A Transparent Absurdity

Colleges and universities quantify applicants’ merits using SAT scores and GPAs. But as a measure of anything that is itself worthwhile—of any meaningful achievement or genuine human excellence—an SAT score or a GPA is not so much imprecise and incomplete, or biased and unfair, as simply nonsensical. Even if individual questions on the test identify real skills, and even if grades on individual assignments or courses reflect real accomplishments, the sums and averages that compose overall SAT scores and GPAs fail to track any credible concept of ability or accomplishment. What sense does it make to treat a person who uses language exceptionally vividly and creatively but cannot identify the core facts in a descriptive passage as possessing, overall, average linguistic aptitude or accomplishment? It is more absurd still to treat someone who reads and writes fantastically well but is terrible at mathematics as, in any way, an ordinary or middling student. But SAT scores and GPAs push inexorably toward both conclusions. Again, even if one sets aside doubts about whether individual skills can be measured by multiple-choice questions or whether particular course work can be accurately graded, these metrics create literally mindless averages—totally without grounding in any conception of how to aggregate skills or accomplishments into an all-things-considered sum, or even any argument that the these things are commensurable or that aggregating them is intelligible.

Applicants, for their parts, measure colleges and universities by rankings, including most prominently those published by US News & World Report. These rankings are, if anything, even less intelligible than the metrics used to evaluate applicants. For colleges, for example, the rankings aggregate many factors: graduation and retention rates (both in fact and as compared to US News’s expectations), an idiosyncratic measure of “social mobility,” class size, faculty salaries, faculty education, student-faculty ratio, share of faculty who are full-time, expert opinion, academic spending per student, student standardized test scores, student rank in high school class, and alumni giving.¹ Once again, even supposing that these factors reflect particular educational excellences and that the data US News gathers measure the factors, the aggregate that it builds by combining them, using weights specified to within one-tenth of one percent, remains incoherent. Berea College, for example, enrolls students who skew more toward first-generation college graduates than Princeton University, and in this way adds more to the education of each student (especially compared to her likely alternatives), but it has a less renowned, scholarly, and highly paid faculty. What possible conception of “excellence” can underwrite an all-things-considered judgment of which is “better”? US News boasts that “our methodology is the product of years of research.”² But the basic question of what this research is studying—of what excellence this method of deciding which colleges and universities are “best” could conceivably measure, or whether any such excellence is even intelligible—remains entirely unaddressed.

In spite of their patent absurdities, the metrics deployed by both sides of the college admissions complex dominate how students and colleges are matched: Schools use test scores and grades to decide whom to admit, and applicants use rankings to decide where to enroll. The five top-ranked law schools, for example, enroll roughly two-thirds of applicants with Law School Admission Test (LSAT) scores in the ninety-ninth percentile.³ And although law schools hold precise recruitment data close, one can reasonably estimate that of the roughly 2,000 people admitted to the top five law schools each year, no more than five (which is to say effectively none) attend a law school outside the top ten.⁴ Law school is likely an extreme case. But instead of being outlandish, it lies at the end of a continuum and emphasizes patterns that repeat themselves (less acutely) across American higher education. Metrics that are literally nonsense drive an incredibly efficient two-way matching system.

When a transparent absurdity dominates a prominent social field, something profound lies beneath. And the metrics that tyrannize university life rise out of deep waters indeed. Elites increasingly owe their income and status not to inherited physical or financial capital but to their own skill, or human capital, acquired through intensive and even extravagant training. Colleges and universities provide the training that builds human capital, and going to college (and to the right college) therefore substantially determines who gets ahead. The practices that match students and colleges must answer the need to legitimate the inequalities this human capitalism produces, by justifying advantage on meritocratic grounds. Even when they are nonsense, numbers provide legitimacy in a scientific age. The numbers that tyrannize university life in America today, and the deformations that education suffers as a result, are therefore the inevitable pathologies of schooling in an age of human capitalism.

