Impact of strengthening non-custodial sanctions: evidence from the reform to alternative sanctions in Chile

Abstract

Objectives This article assesses the impact of the new regime of alternative sanctions implemented in Chile between the years 2012 and 2015 on recidivism. It aims to fill the evidence gap on the effects of structural changes in the non-custodial sanctions’ systems. Methods To identify the causal effect of the new alternative sanctions’ regime on recidivism, we implement a robust quasi-experimental design that combines matching and differences in differences using repeated cross-sectional data. The total sample contains more than 100 thousand observations. This methodology allows controlling for both individual observable characteristics (age, gender, sentence length, type of crime, previous sanctions and specific type of non-custodial sanction) and temporal non-observables. Results The results suggest a statistically non-significant reduction of -0.6% on the recidivism rate attributable to the alternative sanctions’ reform. Although, the comparison of recidivism before and after showed a statistically significant reduction on recidivism (- 2.8%), it was not statistically greater than the recidivism expected time-trend (-2.2%). Conclusions The results suggest that the reform was not able to reduce criminal recidivism. As a policy recommendation, we detect an opportunity to improve the reform performance by reorienting treatment resources over alternative sanctions according to the offenders’ risk of offending. Additionally, the methodology presented opens the opportunity to study criminal justice reforms using a robust quasi-experimental design in the context of more accessible cross-sectional administrative data. Keywords Non-custodial sanctions, Impact assessment, Recidivism, Matching, Difference in Difference