Legacies of Injustice
Alumni preferences threaten educational equity--and no one seems to care.
College-bound high school students do not always lose their chastity before graduation, but they certainly lose their innocence. Nearly every senior who has gone through the admissions mill can recount stories of peers with outstanding academic records—class valedictorians with stellar SATs and perfect GPAs—who were passed over by top colleges while others with far more modest credentials got the nod. The New York Times reports that Harvard turned down 1,100 applicants with perfect 800s on the math SAT this year. Yale rejected several with perfect 2400s on the three-part SAT exam. Princeton said no to thousands with 4.0 GPAs.
To many frustrated parents, one word de-scribes the admissions process at America's elite universities: arbitrary. But that's not the word admissions officials use, as I discovered two summers ago when I toured a dozen or so East Coast campuses with my son, a high school junior at the time. Asked what kind of grades and scores made kids competitive for their schools, officials in university after university insisted, as if reading off the same memo, that the review process was "holistic," "comprehensive," or "individualized." Grades, we were repeatedly told, "are only one among many factors we consider."
Another such factor is race. Nearly every selective college, public and private, gives a sizable edge to underrepresented minorities. Before the U.S. Supreme Court outlawed the University of Michigan's undergraduate admissions criteria in Gratz v. Bollinger (2003), the school relied on a complicated rating system that awarded points for several personal and academic factors, including skin color. Black and Hispanic candidates automatically got 20 points. A great essay counted for only one point; a perfect SAT score, a mere 12.
But as Justice Clarence Thomas observed in his dissent in a companion case, race is not the only factor that distorts college admission decisions. "The entire [college admission] process is poisoned by numerous exceptions to 'merit,'?" he noted.
The Wall Street Journal's Daniel Golden exposes those other exceptions in his 2006 book The Price of Admission. Golden shows that elite schools routinely hand preferences to athletes; to the children of faculty, celebrities, and politicians; to "development cases" whose fabulously wealthy parents offer hefty donations up front; and, above all, to the offspring of alumni. Universities expect the parents of these "legacy" candidates to contribute to their coffers after their children are admitted.
Robert Birgeneau, chancellor of the University of California at Berkeley, told Golden that at one Ivy League school only 40 percent of the seats are open to candidates competing on pure educational merit. According to a 2005 study by the Princeton sociologists Tom Espenshade and Chang Y. Chung, in 1997 nearly two-thirds of all these non-race-based preferences at elite universities benefited whites, even though whites comprised less than half of all applicants that year.
We have a vigorous national movement to eradicate racial or minority preferences, at least in public universities. In 2006 Michigan became the third state in the country after California and Washington to approve a ballot measure imposing a constitutional ban on the use of race in admissions at state-run schools and in government hiring decisions. And this year the author of all those bans—Ward Connerly, a black California businessman—is stepping up his crusade. He has launched petition drives in Oklahoma, Missouri, Colorado, Nebraska, and Arizona to put similar measures before voters in November.
But there's no comparable effort to get rid of legacy preferences. Even more troubling, many prominent opponents of racial preferences greet suggestions to get rid of legacies, the mother of all preferences, with a perfunctory nod—or a gaping yawn.
It shouldn't be that way. Legacy preferences are the original sin of admissions, the policy that fundamentally compromises fair, merit-based standards. Universities can't in good conscience tip the admission scales for the more privileged and then ask the less privileged to compete solely on merit. What's more, eliminating race while keeping legacies will make the admissions process less fair, not more fair, because it will open up minority slots to competition by whites but not vice versa.
Legacy preferences are an especially terrible idea for tax-supported public universities, since they make it possible for rich, white, and less qualified kids to take seats that are at least in part supported by the tax dollars of poor, minority families. Private schools, of course, should be free to admit whomever they want, and it is therefore tempting to ignore their use of legacies. But there are few genuinely private schools in America anymore, thanks to the enormous amount of federal funding they accept. And setting public policy aside: Just as a matter of propriety, should there be room for legacies at institutions that market themselves as bastions of meritocracy? The use of legacies by the Harvards, Yales, and Princetons of the world dilutes the standards of excellence they pretend not merely to uphold, but to embody.
Who Cares About Legacies?
