Legacy preferences, which provide a leg up in college admissions to applicants who are the offspring of alumni, are employed at almost three-quarters of selective research universities and virtually all elite liberal-arts colleges. Yet legacy preferences have received relatively little public attention, especially when compared with race-based affirmative-action programs, which have given rise to hundreds of books and law-review articles, numerous court decisions, and several state initiatives to ban the practice.
The secrecy surrounding legacy preferences has perpetuated a number of myths, including the following:
1. Legacy preferences are just a "tie breaker" in close calls.
While some colleges and universities try to play down the impact of legacy preferences, calling them "tie breakers," research from Princeton's Thomas Espenshade suggests that their weight is significant, on the order of adding 160 SAT points to a candidate's record (on a scale of 400-1600). Likewise, William Bowen, of the Andrew W. Mellon Foundation, and colleagues found that, within a given SAT-score range, being a legacy increased one's chances of admission to a selective institution by 19.7 percentage points. That is to say, a given student whose academic record gave her a 40-percent chance of admissions would have nearly a 60-percent chance if she were a legacy.
The children of alumni generally make up 10 to 25 percent of the student body at selective institutions. The proportion varies little from year to year, suggesting "an informal quota system," says the former Wall Street Journal reporter Daniel Golden. By contrast, at the California Institute of Technology, which does not use legacy preferences, only 1.5 percent of students are children of alumni.
2. Legacy preferences have an honorable history of fostering loyalty at America's great institutions of higher learning.
In fact, as Peter Schmidt, of The Chronicle, notes, legacies originated following World War I as a reaction to an influx of immigrant students, particularly Jews, into America's selective colleges. As Jews often outcompeted traditional constituencies on standard meritocratic criteria, universities adopted Jewish quotas. When explicit quotas became hard to defend, the universities began to use more-indirect means to limit Jewish enrollment, including considerations of "character," geographic diversity, and legacy status.
3. Legacy preferences are a necessary evil to support the financial vitality of colleges and universities—including the ability to provide scholarships for low-income and working-class students.
While universities claim that legacy preferences are necessary to improve fund raising, there is little empirical evidence to support the contention. In fact, several colleges and universities that do not employ legacy preferences nevertheless do well financially. As Golden notes, Caltech raised $71-million in alumni donations in 2008, almost as much as the Massachusetts Institute of Technology ($77-million), even though MIT, which does provide legacy preferences, is five times the size and has many more alumni to tap. Berea College, in Kentucky, favors low-income students, not alumni, yet has a larger endowment than Middlebury, Oberlin, Vassar, and Bowdoin. And Cooper Union, in New York City, does not provide legacy preference but has an endowment larger than that of Bucknell, Haverford, and Davidson.
Moreover, a study included in our new book, Affirmative Action for the Rich, finds no evidence that alumni preferences increase giving. Chad Coffman, of Winnemac Consulting, and his co-authors examined alumni giving from 1998 to 2007 at the top 100 national universities (as ranked by U.S. News & World Report) to test the relationship between giving and the existence of alumni preferences in admissions. They found that institutions with preferences for children of alumni did have higher annual giving per alumnus ($317 versus $201), but that the advantage resulted because the alumni in colleges with alumni preferences tended to be wealthier. Controlling for the wealth of alumni, they found "no evidence that legacy-preference policies themselves exert an influence on giving behavior." After controls, alumni of legacy-granting institutions gave only $15.39 more per year, on average, but even that slight advantage was uncertain from a statistical perspective. Coffman and his colleagues conclude: "After inclusion of appropriate controls, including wealth, there is no statistically significant evidence of a causal relationship between legacy-preference policies and total alumni giving at top universities."
The researchers also examined giving at seven institutions that dropped legacy preferences during the period of the study. They found "no short-term measurable reduction in alumni giving as a result of abolishing legacy preferences." For example, after Texas A&M eliminated the use of legacy preferences, in 2004, donations took a small hit, but then they increased substantially from 2005 to 2007.
Nor can legacy preferences be said to be necessary for colleges to maintain high standards of excellence. It is intriguing to note that, among the top 10 universities in the world in 2008, according to the widely cited Shanghai Jiao Tong University rankings, are four (Caltech, the University of California at Berkeley, the University of Oxford, and the University of Cambridge) that do not employ legacy preferences.
4. After a generation of affirmative action, legacy preferences are finally beginning to help families of color. Pulling the rug out now would hurt minority students.
