How do top colleges treat applicants from different socioeconomic backgrounds, and how much does it matter for a given student to attend a more elite school? These are timely questions, given that the Supreme Court effectively ended racial affirmative action in higher ed last month. Harvard’s Raj Chetty and two coauthors have published a new paper providing some answers.

Like much of Chetty’s work, it’s based on a treasure trove of very private data. Chetty and company don’t just have admissions records from some selective colleges, public and private. Indeed, that’s the weakest link in their dataset, since they don’t have internal records for all the schools they study. But they do have several other sources, all linked together to draw a complete picture of where students entered school in the first half of the 2010s, how much money their parents made, and how they fared in their early years in the job market: “(1) information from parents’ and students’ federal income tax records; (2) college attendance information from the Department of Education; (3) data from the College Board and ACT on standardized test scores.”

Using these data, Chetty and his coauthors can tell how the admissions departments at elite schools treat rich and poor students with similar credentials, as well as how beneficial it really is to go to an elite school. They focus particularly on “Ivy Plus” schools—the “Plus” adds MIT, Stanford, Duke, and the University of Chicago to the grouping—but also cover other top colleges, including public flagships.

As for getting into college: below is the key chart of available data showing admissions rates for students who apply to Ivy Plus and top-tier public flagship schools. It plots parental income against a student’s chances of getting in, after statistically controlling for their test scores. (Recruited athletes are excluded.) Note that the X axis is not divided evenly; it shows much more detail at the top, with the top 5 percent divided into six bins, while the entire bottom 20 percent is placed in a single bin.

Source: “Diversifying Society’s Leaders? The Causal Effects of Admission to Highly Selective Private Colleges,” NBER, p. 82.

Start at the left side of the chart. Both types of schools seem to have some preference for poorer students, but the Ivy Plus preference is stronger. At the Ivies, students below the 40th percentile get a roughly 25 percent boost to their chances, falling to a nearly 25 percent penalty for those between the 90th and 95th percentiles. Some of this is a side effect of now-illegal racial preferences, however, as the bonus and penalty are both a bit smaller in a separate analysis that accounts for race in addition to test scores.

Above the 95th percentile of income, Ivy Plus chances gradually rebound. Those in the top 1 percent but not the top 0.1 percent have about the same chances as a poor applicant. And the top 0.1 percent gets a massive boost, more than doubling their chances.

Almost half of the boost for the top 1 percent at Ivy Plus colleges comes from a preference for legacy admissions, which are especially large for high-earning alumni: “Legacy students from families in the top 1% are 5 times as likely to be admitted as the average applicant with similar test scores, demographic characteristics, and admissions office ratings,” versus three times as likely for those below the 90th percentile. (The authors don’t speculate as to why this might be. The words “donation” and “donor” never appear in the 126-page study, though there is a reference to the “need to maintain alumni relations.”) Another 24 percent of the boost for the top 1 percent stems from athlete recruitment; at these schools, unlike at flagship public schools, recruited athletes are disproportionately found among students from wealthier backgrounds. The remaining 30 percent comes from non-academic credentials—owing in large part, apparently, to wealthy private high schools that emphasize extracurriculars, help polish personal essays, and so on.

On top of the admissions gap, students from richer backgrounds are a bit more likely to apply to elite schools to begin with, and slightly more likely to attend if accepted as well, though these factors are generally outside the schools’ control and account for just a third of the overall enrollment gap.

The huge admissions bonus for the ultra-wealthy would disappear if elite schools wanted it to. How much would it help everyone else if colleges started making changes to practices like legacy admissions? Significantly, but not as much as one might think.

The bottom line is that, for any number of reasons, richer kids tend to have higher test scores and stronger grades by the time they apply to college, and any system designed to select students with the best academics will reflect those inequalities. As Chetty and his coauthors find, “SAT/ACT scores and academic credentials are highly predictive of post-college success,” too. (Legacy and athlete status, not so much.)

At Ivy Plus schools, about 42 percent of students come from parents in the top 5 percent of the income distribution. Chetty, et al. estimate that eliminating legacy preferences, killing the “admissions advantage arising from the higher non-academic ratings obtained by students from high-income families,” andending “the over-representation of students from high-income families in athletic recruitment” would bring that number down only to 33 percent. Alternatively, if all students with the same test scores had the same chance of going to an Ivy Plus (ignoring whether they actually apply or attend), 28 percent of those who made it would be from the top 5 percent of the income scale.

Instead of (or in addition to) eliminating preferences that benefit the wealthy, colleges might give a bigger boost to poorer applicants. Almost anything is possible with this approach: just keep making the preference bigger until you get the numbers you want. Chetty and his coauthors note this fact but, to their credit, don’t seem interested in pushing it to extremes. Their version of “need-affirmative admissions” is limited to students with strong academic credentials and aims simply to match the economic diversity achieved through the legacy/athlete/non-academic-ratings package discussed above, effectively offsetting those boosts for the wealthy rather than eliminating them directly.

The Chetty paper’s reform ideas are sensible, but that’s not to say colleges will find them attractive. Elite colleges have reasons, albeit unseemly ones, for admitting the extremely wealthy. Further, if the effects of these policies on class demographics are not all that impressive, the effects on racial demographics are almost certainly subtler still—a question closer to colleges’ current priorities and studiously avoided in the new paper.

If you’re wondering why it matters who attends a handful of the most elite schools, Chetty and his coauthors have the data to answer that question, too. They do this in several ways—for example, by looking at students put on schools’ wait lists (who, by definition, have borderline credentials) to see if those who happen to get in outperform those who are rejected and must attend a backup school.

As previous studies have shown as well, smart students tend to do fine financially, whether they go to a top school or not. The average earnings boost is modest. But the Ivies do hold the keys to certain influential institutions and extremely lucrative jobs. Kids who go to an Ivy Plus college have twice the chance of going to an elite graduate school, triple the chance of working at a prestigious firm, and a 60 percent higher chance of making it to the top 1 percent of earners. In other words, elite schools may not guarantee a salary boost for everyone who attends, but they do shape the nation’s elite and hike a student’s chances of really hitting it big in the professional world.

The new paper shows, once again, that top private colleges give strong preferences to certain wealthy applicants—especially rich legacies—in addition to the enormous boosts they give (or at least have given historically) to underrepresented minority groups. It also highlights policy changes that could economically diversify elite colleges, and perhaps the nation’s elite along with them.

But only up to a point. Legacy preferences and similar policies make a mockery of fairness and deserve to end, but even without them, elite universities cannot be both academically meritocratic and socioeconomically representative of the broader society. As the authors themselves note, “changes in admissions policies at highly selective private colleges cannot by themselves increase economic mobility substantially,” both because academic gaps by parental income are already large by the time young people apply to college and because Ivy Plus schools enroll fewer than 1 percent of college students. Most colleges aren’t very selective.

Further, when selective schools are hyper-focused on race specifically and are willing to drop test scores to achieve their racial goals, one wonders how much interest a paper that deliberately ignores questions of racial demographics and admits the predictive value of test scores will generate.

Or maybe Chetty and his coauthors are saving the juicy stuff for the next one.

Photo: Boston Globe/Getty Images


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next