The world of higher education is a world of sorting, selecting, and ranking—on both sides of the market. Colleges select students to recruit and then to admit; students choose where to apply and which offer to accept. The sorting process that gets the most attention is in the higher reaches of the market, where it's not too much to say that everybody is ranking and comparing all the time. Viewed from the outside, the activity at the high end, which we call the Little Sort, has a breathless, almost frenzied quality—the stuff, these days, of novels and New Yorker cartoons.
The more significant sorting, though—the one that has a major impact on the lives of many students and on the character of the inequality in American society—is a process most people take for granted. This Big Sort sends young people from low-income families, or those with parents who did not go to college, to community colleges or for-profit institutions—or to the military, or straight into the labor force. Meanwhile, it sends the most talented and generally the most privileged children into the competitive Little Sort.
The extreme and in our view irrational mania for sorting among elite students and institutions is worthy of attention. It wastes valuable financial and human resources and causes damaging levels of stress in some of our most talented young people, while encouraging them to engage in activities that do more harm than good to their emotional and intellectual development. Moreover, the current pressures on public higher education to become more selective in an attempt to move up in the prestige hierarchy threaten to make the frenzy spread to a much larger segment of the market.
But frenzied competition does not reflect the experience of most students or most postsecondary institutions. It's still true that a large fraction of people attending college apply and get in to exactly one institution. In the background of the attention-getting Little Sort that allocates star students among elite places, the Big Sort involves a set of forces working outside the spotlight that wind up determining which students have a plethora of choices about which college to go to and which have few choices or, too often, none at all.
The Big Sort
In 1960, Clark Kerr helped to advance a framework for mass higher education in California—the famous Master Plan—whose basic features have shaped public higher education in most states since. In Kerr's elegant plan, the top third of high school students, as measured by test scores and grades, would be admissible to a four-year college or university, while all other high school graduates could attend a two-year community college. The top third in turn were sorted into the University of California or the California State University (then College) system, depending on whether or not they ranked in the top 12 percent of the population of new high school graduates. Most states lack this much formulaic clarity in their sorting systems (and the California system has become more complicated over time), but the tripartite division within public systems is now nearly universal.
It's hard to argue with the basic logic of this kind of Big Sort. Students have different interests, preparation, motivation, and levels of intellectual engagement. Postsecondary education should not be one-size-fits-all. It makes intuitive sense that students who show more academic promise in high school should go to school longer and that the state should invest more in them than in other students. A selective sorting process, other things being equal, appears both efficient and equitable.
But (and it's a big but) other things aren't anywhere near equal. The odds that a student from a low-income family, or from one in which neither parent is a college graduate, will qualify for admission to a selective college or university are far lower than they are for those from more affluent and educated families. Even among high school graduates with similar academic qualifications, the sorting process leads to dramatically different outcomes depending on family background.
Indeed, the likelihood that those first-generation or lower-income students who do have a good chance of admission to a selective college or university will even apply is quite low. And recent research makes clear that missing out on attendance at a more selective college substantially reduces the likelihood of winding up with a bachelor's degree.
The Big Sort as it exists in the United States is shot through with the unfair consequences of highly—and increasingly—unequal economic and pre-collegiate educational systems. The impoverished educational opportunities offered to disadvantaged students at home, in their neighborhoods, and in their schools ensure that they will typically emerge from high school less well prepared to succeed at college than other students are. They are also ill-equipped to make the choices that will increase their likelihood of having the educational opportunities that are most likely to boost their chances for long-run success. Fixing or at least ameliorating these devastating inequalities is a vital problem of justice for our society.
But even in the context of dramatic inequalities in the circumstances and opportunities of young people, more could be done to diminish the chasm between the postsecondary experiences (or lack of experiences) of people from different backgrounds. We could be sure that both the information and the funding to support the most appropriate educational choices for each individual are universally available. We could provide the necessary resources to ensure that students with inadequate academic preparation can become college-ready in a system that currently makes college success for these students virtually unattainable.
A discussion that does justice to potential creative solutions to these fundamental problems is beyond the scope of this essay. But we should keep the issues at the front of our minds as we focus on the less compelling but still destructive problem of the Little Sort.
