1While research on social policy and welfare arrangements in the context of advanced industrialized democracies has a long tradition, much of the existing literature engaging the post-communist countries is single country-focused, or collections of case-studies, and generally does not engage with scholarly work on other European countries. Further, systematic data on social policy arrangements (with the exception of spending) in the post-communist countries are still missing, and existing welfare regime classifications do not neatly fit the newer European democracies. Finally, as many of these countries are now part of the European Union, and thus expected to adopt policies consistent with the European Social Model (ESM), it is far from clear (1) where do the post-communist countries ‘fit’ in the European welfare models, and (2) how (and if at all) existing welfare typologies and classifications help place them comparatively in the context of countries with longer democratic traditions.

2This paper begins to address the scholarly and data gaps in discussing post-communist social policy arrangements in the context of other European ‘models.’ It does so by proposing an approach for distinguishing among welfare regimes that allows for more detailed comparisons. The goal is not to dismiss existing typologies and classifications, developed mostly from the study of Western democracies, but to go beyond them and to help think of ways to capture the diversity of social policy arrangements both within and between ‘typical’ welfare models in ways that allow for the newer European democracies to be more easily included in the comparison. In particular, this strategy for comparison focuses on two dimensions of social policy : who pays for it, and how are benefits allocated. These two aspects reflect notions of solidarity, personal vs. shared responsibilities, types of societal divisions, and how the latter might be perpetuated by social policy. The paper starts with a condensed overview the numerous ways in which scholars have classified and categorized welfare regimes since the end of World War II. This overview is far from comprehensive but serves to highlight some of the challenges in creating welfare comparisons and in distinguishing among welfare states.

3The next section focuses on discussing five social protection programs—old age pensions, sickness and maternity, unemployment, work injury compensation, and family allowance—across 30 European countries : the 28 EU member states, and Norway and Switzerland. While old age pension programs target retirees, sickness and maternity, unemployment, and work injury compensation policies are designed as (temporary) income replacement for working-age population ; finally, family allowances include primarily subsidies to help raise children up to a certain age. Thus, these programs serve a variety of purposes and including them in the same analysis allows for a more comprehensive comparison among countries. Each program is categorized first in terms of how it is managed, then—by how it is funded, and finally—by the restrictiveness of its eligibility criteria. While this paper overviews social policy arrangements in all of Europe (with the exception of a few island and city-state countries), special attention is given to the post-communist countries, in an attempt to understand where their social policies fit within the typologies of established welfare states. [1] Such an evaluation of the place of post-communist countries helps answer the question of whether there exists a ‘typical’ post-communist welfare arrangement (it does not), and it also raises the more interesting question of what explains the differences among social policies in the new European democracies, given their common historical legacies and external pressures to reform their political, economic and social structures.

Distinguishing among welfare regimes [2]

4In addition to their spending generosity, the welfare arrangements that a country develops over time differ on a number of dimensions. At the most basic level, countries vary by the types of social benefits that they provide. Such differences are indicative of the reasons for the development of each particular welfare and social policy arrangement. While the provision of pensions, as well as health, unemployment, sickness and accident coverage is universal among European countries, differences exist with respect to the scope and eligibility of coverage. Another dimension on which arrangements differ is their sources of funding, i.e. public/state, individual employer/private and the extent to which one of these sources dominates. Finally, countries differ in the outcomes of their social and welfare provisions, i.e. the level of income redistribution and social equality that is achieved, or the level of income substitution that cash benefits provide, or the percentage of the population that is educated, and lives long healthy lives, etc.

