Manufacturing Fair Lending

Steve Simpson, director of separation of powers litigation for the Pacific Legal Foundation, a public interest legal organization that advocates for limited government, recently defended Chicago-based Townstone Financial against a CFPB lawsuit accusing the non-bank lender of redlining. A district court dismissed the case, a decision which the CFPB appealed.

Simpson says lending disparities which regulators call “redlining” differ greatly from the literal practice from which the term originated. “Any disparity in a bank’s lending to one group or another is just held to be de facto redlining,” he says. “I just think that’s flawed.” Noting his position in the intellectual minority, he adds that “very few people” care to address the flawed premise driving the government’s approach to identifying redlining.

He also predicts there is more fair lending litigation to come, given the Supreme Court’s decision last June to overturn race-based affirmative action in college admissions. “The more it seems clear that the Supreme Court is against race conscious policies,” he continues, “the more in favor of those policies regulators become.”

A Modern Theory Of Redlining 

The term “redlining” derives from the literal practice, employed by the federal government and lenders, of drawing red lines around neighborhoods they would not invest in based on those neighborhoods’ racial composition. The practice was enshrined in the Federal Housing Administration’s (FHA) 1938 Underwriting Manual, which prohibited “the occupancy of properties except by the race for which they are intended,” among other racist housing policies.

Redlining lawsuits against non-depository mortgage lenders are a recent phenomenon; traditionally, only depository banks were audited by federal regulators for compliance with the 1977 Community Reinvestment Act (CRA), which outlawed redlining and established a framework for regulators to assess depositories’ progress at meeting the credit needs of their entire communities, including low- to moderate-income and majority-minority neighborhoods.

The primary argument against CRA mandates for non-depository lenders is that those institutions do not receive deposits with which they can reinvest in communities. However, non-depositories have always been subject to other non-discrimination and fair lending laws.

Natural disparities will always exist in lending, Simpson says; inferring redlining or intentional discrimination from statistics compounds the problem by propagating the fantasy of an achievable equilibrium. Accusing lenders who originate a below-average number of mortgages in CRA-eligible census tracts of redlining – without demonstrating a lender’s intent to avoid or otherwise restrict access to mortgage credit in those communities – manufactures perceptions of discriminatory lending, while pushing lenders to manufacture their fair lending compliance.

Lenders’ inability or unwillingness to litigate against the government, given the expense, inflates perceptions of rampant redlining and the effectiveness of regulators’ “combatting” efforts. 

While the CRA speaks to income, not race, Hancock says, “in most of these areas there’s a great overlap between racial concentrations and low- and moderate-income concentrations. So, the government could analyze a bank’s loan distributions and see if you’re intentionally avoiding doing business in the minority communities of your assessment area. That was the theory.”

By relying on lending disparities to establish patterns of discrimination, regulators’ enforcement of that theory corrupts the original intent of the CRA statute; claiming that lower levels of originations in CRA-eligible census tracts equates to redlining makes fair lending enforcement – and lenders’ compliance – a data exercise.

“Under the government’s theory, you can eliminate redlining by just making fewer loans in white areas. You’re not doing any more in minority areas,” Hancock says. “You’re making fewer loans in white areas and somehow that solves your legal problem. That just doesn’t make any sense.”

The reputational harm of being accused of redlining lives with lenders, not regulators, Simpson says, compounding the ill-effects of ill-designed efforts to regulate away historical disparities. Simspon calls regulating away disparities in lending “political” – but not in the sense of Democrats versus Republicans.

“Regulators and everybody out there today in various sectors of the economy and finance being one,” he says, “are just falling all over themselves to look like they’re combating inequality.” The Illinois Community Reinvestment Act (ILCRA, see sidebar), which took effect in January 2024, uses proportional loan distributions to assess depository and non-depository mortgage lenders for fair lending compliance, mirroring a 2007 Massachusetts law for supervising non-banks. 

“They care about how they’re being perceived or feeling like they’re combating a great social ill,” Simpson believes. “When they actually try to implement these policies, they make no sense, and the real experts, like the Mortgage Bankers Association, who knows what it’s doing and understands all this stuff much better than, in my view, the Illinois regulators, no one is paying attention to them.”

Hedging Fair Lending Compliance Risks

No matter whether government overreach is occurring, says Brian Montgomery, co-founder of Gate House Strategies and former deputy secretary of the Department of Housing and Urban Development (HUD), the Biden administration’s focus on redlining – however defined – makes the issue top-of-mind for mortgage lenders. 

