End Discrimination Against Working-Class Students at American Colleges

Dartmouth College, circa 1834
Dartmouth College, circa 1834

After decades of ignoring under-representation of working-class students at elite colleges, some politicians and media-makers have finally taken notice. But discussion has been vague and action has been lacking. We won’t bring about true social diversity at top colleges until we address the problem of discrimination.

In January 2014, President Barack Obama and First Lady Michelle Obama launched a new initiative to increase college access for low-income students. The president said that the initiative had the backing of “college presidents—from state universities and historically black colleges to Ivy League universities and community colleges.” He noted that “more than 100 colleges and 40 organizations are announcing new commitments to help more young people not only go to, but graduate from college.” To document these new efforts, the White House released Commitments to Action, with pledges from individual institutions. (See that and more on the initiative here.)

Unfortunately, the promises offered by top private colleges were either vague, paltry, or both. The only Ivy League college that made a specific, numerical commitment to increase low-income enrollment was Yale. That university’s officials promised to intensify their work with QuestBridge, a national, non-profit organization that seeks to match talented low-income students with leading colleges. According to Commitments to Action, Yale pledged to increase “the number of QuestBridge Finalists it enrolls in its freshman class by 50 percent. Yale has traditionally enrolled 50-60 [QuestBridge] students per year, and is now committing to enrolling 75-80 students who apply through this program for entry in fall 2014 and fall 2015.”

But if the “traditional” baseline for measuring the increase is set at the upper level of 60 QuestBridge students per year, then enrolling 75 such students would constitute only a 25% increase. Enrolling 80 would only amount to a 33% increase.

Yale’s commitment to a 50% increase did not survive the next sentence of the report. At most, Yale is promising to admit 30 more QuestBridge students per year than before. According to the latest federal statistics, 30 students equals 0.5% of Yale’s undergraduate enrollment. When it comes to elite, private colleges, President Obama’s initiative is little more than a public-relations effort.

What would a real campaign for social diversity look like? It would define the problem, demand substantial, measurable improvements, and provide credible sanctions for institutions that fail to comply.

The problem is discrimination.

Politicians, journalists, and “experts” are remarkably consistent in attributing low enrollment of economically disadvantaged students to those students’ own failures or, at worst, innocent oversights by college administrators. But working-class students are so severely under-represented at elite colleges, particularly the most prestigious private ones, that such excuses are insulting.

Private colleges are not required to provide data on social diversity. So our best measurement is the percentage of a college’s undergraduates who receive Pell Grants, a federal scholarship for low- and middle-income students.

As shown by U.S. News & World Report’s ranking of economic diversity at top national universities, only a few elite institutions have percentages of Pell Grant recipients higher than the ’teens. There is an enormous gap between the two leading University of California institutions and all the others on the list. (Columbia University is somewhat more diverse than the other Ivies, but I discovered that its percentage of Pell students is greatly exaggerated, due to an error regarding that institution’s enrollment. The figure should be 23%, not 30%.) UCB and UCLA have achieved Pell percentages in the upper-30s without compromising academic excellence. What excuse do the other 23 colleges have?

The Southern Education Foundation recently reported that the majority of U.S. public-school students are low income, as defined by participation in free- and reduced-rate school-lunch programs. The majority, and most likely a large majority, of public-school students would qualify for Pell Grants, if they went to college. Yet only a small fraction of students at most elite private colleges receive Pell Grants.

Likewise, education scholar John Jerrim has concluded that low-social-status students are under-represented at U.S. colleges, and that the “access gap” between them and high-status students cannot be explained by academic achievement. The disparity is worst at top private institutions. Differences in academic performance account for 60% of the access gap at “elite public sector colleges,” but only 48% of the gap at “elite private sector colleges.” (That study, “Family background and access to ‘high status’ universities,” is available here.)

That is to say, merit, as defined by grades and test scores, cannot explain even half of the social-class gap at top private colleges in the U.S. The barriers facing low-status university applicants are actually lower in England, where academic performance explains 73% of the access gap. I thought England was the country with a class system, not America.

It is absurd to suggest that the staggering social divide in U.S. higher education is caused entirely by working-class students performing poorly or refusing to apply to colleges that would happily admit them. In fact, class bigotry is deeply ingrained in the practices of elite institutions.

