How should inmates be released from prison? An assessment of parole versus fixed-sentence regimes *

Abstract Over the past 30 years, many states have abolished parole boards, which traditionally have had the discretion to release inmates before the expiration of their full sentence, in favor of fixed-sentence regimes in which the original sentence is binding. However, if prison time lowers recidivism risk and if parole boards can accurately estimate inmates’ recidivism risk, then relative to a fixed-sentence regime, parole can provide allocative-efficiency benefits (costly prison space is allocated to the highest-risk offenders) and incentive benefits (prisoners know they must reduce their recidivism risk to gain an early release, so invest in their own rehabilitation). Exploiting quasi-experiments from the state of Georgia, I show that prison time reduces recidivism risk and that parole boards set prison time in an allocatively efficient manner. Prisoners respond to these incentives; after a reform that eliminated parole for certain offenders, they accumulated a greater number of disciplinary infractions, completed fewer prison rehabilitative programs, and recidivated at higher rates than inmates unaffected by the reform. I estimate that eliminating parole for all prisoners would increase the prison population by 10% while also increasing the crime rate through deleterious effects on recidivism.