The Superordinate Working Class

In 2018, the average CEO of an S&P 500 company took home about $14.5 million in total compensation,⁵ and in a recent year, the five highest-paid employees of the S&P 1500 firms (7,500 workers overall) captured total pay equal to nearly 10 percent of the S&P 1500’s collective profits.⁶ In finance, twenty-five hedge fund managers took home more than $100 million in 2016,⁷ for example, while the average portfolio manager at a mid-sized hedge fund was reported to have made more than $2 million in 2014.⁸ The Office of the New York State Comptroller reported in 2018 that the average securities industry worker in New York City made more than $400,000.⁹ Meanwhile, the most profitable law firm in America yields profits per partner in excess of $5 million per year, and more than seventy firms generate more than $1 million per partner annually.¹⁰ Anecdotes accumulate to become data. Taken together, employees at the vice-presidential level or higher at S&P 1500 companies, professional finance workers, top management consultants, top lawyers, and specialist medical doctors account for more than half of the richest 1 percent of households in the United States.¹¹

These and other similar jobs enable a substantial subset of the most elaborately educated people to capture enormous incomes by mixing their accumulated human capital with their contemporaneous labor. This group now composes a superordinate working class. A cautious accounting attributes over half of the 1 percent’s income to these and other kinds of labor,¹² while my own more complete estimate puts the share above two-thirds.¹³ Moreover—and notwithstanding capital’s rising domination over ordinary workers—roughly three-quarters of the increase in the top 1 percent’s share of national income overall stems from the rise of this superordinate working class, in particular a shift of income away from middle-class workers and in favor of elite ones. The result is a society in which the greatest source of wealth, income, and status (including for the mass affluent) is the skill and training—the human capital—of free workers.

The rise of human capitalism has transformed the colleges and universities that create human capital. Two facets of the transformation matter especially. First, education has acquired an importance it never had before. Until only a few generations ago, education and the skills it produces had little economic value. Even generously calculated, the top 0.1 and the top 1 percent of the income distribution in 1960 derived only about one-sixth and one-third of their incomes, respectively, from labor, which is to say by working their own human capital.¹⁴ Moreover, schools and universities did not dominate production of such human capital as there was; both blue- and white-collar workers received substantial workplace training, throughout their careers. In Detroit, for example, young men might quit childhood jobs on their eighteenth birthdays and present themselves to a Big Three automaker, to take up unionized, lifetime jobs that would (if they were capable and hard working) eventually make them into tool-and-die-makers, earning the equivalent of nearly $100,000 per year—all with no more than a high school education.¹⁵ And in New York, a college graduate joining junior management at IBM could expect to spend four years (or 10 percent of his career) in full-time, fully paid workplace training as he ascended the corporate ladder.¹⁶ Small wonder, then, that the college wage premium was modest at midcentury, and that the graduate-school wage premium (captured above what was earned by workers with just a bachelor’s degree) was more modest still.¹⁷ Elite schools and colleges, in this system, were sites of social prestige rather than economic production. Education had little direct economic payoff; rather, it followed, and merely marked, hierarchies that were established and sustained on other grounds. The critics of the old order were clear eyed about this. Kingman Brewster—the president who did more than anyone to modernize Yale University—called the college he inherited “a finishing school on Long Island Sound.”¹⁸

But today, education has become itself a source of income, status, and power for a meritocratic elite whose wealth consists, principally, in its own human capital. The college wage premium has risen dramatically, so that the present discounted value of a bachelor’s degree (net of tuition) is nearly three times greater today than in 1965.¹⁹ The postgraduate wage premium has risen more steeply still, and the median worker with a postgraduate degree now makes well over twice the wage of the median worker with a high school diploma only, and about 1.5 times the wage of the median worker with a four-year degree only. College and postcollege degrees also protect against unemployment, so that the effects of education on lifetime earnings are more dramatic still. Just one in seventy-five workers who have never finished high school, just one in forty workers with a high school education only, and just one in six workers with a bachelor’s degree enjoy lifetime earnings equal only to those of the median professional school graduate.²⁰