Drawing parallels to the heightened focus on cybersecurity given a recent spate of high-profile breaches, “even if there’s a change in administration,” Montgomery affirms, “there will continue to be a focus on this topic, as there should be.” But, a regulatory rat race mostly hurts consumers on account of the added complexities and expenses for businesses’ compliance. 

Some states have the benefit of staying reliably “red” or “blue,” but whipsawing regulations means businesses must remain ready to pivot at all times depending on the regulatory priorities of the day. The headaches and expenses that arise can be exorbitant.

“The lenders here are facing an issue,” Hancock adds. “If there is a change of administrations, will they suddenly be tagged for doing what was demanded of them now?” 

Lenders must hedge their fair lending compliance to satisfy regulators’ present demands for a racial balance while avoiding future legal liability for doing just that. Under a second Trump administration, the DOJ, the Department of Housing and Urban Development (HUD), and the CFPB would likely look very different. “There could be completely different theories espoused as to not only what the law requires, but what the law permits,” Hancock worries. 

Therein lies the untenable nature of the government’s strategy for “combatting redlining.”

“The problem arises when they start inferring discrimination from statistics when they don’t have direct evidence of discrimination,” says Simpson, referencing his work with Townstone Financial. “You end up trying to compel mortgage companies or businesses to lend on the basis of race and on the basis of other factors that they’re not supposed to take into consideration, which just creates an equal protection problem [through] perverse incentives and perverse effects.”

The perverse application of the law of unintended consequences means that “any community, or anybody who operates on a community level can ultimately be accused of discrimination,” he continues, “in some sense precisely because they’re operating locally and helping a given population, or they’re operating by word of mouth and just talking to their circles, their friend circles, their community circles” – i.e., referrals, which is how most of mortgage lending works.

Such disparate-impact discrimination is at the heart of regulators’ modern theory of redlining.

A Decade Of Disparate Impact Legislation

Richard Andreano, partner and Practice Leader of the Mortgage Banking Group at Ballard Spahr, a national law firm, says he “probably would not” have appealed the Townstone Financial case up to the Seventh Circuit. Andreano also leads Ballard Spahr’s Fair Lending Team.

In 2015 the Supreme Court ruled in Texas Department of Housing & Community Affairs v. The Inclusive Communities Project, Inc. that plaintiffs must meet a rigorous standard to establish a clear-cut case of disparate-impact discrimination under the Fair Housing Act (FHA). The justices even noted that claims based on statistical disparities fail without showing robust causation. 

On remand, the U.S. District Court for the Northern District of Texas applied that new standard and found the plaintiff (Inclusive Communities) fell short of showing robust causation. The outcome of these proceedings confirmed that disparate-impact discrimination claims were legitimate under the Fair Housing Act, but that plaintiffs proceeding under a disparate-impact theory face a significant burden of proof.

“It was clear the court was very concerned about the rules in that area,” says Andreano, “and that if you went too far, were you actually in effect mandating the consideration of race in a way that would appear to be contrary to the Constitution.” 

However, in 2013, HUD implemented its own “Discriminatory Effects Rule” to formalize disparate-impact liability. At the time, HUD noted the new rule had “broad remedial intent” and was “imperative to the success of civil rights law enforcement.” Two insurance industry trade groups sued HUD over the rule, arguing the rule exceeded HUD’s authority under the FHA. 

Noting the Supreme Court would soon be taking up the issue in Inclusive Communities, a D.C. Circuit Court provisionally agreed with the insurance industry’s challenge. The Supreme Court acknowledged HUD’s 2013 Rule in the Inclusive Communities decision, but did not defer to it. The D.C. Court had to reconsider its ruling on the 2013 Rule following Inclusive Communities.

Before that could happen, a new HUD rule (the 2020 Fair Housing Act rule) was implemented , during the Trump administration, supplanting the 2013 rule. Fair housing groups challenged the 2020 Rule and won a preliminary injunction. The Biden administration then reinstated the 2013 Rule in March 2023 – without revising for Inclusive Communities.

Andreano calls this HUD’s “calculated error” – they reinstated the 2013 Rule almost verbatim.