With only a few exceptions, both the right and the left have ignored legacy preferences. Sen. John Edwards (D-N.C.) has promised to do everything in his power to end legacy admissions if he becomes president. But for the most part liberals have picked up the anti-legacy mantle only in retaliation against efforts to eliminate racial preferences. Local activists forced Texas A&M and the University of Georgia to abandon legacy preferences, for example, after these universities stopped using race in admissions. Otherwise, liberals seem quite willing to tolerate legacies, presumably because they make it easier to advocate countervailing preferences for their favored groups.
Given that dynamic, you might expect the opponents of racial preferences to go on the offensive against legacy preferences. But if liberals have been opportunistic about legacies, conservatives have been paralyzed.
In part, that's because they're genuinely divided on the issue. Ward Connerly, like Justice Thomas, regards legacies as a fundamental violation of a fair, merit-based standard. He prodded the University of California, where he is a regent, to abandon them in 2000, four years after California voters banned racial preferences. But Terry Pell, who heads the Center for Individual Rights (CIR), the outfit that engineered the lawsuit against the University of Michigan's race-based admissions, has never fought against legacies. Neither has Stephan Thernstrom, who has co-authored several books attacking racial preferences. "Legacy is a far more complicated issue than race," insists Thernstrom, who once served on Harvard's admissions committee.
Conservatives in Pell and Thernstrom's camp argue that racial discrimination is in a class apart, given this country's history of slavery and segregation. What's more, they say, legacy preferences are just not as big a problem as racial preferences, quantitatively speaking. Further, they produce huge benefits for universities that racial preferences don't. Above all, to the extent that legacies are practiced by private rather than public universities, there are no easy or desirable legal cures that aren't worse than the disease.
The last argument is their most powerful one, but it is hardly grounds for ignoring the issue. There are ways to address the issue of private universities' legacy preferences—and racial preferences—that don't involve lawsuits or government action. But the other arguments for why legacies aren't a public policy problem are simply disingenuous and suffer from the same ideological blind spots that afflict defenders of racial preferences.
A Small Problem?
Legacy preferences, like racial preferences, are repugnant because they reward not individual virtue or accomplishment, but an accident of birth that has no relevance for a college education. Moreover, just because they aren't linked with an egregious history of racial abuse does not justify turning a blind eye to them. India has a far uglier record of discrimination by caste than race. Yet no one would argue that it ought therefore to concentrate only on eradicating caste discrimination and treat race as a non-issue.
It is true that the use of legacies is mainly limited to undergraduate programs in the more selective public and private schools. Racial preferences, on the other hand, pervade every aspect of every school—from undergraduate and graduate admissions to faculty hiring and promotion. Moreover, according to a 2007 paper by Princeton's Douglas S. Massey and Margarita Mooney of data from 28 elite universities, while 77 percent of minorities had standardized test scores below the institutional average, about 48 percent of legacies did. In rare exceptions, such as Middlebury College, legacies actually scored higher than the institutional average.
How far below average do those legacies and minorities score? It's impossible to get up-to-date, nationwide data on the subject, given the universities' secrecy, but the late psychologist Richard Herrnstein and the social scientist Charles Murray reported one telling piece of information in their 1994 book The Bell Curve. In 1990, the average student admitted to Harvard scored 697 on the verbal SAT and 718 on the math section. By comparison, legacies scored 674 on verbal and 695 on math—a 47 point difference. Combined minority scores hover at about 100 to 150 points below the institutional average.
What's more, even when universities lower admission standards for legacies, they don't lower them as much as they do for minorities. As mentioned before, the Michigan point system used to award 20 bonus points to under-represented minorities—the equivalent of boosting a 3.0 GPA to a 4.0. By contrast, it handed only four points to children of alumni.
But such statistics don't tell the full story. Given how intense the competition is for the nation's most selective schools, even seemingly small differences in scores translate into significantly higher rates of acceptance for legacies over "unhooked" candidates—admissions lingo for those who don't qualify for any preferences.
According to the October 1996 Brown Alumni Magazine, 40 percent of legacy applicants were accepted to Brown University, as opposed to 19 percent of the total applicants. The Office of Civil Rights similarly found in 1990 that children of alumni were twice as likely to be accepted at Harvard over more qualified students who did not get legacy or athletic or any other preferences. And a study by the Center for Equal Opportunity, a Virginia-based think tank, found that at the University of Virginia, after controlling for test scores, grades, and other academic credentials, a legacy candidate had 4.3 times higher odds of admission than non-legacy applicants in 1999.
