In this paper, we evaluate the impact of unilateral divorce on crime. First, using crime rates from the FBI's Uniform Crime Report program for the period 1965-1998 and differences in the timing in the introduction of the reform, we find that unilateral divorce has a positive impact on violent crime rates, with an 8% to 12% average increase for the period under consideration. Second, arrest data not only confirms the findings of a positive impact on violent crime but also shows that this impact is concentrated among those age groups (15 to 24) that are more likely to engage in these type of offenses. Specifically, for the age group 15-19, we observe an average impact over the period under analysis of 40% and 36% for murder and aggravated assault arrest rates, respectively. Disaggregating total arrest rates by race, we find that the effects are driven by the Black sub-sample. Third, using the age at the time of the divorce law reform as a second source of variation to analyze age-specific arrest rates we confirm the positive impact on the different types of violent crime as well as a positive impact for property crime rates, controlling for all confounding factors that may operate at the state-year, state age or age-year level. The results for murder arrests and for homicide rates (Supplemental Homicide Report) for the 15-24 age groups are robust with respect to specifications and specifically those that include year-state and year-age dummies. The magnitude goes from 15% to 40% depending on the specification and the age at the time of the reform.