Nearly all of America’s top private colleges, and many of the top public ones, grant large admissions preferences to “legacies,” children of graduates. Or, to be more precise, children of rich graduates. In his 1991 study of applicants to Harvard, education scholar David Karen found that legacies who request financial aid when applying lose most of the legacy advantage. (You can read more about that and legacy preferences in general in Daniel Golden’s book, The Price of Admission.)

more about that and legacy preferences in general in Daniel Golden’s book, The Price of Admission.)

Let’s look at a hypothetical case involving two white, male applicants to an Ivy League university. Applicant A is the first in his family to apply to college. His parents are manual laborers and his family’s income is under $30,000 a year. He hopes that a degree from a prestigious college will afford him a better career than either of his parents have. He also hopes to benefit from Ivy League universities’ “no loan” financial-aid policy for low-income students. All the public institutions to which he is applying would require him to take out student loans. He attended an under-funded public high school and worked part-time jobs to help his family pay the bills. Even so, he scored a 2100 on the SAT and achieved a grade-point average of 3.7.

Applicant B is a fifth-generation legacy. His parents do not work, due to their investment income, which is valued at over $10,000,000 annually. Applicant B does not need a career and he will be better off financially than his parents as long as his family’s investments continue to rise in value. Applicant B attended a top private high school, scored 1700 on the SAT, and achieved a grade-point average of 3.5. He did not have a job during high school. The admissions office rejects applicant A and admits applicant B.

Do something like that once and you are a bigot. Make a policy of doing that and you are an especially low order of bigot. That hypothetical is hardly an exaggeration. In their 1998 book, The Shape of the River, former Princeton President William G. Bowen, former Harvard President Derek Bok, and four other authors wrote about legacy preferences in admissions at elite private colleges. They reported that white legacies who scored in the 1100 range on the SAT (using the old 1600-point scale) were nearly as likely to be admitted as white non-legacies who scored 1300-1600. The former had a 22% chance of admission, while the latter had a 24% chance. (Don’t get the idea that such inequities caused the authors to oppose legacy preferences, however.)

The feudal absurdity of legacy bias is not an isolated issue. Like other destructive behaviors, discrimination gets easier with repetition. Legacy advantages are proof of a system of thought that treats the rich and well-connected as more deserving than others. That bias inevitably affects all admissions decisions. No government policy has confronted it directly, though all elite private colleges are tax-exempt.

Discrimination against working-class whites is also discrimination.

In their 2009 book, No Longer Separate, Not Yet Equal, education scholars Thomas J. Espenshade and Alexandria W. Radford analyzed the impact of class and race on admissions at selective colleges. Looking at applicants with the same academic qualifications, the authors found that “for white applicants to private institutions, there is a low SES [socio-economic status] admissions disadvantage.” Whites who described themselves as “lower class” were about one-third as likely to be admitted as equally qualified whites who described themselves as “upper class.” They were also much less likely to be admitted than equally qualified African Americans and Hispanics from their own social class. (There is more detail on Espenshade and Radford’s findings here.)

Invidious discrimination in favor of rich whites and against working-class whites is worse than Espenshade and Radford’s statistics suggest. They compared applicants with the same academic scores. But that ignores the pronounced advantages of wealthy students, which include attending the best schools, being able to obtain private tutoring and test coaching, and having parents who are likely to be familiar with the admissions criteria of elite institutions. If you come from a working-class or low-income household, and you achieve scores that match those of wealthy applicants, you are actually more qualified than they are.

Though few in the media have noticed, discrimination against working-class white applicants reveals that elite college administrators have a warped definition of diversity. Let’s create a better definition.

To end discrimination, abolish preferences for legacies and donors’ children. At the same time, expand affirmative-action programs to include social class.

There is no excuse for hereditary privileges such as those offered to children of wealthy alumni and donors. Likewise, in The Chosen, a study of the history of admissions practices at Harvard, Yale, and Princeton, Jerome Karabel showed the value of affirmative-action programs. Those three universities’ percentages of black students did not increase substantially until the late 1960s and early 1970s. Why then? Because that was when all three institutions implemented aggressive programs of race-based affirmative action. Likewise, gender-based affirmative action proved essential for increasing gender diversity at American colleges. To solve the problem of social exclusion, it is vital that we add social class to existing affirmative-action criteria.

