Community Advantage Program
Nearly 50,000 low- and moderate-income (LMI) and minority homeowners have received low down payment, responsibly underwritten home loans since 1983 via the Community Advantage Program (CAP). The CAP Study, a unique longitudinal research collaboration between the UNC Center for Community Capital and CAP initiative partners, has provided two decades of cutting edge research into the effects of responsible lending practices, mortgage policy frameworks, and low-income and minority homeownership.
The UNC Center for Community Capital has conducted a long-term study of low- and moderate-income (LMI) and minority homeowners to identify specific lending practices that enable and inhibit successful homeownership. Our findings have yielded an extensive collection of reports that provide facts and recommendations policymakers, regulators and the housing finance industry use to safely expand homeownership opportunity in the United States.
The Community Advantage Program Study, launched in 1999, examines data collected since 1999 from homeowners who receive mortgages through the Community Advantage Program (CAP), a path-breaking mortgage initiative of Self-Help, the Ford Foundation, and Fannie Mae that has funded more than $4 billion of home loans to LMI borrowers.
Self-Help, a community development financial institution, teamed up with Ford Foundation and Fannie Mae to provide a secondary market for these affordable mortgage loans. Their goals were two-fold: to expand home lending to LMI households and to demonstrate the creditworthiness and market opportunity the households offer the mortgage industry. The study has provided a unique opportunity to examine the performance of an affordable mortgage program before, during, and after the one of the most turbulent periods in the nation’s housing history.
CAP Study Components
The CAP Study is a long-term research effort that derives data from two primary components:
- The CAP database is a longitudinal record of the Self-Help Credit Union’s CAP mortgage portfolio, containing data on mortgage origination and performance, as well as borrower details and credit scores, for nearly 50,000 mortgage loans in 47 states.
- The CAP Survey (CAPS) is an 11 year panel survey of approximately 5,000 low- and moderate–income homeowners and renters during the period of 2003-2014. The UNC Center for Community Capital has overseen CAPS data collection with generous funding from the Ford Foundation and currently serves as the CAPS data custodian.
Please see our Frequently Asked Questions page for updates and details.
Access the Data
There are two tiers of CAP Study datasets available for research use:
- Tier 1 data is publicly available on the Carolina Digital Repository. This data includes deidentified survey datasets, annual collection reports, questionnaire specifications, and codebooks.
- Tier 2 data is available via a collaborative data-sharing protocol. This data includes matched supplemental data, mortgage origination and performance data, survey weights, limited geographic components, and other additional variables generated by CCC researchers. If you are interested in this data, please fill out this data request Letter of Interest.
Key findings and recommendations are published in two book-length works based off of the Community Advantage Program. The first, Regaining the Dream: How to Renew the Promise of Homeownership for America’s Working Families (Brookings Institution Press, 2011), recounts the history of the U.S. housing finance system, which significantly expanded U.S. homeownership and helped build the American middle class but also excluded many Americans from accessing this key wealth-building tool. The second book, A Place Called Home: The Social Dimensions of Homeownership (Oxford University Press, 2017), is an analysis of the social impacts and non-financial effects of affordable homeownership.
Additionally, The Center for Community Capital published a research brief in 2014 that describes many of the key findings from the first decade of research using this data.