Elections, Appointments, and Public Goods Provision: Assessing Causal Effect

Do elected or appointed leaders provide more public goods? The incentives and accountability mechanisms at work under democratic selection mechanisms should be better at producing leaders who will provide public goods valued by constituents. However, existing research has found it very difficult to cleanly identify the causal effect of elections and appointments, as they are unable to account for selection effects or omitted confounders. In this paper we employ an original dataset composed of 568 mayors of large Russian cities over the years 1990 to 2012. We take advantage of the staggered introduction and cancellation of mayoral elections in these cities over time with a differences-in-differences research design. This allows us to better isolate the causal effect of elections on public goods provision. This dataset also permits us to fully account for personal characteristics of the mayors and city characteristics with a broad set of dependent variables measuring public goods provision.