The Little Sort
There is not a bright line between the problems of the Big Sort and those of the Little Sort, because the highly selective colleges and universities don't manage to enroll even as many “qualified” low-income and first-generation students as they should. According to research by Catherine Hill and Gordon Winston, in 2003 only 10 percent of students in the most selective colleges and universities came from the bottom 40 percent of the income distribution. At the same time, 12.5 percent of the students with SAT (or equivalent ACT) scores of 1250 or higher came from this population. These are all students who have at least considered enrolling in a somewhat selective four-year college and who are likely to be qualified for admission. Elite colleges should strengthen their efforts to enroll these students.
Beyond this question of broadening the range of backgrounds in the student body, the Little Sort in colleges like Williams, Middlebury, Princeton, and Brown raises issues of a different kind. It's hard to believe that it matters very much to society exactly which students end up in which places at the high end, but the current process of matching students and institutions generates a dynamic that is wasteful and destructive.
The underlying force driving this dynamic is what economists call “positional competition,” a contest in which the aim is not to achieve some independently defined goal—like finishing a marathon in under four hours—but a goal that depends on a ranking, a place in line, like finishing a marathon in the top ten. Sports are in fact an excellent metaphor for this kind of positional competition.
Another example was developed by Tom Schelling. Imagine a world, which may in fact not be far in the future, where parents can influence the height of their children through genetic manipulation. How tall would you want your son to be? Schelling suggests most people would not want a very tall son or a very short one. Probably most people would aim for a son whose height was a little above average. But if most people made that choice, the next generation would be a little taller than the present generation, and to satisfy your positional preference you'd want a son a little taller than that. And so on. As the years passed, men's average height would grow without limit until we became a race of giants—an outcome that virtually nobody wants or would ever deliberately choose. (Economist Robert Frank has written extensively on such positional competition, including in higher education.)
The athletic arms races in which selective colleges are involved are not so different from these other examples. Colleges that don't award athletic scholarships or make any money from sports probably don't feel they have to have the best teams, but they fear being below average. So they keep raising the intensity (and costs) of their athletic programs as their competitors do the same. So the average is a moving target.
The athletic competition is attracting criticism for siphoning resources from activities more central to the academic mission. But competition for students with higher test scores, more extra-curricular activities, and more outstanding recommendations is similar.
The dynamic is even more powerful because both sides of the market have positional aims. A college that wants to move up in the rankings aims at improving the measured quality of its admitted students just enough to get ahead of its rivals. But as the rivals below strive to move up, each college has to ramp up a bit just to stay in the same place.
On the other side of the market, students don't need to be aiming to be the best in the world; they (and their parents) may just want to hit that ever-rising threshold for a top-10 or a top-20 institution, according to any of several similar ranking systems. But that means getting a little bit ahead of the other high school seniors with similar goals.
As these students push ahead, the colleges in turn have to decide who among those who would have qualified last year are no longer good enough to make the cut—and the wheel turns again.
Now obviously some of the things that colleges and students do to improve their credentials are of real value to the colleges, to the students, and to society. When colleges with too many large classes reduce class size or create more quiet spaces for study, that's a win all around. Up to some point, better residential accommodation and recreational facilities also add real value. And when students improve their study habits in high school and take more demanding courses that expand their capacities, that's terrific.
But as the process continues, further improvements run into decreasing and maybe even negative returns. Competition-driven amenities that have more to do with impressing applicants than with enriching education can become both wasteful and distracting. Students may add more to their stress levels than to their learning when they feel pressed to take every AP class in sight; the enormous pressure to look good on paper may also cause students to pass on the valuable experience of tackling a subject or project that might lead to failure. And despite their concern for socioeconomic diversity, colleges find themselves rewarding students for multiple summers building houses in developing countries, for musical accomplishments that would be impossible without expensive lessons, and for extra-curricular activities available only in well-resourced high schools.
Any college that turns down a significant number of applicants who are able to do the work is in some measure engaged in this dynamic. But the share of the competitive behaviors that are wasteful or destructive almost certainly increases as you move toward the more elite end of the market. The plausible candidates for admission at Williams or Princeton are terrific secondary students, with strong academic records and great achievements in public service, athletics, literary works, or all of the above. And the top universities and colleges, even after the recent endowment woes, are spending breathtaking amounts of money on every undergraduate they enroll. They are rejecting students who would likely be stars, while they spend more and more on the lucky few without measurably increasing the quality of their educational experiences.
This unhealthy ratcheting is a result of competition, and competition is supposed to be good for both consumers and producers. So what's wrong? In a market where firms are trying to be #1 but consumers are just looking for value, producers who embellish their products with frills that don't add utility lose customers. But as the markets for luxury products reveal, in some cases consumers are buying not just products but the status those products bring. If students and parents were just looking for the best education for the price, they might eschew colleges that waste resources. But they are also looking for status.
It would seem that some prestigious colleges would see an opening for attracting the high-quality students who are looking for value. But the risk of losing prestige that a single institution would take in focusing on value is too great. Moreover, the top-ranked students tend to come from families for whom a few thousand dollars more to get a more-prestigious degree is worth it. And of course the “Chivas Regal effect” is powerful: more expensive means better, particularly because it is so difficult to measure “better” in this market.
Another factor limiting the extent to which competition in this market can be constructive is the fixed supply of places in prestigious institutions. In competitive markets, new firms can enter when they see excess profits and the potential for meeting consumer needs at lower cost. But it's virtually impossible to create a new college at the top of the rankings.
The top institutions have powerful reputations that have been earned over many years. They have impressive lists of alumni. And they have sky-high endowments. It is difficult for a college in 11th or 12th place to move into the top ten. The idea of a new, innovative institution doing that is almost unimaginable. And, of course, as long as the competition is about relative position in a ranking, when a new player squeezes in, somebody else is squeezed out. The names change, but the game goes on.
Further down the food chain, strong colleges and strong students are more likely to have headroom for genuine improvement—even though the competitive dynamic rewards excessive attention to what is easiest to measure, count, or see on a campus tour.
If we could be confident that this phenomenon, at least in its more excessive forms, would stay confined to the relatively boutique-y world of the top students and colleges, improving matters might be ranked as a relatively low-priority problem. But there is reason to worry that this is a virus that will spread.
The presidents of flagship public universities are very focused on positional competition. The great majority seem to share the goal of being (or staying) in the top twenty. Of course score is kept in many ways other than the quality of undergraduate students, notably including research dollars, rankings of graduate programs, and contributions to local economic development. But as state funding dries up, these universities have less political pressure to serve the needs of the broad state population; they find more rewards in impressing donors, recruiting out-of-state students who bring in more tuition and often better credentials, and appearing in the same lists of prestigious places for an undergraduate education as the elite privates.
This strategy makes a good deal of sense for these universities considered one at a time (at least if you put in brackets their fidelity to the historic mission of land-grant universities), but it could not be more obvious that getting into the top-twenty public universities is an aim that is collectively self-defeating for the well over fifty institutions that consider themselves candidates for this status.
It is easy to write off as socially meaningless the question of which students go to Amherst and which go to Vassar. In contrast, the efforts to increase the number of low-income and first-generation students who go to elite colleges and universities—and to reduce the role of legacy status or athletic accomplishment in increasing chances for admission—have the potential to improve the equity of the system and of society in general, and certainly to dramatically improve the lives of some deserving individuals.
Yet modifying these outcomes won't touch the large-scale inequities in postsecondary opportunities in this country. As flagship public universities turn down more and more qualified in-state students and focus increasingly on prestige, there is the potential for the frenetic, status-oriented competition at the upper end of the postsecondary system to endanger the successes we have had in broadening and deepening the role of higher education.
In other words, failure to mitigate the frenetic and wasteful admissions competition may have more serious implications than we realize.
What Is To Be Done?
There is a relentless logic to the two-sided arms race driving the Little Sort that is peculiarly hard to overcome. Simply pointing out that a top-twenty strategy is collectively self-defeating will not curtail pursuit of such a positional goal among public universities, any more than it has that effect on the private universities and colleges with similar goals.
As the example of parents' choosing their children's height suggests, players in the admissions game don't have to make absolute success their first priority – they just need to seek out a modest advantage compared to their nearest rivals. Every president understands that she can't step out of the rat race and just stay put—as others continue to strive, her university or college will inexorably fall behind.
This logic implies that restraining the destructive aspects of this competition requires cooperation among institutions. This is how professional sports leagues restrain wage competition for players. The college draft and salary caps derive from enforceable agreements that keep teams' bidding wars under control. The obvious problem with such agreements is the threat of anti-trust litigation. Remarkably, Congress voted an explicit exemption for organized baseball, while the other major sports rely on enforcement through union contracts, which are protected under existing anti-trust law.
Might elite colleges and universities similarly enter into enforceable agreements to limit competition for students? The track record is not promising. The one clear example of colleges agreeing to restrain competition was the set of “overlap” agreements about financial aid offers across groups of colleges, which aimed to curtail merit-aid bidding wars for desirable students. These agreements provoked an anti-trust prosecution and, although a clear finding of legal violation was never reached (the case ended in a settlement), colleges have been understandably gun-shy about cooperative agreements ever since.
In our view, it would be highly desirable for Congress to provide protection from anti-trust violation for agreements among colleges to restrain socially unproductive positional competition. Admissions officers, for example, should not have to worry about legal jeopardy if they agree to limit the number of Advanced Placement results that can count in students' admissions records. Universities should similarly be able to adopt common constraints on investments in upgraded athletic and recreational facilities, as well as limits on merit-aid offers. Lines would need to be drawn carefully and practices would need to be monitored, but the current assumption that maximum market competition among these non-commercial institutions is always socially desirable is quite unjustified.
Is there any way to make progress short of gaining anti-trust protection? It isn't easy, but some things might be tried.
Our first suggestion arises from noting that a better matching system for students and selective colleges seems almost inevitable in this computer age. A generation ago, it would have been hard to imagine that so many people would be meeting their partners on line after filling out forms reporting their salient characteristics and those of their ideal matches. Travelers now can similarly benefit from aggregation sites like Kayak.com that let them compare air-travel and hotel offers from hundreds of brands at once.
The potential efficiency gains of relieving admissions officers of reading the applications of hundreds of students who are not seriously considering enrolling – and of relieving students from applying to more and more institutions to cover their bets – are large. Evolution toward a more coordinated and efficient admissions processing system across institutions might, if developed on the right lines, provide a vehicle for moderating the destructive aspects of the current competition. The “common application” is an early and relatively low-tech illustration of this trend.
Suppose, to illustrate, that either a third party or a consortium of colleges invited students to apply to a set of institutions as a group, listing their preferences among the colleges. (The University of California system functions in roughly this fashion now.) To invite applicants, a consortium would need to assemble an attractive package of options, in much the same way that students are now encouraged to apply to “stretch” and “safety” schools.
Consortia might make certain commitments to students (for example, guaranteeing admission somewhere in the consortium to all applicants who meet specified requirements) or require certain pre-commitments from them (such as the promise to accept admission to any of their top two or three choices). If this approach caught on, competition would come to take place among consortia more than among individual institutions and might result in a more rational and less frenetic marketplace.
Another potential solution elicits powerful negative reactions from many of the guardians of selectivity—but some signs of movement and a few quieter nods of understanding encourage us to put it forward. There is of course no possibility of easing the competition for a place in a top-ten institution by adding more institutions to the top ten. But what about making the enrollments of the top ten bigger?
When Toyota sees increased demand for Lexus, they expand production. Of course they wouldn't want everyone to drive a Lexus, because the prestige factor would be lost. But there is some room for increased supply of positional goods in an environment where the population is growing and where models are being added at the lower end of the hierarchy.
Twenty of the colleges and universities with endowments per student exceeding about $250,000 and the lowest acceptance rates in the nation enroll a total of about 20,000 first-year students. As their endowments recover the value they lost in the financial crisis, they are likely to lavish yet more luxurious facilities and curricular options on their privileged students. What if instead they maintained spending per student but increased the size of their undergraduate student bodies? They would still maintain their positions as the most-selective and the best-endowed colleges and universities in the country.
Right now, the top colleges expand production only reluctantly. They don't tend to increase the size of their entering classes as the number of students enrolling in college grows and the number of applicants qualified to enroll in the best of the best increases. But the distribution at the top is extremely skewed, so there's lots of room before the next tier of schools catch up. Admitting an extra 2,000 or 5,000 or even 10,000 students into this small circle wouldn't eliminate the high school frenzy. But it would diminish it. And it would share the wealth of the extraordinary opportunities offered to a dwindling fraction of college students in the United States.
Institutions wouldn't do this overnight, certainly, but of late both Amherst and Princeton have begun to increase their enrollments modestly, citing aims related to those that motivate us. And certainly the experience of the formerly men's colleges in the early 70s, when they expanded enrollment to make room for half of the human race, has been largely positive.
It is reasonable to ask why we can't just consider a larger number of institutions “elite.” Why can't students be happy going to number 40 instead of number 20? In fact, students who wind up at number 40 typically end up satisfied with the place where they enrolled – often, we suspect, at least as happy (and well educated) as if they had climbed further up the ladder. But in our prestige- and position-driven world, few students are as happy with the thick envelope from number 40 as from number 20. It will be no less exciting to be accepted at Harvard if they accept 10 percent of their applicants instead of 8 percent. The top 10 institutions could expand their enrollment by 20 percent without threatening their cachet, whereas the notion of the “top twelve” would be a harder sell.
Higher education is often viewed as a powerful instrument for promoting constructive social change. It is that, in some respects, although it could and should be more so. But higher education is also a mirror of the society it serves, and this aspect is prominent in the workings of both the Big Sort and the Little Sort. The broad sorting of students at the end of high school into more and less promising options for further education is heavily and perhaps even dominantly shaped by the deep patterns of inequality that structure opportunity in our society. And the intense scramble of the Little Sort both mirrors and reinforces the values of a society where the hippest car and the fanciest address are seen as marks of worth.
Reversing the trend toward ever-greater material inequality in our society would ameliorate the negatives of both the big and Little Sort. Low parental income is a powerful predictor of college attainment in this country, and the growing disparity between top and bottom incomes tends to create still greater inequality in the next generation.
The Little Sort is fueled in part by the perception that the path to the most prestigious jobs and financial wealth runs through the top colleges and universities. It is also fueled in significant measure by wealthy and ambitious parents whose investment in college credentials begins with the right preschool and proceeds through the very best private-college counselors and beyond.
It is idle to suggest that higher education, on its own, has the power to overcome the negative consequences of inequality and the quest for positional advantage that help to shape it. But higher education is not helpless. Past efforts to extend educational opportunities to previously excluded groups, including blacks and women, show some of the potential here. Determined efforts by higher education leaders both to improve the pre-college preparation of students from disadvantaged backgrounds and to enroll more of them who are qualified can make a difference. At the same time, to whatever extent college officials and faculty can influence students' choice of careers, they should help them gain some perspective on prestige and money as goals and encourage them to focus more on things of intrinsic and social value.
Although prestigious colleges and their leaders are in important ways trapped by the positional competition they feed, they are not without options. Any movement toward cooperation will require those at the top – the big winners in the current scramble—to put some of their advantage at risk and to yield some privilege in the service of a more humane and educationally productive college admissions process.
1. Bowen, W. G., Chingos, M. M. and McPherson, M. S. (2009) Crossing the finish line: Completing college at America's public universities, Princeton University Press., Princeton, NJ.
2. Frank, R. H. (1999) Higher education: The ultimate winner-take-all market?, Cornell Higher Education Research Institute.Paper 2
3. Hill, C. B. and Winston, G. C. McPherson, M. S. and Schapiro, M. O. (eds) (2006) “How scarce are high-ability low-income students?”.. College access: Opportunity or privilege?, The College Board., New York, NY.
4. Hill, C. B., Winston, G. C. and Boyd, S. A. (2005) Affordability: Family incomes and net prices at highly selective private colleges and universities. Journal of Human Resources 40:4, pp. 769-790.
5. Schelling, T. C. (1978) Micromotives and macrobehavior, W. W. Norton & Company, Inc., New York, NY.
Sandy Baum (email@example.com) is an independent higher education policy analyst and consultant, a senior fellow at the George Washington University School of Education and Human Development, and a professor emerita of economics at Skidmore College. She has written and spoken extensively on college access, college pricing, student aid policy, student debt, affordability, and other aspects of higher education finance.
Michael S. McPherson (firstname.lastname@example.org) is the fifth president of the Spencer Foundation. Prior to joining the Foundation in 2003, he served as president of Macalester College. A nationally known economist who focuses on the interplay between education and economics, McPherson spent the 22 years prior to that as professor of economics, chairman of the economics department, and dean of faculty at Williams College.