The discussion on how to accurately identify discrimination in auto lending picked up at the congressional level over the last month, with a House Financial Services subcommittee hearing and the introduction of a bill that would allow lenders to collect demographic data from customers applying for auto financing.
Under new leadership, the Subcommittee on Oversight and Investigations reignited the debate on discrimination in auto lending during a hearing last week.
Industry experts had anticipated a shift when Democrats took over the House this year, and a longtime defender of consumer protection legislation, Rep. Maxine Waters, D-Calif., is delivering on those expectations.
The methodology the Consumer Financial Protection Bureau used to determine race has been widely criticized, but it fueled the CFPB's allegations that at least four auto lenders had discriminated against minority borrowers. The CFPB settled with those lenders from 2013 to 2016.
Last year, the House passed a measure to scrap the CFPB's 2013 auto lending bulletin. The bulletin suggested that variances in dealerships' discretion caused minorities to be charged higher interest rates than their nonminority counterparts with similar credit, even if no discrimination was intended.
Many industry groups, including the National Automobile Dealers Association, said that claim was based on flawed methodology. NADA and others said that by using the Bayesian Improved Surname Geocoding, which used names and ZIP codes to guess the race and gender of credit customers, the bureau had no method of accurately determining whether discrimination occurred.
Last month, Sen. Kirsten Gillibrand, D- N.Y., announced a bill that would require lenders to collect demographic data including age, race, sexual orientation, gender identity, sex and marital status for transactions less than $15,000 or on a repayment period less than three years. The goal of the bill, called the Protections in Consumer Lending Act, would be to furnish the CFPB with more data to determine whether discriminatory practices have occurred so that it may enforce anti-discrimination laws, Gillibrand's office said in a statement.
"We know that discrimination in credit lending is a problem, but we need more data to understand the scope of it," Gillibrand said. "My bill would help create transparency in the credit-lending process and help identify when discrimination is taking place."
Discriminatory dealer reserve rates and F&I product markup were discussed by a panel of attorneys, policy advisers and policy analysts during the House subcommittee hearing. But the specter of the Bayesian Improved Surname Geocoding hung over the proceedings and was challenged by numerous committee members on both sides of the aisle.
In his opening statement, Rep. Al Green, D-Texas, chairman of the subcommittee, said the hearing was "long overdue."
Green cited, among other findings, a 2018 study by the National Fair Housing Alliance, which claimed nonwhite vehicle buyers received costlier loans than their white counterparts, despite being more qualified for financing, 62.5 percent of the time
"These dealer markups, unrelated to credit risk, allow predatory pricing and invidious discrimination," Green said. "The empirical evidence that I've cited — which is but a scintilla of what's available — seems to overwhelmingly indicate that certain persons pay more than others for the same product."
The question of how to legally — and accurately — test for racial discrimination when disclosure of race isn't part of the vehicle financing agreement remained largely unresolved at the hearing, but maybe legislation, such as the bill Gillibrand proposed, will offer a solution.
How far the bill will go with a split House and Senate remains to be seen, but the pursuit of comprehensive data on possible discrimination should not be abandoned based on political motivation. Lawmakers on both sides of the aisle agree that racial discrimination in lending is and should be illegal, and they should work together to find a method that's accurate and fair to dealers, lenders and customers.