By Michael LaForest

Nearly three-fourths of incarcerated individuals are released under parole supervision in the United States. However, relatively little is known about the effects of supervised release. In this work, I first investigate the effects of early release from prison using the quasi-random assignment of interviewers to parole hearings in Pennsylvania. I find that, at the margin of release, individuals initially paroled experience higher rates of post-release recidivism than individuals released at a later date. Second, I separately identify the effects of the three major components of parole supervision – (1) supervision intensity, (2) special conditions such as curfew or placement in a halfway house, and (3) the assigned parole officer who manages supervision – by leveraging three separate quasi-random assignment mechanisms in Pennsylvania. Along most margins, I find that increased supervision leads to additional parole violations with little effect on future arrests or employment.

Draft Paper, 2022. 41p.