In September 2023, a decade after the case began, the D.C. Court determined that in light of Inclusive Communities, HUD’s 2013 Rule does not conflict with the FHA when applied to insurers’ underwriting and rating practices. However, the D.C. Court noted that the 2013 Rule may exceed limits laid out in Inclusive Communities.

“By adopting a rule that was adopted two years before that Inclusive Communities decision,” says Andreano, “and then claim that the rules completely consistent with that future decision, was an error. I think you’re going to see this Supreme Court reign that conduct in, saying, ‘No, you have to actually follow our decisions.’”

That calculated error could be amplified if the Supreme Court reverses or narrows the Chevron principle, a long-standing judicial precedent that says courts should defer to federal agencies’ expertise when interpreting and administering federal statutes and regulations. After hearing oral arguments in a separate case in mid-January, the Supreme Court seems poised to overturn the Chevron principle. 

The CFPB appealed its case against Townstone Financial arguing Chevron deference. 

“All the redlining claims were brought against banks who had their banking regulator twisting their arm to settle. They settled. They sued the first non-bank, and the non-bank didn’t want to settle, and that’s how we’re finally getting court review of this,” Andreano says. “It just never faced a challenge was the issue. I do think the courts, particularly if Chevron’s thrown out, only if you just modify it, I think that could significantly change how courts defer.”

Emerging Fair Servicing Compliance Risks

The same way underwriting standards are set by the government-sponsored enterprises (GSEs), so are default servicing standards. Where fair lending focuses more on the origination side of mortgage lending, fair servicing kicks in after the loan closes, entailing such actions as collection processes, fee assessments, and a waterfall of loss mitigation options.

The problem for many lenders, says Michael Waldron, founding partner of Gate House Compliance and former chief compliance officer at Community Loan Servicing, LLC (formerly Bayview Loan Servicing, LLC) is that fair lending and fair servicing at many mortgage companies remain disparate processes.

“It doesn’t mean that it’s not a priority,” he explains. “It doesn’t mean that it’s not resource-intensive. It’s just not as mature of a structure and mature from a thought-leadership perspective.” Hence, for mortgage lenders, fair servicing compliance can be more difficult to stay ahead of than fair lending compliance.

“The playbook has been out there,” says Liza Warner, a partner with CrossCheck Compliance, “but interpreting the playbook and operationalizing it and incorporating it into a mortgage operation’s day-to-day activities is not always clear.” CrossCheck aided Townstone Financial with their data analysis in the process of their defense against the CFPB. These days, data reigns when it comes to staying out of regulators’ crosshairs, Warner says. 

Regulators can assess servicing data for fair servicing compliance, looking for patterns that could signal discrimination, such as whether protected class borrowers are being foreclosed on more often than non-protected class borrowers. 

“Fair servicing,” Warner continues, “obviously is not as mature of a process of monitoring as it is on the origination side, and you don’t have a set of data like you have the [Home Mortgage Disclosure Act] data to compare results against. It’s a little more challenging that way. You really have to understand what’s happening within the operation in order to conclude on anything with respect to the data.”

Problematically, much of the data available to lenders for analysis offers a rear-view perspective. They know where their own loans are written, “but the government is looking at a comparison with other lenders, and you don’t have other lenders’ data until the following year,” says Hancock. When it comes to peer benchmarking, “there’s no way of knowing.”

Use what you have though, Warner says. “You need to make sure as a company that the data is accurate, first of all, and that as an organization you understand what the data is telling you. Am I in trouble or am I not? If so, what areas do we need to focus on? Somebody internally needs to really take that in and understand what it means and how to make changes.”

Lenders’ failure to mine their own data for insights into potential fair servicing risks underscores the shortcomings of data reliance for fair lending and fair servicing compliance overall. “What you don’t want to do,” Waldron explains, “is take action that inadvertently doesn’t mitigate the very issue that you’re trying to solve for.” 

Drawing bad conclusions from good data can lead lenders to make misguided investments or operational changes, inadvertently increasing long-term risks. 

“As complex as we’ve set it out on the origination side, I would contend on the servicing side it’s a multiple in terms of complexity,” Waldron says. “The interaction is condensed on the servicing side. It’s a continuum, it’s a relationship, and it can touch everything from how you waive fees or give deference to your servicing client, your outreach efforts… how hard are you trying to reach them when an event occurs in the loss mitigation scenario?”