In fact, legacy preferences continue to disproportionately hurt students of color. John Brittain, a former chief counsel at the Lawyers Committee for Civil Rights, and the attorney Eric Bloom note that underrepresented minorities make up 12.5 percent of the applicant pool at selective colleges and universities but only 6.7 percent of the legacy-applicant pool. At Texas A&M, 321 of the legacy admits in 2002 were white, while only three were black and 25 Hispanic. At Harvard, only 7.6 percent of legacy admits in 2002 were underrepresented minorities, compared with 17.8 percent of all students. At the University of Virginia, 91 percent of early-decision legacy admits in 2002 were white, 1.6 percent black, and 0.5 percent Hispanic.
Moreover, this disparate impact is likely to extend far into the future. In 2008, African-Americans and Latinos made up more than 30 percent of the traditional college-aged population but little more than 10 percent of the enrollees at the U.S. News's top 50 national universities.
5. An attack on legacy preferences could indirectly hurt affirmative-action policies by suggesting that "merit" is the only permissible basis for admissions.
The elimination of legacy preferences would not threaten the future of affirmative action, because the justifications are entirely different. Affirmative-action policies to date have survived strict scrutiny because they enhance educational diversity. (For some members of the Supreme Court, though not a majority, affirmative action also has been justified as a remedy for centuries of brutal discrimination.) Legacy preferences, by contrast, have no such justification.
Because they disproportionately benefit whites, legacy preferences reduce, rather than enhance, racial and ethnic diversity in higher education. And rather than being a remedy for discrimination, they were born of discrimination. Affirmative action engenders enormous controversy because it pits two great principles against each other—the antidiscrimination principle, which says we should not classify people by ancestry, and the anti-subordination principle, which says we must make efforts to stamp out illegitimate hierarchies. Legacy preferences, by contrast, advance neither principle: They explicitly classify individuals by bloodline and do so in a way that compounds existing hierarchy.
6. Legacy preferences may be unfair, but they are not illegal. Unlike discrimination based on race, which is forbidden under the 14th Amendment, it is perfectly legal to discriminate based on legacy status, as the courts have held.
Remarkably, legacy preferences have been litigated only once in federal court, by an applicant to the University of North Carolina at Chapel Hill named Jane Cheryl Rosenstock, in the 1970s. A New York resident whose application was rejected, she claimed that her constitutional rights were violated by a variety of preferences, including those for in-state applicants, minorities, low-income students, athletes, and legacies. Rosenstock was not a particularly compelling candidate—her combined SAT score was about 850 on a 1600-point scale, substantially lower than most out-of-state applicants—and she was also a weak litigant. She never argued that, because legacy preferences are hereditary, they presented a "suspect" classification that should be judged by the "strict scrutiny" standard under the amendment's equal-protection clause.
The district-court judge in the case, Rosenstock v. Board of Governors of the University of North Carolina, held that it was rational to believe that alumni preferences translate into additional revenue to universities, although absolutely no evidence was provided for that contention. The decision was never appealed. As Judge Boyce F. Martin Jr. of the U.S. Court of Appeals for the Sixth Circuit notes, the 1976 opinion upholding legacy preferences in Rosenstock addressed the issue "in a scant five sentences" and is "neither binding nor persuasive to future courts."
A generation later, two new legal theories are available to challenge legacy preferences. First, Carlton Larson, a law professor at the University of California at Davis, lays out the case that legacy preferences at public universities violate a little-litigated constitutional provision that "no state shall ... grant any Title of Nobility." Examining the early history of the country, Larson makes a compelling case that this prohibition should not be interpreted narrowly as simply prohibiting the naming of individuals as dukes or earls, but more broadly, to prohibit "government-sponsored hereditary privileges"—including legacy preferences at public universities. Reviewing debates in the Revolutionary era, he concludes: "Legacy preferences at exclusive public universities were precisely the type of hereditary privilege that the Revolutionary generation sought to destroy forever." The founders, Larson writes, would have resisted "with every fiber of their being" the idea of state-supported-university admissions based even in part on ancestry.
Second, the attorneys Steve Shadowen and Sozi Tulante argue that legacy preferences are a violation of the 14th Amendment's equal-protection clause. While the amendment was aimed primarily at stamping out discrimination against black Americans, it also extends more broadly to what Justice Potter Stewart called "preference based on lineage." Individuals are to be judged on their own merits, not by what their parents do, which is why the courts have applied heightened scrutiny to laws that punish children born out of wedlock, or whose parents came to this country illegally.
