It has been nearly three years since the Supreme Court declared the end of affirmative action in Students for Fair Admissions v. Harvard (SFFA) in June 2023. As a result, colleges and universities can no longer offer even a small advantage in admissions to applicants from minority groups underrepresented in their student bodies, such as Blacks, Hispanics, and Native Americans. The SFFA decision overturned multiple precedents from the past half-century that had permitted race-based preferences when they were employed not as quotas but as a modest factor in a holistic assessment of each individual applicant. In 2003 the Court upheld an affirmative action program at the University of Michigan Law School but stated that it “expect[ed]” that racial preferences would no longer be needed in twenty-five years. That was a prediction, not a deadline, yet the SFFA decision cut that period short by five years.
Only two incoming classes have been selected since the decision, so its long-term consequences remain to be seen. But it has already caused dramatic reductions in Black representation in the incoming classes of many of the nation’s most prestigious universities. In 2025 Black students made up only 11.5 percent of Harvard’s first-year class, down from 18 percent in 2023. At Princeton, Black students were 5 percent of the 2025 incoming class, compared with 9 percent two years earlier. Columbia’s proportion of Black freshmen fell from 20 percent to 13 percent, while Amherst’s dropped from 11 percent to 6 percent. Hispanic students have been less affected, while there has been increased enrollment of Asian Americans at many selective colleges, including Harvard, Princeton, Columbia, and Yale. A comprehensive study of the racial makeup of three thousand colleges and universities reported that at the nation’s fifty most selective schools, Black freshman enrollment dropped by 27 percent in just the first year after SFFA was decided, and the Hispanic first-year population decreased by 10 percent.1
To a majority of the Supreme Court justices, this is apparently what equality looks like. From their perspective, it’s not a constitutional problem that Black student enrollment has dropped so precipitously at the nation’s best universities; the problem is that Blacks were ever given any advantage at all. For Chief Justice John Roberts, it’s simple; as he wrote in 2007, “The way to stop discrimination on the basis of race is to stop discriminating on the basis of race.”
But to a majority of the Court for a half-century, it was more complicated than that. The Court recognized that affording underrepresented minorities a modest advantage in order to achieve the educational benefits of diversity was permissible. With Donald Trump’s appointments of Justices Neil Gorsuch, Brett Kavanaugh, and Amy Coney Barrett, however, Roberts had six votes in support of his view that any departure from “colorblindness” was impermissible, even when it is designed to include members of historically disadvantaged groups in a campus community.
The short- and long-term results of this colorblind mandate are likely to be anything but equal. If Black students are significantly underrepresented in top colleges, they will continue to be underrepresented in leadership positions in corporations, the professions, and government, as tomorrow’s leaders are largely drawn from the country’s most selective schools. That in turn will probably ensure that the already unconscionable racial wealth gap not only persists but grows. In 2022 for every one hundred dollars in wealth held by White households, Black households held only fifteen dollars; the median White household had $240,000 more wealth than the median Black household. That’s a peculiar vision of equality.
The effects of SFFA, moreover, are not limited to college admissions. Conservative public interest groups, emboldened by the Harvard case, have filed a spate of new legal challenges to other race-based initiatives, including scholarships, fellowships, and internships established to benefit members of minority groups. Such initiatives, the challengers claim, violate federal laws that prohibit race-based employment and contracting decisions. In their view, the fact that a program is designed to promote equity by giving an advantage to members of underrepresented racial minorities is no defense. SFFA did not change the statutory law that governs these disputes, but it has definitely altered the culture, and opponents of race-based advantages are seeking full enforcement. That in turn will mean fewer opportunities for Blacks and Hispanics to attain jobs and pursue careers in areas where they are already often scarce.2
At the same time, the Trump administration has aggressively misread SFFA in an attempt to eliminate all “diversity, equity, and inclusion” (DEI) programs on the grounds that they discriminate against White people. Trump issued an executive order banning any recipient of federal funds from implementing DEI programs “that violate any applicable Federal antidiscrimination laws” or that “constitute illegal discrimination or preferences.” The Department of Education issued a “Dear Colleague” letter to all schools receiving federal aid along similar lines. All sorts of institutions got the message and eliminated their DEI initiatives, sometimes scrubbing their websites of any references to the terms.
