Closing the Racial Refinancing Gap

Based on research by Kristopher Gerardi, Paul S. Willen, and David Hao Zhang

• White and Asian homebuyers are more likely to refinance their mortgages than Black and Hispanic homebuyers, especially in response to overall market changes.
• The fact that Black and Hispanic homebuyers are less likely to refinance (and face barriers to some refinancing options) costs them $812 million in extra annual interest payments.
• Race-neutral policies could help close this gap.

Black and Hispanic homeowners pay significantly higher mortgage interest rates than white and Asian homeowners. In part, the difference can be explained by factors like the borrower’s credit score, pre-existing debt and whether they faced racial bias when requesting a loan. However, controlling for these factors only slightly reduces the difference in interest rates. So why are paid interest rates so much higher for Black and Hispanic borrowers?

According to Rice Business professor David Hao Zhang, along with Kristopher Gerardi of the Federal Reserve Bank of Atlanta, and Paul S. Willen of the Federal Reserve Bank of Boston, the difference actually lies in refinancing. A new study shows that white and Asian borrowers are more likely to refinance than Black and Hispanic borrowers, especially in response to macroeconomic factors such as falling interest rates. Zhang, Gerardi, and Willen find that this difference in refinancing behavior is the most significant determinant in the higher interest rates that Black and Hispanic borrowers pay.

To measure how large the interest rate gap is, the study looked at a nationally representative sample of loans originated from 2005 through 2015. The loans were for 30-year fixed mortgages, and they included both Federal Housing Administration (FHA) loans and loans from government-sponsored enterprises (GSEs) such as Fannie Mae and Freddie Mac. The study also looked at borrowers’ credit scores, noting that the FHA program is characterized mostly by first-time homebuyers with low income and low credit scores.

To get to the key factor causing the gap in interest rates, the study controlled for several characteristics that might contribute to differences in interest rates at the start of a loan. Before these controls, refinancing rates for white and Asian borrowers were similar, while Black and Hispanic borrowers were less likely to refinance. After the controls, for both GSE and FHA loans, the gap in refinancing rates between white borrowers and Black borrowers decreased significantly.

Zhang, Gerardi, and Willen also looked at the effect of monetary policy on the racial interest rate gap. In November 2008, the federal government announced a plan for a large-scale buyback of mortgage-backed securities, in order to drive down interest rates. The researchers compared refinancing rates during the six months before and the six months following this announcement.

Although the buyback benefited Black and Hispanic homebuyers somewhat, expansionary monetary policy leading to lower interest rates disproportionately benefited white borrowers and exacerbated mortgage rate inequality, according to the study. The probability that white borrowers would refinance their loans increased by 3.2 percentage points (per quarter) after the announcement, compared with an increase of only 1 percentage point for Black borrowers. This led to rates paid by white borrowers dropping significantly after November 2008. Rates paid by Black and Hispanic borrowers also declined, but by much less.

The problem was compounded by the fact that white borrowers appeared to have a stronger response to changes in market rates. For example, in 2006 and 2007, Black and white borrowers refinanced at around the same rate. In 2009 and 2010, after the recession, white borrowers were twice as likely as Black borrowers to refinance. According to the study, Black borrowers were more likely to hold onto higher-than-market interest rates, and held onto these rates for a longer time. The differences didn’t level off until around the second quarter of 2013.

Zhang, Gerardi, and Willem argue that because most of the racial refinancing gap can be explained by observable differences, race-neutral policies may be effective at closing this gap. They provide examples such as expanding the use of adjustable-rate mortgages or encouraging the use of fixed-rate mortgages with prepayment penalties. Recently, partly in order to close this gap, the Federal Housing Finance Agency implemented new rules to raise mortgage fees for homebuyers with high credit scores, and to lower fees for homebuyers with lower credit scores. No matter the method used, the study shines a light on an opportunity to create more equity in personal finance by closing the racial refinancing gap.

David Hao Zhang is an assistant professor of finance at Jones Graduate School of Business at Rice University.

To learn more, please see: Gerardi, K., Willen, P.S. & Zhang, D.H. (2023). Mortgage prepayment, race, and monetary policy. Journal of Financial Economics, 147(3), 498-524. https://doi.org/10.1016/j.jfineco.2022.12.001