Universities claim that legacy status is never a major or decisive factor in their admission decisions. It's only used, they say, as a tie-breaker among otherwise comparable candidates. That's what they claimed about racial preferences too, and that turned out to be false. Indeed, it is hard to really know how much weight universities award to legacies given their stubborn refusal to reveal their admissions data or even talk about their admission policies. (University of Michigan officials, for instance, declined repeated requests to discuss this issue.) But why do legacies deserve any edge, big or small?
Racial preferences, at least originally, were meant to remedy discrimination—both historic and current—against blacks. What is the justification for favoring the offspring of Harvard, Yale, and Princeton alumni? Unlike many inner-city kids, they grow up in families with a strong pro-education ethos. They have access to the finest public or private high schools in the country. Their parents can spring for tutors, standardized test preparation courses, and even consultants to help them write essays and complete their college applications. "These are kids who grow up with every privilege," notes Connerly. "They don't deserve any additional advantage."
Moreover, though the policy of using legacy as a "tie-breaker" among equivalent candidates sounds innocuous, it has perverse consequences for one group in particular: Asian Americans. Asians don't benefit from racial preferences because they are not considered underrepresented minorities. And they don't benefit from legacy preferences because they tend to be the children of first-generation immigrants. Espenshade, the Princeton researcher, found that while legacy and athletic preferences offset the effects of racial preferences on whites, they compound them for Asian Americans.
According to Espenshade's regression analysis of data from a dozen selective colleges, on a 1600-point SAT scale, being black and Hispanic adds up to an advantage of 230 and 185 extra SAT points respectively. The preference for legacies translates into an edge of 160 points. By contrast, being Asian American represents a 50 SAT-point disadvantage.
The CIR's Pell, however, argues that the legacy problem is "self-correcting." Racial preferences have become so ideologically embedded that universities will never abandon them unless forced to by courts or voters, Pell maintains. But as the ethnic mix of the broader population changes so does the composition of the student body. A generation later, then, so will the composition of the beneficiaries of legacy preferences.
But the problem with legacies is not that they never adjust to shifting demographics. It is that they slow the process of adjustment. Legacy policies protect groups that are already in, at the expense of those that are trying to break in.
Practical Benefits?
Conservatives pride themselves on being sensible realists, not starry-eyed utopians eager to stamp out every form of social injustice regardless of consequences. This tendency partially explains their squishiness on the legacy issue. On the one hand, they don't dispute that legacy admissions border on institutionalized nepotism—rewarding children for the accomplishments of their parents and relatives. On the other hand, enforcing a strict merit-based standard seems a tad fanatical given all the practical benefits of legacy policies for universities.
One purported benefit is that legacies are an important source of funding for universities. Not only do more legacies donate to universities, they donate in greater amounts. For instance, according to the Cavalier Daily, the University of Virginia's student newspaper, 65 percent of legacy parents contributed to the university's 2006 capital campaign, compared with 41 percent of non-legacy parents. Moreover, legacy parents on average coughed up $34,759 each whereas non-legacy parents gave only $4,070.
All in all, legacies alone account for over 30 percent of the private donations to most elite colleges. "If mild preferences to legacy students allow universities to maximize their income, is that so objectionable?" asks Thernstrom.
Without such donations, universities claim, they could not invest in high-quality faculty and facilities and remain competitive. Even more important from the standpoint of social justice, universities say they couldn't maintain need-blind admission policies. These policies allow colleges to admit students purely on academic grounds—and then offer financial aid to anyone unable to afford the roughly $50,000 per year it costs in tuition and living expenses to attend a top-notch university these days. Without legacy contributions, such aid would supposedly become more difficult, and elite campuses would truly become playgrounds of the rich.
But Thernstrom and Pell don't buy arguments from social utility when it comes to racial preferences. Like other conservatives, they insist that universities that want to help inner-city minorities need to find race-neutral ways that don't selectively dilute academic standards for some groups. Nor do they believe that the educational benefits of a diverse student body are real or big enough to justify giving minorities a leg up.
