The much-heralded CFPB Auto Finance Guidance issued in March 2013 appears to be losing its momentum in the first part of 2015.
You will recall that the Consumer Financial Protection Bureau (CFPB), which lacks jurisdiction under the Dodd-Frank Act over most franchised automotive dealers, claimed that lenders allowing dealers to mark up lender buy rates on credit sales had a “disparate impact” effect on protected classes of persons, principally women and minorities, under the Equal Credit Opportunity Act (“ECOA”). This, they claim, is credit discrimination. Very little detail was given to back up these assertions but the CFPB settled with a large lender for $98 million in December 2013 and several other lenders in September 2014.
In their recently issued 2014 Fair Lending Report, the CFPB indicated it “investigated a number of indirect auto lenders and has a number of authorized lawsuits.” Yet, no such suits have been filed. Also, the CFPB has announced no confidential supervisory settlements of auto finance credit discrimination since September 2013. That’s a long time for what the CFPB had previously characterized as a priority issue.
Debunking the Auto Finance Guidance
The original Auto Finance Guidance was very general and conclusory. It didn’t explain how the CFPB reached its conclusions. You can’t collect racial and demographic information in automotive finance transactions the way you can with mortgages. It took a while, but the CFPB announced it had identified racial and minority persons and their discriminatory rates by using a “proxy” known as the Bayesian Improved Surname Geocoding or BISG proxy. BISG estimates race and ethnicity based on an applicant’s name and census data. But in November 2014, AFSA released a study by the Charles River Associates Group that essentially blew away the credibility of the BISG as a legitimate and accurate way to identify classes of persons.
The Charles River Study calculated BISG probabilities against a test population of mortgage data, where race and ethnicity are known. Among the findings:
- When the BISG proxy uses an 80 percent probability that a person belongs to an African American group, the proxy correctly identified their race less than 25 percent 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 contracts are associated with the allegedly harmed consumers.
- When appropriately considering the relevant market complexities and adjusting for proxy bias and error, the observed variations in dealer reserve are largely explained. In looking at approximately 8.2 million new and used motor vehicle retail installment contracts originated during 2012 and 2013, the researchers found little evidence that dealers systematically charge different reserves on a prohibited basis and instead found that reserve variations could “largely be explained by objective factors other than race and ethnicity.”
Remember that $98 million CFPB consent decree in December 2013? As if to confirm the Charles River Study’s findings, not one customer refund check has been issued almost 18 months after the fact. Apparently, the CFPB can’t figure out who to send the refund checks to.
Judicial and Congressional Action
The CFPB is also facing a problem in the U.S. Supreme Court. In a case brought challenging the legitimacy of disparate impact credit discrimination under the Fair Housing Act—a law that has exactly the same language as ECOA in prohibiting only intentional discrimination and not disparate impact claims—the Supreme Court is expected to rule on the legitimacy of the disparate impact claim possibly before this article reaches print. In 2005, the Supreme Court ruled in an employment law case, that a disparate impact cause of action had to be legislated by Congress in the very words of the law and not be a product of a regulator’s concoction of what the Congress may have intended. Depending on how the Court rules now, the CFPB may be hard pressed to continue to assert that disparate impact credit discrimination is a viable legal theory.
The Congress has also gotten into the act. Last September, 130 House members of both parties sponsored a bill to repeal the 2013 Auto Finance Guidance. While that bill died at the end of the session, a bill to cut back the CFPB’s budget to a lower portion of the Federal Reserve’s budget (the CFPB is not appropriated by Congress) was recently passed out of Committee and sent to the House floor. Other bills to make the CFPB headed by a panel instead of a single Director and making at least a part of the agency’s funding dependent on Congressional appropriations are also making their way through the Congress.
In the meantime, few lenders have gone to flat fee pricing as the CFPB wants and dealer participation is alive and well. Even the CFPB has hinted that rate markups of 100BP or less have a good probability of not creating statistically significant rate variances.
A Possible Solution
So where will this end? In 2007, the Department of Justice (DOJ) settled two disparate impact credit discrimination claims with automotive dealers and their methodology in doing so may provide the impetus for a resolution now. The DOJ told the dealers to adopt a standard rate markup for all customers. They could deviate downwards (but not upwards) only if there existed a “legitimate business reason” and the DOJ identified seven of them. Every deal jacket would contain a worksheet indicating whether the standard markup or a lower markup was used and, if lower, the legitimate business reason that justified doing so. The DOJ’s Chief of Enforcement confirmed his agency’s support for that solution at a CFPB hearing in November 2013.
NADA has essentially put into words the DOJ’s solution in its Fair Credit Compliance Policy and Program (available at https://www.nada.org/faircredit/) and it is a program that all dealers should consider adopting. As in 2007, you establish a standard rate markup and apply that to all customers unless one of the seven legitimate business reasons applies to justify a lower markup. You document in each deal jacket what the standard rate markup is and whether you used it or a lower amount. If lower, you indicate which of the seven legitimate business reasons justified doing so and keep that worksheet in the deal jacket.
The CFPB is trying to claim that even if every dealer adopted the NADA program, different rate markups by different dealers would create a “portfolio-level disparate impact” for lenders who buy from many dealers. There is no legal authority that supports this position and that argument was pretty well disposed of by the U.S. Supreme Court several years ago in a case against Wal-Mart. In the Wal-Mart case, the Court held that the hiring decisions of individual store managers could not be imputed to Wal-Mart as a whole to prove Wal-Mart was discriminating. In the auto finance context, dealers are independent business people and trying to make a lender liable for rate differences among individual dealers is at least as tenuous as the argument against Wal-Mart for its individual store managers.
So on all fronts—legislative, judicial, statistical—the CFPB is having a hard time defending the 2013 Auto Finance Guidance. That would be consistent with why no actions have been filed or consent decrees issued for almost nine months. The tone of this “disparate impact” issue has definitely changed and is changing and not in a way the CFPB would like.