• Insights

  • Home Equity Lending: A Fair Lending Game Changer

    Published December 22, 2018

    Originally published in MBA Insights on Dec. 10, 2018

    On the eve of reporting expanded Home Mortgage Disclosure Act data, there is justifiable anxiety around these changes and how the new data will be a reflection on each reporting institution. Perhaps the biggest change in the HMDA data is the morphing of the loan purpose (Home Purchase, Refinance, Home Improvement) reporting requirement to the lien-based requirement for consumer transactions. This subtle move is expected to significantly increase the overall number of records that are reported within the industry.

    Not only does this modification require the reporting of all home equity lines of credit that were previously optionally reported, but the change will also incorporate more transactions that previously did not meet the old loan purpose definitions. For example, previous home equity loans (or lines) that were not for the purpose of home improvement and did not satisfy or replace an existing lien were not HMDA reportable. This would include loans and lines for the purpose of debt consolidation, education expenses, automobile purchases and the countless other reasons people take a loan or line against the equity in their home.

    The impact of these additional transactions goes beyond the increased HMDA burden. There are important fair lending considerations, and these additional records are likely to provide greater insight into the controls and credit practices of each entity reporting these applications. This is because these transactions are not typically investor-driven, so the institutions generating these loans could potentially exercise a greater degree of discretion when making credit and pricing determinations. Coupled with new 2018 HMDA reporting data fields (such as interest rate, points & fees, credit score, debt to income and loan to value), the added records could provide regulators and community groups more data for finding apparent differing treatment on a prohibited basis for institutions that generate home equity lending.

    Although there has been exposure to the lending practices of institutions through the reporting of HMDA data in the past, the introduction of the reporting of an entire product set has not occurred since the introduction of HMDA data reporting. The introduction of the Home Mortgage Disclosure Act in 1975 was an absolute game changer with respect to fair lending enforcement and has resulted in numerous changes in the industry. One can only speculate as to the degree in which this data will reveal fair lending issues and change the game within home equity lending.

    Not being HMDA reported in the past means that (in most cases) these transactions were not included in CRA performance evaluations. Thus, one could speculate that the percentage of these originations being generated in low- or moderate-income areas was not a focus for banks or other lenders. And because low and moderate-income census tracts overlap with majority minority census tracts, one could further speculate that there could be an increase in redlining issues for those that participate in home equity lending. This is because regulators typically view an institution's redlining risk as a percentage of applications taken in minority census tracts compared to the peer percentage of applications in those same areas. If home equity lenders are not focused on this distribution of home equity loans and lines, redlining risks are likely to increase.

    Additionally, one could speculate that without the guiding structure of investors such as Fannie Mae and the Federal Home Loan Mortgage Corp., home equity lending may be subject to greater discretion and less consistency in the underwriting and pricing. Without this secondary market infrastructure guiding underwriting and pricing decisions, some lenders may be more likely to grant exceptions to written policies as the product is typically "held for investment." This type of flexibility is great for making loans, but also can result in fair lending issues as only a few exceptions to policy can result in statistically significant differences.

    Clearly there are action steps that can be taken now to reduce the risk exposure associated with this new data. These steps are not new, but in the context of this new information it is a good idea to revisit sound practices that should be a part of every fair lending risk management program. Action steps start with awareness, communication, training and monitoring. This activity should be accompanied by robust data analytics so that an institution can better understand the risks and identify opportunities for risk mitigation.

    Below are three key focal areas for reviewing the home equity business to help identify and assess this evolving fair lending risk:

    Review Home Equity Policies and Procedures

    A review of the home equity underwriting and pricing policies and procedures may seem overly basic, but as this home equity data will be newly exposed to the world of HMDA, it is a good idea to review the policies and procedures associated with this business practice. For determining the basis of what is sound, a simple reference can be made back to Regulation B--is a judgmental or scoring system in place? Are the underwriting or pricing factors "empirically derived, demonstrably and statistically sound?"

    Perhaps even more simplistic, it makes sense to review policies and procedures for vagueness or opportunities to misinterpret any guidance. A good practice is to review for the word "should" within the policies. Take the example guideline of, "an applicant should have a credit score of 620 or greater." What does that really mean, and how is it applied? Can an underwriter approve a loan with a credit score below 620? Could this decision be based on the risk tolerance of the individual underwriter and result in differing treatment on a prohibited basis within the portfolio?

