Social Protection in the Member States

reached by Genschel ( 2004 ), namely, that various institutional solutions can prove economically successful.

on social protection and healthcare is very similar to the fi rst cluster; that is, it can be described as high. Income disparities are moderate, however, making the poverty risk both before and after transfers similarly around average. Th e ratio of pensions to GDP is high. Family and child benefi ts comprise only a small portion of spending on social protection, while provisions for the aged are high. Th e government makes a smaller con-tribution to the fi nancing of social protection expenditures, while the contributions of employers and benefi ciaries are higher than in the fi rst or second cluster.

Th e three Baltic states, the Czech Republic, Slovakia, Romania and Bulgaria, make up the group in which social and healthcare spending is the lowest. With the exception of the Czech Republic and Slovakia, disparities in income distribution are the largest here, while the poverty risk is slightly below average before social transfers and above average afterwards. Although the ratio of pensions to GDP is low, the share of family and child support in social expenditures is moderate, and that of provisions for the aged is high. Th e contribution of employers to spend-ing on social protection is conspicuously high, while the government and benefi ciaries contribute relatively little (Table 3.7 ).

In Fig. 3.5 , the horizontal axis indicates the level of development of social protection. Th e vertical axis permits us to gauge whether old-age and pension provisions or child and family benefi ts dominate the system of social protection.

On the topic of social protection and the welfare state, the VoC lit-erature is intertwined with the work of sociologists and social policy-makers. In the wake of Titmuss’s pioneering works on social policy, Esping- Andersen’s book, Th e Th ree Worlds of Welfare Capitalism ( 1990 ), proved to be a milestone. His formulation of three types of regimes—lib-eral, conservative- corporatist, and social democratic—prompted a long succession of authors who refi ned or refuted these models or expanded their number. Th e most frequent alteration was to group the Latin or Mediterranean countries separately (Kleinman 2002 ). Currently, the divi-sion into four models is accepted, and debate tends to focus more closely on the extent to which social policy can be integrated at the European level and on which model is sustainable. Caminada et al. ( 2010 ) identify a

decade-long convergence among the OMS from the mid-1980s onwards, followed by a divergence thereafter. Pöder and Kerem ( 2011 ), as well as Leibrecht et al. ( 2011 ), place the countries of CEE in a separate group alongside the aforementioned fourfold division. However, based on their empirical investigations, the former authors fi nd a convergence between the Mediterranean and continental (conservative) welfare regimes, while the latter authors determine a similar convergence between the Nordic (social democratic) and conservative welfare regimes.

Table 3.7 Social protection clusters Clusters

1. High level of welfare spending Denmark, Finland, Luxembourg, Sweden Low level of income disparities

Within welfare spending, high proportion of family and child benefi ts

High government contribution to the fi nancing of social protection expenditures

2. Low level of welfare spending Ireland High level of poverty risk, with moderate

income disparities

High government contribution to the fi nancing of social protection expenditures

3. High level of welfare spending Austria, Belgium, UK, France, Greece, Netherlands, Poland, Hungary, Germany, Italy, Portugal, Spain, Slovenia

Moderate income disparities

Within welfare spending, a low proportion of family and child benefi ts

High ratio of pension expenditures to GDP Low government contribution to the fi nancing

of social protection expenditures, high contribution by employers and benefi ciaries

4. Low level of welfare spending Bulgaria, Czech Republic, Estonia, Latvia,

Lithuania, Romania, Slovakia

High level of income disparities Within welfare spending, a moderate

proportion of family and child benefi ts Low ratio of pension expenditures to GDP High contribution by employers to fi nancing of

social protection expenditures, low

contribution by government and benefi ciaries

Within the thematic area of social protection, not only interdisci-plinary overlaps but also shared content were found because the topic is closely connected to the labour market. I examine labour market poli-cies and their expenditures separately and, following Eurostat’s statis-tical system, list only direct social allocations among social protection expenditures. In the study by Aronja et al. ( 2001 )—in accordance with OECD data collection—expenditures on active labour market policy are

Fig. 3.5 Two-dimensional MDS-based representation of social protection

also included in the latter. Sapir ( 2006 ) formulates his social models in two dimensions: welfare redistribution and labour market policy. Th e overlap is unavoidable because the state’s labour market policy and regu-lation determine the framework for operation of the labour market as a partial market, and they also constitute an important element of social protection.

Th e picture emerging from this cluster analysis roughly corresponds to what is found on welfare regimes in the literature in the wake of Esping- Andersen. Esping-Andersen himself ( 1990 ) explains the variety within welfare regimes not on the basis of diff ering expenditures in individual states, but rather on the basis of the type of institutional framework through which welfare provisions can be accessed, an approach later rein-forced by experience. ¹¹ Indicators used in the cluster analysis, meanwhile, clearly refl ect the institutional system that Esping-Andersen attributes in his 2002 study to the Nordic, liberal, and continental models. In his view, the distinctive feature of the Nordic model is that it provides universal income guarantees and well-developed services to children, the disabled, and the elderly in need of support, while its activation policy reduces long-term unemployment.

