Child's Gender, Young Fathers' Crime, and Spillover Effects in Criminal Behavior

This paper studies whether an exogenous reduction in the criminal activity of one individual lowers crimes committed by other young men who live in the immediate neighborhood. Using the randomness of a child's gender, we first show that men who father their first child at a very young age are convicted of significantly fewer crimes in the first years after the birth if the child is a son rather than a daughter. We next show that this leads to behavioral spillovers that significantly reduce criminal convictions among other young men living in the same neighborhood as the father at the child's birth, as well as victimization rates, for at least five years after birth. Evaluating our estimates within a structural model shows that spillovers in crime generate crime multipliers that continue to increase even after the primary impact of the initial shock on the focal individual has dissipated. From the model we further illustrate that crime prevention policies that target high crime individuals at an early stage of their lives are likely to lead to far larger reductions in the cost of crime than suggested by the primary effects alone.