XIII Congreso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Buenos Aires, Argentina, 4 - 7 nov. 2008 Three welfare regimes, three critical paths to strengthen redistribution Juliana Martínez Franzoni 1. Introduction Social policy may promote or deter collective action as well as informal networks, organizations and parties that support democratic processes and place distributive matters in the political (rather than merely technical) arena (Filgueira, 1998). According to Pearson, far more than in the era of welfare state expansion, current struggles over social policy have become struggles over information about the causes and consequences of policy change, that is struggles over policy strategies. As I show below, the effects of social policy upon distribution differs across welfare regimes as does the role of distributional coalitions behind these policies. Esping-Andersen identified three distinct patterns of state intervention in the advanced capitalist countries of the North: social democratic, liberal and conservative. In each case, the state intervened through social and (to a lesser extent) labor-market policies to reduce inequality (both between individuals and across the life-cycle), but the form of that intervention differed in terms of the scale of public expenditure and the extent to which the state displaced the market and family in determining the incomes and welfare of its citizens. At the end of the twentieth century, these different clusters of states responded differently to fiscal, demographic and political pressures (Esping-Andersen 1999, 2002). “A welfare regime can be defined as the combined, interdependent way in which welfare is produced and allocated between state, market and family” (Esping-Andersen 1999, 34-5). Across welfare regimes Korpi and Palme (1998) showed a close relationship between the size of the budget for redistribution (i.e. the public welfare budget) and the extent of income redistribution through transfers and direct taxation (measured through the reduction in the Gini coefficient). The social democratic welfare regimes tended to spend more and redistribute more than the conservative welfare regimes of continental Europe, which in turn tended to spend and redistribute more than the liberal welfare regimes (see also Huber and Stephens 2001, Milanovic 1999, Przeworski and Gandhi 1999, Bradley et al. 2003 in Seekings, 2008a). Redistribution reflected class alliances leading to the central role of left-wing parties in social- democracies, of social-Christian parties in conservative welfare regimes, and of liberal parties in liberal welfare regimes. According to the power resource approach, class coalitions are a necessary condition to promote redistribution as well as policy change. Such coalitions are fairly different across welfare regimes (Esping-Andersen, 1990). The interactions and links between social policy and class formation increase the complexity of the analysis. Class formation shapes social policies and these, in turn, have re- stratification effects other than merely de-stratification ones: public policies have enduring histories. Although neither rigid nor immutable, conditions behind distinctive welfare regimes and the effects these regimes produce influenced the ways in which they later evolved (Seekings, 2008a). Policy feedback stresses the need to “incorporate historical analysis into the study of contemporary politics... today´s policymakers must operate in an environment fundamentally conditioned by policies inherited from the past” (Pierson, 2000:9). Historical trajectories are also central to the Latin American context. Prior to the economic crisis of the 1980s and the implementation of the import substitution model, there were countries that - although stratified - were relatively universal social States (such as Argentina, Chile, Costa Rica and Uruguay); and others that were a combination of stratified universalism in urban areas and exclusion among rural sectors (such as Brazil and Mexico); and others with exclusive States (Filgueira, 1998). During the past 1