5Even though there have been earlier attempts to classify welfare states, [3] Esping-Andersen’s categorization in The Three Worlds of Welfare Capitalism dominates the scholarly literature. [4] Esping-Andersen focuses on the extents of decommodification and stratification when comparing welfare states. Decommodification reflects “the degree to which individuals, or families, could uphold a socially acceptable standard of living independent of market participation,” [5] while stratification accounts for how states distinguish among societal and occupational groups when conducting social policies. Based on these two dimensions, Esping-Andersen’s well-known three types of welfare states, ranked from least to most decommodifying, and from most to least stratifying are : liberal, corporatist and social-democratic. Unlike Esping-Andersen who focuses on the balance of power between different classes in a society, the Castles’ ‘families of nations’ approach looks for broader linguistic, historical and cultural explanations for differences in policy outcomes among nations. [6] Both the Esping-Andersen and the Castles’ classifications focus on the welfare aspects of the broader notion of the European Social Model (ESM) which also includes normative discussions of shared values, citizenship rights, as well as employment and labour policies in addition to social welfare practices and outcomes. [7] The rich ESM literature distinguish among four groups of countries (Nordic, Continental, Anglo-Saxon, Mediterranean), [8] with the post-communist set potentially comprising a fifth subcategory in the ESM. [9] Fundamentally, however, differences between types of welfare regime remain rather subtle, especially when scholars are focused only on few dimensions, and as long as they are discussing the entire welfare ‘regime’ as a single unit of analysis. For example, level of spending does not differ profoundly among different types of welfare regimes, esp. between social-democratic and corporatist. Where differences begin to emerge is in the source and the structure of the spending, i.e. (1) how are social programs funded—taxation or social insurance contribution, (2) how are benefits paid—via means testing or universal eligibility.

6Bonoli moves beyond types and classes of welfare states and instead applies a two-dimensional classification that captures the how much and the how of welfare provision. [10] He distinguishes among countries depending on their level of spending and on where they stand on the Bismarkian-Beveregian continuum, which is based on the historical development of the modern welfare state in Western Europe. Bismarckian social policies use social insurance mechanisms to provide earnings-related benefits for employees. Entitlement is conditional upon a satisfactory employment and contribution record, and financing is mainly based on employer/employee contributions. Beveridgean social policy is characterized by universal provision of benefits with entitlement that is based on residence and/or need. Benefits are typically flat rate and are financed through general taxation. Bonoli then uses the proportion of contribution financing as a measure of the approximation to Bismarckian type. When the source of the contributions is combined with level of spending, it produces four models of welfare states : Beveridgian with high-spending, Beveridgian with low-spending, Bismarckian with high-spending, and Bismarckian with low-spending.

7Yet another way to look at types or models of welfare provisions is Ferrera’s coverage approach that focuses on whether, and the extent to which, social policy distinguishes among different societal and occupational groups. [11] Coverage can be universal, when the entire population is covered by a single scheme, or occupational, or means tested, when different groups in society are covered by different schemes depending on their employment or income status. Universal provision is a typical feature in Scandinavian social policy, as well as (partly) in Britain. By contrast, continental European countries are generally characterized by fragmentation of social protection schemes along occupational lines. Universal coverage is usually related to tax-financed provisions, while occupational coverage is related to contributory social insurance. Similarly and more recently, Lynch discusses whether social and tax expenditure policies across western democracies vary in their age-orientation, i.e. do they emphasize benefits for the elderly or not. [12] Lastly, a very recent Comparative Welfare Entitlements Dataset (CWED) developed at the University of Connecticut in the United States and University of Greifswald in Germany goes beyond classic welfare regime distinctions and creates measures for benefits generosity of a number of policies. CWED is the first welfare policy dataset that includes countries from both Western and Eastern Europe, although its measures on a number of policies are still missing for the post-communist countries, preventing a more widespread use of these data. [13]

8The more limited research on social policy of the post-communist countries tends to emphasize the region’s uniformity, [14] to focus on determinants for policy transformation, [15] including role of ideology, [16] or international agencies, [17] and is often limited to individual countries and/or policy aspects. [18] Much of these works utilize general measures such as spending and/or broad conceptual distinctions, and pay less attention on the more fine-tuned distinctions across post-communist states. An exception is the recent work by Bohle and Greskovits, which analyses the post-communist economies on a number of dimensions, including social policies, to classify these countries into three groups based on “the vigor with which, and the forms in which transformative actors have used state power to build market economies pursuing the goals of neoliberalism, and to simultaneously preserve social cohesion…” [19] What distinguishes this work and its findings is that the authors utilize typologies that are applicable in a wide range of setting, thus acknowledging the diversity among the post-communist countries, and the need to put them in the context of broader European economic developments, categorizing these countries in three groups : neoliberal, embedded neoliberal, and neocorporatist. Overall, though, research that emphasizes differences in the nature and structure of post-communist economies, including comparisons with institutional arrangements in other European countries, is relatively rare. Thus, the section below turns to presenting a way to think about social welfare arrangements in European democracies, which accounts for the diversity among them and allows for broader within and cross-regional comparisons.