Graduates of the top colleges and universities capture yet higher incomes, enjoying more than double the income boost of an average four-year degree, with even greater gains at the very top. The highest-paid 10 percent of Harvard College graduates make an average salary of $250,000 just six years out,²¹ while a recent study of Harvard Law School graduates ten years out reported a median annual income (among male graduates) of nearly $400,000.²² Overall, graduates of top-ten law schools make on average a quarter more than graduates of schools ranked eleventh to twentieth, and a half more than graduates of schools ranked twenty-first to one-hundredth;²³ and 96 percent of the partners at the $5 million-a-year law firm graduated from a top-ten law school.²⁴ More broadly, a recent survey reports—incredibly—that nearly 50 percent of America’s corporate leaders, 60 percent of its financial leaders, and 50 percent of its highest government officials attended only twelve universities.²⁵ This makes elite education one of the best investments money can buy. Purely economic rates of return have been estimated at 13 to 14 percent for college and as high as 30 percent for law school, or more than double the rate of return provided by the stock market.²⁶ Meanwhile, the educational alternatives to college have all but disappeared. According to a recent study, the average US firm invests less than 2 percent of its payroll budget on training.²⁷

A second transformation follows from the first. Education, especially at top-tier colleges and universities, is now distributed in very different ways from before. Colleges, especially elite ones, have never welcomed poor or even middle-class people in large numbers. But once those schools chose students based on effectively immutable criteria—breeding, race, gender—so that while college was exclusive, it was nevertheless (at least among those who qualified) effectively nonrivalrous and not competitive. Even the very top schools routinely accepted perhaps a third of their applicants, and some took much greater shares still.²⁸ As recently as 1995, the University of Chicago admitted 71 percent of those who applied. These rates naturally produced an application process that appears almost preposterously casual today. A midcentury graduate of Yale Law School, for example, recollects that when he met the dean of admissions at a college fair, he was told, based only on their conversation, “You’ll get in if you apply.” An easy confidence suffused the very language of going to college, as the sons of wealthy families did not apply widely but rather “put themselves down for” whatever colleges their fathers had attended. The game was rigged, and the stakes were small.

But today, education is parceled out through an enormous competition that becomes most intense at the very top. Even as poor and even middle-class children have virtually no chance at succeeding, rich children (no matter how privileged) have no guarantee of success. Colleges today—especially the top ones—are therefore both extremely exclusive and ruthlessly competitive. In a recent year, for example, children who had at least one parent with a graduate degree had, statistically, a 150 times greater chance of achieving the Ivy League median on their verbal SAT than children neither of whose parents had graduated high school.²⁹ Small wonder, then, that the Ivy Plus colleges now enroll more students from households in the top 1 percent of the income distribution than from the entire bottom half.³⁰ This makes these schools more economically exclusive than even notorious bastions of the old aristocracy such as Oxford and Cambridge. At the same time, while being born to privilege is nearly a necessary condition for admission to a really elite American university, it is far from sufficient. Last year, the University of Chicago admitted just six percent of applicants, and Stanford fewer than five percent.

These admissions rates mean that any significant failure—any visible blot on a record—effectively excludes an applicant. Rich families respond to this fact by investing almost unimaginable resources in getting their children perfect records. Prestigious private preschools in New York City now charge $30,000 per year to educate four-and-five-year-olds, and they still get ten or twenty applications for every space. These schools feed into elite elementary schools, which feed into elite high schools that charge $50,000 per year (and, on account of their endowments, spend even more). Rich families supplement all this schooling with private tutors who can charge over $1,000 per hour. If a typical household from the richest 1 percent took the difference between the money devoted to educating its children and what is spent on a typical middle-class education, and invested these sums in the S&P 500 to give to the rich children as bequests on the deaths of their parents, this would amount to a traditional inheritance of more than $10 million per child.³¹ This meritocratic inheritance effectively excludes working- and middle-class children from elite education, income, and status.

These expenditures are almost as inevitable as they are exorbitant. When one set of institutions dominates the production of wealth and status in a society, the privileged few set out to monopolize places, and the pressure to gain admission becomes enormous. Human capitalism, moreover, makes schooling a positional good. “The value to me of my education,” the economist Fred Hirsch once observed, “depends not only on how much I have but also on how much the man ahead of me in the job line has.”³² This remains so, moreover, regardless of how much education the person ahead of me and I both possess. Every meritocratic success therefore necessarily breeds a flip side of failure—the investments made by the rich exclude the rest, and also those among the rich who don’t quite keep up. This means that while the rich get sated on most goods (there is only so much caviar a person can eat), they cannot get sated on schooling. Finally, rather than pick schools based on family tradition, applicants make deliberate choices about where to apply, and almost always attend the highest-ranked school that admits them, as when effectively nobody admitted to a top-five law school attends a school outside the top ten.

