The Meritocracy Trap

by Daniel Markovits

  At the roughly 150 most competitive: For the fourteen-to-one number, see Carnevale and Strohl, “How Increasing College Access Is Increasing Inequality,” 137, Figure 3.7.

  For the twenty-four-to-one number, see Jennifer Giancola and Richard D. Kahlenberg, True Merit: Ensuring Our Brightest Students Have Access to Our Best Colleges and Universities (Lansdowne, VA: Jack Kent Cooke Foundation, 2016), 5, Figure 1, www.jkcf.org/assets/1/7/JKCF_True_Merit_Report.pdf.

  These ratios are not outliers but rather typical. Another study of students (in 2004) at the most selective 5 percent of colleges finds 69 percent from the top quartile of the income distribution, 18.7 percent from the second, 8.1 percent from the third, and 4.1 percent from the fourth. For the next most selective 7 percent of colleges, these shares were 66.2 percent, 19.5 percent, 9.3 percent, and 5.0 percent, from the top to bottom quartiles. By contrast, at noncompetitive four-year colleges (which accounted for the bottom 7 percent in selectivity), the shares were 25.9 percent, 29.1 percent, 25.4 percent, and 19.6 percent. Finally, the shares of college-aged young adults to receive no postsecondary education, by household income quartile, were 7.7 percent, 19.1 percent, 31.2 percent, and 42 percent. Michael N. Bastedo and Ozan Jaquette, “Running in Place: Low Income Students and the Dynamics of Higher Education Stratification,” Educational Evaluation and Policy Analysis 33, no. 3 (September 2011): Appendix, Table 6, www.personal.umich.edu/~bastedo/papers/EEPA-Appendix.pdf [inactive]. Hereafter cited as Bastedo and Jaquette, “Running in Place.”

  Finally, Charles Murray, citing Joseph Soares, similarly reports that in the 1990s, 79 percent of students at “tier one” colleges came from the top quartile of the socioeconomic distribution and only 2 percent came from the bottom quartile. Murray, Coming Apart, 59; Joseph Soares, The Power of Privilege: Yale and America’s Elite Colleges (Stanford, CA: Stanford University Press, 2007), 4, Table 1.1.

  most elite institutions: It is worth noting that the exclusion has become more dramatic over recent decades: the share of Pell Grant recipients among the student bodies at both public and private four-year colleges fell by about a third between the early 1970s and the early 2000s. Bishop, The Big Sort, 31. Bishop cites U.S. Department of Education, National Center for Education Statistics, Thomas D. Snyder (Project Director), Alexandra G. Tan, and Charlene M. Hoffman, Digest of Education Statistics 2003, December 2004 (1999–2000 data), 379, Table 325, http://nces.ed.gov/pubs2005/2005025.pdf. The level has remained roughly steady between 2001 and the present. Giancola and Kahlenberg, True Merit, 6, Figure 2.

  between eight and four to one: Carnevale and Strohl, “How Increasing College Access Is Increasing Inequality,” 137, Figure 3.7.

  At elite colleges: These patterns are no less striking when viewed from the perspective of children hoping to get to college. Students from rich families are three times more likely to attend highly competitive colleges than middle-class students and nearly eight times more likely than poor students. Bailey and Dynarski, “Inequality in Postsecondary Education,” 120, Figure 6.2; Reardon, “No Rich Child Left Behind.”

  the early 2000s: Alexander W. Astin and Leticia Oseguera, “The Declining ‘Equity’ of American Higher Education,” Review of Higher Education 27, no. 3 (Spring 2004): 329–30, Figure 1. In 1985, the top quartile of the income distribution filled 46 percent of the places at elite colleges (defined as the top tenth of all colleges), for an overrepresentation of 84 percent. In 2000, the top quartile filled 55 percent of the places, for an overrepresentation of 120 percent. Another study reports that the gap between the shares of rich and poor Americans to earn a BA grew by nearly half between the early 1980s and the early 2000s. See Martha J. Bailey and Susan M. Dynarski, “Gains and Gaps: Changing Inequality in U.S. College Entry and Completion,” NBER Working Paper No. 17633 (December 2011), 26, Figure 3, www.nber.org/papers/w17633.pdf. For the 1961–64 birth cohort, the top quartile of the income distribution had a 36 percent college completion rate by age twenty-four and the bottom quartile had a 5 percent college completion rate, for a difference of 31 percentage points. For the 1979–82 birth cohort, the top quartile of the income distribution had a 54 percent college completion rate and the bottom quartile had a 9 percent college completion rate, for a difference of 45 percentage points.

