Redlining Evaluating Risk and Defending Claims Melanie Brody Partner Mayer Brown mbrody@mayerbrown.com Brian Clark Senior Manager Ernst & Young Brian.Clark@ey.com
Speakers Melanie Brody Partner Mayer Brown mbrody@mayerbrown.com Brian Clark Senior Manager Ernst & Young Brian.Clark@ey.com 2
Overview Background/redlining defined How should lenders evaluate redlining risk? Assess institutional organization Statistical analyses How can lenders defend redlining investigations? Legal defenses Factual defenses Statistical defenses 3
Background/Redlining Defined Redlining historically referred to the practice of denying financial services usually residential mortgage loans to residents of certain geographic areas based on the racial or ethnic makeups of those areas: [D]eclaring black areas off-limits for mortgage lending [is] a practice otherwise known as redlining. Press Release, U.S. Department of Justice, Justice Department Obtains Unprecedented Settlement from D.C. Area Bank [i.e., Chevy Chase] for Allegedly Failing to Service Predominantly Black Areas, (Aug. 22, 1994). The term redlining arose from the practice of literally drawing red lines on maps to indicate neighborhoods in which companies would refuse to do business. 4
Background/Redlining Defined Following the 2007 financial crisis, access to credit has contracted for several reasons: Underwriting standards significantly tightened Many lenders focus on QM lending Many institutions exited wholesale lending Concerns about FHA lending Regulators, including the Consumer Financial Protection Bureau (CFPB), Department of Justice (DOJ), prudential regulators, and Department of Housing and Urban Development (HUD) have expressed particular concern about access to credit for minority consumers. As a result, redlining has become a top fair lending enforcement priority. 5
Background/Redlining Defined The term redlining is now pplied much more broadly, and focuses on whether lenders are making credit available to consumers in minority neighborhoods. Today we are focusing on redlining from an access-to-credit standpoint. Sam Gilford, CFPB Without access to affordable credit, neighborhoods deteriorate in the long shadow cast by unfair lending. Richard Cordray, CFPB Three redlining settlements in 2015 alleged that lenders made a smaller proportion of their loans (or received a small proportion of applications) in minority census tracts as compared to non-minority census tracts within the same Metropolitan Statistical Area ( MSA ), as compared to their peer lenders. 6
Background/Redlining Defined Redlining enforcement has so far focused on depository institutions with Community Reinvestment Act obligations. The key questions have been: Has the assessment area been drawn in a fair and nondiscriminatory manner? Has the lender properly served minority areas within the assessment area? The CFPB now conducts redlining examinations and investigations on non-depository institutions and has indicated that it expects such institutions to monitor their lending for potential redlining risk. The CFPB has also suggested that it may evaluate non-mortgage products for potential redlining, and in its most recent Annual Fair Lending Report (April 2016), stated that it has begun to examine redlining in small business lending. 7
Evaluating Redlining Risk Institutional Organization Assessment area/market area Evaluate CRA assessment area or lending market to ensure it includes diverse areas. Keep a record of rationale. Branch locations Evaluate distribution of branches in non-minority and minority areas. 8
Evaluating Redlining Risk Institutional Organization Advertising/marketing Advertising/marketing content Selection criteria for prescreened solicitations Target markets for television, radio, Internet, and print advertisements Characteristics and location of brokers, realtors, builders, affinity partners Community outreach efforts Workforce diversity Evaluate whether loan origination staff are diverse. 9
Identifying and Assessing Redlining Risk Identifying redlining risk Redlining can exist in overt or comparative evidence Overt evidence: explicit statements or any geographical terms used by the institution that would connote a specific racial or national origin character (e.g. zip code). Comparative evidence: institution s treatment of areas with contrasting racial or national origin characteristics is compared. Examples of potential redlining Intentionally excluding high minority areas from the CRA assessment area. Restricting or limiting access to credit in areas that have a high minority population percentage. Restricting or limiting access to credit in areas that have a low/moderate average household income relative to other areas in the MSA. Giving disproportionately unfavorable treatment to applications from high minority or low/moderate income areas. Restricting or limiting access to credit based on locations that are limited English proficiency areas. Assessing redlining risk Assessing redlining risk Identify high minority areas located just outside the CRA assessment area and consider whether it appears as though any areas were intentionally avoided. Determine whether there is a credible and reasonable explanation for these areas being excluded. Examine marketing materials and product offerings for consistency across geographical areas. Analyze the number of applications from high minority or low/moderate income census tracts received by an institution relative to its peers. Evaluate the diversity of locations in which bank branches, loan officers, and mortgage brokers are located. Examine any branch-specific products or programs that may disproportionately harm or benefit a particular applicant group due to the location of the branch. Analyze the treatment of applications received from high minority or low/moderate income tracts by looking at the proportion of action outcomes relative to other areas. 10
Looking at Redlining Visually: First Federal Minority Percentage 80-100% 50-80% 0-50% First Federal Application Volume Increases with volume size 11
