The Supreme Court's decision on housing discrimination

Texas Department of Housing and Community Affairs v. Inclusive Communities Project, Inc., decided on June 25, 2015, understandably received much less attention than the Supreme Court’s same-sex marriage decision the following day.  But Inclusive Communities is an important case nonetheless. (Disclosure: I assisted two New York Law School colleagues who, together with students, wrote an amicus brief broadly supporting the result the Court reached.)

The question the case addressed was whether the Fair Housing Act forbids both action that deliberately discriminates based on race, and also action that has a racially discriminatory effect. Deliberate discrimination in this context is called “disparate treatment”; action with racially discriminatory effects, but no proven discriminatory purpose, is described as having a “disparate impact.” As lawyers would put it, then, the question was whether the Fair Housing Act creates liability not only for “disparate treatment” but also for “disparate impact.”

It’s been settled since Griggs v. Duke Power Co., decided by the Supreme Court in 1971, that the following language of the Civil Rights Act of 1964 forbade both disparate treatment and disparate impact in employment discrimination cases:

            Section 703(a): It shall be an unlawful employer practice for        an employer …

(2) to limit, segregate, or classify his employees or applicants for employment in any way which would deprive or tend to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual’s race, color, religion, sex, or national origin.

That case is settled law. The Inclusive Communities case was about two sections of a different statute, the Fair Housing Act. The language of the two sections varies a bit, but the central words at issue can be seen here, in section 804(a), which makes it unlawful:

To refuse to sell or rent after the making of a bona fide offer, or to refuse to negotiate for the sale or rental of, or otherwise make unavailable or deny, a dwelling to any person because of race, color, religion, sex, familial status, or national origin.

The Supreme Court, by a 5 – 4 vote with Justice Kennedy writing the majority opinion, concluded that this language did encompass “disparate impact” as well as “disparate treatment” liability. Justice Kennedy writes (at 10) that Griggs and the plurality opinion in a subsequent case, Smith v. City of Jackson (2005), show “that antidiscrimination laws must be construed to encompass disparate-impact claims when their text refers to the consequences of actions and not just to the mindset of actors, and where that interpretation is consistent with statutory purpose.”

Here, the Court says, the language that “refers to the consequences of actions” is the phrase “otherwise make unavailable” (11). That’s a possible reading, but not, it seems to me, an inevitable one. “[O]therwise make unavailable” might be meant, as the Court is suggesting, to refer to acts of unintentional discrimination – in contrast to deliberate “refusals” or “denials.” But one might instead say that “otherwise make unavailable” refers to intentional acts that just aren’t the same intentional acts as “refusals” or “denials.” A “failure to respond” – unaccompanied by an express refusal or denial – might be an example.

If the phrase “otherwise make unavailable” is ambiguous, we might look a few words further to the phrase “because of race.” What does it mean to make housing unavailable “because of race”? The dissent’s answer is that “because of” means “by reason of” or “on account of,” and if there’s any ambiguity in those definitions, the dissent goes on to cite a lower court case spelling out that “the terms [after] the ‘because of’ clauses in the FHA supply the prohibited motivations for the intentional acts … that the Act makes unlawful.” (Dissenting opinion of Alito, J., at 5). Alito maintains that this is simply what “because of” means in our language – and to make his point he cites 14 articles using the phrase in the Washington Post on the day this case was argued. (Id. at 4 – 5 n.2.)

If he’s right that the words “because of” have this meaning, and can only have this meaning, then the FHA must be insisting on actions “motivated by” race – “disparate treatment” – and must not be making actions with merely “disparate impact” unlawful.

But it’s quite clear that “because of” in our law does not always mean “motivated by.” We know that from Griggs itself, where the same “because of” language appears – and did not stop the Supreme Court from finding that the Civil Rights Act of 1964 prohibited employment decisions that had “disparate impact” on particular racial groups.

Still, one might say that Griggs was mistakenly decided, and that its mistake should not be propagated – and this is exactly what Justice Thomas does say in his separate dissent. So it is important to ask whether “because of” could possibly mean, as a matter of English language usage, something other than “motivated by.”

