Thoughts on the Supreme Court's Ruling in the Harvard and UNC Racial Preferences Cases

A preliminary assessment of today's decisions. The majority rightly struck a blow against the use of racial preferences for purposes of advancing "diversity" in education. But there are some flaws in its reasoning.

Today, the Supreme Court ruled against Harvard and the University of North Carolina in cases challenging the legality of their use of racial preferences in student admissions. The decision severely restricts, even if it doesn't completely ban, the use of racial preferences of purposes of achieving "diversity" in educational institutions. Chief Justice Roberts' majority opinion does an excellent job of laying out many of the flaws in diversity preferences, including nebulous goals, reliance on crude racial classifications and stereotypes, and the unconstitutional use of race as a "negative" to disadvantage Asian-American applicants, among others. Justice Neil Gorsuch's concurrence correctly points out that the cases could have been resolved more easily by relying on the plain text of Title VI of the Civil Rights Act of 1964.

On the downside, the Court did a poor job of reconciling its decisions with previous precedents giving much broader leeway for "diversity" preferences, most notably  Grutter v. Bollinger (2003) and  Fisher v. University of Texas II (2016). Some parts of the majority opinion could also potentially enable the continuation of some racial preferences in disguise.

No blog post could do justice to the 237 pages of majority, concurring, and dissenting opinions in this case! But I will try to expand somewhat on several key points.

First, it's important to remember that Harvard and UNC justified their use of racial preferences by reference to the supposed educational benefits of racial and ethnic "diversity." Even if you think affirmative action can be justified on some other basis, such as compensating for historical injustice, today's rulings are focused on the far more dubious diversity rationale.

And, as Roberts and Gorsuch effectively explain, that rationale is so full of holes that it can't possibly pass muster under the "strict scrutiny" imposed on the use of racial classifications. For example, the racial categories into which Harvard UNC divide up applicants (black, white, Latino, Asian, etc.) are extremely crude and verge on simplistic stereotyping. As Gorsuch points out, the "Asian"  category "sweeps into one pile East Asians (e.g., Chinese, Korean, Japanese) and South Asians(e.g., Indian, Pakistani, Bangladeshi), even though together they constitute about 60% of the world's population." The other categories are not much better. WASPS, Jews, and immigrants from Bulgaria and Sweden are all equally "white." "Latino" likewise includes people from a vast range of nations and cultures.  "Black" lumps in native-born African-American descendants of slaves with immigrants and children of immigrants from a wide range of countries in Africa and the Carribbean. The "narrow tailoring" required by strict scrutiny surely compels a far more nuanced assessment.

As Chief Justice Roberts explains, this kind of lumping also inevitably leads to crude stereotyping, based on the assumption that all members of these broad categories have relatively similar views and backgrounds, different from those of all the other broad aggregates. That is pretty obviously false in many cases. For example, an upper-middle class white person probably has much more in common with a native-born African-American from the same economic background in the same city, than either is likely to have with an immigrant from Bulgaria or Nigeria, even though the former is classified as "white," and the latter "black." Along related lines, the exchange between  Clarence Thomas' concurring opinion in today's cases and Ketanji Brown Jackson's dissent powerfully demonstrates how two native-born African-Americans from southern states can have vastly different perspectives on the the black American experience, its history, and what that history implies for today.

The crudeness of the racial and ethnic categories used by Harvard and UNC also undercut Justice Jackson's otherwise powerful appeal to the historic disadvantages faced by African-Americans. She is obviously right that blacks are on average worse off than whites on various social and economic dimensions, and that the legacy of slavery, Jim Crow, and other discrimination is a large part of the reason why.

But even if blacks are worse off, on average, that doesn't mean that all or even most black applicants to elite institutions like Harvard have suffered greatly from discrimination. Many are relatively affluent members of the upper middle class. Conversely, many of those discriminated against by affirmative action themselves come from groups with their own histories of disadvantage and discrimination (most notably Asians).

One of my black classmates at Yale Law School was the son of the attorney general of his state. That does not mean his life was free of racism (for example, he still experienced racial profiling by police, which is a serious injustice opponents of affirmative action, including my fellow libertarians, should pay more attention to). But it seems unlikely he was, overall, as disadvantaged as, say, a recent Asian immigrant, or a poor white applicant from Appalachia.

The horrific historic injustices suffered by African-Americans do not justify lumping them all into one group for purposes of racial preferences. The same goes for whites, Latinos, and others.

If the racial and ethnic categories Harvard and UNC use are nebulous and crude, the same applies to the goal these categories are supposed to serve. As Roberts also explains in detail, it is hard to say what is meant by "diversity," what the educational benefits of it are, and how we can measure whether and to what extent they have been achieved.

None of this matters much if you think universities should be given broad discretion to use racial preferences, so long as it is for seemingly good motives. But even supporters of affirmative action usually acknowledge that there needs to be at least some rigorous scrutiny of government's use of racial classifications. They can't just be given a pass, like run of the mill government policies. That's true whether you are an originalist, a living constitutionalist, or some combination of both.

Consider, for example, Justice Ruth Bader Ginsburg's statement that "[t]he mere assertion of a laudable governmental purpose, of course, should not immunize a race-conscious measure from careful judicial inspection…. Close review is needed 'to ferret out classifications in reality malign, but masquerading as benign,' Adarand, 515 U.S., at 275 (Ginsburg, J., dissenting), and to 'ensure that preferences are not so large as to trammel unduly upon the opportunities of others or interfere too harshly with legitimate expectations of persons in once-preferred groups.'" The Harvard and UNC policies can't possibly survive a genuine "careful judicial inspection."

