December 1, 2015
Congressional Staff Report Reveals Sad Inner Workings of CFPB Fair Lending Attack on Dealer Compensation
Just before Thanksgiving, Capitol Hill lobbed a bomb at the Consumer Financial Protection Bureau. The explosion was heard across Washington, and the aftershocks were felt throughout the indirect auto finance world.
On November 24, the House Committee on Financial Services released a 54-page report by its Republican staff, entitled "Unsafe at Any Bureaucracy: CFPB Junk Science and Indirect Auto Lending." The report is a devastating attack on the Bureau's pursuit of discrimination claims under the Equal Credit Opportunity Act against indirect auto creditors for alleged pricing differences in contracts they buy from dealers.
Despite the title of the report, its tone is restrained and even academic. The shock is provided by the report's liberal quotations from Bureau internal documents. Quotes from leaked CFPB confidential documents began appearing in the press last summer, which prompted renewed requests from both Democratic and Republican Congress members for the Bureau to provide the Committee with answers to questions and internal documents related to its ECOA enforcement program regarding indirect auto finance.
The CFPB documents cited in the report paint a picture of a staff that is well aware of the weaknesses of its legal theory and of the fact that its evidence is based on a seriously flawed methodology. One hardly knows whether to be pleased or dismayed that the CFPB staff discussed these problems candidly in memoranda to Director Cordray-pleased because the staff was frank about the weaknesses in both the law and the evidence, and dismayed because the CFPB proceeded with the cases nonetheless.
From my many years of enforcing the ECOA at the Federal Trade Commission, I know the Commission would have refused to authorize an enforcement action if presented with staff admissions of such a flimsy case. Government agencies do not like to lose cases, and they have plenty of strong cases to prosecute. So, why did Director Cordray sign off on these fair lending cases? The most troubling part of this sordid tale appears in a final decision memorandum seeking authority for the enforcement action against Ally Financial Inc. and Ally Bank. The staff conceded that the Bureau's claims against Ally presented "issues," such as the use of proxies and the disparate impact theory, "that would pose litigation risk of enough significance to merit serious consideration prior to taking administrative action or filing suit in district court."
Yet the staff recommended, and Director Cordray approved, the enforcement action because it had "powerful" political leverage on Ally, making such weaknesses in its case likely irrelevant. "Ally might have a powerful incentive to settle the entire matter quickly without engaging in protracted litigation," the Bureau staff wrote in reference to Ally's pending application for financial institution holding company status. If the Federal Reserve Board did not approve Ally's application by December 24, 2013, Ally would be forced to divest its insurance and used-car remarketing operations. "Settlement of the Bureau's fair lending investigation was a prerequisite for Ally's status change," the Committee staff's report observed. Ally signed the settlement, costing it $98 million, five days before its deadline to achieve financial institution holding company status. In short, the agencies had Ally over a barrel, so Ally was the perfect target for the ECOA enforcement action that the CFPB hoped would change the way banks and finance companies compensate auto dealers for the credit contracts they buy from dealers.
The Committee staff's report details the significant shortcomings of both the Bureau's legal theory and the evidentiary basis for the CFPB's attack on dealer compensation. Any one of these shortcomings would be sufficient to defeat the CFPB's fair lending case against indirect auto creditors. Let's take a quick tour of them.
Disparate Impact. The CFPB's legal case relies on a controversial theory of disparate impact, which challenges a practice that is discriminatory in effect because it has a disproportionately negative impact on a prohibited basis, even though the company has no intent to discriminate and the practice appears neutral on its face. Here are the problems with disparate impact.
- Supreme Court decisions cast doubt on whether a disparate impact theory is valid under the ECOA. The statutory language that recent Supreme Court decisions have relied on to recognize disparate impact liability under other laws is absent in the ECOA.
- Dealer "discretion" in pricing is not a finance company "policy or practice" required for disparate impact liability. CFPB internal memoranda acknowledge the Supreme Court's rejection of "discretion" as policy giving rise to potential disparate impact.
- Factors other than dealer discretion are likely to be the real cause of markup differences. The Supreme Court has stressed there must be a "robust causality" between the challenged practice and the adverse impact on minorities. The Bureau has refused to control for factors such as credit tier, term, new/used, and LTV that might explain the pricing differences, although the Bureau staff acknowledged internally that refusing to control for race-neutral factors might be unfair to assignees of auto finance contracts.
- Indirect auto creditors have a legitimate business justification for current dealer compensation policies. CFPB internal memoranda acknowledge that indirect creditors that have gone to flat fees have lost significant market share and refer to the compensation shift as "corporate suicide." The memos acknowledge that Honda and other indirect creditors "may be able to convince a court that they had a legitimate business justification for their policies."
Indirect auto finance companies as ECOA "creditors." Under the ECOA, an assignee of a credit contract is a "creditor" only if it participates in the credit decision.
- Often, assignees are not ECOA "creditors." The CFPB acknowledges internally that indirect lenders are not ECOA creditors in spot deliveries because the terms are set before the assignee's involvement. Even in other transactions, the dealer often receives many wholesale rate quotes and may have set the retail rate based on factors other than the assignee's quote. Banks and finance companies generally do not know if a transaction was a spot delivery or why the dealer contracted with the customer for a specific rate. It is the Bureau's burden, as plaintiff, to prove that the assignee was an ECOA creditor, but the Bureau has no way of doing this in any given transaction.
Use of BISG to proxy race and national origin. The Bureau has staunchly promoted the use of BISG methodology, which combines the odds of being a member of a minority group based on surname and geographical location. Yet it has admitted, in internal memos, that other proxy methodologies have been proven more reliable. In fact, internal memoranda argue against revealing the CFPB's proxy methods because companies targeted for enforcement will "show how our methods are inferior to other proprietary proxies."
- BISG proxy methods are inferior to proprietary data. The Bureau staff defended the use of BISG over more accurate proxies by arguing that companies could save money by not having to buy proprietary products. As the House Committee staff noted, the concern about saving regulatory costs is "undoubtedly cold comfort to creditors publicly labeled as engaging in racial discrimination brought by the Bureau using a methodology it knew to be faulty and unreliable."
- BISG proxy methods are highly unreliable, especially for African Americans. The CFPB's 2014 public white paper on proxy methodology admitted a 20% overestimation of African Americans. But its internal memos acknowledged that data from an outside source demonstrated that for every 100 African-American consumers, BISG could identify only 19 of them as African Americans. Statisticians refer to these errors as false negatives; there are also whopping false positives. "Out of 100 applicants that are identified by the proxy methodology as African-Americans, only 54 of them are actually African-Americans according to the HMDA data," conceded an internal memo.
The House Committee on Financial Services staff report provides a rare look into the inner workings of the CFPB. That picture is grim. In a sense, nothing the report revealed is different from many in the industry have suspected. But the frank admission of the weaknesses in the Bureau's legal theory and evidence surprised me. It suggests an adherence to an approach the staff acknowledges is seriously flawed. Although I expected to feel anger on reading the Committee staff's report, I felt sad. For more than a decade, I directed the FTC's enforcement of the ECOA. Like most creditors, I believe that a commitment to nondiscriminatory practices is not only good for business but also required by our legal system and our values. I also believe that the role of government is to do what is just, not to "win" at any cost.
Should the reaction to the Committee staff's report be outrage? I think of Horatio's response to Hamlet that the ghost of Hamlet's father had "a countenance more in sorrow than in anger." Read the report. If your response is anger, I understand that. Right now, I'm feeling sorrow.