The government must require that institutions provide detailed statistics on social diversity, among graduate and professional students as well as undergraduates. Pell Grants are an inadequate measure of the percentage of working-class students on campus.

To make affirmative action work, the government needs to obtain reliable statistics from colleges on the social backgrounds of students. Private institutions are not required to supply such information. Consequently, the only verifiable measure of low-income students at any private college is its number of Pell Grant recipients. But Pell Grants are normally only given to undergraduates, and eligibility for that program is no longer a reliable indicator of low income.

Writers on education policy have long followed the rule of thumb that being eligible for a Pell Grant places a student in roughly the bottom half of the U.S. income scale. In most cases, that would be an accurate assumption.

However, the Pell program has expanded dramatically since 2007. During that time, the number of Pell recipients has risen by 73%, and the number of recipients with family incomes above $60,000 per year has risen by nearly 900%. In fact, the most recent government statistics show that more than 2,000 students from households with incomes over $100,000 per year obtained Pell Grants. If admissions officers worked the numbers right, the eight Ivy League colleges could easily quintuple their number of Pell recipients without admitting a single applicant with a family income below $70,000 per year. That would still leave tens of thousands of Pell students in the over-70,000 income category, allowing other elite colleges to do the same.

The Crash of 2008 and the Great Recession muddied the issue further. Many investors and professional/managerial staff lost wealth or even jobs as a result of the crisis. Some of them are now working lower-paying jobs or relying on savings, factors that could make their children newly eligible for Pell Grants. Students from such families may not be as well-to-do as they were previously, but it is misleading to place them in the same category as first-generation collegians whose parents are low-paid, manual laborers. If a college’s increase in Pell recipients is slanted toward students in the former category, exclusion of working-class and genuinely low-income students may be as bad as ever. We need to know if that is the case.

There is a similar problem with regard to recruited athletes. If even a few Pell recipients at a particular college were recruited by the athletic department, that school’s Pell statistics will exaggerate social diversity among students who are not sports stars. But there is now no independent means of determining how many Pell students are recruited athletes.

Nor is income the only indicator of economic disadvantage. There is evidence that having parents who did not attend college is a larger obstacle to college admission than low income. In a study of admissions at 19 selective colleges published in 2005, former Princeton president William G. Bowen and two other education scholars found that those institutions were less likely to enroll first-generation college students than students from the bottom income quartile. They reported that 6.2% of students were in the former category, 10.8% in the latter, and only 3.1% in both. We require data that pertains to all categories of class disadvantage—low income, first-generation college student, child of manual laborers—in order to end admissions discrimination based on class.

Likewise, social diversity is essential in graduate and professional schools, as well as undergraduate studies. Recent discussions of that issue have largely ignored students at the post-bachelor’s degree level. That omission is unacceptable.

For colleges that fail to achieve social diversity, penalties must include a loss of tax exemptions, rather than Pell Grant cuts. We must also amend laws against educational discrimination to take account of social class.

The Obama administration has called for federal funds to be “steered toward high-performing colleges that provide the best value.” The president wants to create a federal ranking system for colleges. Criteria for the rankings would include affordability and access, with the latter measured by “the percentage of students receiving Pell grants.” Obama has promised to propose legislation that ties federal funding to the rankings, but has not yet released a detailed plan for ranking colleges. According to a White House press release from October 2013:

The Administration will seek legislation using this new rating system to transform the way federal aid is awarded to colleges once the ratings are well developed. Students attending high-performing colleges could receive larger Pell Grants and more affordable student loans. 

The language (“could receive”) is tepid. But the president’s proposal holds out the possibility of funding cuts, or at least smaller funding increases, for colleges that rate low on access and affordability.

Nonetheless, it would be a mistake to expect that implicit threat to have any impact on admissions at the most socially exclusive colleges. Given that the president was impressed by the vague and small-scale promises offered by elite colleges in Commitments to Action, there seems to be little danger of institutions getting into trouble with this White House over lack of access for low-income students. But even if the president and Congress started cutting Pell funds to colleges that fail to improve social diversity, that action would achieve nothing.

