The Meritocracy Trap

by Daniel Markovits

  The pattern concerning colleges that are not just selective but highly selective is more extreme still. Nothing could ensure high school graduates from rich families a spot at a truly elite college—there are too many rich families and too few elite colleges for that to be possible—but selectivity effectively does ensure that high school graduates from poor and middle-class families will not attend a really elite college. From the high school class of 2004, for example, about 15 percent of high-income students but only 5 percent of middle- and 2 percent of low-income students enrolled in highly selective colleges. These are large differences, and, once again, the gap between the rich and the middle class massively exceeds—it more than triples—the gap between the middle class and the poor.

  The rates at which parents in each income bracket send their children to college of course determine the shares of students in college who hail from each income bracket. Small wonder, then, that college student populations skew spectacularly toward wealth. About 37 percent of all college students now come from households in the top quarter of the income distribution, compared to about 25 percent from each of the middle two quarters and 13 percent from the bottom quarter. The skew toward wealth within college student bodies has, once again, increased over time, especially since meritocracy’s early, democratic years. In addition, because graduation rates increase with household income, the skew to wealth among college graduates is greater still than among students. The shares of all bachelor’s degrees awarded to students from the bottom quarter of the income distribution, for example, was just 10 percent in 2014 (having declined from 12 percent in 1970).

  These inequalities, moreover, are greatest at elite colleges, and the skew toward wealth among students at the most elite colleges and universities is simply amazing. At the roughly 150 most competitive and selective—and therefore most elite—colleges, students from households in the top quarter of the income distribution outweigh students from households in the bottom quarter by a factor of fourteen to one according to one study; and at the 91 most competitive colleges, the top outweigh the bottom by twenty-four to one, according to another. These numbers entail that 72 percent of students at elite colleges come from the top quarter and only 3 percent come from the bottom quarter.

  The tiny share from the bottom is distressing but perhaps not surprising. The poor have never, to be sure, figured prominently among populations of any society’s most elite institutions. But the skew toward wealth appears, shockingly, even within the top part of the income distribution. Across selective colleges, students from households in the top quarter of the income distribution outweigh students from each of the middle two quarters by between eight and four to one. At elite colleges, rich students utterly dominate not just poor students but also students from the broad middle class. Once again, these imbalances have been rising over time, and especially over the course of meritocracy’s career—unsurprisingly, given rising educational inequality in early childhood and in high school. According to one study, the overrepresentation of the rich at elite colleges increased by roughly half between the late 1980s and the early 2000s. The abstract numbers reflect facts about concrete walks of life. A 2004 study of the most selective private universities, for example, found more freshmen whose fathers were medical doctors alone than whose fathers were hourly workers, teachers, clergy, farmers, and soldiers combined.

  The skew toward wealth becomes sharpest and most disturbing at the very top of the educational hierarchy. The administrations of the very most elite colleges and universities do not publish systematic and comprehensive data concerning the class backgrounds of their student bodies, but students at some of them have begun to collect and report data about themselves. Student reporting at both Harvard and Yale Colleges reveals that for recent classes, the share of students from households in the top quintile of the income distribution exceeds the share from the bottom two quintiles combined by a ratio of about three and a half to one. More distressingly still, across the Ivy League, the University of Chicago, Stanford, MIT, and Duke, more students come from families in the top 1 percent of the income distribution than from the entire bottom half. The scale of this skew toward wealth is simply outlandish. Even Oxford and Cambridge, long-standing symbols of the intersection between social class and elite education, today enroll student bodies with substantially greater economic diversity than Harvard and Yale.

  These facts, taken together, paint a stark overall picture. Being born to rich parents is nearly a sufficient condition for getting a BA, and it is nearly a necessary (although not a sufficient) condition for getting a BA from an elite college. College dominates the post-high-school lives of rich students, and children of rich parents dominate the student bodies of elite colleges. Whatever its origins and purposes, meritocracy now makes college a rich person’s affair.

  College itself exacerbates the concentration of training in the elite—extending education’s special focus on the rich into adulthood and further increasing the gap between the investments in human capital that middle-class and rich people receive. Whereas organized investment in the human capital of young Americans from poor and middle-class families mostly ceases at high school graduation, college initiates a new round of investment in almost all rich youth. The special investments associated with the education provided by the most competitive colleges go almost exclusively to rich youth. These investments, moreover, are massive.