But the vast majority of DEI initiatives do not even conceivably violate federal law. It is entirely lawful, for example, to conduct training in how to understand people across racial differences, to celebrate Black History Month or Pride, or to promote student or employee groups that focus on the interests of Black, Hispanic, female, or LGBT persons, as long as they are open to all. Such classic DEI efforts are designed to make everyone feel included and do not burden or benefit anyone because of their race. The “colorblind” mandate does not require us not to talk about race or racism, and that’s all that most DEI initiatives do. A federal court enjoined the “Dear Colleague” letter because it went further than the law authorized, and in January the Trump administration dropped its appeal of that order. Still, far too many institutions have reacted out of fear and abandoned even wholly lawful measures designed to welcome students and workers of all races.3
The affirmative action debate is often cast as a clash between formal and substantive conceptions of equality, and that is how the majority and dissents on the Supreme Court have divided. To the majority, the Equal Protection Clause treats all race-based considerations as suspect, regardless of whether they are intended to benefit or to burden a historically disadvantaged minority group. According to this formal understanding, giving Black students an advantage in the application process based on their race is just as suspect as giving them a disadvantage on that basis. Any consideration of race must satisfy “strict scrutiny,” which requires that it be necessary to further a “compelling” government interest, the most demanding constitutional standard the Court uses.
To supporters of affirmative action, including Justices Sonia Sotomayor, Elena Kagan, and Ketanji Brown Jackson, who dissented in the SFFA case, equality requires more than mere colorblindness. They would treat efforts to benefit historically disadvantaged minority groups as less inherently suspicious than efforts to burden those same groups. From this substantive equality perspective, the persistent subordination of Black Americans is a problem that institutions should be permitted to seek to ameliorate. Insisting on formal equality in a fundamentally unequal society will only reinforce existing inequality and frustrate efforts to redress the lingering effects of historical and structural disparities. Thus the dissenting justices emphasized the continuing challenges faced by Black, Hispanic, and Native American populations and the historical roots of today’s disparities. A commitment to equality, in their view, requires us to recognize those disparities, not close our eyes to them.
But as Justin Driver, a professor of constitutional and education law at Yale Law School, shows in The Fall of Affirmative Action, it’s not nearly so simple, even if one adopts a substantive view. Advocates and justices on both sides of the affirmative action debate invoke substantive notions of equality; they just draw radically different conclusions about what furthers it. Justice Clarence Thomas, for example, has long argued that affirmative action harms Blacks, because it marks every Black student with the stigma of not being qualified on merit alone and thereby reinforces stereotypes of racial inferiority. Others have argued that affirmative action disserves Black students by admitting them to schools where they are less likely to be competitive, thereby setting them up for failure. According to this “mismatch” theory, Black students would be better off at a college where they would be more likely to compete on an equal footing with their peers.
And of course, Black students are not the only ones interested in substantive equality. The plaintiffs in the SFFA case were Asian Americans who claimed that their representation in Harvard’s class was depressed in order to make room for less qualified Black and Hispanic students. The litigation revealed that on average, Asian Americans did better than all other racial groups on grades, test scores, and extracurricular activities but generally received lower “personality” scores, a subjective assessment susceptible to stereotypes and prejudice.
This is the difficulty with conceptions of substantive equality: What’s the baseline for measuring equality, and whose substantive equality matters most? Asian Americans are overrepresented in the nation’s top colleges compared to their share of the general population. But if as a group they perform better than all other groups on the traditional measures of merit, they are likely underrepresented as compared to applicants with equal “merit”—itself a contested term. Should the baseline include merit or not? And it’s not as if Asian Americans have not suffered historical and ongoing discrimination and prejudice; consider, for example, the internment of Japanese Americans during World War II or the long history of discrimination against Chinese immigrants. Many supporters of affirmative action condemned Students for Fair Admissions, the conservative nonprofit that sued Harvard, for opportunistically focusing on the plight of Asian Americans when what it really cared about were Whites; Driver joins that chorus. But even if the organization was acting opportunistically, that doesn’t justify disregarding the burdens borne by Asian Americans. As a substantive equality matter, is it fair to sacrifice one minority group’s interests for another’s?