Looking at Redlining Visually: Hudson City Minority Percentage 80-100% 50-80% 0-50% Hudson City Application Volume Increases with volume size 12
Penetration Analysis: Illustrative Example One example of analysis is comparing market penetration in high minority and low income areas relative to peers. If peers can successfully capture a market share in high minority areas relative to their market share in low minority areas, it would raise questions if your institution was an outlier. Bank ratios can be calculated, comparing market share in high minority (>80% minority) vs. low minority (<50% minority) census tracts as compared to selected market peers. The penetration analysis can be performed at varying degrees of granularity, most commonly at the MSA level. Market share in low minority (<50%) areas Market share in high minority (>80%) areas Ratio of high minority market share to low minority market share Peers Bank % Peer % Bank % Peer % Peer % Bank % Ratio (Peer/ Bank) Market Peer 1 6% 8% 133% Market Peer 2 20% 12% 15% 9% 75% 75% 1.00 Market Peer 3 5% 3% 60% 0.80 The target bank for this analysis has 20% market share in low minority census tracts and 15% market share in high minority census tracts. 1.78 The target bank s ratio of high minority to low minority lending is.75 (15%/20%). This ratio is calculated similarly for each peer institution. The final ratio is calculated by dividing each peer s ratio of high/low lending by the target bank s ratio of high/low lending. Ratios greater than 1 indicate the peer is penetrating high minority areas relative to low, stronger than the target bank. 13
Action Taken by Area Analysis: Illustrative Example Another example observed is comparing action taken by neighborhoods by calculating the percentage of each action in high minority and low income areas, and potentially compare those relative to peers. If the bank has more application withdrawal in high minority areas than low minority areas and/or more withdrawal in low income areas than higher income areas, it would raise questions if your institution s application was more difficult in high minority and/or low income areas. Below is a representative example of action taken analysis comparing high minority to non-high minority census tracts. High-Minority Census Tracts 14 Non-High Minority Census Tracts Action Type Count Percentage Count Percentage Overall Count Origination 2,000 43.91% 12,000 54.73% 14,000 Approved But Not Accepted 1,000 21.95% 5,000 22.81% 6,000 Denied 1,000 21.95% 4,000 18.24% 5,000 Withdrawn 500 10.98% 750 3.42% 1,250 Incomplete 5 0.11% 100 0.46% 105 Purchased 50 1.10% 75 0.34% 125 Totals 4,555 100.00% 21,925 100.00% 26,480 The representative bank in this analysis has 10.98% of applicants in high minority census tracts withdraw their applications, whereas only 3.42% withdraw their applications in non-high minority census tracts.
Redlining for Non-HMDA Products Redlining does not only apply to HMDA products. Redlining risk can be assessed for credit cards, small business cards, and small business loans. Credit Cards Small Business Loans To evaluate potential redlining risk for credit cards, the distribution of applications processed and approved can be examined across geographies. The analysis can be viewed at the overall or MSA level. In the below example, approximately one third of applicants come from census tracts with less than 20 percent minority concentration. Another one third of applicants come from census tracts with 20 to 50 percent minority concentration. The remainder are roughly equally distributed between 50 to 80 percent and 80 percent and higher minority concentration census tracts. When comparing the application and account distribution to the overall census distribution, the chart shows the lender lends proportionally more in high minority concentration areas. Assessing redlining risk in small business creates a unique challenge in that the distribution of the general population is not necessarily reflective of the distribution of small business owners. Therefore, one could leverage public data on small businesses in geographical locations, for example, Community Reinvestment (CRA) data, to represent the existence of small businesses in the location. Limitations still exist, but it can be a precursory view into penetration against peers, relative to the number of businesses in that area. Once the CFPB issues guidance with respect to small business data collection, the ability to perform redlining analytics will be increased. Even in the absence of guidance, questions have been raised in this area. 50% 40% 30% 20% 10% 0% Less than 20% Between 20% and 50% Between 50% and 80% Greater than or equal to 80% Total Population from 2014 Census Data Credit Card Applications Credit Card Approved Accounts 15 Section 1071 of Dodd-Frank amended ECOA to require financial institutions to compile, maintain, and submit to the Bureau data on lending to small, minority-, and women-owned, businesses.rulemaking required by the Dodd-Frank Act s small business data collection provision is now in the pre-rule stage. - 2016 Fair Lending Report of the CFPB
Defending Redlining Claims Legal Defenses Disparate treatment or disparate impact? The term redlining implies intentional conduct, and the practice historically was viewed as a disparate treatment violation. Redlining is a form of illegal disparate treatment whereby a lender provides unequal access to credit, or unequal terms of credit, because of the race, color, national origin, or other prohibited characteristic(s) of the residents of the area in which the credit seeker resides or will reside or in which the residential property to be mortgaged is located. Consumer Compliance Handbook, Fair Lending Overview (1/06). 16
Defending Redlining Claims Legal Defenses In litigation, a defendant may be able to challenge whether the elements of disparate treatment and/or disparate impact have been satisfied. Disparate treatment: Charging party must show that the lender avoided certain geographies because of race/ethnicity. Disparate impact: Statistical disparity must be linked to a specific policy or practice, and the lender may have a legitimate business justification. 17