Beyond welfare regime types

9Keeping in mind existing research, as well as the tremendous diversity of social and welfare policies, this section introduces a new dataset that characterizes the social provisions in European countries in a number of ways. Similar to Kopri and Palme, the dataset discussed here focuses on the institutional arrangements of social policies across the European democracies, as a fundamental component of scholarly analysis of differences between and within welfare regime types. [20] Social policies are compared along two dimensions, source of funding and restrictions on eligibility, for the following reasons. The distribution of the responsibilities for financing each social policy is a fundamental component of distinguishing among welfare regimes. Who pays for each policy, in addition to how much, reflects an understanding of whether government, employers, or individuals should bear the primary responsibility for citizens’ well-being. Normatively, it also speaks to a notion of societal and cross-generational solidarity that helps provide legitimacy for these redistributive programs. [21] Despite the labels attached to various social policy regimes, which group finances welfare spending remains a critical question, one that is especially relevant for the post-communist countries, which rebuilt their modern social policies in the last 25 years. Much of the existing welfare state/regime classifications could hardly be applied to these countries ; yet comparing them by the source of social spending is a good way to put them in the context of other European countries. In addition to differences in sources of funding, the section also discusses variation among countries in terms of the eligibility requirements for each program, i.e. how difficult it is for a recipient to qualify for benefits. Whether or how countries restrict who can receive what benefits reflects the inclusiveness of their social and welfare policies, which has a direct relevance to the desired outcomes of the ESM.

10Comparisons based on these two dimensions, solidarity (how is the burden of cost shared) and inclusiveness (how is eligibility restricted) are conducted across three sets of social protection programs : old age pensions, income replacement (sickness, maternity, work injury and unemployment compensations), and family allowance. To conduct this comparison, the paper relies on information primarily from the United States Social Security Administration’s (SSA) edition of Social Security Programs Around the World, which reviews these programs (and others) in the majority of the world’s countries. [22] Information available through the SSA was coded to account for program level diversity across countries, with the coding scheme available in Appendix A. [23] The European Commission’s Eurostat division was used for data on social protection funding.

Who pays—overall social protection funding

11Figure 1 shows 30 European countries (the 28 EU member states, and Norway and Switzerland) ranked by the percentage of government contribution for overall social protection, using Eurostat data for 2011. [24] Most countries have five sources of funds towards their annual social protection expenses, central government, employers, employees, self-employed individuals, and so-called protected groups (pensioners and people receiving social support). Central government, employers, and employees are the main contributors to social protection funds, with the other two groups representing an average of 3-5% of contributions, and for presentation purposes, receipts from self-employed and protected individuals are excluded from the graph in Figure 1. [25] The data reveal that only in two countries, Denmark and Ireland, the central government contributes most of the funds towards social protection, with six other countries (Bulgaria, Cyprus, Malta, Norway, Romania, and Sweden) where the government contributes around or a little over 50% for social protection. In the rest of the countries, contributions by employers and employees finance the majority of spending for social protection.

