Benefit–Cost Analysis of an Innovative Program for Self-Sufficiency and Homeownership

• Published in: Evaluation review 2019-02, Vol.43 (1-2), p.3-40
• Main Authors: Galster, George C., Santiago, Anna Maria, Smith, Richard J., Leroux, Joffre
• Format: Article
• Language: English
• Subjects: Bias; Cost analysis; Cost benefit analysis; Cost Effectiveness; Deviation; Earnings; Economic models; Health technology assessment; Home ownership; Housing; Inferences; Monte Carlo Methods; Monte Carlo simulation; Ownership; Poverty; Propensity; Quasi-experimental methods; Research Design; Self sufficiency; Simulation; Social sciences; Specification; Wealth
• ISSN: 0193-841X; 1552-3926
• DOI: 10.1177/0193841X19846697

Background: Federal policy has increasingly sought to build financial capability, earnings, and assets of subsidized housing recipients. Objective: We conduct a benefit–cost analysis of the Denver Housing Authority’s (DHA) innovative Home Ownership Program (HOP), which incentivizes participants to increase earnings, build wealth, and purchase homes. Research design, subjects, and measures: In assessing HOP participant benefits (earnings, home-buying, and positive exits from DHA), we use parameter estimates from quasi-experimental methods (i.e., propensity score matching) that permit drawing causal inferences of program impacts. Impact estimates are robust to alternate model specification and mostly insensitive to omitted variable bias found in the social sciences. We deploy a comprehensive accounting framework, distinguishing benefits and costs accruing to program participants, nonparticipants (other citizens, taxpayers, and governments), and society as a whole. We use Monte Carlo simulation techniques to approximate distributions of benefit and cost parameters, thereby ascertaining how reliably participation in HOP yielded net benefits compared to if families had continued to receive housing assistance during the same period. Results: We estimate a net social benefit from HOP of US$6,015 per participant. The simulated standard deviation was only a third of this value and 99.9% of simulations returned positive net social benefits. Conclusion: We conclude with a high degree of statistical confidence that HOP produced substantial net benefits to society as a whole, program participants, and nonparticipants alike. HOP offers strong potential for poverty alleviation among housing subsidy recipients and should be replicated.