"A recurrent critique on the welfare state is that it crowds out social capital (networks, trust and norms). However, the empirical evidence on the crowding out hypothesis is still scarce, findings are sometimes contradictory, and there is variation in the measurement of social capital. In this article we explore the crowding out hypothesis on the basis of data from the European Values Survey wave 1999/2000 for 23 European countries. Compared to (the few) other comparative studies on the hypothesis, our study contains more recent data and for a larger number of countries. Instead of focusing on a single dimension of social capital, we use an eight-scale measurement model of social capital; we explore the relationship between welfare (regime type, social spending) and social capital at both country and individual level, and we control for confounding factors. At the aggregate country level we found no evidence at all in favour of the hypothesis. At the individual level we found that it does matter for people's social capital in which type and size of welfare state they live. However, there is only evidence for a crowding out effect in case of people's trustworthiness. With regard to other forms of social capital, there is at best mixed evidence, but mostly our findings contradict the crowding out hypothesis altogether."

Support: Digital