Clinging to Racial Preferences

Too many on the left stubbornly hang on to a failed and unconstitutional approach.

Democratic politicians grappled with the issue of racial preferences for decades without much success at reconciling competing beliefs. On the one hand, the American public has long been against the practice. In 2020, even as liberal California voters supported Joe Biden over Donald Trump by an overwhelming 29 points, an effort to reinstate racial preferences was soundly defeated by 14 points. On the other hand, Democratic interest groups in Washington, D.C., have been diehard supporters of racial preference policies, driving positions on these issues that many politicians feared to challenge.

So, for years, Democratic politicians spoke one way, then acted another. In 1995, President Bill Clinton launched a trial balloon, saying he wanted to shift the basis of affirmative action from race to economic need, but he backed down after interest groups rebelled. More than a decade later, presidential candidate Barack Obama said he thought his own daughters did not deserve racial preferences in college admissions and that working-class students of all races did. When I told a top Obama staffer after the election that I would like to help the new administration develop a class-based affirmative action program, however, I was told there was no way Obama could go against powerful Democratic interest groups. The courts would have to force him to make the shift.

The Supreme Court’s Political Gift to Democrats

In 2023, after years of waffling, the U.S. Supreme Court acted decisively on the matter. In Students for Fair Admissions v. Harvard, the justices upended 45 years of precedent and struck down racial preferences in college admissions. While Democrats expressed outrage at the time, the Court indirectly liberated them from their political confines. Democrats could tell interest groups privately that they supported racial preferences, but their hands were tied. The courts had ruled. A political albatross having been removed, Democrats could instead champion the broadly popular idea of affirmative action based on economic need, for which Clinton and Obama had articulated support but been unable to deliver.

Two years after the Students for Fair Admissions decision, however, many on the left still don’t want Democrats to accept the Supreme Court’s political gift. Exhibit A is Yale Law professor Justin Driver, who argues in a new book for preferences for descendants of slaves, immigrants, and members of legally recognized tribal nations, a special treatment that maps closely onto the black, Hispanic, and Native American categories of students who benefited from the old system of racial preferences.

A public intellectual who writes frequently for the New York Times, Driver is a smart observer of the law and a fluid writer. But his book provides further evidence that damaging ideologies about race die slowly. The Fall of Affirmative Action: Race, the Supreme Court, and the Future of Higher Education (New York: Columbia Global Reports, 2025) is a book that is stuck in the past. Troublingly, it appears to be gaining an audience among some higher education leaders. Michael Roth, the president of Wesleyan College, for example, gave Driver’s race-centric book a glowing review in the New York Times, lauding the volume as “insightful” and “excellent.”

My quarrel with Driver is over means, not ends. I agree with him that universities should find new paths to economic and racial diversity on campus. Racially and economically integrated student bodies at selective institutions of higher education are good for students and important for the future of the nation’s leadership class. But his suggested approaches aren’t based so much on a desire for fairness in the process—rewarding hard-working students who have done well despite obstacles—as on a desire for a prescribed racial outcome. As a result, he presents clever legal arguments that are politically tone-deaf.

Preferences for Descendants of Slavery, Immigrants, and Members of Indian Tribes

Driver’s leading proposal is to provide “an admissions preference to all applicants who identify as the descendants of slaves.” Because not all black people would benefit, the preference is not strictly racial in character. It’s consistent, Driver argues, with the Freedmen’s Bureau, created after the Civil War, to compensate slaves with 40 acres and a mule. Driver cites as precedent a Georgetown University program providing preferences to descendants of those enslaved people with a direct connection to the university. He complains, however, that Georgetown has only helped about three students per year. Driver would like to supercharge the program to apply to all descendants, not just those whose ancestors were enslaved by particular institutions of higher education.

