The university’s reputation for fairness and impartiality emerges bruised
ABBOTT LAWRENCE LOWELL, the president of Harvard from 1909 until 1933, thought the university had too many Jews. In the first year of Lowell’s presidency, they made up 10% of the student body. By 1922 their numbers had more than doubled. To address what he called “the Hebrew problem”, Lowell proposed an explicit Jewish quota of 15%. When that proved controversial, he set about making “a rule whose motive was less obvious on its face” to deny admission to students suspected of being Jewish. Admission to Harvard, previously granted by meeting a clear academic cut-off, became more nebulous—based more heavily on the “character and fitness” of applicants. The new “holistic” admissions policy worked as intended, successfully suppressing Jewish admissions.
Harvard, like many of America’s top universities, retains a holistic admissions process. Unlike elite universities in most other countries, American colleges do not simply select the cleverest pupils—they also take into account extracurricular activities, family wealth and race. To critics, this system still operates as an engine of unfairness, except that the victims have now become Asian-Americans, who outperform their white peers on academic measures but still face stiffer odds when applying to Ivy League colleges. Students for Fair Admissions (SFFA), an organisation founded by Edward Blum, a conservative activist opposed to race-based affirmative action, filed a lawsuit against Harvard alleging discrimination against Asian-American students in 2014. Despite a furious effort to quash the suit, Harvard was forced to turn over 90,000 pages on its tightly guarded admissions process. On June 15th both sides revealed duelling statistical analyses of admissions-decision data in court filings. Harvard’s reputation for fairness and impartiality emerges bruised.
By the admission office’s own ratings, Asian-Americans rank higher than white applicants in both their academic prowess and the quality of their extracurricular activities. Yet their admission rates are much lower. For Asian-Americans in the top decile of academic skill, just 13.4% are admitted, compared with 18.5% of whites (see chart). Asians are scored much worse on another measure of applicant quality—the “personal rating”—by admissions officers. Unlike the other two metrics, personality is judged subjectively and is decided by admissions officers who have not met the applicants. The alumni who conduct in-person interviews rate Asian-Americans as highly as white applicants. To SFFA, this constitutes clear proof of discrimination.
Peter Arcidiacono, an economist at Duke University employed by the plaintiffs, built a statistical model of the effect of race on admissions. He estimates that a male, non-poor Asian-American applicant with the qualifications to have a 25% chance of admission to Harvard would have a 36% chance if he were white. If he were Hispanic, that would be 77%; if black, it would rise to 95%. Damningly for the defendants, an internal report by Harvard’s research arm, obtained during discovery, reached the same conclusions. Harvard officials claim that the report was incomplete and the analysis oversimplified.
Fighting statistics with statistics, Harvard’s lawyers hired David Card, a prominent labour economist at the University of California, Berkeley. His model includes factors like the quality of a candidate’s high school, parents’ occupations and the disputed personal rating. Under these controls, Mr Card claims that Asian-American applicants are not disadvantaged compared with whites. But given that these factors are themselves correlated with race, Mr Card’s argument is statistically rather like saying that once you correct for racial bias, Harvard is not racially biased.
The duelling economists disagree because they cannot agree on what constitutes fairness. Susan Dynarski, an economist at the University of Michigan, argues that Mr Arcidiacono’s model tests for racial bias in an idealised system. Mr Card’s model searches for racial bias in the context of how Harvard actually operates.
For those unconvinced by fancy maths, the basic statistics also look worrying. Harvard insists that it has no racial quotas or floors, which would fall foul of Supreme Court rulings and jeopardise the university’s federal funding. Yet the share of Asian-Americans it admits has stayed near 20% over the past decade. This is true even as the number of Asian-Americans in high schools has increased. Caltech, a top university without race-based affirmative action, saw its share of Asian-Americans increase dramatically over the same period.
Court filings also reveal how legacy preferences, which give significant advantages to the relatives of alumni, skew Harvard’s admissions system. A suppressed internal report shows that the preference is the same size as that given to black applicants. Roughly 34% of legacy applicants are admitted—more than five times the rate of non-legacy applicants. This is tantamount to affirmative action for well-off white students. According to a survey of freshmen conducted by the Harvard Crimson, the college newspaper, 88% of legacy students come from families making more than $125,000 a year. Recruited athletes, which Harvard admits in droves to fill its lacrosse teams and rowing eights, are also disproportionately white. By Mr Arcidiacono’s reckoning, 22% of white students are legacies and 16% are recruited athletes.
Even if Harvard does not actually discriminate against Asian-Americans, its insistence on preserving hereditary preferences undermines its case. Rakesh Khurana, the dean of Harvard College, justified the policy on the grounds that it would bring those with “more experience with Harvard” together with “others who are less familiar with Harvard”. Others say that it is necessary to ensure fundraising. Aside from the moral questions this poses, it is worth noting that nearby MIT, which does not favour legacy applicants, manages to do just fine.