Anyone who has been following enforcement activity in consumer financial services knows that fair lending is a key focal point for federal regulators, with recent huge monetary settlements and more likely to occur. It’s not just a large bank issue; community banks have also been targeted. What can bank boards of directors and management do to avoid or mitigate such regulatory actions? Identify the risks and address and resolve them before they become big risks.
Over the past year, the U.S. Department of Justice (DOJ) has entered into the three largest fair lending settlements in history – all of which carried multimillion dollar price tags to resolve allegations of discrimination in retail and wholesale mortgage lending by major lenders. The DOJ, Consumer Financial Protection Bureau (CFPB), U.S. Department of Housing and Urban Development (HUD) and prudential banking regulators (FDIC, Federal Reserve, and OCC) are all aggressively pursuing, and in some cases actively soliciting, fair lending cases. Many of these cases are based on the controversial “disparate impact” theory of discrimination, which narrowly escaped review by the U.S. Supreme Court in early 2012, to determine whether this legal theory is even cognizable under federal fair lending laws. Notwithstanding the unresolved question over the use of disparate impact in the fair lending context, the federal banking regulatory and enforcement agencies have uniformly stated that they will prosecute fair lending cases under this legal theory.
Fair lending allegations are not limited to large national retail banks. Community banks have also been targeted in these investigations and complaints. For example, in June 2011, DOJ reached a settlement with Nixon State Bank to resolve allegations that the bank had violated the Equal Credit Opportunity Act (ECOA) by charging higher prices on unsecured consumer loans made to Hispanic borrowers, which required Nixon to pay approximately $100,000 in restitution. Nixon State Bank did not maintain written loan pricing guidelines for its unsecured consumer loans; instead, the bank’s loan officers were granted broad discretion in handling all aspects of the unsecured consumer loan transaction. DOJ alleged that this policy had a disparate impact on Hispanic borrowers. In a more recent case from September 2012, Luther Burbank Savings Bank (discussed below) agreed to settle with DOJ for $2 million for setting a minimum residential mortgage loan amount that adversely impacted African American and Hispanic borrowers.
Regulators or plaintiffs typically demonstrate “disparate impact” through a burden-shifting test that begins with an allegation that a lender applies a facially neutral policy or practice consistently to all credit applicants, but the policy or practice has a disproportionately adverse impact on members of a group protected under ECOA or the Fair Housing Act, the federal fair lending laws (protected class group).* If the first prong of the analysis is satisfied, the burden shifts to the lender who must prove that there was a legitimate, non-discriminatory business justification for the policy at issue. If this prong is satisfied by the lender, the burden shifts back to the regulator or plaintiff to prove that there is a less discriminatory alternative that achieves the same result. Under the disparate impact theory, no evidence of intentional discrimination is required.
Example of disparate impact:
- A community bank maintains a residential mortgage lending policy that precludes mortgages on homes that are more than 50 years old. The area of town where homes tend to be over 50 years in age is predominantly Hispanic. As a result, the bank’s lending policy, which appears facially neutral, has a disproportionately adverse impact on Hispanic borrowers as a protected class group and has the effect of restricting access to credit for Hispanic borrowers. The bank is unable to offer a legitimate, nondiscriminatory business justification for the policy.
Set forth below are six “red flags” for fair lending risk that you should be aware of so you can determine whether your institution is taking appropriate action.
Red Flag #1: Disparate impact that may not be readily apparent.
- Potential disparate impact is often difficult to detect, so it is important to proactively review your institution’s policies and procedures to determine if a risk of disparate impact may exist.
- To mitigate the risk of a disparate impact case against your institution, review marketing, advertising, underwriting, pricing, compensation and other related policies and procedures to evaluate whether they could have a disproportionately adverse impact on a protected class, even if the policy or procedure is facially neutral. If there is a potential disparate impact, you should determine whether the level of risk merits performing a formal, documented disparate impact analysis, which would include a determination of whether an adequate business justification for the policy or procedure exists and whether there is a less discriminatory alternative that could achieve the same results.
Red Flag #2: Lending policies that require a minimum loan amount.
- Underwriting guidelines that impose minimum loan amounts have been an area of fair lending scrutiny. Although these policies are facially neutral, they have been the subject of numerous fair lending complaints and settlements because of the perception that minimum loan amounts have a disparate impact on minority borrowers who are more likely to live in neighborhoods and communities with lower property values and, therefore, require lower loan amounts.
- For example, on September 12, 2012, DOJ entered into a $2 million settlement with Luther Burbank Savings Bank for setting a minimum loan amount of $400,000 for its single-family residential mortgage lending program, which resulted in very few mortgage loan originations to African American or Hispanic borrowers. Despite being aware that the low level of lending to minorities was attributable to the minimum loan amount, the bank continued to enforce the policy, which restricted credit to those protected class groups.
- Although most complaints involving minimum loan amounts involved mortgage lending, the same principle applies to all consumer lending if the minimum loan amount makes a product inaccessible to members of a protected class. To understand your fair lending risk in this area, you should review your lending policies and procedures to identify any minimum loan amounts.