Whether or not adopting such a policy would be legally viable, it would surely be a politically unwise move for institutions of higher education that are already facing substantial public distrust. The obvious problem with the Freedman’s Bureau analogy is that the 19th-century beneficiaries were the actual victims of a brutal policy on chattel slavery, and the disadvantages of race and economic class were neatly joined. The vast majority of enslaved people lacked wealth.

Several generations later, that dynamic has changed for the better. Today, there is a vibrant black elite, equipped with college educations and making six and seven figures. Research shows the economically privileged offspring of these families are the very ones who tend to benefit from preferences to elite colleges. At Harvard, for example, the litigation revealed that 71 percent of the black, Hispanic, and Native American students on campus were from the top socioeconomic 20 percent of the black, Hispanic, and Native American populations nationally.

Driver’s call for reparations is deeply unpopular. The idea that “descendants of people enslaved in the U.S. should be repaid in some way” is on the short end of a 68 percent to 30 percent split, according to Pew. This reality may explain why the nation’s only black governor, Wes Moore of Maryland, recently vetoed a reparations bill passed by the state legislature. Which does not mean that there aren’t other ways to take steps to repair our nation’s history. Dr. Martin Luther King called for reparations in the form of a “Bill of Rights for the Disadvantaged” of all races, a disproportionate share of whom would be black. (See more below.)

Driver’s second proposal is “preferences for immigrants.” Tellingly, Driver doesn’t make a case for doing so as a matter of justice. He does not argue, for example, that the academic record of immigrants from non-English-speaking families should count because of the obstacles associated with having to learn in a new language. Instead, the rationale is racial and ethnic. He cites the fact that “in 2018, white people accounted for less than one in five immigrants to the United States” as a central rationale for his policy.

Driver’s third proposal is for “admissions preferences on the basis of official tribal membership.” Again, Driver doesn’t make a moral argument based on “true merit”; he does not, for example, cite statistics about the economic struggles of the average Native American applicant. Instead, the idea is to boost representation of Native American students in a way that draws upon a long line of Supreme Court cases that labels tribal membership a “‘political’ rather than ‘racial’ classification.”

Driver is less enthusiastic about a fourth, much discussed idea: exploiting the so-called “personal essay loophole” of the Students for Fair Admissions decision, which allows a university to factor in discussions of race in essays. Driver cites Justice Sonia Sotomayor’s biting dissent, in which she called the provision “lipstick on a pig.” Given Driver’s focus on race, rather than deservedness, his skepticism makes sense. After all, the decision requires that discussions of essays be applied in a race-neutral fashion. If an admissions committee wants to credit a black student who overcame discrimination (as I believe it should) because the student showed perseverance, it would also have to credit an Asian American student who faced discrimination or a poor white student who overcame lousy schools, because those examples also provide evidence of grit.

Avoiding Class Inequality

What is particularly disappointing about Driver’s big three ideas is that they would allow universities to continue to elide the great problem of class inequality. Under a system of racial preferences, Harvard’s student body was racially integrated but had 23 times as many students who were wealthy as low-income. At the University of North Carolina, Chapel Hill, the subject of a parallel lawsuit, the ratio of rich to poor was 16 to 1. There is little reason to think Driver’s preferred ideas would not tend to benefit the wealthiest descendants of slavery, the most affluent immigrants, and the richest members of Indian tribes.

Driver is lukewarm about the one proposal that is grounded in fairness, has broad political support, and would foster racial diversity while tackling the class divide head-on: the idea Clinton, Obama, and King proposed of affirmative action for the economically needy of all races.

Driver notes that class-based affirmative action has been supported by Supreme Court justices across the ideological spectrum, from the liberal William O. Douglas to the conservative Antonin Scalia. But he raises two big concerns: students from disadvantaged and working-class backgrounds “may be much less well prepared to succeed” than wealthier students, and the program could end up mostly helping poor white students.