Figure 1

Receipts for social protection, by category, for 30 European countries, 2011 data

Receipts for social protection, by category, for 30 European countries, 2011 data

Data for government and employers contributions only

12In Estonia employee contributions are minimal, and those are also relatively small in Italy and Spain, as well as in Sweden. Countries such as Croatia, Slovenia, and to some extent Switzerland, rely on relatively equal distribution of contributions among the three groups. Overall, Figure 1 shows the diversity across European countries—within and across traditional social models—as to who is responsible for funding social protection spending. The figure also highlights that in terms of the source of funds for social spending, there is no single post-communist model of social protection. Bulgaria and Romania are closer to some of the Scandinavian countries, while Hungary, Latvia, and Slovakia resemble Southern European countries such as Spain and Greece. On the other hand, Croatia, Lithuania, and Slovenia have adopted more of a continental model (resembling Germany, France, Austria), at least regarding the share of government contributions for social protection. At the lower end of the government share of contributions are the Czech Republic, Estonia, and Poland, resembling Switzerland.

13Figure 2 presents the same data in a different way, focusing on the ratio of employer to employee contributions to highlight how is financing for social protection shared, in addition to the central government contributions. [26] It is assumed and expected that in all European countries, government has a responsibility to finance at least part of the spending for social programs, although as discussed above, countries differ in the extent to which government serves as the main source of finding. Who else, in addition to government, and to what extent, finances social protection is often ignored in the literature, yet it constitutes a fundamental component of the structure of each country’s social model. In only three countries, Croatia, Slovenia, and Switzerland, the share of employer contributions is lower than that of employees, while in the rest of the European countries, employer contributes a greater share than the employee although there is much diversity in the ratio between the contributions of the two groups.

Figure 2

Ratio of employer to employee contributions, for 30 European countries, 2011 data

Ratio of employer to employee contributions, for 30 European countries, 2011 data

The Netherlands is excluded due to lack of employee data. Estonia was not included in the figure as its ratio of 41 skews scale on which the data are shown.

14Denmark and Luxembourg have relatively equal shares of employer – employee ratios, around or less than 1.5, indicating a notion of ‘shared responsibility’ between the two groups for the financing of social protection. In the majority of the remaining countries, employers contribute between 1.5, 2 and 3 times more than employees, while a small number of countries (Finland, Italy, Lithuania, Spain, and Sweden) the share of employer contributions exceeds that of employees more than three times. [27] Figure 2 shows again the hidden diversity within and between classic welfare models. While in terms of share of government contributions, Sweden and Norway, Germany and Austria, and Italy and Spain are relatively similar to each other (Figure 1), such groupings are much harder to identify in Figure 2. Denmark, Norway, and Sweden differ greatly in the extent to which employers and employees finance social protection, as do countries in the corporatist model such as Germany and France. Also, there is again clearly no post-communist model of distribution of social protection funding, with countries as different in their employer – employee ratios as Slovenia (less than 1) and Slovakia (closer to 3).

15Next, the paper turns to comparing the individual components of social policies among European countries, focusing on the type of program, the funding sources, and how restrictive are the eligibility criteria, e.g. employment requirement, means testing, age, etc. Using information from the United States Social Security Administration 2012 edition, the section below focuses on three groups of social programs : old age pensions, income replacement, which includes sickness, maternity, work injury and unemployment, and family allowance. For comparison purposes, the discussion is organized based on the program type, funding source, and eligibility restrictions, across the three groups of social programs.

Type of programs

16In terms of how old age pension, income replacement (sickness and maternity, unemployment, and work injury), and family allowance aspects of the social protection systems are organized, countries rely on social insurance, social assistance (means tested), universal (residency), mandatory individual accounts, and employer liability programs, or a combination of these. The data in Table 1 indicates that with the exception of the family allowance income supplementation, countries rely primarily on social insurance types of programs (most often combined with a means tested social assistance) to manage old age pensions and income replacement aspects of social protection. [28]