In these ways, human capitalism creates an educational competition in which the stakes are immense and everyone competes for the same few top prizes. Whereas aristocracies perpetuated elites by birthright, meritocratic inequality establishes school and especially college admissions committees as de facto social planners, choosing the next generation of meritocrats. Education becomes a powerful mechanism for structural exclusion—the dominant dynastic technology of our enormously unequal age. This places extreme pressure on the schools, and especially admissions committees, which must decide which people to privilege, using what criteria, and to what ends.

Bohr’s Lucky Horseshoe

What happens to schools when the degrees they grant grow so valuable that the demand for them outstrips their supply, and when admissions decisions make or break applicants’ life plans and determine who gets ahead in society? How have schools and colleges responded to their admissions decisions’ raised stakes? And what has the rise of human capital, its dominant role in wealth (even among the rich), done to the nature of education itself—to education’s aims, and to the standards by which it determines success? Measurement, and the tyranny of numbers, turns out to play a central part in the answer to all these questions—and for reasons not just shallow but deep. The manifestly absurd metrics that dominate university life are direct consequences of the role that schooling plays in our present economic and social order.

That which is measured becomes important. But at the same time, that which is important must be measured—and on a scale that allows for the sort of confident and exact judgments and comparisons that numbers yield. In a technocratic age—suspicious (for good reasons as well as bad) of humanist, interpretive, and therefore discretionary judgments about value—the demand for certainty and precision becomes irresistible when the stakes get high enough. The rise of human capitalism therefore makes it essential to construct metrics that schools and colleges might use to assess human capital and to compare the people who possess it, in order to determine whose human capital should receive additional investments.

The problem becomes more pressing still because education is lumped into standardized units called degrees, so that schools (especially the most exclusive ones, which have no part-time students or “honors colleges”) cannot hedge their bets by offering applicants varying quantities or qualities of training, but must instead make a binary choice to accept or to reject, full stop. The metrics that admissions offices use must therefore be able to aggregate across dimensions of skill and ability, in order to construct a single, all-things-considered measure of ability and accomplishment capable of supporting a “yes” or a “no.” This task becomes especially demanding in a world that has rejected the unity of the virtues and insists instead that people and institutions may excel in some ways even as they fail in others. GPAs and standardized test scores, especially on the SAT, as well as university rankings as provided by US News & World Report, provide the required metrics—comprehensive and complete orderings that can make fine distinctions that all who accept the metrics must agree on. Averages, scores, and rankings operate as prices do in economic markets, corralling judgments made unruly by normative pluralism and fragmentation into a single, public, shared measure of value.

These metrics—especially the SAT—are of course themselves disputed, sometimes vigorously. Certainly, they rest on arbitrary assumptions, and precision comes only at the cost of simply ignoring anything intractable, no matter how important. Nevertheless, even challenges to particular measures of human capital often accept the general approach that lies behind them all, and therefore give away the evaluative game—as (once again) when the SAT is criticized for lacking much power to predict GPAs. And even when they are contested, metrics like the GPA and SAT suppress ambiguities that they cannot eliminate, by pushing contestation into the background, far away from the individual cases and the evaluation of particular applicants. We may disagree about the validity of the SAT, and indeed harbor doubts about the test’s value, but we will nevertheless all agree on who has the highest score. In this sense, GPAs and SATs are like Niels Bohr’s lucky horseshoe—they work even if you don’t believe in them. In a world in which people cannot possibly agree on any underlying account of virtue or success, but literally everything turns on how success is measured, numerical scores allow admissions committees to legitimate their choices of whom to admit.

