Disparate Impact 2.0

There is a knock on your favorite dealer’s door. He opens it to find a representative from his primary indirect lender, who announces they are going to do an unannounced deal-jacket audit to check for ECOA compliance. This could turn out to be a very long day, depending on the compliance program the dealer has in place.

Are your dealers ready for the next knock on the door? They should be, because federal regulators are putting immense pressure on banks and finance companies, and dealers are feeling it. Let’s discuss what auditors are looking for and how a clear understanding of the theory of “disparate impact” can help you prepare your dealers for any inquest.

Just a Theory

By way of refresher, the Consumer Financial Protection Bureau (CFPB) and the U.S. Department of Justice (DOJ) use disparate impact to go after indirect automotive lenders under ECOA, which is shorthand for the Equal Credit Opportunity Act. The ECOA generally makes it illegal to discriminate based upon race, gender, age, national origin, religion and other factors. Many car buyers are considered members of one or more protected classes under the law.

Under the disparate impact theory, an analysis is needed to determine if members of protected classes are being treated fairly compared to similarly situated individuals who are not in a protected class. To determine whether protected class members were involved in automobile loan transactions, the CFPB uses something called the Bayesian Improved Surname Geocoding (BISG) methodology.

The BISG theory is based on census data, and census data, in turn, is based on citizens making their own (unverified) report of their own ethnic background and providing their last name. BISG takes a portion of a ZIP code and list of surnames and concludes (arbitrarily) that if 80% or more of the census group were in a protected class, then 100% of their neighbors are deemed to be in the protected class as well.

The assumption here is that 80% somehow equals 100%. The further assumption is, for instance, that if a certain percentage of members of a protected class have a certain surname, then that percentage is present in the ZIP code being analyzed.

This whole process is sometimes referred to as using a “proxy,” since indirect automobile lenders cannot directly collect this information. Quite understandably, this use of BISG/proxies has been referred to as “junk science” by no less than the House of Representatives’ Financial Services Committee. In fact, the chairman of that committee, Rep. Jeb Hensarling (R-Texas), has gone so far as to refer to the CFPB as a “dangerously out of control agency” and said the CFPB is essentially “inventing” discrimination by using these methods.

While the withering criticism of the CFPB’s use of shaky theories to establish a disparate impact case is encouraging, it does not eliminate this practice, and disparate impact claims continue to exist in 2016. … Or do they?

Reason for Hope

Last June, the U.S. Supreme Court issued a decision in the Texas Department of Housing and Community Affairs v. Inclusive Communities Project Inc. The issue at hand was whether disparate impact theories can be used in a case arising out of the Fair Housing Act (FHA). Proponents of disparate impact theories and detractors of disparate impact theories both thought that the case may finally lay to rest any doubts about the validity of this theory.

In a 5-4 decision, the Supreme Court found that Congress intended to include disparate impact in the FHA. The CFPB might have claimed a total victory if the justices hadn’t gone on to say that mere statistical evidence is not enough to sustain a disparate impact claim. On the contrary, the Supreme Court imposed what they called a “robust causality requirement,” demanding proof that a particular policy caused the statistical disparity regarding the protected class.

This causation requirement gave renewed optimism to those seeking to eliminate the disparate impact theory from the CFPB’s arsenal. The Supreme Court also described a “valid interests” defense: If the underlying policies or policy was necessary to achieve a “valid interest,” then the disparate impact claim could be defeated. Keep in mind that the Inclusive Communities case was decided under the auspices of the FHA and not the ECOA, as applied to indirect automotive lenders.

So what is the final takeaway? Disparate impact claims appear to have survived the FHA case, although defenders of these claims have gained some insight into valid defenses, too. Neither side can claim complete victory. That leaves you, the agent, with two key questions to ask of all your dealer clients:

• Does your dealership have written policies and procedures regarding your credit policy?
• Does your dealership maintain written documentation of valid business reasons for deviating from your written credit policy?

If your dealers consistently apply and document their credit policies as part of a comprehensive compliance management system, their next unannounced audit visit from an indirect lender will go much more smoothly.