Updates from the DOJ | By Peter Weinstock and John Delionado
In response to the recent housing and foreclosure crisis, the Department of Justice (DOJ) has made fair lending issues a top priority. In 2010, the DOJ established a Fair Lending Unit in the Civil Rights Division’s Housing and Civil Enforcement Section in order to devote more resources to fair lending enforcement.
In the attorney general’s 2010 Annual Report to Congress Pursuant to the Equal Credit Opportunity Act dated April 5, 2011, the DOJ indicated that in 2010 it had 60 active investigations and received 49 referrals from regulatory agencies involving a possible pattern or practice of discrimination – more than the DOJ received in the last 20 years. To compare, from 2001 through 2008, the DOJ received only a combined total of 30 referrals involving race or national origin discrimination. Of those 49 referrals in 2010, 26 involved allegations of discrimination based on race or national origin. The increased number of referrals to the DOJ is a direct result of deliberate and aggressive steps taken by these financial institution regulators within the fair lending context.
The DOJ is targeting specific discriminatory lending practices, including, but not limited to:
Discrimination in the underwriting or pricing of loans.
Redlining through the failure to provide equal lending services to minority neighborhoods.
Reverse redlining through the targeting of minority communities for predatory loans.
Steering minority borrowers into less favorable loans.
Marital status, gender and age discrimination in lending.
The discriminatory lending practices at issue can take the form of either intentional discrimination (e.g., discriminatory underwriting criteria, such as requiring that borrowers speak English) or discrimination manifested by the disparate treatment of borrowers (e.g., statistical variations showing that borrowers of a protected class received less favorable loan terms when compared to borrowers of a non-protected class).
Many of the investigations that have been referred from financial institution regulators to the DOJ have been based upon a review of data available under the Home Mortgage Disclosure Act (HMDA). This HMDA data has allowed regulators to isolate issues relating to the pricing of mortgage loans, the fees of mortgage loans and the difference in loan products, as well as issues relating to the location of the borrowers, otherwise known as red-lining. However, the review by financial institution regulators is not limited to only HMDA reported loans, but also includes unsecured and secured (whether by vehicle or some other collateral) loans.
Notwithstanding the number of issues that can potentially arise from a fair lending examination, we think that there are a few practical steps a financial institution should take to help prepare itself for the examination and minimize the potential for referral to the DOJ and an enforcement action.
We generally divide the fair lending examination into three phases. The first phase involves the regulatory examination process which includes the extraction of loan data, interviews with the financial institution’s personnel and the formal exit interview conducted by the financial institution regulator. The second phase includes the financial institution’s response to a suspected fair lending issue, otherwise known as responding to a “15-day” letter. The third phase of the fair lending process involves production and negotiation with the DOJ in the event a referral has been made by the financial institution regulator. In each phase the financial institution is given the opportunity to correct any deficiencies and demonstrate to the regulators that it is not engaging in discriminatory practices. We discuss some of the steps that a financial institution can take in each phase below:
The First Phase: Examination Process, Extraction of Data and Interviews
During the first phase of the fair lending examination, it is critical for the financial institution to understand what the financial institution regulators are looking for. First and foremost, they are looking for accurate data. In order to assess whether or not the financial institution has accurate data available to it, the financial institution should go back and do a spot review of its electronic data against some of the actual loan files.
We have often found that a financial institution’s electronic system automatically updates and replaces key elements of the loan data without the financial institution’s knowledge. For instance, when loans are renewed and a new FICO score is recorded the financial institution’s electronic system often retains only the new FICO score, thereby overriding the historical score(s) that may have been previously recorded over the course of the lending relationship. The impact of such data changes can be significant and result in inaccurate analysis of the data by the financial institution regulators.
The financial institution should also compile a complete list of all the materials and factors that are used by the financial institution in setting pricing. An incomplete list will not serve the financial institution’s best interests, potentially resulting in an inaccurate assessment by the regulator. Unless the financial institution informs the regulators that it uses a particular risk factor in making and setting its prices, the regulators will not know it and will not use that risk factor in analyzing the data.
The financial institution regulators will also seek to conduct interviews of the financial institution’s personnel, in particular the loan officers. It is important that the financial institution’s loan officers are familiar with the policies and processes used at the financial institution to approve loans to ensure that they are not communicating inaccurate information to the regulators.
It is critical that the loan officers convey all of the factors and materials actually considered and used by the loan officers in setting price.
At the time of the exit interview, the financial institution regulators will express some of their views on the financial institution’s policies and the results of the personnel interviews. The regulators will also ask the financial institution’s management to validate the regulators’ findings and/or address any issues the financial institution may have with the conclusions reached by the regulators.
