Comparing Public Housing and Housing Voucher Tenants With Bayesian Propensity Scores

Brent D. Mast , U.S. Department of Housing and Urban Development

As with the articles in this issue, this introduction reflects the views of the authors and does not necessarily reflect the views of the U.S. Department of Housing and Urban Development.

The U.S. Department of Housing and Urban Development (HUD) currently has no administrative data to compare housing quality of public housing units with that of Housing Choice Voucher Program (HCVP) units. The American Housing Survey (AHS) provides the only data available to compare subjective housing and neighborhood quality assessments in HUD’s largest rental assistance programs.

Quality comparisons based on AHS data are problematic because the AHS overrepresents public housing and underrepresents the HCVP.

HUD administrative data, however, are an excellent source of prior information for the expected proportion of households in public housing. In this study, I explore Bayesian methods using prior information on variables such as income and rents to estimate propensity scores for program participation. I then use the Bayesian propensity scores to improve the reliability of AHS-based quality comparisons. Results indicate that, after adjusting for program participation propensities, little difference exists in AHS household and neighborhood quality ratings between public housing and voucher households.