At last federal count, Harvard’s endowment exceeded $30 billion, while Yale’s was over $19 billion. It is ridiculous to expect administrators at such institutions to be frightened by the threat that some of their students might lose federal grant funds, especially since that occurrence would furnish an excuse for admitting fewer low-income students. The loss of tax exemptions is a far more effective sanction, and effective sanctions are long overdue. It would be doubly useful if the proceeds from taxes on elite (and elitist) institutions were channeled into need-based scholarships and funding for public education in general.

Our ultimate goal must be to end class-based educational bias by making it illegal. There are areas of the economy (home-buying, for instance) in which it would be impossible to ban discrimination based on social class. Education is not one of those areas. (For more on that issue, see this article.)

It is up to the people to fix America’s socially segregated system of higher education. That task cannot be left to college administrators.

Officials at top colleges have repeatedly declared mission accomplished on social diversity, and they have not been swayed by evidence to the contrary. In 2011, The Chronicle of Higher Education published a list of the percentage of Pell Grant recipients at the 50 richest colleges in the U.S. The three most eminent Ivy League universities, Harvard, Princeton, and Yale, each ranked in the bottom ten, with percentages in the single digits. Harvard ranked the lowest of the three, 49th, with only 6.5% of its students receiving Pell Grants.

William R. Fitzsimmons, dean of admissions and financial aid at Harvard, and Sarah C. Donahue, Harvard’s director of financial aid, wrote a letter to the editor stating that their institution “has played a leadership role among colleges and universities to attract a diverse array of students, including those eligible for Pell Grants.” Consequently, Fitzsimmons and Donahue were “disappointed” by the “misleading” 6.5% figure. That percentage is only accurate, they wrote:

if you count both students attending Harvard College, the institution that comes to mind when most people think of undergraduates attending Harvard, and also the thousands of nontraditional students who took even one undergraduate class at the Harvard Extension School that year.

The two administrators explained that the majority of extension-school students were not eligible for Pell Grants because they were not enrolled in degree programs. If all extension-school students were excluded from the sample, Fitzsimmons and Donahue argued, Harvard’s percentage of Pell Grant recipients in 2008-09 would rise to 12.8%.

The Chronicle responded that it was not feasible to exclude Harvard’s extension school because “to do so would have compromised the analysis by basing it on data that institutions had submitted individually, excluding some students, rather than on national baseline data collected and policed by the U.S. Education Department.” The editors added that “the article noted that the enrollment figure for Harvard included its extension school.”

The Chronicle was on firm ground, but the main point is that the difference between the two percentages—6.5 and 12.8—is less significant than Harvard officials pretend. Under either measure, Harvard’s percentage of Pell Grant recipients was lower than its percentage of students who were legacies, as reported by Daniel Golden of the Wall Street Journal in 2006: 13%.

Harvard’s dispute with The Chronicle of Higher Education continued in 2013, when that journal published another article listing percentages of Pell Grant recipients at top colleges. The Chronicle placed Harvard’s figure at 11%. William Fitzsimmons wrote another letter to The Chronicle’s editor stating that, if Harvard’s extension school were excluded from the statistics, Harvard’s percentage of Pell Grant recipients would be 16.9% in the year used for The Chronicle’s study (2011) and 17.2% the following year. Fitzsimmons wrote: “This represents an 81-percent increase in the number of Pell Grant recipients since 2004, when we launched a targeted initiative to encourage talented low-income students to apply to, and attend, Harvard College.”

However, Fitzsimmons neglected to note that the latest federal data then available showed an 80% increase in Pell Grant eligibility nationwide since 2007. He also failed to note the nearly-900% increase in Pell Grants to students with annual family incomes over $60,000 during the same time frame. How much credit should Harvard receive for increasing its percentage of Pell students by a rate that almost exactly matches the overall national increase in eligibility for that program? Like Fitzsimmons, other administrators at elite colleges have boasted of recent, dramatic increases in enrolling Pell Grant recipients. Those assertions must also be weighed against the expansion of the Pell program nationwide.