The early meritocrats understood this. At Harvard, James Bryant Conant, president from 1933 to 1953, introduced the SAT into college admissions with the specific purpose of identifying deserving applicants from outside the aristocratic elite. (James Tobin, who would serve on President John F. Kennedy’s Council of Economic Advisers and win a Nobel Prize, was an early success story.³³) Yale came to meritocracy later, but (perhaps for this very reason) embraced the logic of numbers-based meritocratic evaluation more openly and explicitly. Kingman Brewster, president from 1963 to 1977, called himself an “intellectual investment banker” and encouraged his admissions office to compose Yale’s class with the aim of admitting the students who would maximize the human capital that his investments would build. R. Inslee “Inky” Clark, Brewster’s dean of undergraduate admissions from 1963 to 1969, called his selection process “talent searching” and equated talent with “who will benefit most from studying at Yale.” The new administration, moreover, deployed test scores and GPAs not just affirmatively, to find overlooked talent, but also negatively, to break the old aristocratic elite’s monopoly over places at top colleges. Clark called the old, breeding-based elite “ingrown,” and aggressively turned Yale against aristocratic prep schools. In 1968, for example, when Harvard still accepted 46 percent of applicants from Choate and Princeton took 57 percent, Yale accepted only 18 percent.³⁴

The meritocrats aimed by these means to build a new leadership class. The old guard recognized the threat and resisted, both privately and even publicly. Brewster’s predecessor had scorned Harvard’s meritocratic admissions, which he said would favor the “beetle-browed, highly specialized intellectual.” When Brewster’s revolution was presented to the Yale Corporation, one member objected, “You’re talking about Jews and public-school graduates as leaders. Look around you at this table. These are America’s leaders. There are no Jews here. There are no public-school graduates here.” And William F. Buckley lamented that Brewster’s Yale would prefer “a Mexican-American from El Paso High…[over]…Jonathan Edwards the Sixteenth from Saint Paul’s School.” Just so, the meritocrats replied.³⁵

They added that their meritocratic approach to building an elite—because numbers measure ability and, just as important, block overt and direct appeals to breeding—would launder the hierarchy that it produced. Prior inequalities—especially aristocratic ones—were prejudicial, malign, and offensive. But meritocracy purports to be wholesome: backed by objective numbers, open to all comers, and resolutely focused on earned advantage. Indeed, meritocracy aspires to redeem the very idea of inequality—to make unequal outcomes compatible with equal opportunities, and to render hierarchy acceptable to a democratic age. In this way, the early meritocrats combined stark criticism of the present with a profound optimism about the future.

The Soldier, the Artist, and the Financier

The meritocrats’ optimism fell, if not at once, then soon. And it fell at hurdles erected by their own reliance on numbers. The metrics that the meritocrats constructed, and that now dominate education, turned out to be not just absurd but destructive.

To begin with, numerical metrics of accomplishment naturally inflame ruthlessly single-minded competition. There is no general way to rank learning, or creativity, or achievement—merit—directly. There is no way to say, all things considered, who has better skills, wider knowledge, or deeper understanding, much less who has accomplished more overall. People value different things for different reasons. We disagree with one another about what is most valuable: the entrepreneur’s resourcefulness, the doctor’s caring, the writer’s insight, or the statesperson’s wisdom. Moreover, each of us is unsure, in our own judgments, about how best to balance these values when they conflict—unsure, to pick a famous example, about whether to pursue a life of politics or of reflection, to pursue the executory or the deliberative virtues. The agreement and repose needed to sustain a stable direct ranking simply can’t be had. This is not all bad: Ineliminable uncertainties about value diffuse and therefore dampen our competition to achieve. The soldier and the artist simply do not compete with each other, and neither competes with the financier.

By contrast, numerical metrics—again including especially GPAs and SAT scores—aggregate across incommensurables to produce a single, complete ranking of merit. Indeed, producing this ranking is part of such metrics’ point—the thing that makes them useful to admissions offices. But now, competition whose natural state is disorganized and diffuse becomes highly organized and narrowly focused. Aspiring businesspeople, doctors, writers, and statespeople all will benefit, in reaching their professional goals, from high SAT scores and GPAs, and, accordingly, they all compete to join the top ranks. The numbers on which admissions offices rely to validate their selections therefore create competition and hierarchy where the incommensurability of value once made rank unintelligible. SATs and GPAs do to human capital what prices earlier did to physical or financial capital—they make it possible to say, all things considered, who has more, who is richest. Unreasoning accumulation and open inequality follow inexorably.