In the majority opinion, Chief Justice Roberts also emphasizes that race cannot be used as a "negative" in university admissions. In one sense, as he also points out, any use of racial preferences creates such a negative for non-preferred groups. That's an inevitable feature of the zero-sum nature of college admissions at selective institutions.  But he also notes evidence that Harvard specifically tried to restrict the percentage of Asian-American applicants admitted, through the use of various devices, such as giving them lower personal ratings. In different ways, Roberts, Gorsuch, and Thomas note that discrimination against Asian applicants makes a mockery of the "diversity" rationale, and also of the idea that racial preferences are supposed to benefit historically disadvantaged groups. After all, Asians themselves have a long history of being victimized by state-sponsored discrimination, of which the detention of Japanese-Americans in internment camps during World War II is just one of many examples.

The Court could, however, have avoided the need to go into the details of the Harvard and UNC programs if it had simply decided these cases based on Title VI of the Civil Rights Act of 1964, which states that "No person in the United States shall, on the ground of race, color, or national origin, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any program or activity receiving Federal financial assistance."

There is no exception here for racial and ethnic preferences adopted for purposes of promoting diversity, or indeed for any other reason. Harvard, a private institution, is actually covered only by Title VI; it is not constrained by the anti-discrimination requirements of the Equal Protection Clause of the Fourteenth Amendment, which only applies to government entities (including UNC). Gorsuch is right to argue that the Harvard and UNC cases could have been resolved based on Title VI alone. However, only Thomas was willing to join Gorsuch's concurring opinion on this point.

That may be because previous Supreme Court decisions have ruled that Title VI's antidiscrimination standards are identical to those of the Equal Protection Clause, and the Supreme Court has a high bar for overruling statutory precedents, reaffirmed just recently in Allen v. Milligan. To my mind, the deviation from the plain text of Title VI is so egregious and so poorly reasoned that overruling statutory precedent would have been justified here. But the majority of justices clearly don't agree.

Less excusably, the Roberts' majority opinion and Brett Kavanaugh's concurrence play fast and loose with the Court's affirmative action precedents, such as Grutter and Fisher II. They contend that today's majority opinion is completely compatible with those previous precedents and doesn't require any significant modification of them. I won't try to go over the various convoluted details here. But I think the dissents by Justices Jackson and Sotomayor effectively point out that these precedents give far greater deference to university decision-making on affirmative action policy than does today's majority opinion. The majority would have done better to overrule Grutter, or at least significantly limit its scope.

The failure to overrule or limit Grutter leaves open the possibility that "diversity" might still be a compelling state interest that could justify the use of racial preferences in admissions, in at least some circumstances. Justice Thomas says that "[t]he Court's opinion rightly makes clear that Grutter is, for all intents and purposes, overruled." I don't think so. Otherwise, there would be no need for Roberts' and Kavanaugh's elaborate efforts to square their reasoning with that precedent. That said, the majority opinion does make it very hard to justify anything like the kinds of crude racial classifications used by many universities today.

In a footnote, Chief Justice Roberts notes that today's decision does not apply to the special context of "the nation's military academies." There are indeed special justifications for affirmative action in the military context that probably don't apply elsewhere. Still, this is another sign that the decision doesn't categorically ban all racial preferences, even in higher education.

The majority also emphasizes that universities can still consider applicants' experiences of racial discrimination and other effects that racial or ethnic identity may have had on their lives:

[N]othing in this opinion should be construed as prohibiting universities from considering an applicant's discussion of how race affected his or her life, be it through discrimination, inspiration, or otherwise…. But, despite the dissent's assertion to the contrary, universities may not simply establish through application essays or other means the regime we hold unlawful today. (A dissenting opinion is generally not the best source of legal  advice on how to comply with the majority opinion. "[W]hat cannot be done directly cannot be done indirectly. The Constitution deals with substance, not shadows," and the prohibition against racial discrimination is "levelled at the thing, not the name." Cummings v. Missouri, 4 Wall. 277, 325 (1867). A benefit to a student who overcame racial discrimination, for example, must be tied to that student's courage and determination. Or a benefit to a student whose heritage or culture motivated him or her to assume a leadership role or attain a particular goal must be tied to that student's unique ability to contribute to the university. In other words, the student must be treated based on his or her experiences as an individual—not on the basis of race.

Roberts is right that universities can legitimately consider applicants' individual experiences with racial discrimination, and other ways in which their racial and ethnic backgrounds may have affected them. He's also right that such consideration should not become a smokescreen for reintroducing racial preferences by the back door. But I worry, nonetheless, that many institutions will try to do the latter under the guise of the former.

More generally, it is likely that many institutions will try to replace explicit racial preferences with seemingly "race-neutral" alternatives that try to target characteristics that correlated with membership in a particular racial or ethnic group. Such subterfuges were used on a large scale to try to resist desegregation after Brown v. Board of Education. And we already see them in recent efforts to preserve racial preferences for blacks and Latinos, and keep down the percentage of Asian students at selective institutions.

It is also not entirely clear what the implications of today's ruling are for racial preferences outside education, such as in the field of government contracting. The conservative majority on the Court is likely to take a dim view of those preferences, as well. But exactly how dim is hard to tell.

Despite these and other caveats and shortcomings, today's decisions are an important step in the right direction. They won't put an end to all use of racial preferences. But they reaffirm and extend the fundamental principle that such discrimination is deeply unjust, and at least presumptively unconstitutional.

UPDATE: I wrote and posted this before seeing Will Baude's insightful post on the same topic. I agree with most of what he says there.