Table 1

Types of social protection programs in 30 European countries

Old age pensions Sickness and maternity Unemployment Work injury Family allowance SI AT, BE, CZ, DE, EL, HU, LU, ES, NL AT, BE, BG, HR, CY, EE, FI, FR, DE, HU, LV, LT, LU, PL, RO, SK, SL, ES, SE, CH, NL AT, BE, BG, HR, CY, CZ, EL, HU, IT, LV, LT, LU, PL, RO, SK, SL, ES, CH AT, BE, BG, CY, CZ, EE, FR, DE, EL, HU, IE, IT, LV, LT, LU, MT, PL, RO, SK, SL, ES, SE, CH BE, EL, IT, CH SA BG, HR, EE SI, SA CY, IE, LT, MT, PT, SL, UK UK DK, EE, FI, FR, DE, IE, MT, PT, SE, NL, UK UK MT, ES SI, MI EE, IT, RO, SK, SE, CH SI, U DK, NO CZ NO NO SI, SA, MI BG, HR, FR, PL DK SI, SA, U PT SA, U DE, IE, UK U AT, CY, CZ, FI, FR, HU, LV, LT, LU, NO, PL, PT, RO, SK, SL, NL EL DK SI, EL IE, IT, MT, NO PT MI, EL FI

Types of social protection programs in 30 European countries

SI : social insurance, SA : social assistance, U : universal, MI : mandatory individual, EL : employer liability
Country codes available here : http://epp.eurostat.ec.europa.eu/statistics_explained/index.php/Glossary:Country_codes

17Universal type of programs, based on residency are relatively rare and dominate family allowance policies, although Denmark, Czech Republic, and Norway have such programs for old age pensions, sickness and maternity, and unemployment, respectively. A handful of countries, among them Bulgaria, Croatia, Denmark, Estonia, Italy, and Poland require mandatory insurance schemes for old age pensions and/or work injury protection, while very few countries, e.g. Denmark, Italy, Malta use employer liability for sickness and maternity, for example. Overall, countries show the greatest diversity in how their sickness and maternity programs are managed, and the least—with respected to unemployment compensation.

Who pays—at the individual program level

18Table 2 lists the ratio of employer to employee contributions, in cases where both groups contribute towards social protection, for each of the five programs in the 30 European countries. [29] A ratio of less than zero indicates that employees contribute a greater share than employers, and vice versa. A ratio of one is indicative of equally shared costs between the two groups. The share of contributions by central government is excluded from the table due to unavailability. In almost all countries and programs, government funds any program deficits or makes a subsidy but their share of the overall contributions is not announced. Table 2 though indicates when certain programs are fully funded by government, by employers, or by employees only. The data in Table 2 thus supplements information presented in Figure 1, which shows how the share of each group’s contribution to the overall cost of social protection is distributed. Looking how costs are shared at the level of each program is another—more detailed—indicator of how societies view responsibilities for funding social protection and if and how such views differ across types of social programs—pensions, income replacement, and subsidies.

Table 2

Ratio of employer and employee share of contributions, by social protection program in 30 European countries^{*,**,***,*****}

Country Old age pensions Sickness and maternity Unempl. Work injury Family allowance Austria 1.2 0.9 1.0 *** *** Belgium 1.2 1.9 1.7 *** *** Bulgaria 1.3 1.5 1.5 *** ** Croatia * *** *** *** ** Cyprus 1.0 1.0 1.0 1.0 ** Czech Republic 3.3 *** *** *** ** Denmark ** *** **** *** ** Estonia *** *** 0.5 *** ** Finland 3.0 2.4 ** *** *** France 1.2 3.7 1.8 *** ** Germany 1.0 0.9 1.0 *** ** Greece 2.0 2.0 1.7 2.0 1.0 Hungary 2.7 0.3 0.7 2.7 ** Ireland 2.7 2.7 2.7 2.7 ** Italy 2.6 *** 5.4 *** *** Latvia 2.2 2.2 2.2 2.2 ** Lithuania 7.8 *** *** *** ** Luxembourg 1.0 1.0 * *** ** Malta 1.0 1.0 1.0 1.0 1.0 Netherlands * 1.3 0.3 *** ** Norway 1.8 1.8 1.8 *** ** Poland 1.0 * *** *** ** Portugal 2.1 2.1 2.1 *** 2.1 Romania 2.0 *** 1.0 *** ** Slovak Republic 3.5 1.0 1.0 *** ** Slovenia 0.6 1.0 0.4 *** ** Spain 5.0 5.0 3.5 *** ** Sweden 1.5 *** ** *** ** Switzerland 1.0 * 1.0 *** *** United Kingdom 1.3 1.3 1.3 1.3 **

