After leaving the public, press, regulators, and lenders lingering for six months, the Supreme Court finally produced its 5-4 opinion in Texas Department of Housing & Community Affairs v. The Inclusive Communities Project, Inc., a blockbuster case in which the Court concluded that the Fair Housing Act permits statistically-based disparate impact claims.[1]  The Supreme Court’s opinion is important for two reasons.  First, it finally puts the Supreme Court’s seal of approval on FHA disparate impact claims, a theory of liability that has been universally recognized by the federal courts of appeals.  Second, it will likely embolden plaintiffs to file more suits using a disparate impact theory, which means lenders should evaluate their current policies and practices, especially discretionary pricing policies, to avoid engaging in practices that could be interpreted as discriminatory.  

But before going into Justice Kennedy’s rationale for the majority, it is important to understand disparate impact liability and the legal landscape that existed before the Court’s opinion in Texas Department of Housing & Community Affairs

What is disparate impact?

The theory of disparate impact originated with the Court’s opinion in the Title VII employment discrimination case of Griggs v. Duke Power Company.[2]  In that case, the Court resolved whether an employer’s requirement that employees have a high school diploma or pass a standardized general intelligence test for promotion disproportionately affected African-Americans and kept them from advancement.  Relying upon the catch-all phrase “otherwise adversely affect” in 42 U.S.C. § 2000e-2(a)(2), the Court concluded that Title VII “proscribe[d] not only overt discrimination but also practices that are fair in form, but discriminatory in operation.”[3] 

Disparate impact claims, unlike disparate treatment claims, do not require a showing of discriminatory intent.[4]  Because intentional discrimination is often nearly impossible to prove, disparate impact liability provides an alternative for addressing “facially neutral” policies that have a disproportionate effect on a protected class of individuals.  The most common method in which plaintiffs show disparate impact is through statistical disparities that suggest the defendant is engaging in a discriminatory practice.[5]  To show statistical disparities under a disparate impact theory, courts have generally used regression analysis to assess the influence of race and other factors in the defendant’s decision-making process.[6] 

To show a prima facie case of disparate impact, plaintiffs must first establish that they are members of a statutorily protected class.[7]  Next, they must show that the facially neutral policy of the defendant “actually or predictably results in . . . discrimination,” which is generally established through regression analysis.[8] 

Once a plaintiff has made that showing, the burden then shifts to the defendant to show that its policy was necessary to promote a legitimate, non-discriminatory governmental interest (of a public sector defendant) or business objective (of a private sector defendant).[9]  If the defendant shows that its policy was justified, then the burden shifts back to the plaintiff to show that the defendant’s justification was false or pretextual.[10] The plaintiff can also present a less discriminatory alternative that still achieves the defendant’s legitimate governmental interests or  business objectives.[11] A plaintiff’s failure to make either showing can be dispositive of a disparate impact claim.  

What changed with Texas Department of Housing & Community Affairs

Disparate impact liability under the FHA has been recognized for nearly 40 years.  The theory was first recognized under the FHA by the Eighth Circuit in 1974, just six years after the statute’s enactment.[12]  In the next four decades, nine other federal courts of appeals concluded that the FHA encompassed disparate impact claims.[13]  Even the Department of Housing and Urban Development, the administrative agency responsible for enforcing the FHA, had recognized the theory in administrative adjudications for more than 20 years before Texas Department of Housing & Community Affairs.[14]

But disparate impact liability remained controversial because of changes in other areas of the law.  In the mid-1970s, the Court decided two cases under the Fourteenth Amendment’s Equal Protection Clause where it explicitly rejected disparate impact liability and required a showing a discriminatory intent.[15]  The general notion that discriminatory intent is required to pursue a discrimination claim was soon incorporated by lower federal courts into federal anti-discrimination statutes.[16]  Indeed, the government defendant in Texas Department of Housing & Community Affairs argued for an intent-based standard under the FHA.[17]

Further, disparate impact claims purport to impose liability for actions that may be incidental and not entirely foreseeable.  For example, a mortgage lender who asks about the criminal history of a loan applicant could face FHA liability if the question has the effect of disqualifying the majority of African-American applicants.  Similarly, a mortgage lender who considers the characteristics of loan applicants’ neighborhoods in its underwriting process may encounter legal scrutiny if a plaintiff shows that some loans were denied because of the racial composition of the applicants’ neighborhoods.  But both instances might be justified on legitimate business grounds: in the first example, an applicant’s criminal history can be relevant in determining the applicant’s employment prospects, ability to pay the loan, and truthfulness; in the second example, deterioration of the surrounding neighborhood could affect the value of the property used to secure the loan.  

