This policy brief discusses how voters respond to public policies with a particular focus on crime control policies. We present the main methodological issues that the researcher faces to disentangle the response of voters to the consequences of policies from the other factors. We present the results of a study of the consequences of the 2006 Italian collective pardon. Using randomness in local variation in recidivism rates we show that Italian voters held the incumbent government responsible for local variations in recidivism caused by the collective pardon. We conclude by discussing the consequences for the evaluation of the electoral response to public policies.