Circles of Solidarity: The National Contexts of Diversity and Redistribution in Developed Democracies

Abstract

In recent years, as many Western democracies have been facing growing racial, ethnic, and religious diversification and tensions, the social policy implications of these cleavages have come to the fore. This dissertation explores how ascriptive diversity—heterogeneity in ethnicity, race, religion, or language—shapes income redistribution amongst and between different identity groups in developed democracies. The common consensus among social scientists asserts that diverse countries are less solidary and thus maintain smaller welfare states than homogeneous ones. However, the data reveal unexplained variation in the aggregate redistribution levels of similarly diverse democracies. Furthermore, additional research, which focuses on individual-level behavior, finds that intergroup relations depend on the contextual environment, particularly socioeconomic inequality and geographic segregation between different identity groups.

Addressing this gap, I propose a new theoretical framework that explains how complex social structures, particularly inequality and geography, shape income redistribution between ascriptive identity groups in democratic societies. I argue that greater ascriptive diversity suppresses redistribution only and insofar as it is reinforced by either higher intergroup inequality or higher regional segregation. Each of these factors, however, fractures national solidarity along different social categories and, subsequently, creates different redistributive preferences and outcomes. When identity groups are highly unequal socioeconomically, better-off identity groups can minimize intergroup redistribution through regressive and exclusionary welfare policies that underprioritize the needs of the poor. By contrast, when identity groups cluster in different parts of the country, their members can minimize intergroup redistribution by decentralizing national welfare programs to the regional level. Hence, these factors determine both the extent to which ascriptive diversity dampens redistribution and whether it would result in more regressive or more decentralized policies.

My empirical analysis corroborates these expectations using a multi-method approach. First, I analyze cross-sectional time-series data from 22 democracies over a period of 31 years and find broad empirical patterns that fit my expectations. All else equal, democracies with higher ascriptive diversity invest less in redistributive programs when their various identity groups are more unequal and/or more regionally segregated. The conditional effect of class and geography, moreover, correlates with different sets of outcomes as predicted by my theoretical framework. Intergroup inequality suppresses investment in programs that target lower-class needs and deems cross-class programs more exclusionary. Intergroup regional segregation, by contrast, implicates all types of national programs and correlates with higher redistributive decentralization. Second, I analyze three representative case studies in greater depth: The United States, Belgium, and Spain, each illustrates ascriptive diversity under a different configuration of class and geography. The three cases provide additional support for my hypotheses, place the different theoretical components on a single coherent continuum, and contextualize them in richer and noisier settings. Moreover, they add important insights about the underlying mechanisms that link preferences and policies and about the implications of long-term changes.

My dissertation, therefore, improves our understanding of the complex relationship between ascriptive diversity and social policy. In doing so, it also ties together separate disciplinary discussions on these issues and contributes new theoretical and empirical insights on the roots of existing policies and on future developments in growingly diverse democracies.