Running to Stand Still

Abstract

Research and policy on the geography of assisted housing is dominated by a powerful conventional wisdom: Project-based subsidies are presumptively bad because they anchor assisted households in poor, racially segregated neighborhoods, while vouchers are inherently good because they promote deconcentration and integration through tenant choice. Unfortunately, this consensus is based on geographical assumptions that have been subverted by the dramatic restructuring of cities with tight housing markets over the last generation. In this study, we use the case of New York City to analyze these spatial contradictions. Project-based subsidized housing is disappearing from yesterday's poor neighborhoods that have been remade by gentrification at the urban core, while recipients of Housing Choice Vouchers are concentrated in today's poor neighborhoods of color farther from the city center. If the policy goal is to break the link between housing assistance and the stereotypes of “projects” in the worst neighborhoods, then in the case of tight, expensive urban housing markets, voucher-driven deconcentration will be less successful than the preservation of the existing project-based housing stock.