The Effects of School Desegregation on Crime

In this paper we provide evidence suggesting that court-ordered school desegregation generates large reductions in homicide offending and victimization for blacks, and reductions in victimizations for whites as well. Our research design exploits the fact that from the late 1960s through early 1980s, most of the nation's largest school districts were forced to desegregate by local Federal court order. Within the set of districts ever subject to court order, the timing of the orders is plausibly exogenous to trends in other determinants of youth outcomes. For blacks, we find that homicide victimization rates decline by around 25 percent and homicide arrest rates decline by about 15 percent in the long-run, effects that are caused in part by improved schooling outcomes for blacks. We find white homicide victimizations also decline, due perhaps in part to a decline in the rate at which blacks kill whites and an increase in school and police spending in response to these school desegregation orders. Whether there is a behavioral response by whites to commit fewer homicides is less clear in our data.