Shadowen and Tulante argue that legacy preferences at private universities, too, are illegal, under the Civil Rights Act of 1866. Unlike Title VI of the 1964 Civil Rights Act, which outlaws discrimination only on the basis of "race, color, or national origin," the 1866 law prohibits discrimination on the basis of both "race" and "ancestry."
7. Legacy preferences—like affirmative action, geographic preferences, and athletics preferences—are protected by academic freedom, especially at private universities and colleges.
It is true that the courts have recognized that colleges and universities should be given leeway in admissions in order to promote academic freedom. But that freedom is not unlimited, even at private institutions. As Peter Schmidt notes, the Supreme Court held, in Runyon v. McCrary (1976), that private schools could not engage in racial discrimination in admissions. In Regents of the University of California v. Bakke (1978), it struck down the use of racial quotas. And in the 2003 Gratz v. Bollinger decision, the court invalidated a policy that awarded bonus points to minority students. Ancestry discrimination—providing a leg up in admissions based not on merit but on whether a student's parents or grandparents attended a particular university or college—likewise falls outside the protected zone of academic freedom.
8. Legacy preferences have been around a long time and are unlikely to ever go away, because powerful political forces support them.
In fact, legacy preferences are not only legally vulnerable; they are politically vulnerable as well. Polls find that Americans oppose legacy preferences by 75 percent to 23 percent, and in the past decade or so, 16 leading institutions have abandoned them. As affirmative-action programs come under increasing attack, legacy preferences become even harder to justify politically.
Moreover, as a matter of tax law, legacy preferences are fundamentally unstable. Assuming it is true that they entice alumni to provide larger donations than they otherwise would—a claim that has not been empirically proven—then IRS regulations raise questions about whether those donations should be tax deductible. If universities and colleges are conferring a monetary benefit in exchange for donations, then the arrangement, writes the journalist Peter Sacks, "shatters the first principle underlying the charitable deduction, that donations to nonprofit organizations not 'enrich the giver.'" The IRS regulations place universities in a legal Catch-22: Either donations are not linked to legacy preferences, in which case the fundamental rationale for ancestry discrimination is flawed; or giving is linked to legacy preferences, in which case donations should not be tax deductible.
9. Legacy preferences don't keep nonlegacy applicants out of college entirely. They just reduce the chances of going to a particular selective college, so the stakes are low.
True, legacy preferences don't bar students from attending college at all. But the benefits of attending a selective institution are substantial. For one thing, wealthy selective colleges tend to spend a great deal more on students' education. Research finds that the least-selective colleges spend about $12,000 per student annually, compared with $92,000 per student at the most-selective ones. In addition, wealthy selective institutions provide much greater subsidies for families. At the wealthiest 10 percent of institutions, students pay, on average, just 20 cents in fees for every dollar the college spends on them, while at the poorest 10 percent of institutions, students pay 78 cents for every dollar spent on them. Furthermore, selective colleges are better than less-selective institutions at graduating equally qualified students. And future earnings are, on average, 45 percent higher for students who graduated from more-selective institutions than for those from less-selective ones, and the difference in earnings ends up being widest among low-income students. Finally, according to research by the political scientist Thomas Dye, 54 percent of America's corporate leaders and 42 percent of governmental leaders are graduates of just 12 institutions. For all those reasons, legacy preferences matter.
10. Everyone does it. Legacies are just an inherent reality in higher education throughout the world.
In fact, as Daniel Golden writes, legacy preferences are "virtually unknown in the rest of the world"; they are "an almost exclusively American custom." The irony, of course, is that while legacies are uniquely American, they are also deeply un-American, as Michael Lind, of the New America Foundation, has argued.
Thomas Jefferson famously sought to promote in America a "natural aristocracy" based on "virtue and talent," rather than an "artificial aristocracy" based on wealth. "By reserving places on campus for members of the pseudo-aristocracy of 'wealth and birth,'" Lind writes, "legacy preferences introduce an aristocratic snake into the democratic republican Garden of Eden."
For the most part, American higher education has sought to democratize, opening its doors to women, to people of color, and to the financially needy. Legacy preferences are an outlier in that trend, a relic that has no place in American society. In a fundamental sense, this nation's first two great wars—the Revolution and the Civil War—were fought to defeat different forms of aristocracy. That this remnant of ancestry-based discrimination still survives—in American higher education, of all places—is truly breathtaking.