This is the financial institution’s opportunity to clear up any misconceptions and ensure that the regulators are aware of all the factors that should be considered as part of the fair lending examination. It does not serve the financial institution’s best interest to oversimplify things to please the regulators. If the financial institution engages in a multi-staged process in setting pricing that allows for discretion within certain parameters, the financial institution should expressly say so.
Ultimately, the financial institution regulators will compile the information from the written documentation and interviews to create a package of information for the agency’s fair lending task force. This information will be reduced into a series of variables in a regression analysis that the financial institution regulators will use to assess whether or not there is a disparate impact on protected classes of individuals.
The regulators are obligated to include in this regression analysis all factors that the financial institution considers in setting pricing. That is why it is essential that the financial institution provides the regulators with a complete list of all the factors considered by the financial institution. The goal of the regression modeling analysis is to control for all legitimate pricing factors and test whether the financial institution is applying those legitimate pricing factors in a uniform way such that members of protected classes are treated the same as members of non-protected classes.
In the event that the regression analysis comes out with a statistically significant differential between the protected and non-protected classes that is not explained by legitimate pricing factors, the regulators will likely share these results with the financial institution and ask the financial institution for an explanation in writing.
The Second Phase:
Responding to a ‘15-Day’ Letter
In the event that a financial institution receives a “15-day” letter from the financial institution regulators conveying that the regulators have seen an unexplained differential in the price between the protected group and non-protected group, the financial institution should immediately begin preparing a detailed response. Among the things a financial institution should consider is retaining a firm that has the ability to run its own regression analysis to test the appropriateness and accuracy of the financial institution regulator’s model.
The financial institution should request from the regulator not only the back-up documentation used to reach the financial institution regulators’ results, but also the back-up “codes” and the “model” that the regulators purported to use. The models that the regulators use are fairly similar and involve an attempt to normalize a number of legitimate pricing factors within a single regression model per portfolio of loan.
The financial institution should first review the data the financial institution regulators used to build their model to satisfy itself that the data used is what the financial institution actually provided and is complete and accurate. If the data is inaccurate, then the results of the regression analysis, and therefore, the conclusions by the financial institution regulators will also necessarily be inaccurate.
The financial institution should then seek expert guidance as to whether the model the regulators have built accurately captures all the factors and information the financial institution conveyed to the regulators during the examination process culminating with the exit interview. As with inaccurate data, an inaccurate model can also lead to an inaccurate assessment of the financial institution’s lending practices.
The Third Phase: Referral to the DOJ
In the event that the financial institution regulators find a potential pattern or practice of discrimination, the regulators will refer the matter to the DOJ. Upon receiving a referral, the DOJ must determine whether to file a lawsuit in federal court or return the matter to the financial institution regulators for administrative enforcement.
The DOJ will receive from the financial institution regulators all of the compiled information from the fair lending examination along with the results of the regression analysis. The DOJ will contact the financial institution and initiate its own investigation and analysis, resulting in the production of additional information by the financial institution. During this phase, as with the earlier phases, the financial institution must engage in a process of validating the accuracy of the additional data produced to the DOJ.
The financial institutions must normalize the additional data to make sure that this data is consistent with the information previously provided and consistent with the actual factors employed by the financial institution in making pricing decisions. And, as with the second phase, the financial institutions should engage expert consultants to analyze the data and determine the affects, if any, the additional data may have upon the models used by the DOJ. After a full analysis of the data is completed, the financial institution should assess whether a meeting with the DOJ would be beneficial to make an affirmative presentation.
In sum, the DOJ’s current emphasis on targeting financial institution’s for potential fair lending issues is unprecedented. Fair lending investigations and enforcement actions are nuanced and involve a unique focus upon statistical regression analysis in assessing whether or not there is a disparate impact on protected classes of individuals. As such, great care must be taken by financial institutions to ensure that both the regulators and the DOJ are provided a complete and accurate picture of the lending process and pricing criteria employed by the financial institution. Further, engaging consultants familiar with the statistical models and the investigation practices employed by the regulators and the DOJ is highly recommended. ■
Peter G. Weinstock is the practice group leader of the financial institutions corporate and regulatory practice group at Hunton & Williams LLP. He writes and speaks frequently on topics of interest to community bankers. He may be reached at (214) 468-3395 or firstname.lastname@example.org. John J. Delionado is s a partner in the litigation and intellectual property group at Hunton & Williams LLP in Miami, Florida. He may be reached at (305) 536-2752 or email@example.com.