Admission to a selective college, we can probably all agree, should be fair, because the stakes are high and the public support we provide elite colleges is incredibly generous. The twelve universities that educate roughly half of America’s government and corporate leaders (and serve less than 1 percent of undergraduates) receive public subsidies of up to $105,000 per pupil, on the grounds that nonprofit universities serve the public interest. College admissions are laden with unfair preferences for the children of alumni, for athletes in a variety of sports, and for black and Latino students, most of whom also come from middle- or upper-class backgrounds. The one group that generally receives no preference? Economically disadvantaged students of all races. That will almost certainly change if the Supreme Court curtails the use of racial preferences in deciding Fisher v. University of Texas II, in which a white student, Abigail Fisher, has challenged the use of race in admissions. In states where racial preferences have been banned (usually by voter initiative), states and universities have responded by vigorously recruiting, and boosting financial aid to, less-advantaged kids. Community college students are encouraged to transfer, and colleges downplay standardized test results, and/or eliminate preferences for the generally wealthy children of alumni.
Race, Class, and
by Sigal Alon
Russell Sage Foundation, 348 pp.
The Fisher case will likely turn on an empirical question: whether the Supreme Court believes that universities can achieve racial diversity by alternative means. The death of Justice Antonin Scalia, an almost certain vote against racial preferences, might have been balanced out by Justice Elena Kagan, a supporter of preferences. But she has recused herself from the case. Justice Anthony Kennedy, typically the swing vote, likes racial diversity but dislikes policies that favor students based on skin color or surname. When the Fisher case first came before the Supreme Court, in 2013, Kennedy wrote that diversity is a compelling interest but that universities bear “the ultimate burden of demonstrating, before turning to racial classifications, that available, workable race-neutral alternatives do not suffice.”
Into this fray jumps Sigal Alon, a sociologist and anthropologist at Tel Aviv University, with her new book, Race, Class, and Affirmative Action. In a recent New York Times op-ed, Alon claimed to have “conclusive” evidence that “race-neutral models of preferential treatment cannot match the level of racial and ethnic diversity that race-based affirmative action can achieve.” Her conclusion is at odds with a wide body of research showing that when class disadvantage is properly defined—to include not just family income, but also family wealth and whether students grow up in disadvantaged neighborhoods—it can produce a vibrant level of racial and ethnic diversity. Most surprisingly, her broad-brush summary conflicts with her own book’s finding that sometimes, “race-neutral class-based affirmative action could not only replicate the current level of racial and ethnic diversity at elite institutions but even increase it.” Alon makes two arguments as to why class-based affirmative action is generally a deficient substitute for racial diversity policies. First, she turns to her home country, Israel, where race-blind programs for economically disadvantaged students were instituted at selective colleges a decade ago. The result: only half of the benefits went to Jews of Asian or African origin and Arabs, whom she analogizes to American black and Latino students as marginalized populations. The other half went to Jews of European ancestry from poorer neighborhoods or high schools. Cross-national comparisons are always tricky, but Alon says hers is justified because, “with the exception of a few sporadic experiments, class-based affirmative action has never been implemented in the United States.” Meanwhile, she says, Israel is “large enough to study.”
This is a doubly curious claim. In the United States, eight states, with a combined population of eighty million—ten times Israel’s population—have banned racial preferences at public universities and turned instead to experiments with class-based affirmative action. In a 2012 analysis, my colleague Halley Potter and I found that at seven of ten leading American public universities where race was banned in admissions, colleges created class-based alternatives that allowed them to meet or exceed both African American and Latino representation levels previously achieved with racial preferences. These institutions include University of Texas Austin, Texas A&M, University of Washington, University of Florida, University of Georgia, University of Nebraska, and University of Arizona.
Three outliers were not successful in preserving diversity—University of California, Berkeley, UCLA, and University of Michigan. These schools recruit from a national rather than regional pool of students and therefore compete for bright minority students with other national universities that can continue to use racial preferences. It isn’t surprising that their number of minorities declined.
In addition, the Israeli experiment is an oddly truncated one to study. Israeli universities look primarily at whether students come from disadvantaged neighborhoods or high schools—not the actual income of applicants’ families.
Alon’s second major approach is to use statistical models to estimate what would happen if elite U.S. universities moved from their current emphasis on race-based affirmative action to one based on a variety of class indicators, an approach that she says has “never been tested.” But, in fact, the Georgetown University researchers Anthony P. Carnevale, Stephen J. Rose, and Jeff Strohl conducted precisely such a simulation for the Century Foundation in 2014 and concluded that combined black and Hispanic representation would rise under class-based affirmative action compared with current race-based programs. The findings were not hidden from sight; they were published in the New York Times.
Moreover, Alon’s very own analysis suggests that if the least academically prepared students at selective colleges were replaced by academically qualified but economically disadvantaged students, racial, ethnic, and socioeconomic diversity would rise. Alon calculates that African American enrollment would go from 7 percent to 8.6 percent, Hispanic enrollment from 8.7 percent to 11 percent, and economically disadvantaged student enrollment from 8.9 percent to 19.2 percent. Because mean SAT scores would stay steady, she says, “all this could be done without jeopardizing academic selectivity.”
Alon discounts the real-world implications of these findings because she presumes that many of those in the bottom quartile who would be displaced by high-ability economically disadvantaged students are probably athletes or legacies. As a political matter, she argues, preferences for these students are unalterable. But that assumption is problematic at two levels.
First, in the real world, universities that were faced with a ban on racial preferences did often drop special consideration for legacy kids. The University of Georgia, University of California, Berkeley, UCLA, and Texas A&M all did so. Others might follow suit, bolstered by 2010 research finding that legacy preferences don’t actually increase alumni giving. Universities are unlikely to drop preferences for recruited athletes for popular sports like football or basketball, but they might well for squash or skiing. Second, as a legal matter, the Supreme Court says race can only be used as a last resort, when it is absolutely “necessary.” Alon’s data clearly shows that racial preferences are not necessary; there are other paths to diversity. It seems hard to believe that the Supreme Court would grant the extraordinary power to use race in admissions as a nod to the need to perpetuate discrimination against non-legacy candidates and provide athletic preferences to stock the sailing or equestrian teams.
To her credit, Alon is not oblivious to the enormous economic disparities in America’s selective colleges. She notes that between 1972 and 2004, though the share of underrepresented minorities at these institutions almost doubled, the share of students from the bottom socioeconomic quartile actually declined, from 9 percent to 7 percent. Her solution is to call for universities to employ affirmative action by race and class.
In practice, however, universities almost never seriously pursue a dual race-plus-class strategy. While colleges have long claimed that they “give significant favorable consideration” to the less well off, in fact four careful studies—led by William Bowen, Anthony Carnevale, Thomas Espenshade, and Sean Reardon—find that universities do not do so, so long as direct racial preferences are available to them. Bowen, for example, found that being an underrepresented minority increases one’s chance of admission by 27.7 percentage points but being in the bottom income quartile has no positive effect.
But universities do employ class-based affirmative action when racial preferences become unavailable, because it becomes a means for indirectly promoting racial diversity. It’s revealing that nine of ten top public institutions identified by the New York Times as doing the most for low-income students are in states where racial affirmative action has been banned. Meanwhile, almost all the low-performing public universities are in states where racial preferences are available.
The good news is that if class-based affirmative action is carefully constructed, in order to acknowledge the legacy and ongoing reality of racial discrimination, it can produce both racial and socioeconomic diversity in a way that race-based programs never have.