Even if one focuses solely on the groups that affirmative action was designed to help, how does one weigh the harms to African Americans that Justice Thomas and the “mismatch” advocates have identified against the benefits of attendance (for themselves and their classmates) that proponents cite? If some African Americans are benefited and others are burdened by affirmative action, what does a commitment to “substantive equality” demand? And should the equality assessment also consider the White students who are denied admission at least in part because of their race?
Driver candidly criticizes liberals for too often failing to take on these critiques of affirmative action. He takes them seriously, which is more than many liberals do. But once one acknowledges that affirmative action has nontrivial costs as well as benefits for members of historically disadvantaged minority groups, including those it is designed to assist, a credible proponent of affirmative action must show why the benefits nonetheless outweigh the costs.
To my mind, the benefits do outweigh the burdens, and a significant part of the answer lies in the sort of society we should seek to build if we believe in equality—one that fosters cross-racial equity and understanding. When our nation’s most influential institutions use racial preferences to include minority groups that have been and remain underrepresented there, they seek to promote a future in which racial caste and subordination are no longer built into our society. We are not there yet. Not even close. And the interest in eliminating such persistent inequality dovetails with, though is distinct from, the university’s particular interest in the educational benefits of a diverse student body.
Given how racially divided our society still is and that segregation has long been a central feature of inequality, we should foster integration where we can. Because colleges select their student bodies from a large number of diverse applicants, they have the opportunity to create integrated communities for young Americans at an especially important point in their personal development. That integration in turn will facilitate cross-racial understanding that will help break down the divisions that too often plague our multicultural nation.
For those who believe that the disproportionately low numbers of Black, Hispanic, and Native American students in the nation’s most selective schools remain a problem to be solved, Driver has some excellent suggestions. He proposes a series of “race-neutral” means of maintaining a racially diverse student body. They include assigning preferences in admissions to applicants from underrepresented urban neighborhoods or public schools and from federally recognized tribes, children of first-generation or second-generation immigrants, individuals who are the first in their family to attend college or come from families with low net wealth, and the descendants of slaves.
These preferences are all as a formal matter race-neutral because they do not offer an advantage on the basis of race itself. Even offering a benefit to descendants of slaves is race-neutral because the category is not defined by race. The descendants of slaves will, of course, virtually all be Black, but plenty of Black applicants would not qualify, and members of other racial groups might. It is not race that determines the benefit but whether one’s ancestors were slaves. Driver’s other proposals are even more clearly formally neutral as to race.
All these measures will disproportionately assist members of minority groups, but that effect in itself is insufficient to call them into question as a constitutional matter. Under the same commitment to formal equality that doomed race-conscious admissions in SFFA, the Court has long ruled that the fact that a policy that is race-neutral on its face has a disparate impact on a racial group is insufficient to violate equal protection; it interprets the Equal Protection Clause to prohibit only intentional discrimination. And the Court has defined “intent” as requiring much more than foreseeable consequences. Rather, a plaintiff must show that a policy was adopted “‘because of,’ not merely ‘in spite of,’” its disparate effects. On this reasoning, the Court has upheld a civil service test that Blacks failed at higher rates than Whites and a Massachusetts rule giving veterans preference for civil service jobs that foreseeably disadvantaged women, because the challengers could not show that the policies were adopted for the purpose of disadvantaging individuals based on race or sex.
But here’s the problem: Driver’s avowed purpose in proposing the “race-neutral” preferences he outlines is to promote Black, Hispanic, and Native American enrollment. If universities adopted them for that purpose, they would arguably be doing so “because of, not merely in spite of,” their disparate racial effects. And under the Court’s equal protection jurisprudence, facially neutral measures adopted for a racial purpose are just as suspect as explicitly race-based measures. That means strict scrutiny would apply, and under the SFFA decision, it’s not clear that they would survive that scrutiny.
