The Political Cost of Being Soft on Crime: Evidence from a Natural Experiment

We study voters’ response to criminal justice policies by exploiting a natural experiment. The Italian collective pardon, promoted and implemented by the national government in 2006, unexpectedly released about one third of the prison population. The collective pardon law created idiosyncratic incentives to recidivate across pardoned individuals. Municipalities where resident pardoned individuals had a higher incentive to recidivate experienced a higher recidivism rate. We show that in those municipalities voters “punished” the incumbent national government in the 2008 parliamentary elections. In particular, our estimates suggest that - in an average municipality – an additional crime by pardoned individuals led to a drop of 272 votes (1.77% of eligible voters) for the incumbent national government relative to the opposition coalition. We also provide evidence of newspapers being more likely to report crime news involving pardoned individuals and of voters hardening their views on the incumbent national government's ability to control crime. Our findings indicate that voters keep politicians accountable by conditioning their vote on the perceived effects of public policies.