Transportation and environmental burdens in affordable housing: Evidence from federal housing programs in the Washington metropolitan area

This study examines how accessibility to employment and everyday services and exposure to environmental and health burdens are associated with the geographic presence and concentration of federally subsidized housing in the Washington metropolitan area. Affordable housing programs allocate units through distinct institutional mechanisms that shape residential location patterns. Little is known about how these allocation structures correspond jointly with accessibility and environmental exposure. The analysis contrasts the Low-Income Housing Tax Credit (LIHTC) program with the Housing Choice Voucher (HCV) program. The LIHTC program supports place-based developments selected by private developers through project-level allocation processes. The HCV program provides tenant-based subsidies that enable households to lease units in the private rental market, where affordability constraints and landlord participation shape residential options across a broader set of locations. This institutional contrast enables the examination of whether accessibility and environmental exposure are associated differently under project-based versus tenant-based housing assistance. Census tract–level measures of accessibility and environmental exposure are incorporated into two-step models that estimate both the likelihood of program presence and the concentration of program units. Three findings emerge. First, employment accessibility aligns more strongly with HCV than with LIHTC. Employment access increases both the likelihood of HCV presence and the share of HCV units, whereas its association with LIHTC remains confined to the probability of presence and does not extend to the project scale. Second, environmental and health burdens correspond differently across programs. LIHTC developments are more strongly associated with census tracts characterized by traffic exposure, lower life expectancy, and chronic health conditions, while HCV households exhibit a broader but more heterogeneous exposure profile. Third, accessibility and environmental exposure are not independently distributed across affordable housing programs. Higher employment accessibility in LIHTC tracts corresponds with elevated environmental exposure, whereas the employment accessibility gradient for HCV weakens as environmental exposure increases. These conditional patterns demonstrate that accessibility and environmental exposure correspond in distinct program-specific configurations across neighborhoods.