Yet they uncritically accept the business and social case for legacy preferences. And it is far from clear that universities lack "legacy-neutral" tools to—as Thernstrom puts it—"maximize their profits." They could conceivably rake in more money by auctioning off a certain number of freshmen seats every year to the highest bidders. But elite universities would never entertain a scheme like that, because it could cost them their "elite" reputations. It would expose precisely how much they are diluting their admission standards for how many and for how much. This kind of information would erode their aura of selectivity—the very thing that makes them attractive to legacies and everyone else.
Connerly, after spending years on the University of California board, is not convinced that alumni will stop contributing to their alma maters if their kids don't get preferential treatment. Indeed, as Golden noted in The Price of Admission, Caltech is able to tap alumni money without offering any edge to their children. For instance, Caltech in 2001 obtained a $600 million pledge—the largest gift in the history of higher education at the time—from Gordon Moore, cofounder of Intel, neither of whose two sons attends the university. Caltech's commitment to high standards and excellence is a core part of its sales pitch to raise money from alumni and non-alumni alike.
Golden offers other examples, albeit isolated ones, of schools that have built sizable endowments through business strategies that don't rely on legacy preferences. Cooper Union, a highly prestigious and selective art school in New York that offers a free education to everyone admitted, for decades lived off income generated through its investments in real estate. Berea College, a small college in Kentucky exclusively targeted toward low-income kids, has accumulated a startlingly large endowment by making its progressive credentials a selling point to potential donors: It is the South's first inter-racial, co-educational college and was founded by an abolitionist minister. Its mission is to educate and uplift impoverished Appalachian families.
Legacy money doesn't seem to boost the presence of low-income kids on elite campuses by subsidizing their educations, either. The schools that get the most legacy money—Harvard, Yale, and Princeton—are among the worst when it comes to the economic diversity of their students. In his 2005 book The Chosen: The Hidden History of Admission and Exclusion at Harvard, Yale and Princeton, Berkeley sociologist Jerome Karabel reported that among the top 40 schools, Princeton and Harvard are ranked at 38th and 39th, respectively, when it comes to such diversity, and Yale 25th.
For a contrast, look at Caltech. It is the nation's most meritocratic private university that eschews all preferences, and it is among the 10 most economically diverse schools. Nor is it hard to understand why. Admissions are a zero-sum game with many candidates vying for a finite number of seats. The crucial determinant of economic diversity on campus therefore becomes not how much largesse legacies expend on poor kids but how many seats they take away from them.
If elite colleges were serious about offering equitable access to genuinely talented students, they could find business models that don't involve legacy preferences. If they have not done so, it is because the government won't—and market forces can't—hold them accountable.
Is There Any Rationale for Legacies at Public Schools?
The core mission of taxpayer-funded public universities is not to conduct research, promote economic growth, or correct broader social problems. It is to expand higher education opportunities. That, at any rate, is what the general public believes: Respondents in a 2003 survey conducted by The Chronicle of Higher Education overwhelmingly picked "offering a general education to undergraduates" as the top priority among 21 different roles that public universities could play.
Taxpayers perceive different public universities as fulfilling this educational mission in different ways. They regard land-grant universities as catering to rural kids, urban universities to commuters who can't live on campus, community colleges to students not served by traditional four-year colleges.
There is something problematic, even oxymoronic, about the very idea of "elite" public universities whose doors are by definition shut to the vast majority of taxpayers who fund them. If they must exist, they should exist to serve academically gifted kids. Thus the only defensible admission policy for these universities is one that allows all gifted kids an equal shot at admission.
This is precisely what legacy and other preferences don't allow. They reduce the fate of applicants to the discretion of admissions bureaucrats, eliminating clear-cut standards applied equally to all. Preferences replace the rule of law with the rule of men.
There are no good legal tools to mount court challenges against legacies in either public or private universities. The Constitution requires public entities to award everyone "equal protection under the law." But when a student tried to use this guarantee to mount a legal challenge against legacy preferences, she failed: In Rosenstock v. Governors of University of North Carolina (1976), an out-of-state applicant who was denied admission to the University of North Carolina argued that preferential treatment for in-state residents and children of alumni violated her right to equal protection. The court ruled that the state had no compelling interest in barring discrimination on the basis of alumni status, even at a public university.
But if the courts can't or won't ban legacy preferences, state voters certainly can. The simplest way to do so would be to append them to Connerly's ballot initiatives banning "race, gender, color, ethnicity, and national origin" preferences. Even the CIR's Pell acknowledges that it would be entirely appropriate for state voters to ban legacy preferences at public universities.
