Abstract
Since being created in the 1970s, housing vouchers have become the primary mode of federal housing support for low-income households in the US. The voucher programme was designed to provide recipients with the mobility needed to secure higher quality housing in neighbourhoods of their choice. Decades of analysis suggest that the programme has failed to produce the favourable outcomes envisioned by policymakers. To add to our understanding of the outcomes of this important federal programme, this paper seeks to underscore the importance of context-dependent policy analysis. In particular, this study analyses the impact of housing market conditions on the outcomes achieved by voucher recipients. Using neighbourhood and housing outcome data from the American Housing Survey, and median rent and rental market vacancy data, this paper demonstrates the important role that market conditions play in programme outcomes. The results from this study suggest that voucher recipients are successful at improving housing unit quality outcomes regardless of market conditions, but the ability to move to a better neighbourhood is a function of vacancy rates.
Disclosure statement
No potential conflict of interest was reported by the author.
Notes on contributor
Gregg Colburn is an assistant professor of real estate at the University of Washington. He studies housing policy, housing markets, and homelessness.
Notes
1 As a robustness test, propensity score matching (PSM) was also used to predict the outcomes of interest using the same model terms. PSM can only be applied to models without the interaction term between voucher and vacancy rate. The outcomes of the PSM models were consistent with the logit models that appear in the following analyses.
2 To test the robustness of this finding, I use an objective measure of housing adequacy (zadeq in the AHS) as a different dependent variable. The results are very similar to the better housing unit variable that serves as the primary dependent variable in this analysis. The results suggest that this finding holds for both objective and subjective measures of housing quality.
3 Because of low response rates for objective measures of neighbourhood quality in the AHS, this analysis is limited to the single, subjective measure of neighbourhood quality.