  hourly workers, teachers, clergy, farmers, and soldiers combined: David Leonhardt, “Top Colleges That Enroll Rich, Middle Class and Poor,” New York Times, September 8, 2014, https://www.nytimes.com/2014/09/09/upshot/top-colleges-that-enroll-rich-middle-class-and-poor.html; David Leonhardt, “As Wealthy Fill Top Colleges, Concerns Grow over Fairness,” New York Times, April 22, 2014, www.nytimes.com/2004/04/22/us/as-wealthy-fill-top-colleges-concerns-grow-over-fairness.html.

  a ratio of about three and a half to one: David Freed and Idrees Kahloon, “Class of 2019 by the Numbers: Makeup of the Class,” Harvard Crimson, http://features.thecrimson.com/2015/freshman-survey/; Laya Anasu and Michael D. Ledecky, “Freshman Survey Part II: An Uncommon App,” Harvard Crimson, September 4, 2013, www.thecrimson.com/article/2013/9/4/freshman-survey-admissions-aid/; Stephanie Addenbrooke and Emma Platoff, “2019 by the Numbers: First Impressions,” Yale Daily News, http://features.yaledailynews.com/blog/2015/08/28/2019-by-the-numbers-first-impressions/.

  The Harvard survey data suggest that 53 percent of students entering into the class of 2017 had parents making more than $125,000 per year (just above the lower limit for the top quintile). Those from families making more than $250,000 per year (roughly the top 5 percent of income earners in the country) made up 29 percent. Those from families making less than $40,000 per year (roughly the bottom two quintiles) represented just 15 percent of the students.

  The Yale survey data suggest that students from families in the top quintile (over $125,000) represented 56 percent of the class, while those from the bottom two quintiles (less than $40,000) represented 14 percent. Students from families in the top 5 percent of income earners (over $250,000) represented 35 percent of the class. The mismatch between the categories in the surveys and income quintiles requires some interpolation, which explains the “about” in the main text. Finally, both surveys rely on self-reporting and thus undoubtedly reflect both selection biases and student errors about parental incomes. But the patterns that they report are so stark that they are revelatory nevertheless.

  the entire bottom half: See Raj Chetty et al., “Mobility Report Cards: The Role of Colleges in Intergenerational Mobility,” NBER Working Paper No. 23618 (July 2017), 1, 14, Figure 1, www.nber.org/papers/w23618; Gregor Aisch et al., “Some Colleges Have More Students from the Top One Percent Than the Bottom 60,” New York Times, January 18, 2017, www.nytimes.com/interactive/2017/01/18/upshot/some-colleges-have-more-students-from-the-top-1-percent-than-the-bottom-60.html. See also David Freed and Idrees Kahloon, “Class of 2019 by the Numbers: Makeup of the Class,” Harvard Crimson, http://features.thecrimson.com/2015/freshman-survey/; Stephanie Addenbrooke and Emma Platoff, “2019 by the Numbers: First Impressions,” Yale Daily News, http://features.yaledailynews.com/blog/2015/08/28/2019-by-the-numbers-first-impressions/.

  Other studies produce compatible results, reporting, for example, that Harvard and Princeton Colleges admit perhaps twenty-five rich students for each poor one. Elizabeth Stoker and Matthew Bruenig, “The 1 Percent’s Ivy League Loophole,” Salon, September 9, 2013. To be sure, some elite colleges do better at enrolling students from poor families than others. Thus at Berkeley and UCLA, 34 and 36 percent of students get Pell Grants (this is roughly the same as the national average across all schools), whereas at Michigan, only 16 percent do. See Richard Pérez-Peña, “Income-Based Diversity Lags at Some Universities,” New York Times, May 30, 2013 accessed November 20, 2018, www.nytimes.com/2013/05/31/education/college-slots-for-poorer-students-still-limited.html?pagewanted=all.

  substantially greater economic diversity than Harvard and Yale: According to one study, children from the top quarter of the socioeconomic distr
ibution in the United States fill 79 percent of Yale’s slots; to fill 79 percent of Oxford’s slots requires expanding to the top 40 percent of the United Kingdom’s socioeconomic distribution. Joseph A. Soares, The Power of Privilege: Yale and America’s Elite Colleges (Stanford, CA: Stanford University Press, 2007), 14. Oxford has established a scholarship fund expressly and exclusively for undergraduates from lower-income families. See Chris Cook, “Oxford Sets Up £300 Million Scholarship Fund,” Financial Times, July 11, 2012, www.ft.com/cms/s/0/c5c7835e-cb55-11e1-916f-00144feabdc0.html#axzz4G63MSRuk. Similarly, following transformations in admissions practices in the 1960s, Oxford has had four times Harvard’s share of students from blue-collar families. See Joseph A. Soares, The Decline of Privilege: The Modernization of Oxford University (Stanford, CA: Stanford University Press, 1999), 4, Table 1.1.

  from an elite college: This state of affairs, in which rich parents are an effectively necessary condition for acquiring an elite education without being in any way a sufficient condition, follows directly from the basic combinatorics of social stratification.