The competition that the numerical metrics create, moreover, aims at foolish and indeed fruitless ambitions. SAT scores and GPAs, once again, do not measure any intelligible excellences, and high scores and averages therefore have no value in themselves. At best, pursuing them wastes effort and attention and almost surely deforms schooling, by diverting effort and attention from the many genuine excellences that education can produce. This is even more vividly true on the side of colleges and universities, with respect to the wasteful and even destructive contortions they put themselves through in pursuit of higher US News rankings.

The numbers-based distortions induced by students’ pursuit of higher test scores and institutions’ pursuit of higher rankings both may be given a natural framing in terms of the distinction between excellence and superiority. Excellence is a threshold concept, not a rank concept. It applies as soon as a certain level of ability or accomplishment is reached, and while it can make sense to say that one person is, in some respect, more excellent than another, this does not eliminate (or even undermine) the other’s excellence. Moreover, excellence is a substantive rather than purely formal ideal. Excellence requires not just capacity or achievement, but rather capacity and achievement realized at something worthwhile. It is a moral error to speak of excellence in corruption, wickedness, or depravity. Superiority, on the other hand, is opposite in both respects. It is a rank—rather than a threshold—concept, and one person’s superior accomplishment undoes rather than just exceeds the superiority of those whom she surpasses. In addition, superiority is purely formal rather than substantive. It makes perfect sense to speak in terms of superiority at activities that are worthless or even harmful.

When the numbers that rule over the processes that match students and schools under human capitalism subject education to domination by a single and profoundly mistaken conception of merit, they depose excellence, installing in its place a merciless quest for superiority. Human capitalism distorts schooling in much the same way that financialization distorts for-profit sectors of the real economy. Once, firms committed to particular products (General Motors to cars, IBM to computers) might view profits as a happy side-effect of running their businesses well. But in finance, whose only product is profit, the distinction between success and profitability becomes literally unintelligible, and financialization therefore subjects the broader economy to a tyranny of profit. Similarly, flourishing schools and universities will view their reputations and status as salutary side-effects of one or another form of academic excellence. But human capitalism shuts schools off from these conceptions of excellence and enslaves them to the pursuit of superiority. Schooling in an age of human capitalism thus becomes subjected to a tyranny of SATs, GPAs, and college rankings.

All these consequences, moreover, are neither accidents nor the result of individual vices: the shallowness of applicants or the vanity of universities. Rather, a social and economic hierarchy based on human capital creates a pitiless competition for access to the meritocratic education that builds human capital. Working- and middle-class children lack the resources to compete in the educational race and so are excluded not just from income and status but from meaningful opportunity. Rich children, meanwhile, are run ragged in a competition to achieve an intrinsically meaningless superiority that devours even those whom it appears to favor. And the colleges and universities that provide training, and administer the competition, are deformed in ways that betray any plausible conceptions of academic excellence. The Varsity Blues scandal exposed this corruption alongside the frauds that conventional responses emphasized. Why would intelligent and otherwise prudent people—one of the culprits was cochair of a major global law firm—pursue such a ham-fisted scheme other than from a desperate fear of losing meritocratic caste? No one escapes the meritocracy trap.

The only way out—for schools as well as for students—involves structural reforms that extend well beyond education, to reach economic and social inequalities writ large. But although reforms cannot end with schools, colleges, and universities, they might begin there. In particular, the familiar hope that making standardized tests less biased and more accurate and making rankings more comprehensive—that is, perfecting meritocracy—might more effectively launder social and economic inequalities without diminishing them is simply a fantasy. Colleges and universities, in particular, cannot redeem their educational souls while retaining their exclusivity. Instead, elite schools must become, simply, less elite.

If it mattered less where people got educated, applicants could pursue different paths for different reasons. And schools and colleges, freed from the burden of allocating life chances, could abandon their craving for superiority and instead pursue scholarly insight, practical innovation, community engagement, and a thousand other incommensurable virtues. Along the way, by freeing themselves from superiority’s jealous grasp, universities might redeem the very idea of excellence.