Association for Public Policy Analysis & Management

Parental housing decisions determine the quality of amenities and opportunities their children will access. Yet the effects of a single housing choice are not limited to an individual family; residential choice also affects the price and supply of choices available to other families, as well as the composition of the neighborhood and local schools. As a result, housing selection can both express and reinforce intergenerational inequality. Families with more resources leverage their advantage in the housing market to secure opportunities for their children, driving up housing prices in high-demand, highly resourced communities and ultimately creating prohibitive cost barriers for less-advantaged families.

Prior work shows that residential choice is not merely a story of economic sorting. White households have distinct and pronounced preferences for mostly white neighborhoods and school districts, net of correlated geographic, housing, economic, and educational factors. This is particularly true for the parents of school-aged children.

While considerable work has investigated the consequences of white racial homophily in the housing market, we know much less about how school-related racial residential choices vary according to family economic resources. Is school district racial homophily a pattern driven primarily by affluent Whites, or does the selection process span across the income distribution? How do house prices and rents fluctuate vis-à-vis this demand structure? And how would sorting outcomes change if group differences between Black and White family resources were equalized?

To address these questions, I evaluate school district sorting using the Panel Study of Income Dynamics from 2009 to 2015. With restricted geocodes, I observe where White and Black families choose to live (or stay) given the array of school district options available in their local metropolitan area. My research design accounts for individual family resources as well as a rich set of contextual measures drawn from the Decennial Census and American Community Survey, the National Center for Education Statistics Common Core of Data, and the Stanford Education Data Archive. I focus on four school district characteristics potentially driving demand: academic achievement, financial resources, housing stock, and racial composition. Using a McFadden conditional logit regression model, I evaluate how variation in household wealth and income moderates the influence these contextual characteristics have on where White and Black families choose to live, while accounting also for other neighborhood-level demographic and economic features.

I conclude with an estimation of the mean rent or home price that low-opportunity families would have to pay in equilibrium in order to access the highest performing school district in their metropolitan area. The results reveal an underlying demand structure reproducing spatial segregation at a wide school-district-level geographic scale, and they identify potential roadblocks to achieving integration in highly unequal housing markets.