2015 Series • No. 15–11
Research Department Working Papers
Do Increases in Subsidized Housing Reduce the Incidence of Homelessness? Evidence from the Low-Income Housing Tax Credit 
The provision of affordable housing for low-income families is often cited by policymakers and advocacy groups as a necessity for ending homelessness. The U.S. government spends a considerable amount on housing programs for the nation's poor, and the use of federal housing programs to mitigate homelessness has attracted increasing interest following the recent financial downturn and housing market crisis. While important for housing policy, however, the question of whether subsidized housing is effective for combating homelessness remains unresolved. In this paper, the authors examine the impact of subsidized housing on homelessness using the Low-Income Housing Tax Credit (LIHTC), the largest place-based housing program in the United States.
Key Findings
- Using data from the U.S. Census and HUD, the authors find that while the LIHTC program increases the local (tract-level) stock of low-income rental units, there is no evidence that these increases reduce local area homelessness in moderately poor neighborhoods.
- At the county level, however, LIHTC project installation does significantly reduce homelessness.
- The analysis suggests that the mobility of the homeless across neighborhoods helps to explain these results.
Implications
These findings are consistent with the impacts of place-based policies crossing local boundaries and homeless housing demand that is sensitive to rental prices.
Abstract
We examine the impact of subsidized housing on homelessness using the Low-Income Housing Tax Credit (LIHTC), the largest place-based housing program in the United States. To generate quasi-experimental variation in housing placements, we exploit a discontinuous increase in the amount of tax credits available to projects placed in certain high-poverty neighborhoods. Using data from the U.S. Census and HUD, we find that LIHTC project installation has no significant impact on neighborhood homelessness but does significantly reduce county-level homelessness. Our analysis suggests that the mobility of the homeless across neighborhoods helps to explain this result. These findings are consistent with the impacts of place-based policies crossing local boundaries and homeless housing demand that is sensitive to rental prices.
JEL Classifications: H20, H31, I32, R21, R31
Keywords: low-income housing, tax credits, homelessness, regression discontinuity