Association for Public Policy Analysis & Management

Policymakers have long been preoccupied with the degree of student segregation across schools. As the busing and desegregation orders meant to counter residential sorting patterns have become less prevalent in recent years, school segregation has been on the rise and addressing it has increasingly fallen under the purview of elected local school boards, principally through the drawing of attendance zone boundaries. Yet, despite the documented importance of schools and peers to educational outcomes, there exists little evidence (causal or otherwise) about the role of school boards in the allocation of students to schools. In this paper, we examine the causal effect of their decisions on student segregation. The key identication challenge is that the composition of a school board is potentially correlated with household sorting that occurs within district. We overcome this issue using a regression discontinuity design at the electoral contest level, exploiting quasi-random variation from narrowly-decided elections. Such an approach is made possible by exploiting a unique dataset, which combines matched information about North Carolina school board candidates, including vote shares and political affiliation (obtained from voter registration records), with time-varying district-level racial and economic segregation outcomes (calculated using each student's residential location and school attended). Focusing on the political composition of school board members, wend that (relative to their non-Democrat counterparts) Democrat board members decrease racial segregation across schools: an electoral victory that shifts the board to majority Democrat causes a reduction in the black dissimilarity index across schools of 8 percentage points, while the election of even a single Democrat in the minority leads to a reduction of 15 percentage points. These two-stage least squares estimates significantly differ from their OLS counterparts, indicating that the latter are biased upward (understating the effects). To establish the key mechanism underlying these effects, we then use student addresses to construct a novel measure of attendance zone shifts, without needing to observe the exact geocoded boundaries. Our results show that such shifts increase by 0.15 standard deviations when an additional Democrat (relative to non-Democrat) is elected, consistent with them reversing the effects of neighborhood sorting. We identify two associated knock-on effects in districts with high proportions of black students: the initial boundary adjustments are somewhat counteracted through additional neighborhood sorting and "white flight" out of the district.