Evaluating the Effects of Early Release from Prison: A Natural Experiment

• In Event: The Effects of Imprisonment on Recidivism

Abstract

In 2015, the Washington State Department of Corrections (DOC) discovered a computer coding error in its data system resulting in the unintended early release of 3,200 inmates over a 13-year period. This unfortunate mistake created a natural experiment, offering researchers unexpected and ideal conditions to study the effects of early release on recidivism. Very little direct empirical evidence exists, to date, on the broad policy implications of sentencing, deterrence, and the returns of incarceration. Utilizing a comparison group of inmates, we examine the impact of length of stay in prison on public with regard to the commission of new criminal behavior, violence during the early release period, as well as the timing and prevalence of crime among released offenders. Although the state saved an estimated $200 million through the reduced need for prison beds, we explore whether these savings outweigh the costs of crimes incurred through reduced incapacitation. Findings from this study will help guide potential sentencing reform as well as identify potential areas of efficiency and cost savings within the system.

Authors

• Zachary Hamilton, Washington State University

• Elizabeth K. Drake, Washington State Institute for Public Policy