UC Irvine

This study utilized a unique dataset that combines information on parolees in the city of Sacramento, CA over the 2003-06 time-period with information on monthly crime rates in Sacramento census tracts over this same period, providing us a fine-grained temporal and geographical view of the relationship between the change in parolees in a census tract and the change in the crime rate. We find that an increase in the number of tract parolees in a month results in an increase in the crime rate. We find that more violent parolees have a particularly strong effect on murder and burglary rates. We find that the social capital of the neighborhood can moderate the effect of parolees on crime rates: neighborhoods with greater residential stability dampen the effect of parolees on robbery rates, whereas neighborhoods with greater numbers of voluntary organizations dampen the effect of parolees on burglary and aggravated assault rates. Furthermore, this protective effect of voluntary organizations appears strongest for those organizations that provide services for youth. We show that the effect of single parent households in a neighborhood is moderated by the return of parolees, suggesting that these re-united families may increase the social control ability of the neighborhood.