Connerly considered doing just that when drafting Proposition 2, the ballot amendment that banned racial preferences in Michigan in 2006. But he ultimately dropped the idea, he says, because constitutional amendments ought to be reserved for things that are "sacred for now and forever." He wasn't sure that alumni preferences were in that category.
But a further reason, some people close to Connerly admit, was that including legacies risked alienating whites in addition to blacks, making it harder to pass the initiatives. Strategically, it made more sense to deal with the two issues separately, reserving the ballot amendment process for race while looking for other ways to address legacies. Yet beyond Connerly's own personal crusade against what he calls "fat cat preferences" at the University of California, there has been almost no action on a national level against legacies.
If voter bans against legacies won't work, another way to force public universities to adhere to a stricter version of merit might be by requiring them to post—and adhere to—straightforward admission criteria like universities elsewhere in the world do.
For instance, Oxford, one of Britain's most prestigious universities, states unambiguously on its website precisely what scores and grades applicants need in order to gain admission. U.S. kids, it notes, need a combined SAT score of 2100 or a composite ACT score of 32 to 36—comparable to what kids from England, Wales, Scotland, and Ireland need. In order to remain true to its mission of creating an intellectually rigorous academic environment, Oxford, at least on paper, maintains the same admission standards for British students as for applicants from elsewhere. By contrast, it is accepted practice for elite American public universities to lower the bar for in-state students. To the extent that the tax contributions of the parents of these students fund the universities, they certainly deserve a break over out-of-state students. But that break ought to only involve lower tuition fees, not lower standards.
Openly publishing admissions criteria ensures transparency in the admissions process and serves as a sort of guarantee to prospective students that those who score below these minimum requirements won't be admitted ahead of those who do.
Connerly could use the contacts and machinery he has built in various states in the course of pursuing his anti–racial preferences amendments to push admission reform laws requiring public universities to set open and objective admissions standards. Universities will no doubt wail about the loss of academic freedom. But the rule of settled and transparent laws is no loss to freedom. It would only hem in the discretionary power of bureaucrats who wield it in an arbitrary way to offer access for their own self-serving purposes.
What About Market Forces?
There is a strong civil libertarian argument against applying such laws to private universities. Such schools ought to be allowed to admit whomever they want for whatever reason they want, as part of their right of voluntary association. Some schools might exercise this right for pernicious ends. But just as tolerating odious speech is essential for the sake of protecting broader freedoms, so, arguably, is tolerating odious forms of voluntary association.
This thinking has restrained conservatives from challenging racial preferences at private schools, even though they have powerful legal tools to do so. For instance, Title VI of the 1964 Civil Rights Act bars racial discrimination by any school that receives federal funds—a category that includes practically every private university given the ubiquity of federal research grants, scholarship aid, student loan guarantees, and countless other forms of direct and indirect financial assistance. Yet Michael Greve, former executive director of CIR, told the ABA Journal some years ago that his organization had no plans to go after private universities for racial preferences—a policy that his successor, Terry Pell, also adheres to. If Harvard, Stanford, and Yale want to discriminate for any reason, Greve said, that's their business.
It is conceivable that laws requiring public universities to set open and objective admission standards might help trigger industry-wide change, including at private schools. But why hasn't the higher education industry reformed its own admission practices? The market appeal of Harvard, Yale, and Princeton rests on the impression that getting through their door puts you in a league of intellectual superstars. If these schools dilute their standards—if they turn away incandescent intellects in favor of the merely bright with good family connections—how are they still able to maintain their luster? In a functioning marketplace, you would expect more Caltechs to emerge: Elite schools that market their uncompromising adherence to standards of excellence to snag students interested in being part of a true meritocracy. In the process, they would force Harvard and others to either reform their admission practices or relinquish their niche.
So far, instead of elite schools losing their niche, smart kids are losing their shot at an elite education. "Higher education is the only industry that is rewarded for turning away customers," observes Richard Vedder, an economist at Ohio University.
There are both demand and supply reasons for this peculiar state of affairs. The demand for colleges with established reputations is artificially inflated, notes Vedder, because of the absence of any meaningful metrics of educational quality, leaving students with nothing but the prestige factor to go by. Meanwhile, instead of accommodating this demand by expanding their supply, colleges have every incentive to ignore it: Their ranking in the annual U.S. News and World Report college ratings depends in large part on their "selectivity"—on what percentage of applicants they reject.
