A few days before the opening of its new term, the Supreme Court accepted for review a case from Texas that could prove one of the Court’s most important this year—provided that the justices actually get to decide it.
The case is Texas Department of Housing and Community Affairs v. The Inclusive Communities Project, and it concerns the Fair Housing Act of 1968, which prohibits racial discrimination in housing, as well as discrimination on other grounds including ethnicity, sex, and religion.
Under the statute, it’s clear that claims of “disparate treatment”—where individuals are treated differently in some housing matter because of their race—may be litigated, with the accusing party required to prove discriminatory intent or motive.
But—and this is the single question presented in the Texas case—does the FHA also authorize claims of “disparate impact,” in which plaintiffs need show only that a practice neutral on its face and nondiscriminatory in its intent has a disproportionate effect on some group of persons defined in terms of one of the prohibited characteristics?
If the issue sounds familiar, some recent history is worth recalling. In 2011, the Supreme Court took a case from the Eighth Circuit (Magner v. Gallagher) that presented the same question the Texas case does now. But the Court was unable to decide Magner because the case was withdrawn before the oral argument. In 2013, the Court again accepted for review a case raising the issue, one from the Third Circuit (Township of Mount Holly, New Jersey v. Mount Holly Gardens Citizens in Action). This case was settled as the argument drew near, again leaving the Court nothing to resolve.
Those two circuit courts are among the many that have decided matters involving disparate impact claims brought under the FHA. Only the D.C. Circuit has lacked occasion to rule on whether the FHA authorizes such claims. The other circuit courts have concluded or assumed it does, while differing markedly on the standards and burdens of proof that the courts should apply in reviewing the claims.
As for the Supreme Court, it has had two FHA cases involving disparate impact theory. But in neither did the Court consider the threshold question whether the law actually allows such lawsuits. The justices seem prepared to take up that question—have been, in fact, since 2011.
The issue in the Texas case is how the state’s housing department distributes federal tax credits to developers who build low-income housing. The allegation is that between 1999 and 2008 the department disproportionately allocated tax credits to projects in minority-populated areas, thus also disproportionately denying tax credits to projects in mainly white neighborhoods—in violation of the FHA’s prohibition of racial discrimination.
Statistics are at the heart of all disparate impact claims, and in the district court, the statistics offered in support of the Texas plaintiff’s claim were these: that during the years at issue the department approved tax credits for 49.7 percent of proposed non-elderly units in areas between 0 percent and 9.9 percent white, but only 37.4 percent of such units in areas between 90 percent and 100 percent white. For the district court, that statistical disparity was enough to establish a “prima facie case” of discrimination, thus flipping to the housing department the burden of proving that it had not discriminated on the basis of race.
The department offered an explanation for the statistics: Federal and state laws require it to distribute the credits according to fixed criteria, and some of these correlate with race. But the district court held that the department was unable to show its inability to narrow the disparity in approval rates. The court held for the plaintiff on its disparate impact claim, a judgment that was affirmed by the Fifth Circuit.
There is reason to think the Supreme Court will side with Texas. This is a court interested in getting right the meaning of texts, and the text of the FHA plainly understands discrimination as a specific act of intentional discrimination. It contains no language establishing disparate impact liability.