With analogous case law and the practical problems with disparate impact claims, why did the Court decide to recognize disparate impact liability under the FHA?  First, it was readily apparent that the historical circumstances surrounding the passage of federal anti-discrimination statutes during the Civil Rights Movement and the remnants of de jure residential segregation had some effect on the Court’s decision.  Several pages of the Court’s opinion are dedicated to explaining the FHA’s broader remedial purpose.[18]  Second, the Court relied upon the similarities between the catch-all phrase at issue in Griggs and a catch-all phrase in the FHA.[19]  The Court highlighted that, similar to Title VII in the employment context, the FHA makes it illegal      “[t]o . . . otherwise make unavailable or deny, a dwelling to any person because of race, color, religion, sex, familial status, or national origin.”[20] 

Third, the Court considered the fact that Congress failed to specifically prohibit disparate impact claims in the 1988 amendments to the FHA.[21]  The Court found that Congress was aware of unanimous precedent recognizing such claims from the federal courts of appeals when it amended the statute, and that Congress had rejected an amendment that would have eliminated disparate impact claims in certain circumstances.[22]  In the Court’s view, “the 1988 amendments signal[ed] that Congress ratified disparate-impact liability.”[23]

But Texas Department of Housing & Community Affairs answers some broader concerns about disparate impact liability, and leaves some opportunities for private lenders and government actors.  The Court first explained that disparate impact liability could play “a role in uncovering discriminatory intent” by allowing “plaintiffs to counteract unconscious prejudices and disguised animus that escape easy classification as disparate treatment.”[24]  The Court then noted that the FHA does not require “a particular vision of urban development; and it does not put housing authorities and private developers in a double bind of liability, subject to suit whether they choose to rejuvenate a city core or to promote new low-income housing in suburban communities.”[25]

The Court also stated that a mere statistical disparity is insufficient for disparate impact liability under the FHA “if the plaintiff cannot point to a defendant’s policy or policies causing that disparity.”[26]  This “robust causality requirement” requires a plaintiff to allege facts at the pleading stage or produce statistical evidence “demonstrating a causal connection” between the defendant’s policy and the alleged discriminatory effect.[27]  Disparate liability under the FHA cannot be imposed for “a one-time decision,” which suggests that only those policies or decisions that are continuous and have a systemic discriminatory effect are actionable.[28]

How does the Court’s recognition of a business necessity defense help potential FHA defendants? 

In addition, the Court recognized a business necessity defense similar to the defense recognized in employment cases under Title VII.[29]  The Court did not outline the contours of the defense, but lower court decisions have long recognized a business necessity defense.[30]  Because the business necessity defense requires an inquiry into the specific objectives of the defendant, raising the defense requires a fact-intensive analysis.[31]  But potential defendants, especially private lenders, should be aware of those business necessity defenses that have been accepted or rejected by lower courts under the FHA. 

Arguments based upon general “market forces” have been repeatedly rejected by federal courts.[32]  One federal court noted that “[i]t is precisely because the market could not self-correct for discrimination that statutes like Title VII, the FHA, and [the Equal Credit Opportunity Act] were necessary.”[33]  The court continued that because the market is impacted by “individual preferences,” “those preferences derived from racial and gender-based stereotypes,” which can lead to subjective criteria that has no relationship to a borrower’s creditworthiness.[34]  So, for example, if fees are included in a loan application that are unrelated to “objective indicators of creditworthiness,” then a plaintiff may have an actionable FHA disparate claim if they can show a disproportionate impact on a protected class.[35]

Contract formation defenses, however, have been successful in turning back FHA claims.  For example in Mitchell v. Century 21 Rustic Realty, the defendants were able to defeat a FHA claim because the plaintiffs failed to satisfy the contractual terms to purchase a home by the required deadline.[36]  The court found that because the defendants were later able to negotiate a contract with another purchaser who was offering a higher price and better terms, the defendants’ actions “were perfectly legitimate and completely within their rights.”[37]

Similarly, courts have rejected FHA claims where plaintiffs failed to show that they would have qualified for a loan under objective criteria set by the lender.[38]  InThomas v. First Federal Savings Bank of Indiana, the court rejected a FHA claim because the lender had a legitimate, non-discriminatory reason to refuse a refinance loan based upon the loan-to-value ratio of the plaintiffs’ home.[39]  The court in Thomas also highlighted “legitimate business criteria” that lenders can consider without fear of a potential FHA claim.  That criteria includes the creditworthiness of the borrower, the marketability of the property offered as security for the loan, and the potential resale value of that property, including the aesthetics of the neighborhood in which it is located.[40] 

But the most commonly utilized defense is to convincingly demonstrate that the plaintiff failed  to produce reliable statistical analysis or to highlight deficiencies in any statistical evidence produced by the plaintiff.[41] Claims that rest “entirely on conclusory analytics of highly generalized data” are likely to be rejected by courts.[42]  Yet statistical evidence of disparate impact coupled with other evidence will likely get a plaintiff past the summary judgment stage and create a question of fact for trial.[43]  For example, statistical evidence and slight direct evidence that a lender singled out a protected class for subprime mortgages or that it limited loans based upon the racial composition of certain neighborhoods will likely go before a jury.[44]  Thus, it important for lenders to be able to address a plaintiff’s statistical analysis as well as provide their own direct and statistical evidence showing that any disparate impact is caused by non-discriminatory factors.  

What’s next?

Because disparate impact has been a recognized theory under the FHA for four decades, lenders should have a set of best practices to minimize risk and reduce the possibility of litigation.  Those practices should be aimed at using underwriting criteria only related to a borrower’s creditworthiness and the likelihood of repayment on the loan.  Lenders should also ensure that loan officers and other employees are making decisions based upon clearly defined objective criteria and that subjectivity is minimized to the greatest practical extent.  This does not mean that lenders have to make bad loans or ignore the telltale signs that a borrower is a less than ideal candidate for a loan.  But lenders should be prepared to justify every lending criteria and decision using legitimate, non-discriminatory reasons.