Defending Redlining Claims Factual Defenses Rationale for assessment area/market area Be prepared to explain how your institution chose to delineate its assessment area or market to rebut claims that it intentionally avoided minority tracts. Rationale for branch locations Similarly, be prepared to explain how your institution chose its branch locations. Advertising/marketing efforts Advertising in and marketing to minority neighborhoods can help refute claims that an institution intentionally avoided doing business there. 18
Defending Redlining Claims Factual Defenses Business strategy Niche Lenders: Provided the lender did not adopt its niche under circumstances from which a discriminatory purpose could fairly be inferred, and that the lender s niche marketing is consistently applied, [, ] would not generally expect the Department [of Justice] to prosecute on the basis of such niche lending practices. Letter from Deval L. Patrick, Assistant Attorney General, U.S. Dept. of Justice, Department of Justice Fair Lending Enforcement Program, (Feb. 21, 1994). Product/geographic expansion plan Diverse workforce Diverse sales teams/mortgage brokers can help demonstrate a desire to reach diverse customer base. 19
Defending Redlining Investigations Statistical Analyses The institution is being investigated in regards to redliningwhat can you do? Definition of peer Sensitivity analyses Develop analytics to reflect business process 20
Redlining Analytics: Peer Selection When faced with an investigation, consider whether the peers used in the analyses are actually peers: Consider comparing against an individual peer vs. comparing against the "aggregated peer" There are two general characteristics to consider when identifying peers: Volume Market type To be considered a peer, an institution should decide on a similar number of applications within the MSA. A much smaller institution likely has fewer branch locations and less resources that can be dedicated to marketing. A much larger institution likely has more local presence and pull in the region, as well as additional incentive to invest in lower-income areas. A good threshold range is 50-200% application volume. Additionally, peers that have an average loan size of over $1M are removed. To be considered a peer, an institution should operate in the same market within an MSA. Loan type: The percentage of government loans should be considered and if vastly different the business should not be considered a peer. Property types - The distribution of property types is likely to vary across geographical areas. For this reason, a lender dealing primarily in Jumbo loans may be more limited to lending in certain areas than a lender dealing primarily in conforming loans. Risk appetite - An institution that has an objective to take high risk and earn high return will seek riskier applicants and therefore face potentially different geographical markets. For example, sub-prime lenders may be more limited to lending in certain areas. 21
Illustrative Sensitivity Analyses 5 1 Census tract or self-identified 2 Loan action type Sensitivity analyses can be conducted testing for the difference in redlining using census tract minority data or selfidentified minority data. Deposit analysis An analysis could be created to reflect the anecdotal stories relative to your specific business. An example of this type of analysis is represented on the next slide as it relates to lending relative to deposits. 5 Example sensitivity analyses The HMDA data includes action taken type. A sensitivity analysis can be conducted to include only originations or all applications or additionally consider loans purchased by the financial institution. 3 Purpose type The HMDA data includes purpose of the loan. Sensitivity analyses by loan purpose can be conducted, for example, separating home purchase from refinance. 4 Owner occupancy The HMDA data includes owneroccupancy status. Sensitivity analyses by owner occupancy can be conducted. 22
Deposit Analysis: An Illustrative Example When faced with an investigation, consider analyses that will reflect the specifics of your business, for example, consider lending relative to deposits received if physical presence isn t there. $0.50 $0.40 $0.30 Dollars lent (annual) per $1 deposit The dollars lent per dollar of deposits can be compared in high minority and low-income areas relative to peers. The institution can argue, that relative to deposits, it is lending the same or more than its peers. High Minority Tract $0.20 $0.10 $0.00 Overall High Minority Tract Non-High Minority Tract Bank A Bank B Bank C Bank D Deposits Loan Count Average Loan Size Loans per Deposit Bank A $ 5,000,000,000 5,000 $ 200,000 1.00 $ 0.40 Bank B $ 12,000,000,000 6,000 $ 240,000 0.50 $ 0.12 Bank C $ 20,000,000,000 3,000 $ 210,000 0.15 $ 0.03 Bank D $ 5,000,000,000 4,000 $ 175,000 0.80 $ 0.14 Annual Dollars Lent Per Dollar Deposit The bank in question, Bank A, lends proportionally more money per deposit than its peer institutions. 23
QUESTIONS
Mayer Brownis a global legal services provider comprising legal practices that are separate entities (the "Mayer Brown Practices"). The Mayer Brown Practices are: Mayer Brown LLPand Mayer Brown Europe Brussels LLP, both limited liability partnershipsestablished in Illinois USA; Mayer BrownInternational LLP, a limited liability partnership incorporated in England and Wales (authorized and regulated by the Solicitors Regulation Authority and registered in England and Wales number OC 303359);Mayer Brown, a SELAS established in France; Mayer Brown JSM, a Hong Kong partnership and its associated legal practices in Asia; and Tauil & Chequer Advogados, a Brazilian law partnership with which Mayer Brown is associated. Mayer Brown Consulting (Singapore) Pte. Ltd and its subsidiary, which are affiliated with Mayer Brown, provide customsand trade advisory and consultancy services, not legal services. "Mayer Brown" and the Mayer Brown logo are the trademarks of the Mayer Brown Practices in their respective jurisdictions.