  A thought experiment illustrates this logic. Suppose a society maintains a steady overall size, with 10,000 people in each generation, who all marry and stay married, with each of the 5,000 households constructed in this way including two children. If just under one-third of each generation graduates college, there will be about 3,000 places in the society’s colleges, and about 30 places in the most competitive 1 percent of the society’s colleges. Now imagine that just under one-sixth of the children from households in the top 1 percent by income graduate from the most competitive 1 percent of colleges and universities. The richest 1 percent of households, in this society, collectively include 100 children, so one-sixth of these children—or roughly 15 students—take up half the places at the most competitive 1 percent of colleges. The remaining half of the places—that is, 15 places—would be distributed among the 9,900 students raised in households from the remaining 99 percent of the income distribution. These non-one-percenter students would have only a fraction of a percent chance of making it to the most competitive colleges. And having elite parents would thus be an effectively necessary condition for getting an elite BA without being anywhere near a sufficient condition.

  Something like this combinatoric pattern applies in the United States today.

  total educational expenditures in the United States: The United States spends $336 billion annually on public universities, compared to $668 billion annually on public K–12 schools, and 336/(336 + 668) = 0.334. See National Center for Education Statistics, “Fast Facts: How Much Do Colleges and Universities Spend on Students?,” https://nces.ed.gov/fastfacts/display.asp?id=75, and Center for Education Statistics, “Fast Facts: How Much Money Does the United States Spend on Public Elementary and Secondary Schools?,” http://nces.ed.gov/fastfacts/display.asp?id=66.

  The shares of total (public and private) expenditures devoted to postsecondary education were 39.8 percent in 1996–97, 39.7 percent in 2008, and 43.75 percent in 2012. Thomas D. Snyder, Charlene M. Hoffman, and Claire M. Geddes, Digest of Education Statistics 1997, U.S. Department of Education, National Center for Education Statistics (December 1997), 35, Table 32, www.finaid.org/educators/educstat.pdf; Susan Aud et al., The Condition of Education 2012, U.S. Department of Education, National Center for Education Statistics (May 2012), 200, Table A-22-1, https://nces.ed.gov/pubs2012/2012045.pdf; Grace Kena et al., The Condition of Education 2016, U.S. Department of Education, National Center for Education Statistics (May 2016), 141, Figure 3, https://nces.ed.gov/pubs2016/2016144.pdf.

  in 1970: National Center for Education Statistics, “Expenditures of Educational Institutions Related to the Gross Domestic Product, by Level of Institution: Selected Years, 1929–30 through 2014–15,” Table 106.10, https://nces.ed.gov/programs/digest/d15/tables/dt15_106.10.asp?referrer=report. In 1970, the amount of spending for all degree-granting postsecondary institutions was $23 billion in current dollars, which is $142 billion when adjusted to constant dollars based on the 2014 CPI. The table cites the following sources: U.S. Department of Education, National Center for Education Statistics, Biennial Survey of Education in the United States, 1929–30 through 1949–50; Statistics of State School Systems, 1959–60 through 1969–70; Revenues and Expenditures for Public Elementary and Secondary Education, 1970–71 through 1986–87; Common Core of Data (CCD), “National Public Education Financial Survey,” 1987–88 through 2012–13; Higher Education General Information Survey (HEGIS), Financial Statistics of Institutions of Higher Education, 1965–66 through 1985–86; Integrated Postsecondary Education Data System (IPEDS), “Finance Survey” (IPEDS-F:FY87–99); and IPEDS Spring 2001 through Spring 2015, Finance component. U.S. Department of Commerce, Bureau of Economic Analysis, National Income and Product Accounts Tables, retrieved January 29, 2016, from http://www.bea.gov/iTable/index_nipa.cfm. All figures in current dollars. Table prepared January 2016.

  investment in nonresidential physical capital: See George E. Johnson, “Investment in and Returns from Education,” in The Level and Composition of Household Saving, ed. Patric H. Hendershott (Cambridge, MA: Ballinger, 1985). See also Langbein, “Twentieth-Century Revolution,” 732.

  the entire nation’s 1840 investment in education: Yale’s fiscal year 2018 budget was $3.765 billion. See Yale University Office of the Provost, “Data at a Glance,” https://provost.yale.edu/budget/data-glance. Total investments in education in the United States in 1840 were $9.2 million (roughly $250 million in 2015 dollars). See Albert Fishlow, “Levels of Nineteenth-Century American Investment in Education,” Journal of Economic History 26 (1966): 418, 420. See also Langbein, “Twentieth-Century Revolution,” 730.

  on postsecondary education: See Grace Kena et al., The Condition of Education 2016, U.S. Department of Education, National Center for Education Statistics (May 2016), 141, Figure 3, https://nces.ed.gov/pubs2016/2016144.pdf. “Direct expenditures on education as a percentage of gross domestic product (GDP) for Organization for Economic Cooperation and Development (OECD) countries with the highest percentages, by level of education: 2012,” at p. 141, https://nces.ed.gov/pubs2016/2016144.pdf

  more rapidly than enrollments since 1970: Expenditures (in constant 2009–10 dollars) were $461 billion in 2010, compared to $120.7 billion in 1970, which yields a growth factor of 3.8 over forty years. Enrollments were 8,581,000 in 1970, compared to 20,583,000, which yields a growth factor of 2.4. U.S. Census Bureau, “Section 4: Education,” Statistical Abstract of the United States no. 131, 2012, www.census.gov/library/publications/2011/compendia/statab/131ed/education.html (see files 219, “School Enrollment,” and 220, “School Expenditure by Type of Control and Level of Instruction in Constant (2008–2009) Dollars”). (Note that the constant dollars referred to in the file name and within the file itself are different.) Similarly, the National Center for Education Statistics reports that total postsecondary enrollment was 20,453,000 in 2007 and 8,005,000 in 1969, for a ratio of 2.56. See Digest of Education Statistics 2017, U.S. Department of Education, National Center for Education Statistics (2017), Table 105.30, https://nces.ed.gov/programs/digest/d17/tables/dt17_105.30.asp.

  increased by nearly 60 percent: 3.8 ÷ 2.4 = 1.58.

  between just 2001 and 2015: Median fall enrollments at Ivy League universities grew from 12,230 in 2001 to 13,702 in 2015, a 12.04 percent increase, while median expenditures grew from $944,755,880 to $1,904,823,037 (in constant 2015 dollars), an increase of 101.62 percent. See U.S. Department of Education, National Center for Education Statistics, “Use the Data,” https://nces.ed.gov/ipeds/Home/UseTheData. Real dollars calculated using BLS CPI GDP Deflator. “The CPI inflation calculator uses the Consumer Price Index for All Urban Consumers (CPI-U) U.S. city average series for all items, not seasonally adjusted. These data represent changes in the prices of all goods and services purchased for consumption by urban households.” “CPI Inflation Calculator,” https://data.bls.gov/cgi-bin/cpic
alc.pl/.

  at the least selective ones: Hoxby, “Changing Selectivity,” Figure 2. Hoxby develops comparable numbers across time by ranking schools according to their selectivity in 1962. Since the identities of the very most selective schools, in particular, have not changed over the past fifty years, this technique captures the changing effects of selectivity and not just spending by contingently chosen named schools. See also Highlights from Rewarding Strivers, ed. Richard Kahlenberg (New York and Washington, DC: Century Foundation Press, 2010), 4, accessed November 18, 2018, www.tcf.org/assets/downloads/tcf_rewarding.pdf [inactive]. Hereafter cited as Highlights from Rewarding Strivers.

  The difference in per-student spending between competitive colleges and noncompetitive colleges partly reflects the fact that the increasingly rich students at competitive colleges demand increasingly luxurious living conditions. The former president of Macalester College—an elite, competitive liberal arts college—observes that in order to succeed as a college president, “you have to recruit some affluent students, and part of the way you recruit affluent students is by having symbols of excellence, like an up-to-date campus center and up-to-date athletic facilities. . . . It’s a visible demonstration that you’re not about to go broke. . . . A few years ago we started investing in athletic facilities because we were being told by parents that the facilities weren’t as good as the facilities in their high schools, and enterprises.” Dylan Matthews, “The Tuition Is Too Damn High, Part VIII: Is This All Rich Kids’ Fault?,” Washington Post, September 4, 2013, accessed November 18, 2018, www.washingtonpost.com/news/wonk/wp/2013/09/04/the-tuition-is-too-damn-high-part-viii-is-this-all-rich-kids-fault/.

  Note, relatedly, that “auxiliary spending” accounted for 41.2 percent of total increased costs in public research universities between 2000 and 2010. Matthews, “The Tuition Is Too Damn High, Part VIII.” For more on college spending per student and revenues, go to the Delta Cost Project, www.deltacostproject.org/. For a list of data sources on colleges, see also Stacy Berg Dale and Alan B. Kreuger, “Estimating the Payoff to Attending a More Selective College: An Estimation of Selection on Observables and Unobservables,” NBER Working Paper No. 7322 (August 1999).