Q&A: Brent Ambrose

The Smeal College of Business professor's research reveals racial disparities in the mortgage market.

Mortgage illustration

Q: How did you study the link between mortgages and race?
Ambrose: We used a database of high-risk subprime loans purchased via the Institute for Real Estate Studies and made by New Century Financial, a real estate investment trust. The data identified the borrowers’ races, who originated the loan, and the fees borrowers paid their brokers. Based on where each mortgage broker was located and their name, we were able to guess their race and put everything together.

Q: What did you find out?
: In the raw data, there’s a pattern of minorities paying much higher origination fees than white borrowers, but this isn’t evidence of discrimination per se. Perhaps minorities selected mortgages that are more difficult to originate, which drove the fees up. By controlling for those kinds of variables through statistical regression and drilling down to get to the same type of mortgages being originated around the same time by the same broker, we cut the headline “discrimination” down by about half—and still we identified cases where minorities were being charged more fees than similar white borrowers.

Q: Could you quantify this?
Getting as close as possible to all variables being controlled for, we found that Black borrowers getting loans from a white broker were paying about 8% more than white borrowers. Hispanic borrowers paid a 5% premium relative to white borrowers. We also found that Hispanic borrowers paid the same premium to a Hispanic broker as white borrowers did.

Q: Interesting. Why?
: We can only conjecture that perhaps Hispanic brokers were using the Spanish language as a way to extract more fees.

Q: Are there better controls in today’s mortgage market?
Since the financial crisis, some policymakers have wanted to restrict brokers from earning dual compensation, a fee from the borrower and a rebate from the lender. But applying this to our dataset, we found the elimination of dual compensation didn’t change things in any meaningful way, which means it is unlikely to eliminate racial price disparities. The Consumer Financial Protection Bureau (CFPB) also introduced regulations limiting fees charged on “qualified” mortgages, to make them more affordable and reduce broker discretion in pricing. However, our analysis shows that that may also potentially reduce access to mortgages for borrowers requiring more effort. We estimated that about 20% of Hispanic borrowers and 29% of Black borrowers who obtained mortgages from white brokers would have been at risk of losing access to credit if the qualified mortgage policy were in place, so these rules alone are unlikely to eradicate minority premiums and may place borrowers at risk of credit rationing.

Brent Ambrose is the director of the Institute for Real Estate Studies in the Smeal College of Business.