Ross Research: Minorities Get Worse Deals From Banks Thanks In Part To Rules Designed To Help Them


New research by Professor Amiyatosh Purnanandam and PhD alumnus Taylor Begley shows poor and minority customers see inferior mortgage services from banks.

Bank lending practices to minorities have come under scrutiny recently, highlighted by the city of Philadelphia’s lawsuit against Wells Fargo that accuses the bank of selling inferior products in minority areas.

New research by Michigan Ross Professor Amiyatosh Purnanandam and Ross PhD alumnus Taylor Begley, assistant professor of finance at Olin Business School at Washington University in St. Louis, suggests minority customers indeed experience inferior and possibly unfair service from retail banks in general on mortgage products. But it also suggests the driver could be legislation intended to increase credit access to those very same customers.

They analyzed data from the Consumer Financial Protection Bureau, which collects consumer complaints against banks. Examined by zip code, Purnanandam and Begley found that complaints on mortgage deals were significantly higher in zip codes with lower education rates and incomes. Even after controlling for income and education, complaints were still higher in zip codes with a significant minority population.

For example, a one-standard deviation increase in a zip code’s minority population leads to a 16-percent jump in complaints to the CFPB, while the effect for income is 2 percent and education 6 percent. The effect also increases exponentially as the minority population rises. Complaints to the CFPB are considered significant because they concern issues that weren’t resolved between the bank and customer, and because it takes some time and effort to file the complaint.

They focused on mortgage-related complaints because it’s economically more meaningful and mortgage-related misconduct has been the focus of regulator scrutiny and research.


“This was a pervasive effect, and we looked at it very carefully in many different ways,” says Purnanandam, professor of finance. “It can’t be explained away by lower income and low education rates.”

The authors also ruled out high foreclosure rates, declining home values, and the theory that low-income customers have a lower financial threshold for filing a complaint as reasons for the disparity.

Instead, an unintended consequence of regulations such as the Community Reinvestment Act may be driving banks to sell low-income and minority customers products with less focus on the quality or fit of those loans.

The Community Reinvestment Act was created to remedy the unequal access to credit for low-income and minority areas. The authors say that’s an important goal, but one that is incomplete since the legislation measures only quantity, not quality.

And that might create the wrong incentive for banks, says Begley. The authors found an adverse effect of the regulation on the quality of mortgage products and services across the country. But this effect was substantially stronger in minority zip codes targeted by the CRA. That suggests banks may be making unsuitable loans to minority consumers in these zip codes to satisfy quantity requirements.

“When you’re writing rules and policies, you tend to rely on things that are easy to measure,” says Begley. “The Community Reinvestment Act, though well-intentioned, focuses on certain quantity metrics for its enforcement. Those are important because we want to improve access to credit for truly creditworthy borrowers, but it provides an incomplete picture. We also need to be looking at quality. All in all, the act may not be bad, but there’s an important hidden cost we think should be monitored.”

Because banks will manage to what they’re measured by, the data suggests not enough thought goes into understanding customers in low-income and minority markets, the authors say. As a result too many customers in those areas wind up with loans that aren’t a good fit.

“We’re not saying banks get together and decide to sell poor people and minorities bad products, but those populations are seeing worse outcomes and we need to find out why,” says Purnanandam. “The policy takeaway from our research is that we need to start thinking about quality based regulations.”



Media Contact: