Larry Summers’s Unsatisfying Proposal

In a Washington Post column today on “Our inequality of opportunity,” former Harvard president Lawrence Summers proposes that top schools do more to recruit low-income students, showing the same commitment to economic diversity that universities have to racial diversity. He then couples this powerful challenge to include disadvantaged students with a jarring defense of admissions preferences for the children of alumni.

“It is unrealistic to expect that schools that depend on charitable contributions will not be attentive to offspring of their supporters,” he writes. “Perhaps, though, the custom could be established that for each ‘legacy slot,’ room would be made for one ‘opportunity slot.’”

This formula—to provide a preference for poor applicants equivalent to a preference for legacy applicants—was also advanced by former Princeton president William Bowen in his 2005 book, Equity and Excellence in American Higher Education. In that volume, Bowen and his co-authors Martin Kurzweil and Eugene Tobin asked, “Why not give the least privileged group of applicants, who presumably have had to overcome many obstacles to become qualified candidates for admission, the same advantage in the admissions process that is now conferred on those whose parentage has given them a special place in the competition for admission?” Doing so, the authors suggested, would have “a nice kind of symbolic symmetry associated with it.”

Bowen and Summers are both brilliant economists and leaders who have done a lot to advance the cause of low-income students, but the parallel they suggest is nothing short of bizarre. From a moral or meritocratic perspective, there is no symmetry whatsoever between giving a leg up to low-income and legacy candidates. For one set of students, who have faced numerous impediments to success, their raw grades and tests scores surely underestimate their long run potential in life; consideration of economic disadvantage is a corrective. Legacy candidates, by contrast, have mostly likely enjoyed considerable unearned benefits in their young lives, and providing them yet another preference is to advantage the already advantaged. The moral difference presented by low-income and legacy applicants may help explain why the public supports affirmative action for low income students by 2:1, but opposes affirmative action for legacies by 3:1.

In his article, Summers casually implies that charitable contributions would decline if alumni children weren’t provided preference in admission. Bowen and his colleagues, meanwhile, suggest colleges and universities have a legitimate interest in using legacy preferences as a way of “maintaining a real sense of historical continuity on their campuses and in attracting the resources that they need to pursue their excellence objectives.”

As former presidents of very successful and wealthy institutions, people who spent a lot of time fund raising, one might be tempted to take their word for it. But Bowen and Summers are hard-nosed economists, data-hounds, who would normally want some empirical support for the link between legacy preference and alumni giving.

Astoundingly, there appears to be no publicly available research to buttress the idea that the existence of legacy preferences increases donations. A 2010 study by researchers Chad Coffman, Tara O’Neil, and Brian Starr (in a volume I edited) noted, “there is no well-accepted systematic empirical analysis that establishes a causal relationship between legacy preferences and alumni giving behavior.” To fill that gap, Coffman and his colleagues examined alumni giving at the nation’s top 100 research universities from 1998 to 2007 and found that once they controlled for the wealth of an institution’s alumni, there was “no statistically significant evidence of a causal relationship between legacy preference policies and total alumni giving.”

Even if studies were able to confirm a relationship, surely Bowen and Summers, as economists, would recognize that there are more direct ways to entice donations than to provide a preference to all the children of alumni. As Michael Dannenberg has suggested, if would be far more efficient to auction off spots to immensely wealthy parents, irrespective of their alma mater. (To uphold standards, universities could limit the auction to minimally qualified candidates.)

So in place of the Summers and Bowen proposals for a strict symmetry between legacy and low-income preferences, I would suggest an alternative. For every one slot auctioned off to the child of immensely wealthy parents, offer 10 for low-income students who beat the odds and show tremendous promise. The proposal for a one-to-one ratio of opportunity and legacy slots hardly belongs in an article proposing to curtail “the perpetuation of privilege” and seeking “to equalize opportunity.”

Return to Top