & Gyorgy, I. (1998). Tax awareness and reform of the welfare state: Hungarian survey results. Economics of Transition, 6, 287-312.
Duch, R. M. (1995). Economic chaos and the fragility of democratic transition in former communist regimes. Journal of Politics, 57, 121-158.
Evans, G. (1996). Cross-national support for welfare and redistribution: An analysis of competing theories. In B. Taylor & K. Thomson (Eds.), Understanding change in social attitudes. Darmouth: Aldershot, UK.
Evans, G. (1998). Britain and Europe: Separate worlds of welfare? Government and Opposition, 33, 183-198.
Evans, G., & Whitefield, S. (1993). Identifying the bases of party competition in Eastern Europe. British Journal of Political Science, 23, 521-548.
Evans, G., & Whitefield, S. (1996). The social bases of electoral competition in Eastern Europe. Paper presented at the Annual Meeting of the American Political Science Association, August Chicago.
Free, L. A., & Cantril, H. (1967). The political beliefs of Americans: A study of public opinion. New York: Simon & Schuster.
Gibson, J. L. (1998). A sober second thought: An experiment in persuading Russians to tolerate.
American Journal of Political Science, 42, 819-850.
Haller, M., Höllinger, F., & Raubal, O. (1990). Leviathan or welfare state? Attitudes toward the role of the government in six advanced Western nations. In J. W. Becker, J. A. Davis, P. Ester, & P. P. Möhler (Eds.), Attitudes in inequality and the role of government. Rijskijk, Netherlands: Social and Cultural Planning Office.
International Social Survey Program (ISSP). (1996) International Social Survey Program: Role of government III [Computer file]. Cologne, Germany: Zentralarchiv fuer Empirische Sozialforschung [producer], 1999. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 1999.
Jacoby, W. G. (1994). Public attitudes toward government spending. American Journal of Political Science, 38, 336-361.
Kim, J. O., & Mueller, C. W. (1978a). Factor analysis: Statistical methods and practical issues. Newbury Park, CA: Sage Publications, Inc.
Kim, J. O., & Mueller, C. W. (1978b). Introduction to factor analysis: What it is and how to do it. Newbury Park, CA: Sage Publications, Inc.
Kitschelt, H. (1995). Formation of party cleavages in post-communist democracies: Theoretical propositions. Party Politics, 1, 447^172.
Kitschelt, H., Mansfeldova, Z., Markowski, R., & Töka, G. (1999). Post-communist party systems. Competition, representation, and inter-party cooperation. Cambridge, United Kingdom: Cambridge University Press.
Lipsmeyer, C. S. (2000). Reading between the welfare lines: Politics and policy structure in post-communist Europe. Europe-Asia Studies, 52, 1191-1211.
Lipsmeyer, C. S. (2002). Parties and policy: Evaluating economic and partisan influences on welfare policy spending during the European post-communist transition. British Journal of Political Science, 32, 641-661.
Mason, D. S., & Kluegel, J. R. (2000). Marketing democracy: Changing opinion about inequality and politics in East Central Europe. Lanham, MD: Rowman & Littlefield Publishers, Inc.
564
Policy Studies Journal, 31:4
Miller, A., Erb, G., Reisinger, W. M., & Hesli, V. L. (2000). Emerging party systems in post-Soviet societies: Fact or fiction? Journal of Politics, 62, 455-490.
Miller, A., Hesli, V., & Reisinger, W. (1997). Conceptions of democracy among mass and elite in post-Soviet societies. British Journal of Political Science, 27, 157-190.
Rabusic, L., & Sirovafka, T. (1999). Czech welfare state and its legitimacy. Polish Sociological Review, 2, 239-263.
Roller, E. (1999). Shrinking the welfare state: Citizens' attitudes towards cuts in social spending in Germany in the 1990s. German Politics, 8, 21-39.
Rose, R., & Haerpfer, C. (1994). Mass response to transformation in post-communist societies. Europe-Asia Studies, 46, 3-28.
Rose, R., & Makkai, T. (1995). Consensus or dissensus about welfare in post-communist societies? European Journal of Political Research, 28, 203-224.
Shapiro, R. Y., & Young, J. T. (1989). Public opinion and the welfare state: The United States in comparative perspective. Political Science Quarterly, 104, 59-89.
Social Security Administration. (1991, 1993, 1995, 1997). Social security programs throughout the world. Washington, DC: U.S. Department of Health and Human Services.
Taylor-Gooby, P. (1985). Public opinion, ideology and state welfare. London: Routledge & Kegan Paul.
Whitefield, S. (2002). The political economy of post-communist welfare reform: Public support for targeting and poverty alleviation in Ukraine. Post-Soviet Affairs, 18, 71-85.
Whitefield, S., & Evans, G. (1994). The Russian election of 1993: Public opinion and the transition experience. Post-Soviet Affairs, 10, 38-60.
World Bank. (2000). Balancing protection and opportunity: A strategy for social protection in transition economies. Washington, DC: Author.
Wyman, M., White, S., Miller, B., & Heywood, P. (1995). Public opinion, parties and voters in the December 1993 elections. Europe-Asia Studies, 47, 591-614.