Generalized Trust, Preferences for Redistribution and Institutions

Abstract: We study how institutional quality moderates the relationship between generalized trust and preferences for redistribution. It has been well established in the literature that generalized trust is conducive to greater support for redistribution because it reduces expectations of cheating and free-riding among others. Following Algan et al. (2016), we hypothesize that the effect of trust on preferences for redistribution is conditional on the quality of the institutional environment. Trusting individuals, that is, are hypothesized to be more supportive of redistribution in favor of target groups with a high propensity of free-riding (like the poor and the unemployed) if the institutional environment is more likely to detect and penalize free-riders. We test this hypothesis with data from the Life in Transition II survey that contains 38,000 respondents from 35 transition and developed countries. We find that the effect of individual trust on supporting redistribution in favor of the poor and the unemployed depends indeed on the quality of the formal institutions, like the country-level control of corruption and the rule of law. Trust and formal institutions are therefore complements with respect to their effect on preferences for redistribution. This relationship is not observed for groups conventionally thought of as unambiguously deserving or delineated with free-riding proof eligibility criteria, i.e. for the disabled and families with children. Less trusting people are also willing to contribute as long as they observe a high level of neediness in their direct environment, implying that the observed level of neediness and the level of trust are substitutes in the decision to contribute.