Back in 2004, William G. Bowen corresponded with Fitzsimmons about social diversity at Harvard. The dean informed Bowen that Harvard was in the midst of a campaign to recruit low-income applicants. Summarizing Fitzsimmons’s letter to him, Bowen wrote: “A large part of this outreach effort is directed toward what Fitzsimmons calls ‘non-traditional’ families. Major efforts are made to demystify the application process and to emphasize the much greater availability of financial aid today.”

Those must have been some efforts. Harvard earned The Chronicle’s ranking of 49th most socially diverse college among the 50 richest four and a half years after Fitzsimmons’s reassuring missive to Bowen. A constant feature of debates about higher education is Harvard officials’ unshakeable conviction that they are doing a great job of increasing social diversity.

Harvard is by no means alone on that score. In 2003, the U.S. Supreme Court heard legal arguments in Grutter v. Bollinger, in which the plaintiff sought to end race-based affirmative action. Eight top colleges—Brown, Chicago, Dartmouth, Duke, (yes) Harvard, Pennsylvania, Princeton, and Yale—joined to present a friend-of-the-court brief in defense of the policy. The authors of the brief depicted preferences based on race as just one of many categories of admissions advantages granted by college administrators. Responding to a suggestion from the plaintiff’s lawyers that social background should replace race as a category of affirmative action, the lawyers for the eight colleges wrote:

Although petitioners suggest that universities should consider factors like economic circumstances and personal hardships, the truth is that those factors are already taken into account in the typical selective admissions process.

That is a remarkable assertion. Lawyers for the eight colleges replied to a suggestion that they apply class-based affirmative action by stating that institutions already practiced some form of that policy. Specifically, they implied that their institutions practiced class-based affirmative action. Then there is this passage, in which class-based affirmative action is given equal standing alongside preferences for legacies and recruited athletes.

Admissions officials give special attention to, among others, applicants from economically and/or culturally disadvantaged backgrounds, those with unusual athletic ability, those with special artistic talents, those who would be the first in their families to attend any college, those whose parents are alumni or alumnae, and those who have overcome various identifiable hardships.

It is beyond doubt that recruited athletes and legacies received—and still receive—preferences. In 2005, William G. Bowen published a study of admissions data (from 1995) at 19 selective colleges, including the eight that submitted the Grutter friend-of-the-court brief. He reported that being a recruited athlete improved an applicant’s admissions chances by 30 percentage points. For legacies, the advantage was 20 points.

For low-income applicants, however, Bowen found “no perceptible difference in the chance of being admitted.” Other evidence tells the same story. In 2004, the Century Foundation published a listing of the percentage of Pell Grant recipients at leading colleges, based on data from the 1999-2000 academic year. At the University of California at Berkeley, a leading institution that practices a form of class-based affirmative action, 32.4% of undergraduates received Pell Grants. At the eight colleges that filed the Grutter brief, the percentages of Pell recipients ranged from 6.8% to 12.4%, and five of the eight institutions were in single digits. If those latter colleges “gave special attention to” economically disadvantaged applicants, there was no evidence of that in their enrollment statistics.

An extremely charitable interpretation of the eight colleges’ Supreme Court brief would be that officials at those schools simply forgot that, while they granted preferences to legacies, recruited athletes, and under-represented racial minorities, preferences for “economically disadvantaged applicants” remained a theoretical concept on their campuses. A less charitable interpretation would be that those officials misled the Supreme Court of the United States.

Despite the secrecy that shrouds admissions practices at top private colleges, it is easy to discern a pattern of deceit. Administrators at those schools are quick to proclaim commitments to social diversity and just as quick to support those claims with false assurances and misleading statistics. They have every motive to distort the reality of their institutions’ practices, and—so far—no reason to fear consequences if they do.

Most of America’s elite colleges have long histories of class discrimination and equally long histories of covering it up. Genuine reform will begin only after we acknowledge, and take action against, discrimination. Action starts with the end of tax exemptions for wealthy, socially exclusive colleges, those that fail to meet UC Berkeley’s standard for enrolling Pell students (36% of undergraduate enrollment).

Then we must demand detailed information on social diversity from colleges, and craft new laws to address discrimination revealed by such data. However appealing their rhetoric of inclusion, politicians, journalists, and other authorities who reject those measures are opponents of social justice in American education.