U.S. News' near-exclusive focus on inputs rather than outcomes when ranking universities perverts the admissions process in an even more direct way. One of the factors in its ranking is the extent of alumni giving, which is supposed to indicate alumni satisfaction with the education they received. Though this sounds reasonable, in practice it hands universities one more incentive to dole out more legacy preferences to shake down its alumni—and avoid a search for less compromising fundraising alternatives.
If market forces seem unable to hold elite universities accountable, it is because prospective students don't have good information about their educational outcomes to truly gauge whether these colleges are worth six figures. "Competition succeeds only to the extent that customers…can define success in some legitimate way in order to establish a standard and reward those who best achieve it," Derek Bok, Harvard's president from 1971 to 1991, has noted. "In education, at least at the university level, this ability is lacking."
So what did Bok—whose wife, the philosopher Sissela Bok, has written an entire book decrying institutional secrecy—do to make more information available to "customers" as Harvard president? Nothing. In fact, he tenaciously resisted repeated calls to reveal Harvard's admissions and other data.
Bok's reaction is typical. In 2003, Sen. Ted Kennedy (D-Mass.) tried to pass a mandate requiring both public and private universities to reveal their admissions data and show how many legacies they were admitting. Political pressure from universities killed the plan. Earlier this year, they opposed the recommendations of a commission convened by Education Secretary Margaret Spelling. The commission wanted all universities to report, among other things, their graduation and retention data.
Vedder, an ardent free-market advocate who served on that commission, is sympathetic to concerns about extending the government's reach into private universities. But he also argues that these universities are only nominally private, given the enormous amount of federal research subsidies and student aid they receive. "If we are going to drop planeloads of money to these universities," he asks, "why is it unreasonable to require them to report some basic information?"
Can Meritocracy Prevail?
Universities resist not just outside efforts to force more transparency, but also efforts from within the industry itself. Over the last decade, two serious efforts have emerged in the higher education marketplace to measure "outcomes"—the quality of education that colleges provide. In 2000, the Pew Charitable Trust and Indiana University launched the annual National Survey of Student Engagement (NSSE), which polls students about their college experience and, based on this feedback, ranks each college against its peers. Meanwhile, the Council for Aid to Education has developed the Collegiate Learning Assessment survey (CAL). While the NSSE measures only subjective student opinions about their college experience, the CAL actually offers extensive exit exams to students to measure what they have actually learned.
Making information about student satisfaction and learning available to consumers might revolutionize the way they make decisions about colleges. But elite colleges, having little incentive to have their reputation questioned by actual data, have refused to participate in either survey.
America's fundamental promise is that individuals ought to control their destiny through hard work and talent, not arbitrary accidents of birth. Legacy preferences are no less damaging to this promise than racial preferences. Those who oppose race as a factor in admissions but ignore legacies open themselves to accusations of inconsistency and hypocrisy. But, worse, to the extent that they succeed in dismantling race while leaving legacies intact, they risk putting in place a less—not more—fair admissions system.
As their battle against racial preferences heats up this year, they need to open another front against legacy preferences. The U.S. Constitution and courts do not offer ready weapons for the new battle. But that hardly justifies laying down arms without a fight.
There are plenty of ways at the state level to stop the use of legacies at public universities, from constitutional bans to state mandates requiring more transparent admission policies. Government can't ban private universities from using preferences, legacy or racial or any other, without running afoul of the Constitution. But that doesn't mean that moral suasion can't be used to prod them toward fairer admissions policies. Public outrage recently forced Harvard to give up its early decision program. The program, which overwhelmingly benefited the rich and connected, effectively lowered the bar for students who applied early and promised to accept its admission offer.
Most of all, we need policies to strengthen market accountability. We need to end the cartel-like character of the higher education industry, where private universities can keep consumers in the dark about their admission practices and educational product and still charge exorbitant prices without worrying that a competitor will emerge to challenge their market dominance with a cheaper and better product. An honest and straightforward recognition of the dangers of legacy preferences will go a long way toward bringing about such reforms.
Shikha Dalmia is a senior analyst at the Reason Foundation.