But the weight of the evidence runs in the opposite direction. Since the 1990s, the University of Texas at Austin has admitted most students through the “Top Ten Percent” plan for those with the top GPAs in high schools throughout the state, including those in very poor high schools. Those students, who were much more likely than other students to come from working-class backgrounds, have excelled academically. And a simulation using actual applicants to Harvard University in the Students for Fair Admissions litigation showed that if Harvard switched from race- to class-based affirmative action, the average student body SAT score would stand at the 98th percentile.

Those same simulations demonstrated that Harvard could achieve almost as much racial diversity using economic preferences. And since the Supreme Court has ruled, the evidence has grown. A substantial number of universities have announced a host of new economic affirmative action programs (including more generous financial aid and outreach), record levels of first-generation college students and students eligible for federal Pell grants, and high levels of racial diversity.

In the first year after Students for Fair Admissions, Princeton, Dartmouth, the University of Pennsylvania, the University of Virginia, and Emory all announced they had succeeded in keeping racial diversity at roughly the same level as they had achieved in the past employing racial preferences. At Yale, black and Hispanic representation stayed even at 14 percent and 19 percent respectively. Duke actually saw an increase in black enrollment from 12 percent to 13 percent and in Hispanic enrollment from 13 percent to 14 percent. Williams, Bowdoin, Bates, and Caltech all increased black enrollment as well.

Misleadingly, Driver downplays these successes, saying only that Duke, Princeton, and Yale “did not experience a stark contraction of Black enrollment” and instead focuses mostly on the outliers—such as MIT, Amherst, and Brown—which did see large declines in black representation, implying that they are the norm. They are not. A 2025 College Board analysis of 60 selective public and private colleges and universities found that black enrollment fell only 0.1 percentage points between 2023 and 2024 (from 7.5 percent to 7.4 percent).

Nevertheless, Driver calls the SFFA decision “a veritable catastrophe” and makes the hyperbolic claim that just as in the early 1960s, when just one half of one percent of students at elite colleges were black, university leaders will once again be forced to ask, “Where are the black students?”

While Driver is not particularly forthcoming about the balance of successes and failures among colleges seeking to sustain racial diversity, at other times he shows a candor that is refreshing in affirmative action debates. He acknowledges that college officials at UNC spoke about racial preferences in “raw, objectionable fashions” in text messages (e.g., “If it’s brown and above 1300 [SAT] put them in for [the] merit/Excel [scholarship].”) He does not try to excuse Harvard’s anti-Asian bias, as some supporters of affirmative action did, and is critical of Harvard’s practice of rating Asian-American applicants lower on subjective traits like “kindness” and “empathy.” He notes that Harvard attorney Seth Waxman’s defense of the personal score before the U.S. Supreme Court “can easily be construed as further entrenching ugly stereotypes” that Asian Americans “do not possess lively personalities…because they are too busy grinding away at schoolwork and extracurriculars.”

Driver is also correct on another important emerging legal matter: the legality of economic affirmative and geographic action programs like the “Top Ten Percent” plan that are aimed in part at promoting racial diversity on campus. The Trump administration, and a few very conservative justices, have advanced what Driver properly calls a “highly dubious interpretation of SFFA” that such programs are illegal if racial integration is a partial motivation. Driver points out, “SFFA did not in fact outlaw the goal of diversity; rather, it simply curtailed the means by which universities might pursue that goal, namely reliance upon racial classifications.”

By championing preferences for descendants of slavery, immigrants, and members of tribal nations, however, Driver would allow Trump to regain the upper hand. Trump would love nothing better than for Democrats to support a set of policies nakedly motivated by race rather than fairness. Let’s hope Democrats reject Driver’s advice, accept the political gift the Supreme Court delivered on racial preferences, and return to King’s bracing vision of policies that advance low-income and working people of all races. That is the only path to King’s dream of a multiracial coalition that puts hard-working low-income and working-class Americans at the very center.

Richard D. Kahlenberg is Director of the American Identity Project at the Progressive Policy Institute and author of Class Matters: The Fight to Get Beyond Race Preferences, Reduce Inequality and Build Real Diversity at America’s Colleges.