The answer, it seems to me, is clearly yes. Suppose we ask, for instance, why black students as a group do less well than whites on the SAT exam. We might answer, “because of race.” What we would mean is not that the SAT deliberately discriminated against black people, but that the SAT’s negative impact on black students as compared to whites results from the many ways that race in our society contributes to social and educational disadvantage. “Because of race,” when used this way, is shorthand for “as a result of the interaction between the action taken and the social conditions that race has shaped.” And this is anything but fanciful: the massive, systematic socioeconomic disadvantages facing black people in the United States today can’t plausibly be explained as the sum of acts of conscious, purposeful racial discrimination, persistent as that kind of discrimination is; there’s much more to it than that, and that “much more” is encompassed in the words “because of race.”

There’s also more to the Inclusive Communities case, but it’s important to recognize, right away, that speakers of English do use the words “because of race” to refer to “disparate impact” and not only to “disparate treatment.” So the plain meaning of the statute’s words did not rule out the Court’s interpretation.

Nontextual statutory interpretation -- American and South African

One might say that the difference between the two statutory interpretation cases I've recently discussed is the difference between an American jurisprudence obsessed with text and with circumscribing power, and a South African jurisprudence committed to shaping law to transform a nation and therefore determining the meaning of law in light of its purpose rather than its words. And this idea does, indeed, capture an important difference between U.S. and South African law.

But it would be wrong to think that nontextual statutory interpretation is unknown in the United States. Certainly we have a history of such interpretation, dating back to the classic case of Rector of Holy Trinity Church v. United States, 143 U.S. 457 (1892). Holy Trinity asserted -- and not without precedent -- that "[i]t is a familiar rule that a thing may be within the letter of the statute and yet not within the statute, because not within its spirit nor within the intention of its makers." 143 U.S. at 459. But Holy Trinity has largely fallen from favor in recent years. In an opinion concurring in the judgment (but not in the more free-wheeling majority opinion) in the case of Public Citizen v. United States Department of Justice, 491 U.S. 440 (1989), Justice Kennedy noted that Holy Trinity affirmed that a statute should be interpreted in light of the supposed fact that the United States is a "Christian nation." He responded, 491 U.S. at 474, that "I should think the potential of this doctrine to allow judges to substitute their personal predelictions [sic] for the will of the Congress is so self-evident from the case which spawned it as to require no further discussion of its susceptibility to abuse."

Still, reports of the demise of nontextual reading of statutes in the United States are premature. Broadly speaking, it seems to me that the text fades in significance when the constitutional stakes go up. The more a statute involves constitutional values, the more likely it will be read in light of those values rather than in light of its text.

For a striking recent example, consider Ricci v. DeStefano, 557 U.S. 557, 129 S. Ct. 2658 (2009). This is the well-known case from 2009 in which the Supreme Court addressed a decision by the City of New Haven, Connecticut to set aside the results of a test taken by firefighters seeking promotion, a test under which no blacks at all would have been eligible for promotion. New Haven maintained that it could not use the test results because it "had a good-faith belief" that doing so would itself have constituted illegal discrimination against minorities. 129 S. Ct. at 2671. The Supreme Court, however, decided that the city's decision was actually illegal discrimination against those firefighters -- all of them white, except for two Hispanics -- who had done well on the tests.

As the Court explains, the federal statute that bans employment discrimination on the basis of race, Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq., actually forbids two different kinds of discrimination. One is disparate impact discrimination -- unequal results that lack a justification in "business necessity," whether or not they are themselves the result of purposeful bias. The other is disparate treatment discrimination, which is precisely the result of purposeful bias. Here, New Haven said it had acted in order to avoid the risk of falling into disparate impact discrimination, as it would have if the promotional test it used -- and on which whites did far better than blacks -- was not one that fairly measured applicants' qualifications. But in so doing it in a sense discriminated against the whites who had done well on the test. So the question was -- or at least this was one way to view it -- when is an employer legally entitled to deliberately disadvantage whites in order to avoid discriminating unintentionally against blacks? See 129 S. Ct. at 2674.

