We used a retrospective quasi-experimental design to assess the impact of prison labor on institutional misconduct, post-prison employment, and recidivism among 6,144 offenders released from Minnesota prisons between 2007 and 2011. In addition to using multiple measures of prison labor, we relied on propensity score matching to minimize observable selection bias. Participation in prison labor significantly improved post-prison employment outcomes, but it yielded mixed results for prison misconduct and had little overall impact on recidivism. The best outcomes were observed for those who spent a greater proportion of their overall confinement time working a job in prison. As the percentage of prison time spent working increased, we found significant improvements in prison misconduct, post-prison employment, and several measures of recidivism.