The difficulty of finding a comparable situation for an assessment’s sake makes fair servicing compliance that much more difficult for lenders and regulators to assess, too. Every borrower’s situation is unique, especially when they go into default. In a similar way that regulators can assess lending data to identify redlining, regulators can assess servicing data and draw conclusions about servicing discrimination. 

Ultimately, lenders have to get to their data before regulators do. When regulators are the first to identify patterns of potential discrimination in a lender’s servicing data, that lender has already failed to demonstrate an awareness of their own potential shortcomings.

“The government would like to find a case of discrimination and servicing,” says Hancock, “and there hasn’t yet been a major case. There might be one-off cases someplace, but there hasn’t been a case of discrimination and servicing like there has been in discrimination and origination of loans or marketing such as redlining.” 

Illinois Regulators Ignore Industry Concerns As State-CRA Law Takes Effect

Mirroring a state-level community reinvestment law that took effect in Massachusetts in 2007, the Illinois Community Reinvestment Act (ILCRA) was passed in March 2021, but not implemented until January 2024, when regulations for its implementation were passed.

Matt Rohl, who previously served as senior vice president of CRA Development and Emerging Markets for the recently-acquired Draper and Kramer Mortgage Corporation, was installed in January 2024 on the Illinois Mortgage Bankers Association’s (MBA) Board of Directors. He claims Illinois regulators ignored industry concerns about the ILCRA, submitted during the comment period lasting from March 2021 until January’s implementing regulations took effect.

The ILCRA is modeled off the federal CRA, but expands the scope of covered financial institutions to include credit unions and entities licensed pursuant to the Residential Mortgage License Act of 1987 which lent or originated 50 or more mortgages in the previous calendar year which are not covered pursuant to federal law, like independent mortgage banks (IMBs).

Now that regulations implementing the ILCRA have taken effect, the Illinois MBA’s concerns are being reviewed by state regulators – which is too little, too late, Rohl says.

Who Gets CRA Credit?

A signature issue with the ILCRA, Rohl explains, is the narrowing of who receives credit for originating mortgages in CRA-eligible census tracts. Under the federal CRA statute, buying loans underwritten for borrowers living in low- to moderate-income or majority-minority census tracts – i.e., CRA loans – grants CRA credit to buyers of those loans. 

Fearing “churning” – a CRA loophole exploited by repeatedly buying and selling the same CRA-eligible loans – the ILCRA says only one entity can receive CRA credit for loans underwritten for CRA-eligible census tracts in Illinois. 

Despite perceptions of CRA pools being riskier investments on account of weaker collateral, CRA loans are highly sought after by investor-entities subject to CRA requirements, Rohl says. Competition for CRA-eligible loans in the secondary market tightens the investor spread, improving pricing in the primary market for borrowers applying for mortgages in those areas.

Eliminating the ability for buyers of CRA loans to receive CRA credit undermines the added value of CRA loans to CRA-minded investors, making for less competitive pricing for these loans in the secondary market. “Saying only the first person can get credit for it actually makes the terms worse for the borrower,” says Rohl.

And the reverse can happen for brokered loans, he says. 

Mortgage bankers who broker loans – meaning the loans are underwritten and purchased by a different funding entity – would not receive CRA credit for originating CRA-eligible loans on account of the single-buyer restriction. Only the entity that underwrote and bought the loan would receive credit. Hence, IMBs brokering loans for borrowers in CRA-eligible census tracts, though expanding access to mortgage credit in those communities, would not receive commensurate CRA credit. 

“In that situation, the first person that actually dealt one-on-one with the borrower would not get credit for that CRA loan,” Rohl explains. “If I’m an independent mortgage banker and I’m letting you underwrite it, I would get no credit at all,” despite that originating entity directly helping to expand access to mortgage credit in a CRA-eligible community.

The Examination Process

Whereas banks and credit unions with more than $10 million in assets pay an annual fee to regulators that cover expenses related to regulatory audits and examinations – and those with assets under that threshold pay no annual fee – the ILCRA will charge IMBs who originate 50 or more mortgages in Illinois in a calendar year a daily rate of $2,200 for them to be audited.

Those audits could last a day, a week, a month, or more. IMBs not headquartered in Illinois, but who originate more than 50 loans in the state on an annual basis, will also have to foot the bill for auditors’ travel expenses. An audit that lasts a month would cost more than $60,000, due 30 days from the end of the audit, and audits are expected to happen every two years. 