The problem was that the statute offered absolutely no express answer to this question. So far as the justices' opinions reflect, Congress never considered the potential for such a conflict arising; quite possibly, as Justice Ginsburg suggested in dissent, 129 S. Ct. at 2699, Congress considered its two prohibitions -- of disparate impact and disparate treatment discrimination -- to be congruent and mutually reinforcing, rather than potentially in conflict with each other. A majority of the Supreme Court seems to have been uncomfortable with the "disparate impact" idea for two decades (a history recounted by the dissent, 129 S. Ct. at 2699, though unmentioned by the majority). But Congress may simply not have seen good faith efforts by employers to avoid disparate impact on minorities as raising any issue at all of disparate treatment of whites.

Whatever the history, however, both the majority and the dissent in Ricci acknowledged the possibility that the two duties -- not to create disparate impact discrimination and not to create disparate treatment discrimination -- could come into conflict. The majority, in an opinion written by Justice Kennedy, decided that an employer could legally disadvantage whites in order to avoid disparate impact discrimination against blacks, but only if it had a "strong basis in evidence" for its claim that unless it did so it would in fact be in violation of the prohibition on disparate impact discrimination. 129 S. Ct. at 2675-76. That was a standard the majority took quite directly from one of its cases construing the equal protection clause of the Fourteenth Amendment (a central constitutional protection against race discrimination) to limit affirmative action. The dissent responded that that test was much too stringent, and that all that should be required was that "[t]he employer must have good cause to believe the device [here, the promotion test] would not withstand examination for business necessity." 129 S. Ct. at 2699.

One could argue about which standard is more appropriate. My point, though, is that the statute didn't say. In prohibiting disparate impact and disparate treatment, the statute ultimately left it to the Supreme Court to say when each of those arose, and how to handle collisions between them. The judges responded by importing their constitutional views into the shaping of the answer. (I should acknowledge that Justice Ginsburg responds to the majority's use of constitutional precedent by saying that "equal protection doctrine is of limited utility." 129 S. Ct. at 2700. But I take that statement less as a suggestion that Title VII should be interpreted without regard to the constitution than as a reflection of the dissenters' view that current constitutional equal protection doctrine is itself problematic. Race discrimination in the United States, it seems to me, is always an issue with constitutional overtones.)

So, too, South Africa's Supreme Court of Appeal brought constitutional considerations directly to bear on the statutory text at issue in the Democratic Alliance case. In doing so, the Supreme Court of Appeal sought to constrain executive power, and it seems clear that its central goal was to find legal tools to combat the danger of corruption in South African public life -- a noble objective. The United States Supreme Court in Ricci used a similar technique to challenge what it saw as the perpetuation of racial division and racial discrimination in the United States -- another noble goal, phrased that way.

But many people see the Ricci case as part of the Supreme Court's rejection of affirmative action and race-conscious remedies for the discrimination against blacks and other minorities that has so deeply injured the United States -- and so they do not see the Court's effort in Ricci as a wise one at all. It might similarly be argued that the Democratic Alliance decision interfered with legitimate executive authority, and that in taking this step, even for the best of reasons, the court imperiled the position of the judiciary in South Africa's polity, potentially with results even worse in the long run than those that might have resulted from the appointment at issue in that case.

These are big issues, and reasonable people can disagree about them. What I want to emphasize is  that when a court reads a statute based on its understanding of constitutional values rather than of the words of the statute, it has taken on a profound responsibility for shaping the nation. If it is clear what the constitution commands, of course, one can fairly maintain that the courts are honoring the fundamental democratic choices of the country by reading them into the less fundamental words of ordinary statutes. But if it isn't clear what the constitution commands -- and sooner or later the constitution's meaning does become a subject of debate and controversy, as for example in the case of the Fourteenth Amendment's equal protection clause in U.S. law today -- then the choices being implemented are no longer necessarily those of the people and may instead be those of the judges. Such choices are, I think, sometimes needed and right. They may also be unavoidable, notably when the statute simply does not answer the question that it has now come to be seen as posing. But it's worth remembering that when such judicial choices are made they take the fate of the nation, to a certain extent, out of the democratic process.