Red Flag #3: Credit overlays to underwriting guidelines.
- Another area of fair lending scrutiny by both regulators and consumer advocacy groups has been credit overlays that go beyond the minimum credit standards or underwriting guidelines established by Fannie Mae and Freddie Mac (for conforming loans) and HUD (for Federal Housing Administration (“FHA”) guaranteed loans). Examples of such practices that have resulted in fair lending complaints, investigations and settlements include:
- Imposing higher minimum FICO scores for FHA loans than what is required by HUD;
- Requiring female applicants on maternity leave to return to work before closing a mortgage loan (even when the applicant satisfies the income requirements while she is on maternity leave); and
- Requiring applicants receiving permanent disability benefits from the Social Security Administration to provide proof of continuation of the disability with a letter from a physician attesting to the nature of the disability and the expected duration of the disability.
- To understand your fair lending risk in this area, you should review your underwriting policies and procedures to identify any credit overlays and assess the likelihood that such modifications would have a disparate impact on members of a protected class. Where risk is identified, determine whether a formal, documented disparate impact analysis is appropriate.
Red Flag #4: Lack of periodic risk assessments to evaluate level of consumer compliance risk (including fair lending risk).
- Federal regulators expect depository institutions of all sizes to conduct periodic risk assessments of its operations to evaluate its level of compliance with applicable laws and regulations. Not conducting such risk assessments may cause the regulators to do a more in-depth assessment during examinations and raise questions about a bank’s commitment to consumer compliance. Among the areas that should be covered in these periodic risk assessments is fair lending. Fair lending risk assessments should not only include the technical requirements under ECOA and Regulation B (e.g., adverse action notices, spousal signature), but also the prohibitions on discriminatory practices.
- The methodology for the risk assessment should be reviewed periodically to include changes to applicable laws and regulations and your own operations, products or services. The fair lending risk assessment should be used to ensure that proper controls are in place (especially for high-risk areas) and to help allocate your compliance staff, budget and resources.
Red Flag #5: Use of discretion in consumer lending.
- Policies and procedures often permit loan officers, underwriters or agents (e.g., auto dealers) to exercise discretion in pricing or credit decisioning. Discretionary practices may also arise in areas such as account maintenance (e.g., fee waivers), collections, and loan modifications.
- Wherever discretion is afforded in the loan origination process, it is important to ensure that appropriate controls are in place and to perform ongoing monitoring and testing of the discretionary activity. Controls, monitoring and testing help ensure that discretion is exercised consistently; if not, appropriate remedial action can then be taken so that the consumer is not harmed.
Red Flag #6: Failure to evaluate fair lending risks and controls for non-HMDA loans.
- Although fair lending exams and enforcement actions have historically focused on mortgage lending, regulatory and enforcement agencies have begun to direct their fair lending scrutiny to non-HMDA lending (e.g., indirect auto lending, student lending, credit cards, and unsecured consumer lending). In the absence of any HMDA data to serve as the basis of a fair lending claim, the government has relied on “proxy” data (e.g., geocoding, surname) to conduct quantitative analyses.
- Although it is unclear whether it is a regulatory expectation to conduct your own fair lending testing of non-HMDA loan data using proxy data, it is important to understand your fair lending risk in this area. If your institution conducts periodic fair lending risk assessments, you could rely on the findings to determine which non-HMDA products present the greatest fair lending risk. Depending on the level of risk and availability of suitable proxy data, you can determine whether it is appropriate to perform some preliminary testing. At a minimum, all institutions should review their preventive and detective controls for its non-HMDA lending channels to determine whether proper controls are in place.
All of the fair lending risks noted above lead to the overarching recommendation that institutions develop a comprehensive and robust Fair Lending Program that is part of its compliance risk management program. Depending on the size and risk profile of your institution, it may be appropriate to designate a Fair Lending Officer to manage and oversee implementation of your Fair Lending Program. Whether your institution has a Fair Lending Officer or other compliance professional assigned to manage your Fair Lending Program, the individual should be vested with the experience, resources and authority to effect change when needed.
The board of directors should provide oversight for your institution’s Fair Lending Program and its overall compliance risk management program. At minimum, the board should dedicate one of its standing committees, such as the audit committee or risk management committee, to routinely discuss matters and receive reports related to consumer compliance risk. That committee’s agenda should also periodically include items related to fair lending issues, risks and controls.
* ECOA and the Fair Housing Act prohibit discrimination in lending on specified bases. Both ECOA and the Fair Housing Act prohibit discrimination on the basis of race, color, religion, national origin, and sex. ECOA further prohibits discrimination on the basis of marital status, age, receipt of public assistance, or good faith exercise of a right under the Consumer Credit Protection Act. The Fair Housing Act includes familial status and handicap as additional prohibited bases.CFPB, Disparate Impact, DOJ, ECOA, Fair Housing, Fair Lending, HUD