The expense is disproportionate for smaller IMBs, Rohl says, compared to their asset-heavy banking peers. IMBs don’t take or hold deposits. A more fundamental issue with the examination process, however, is the manner in which IMBs are assessed and judged to be either compliant or non-compliant with ILCRA. 

By failing to engage with the concerns of the Illinois MBA before the implementing regulations took effect, which “activated” the law passed in 2021, regulators neglected to clarify how IMBs in the state are expected to achieve compliance with the ILCRA, besides proportional distribution.

According to Rohl, “They just put this law into place and said, ‘Well, we’ll talk about it after it’s in place.’” He asks, rhetorically: “Who’s my peer group? What are the percentages I should be doing? That, again, is subjective and wouldn’t really have any precedent until the audits start happening.” He and other fair lending experts fear regulators’ reliance on peer benchmarking can manufacture the illusion of redlining, spurring allegations of non-compliance.

Pretend, for example, an out-of-state IMB wants to open its first branch in Illinois. The IMB gets licensed, recruits a small group of loan officers, and opens an office. The branch originates 100 loans in the following year, and another 100 more loans the year after. Due to where the branch is located and due to where the loan officers live and lend, only a handful of originations each year were from CRA-eligible census tracts.

From regulators’ perspective of proportional distribution, that IMB engaged in redlining. “What should the number be?,” Rohl asks. “What should the percentage of CRA loans versus non-CRA loans be? That’s not in writing. It’s a little subjective.” In this scenario, “redlining” was manufactured by the out-of-state IMB starting to do business in Illinois in the first place.

“Under the government’s theory, you can eliminate redlining by just making fewer loans in white areas,” says Paul Hancock, a civil rights attorney who spent more than two decades at senior levels of the Department of Justice (DOJ) developing and implementing the DOJ’s fair lending enforcement program. “You’re not doing anymore in minority areas. You’re making fewer loans in white areas and somehow that solves your legal problem. That just doesn’t make any sense.”

Given lenders’ ongoing profitability crisis, the threat of unreasonable fair lending enforcement could decrease competition by making the state unattractive for doing business, says Rohl, undermining efforts to reach underserved borrowers, likely making mortgage financing costlier.

“To say that I’m not paying attention to an underserved community,” Rohl explains, “it’s just simply because I don’t have anybody in that community. But, what this law says is, I need to hire and open an office there. That’s the only way I would be able to serve that community.”

“My concern,” Hancock adds, “is that the government has kind of drifted much closer to a demand for racial balance in loan originations, rather than attacking lenders that really are refusing to do business in minority areas.” Demands for racial balance are unconstitutional, he says.

Policing Appraisal Bias

A third area of the Illinois MBA’s concern over the ILCRA concerns appraisal bias. Rohl says the ILCRA creates regulatory confusion surrounding the federal enforcement of the Dodd-Frank Act Wall Street Reform and Consumer Protection Act (Dodd-Frank Act). 

In addition to establishing the CFPB, the Dodd-Frank Act was passed in the aftermath of the Great Financial Crisis to implement safeguards in mortgage lending, among other reforms. One such Dodd-Frank safeguard prohibits mortgage lenders from influencing appraisals, in any capacity. The ILCRA makes it the responsibility of IMBs and credit unions to police appraisal bias by denying loans for properties with appraisals that the lender suspects of bias. 

However, asking lenders in Illinois to police an aspect of mortgage lending outside their regulatory and professional purview could do more harm than good, Rohl fears. 

Differentiating less-than-blatant bias from laziness or inexperience can already be difficult for those well-trained in the discipline. Denying CRA-eligible loans suspected of appraisal bias could open up lenders to accusations of lending bias for wrongly denying applicants. Approving applications for loans they suspect of appraisal bias – but do not deny so as not to court fair lending risk – could make lenders vulnerable to lending risks like repurchase demands.

Ultimately, lenders face a conflict of interest because the ILCRA pushes lenders to write more loans in CRA-eligible census tracts, which are historical hotbeds of appraisal bias. “It’s counterintuitive to the actual federal law, in that case,” Rohl says. “They’re saying that you have to police bias on appraisals, yet you’re not supposed to influence or have anything to do with the appraisal.”

Where Rohl believes the spirit of the ILCRA represents a desire to erase racial disparities in homeownership and mortgage lending, the ILCRA will likely undermine that ambition by pushing IMBs to lend where they are not and adding operational complexities and costs.