Association for Public Policy Analysis & Management

Urban school districts across the United States have increasingly relied on closing public schools as an education reform strategy to address declining student enrollment, fiscal constraints, poorly maintained school infrastructure, and low academic performance. Federal policies, including more recent school turnaround initiatives, have prompted many of the nation’s largest urban school districts to endorse school closings as a means of offering students better educational options. Between 2000 and 2010, no fewer than 70 urban school districts closed an average of 11 traditional public schools (Engberg et al., 2012). Closing schools as a matter of policy is controversial, as critics of school closures point to the disproportionate impact this has on economically disadvantaged minority students and the adverse influence student mobility has on academic performance (de la Torre and Gwynne 2009). Families also voice concern that closing schools will cause crime to go up in neighborhoods as schools represent neighborhood anchors and their abandonment may lead to additional blight and crime. There is limited evidence on the neighborhood impact of closing schools on crime (Brinig and Garnett 2010).

We address this dearth of evidence by examining the impact of major rounds of school closures on neighborhood crime. We capitalize on two major rounds of school closures that occurred in Philadelphia in the 2011-12 and 2012-13 school years. During these two years the school district closed nearly 30 schools and displaced nearly 10 percent of its traditional public school students. Relying on a unique dataset that matches monthly reports of all crimes in a census block to a school’s location, we employ a difference-in-difference strategy to estimate the impact of school closures on neighborhood crime. Specifically, we compare the changes in monthly crime patterns in blocks where schools close to those where schools remain open or where schools were never located. We also examine how the estimates vary by grade level of schools closed, as it is likely that closing elementary schools only removes teachers and creates vacant buildings and has no impact on displacing criminal offenders. Initial evidence suggests that school closures have no effect on total reported crime, but closures do lead to reductions in assaults, disorder, and drug-related crimes. The weight of the evidence suggests that closing under-performing schools helped reduce serious crime in neighborhoods where closed schools were formerly located.