Electoral Incentives and Economic Policy across Political Regimes

This paper provides a direct test of the causal link from electoral rules to economic policy. Our theoretical model delivers unambigous predictions on the interaction between institutions and a time varying event, namely the unemployment rate in pivotal and non-pivotal districts. We use local level data on unemployment rate and political competition to obtain an empirical specification which matches our model. First, we test the effect of electoral incentives under majority rule, by analyzing the US House representatives voting records on the 2009 Emergency Unemployment Compensation Extension Act, which increased unemployment benefit coverage and generosity. Second, we exploit the time-varying dimension of our theoretical prediction to test the causal effect on panel data. We use a dataset with local information on electoral competitiveness and unemployment rates for 29 OECD countries in 1980-2001 and employ panel analysis on different measures of UB generosity. The empirical evidence strongly supports our theoretical predictions.