Ratio of employer and employee share of contributions, by social protection program in 30 European countries^{*,**,***,*****}

* Employers do not contribute to this program
** Neither employees nor employers contribute to this program
*** Employees do not contribute to this program
**** Employer contributes at own discretion

19Looking across programs, for three of them, old age pensions, sickness and maternity, and unemployment, both employers and employees contribute to financing (in addition to central government), while employers and government most often fund work injury programs. Family allowance programs are funded either by central government only, or in combination with employer contributions, and only in a few countries, Greece, Malta, Portugal, employees also pay a portion of their income to fund this program. Looking across countries, some conclusions stand out. Croatia and Denmark are the only countries with no programs where both employers and employees contribute, and their social protection spending is funded either by government alone, or by employers, or by employees (depending on program). Estonia, Lithuania, Poland, and Sweden only have one program each that is funded by shared contributions, and the rest are financed as in Croatia and Denmark.

20At the other extreme, Greece and Malta rely on funding from both employers and employees for all programs. These two countries, along with Germany, Luxembourg, Poland, and Switzerland, also stand out with having equal shares of employer – employee contributions. Further, in only a few cases, e.g. old age pensions in Slovenia, sickness and maternity in Austria and Hungary, and unemployment in Estonia, Hungary, and the Netherlands, employers contribute less than employees, as indicated by the ratio of less than zero. In the rest of the countries, employers share of the contributions exceed that of employees, although by varying degree. Lithuania (for old age pensions) and Spain (for old age pensions and sickness and maternity) stand out as countries where the share of employer contributions far exceeds those of employees, e.g. 5 times or more. Czech Republic, France, Finland, Slovakia, and Spain (for unemployment) also have programs with employer contributions three or more times those of employees. For the rest of the countries, where employers pay more than employees, the difference is two times or less.

Eligibility requirements at program level

21Table 3 presents the diversity among social protection programs in Europe in yet another way, focusing on how restrictive the eligibility conditions are, i.e. how difficult it is for someone to qualify for receiving benefits under each of the programs. In addition to generosity (which is not discussed in this paper), eligibility is another good indicator of whether a program acts as a redistributive policy tool. For example, the predominance of programs with several eligibility requirements, many of them tied to employment would be indicative of a societal understanding of benefits as ‘earned’ rather than as ‘rights.’ Capturing differences in levels of restrictiveness is challenging, however, as each social protection program has different sets and number of eligibility criteria.

Table 3

Relative restrictiveness of benefits eligibility, by social protection program, across 30 European countries

Country Old age pensions Sickness and maternity Unemployment Family allowance Austria 75% 25% 25% 33% Belgium 50% 25% 50% 33% Bulgaria 50% 50% 50% 67% Croatia 50% 0% 50% 33% Cyprus 50% 50% 50% 67% Czech Republic 50% 50% 50% 33% Denmark 75% 50% 50% 33% Estonia 75% 0% 75% 67% Finland 75% 50% 50% 67% France 50% 50% 75% 67% Germany 75% 0% 50% 67% Greece 50% 50% 50% 67% Hungary 50% 50% 50% 67% Ireland 50% 75% 75% 67% Italy 75% 25% 75% 67% Latvia 50% 0% 50% 33% Lithuania 50% 50% 50% 67% Luxembourg 50% 0% 25% 33% Malta 75% 50% 75% 67% Norway 75% 50% 50% 33% Poland 50% 25% 75% 67% Portugal 50% 75% 50% 67% Romania 50% 50% 50% 67% Slovak Republic 50% 25% 25% 33% Slovenia 50% 0% 25% 67% Spain 75% 25% 25% 33% Sweden 50% 50% 25% 67% Switzerland 50% 25% 25% 33% The Netherlands 75% 25% 50% 67% United Kingdom 50% 25% 25% 33%

Relative restrictiveness of benefits eligibility, by social protection program, across 30 European countries