Th e cluster of Nordic countries in my analysis is clearly separate from that of the continental countries. Th e level of expenditures is somewhat higher than the average in the continental countries, although—thanks to the reforms of the 1990s—the quantitative diff erence is no longer conspicuously great. Structural diff erences are more noticeable, indicat-ing varyindicat-ing institutional systems. Th e proportion of old-age provisions is great in the continental countries, conforming to the traditional, status- oriented model that makes it possible to maintain social status through-out the various stages of the life cycle. By contrast, in the Scandinavian countries, family and child support is signifi cant, which—following the universal model—aims to level out income across the family’s life cycle in the spirit of an egalitarian ethos. A benefi cial eff ect of this result is that child poverty is negligible. Pension and healthcare expenditures are lower than the EU-15 average. Kiander ( 2004 ) explains that healthcare is cheaper than in the other EU member states because it is provided in public institutions, as opposed to the more expensive combination of social insurance and private care. Th e effi cient operation of welfare

insti-tutions and strong competition within the economy can be attributed to the fact that there is a high poverty risk after market income distribution, while the risk will be the lowest among the four clusters after welfare transfers.

Th e classifi cation of the UK and Ireland among welfare regimes generally elicits uncertainty in the literature. Esping-Andersen ( 1990 ) labels the English-speaking countries as liberal, residual welfare regimes, where the ideal type is the USA. From his book, it also emerges that the UK does not entirely fi t this model because healthcare is a universal ser-vice fi nanced from taxes, just like education. Local governments hold a large supply of rented social housing. In 1990, the author classifi ed the welfare regime in Ireland as belonging to the corporatist-conservative model of the continental countries. A decade later, he assigned both countries—albeit in inverted commas—to the “liberal” welfare model, which, similar to the USA, supports market solutions and limits pub-lic accountability to acute market failures. Th e role of means testing has strengthened, and in parallel, the emphasis has shifted from the traditional assessment of needs to work-conditional welfare provision (Esping-Andersen 2002 ).

In the cluster analysis, the UK appears as a borderline case among the continental countries. Within the structure of expenditures, the diff er-ence compared to the continental countries is apparent not so much in child and family support, but rather in strikingly high housing benefi ts.

Th e poverty risk and income disparities are high. In the MDS diagram, this ambiguous situation is clearly expressed in the UK’s relative sepa-ration from the other countries of the third cluster. At the same time, Kleinman ( 2002 ) points out that even in the Th atcher era, the welfare regime did not become “truly” liberal; the changes were more radi-cal in rhetoric than in reality. Pierson ( 1996 ) attributes this result to resistance in society and political unpopularity forcing governments to backtrack.

In this study, Ireland tends to display the characteristics of an Anglo- Saxon regime like the UK itself. Callan et al. ( 2008 ) likewise categorise their own homeland under the Anglo-Saxon model, given that the insti-tutional system resembles that of the UK. Th ey nevertheless note that spending on social protection is not as low as the GDP-proportionate fi

g-ure would suggest because, in Ireland’s case, there is a large gap between GDP and GNI (in 2004, the latter was just 85 per cent of the former), and GNI is a more realistic basis for comparison. However, even after making this correction, social expenditures are still considerably lower than the average in the OMS.

Esping-Andersen ( 2002 ) includes the Mediterranean countries in the continental welfare model, in the majority of which the traditional responsibility of family in the welfare regime has remained, especially in Southern Europe, but to the least extent in France and Belgium. Coupled with this is the dominance of social insurance, which provides good pro-tection for those with a stable job throughout their working lives but which is unable to adequately manage the risks that accompany the spread of atypical employment.

Th e Mediterranean countries do not stand out among the continen-tal countries in this study (Table 3.7 ) or in the European Commission report on industrial relations as it relates to welfare regimes (Table 3.6 ).

Studies arguing in favour of an independent welfare regime cite charac-teristics such as the importance of family in social protection, the ineffi -cient operation of welfare institutions, their clientelist and particularist nature, and the strength of party-political infl uence (Kleinman 2002 ).

Th ese characteristics naturally cannot be expressed by the indicators used in this analysis. In the past decade, the Mediterranean countries have strived to improve the effi ciency and sustainability of their wel-fare regimes through a series of reforms (Guillén 2007 ; Sacchi 2007 ; Sakellaropoulos 2007 ).

Th e European Commission report on industrial relations (European Commission 2009c ) questions whether the new, post-socialist countries belong to the Anglo-Saxon residual model or to the segmented continen-tal model. Th is cluster analysis acknowledges that these member states could not be grouped under a single welfare regime (Table 3.7 ). Poland, Hungary, and Slovenia belong among the continental countries. As can be seen in relation to the transformation in CEE, Poland and Hungary principally owe this “illustrious” position to the fact that they handled the joblessness that came with structural transformation by pensioning people off . Th e other post-socialist countries, by limiting the role of the state, display more characteristics of a residual regime, although they are

also set apart from this regime by the tradition of continental social insur-ance, where the contribution of employers to fi nancing is high.