    Policies and procedures should also be reviewed with respect to inflated minimum loan amounts or high minimum credit scores to qualify. Consider, for example, a minimum line of credit or loan amount requirement of $25,000. Whether this is an acceptable minimum depends on the context. In a market such as Detroit, Michigan, where median home prices have been between $30,000 and $40,000 over the past year, a minimum home equity loan or line amount of $25,000 or even $10,000 may have a significant impact on the ability to originate loans or lines of credit and create fair lending issues with equal access to credit and redlining claims.

    Lastly, there should be a review of the policies that govern exceptions. To reduce fair lending risks associated with home equity, or any other line of business, detailed policies should exist to establish a structure around this activity. Effective policies governing exceptions should be detailed and limit the authority of who can make exceptions, escalate levels of authority, and quantify the maximum number or percentage of exceptions that can be made. All exceptions should be fully documented so that the business justification is obvious, and exceptions should be monitored for trends with reporting to senior management.

    Analyze the Data for Underwriting and Pricing Issues

    As inclusion of Home equity data is new to the institution, it should be analyzed in a manner that can identify pricing and underwriting issues. Regression analysis is a good way to pinpoint issues within larger pools of applications (i.e. datasets). It is important not get confused with this process. If underwriting and pricing policies differ from the other HMDA reportable applications, then separate analyses should be conducted for these applications to identify potential fair lending issues.

    The timing of the analysis is also important. All too often, institutions perform fair lending analysis only after the submission of the HMDA data. Unfortunately, once the data has been submitted, there is nothing a manager can do to change that data. However, by managing the fair lending risks periodically throughout the year, the data can provide indications of risk that can be corrected before the data is submitted. Is there a loan officer providing too many exceptions? Is there an underwriting policy that is vague or open to interpretation? By correcting these issues prior to reporting the data, disparities and statistically significant findings could be minimized or avoided. This tactic is much easier than trying to explain away a bias in the data that has already been reported.

    Know the Redlining Risk Factors

    In recent years, there have been a significant number of fair lending actions taken against organizations for redlining. The growing foundation of analysis for these actions has been the HMDA institutional data collected by the CFPB (formally collected by the FFIEC). Statistically significant differences between an institution and this "peer" data within an area (such as an MSA) could result in a claim of discrimination. The measure has typically been the percentage of loan applications taken by an organization within minority areas as compared to the collective percentage of business obtained by the peer group in that same area. This can include the minority community as a whole or individual minority neighborhood classifications, such as majority Hispanic census tracts.

    As stated, the changes that result in the inclusion of home equity loans or lines within the HMDA data for 2018 and beyond will likely result in a significant increase in the overall reportable records. For organizations that will see a large increase in their reporting volume, it is highly recommended that analysis be performed on these home equity transactions to determine "where" these loans are being made. This approach will allow for the understanding of the impact on redlining risks resulting from the inclusion of these additional applications into the HMDA data. Institutions should recognize if there are gaps in the number of applications taken within minority communities as compared with peer organizations.

    As important to the analysis of redlining is the focus on the marketing efforts that are made within each community in which an organization operates. This includes both those communities that are considered "in" and "out" of the perceived footprint of the lender. Historically, redlining claims focused on banking branch locations and the construction of the CRA assessment areas. Now the process has evolved to include any market in which an institution is conducting business, as illustrated by the use of the Reasonably Expected Market Area that is emphasized by some regulators.

    Analysis of application information and marketing efforts go hand in hand. Where there is a void in applications in minority areas, marketing efforts need to be reviewed. A lack of effort (typically in terms of marketing or advertising) to close the gaps in lending identified in minority areas could be perceived as a willful attempt to exclude an area. Because of this, it is critical to understand how, where and to whom the home equity product is advertised and what the home equity application percentages look like in minority communities.

    What is important to remember is that the expanded HMDA data reporting requirements represent change. As in the past when HMDA data was introduced and as subsequent changes have been made to the HMDA data reporting requirements, changes were made in the industry to reduce fair lending risks. On the eve of the reporting of a new product set, once more there can be speculation that changes will follow to reduce fair lending risks. While the practices of managing fair lending risks remains the same, the introduction of a new product into HMDA reporting is likely to change the home equity "game" in the future.

  • Please take a moment and tell us what you think of our content.