22For example, all countries restrict old age pensions by the age of the recipient ; additionally, some countries require a minimum number of years of ‘contributions’ as a working adult, income means testing, or residency. Eligibility criteria for sickness and maternity, and for unemployment benefits include employment, duration of contributions, means testing, and/or residency. Criteria for family allowance eligibility include age, means testing, and/or residency. Work injury compensation is an exception from the other social protection program as—given its nature—it generally does not require minimum years of contribution, means testing, or a waiting period. Only Malta requires a period of contributions before eligibility starts, and Greece and Ireland require that accidents are reported within a certain time frame, and that an independent medical board assesses the injured worker before compensation is paid. Thus, work injury eligibility is excluded from Table 3, and for comparison purposes, the restrictiveness scores for the other programs are calculated as a percentage of all possible eligibility restrictions.

23Table 3 shows that old age pension programs are among the most restrictive of the social protection programs examined here, with all countries having at least two eligibility requirements, and a third of the countries—three such restrictions. Family allowance subsidies, which are mostly managed as universal programs also tend to have a relatively high restriction threshold, with 20 of the 30 countries imposing 2 (out of possible 3) eligibility restrictions. Sickness and maternity benefits, on the other hand, are among the less restrictive with several countries such as Croatia, Estonia, Germany, or Latvia not having any eligibility restrictions for enrolled individuals. Looking across countries, Finland, Italy, Ireland, Malta, and Portugal are among the countries with relatively high number of restrictions across all programs, at least 50% of all possible eligibility criteria.

Conclusions

24This paper provides an overview of five social protection programs across 30 European countries, on two basic dimensions : how is each program funded, and how restrictive are the eligibility requirements. Countries are compared across old age pensions, sickness and maternity, unemployment, work injury compensation, and family allowances programs. Going beyond classic and emerging typologies of welfare regimes, and focusing on aspects of social protection programs that relate to notions of solidarity and inclusiveness, reveals the diversity that still exists both between and within regime types, a diversity that is often difficult to discern when discussing overall regime comparisons, and/or spending levels. In conducting such a comparison, one of the goals of this paper was to try to identify where to the post-communist countries fit in the broader European context. It is obvious from the data presented here that there is hardly a post-communist social welfare ‘model,’ as there are many differences in the ways these countries have arranged their social protection policies, how they fund them, and how they control/oversee benefits eligibility.

25One possible exception from this conclusion is the post-communist countries’ overall lower reliance on government funding for social protection programs, as revealed in Figure 1. A number of Eastern European countries, e.g. the Czech Republic, Estonia, Lithuania, Poland, cluster at the lower end of the spectrum with central government contributing around or less than one third of social protection finances. Yet, the rest of the post-communist countries resemble a continental model (e.g. Croatia, Slovakia), a Southern one (e.g. Hungary), or a Scandinavian one (e.g. Bulgaria, Romania), at least in terms of level of government contribution to social protection. Focusing on the level of individual programs reveals an even greater diversity, as discussed in the paper.

26Overall the paper presents an approach of thinking about social policy in the European context that goes beyond classic typologies, or trying to place each country in a specific predefined category. Instead, by focusing on specific programs, the paper highlights much of the underlying diversity of social protection policies between and within countries, making it obvious that the same country might adopt different approaches of funding and spending, depending on the policy in question. Finally, by focusing on two of the dimensions of each policy, the paper provides a starting point for a discussion about how different societies distribute responsibilities for funding social programs, and how restrictive are their social protection policies. Thus, with respect to the ESM, the data presented in the paper highlights the differences rather than the—often ideological—similarities among countries ; additional historical data are needed to address the question of whether European countries are converging on a common model. The scope of this paper does not allow it to address two important questions. What are the underlying societal, legacy, or institutional reasons for the kind of programmatic diversity revealed by the data presented here, especially in the context of the post-communist countries ? Why do, despite differences in institutional arrangements, countries continue to display similarities in social policy outcomes, such as spending, employment levels, social inequality, etc. ? These and other questions are left for future research.