Appeal of Potential Violation of the Equal Credit Opportunity Act and the Fair Housing Act - (First Quarter 1999)
A large bank filed a formal appeal concerning a potential violation of the Equal Credit Opportunity Act (ECOA) and the Fair Housing Act (FHAct). The bank received correspondence stating that the OCC had a reason to believe that the bank had engaged in a pattern or practice of discrimination by treating black applicants for home improvement loans less favorably than similarly situated white applicants. The bank did not agree that it engaged in a pattern or practice of discrimination or that a violation of the ECOA and the FHAct occurred.
1) The bank believed that the OCC's decision was based solely upon the results of a multivariate regression analysis. The bank asserted that the multivariate analysis ignored the inconclusive results of the OCC's univariate analysis and comparative file review, along with the two bank-prepared statistical analyses. The two bank-prepared analyses concluded there was no discrimination.
2) The bank also believed that the statistical regression analysis was flawed because the OCC model:
3) The bank asserted that the OCC relied solely on a multivariate regression analysis, which is inconsistent with both its fair lending examination procedures and its published agency staff commentary on appropriate examination practices.
4) The bank felt that the OCC finding directly conflicted with safety and soundness considerations.
5) Finally, the bank was of the opinion that the OCC had not provided the bank with any comparably unqualified white applicants who received credit.
To evaluate the bank's fair lending performance, the OCC conducted a comparative file analysis and univariate and multivariate regression analyses using statistical modeling. The file analysis compared the treatment of black and white applicants for home improvement loans. The regression model also compared the treatment of black and white applicants. In this case, the findings of the comparative file review were inconclusive, yet the OCC was still troubled by inconsistencies noted with loan outcomes among white and black applicants.
The bank used credit scoring in its loan decision process. However, since bank underwriters did not strictly adhere to the credit score analysis, nor written underwriting guidelines for credit scoring, and used multiple and varying criteria for making approval/denial decisions, the OCC concluded that statistical sampling and modeling was the most beneficial way to address the concerns identified during the comparative file review. The multivariate regression analysis revealed that, after consideration of other factors, race was a significant factor in the probability of denial of credit-scored home improvement loan applicants. The analysis indicated that black applicants were approximately four times more likely to be denied credit than similarly situated white applicants. It was determined that a pattern or practice of discrimination may exist as black applicants appeared less favorably treated than similarly qualified white applicants.
The bank hired an outside consultant to conduct its own statistical regression analysis. The bank's consultant contended that the OCC multivariate regression analysis was flawed and should be discounted. The consultant also performed its own regression analysis, using a different approach, and concluded that there was no potential discrimination.
After evaluating all the evidence, including the bank-prepared statistical analyses and several bank responses, the OCC concluded there remained a reason to believe that the bank had potentially engaged in a pattern or practice of discrimination by treating black applicants less favorably than similarly situated white applicants for home improvement loans. Therefore, the OCC concluded it was obligated to refer this matter to the U.S. Department of Justice and to notify the U.S. Department of Housing and Urban Development.
The ECOA, 15 USC 1691(a), prohibits a creditor from discriminating against an applicant on a prohibited basis regarding any aspect of a credit transaction. The implementing regulation 12 CFR 202 (Regulation B) defines prohibited basis as:
Prohibited basis means race, color, religion, national origin, sex, marital status, or age (provided that the applicant has the capacity to enter into a binding contract); the fact that all or part of the applicant's income derives from any public assistance program; or the fact that the applicant has in good faith exercised any right under the Consumer Credit Protection Act or any state law upon which an exemption has been granted by the Board (12 CFR 202.2 (z)).
The FHAct, 42 USC 3605(a), prohibits a lender from discriminating on a prohibited basis in a residential real-estate-related transaction (including the making of loans) or in the terms or conditions of the transaction. The implementing regulation, 24 CFR 100.130, states it shall be unlawful for any person or entity engaged in the making of loans or on the provision of other financial assistance relating to the purchase, construction, improvement, repair, or maintenance of dwellings, or which are secured by residential real estate, to impose different terms or conditions for the availability of such loans or other financial assistance because of, among other factors, race.
While the ECOA and the FHAct do not define the term "pattern or practice," the "Interagency Policy Statement on Discrimination in Lending" offers guidance on the meaning of a pattern or practice. The policy statement states that "repeated, intentional, regular, usual, deliberate, or institutionalized practices will almost always constitute a pattern or practice" of lending discrimination but "isolated, unrelated, or accidental occurrences will not." In assessing whether a pattern or practice exists, the OCC considers the totality of the circumstances, including the following factors:
This list of factors is not exhaustive and whether the OCC finds evidence of a pattern or practice depends on the egregiousness of the facts and circumstances involved. Each inquiry is intensively fact-specific and there is no minimum number of violations that will trigger a finding of a pattern or practice of discrimination.
The ombudsman thoroughly reviewed the issues highlighted in the appeal letter, OCC documents, and analyses of the issues raised both prior and subsequent to the appeal submission. Discussions were held with the bank's outside attorney, bank personnel, and appropriate OCC staff. The ombudsman acknowledged the bank's arguments, but was not persuaded that the supervisory office conclusion was in error or that the OCC statistical sampling and modeling techniques lacked integrity.
Therefore, the ombudsman concluded that sufficient information existed to support the examination conclusion that there was a reason to believe that the bank engaged in a pattern or practice of discrimination in violation of the ECOA and the FHAct and, as such, remanded to the OCC's supervisory office the matter of referral to the U.S. Department of Justice and notification to the U.S. Department of Housing and Urban Development.
The ombudsman also addressed the bank's concern regarding the conduct and planning of the examination. While the examination took significantly longer than it should have and contributed to misunderstandings and questions regarding OCC examination goals and objectives, the ombudsman found that both the bank and the OCC were responsible for the time it had taken to complete this examination.