This study was conducted by staff from Abt Associates, which co-administers CLEAR. The review of this study was conducted by ICF, which co-administers CLEAR and is trained in applying the CLEAR causal evidence guidelines.
Citation
Moulton, S., Freiman, L., & Lubell J. (2021). Quasi-Experimental Impacts of Family Self-Sufficiency Programs Administered by Compass Working Capital in Partnership with Housing Agencies in Cambridge, Boston, and Lynn, MA. Cambridge, MA: Abt Associates.
Highlights
- The study's objective was to examine the impact of Family Self-Sufficiency (FSS) programs on households’ earnings and public benefits receipt outcomes.
- The study used a nonexperimental design to compare the outcomes of households enrolled in a FSS program to a matched comparison group of households served by other programs in urban areas. The authors used administrative data and statistical models to compare the outcomes of the treatment and comparison households.
- The study found a significant relationship between FSS program participation and increased short- and long-term earnings and reduced short- and long-term public-benefits receipt.
- This study receives a moderate evidence rating. This means we are somewhat confident that the estimated effects are attributable to Family Self-Sufficiency (FSS) programs, but other factors might also have contributed.
Intervention Examined
Family Self-Sufficiency (FSS)
Features of the Intervention
Family Self-Sufficiency (FSS) programs, established in 1990 by Congress, were designed to increase the economic security of households participating in rental assistance programs. FSS programs were funded through private-public collaborations, which may include housing agencies, private owners, and grants from philanthropic organizations. The FSS programs were offered by housing authorities in Cambridge, Boston, and Lynn, Massachusetts. The programs included affordable rental housing, along with 1) case management, one-on-one financial coaching, and/or coordinated services and 2) an escrow savings account that increased with participants’ earnings and rent contributions.
Features of the Study
The study used a nonexperimental design to examine the impact of FSS programs on households’ earnings and public benefits receipt outcomes. The study treatment sample originally included 792 households who volunteered to participate in an FSS program at the Cambridge Housing Authority (CHA), Metro Housing Authority (Metro Housing), and Lynn Housing Authority and Neighborhood Development (LHAND). The analysis ultimately included 564 households that completed the program, excluding those with incomplete data. The primary data source was administrative data provided by the U.S. Department of Housing and Urban Development (HUD) for the period 2007 to 2020. A matched comparison group of households was created by pairing each treatment household with three similar households from other HUD rental assistance programs in urban areas of Massachusetts, Rhode Island, and Connecticut, resulting in a total sample of 2,256 households receiving rental assistance. The authors used statistical models to compare the outcomes of FSS participant households with households enrolled in other HUD programs.
Findings
Earnings and wages
- The study found that FSS program participation significantly increased earnings in both the short-term (approximately 1.5 years) and long-term (approximately 3.2 years).
Public benefits receipt
- The study found that FSS program participation significantly reduced the receipt of both short-term and long-term public assistance, including Temporary Assistance for Needy Families (TANF) and Supplemental Security Income (SSI).
Considerations for Interpreting the Findings
The authors note that the findings rely on administrative data from housing agencies, which may vary in accuracy based on the completeness of public housing agency income certifications. The authors also estimated impacts on multiple related outcomes related to public benefits receipt. Performing multiple statistical tests on related outcomes makes it more likely that some impacts will be found statistically significant purely by chance and not because they reflect program effectiveness. The authors did not perform statistical adjustments to account for the multiple tests. As such, it is possible that the number of statistically significant findings may be overstated.
Causal Evidence Rating
The quality of causal evidence presented in this report is moderate because it was based on a well-implemented nonexperimental design. This means we are somewhat confident that the estimated effects are attributable to FSS programs, but other factors might also have contributed.