Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
How harshly should society punish young lawbreakers? Through a fuzzy regression discontinuity design, I compare harsh and rehabilitative criminal incarceration practices. Young offenders sent to the more rehabilitative youth facilities become less likely to reoffend. On the contrary, offenders at the margin of the age cutoff and exposed to the harsher youth facilities are 27 percent more likely to recidivate in the eight years subsequent to their custody, and they are more likely to commit violent offenses, thefts, and criminal damages. Keeping young offenders separate from their older peers in prison seems effective, but only in institutions not solely focused on punishment.