Abstract
This article focuses on the notion that the policies and politics of states and nations constitute distinct worlds or clusters. We begin by examining the concept of clustering as it has emerged in the literature on policy regimes and families of nations. We then address a series of empirical questions: whether distinct worlds persist in an era of policy convergence and globalisation, whether policy antecedents cluster in the same ways as policy outcomes and whether the enlargement of the EU has led to an increase in the number of worlds constituting the wider European polity. Our main conclusions are that country clustering is, if anything, more pronounced than in the past, that it is, in large part, structurally determined and that the EU now contains a quite distinct post-Communist family of nations.
Notes
1. An instance is the article by Castles (Citation1978), appearing in the first number of West European Politics, which sought to identify the political antecedents of the Scandinavian countries' outstanding welfare state performance.
2. See, for an explicitly Rokkanian analysis of contemporary European welfare state development, Ferrera (Citation2005).
3. For instance, Castles (Citation1993) identifies commonalities amongst the English-speaking nations in respect of poor economic performance, low welfare spending and high divorce rates justifying an earlier attribution of the 'awfulness of the English(-speaking nations)' in Castles and Merrill (Citation1989).
4. Essentially, Esping-Andersen's three worlds of welfare plus Southern Europe. The debate on whether there is a distinctive Southern European welfare state type has been an important theme in the comparative social policy literature since the early 1990s (see Leibfried Citation1993; Esping-Andersen Citation1993; Castles Citation1995; Ferrera Citation1996).
5. See, for instance, Stockard and O'Brien's (Citation2006) distinction between ‘English-speaking’, ‘Romanist’, ‘Germanic’ and ‘Nordic’ families.
6. Outcomes are grouped according to whether they are within one standard deviation of the mean, more than one standard deviation above or more than one below.
7. Exceptions are studies like those of Gough (Citation2001) and Powell and Barrientos (Citation2004), which use cluster analysis techniques to test for coherence. The latter study of types of ‘welfare mix’ focuses on different dimensions and measures of welfare from those featuring in the three worlds typology, but derives a very similar regime clustering and one which is shown to be persistent over time.
8. Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Japan, Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, United Kingdom and United States.
9. Note that direct democracy is not included in the set of variables underlying the analysis of regime origins.