  The distinctive investments that colleges make in educating students from rich backgrounds have been increasing steadily over recent decades. Higher education makes up 33 percent of all public education expenditure in the United States today, and when private expenditures are added in, colleges and universities account for 45 percent of total educational expenditures in the United States. The sums are staggering, both absolutely and relative to the rest of economic life: in 2014, postsecondary institutions spent $532 billion, or 3.1 percent of GDP (compared to $142 billion, or 2.2 percent of GDP, in 1970); and total investment in education in the United States approximately equals the total investment in nonresidential physical capital. Yale University alone now spends many times more than the entire nation’s 1840 investment in education. (Strikingly, compared to other OECD countries, the United States spends a smaller than average share of GDP on elementary and secondary education but nearly twice the average share on postsecondary education.) Moreover, expenditures have grown significantly more rapidly than enrollments since 1970, which entails that real expenditures per student have increased by nearly 60 percent.

  Expenditures have increased most rapidly, and enrollments most slowly, at the most elite schools. Median real per-student expenditures in the Ivy League, for example, increased by 80 percent between just 2001 and 2015. Competitive colleges quite generally spend much more on training their relatively richer students than noncompetitive colleges spend on training their relatively less rich students: $92,000 on student-oriented programs per student per year at the most selective colleges compared to only about $12,000 at the least selective ones; and this is five times the expenditure gap in the 1960s.

  A part of these rising expenditures is financed from the rising incomes of parents of the students who attend elite colleges. But the larger part in fact comes from subsidies paid from outside the students’ families: from rich colleges’ enormous endowments, and from public monies (including tax subsidies associated with colleges’ charitable status). Overall, the generally rich students at the richest 10 percent of colleges pay just 20 cents for every dollar spent on their educations, whereas the generally poor and middle-class students at the poorest 10 percent of colleges pay 78 cents on the dollar. Finally, both the subsidy and especially the gap between what ordinary and elite students receive have grown dramatically over the past fifty years. In 1967, the average annual subsidy per student was about $2,500 at the least selective colleges and about $7,500 at the most selective colleges; by 2007, the average at the bottom had grown to only about $5,000, while th
e subsidies for schools in the 99th percentile for selectivity had ballooned to about $75,000. Once again, the skew to wealth among elite student bodies entails that the largest subsidies go to the richest students.

  College, simply put, not only increasingly concentrates training in students from rich households but also increasingly subsidizes the training that the rich receive. The size of both these elements of the meritocratic inheritance is staggering.

  GRADUATE AND PROFESSIONAL SCHOOL

  According to a common narrative, college graduation marks the end of youth and the commencement (hence the name of the ceremony) of the earnest of adult life. The college graduate, on this account, leaves the nurture of the schoolhouse forever behind. Whatever else she learns or becomes, she must do it in the harsher circumstances of the “real” world.

  Life today defies this telling of it, however, especially among the economic elite; and the gap between imagination and reality grows steadily. At least for increasingly educated and massively trained superordinate workers, college graduation lights a path (if not at once, then foreseeably soon) not to “real” life but to further schooling. Indeed, in the minds of typical students at the most elite American universities, college serves as a conduit to postgraduate schooling in almost the same way in which high school was earlier a mere conduit to college. This additional education further focuses investments in human capital onto an increasingly skilled but also increasingly narrow elite, further expanding the gap between the investments made in the human capital of people born to rich parents and people born to everyone else. Graduate and professional school extends the meritocratic inheritance deeper still into adulthood.

  Graduate and professional education is a relatively recent phenomenon, and its prominence among elite workers is new. Indeed, elaborate graduate training was, until strikingly recently, not strictly required for getting elite jobs, including in the professions. Professional schools—law schools and medical schools—generally did not become graduate schools (requiring their students to have earned a BA prior to admission) until the early twentieth century. Most important, because they are both so numerous and so well paid, elite bankers, consultants, and corporate executives long practiced their trades without any formal graduate education in business administration. As Nitin Nohria, the dean of Harvard Business School, has observed, the midcentury American managerial elite was bound together not by university degrees but rather family networks and religious ties. In 1900, fewer than one in five business leaders had completed college.

  The American professional elite managed without graduate education because it received extensive on-the-job training from its employers. Doctors acquired specialized skills as they treated patients. Lawyers apprenticed to the offices and chambers of senior lawyers and judges. Most important, once again, managers, including elite executives, received systematic and substantial workplace training as they advanced through the elaborate managerial hierarchies that administered midcentury American firms.

  At IBM, for example, training for new executives began at the firm’s intense Armonk training center and in a sense never ceased. Management employees typically devoted two years of their early careers to a rotating practicum through staff positions at Armonk. And they subsequently received three additional weeks of training at Armonk annually throughout their entire careers with field training added on top. A career IBM man, retiring after forty years’ service, might have spent more than four years, or 10 percent, of his work life being trained by his employer. At Kodak, another leading light of midcentury American business, new employees received such quantities of training that the firm effectively never recruited employees over age twenty-five. Nor were these firms outliers. The leading midcentury study of executives observed that new workers sought out workplace training and chose their firms with training in mind, and that midcentury firms answered the call: the basic executive training program at the firms that the study considered lasted fully eighteen months.

