The 1954 decision of the Supreme Court in Brown v. Board of Education launched one of the most important social policy changes of the 20th century. Most of the research on school desegregation has focused on student academic outcomes. Yet impacts on non-academic outcomes may be at least as important for understanding social welfare or distributional consequences. Crime in particular is of central interest given the enormous costs to society, perhaps as much as $2 trillion per year. Ignoring this outcome may substantially distort benefit-cost analyses. This is the first project to examine school desegregation effects on crime using a plausibly exogenous source of identifying variation.
This project seeks to estimate the causal effects of court-ordered school desegregation on crime. Theory yields ambiguous predictions about the expected effect; desegregation orders could expose minority children to more developmentally productive schools or peer groups, and some desegregation orders required school districts to increase educational spending, which together could reduce crime by black and white youth by improving cognitive and non-cognitive skills and legal earnings prospects. On the other hand court desegregation orders could also exacerbate racial tensions. To examine this question the project exploits the fact that since the Supreme Court transformative 1954 Brown decision, the majority of the largest school districts were subject to mandatory, court-ordered desegregation plans. The timing of when these plans went into effect is idiosyncratic and plausibly exogenous to other determinants of youth outcomes. Support for this claim comes from preliminary analyses of annual county-level homicide victimization data from the Vital Statistics, which reveal no pre-existing trends in homicide before these court orders went into effect. However preliminary analyses suggest that the implementation of these orders reduce homicide victimization by around 25 percent for both black and white youth. Funding is requested to: (1) Examine school desegregation effects on homicide offending as well as victimization. While information about offenders is not available for crimes without an arrest or at least identified suspect, these data still may help us understand behavioral responses and distributional effects. (2) Examine impacts on other types of crimes besides homicide. Since these offenses are far more common, analyses of other crimes provide assurance that the homicide results are not driven by small numbers of cases. Other crimes taken together also impose large social costs. (3) Learn more about the behavioral mechanisms through which school desegregation influences crime. For example the preliminary analyses finds desegregation effects on homicide victimization rates are about as large during the summer months as during the school year, and impacts on homicide offending behavior may persist well into adulthood. This project also examines how impacts vary by the design features of the local school desegregation plan.
Broader Impacts: Since the 1990s, federal courts have terminated many local desegregation plans and in two major cases the Supreme Court recently struck down plans in Seattle and Louisville. The findings on desegregation effects on crime may help inform the litigation and legislative debates that are sure to follow, and may help stimulate new research on how school desegregation impacts a wide range of other non-academic outcomes. The results may also shed light on the broader question about the degree to which social policies can provide cost-effective alternatives to mass incarceration for crime control. The project will produce a scientific paper targeted at a top-tier economics journal, a public-use longitudinal dataset on crime and arrests going back to 1960, and an accessible research brief.
