ABA Bank Compliance - September/October 2013

Indirect Automotive Market

ON MARCH 21, 2013, the Consumer Financial Protection Bureau (CFPB) issued a bulletin on compliance with the Equal Credit Opportunity Act (ECOA) in the indirect automotive finance market. This bulletin prompted market participants to take a sharp look at fair lending compliance and analysis in their indirect automotive finance products. 1

Most banks in the mortgage market have established compliance programs and analytical frameworks that meet regulatory standards. While such existing programs provide some benefit when building a corresponding program in the indirect automotive space, there are a number of important differences that must be addressed including the types of analyses performed, the methodologies employed, the interpretation of results, and any subsequent actions taken.

determination of the Prohibited Basis

One key to indirect automotive fair lending analysis is determining which applicants reflect the “prohibited basis” since banks and finance companies rarely collect this information under the self-testing provisions of ECOA/ Regulation B. As a result, the analysis must resort to using proxies based on the probability that an applicant is of a particular race, ethnicity, or gender.

Senior officials at the CFPB have suggested that they
intend to issue guidance on the use of proxies and have
referred practitioners to academic research that adopts
the Bayesian Improved Surname Geocoding (BISG) proxy
method. 2 BISG can be used to proxy race and ethnicity
(but not gender) and will typically increase the number of
applicants in each race/ethnicity group relative to a tract-
only proxy method. However, there are still high rates of
false positives and negatives with BISG. 3 Another recent
development regarding the proxy issue is the CFPB’s use
of continuous measures of race/ethnicity rather than a
threshold-based method. Specifically, this uses the indi-
vidual probability of each applicant’s race/ethnicity rather
than simply including applicants with probabilities above
a threshold value such as 80 percent. While this has some
theoretical advantages when dealing with lower application
volumes, banks and finance companies should take great
care in the interpretation of results derived by this method.

Regular Analysis of loan data

Once a proxy methodology has been identified, “regular analysis of loan data” must be conducted. 4 The CFPB expects banks and finance companies to monitor underwriting and pricing as part of a “robust fair lending compliance management process.” 5 While underwriting and pricing are familiar to the compliance officer, there are important nuances in the indirect automotive market that may not be well understood.