Examining Prison Effects on Recidivism: A Regression Discontinuity Approach

Abstract

The “get-tough” era of punishment led to exponential growth in the rate of incarceration in the United States. Recent reviews of the literature indicate, however, that limited rigorous research exists examining the effect of imprisonment on the likelihood of future offending. As a result, scholars have called for assessment of this relationship, while using methodologies that can better account for selection effects. This study addresses these calls directly by applying regression discontinuity, a methodology well suited to account for selection bias, on a cohort of felony offenders in Florida. Results suggest that prison, as compared to non-incarcerative sanctions, has no appreciable impact on recidivism. Although no differential effects surfaced across race/ethnicity, the analyses indicated that imprisonment exerts a differential effect by gender with the effect being more criminogenic among males than females.