  Firms today provide nothing remotely similar. When IBM abandoned its training-backed model of lifetime employment in the early 1990s, the shock at its headquarters was so great that company officials asked local gun shop owners to close their stores. And Kodak now expressly aspires to staff no more than one-third of its core management positions with internally trained workers. The transformation belongs to contemporary management lore—a collection of middle-aged insurance executives recently reminisced that while their own training a generation earlier had typically lasted a full year, none of their firms today any longer possesses any training program at all. The lore reflects reality, as measured by data: overall, the average U.S. firm today invests less than 2 percent of its payroll budget on training.

  Workplace training provided the fuel for the classic midcentury career arc, which focused on mobility within a single firm—“from the mail room to the corner office,” as the saying went. (A survey that Fortune magazine commissioned in 1952 reported that two-thirds of senior executives had worked for their current firms for over two decades.)

  Today, the fuel is spent. The character of elite work has changed in ways that reduce the value of firm-specific knowledge and increase the value of general skills, and over the same period, the structure of elite labor markets has changed in ways that reduce the commitments between firms and their employees. Workplace hierarchies are organized by occupations rather than by firms or even industries. And employers have abandoned implicit promises, once standard among managers, that competent work would merit lifetime employment and steady promotions. Instead, they offer, in the words of Apple Computer’s statement to its employees, “a really neat trip while you’re here” during “a good opportunity for both of us that is probably finite.”

  These changes, critically, all conspire to make university degrees rather than on-the-job training confer occupational access and advancement. Effectively every ambitious young doctor pursues not just the one-year internship traditionally required in order to obtain a general license to practice medicine but also longer and more intensive residencies, some of which (for example, in neurosurgery) last as long as seven years. Indeed, many specialties today require further formal, full-time training beyond the residency. Young lawyers similarly require three years of post-BA university training in law schools before they may practice law, and the nation’s law schools have produced on average, roughly forty thousand new JDs each year over the past two decades. And elite workers entering finance, consulting, and management today almost universally spend two post-BA years in university training at business schools, which produce over one hundred thousand new MBAs each year. Whereas a pioneering 1932 study found that 55 percent of top corporate managers had not even attended college, nine in ten have completed college today, and elite managers now overwhelmingly hold MBAs or JDs. These patterns have by now become so thoroughly established—so deeply entrenched—in the career paths of superordinate workers that they are taken for granted, as part of the assumed background of elite life. In fact, however, they represent a profound innovation—they are less than a generation or two old.

  This transformation has important consequences for the distribution of training—of investments in human capital—across American society. Post-BA training has long represented a substantial additional investment in the human capital of workers, and in particular of elite workers. The midcentury American employers who provided multiple years of formal training over the course of an elite career spent substantial sums in doing so. University-based graduate and professional schools make, if anything, larger investments in their students: expenditures per student per year at Harvard Business School have exceeded $350,000.

  Shifting training for elite jobs out of the workplace and into the university changes the socioeconomic composition of the people who receive the training and the investment in human capital that it imparts. Employer-provided training likely always skewed somewhat toward wealth, as the better entry-level jobs, which
provided the most training, went to applicants from more elite colleges and therefore from richer families. But university-based professional training skews dramatically toward wealth, as the disproportion of rich students at elite graduate and professional schools matches and even exceeds the socioeconomic imbalance among elite college students. (The one form of workplace training that survives and indeed thrives today—the unpaid internship—similarly favors young workers from wealthy backgrounds, who are disproportionately able to afford working for free.)

  This should not come as any surprise. Most immediately, graduate and professional schools are academically competitive, and the most elite schools are immensely competitive—indeed, more competitive than even the most elite colleges. The median student at Yale Law School, for example, earned effectively straight As in college (for a 3.9 GPA) and scored above the 99th percentile on the LSAT. The median student at Harvard Business School has a 3.7 college GPA and a GMAT score in the 96th percentile. And the median student at Stanford Medical School has a 3.85 GPA and an MCAT score in the 97th percentile. These students, moreover, overwhelmingly received their grades at elite colleges; and the students at the most elite graduate and professional schools overwhelmingly received their BAs at the most elite colleges. Forty percent of the Yale Law School student body attended an Ivy League college and fully 25 percent attended Harvard, Princeton, or Yale. The student bodies of these colleges, of course, themselves skew massively toward wealth. And graduate and professional schools that draw overwhelmingly from them cannot help but replicate the skew.