The SFFA decision rejected Harvard’s asserted end of achieving the educational benefits of a racially diverse student body. The Court reasoned that the benefits claimed were simply too amorphous to be measured and therefore resisted application of strict scrutiny. Courts applying strict scrutiny must determine whether a challenged policy is narrowly tailored to further a compelling interest, and if the compelling interest resists a clear definition, it can be difficult to determine what constitutes a close fit between means and end.4 During oral argument in SFFA, the justices repeatedly asked how schools or courts would know when racial preferences are no longer necessary to achieve the educational benefits of diversity, and the majority found the responses inadequate. But if racial diversity is therefore not a compelling interest, then it’s not clear how race-neutral measures adopted to further that end could satisfy strict scrutiny.
Opponents of affirmative action have already set their sights on just such race-neutral means. The Court had an opportunity to address the issue in 2024, when a Virginia school district changed the admissions process for a public magnet school in order to increase its numbers of Black and Hispanic students. Predictably, Asian American enrollment dropped and Black and Hispanic enrollment increased under the new system, and parents of Asian American students sued. The new admissions program was race-blind; indeed, the applications themselves were stripped of anything that would permit identification of an applicant’s race. But the school district changed the criteria it used for a racial purpose. The US Court of Appeals for the Fourth Circuit upheld the program over a vigorous dissent. The Supreme Court denied review, with Justices Thomas and Samuel Alito dissenting.
The Supreme Court’s decision not to take up the Virginia case could mean one of two things. It is possible that a majority of the Court is comfortable with schools adopting race-neutral means to pursue diversity. But the Court might instead have felt that ruling on the question less than a year after SFFA was premature, especially in the absence of a clear disagreement in the circuit courts on the issue. It’s one thing to tell colleges they have to stop giving even modest racial preferences; it would be another matter entirely to prohibit even race-neutral efforts to ensure a diverse student body. If schools are barred from deploying race-neutral measures, the resulting reduction in enrollment of historically underrepresented groups would make the current declines look gradual. Several states, including California and Michigan, prohibited race-based affirmative action as a matter of state law long before SFFA, and the only way the top schools in those states maintained any meaningful diversity was by adopting race-neutral measures designed with that goal in mind.
Prior decisions of the Supreme Court seemed to find no problem with such initiatives. When the Court considered a challenge to affirmative action at the University of Texas at Austin, for example, it focused on whether the school could consider race directly; the Court did not question the school’s practice of guaranteeing admission to any student in the top 10 percent of their public school’s graduating class, a practice that, because Texas high schools are quite segregated, offered admission to a significant number of Black and Hispanic students who performed well at their predominantly Black or Hispanic high schools.
The race-neutral measures that Driver recommends and that many schools already employ could avoid strict scrutiny if they are adopted for purposes other than increasing racial diversity. A college might choose to give an admissions edge to inner-city residents, low-income students, children of immigrants, or even the descendants of slavery to promote socioeconomic or other forms of nonracial diversity or equity. In that event, no special judicial scrutiny would be necessary, because as we have seen, the Equal Protection Clause subjects to strict scrutiny only efforts adopted for a racial purpose, not those adopted for other purposes that happen to have a racial impact. Progressives long ago lost the battle to subject to skeptical judicial review laws that have a disproportionate effect on racial minorities, but ironically, that means colleges should have a free hand to take measures that give an advantage to groups defined by wealth, income, geography, or heritage, rather than by race.
But what happens if a court concludes that an admission policy was in fact adopted for the purpose of increasing racial diversity? To uphold such measures, the Court will either have to rethink the permissibility of diversity as a goal or soften its insistence that the Constitution is just as suspicious of measures designed to help disadvantaged minority groups as those intended to harm them. There is no doubt that a facially race-neutral policy intentionally adopted to disadvantage Black students, such as giving preference to students who are not descendants of slaves or residents of impoverished inner-city neighborhoods, would not survive strict scrutiny. If race-neutral measures designed to increase Black enrollment are to be tolerated, and at least for the time being they are, that must mean either that racial diversity is a compelling interest while racial exclusion is not or that the Equal Protection Clause does not actually assign equal suspicion to measures helping or harming members of disadvantaged minority groups. Either conclusion would make eminent sense, but they would both be in some tension with SFFA. Ultimately the question is whether a majority of the Court would be comfortable, in the pursuit of “formal equality,” with eliminating all but a minuscule representation of Black students at the nation’s top universities.
English (US) · 