The trend ... to have a student body with proportionately fewer well-off students and relatively more black and Hispanic students was partially reversed after the antitrust case.
For the academic community, an antitrust case that was launched more than a decade ago was a shock. From 1989 through 1991, the Department of Justice (DOJ) investigated a number of private, selective colleges for price fixing. The investigation eventually settled on the "overlap group," comprised of about half of the most selective private colleges in the United States. The group included 23 colleges, from small liberal arts schools like Colby, Vassar, and Middlebury to larger universities like Princeton and MIT. Some students applied to more than one of the 23 schools and, each spring, officials from these institutions met to coordinate the exact calculation of such students' financial need. The case broke new ground in antitrust theory. Although the colleges denied the price-fixing allegation, they discontinued their annual meetings in1991. Nor did they resume, even after a federal Court of Appeals rendered a decision in their favor.
In Benevolent Colluders? The Effects of Antitrust Action on College Financial Aid and Tuition (NBER Working Paper No. 7754), Caroline Hoxby looks at the economic merits of the Justice Department's case and the colleges' case. The DOJ alleged that the meetings enabled the colleges to collude on higher tuition and to increase their tuition revenue. The colleges defended their meetings by saying that they had to have some coordination in order to successfully implement their commitment to fully cover the need of any student they admitted. The colleges wanted to pull needy, able students into the pool of students who applied to selective private colleges because they saw such students as a "public good" for all their students. Yet, no college wanted to end up with a disproportionate share of the needy students simply because it had unintentionally made more generous need calculations than the other colleges. (All of the colleges attempted to use the same formula for need, but varying and difficult-to-interpret information from parents introduced some actual variation in their calculations.)
Hoxby compares the overlap colleges to a "control" group of private colleges that were equally selective but did not participate in the overlap meeting. She finds no evidence that the overlap colleges were colluding to raise tuition, raise net tuition revenue, or to save expenditures on grants. Both the overlap colleges and the control colleges raised tuition about 4 percent a year both before and after the antitrust action.
Hoxby does find however that, as a result of the suit, financial aid at the overlap colleges became less progressive with respect to parents' income and more sensitive to the merit of students, as measured by standard aptitude tests. That is, grants and aid became less sensitive to parents' income and, within any given level of need, more able students obtained more aid. According to Hoxby, this occurred because, with the end of coordination, each college had an incentive to continue to claim to continue to distribute aid on the basis of need (in order to get the "public goods" benefits associated with seeing a full range of applicants) while actually bending their need calculations so as to give out more merit-based aid and less need-based aid than rival colleges (thereby ending up with only a small share of the costs associated with the "public good.") She finds that, as a consequence, the trend in overlap colleges to have a student body with proportionately fewer well-off students and relatively more black and Hispanic students was partially reversed after the antitrust case. So far, the effects on the composition of the student body at these colleges after the end of the overlap meetings have been modest, the author indicates.
Interestingly enough, even among middle-income and upper-income families for whom aid grew slightly more generous after the antitrust action (compared to the control colleges), families became 6 percent less likely to accept an overlap college offer (rather than a control college offer) after the antitrust action. Hoxby concludes that this is weak evidence that need-based aid is not purely an act of charity, but a policy that results in student body composition that is valuable to families.
In her study, Hoxby notes that some of the confusion surrounding the antitrust action resulted from the fact that the Department of Justice found it difficult to apply standard economic analysis to college education. In particular, although students are consumers of education, they are also producers in sense that a students' peers affect his experience. They can, for instance, facilitate one another's learning. "Many students," Hoxby notes, "would like colleges to maintain policies of need-based aid for others while making exceptions for them, awarding them grants for which they would not qualify based on need. Yet the same students might prefer a regime of need-based aid, knowing that it would apply to them, because basing aid on need affects colleges' selectivity and diversity." In other words, students may value having peers who are diverse and able even if they know that the cost of having such peers is having a college that distributes aid more on the basis of need than on the basis of merit.
-- David R. Francis