Ed. #14: Fair Lending; Complexity and Disparate Impact's Impact
August 3, 2020|CFPB, Fair Lending, Mortgage Industry
CFPB Issues RFI on Fair Lending
On July 28, 2020 the CFPB issued a request for information (RFI) relative to fostering innovation and preventing credit discrimination. Among other things, the RFI asks whether and how the CFPB should issue guidance on affirmative advertising to disadvantaged groups. Yet, just a week earlier, the CFPB filed its first ever fair lending case alleging violations arising from the failure to affirmatively market to all communities. Shoot first, ask questions later? Meanwhile, arguably, the most controversial aspect of the RFI asks regarding the use of disparate impact analysis (DI) to assess fair lending compliance.
I hate kale
Many social scientists believe hate is a learned trait. There are disturbing studies, however, concluding that humans desire to seek bonds with others through common hate. To that end, I have been an anti-kale activist for several years and would like some company. My wife has not joined my crusade, but our marriage remains strong.[1] My grievances are simple: kale is a lousy tasting vegetable that in all instances would be better off being lettuce or spinach: it should be relegated back to garnish-only status. Kale is an abomination on pizza or in pasta and why would you ever massage a vegetable? The pro-kale forces, with claims that it will help you live longer[2], have convinced otherwise able chefs to use this former plate decoration in recipes. Popeye ate spinach out of a can and even that tastes better than kale (and look at the results![3]). I don’t claim I have a flair for finishing food fads[4], but I used to be anti-pomegranate. Just sayin’. I’ve got my eye on you now, kombucha. Who’s with me?
Complexity is a subsidy
Does that kale talk have anything to do with fair lending and DI? Probably not, but have you ever heard that complexity is a subsidy? Complexity refers to laws, government regulation and other institutional restraints on free trade that limit how business can operate. I’ll explain the subsidy part shortly, but I think we can all agree that dealing with state real estate conveyancing laws, compliance with the alphabet soup of state and federal regulations (and regulators), licensing, counterparty, underwriting and secondary market guidelines etc., makes mortgage lending an incredibly complex business.
So, how is that a subsidy? I’ll let conservative columnist Jonah Goldberg[5]explain,
The more that financial success depends on ….., lawyers, lobbyists, and accountants; the more onerous regulations become for men-with-strong-backs to find work or for entrepreneurs to start businesses – then the more we move towards a society where the government rewards people based on their ability to navigate paperwork or fulfill quotas on a political to-do list. Complexity benefits statists because increasing complexity allows statists to claim we need more government to help people navigate through these complex times. In the process of helping, they make the government more complicated, creating new services for “fixers” of all stripes to solve problems the statists created in the first place. The more you look around at spots where society and government intersect, the more you can see how pervasive and pernicious this dynamic is. The more rules you have, the more power you bequeath to the people well-suited to make or manipulate the rules.
Complexity subsidy and big banks vs. smaller lenders
Lenders need expensive and knowledgeable people and systems to compliantly and responsibly make mortgage loans and to “protect” the mortgage lender from claims they are violating the law.[6] Meanwhile, DI has been the single biggest challenge for fair lending compliance since the 1990’s. DI is a proverbial compliance trick box, imposing liability for violations without any intent to discriminate and regardless of any ability to control the lending patterns.[7] With DI, lenders fear that any business decision might impact protected classes differently such that discrimination can be found purely based on the impact of the decision. Intent to discriminate is irrelevant to DI. Like other mortgage legal and compliance complexities such as LO Comp, QM and limited English proficiency issues, DI causes many lenders to struggle with being able to be innovative or make nimble local decisions to meet competition such as lowering price or expanding underwriting guidelines.[8] DI requires lenders to assess how that decision may impact lending patterns regardless of any intent to discriminate.
So, many of you may have been scratching your head when you heard that Wells, B of A, Quicken Loans, and a few other large lenders (and David Stevens and NAR) recently asked HUD to delay and reconsider a 2019 proposed rule that would have limited DI [9]. Smaller lenders had different views about it. Why are the bigger lenders vocalizing support for DI, but not the smaller ones? Is it because the big lenders believe DI is the true legal and empirical measure of fair lending, or do they believe DI enforcement is effective in lowering homeownership barriers to minorities?[10] Cynics might say it is just to placate consumer and minority groups in this important moment of racial reassessment and unrest.[11] Well, everyone has their own motivations and motivations are often mixed, but the complexity subsidy might explain the big vs. small lender dichotomy.
DI is complex and that subsidizes big lenders
Only the biggest lenders can invest in the enormous compliance oversight regimes required by DI to effectively measure and control for DI in lending. Smaller lenders don’t have (i) enough loans over a wide enough geographical footprint to avoid obvious outliers, (ii) enough data to anticipate consequences over a broad market area, and/or (iii) the economic scale to afford the compliance resources to model, anticipate and manage DI risk. As a result, DI’s complexity subsidizes larger lenders by commoditizing the mortgage business. Commoditization of mortgage lending benefits companies with larger scale, but at the same time, stifles innovation and poses barriers to disrupters to enter the market.
Fair lending laws haven’t worked
It really should be unnecessary to have laws prohibiting discrimination against minorities and others in lending[12]. As proven by stated income lending in the pre-meltdown era, mortgage originators will make loans to virtually anyone if they qualify. Meanwhile, over the roughly 50 years since enactment of Fair Housing Act and ECOA, fair lending enforcement has been at times lax and at other times aggressive, but the homeownership gap for minorities hasn’t changed.
We need innovation
Despite the good intentions behind fair lending laws and those that enforce them, we often are unable to agree when illegal discrimination is occurring. Moreover, we seem unwilling to recognize that the “neutral” creditworthiness underwriting criteria we commonly use (i.e., complexities created by GSE and secondary market guidelines) may institutionally foster the persistent discriminatory impact in the homeownership gap. I'm not offering a better idea, just saying that the complexity subsidy of many laws is preventing innovation. It will be interesting to see what this RFI process produces in terms of action from the CFPB to facilitate innovation to increase minority homeownership as opposed to just the furtherance of fair lending enforcement. 50 years of fair lending law enforcement without moving the needle means we need to do something different if we really want to impact minority homeownership rates.
[1] She used to hide kale in otherwise tasty dishes. I really love her brussels sprouts, but even bacon can’t help kale.
[2]This apparently is a strong motivator for humans as well.
[3]Not the Popeye video you were expecting. Toot toot!
[4] I am quite fond of alliteration though (See, e.g., Mortgage Musings)
[5] I am a huge fan of Jonah and the new digital news media company he, David French (referenced in a previous Musings here), Sarah Isgur, Steven Hayes and other brilliant writers started less than a year ago called the The Dispatch. It’s a refreshing break from the outrage cheerleading of CNN/Fox/MSNBC etc. Check it out.
[6] It is not lost on me that the same complexity is how I make a living. I am grateful for the paid work I get.
[7]See e.g., 1996 Long Beach Settlement Agreement
[8] If I had a nickel for every time I had to say, “consider the fair lending implications of that decision”…., Never mind, I do get nickels for saying that.
[9]Actually, DI in housing finance was already limited by the Supreme Court in 2015. HUD is just getting around to conforming its DI rules to the current law.
[10] See below. Empirical data on this question is needed given the continued racial gap in homeownership.
[11] I don't recall the same big lenders seeking to file amicus briefs in support of strong DI when the issue was at the Supreme Court. Given the legal issue was decided in 2015, are these letters just a loud barn door closing after the horse is out?
[12] But, such laws are necessary.