Justice Anthony Kennedy continues to surprise. The swing justice wrote an important liberal opinion, holding Thursday for a 5-4 court that the Fair Housing Act prohibits not only intentional discrimination but also policy decisions that discriminate by having a disparate impact on minorities.
The Reagan appointee occasionally shows vestiges of his one-time conservatism, as in one half of his split votes in two recent death penalty cases. But as his vote in the Affordable Care Act case suggests, he's increasingly becoming a confirmed liberal vote. When (and if) he declares a constitutional right to gay marriage in the next few days, he'll enter the pantheon of great justices -- as a liberal.
The issue in the Texas housing case is a fundamental one for race relations today: What counts as discrimination? Under the Constitution, the U.S. Supreme Court has long held, the only discrimination that's prohibited is intentional discrimination, the kind where somebody consciously treats people differently on the basis of race or another prohibited classification.
But statutes can do what the Constitution does not. Under the Civil Rights Act, for example, it's unlawful to discriminate overtly -- but the law also bars discrimination in operation. The lawyers' name for that kind of practical discrimination is "disparate impact." Disparate impact discrimination can be demonstrated without any showing of intent. It's enough for there to be patterns of results that affect minorities differently.
The question before the court was whether the Fair Housing Act incorporates disparate impact discrimination. Kennedy said yes. His judgment rested on a close reading of the statute -- and he compared it to the Civil Rights Act as well as the Age Discrimination in Employment Act of 1967, both of which have been interpreted to allow for disparate impact claims.
What's important about the decision isn't the rationale but the real-world effects. Housing discrimination is a significant legacy of the broader history of racial discrimination throughout the U.S., in North and South alike. Proving intentional discrimination, however, is difficult to do -- especially when a faceless institution is involved.
In this case, the plaintiff alleged that the Texas Department of Housing and Community Affairs was allocating too many tax credits for low-income housing to black neighborhoods, as opposed to white ones. It's entirely possible that no one in the Texas office was consciously or knowingly engaged in racial discrimination. Yet the effects on housing markets, and by extension on racial equality, could be real, regardless of intent.
When the law is interpreted to allow disparate impact claims, it permits statistically based challenges to the reality of discrimination. It draws attention away from the fantasy that behind every discriminatory act lies some identifiable racial animus.
This kind of pragmatic approach is particularly desirable in this historical moment, when police discrimination has been so much a matter of public attention. Frequently, police officers have no conscious intent to treat blacks differently from whites. That's one reason discrimination can happen even when officers are black -- and even when a department is led by a black chief. Yet discriminatory effects can nevertheless be felt by blacks even if no one knowingly means to discriminate.
This outcomes-oriented view of discrimination doesn't necessarily have to be seen as liberal -- but in judicial practice, it has been. Justice Samuel Alito wrote the dissent in the fair housing case for the court's four conservatives. Naturally he rejected Kennedy's reading of the law and of the precedent. But strikingly, the centerpiece of his argument -- with which he began and ended -- was that the broader application of the Fair Housing Act would produce bad practical results, and "will have unfortunate consequences for local government, private enterprise, and those living in poverty." Alito's position is that disparate impact application of the law hamstrings governments that are trying to make things better for minorities. As one example, he cited a case where the Fair Housing Act was used to attack efforts by local governments to enforce housing condition ordinances aggressively in poor neighborhoods. The effect of the enforcement was to drive up rents in those neighborhoods, as landlords passed on the cost of improvements to their tenants. Because those neighborhoods were disproportionately inhabited by minorities, there was a disparate impact.
Alito also cited the facts of the current case as a source of the problem. If Texas allocates more tax credits to poor neighborhoods, then more people who live in the neighborhoods will be able to get housing. If it allocates more to rich neighborhoods, some poor people will be able to move into those neighborhoods. Either policy path was defensible; "either path, however, might trigger a disparate-impact suit." In practice, Alito acknowledged, the solution to such problems lies with the federal Department of Housing and Urban Development, which enjoys "enormous discretion" under federal regulations designed to implement the Fair Housing Act. To Alito and the conservatives, this discretion is a bad thing. But, of course, your view of the discretion depends entirely on your view of HUD's capacity to improve housing quality rationally.
Justice Clarence Thomas wrote separately to say that even the Civil Rights Act shouldn't be interpreted to allow disparate impact. This anachronistic view is more extreme than that of the other conservatives. But all of them seem locked in an ideal of anti-discrimination law that doesn't correspond to a contemporary reality. Discrimination is still with us even as conscious, knowing racism declines. Disparate impact remains a crucial tool for producing more equal outcomes in a more equal America.
Noah Feldman, a Bloomberg View columnist, is a professor of constitutional and international law at Harvard and the author of six books, most recently "Cool War: The Future of Global Competition."