The fight between auto dealers and the Consumer Financial Protection Bureau could reach the 10th round in 2015.
Don’t look for a knockout punch or blood on the mat. But this is the year we may see a decision on a serious point of contention. It’s about time. The dispute has been going on for a few years now. Dealer groups have testified and lobbied hard in Washington.
Dealers and the CFPB agree on one thing: Dealers should get paid for setting up auto financing agreements between their customers and third-party lenders.
“Dealers provide important services, and are entitled to be compensated,” Patrice Ficklin, a CFPB assistant director, told an automotive financing conference.
So getting paid for services rendered isn’t the issue. The dispute centers on how dealers should get paid. The CFPB dislikes the current prevailing method. The federal agency is pressing the lending institutions it regulates to end that. The CFPB doesn’t directly oversee dealers but to their dismay it has tried to indirectly regulate them through lenders.
Dealer-assisted financing represents about 85% of U.S. car loans. Typically, dealers add a percentage point or two to a lender’s wholesale interest rate or “buy rate.” So if that initial rate is 4% and a dealer adds 1%, the customer pays 5% interest.
The CFPB disdains that system. The bureau is pressing for a flat fee or something like it.
Here’s the CFPB’s complaint, and bear with me for using a few inside-baseball terms: The regulator claims so-called “dealer reserve,” or the percentage add-ons, has caused “disparate impact” or “unintended discrimination” against minority borrowers.
The bureau says a study it did proves its claim. Largely because of the study results, the CFPB pressed Ally Financial to pay nearly $100 million to settle a discrimination claim. That got everyone’s attention. It scared a lot of dealers and lenders who did not want to be next.
The mood is changing though. Before, affected parties squawked about how misguided they thought the government study was. That criticism didn’t go far in convincing the CFPB its methodology was flawed.
But an independent review commissioned by American Financial Services Assn. offers strong evidence. It challenges CFPB claims of unintended discrimination. The AFSA analytically makes its case.
“Alleged pricing discrepancies between minorities and non-minorities for auto financing rates are simply not supported by data,” AFSA President Chris Stinebert says of the analysis his association commissioned.
Central to the CFPB discrimination claims is the Bayesian Improved Surname Geocoding (BISG) proxy methodology. The CFPB used it in an attempt to determine disparate impact to legally protected groups.
BISG estimates race and ethnicity based on an applicant’s name and census data. In doing its analysis, the CFPB and Department of Justice did not know for sure which borrowers were minorities. Instead, the agencies assigned race and national-origin probabilities.
So if your name is Martinez, the weighted assumption is you’re Hispanic. If you live in a particular urban zip code, the guess is you’re African-American. Those presumptions may or may not be correct. But the CFPB has run with them.
The AFSA study calculated BISG probabilities against a test population of mortgage data, where race and ethnicity are known. Among AFSA findings:
- When the proxy uses an 80% probability that a person belongs to an African American group, the proxy correctly identified race less than 25% of the time.
- Applying BISG on a continuous method overestimates the disparities and the amount of alleged harm, and provides no ability to identify which loans (also known as retail-installment contracts) are associated with the allegedly harmed consumers.
Using BISG methodology the way the CFPB did is a misapplication. Health-care organizations typically use it to get random statistical information on different ethnic groups. They don’t wield it as a weapon to take millions of dollars in settlement money from financial institutions.
BISG has a fairly wide margin of error, as the AFSA analysis notes. Yet the CFPB utilized it to allege fractional differences in auto-loan interest rates for minorities.
The government says loan basis points were higher by 29 (0.29%) for African Americans, 22 (0.22%) for Asians and 20 (0.20%) for Hispanics. So, even if the CFPB were accurate in its over-reaching assumptions, we’re talking about a few bucks.
Holman W. Jenkins Jr. of The Wall Street Journal wrote about the oddity of it all in a column headlined “Finding Racism Where It Isn’t.”
He says of the situation: “No discrimination is evident at the dealer level, and can’t exist at the lender level since the lender can’t see the face of the borrower. Discrimination is charged based solely on disparities that arise because dealers have different business models.”
For example, he says, a small urban dealership selling low-end cars with tight profit margins may rely more on credit terms to make money. In contrast, a big suburban dealership might offer more attractive financing when selling pricy high-margin vehicles.
The disparity between the two is something regulators jump on as being race-related, Jenkins says.
Everyone involved in this dispute denounces discriminatory lending practices.
“There is no room for discrimination in financing or any other aspect of buying a car,” says Andy Koblenz, the National Automobile Dealers Assn.’s executive vice-president-legal and regulatory affairs.
But dealers and lenders get defensive when the government says they’re unwittingly doing something wrong, presents questionable evidence as proof and then demands a major change in the way they do business.
“It is tempting to use a hammer, and treat everything like a nail,” Stinebert says “The CFPB seems to be wielding a hammer.”