Econometrics Seminar - Giulio Zanella (University of Adelaide)

Title: Prison work and recidivism

Abstract: Does employing inmates in prison jobs reduce post-release reincarceration? This question is investigated applying reduced-form and structural methods to administrative data on the universe of convicts released from Italian prisons between 2009 and 2012, their work records in prison jobs, and their criminal records for three years following the release. The relatively high wage rate in prison jobs in Italy induces a form of rationing that offers a lever for identification. I find that a standard deviation increase in average annual hours spent in a prison job (about 200 hours per year) reduces the reincarceration rate by about 10 percentage points one year of release due to a combination of liquidity, rehabilitation, and training effects. This effect persists after three years of release. The implied rate of return on government funds allocated to prison work programs in Italy is almost 40%.