Association for Public Policy Analysis & Management

Historically, housing and education have been inextricably linked in the United States, with access to high quality schools often limited to middle and upper income families while poor children are “stuck” living in poor neighborhoods with low quality schools. The notion of weakening this link and increasing neighborhood integration through policies such as increased school choice has strong intuitive appeal, yet there is limited research examining the relationship between school choice and residential sorting in a real-world context. Research and theory suggest that the increased choice provided by charter schools may lead to less segregated neighborhoods as families choose to locate in less expensive neighborhoods with lower quality zoned schools. At the same time, prior work shows that white students tend to attend charter schools with higher portions of white students than their zoned schools. Therefore, it is possible, that housing integration may come at a cost. As the charter school sector continues to expand, it is important to understand the possible unintended consequences of increased choice on residential and school segregation.

In this paper we examine the effects of charter schools on neighborhood and school segregation in an urban context. We focus on New York City, an ideal setting to study this phenomenon given both historically high levels of residential segregation and a rapidly expanding charter sector.

First, we focus on sorting within Community School Districts (CSDs), comparing the characteristics of school zones within the same CSD over time, and examining whether and how families in the same CSD re-sort into school zones after a charter school opens. We classify school zones within a CSD into terciles (low, average, and high quality schools) and interact these terciles with measures of charter penetration to examine the effects of charter schools on various outcomes, including median household income, percent white, and exposure indices.

Second, we exploit the fact that students living in the same CSD as a charter school often receive an admission preference to employ a boundary discontinuity approach comparing the characteristics of neighborhoods on either side of CSD boundaries. The intuition is that neighborhoods on either side of the CSD boundary should be otherwise similar except in their differential access to charter schools. This boundary discontinuity approach allows us to examine how neighborhood characteristics vary in response to increased charter school penetration.

Finally, we explore the effect of charter schools on school segregation by examining the socioeconomic and racial composition that NYC public school students would experience if they all attended their zoned school versus the composition of the schools they actually attend.

This study adds to the existing literature by examining the effects of charter schools on the neighborhoods in which they operate, and on the demographic and socioeconomic composition of their surrounding schools. Further, it provides empirical evidence of the general equilibrium effects of charter schools complementing existing structural model estimates of the effects of school choice on residential sorting. Finally, it provides evidence of the broader implications of school choice policies in an urban setting.