Fannie Mae Study Demonstrates Strong Performance of HFA Homeownership Loans
Single-family loans originated through state HFA programs perform substantially better than similar loans to low- and moderate-income (LMI) borrowers, according to a working paper published last week by Fannie Mae. The paper also credits HFAs for serving lower income borrowers than the market as a whole and suggests that HFA programs offer Fannie Mae a way to meet its affordable housing goals without taking on increased risk. Fannie Mae has also published a commentary summarizing the paper’s findings.
The working paper, written by Stephanie Moulton of The Ohio State University, Matthew Record of San Jose State University, and Erik Hembre of University of Illinois, Chicago, examines data on more than 1 million Fannie Mae-guaranteed loans to first-time LMI homebuyers from 2005 to 2014, around 10 percent of which were originated through HFA programs. Comparing the performance of the HFA and non-HFA loans, they find that HFA program loans were 20 percent less likely to experience a long-term default and 30 percent less likely to be foreclosed.
To account for market fluctuations, the paper further breaks down loan performance into three different cohorts that reflect when the loans were originated: 2005-2007, 2008-2011, and 2012-2014. It finds that HFA loans originated in both the 2005-2007 and 2008-2011 periods outperform measurably the non-HFA loans. HFA and non-HFA loans originated in 2012-2014 have similar default and foreclosure rates, though HFA performance remains strong.
Comparing attributes of HFA loans to non-HFA loans, the paper suggests that HFA loans’ superior performance can in part be attributed to the high proportion of HFA loans originated by retail lenders as opposed to brokers, HFAs’ use of Fannie Mae Community Second loans instead of other subordinate financing, and HFAs requiring full documentation during underwriting.
The paper also uses data from NCSHA’s HFA Factbook to identify additional factors that improve HFA loan performance. Based on this analysis, the authors find that the use of in-house servicing and homeownership counseling for borrowers explains nearly three-quarters of the reduced default and foreclosure risk for HFA loans. The report concludes there is no difference in performance between HFA program loans originated through the HFA directly and those loans originated through participating lenders and then sold to the HFA.
The paper also notes some differences between borrowers with HFA loans and borrowers with non-HFA loans in the sample. On average, HFA program borrowers earned 70 percent of area median income (AMI), compared to 84 percent AMI for non-HFA borrowers. HFA borrowers also had slightly lower credit scores, took out smaller loans, and had higher loan-to-value ratios.
The paper concludes by noting that further research is needed, with a larger sample of loans, to fully understand the differences between HFA and non-HFA loans. NCSHA will further study the working paper and look to contribute to further research on HFA lending in the future.