Same scores, double the odds

Two applicants submit the same SAT score to the same Ivy-Plus college this fall. One comes from a family in the top 1% of the income distribution. According to a study published earlier this year in the Quarterly Journal of Economics, that applicant is more than 2 times as likely to get in, and the researchers can now tell you almost exactly why.

Share

This newsletter grows from your shares. Send it to a fellow colleague today!

In today’s issue (free analysis):

• What Raj Chetty, David Deming, and John Friedman actually measured, using anonymized admissions files linked to tax records and test scores

• The 3 specific mechanisms that create the wealth premium at elite colleges, and the exact share of the gap each one explains

• Why the same premium disappears completely at flagship public universities, and what the researchers found when they checked whether any of it predicts a graduate’s actual success

For paid subscribers:

• How to walk a full-pay client through this data honestly in an intake conversation, without sounding cynical about a process you’re being paid to help them win

• Which of the 3 mechanisms is worth a family’s time and money, which one is already saturated among applicants with real hooks, and which one cannot be bought at any price

• The 1 piece of the gap that has nothing to do with any admissions office, and the specific, sourced case it gives you for treating a flagship public as a real contender rather than a backup

Upgrade to paid

Paid subscribers get the full analysis every issue, including the practitioner frameworks in the section below the paywall. If this newsletter helps you stay ahead in your work, a paid subscription is one of the better professional investments you can make. Many readers expense it as professional development.

Upgrade to paid

That gap is not news to anyone running an IEC practice. Every counselor who has sat across from two equally strong students, one who got in and one who didn’t, has a private theory about why. What’s new is that 3 economists just told you which parts of that theory are correct, which parts are not, and exactly how much each true part is worth.

Raj Chetty, David Deming, and John Friedman are not guessing from enrollment photos. Chetty runs Harvard’s Opportunity Insights lab, and for more than a decade his team has built its reputation by linking IRS records to institutional data that colleges would generally prefer to keep to themselves. For this study, the researchers obtained anonymized admissions files from several private colleges and matched them to federal income tax records and SAT or ACT scores. That match is what separates this paper from the familiar observation that wealthy kids end up at better schools. Most prior work compared students who were already enrolled and inferred the rest. This one can take 2 applicants with the identical test score, one from a family in the top 1% and one from a middle class family, and trace what actually happened to each application.

The schools in the study are what the authors call Ivy-Plus: the 8 Ivy League schools, plus Stanford, MIT, Duke, and Chicago, 12 institutions in total. The comparison group is flagship public universities. That comparison turns out to matter as much as anything else in the paper, and we’ll get to why in a moment.

The team first circulated an early version of this analysis in 2023. The peer-reviewed version, revised and finalized, ran in the February 2026 issue of the Quarterly Journal of Economics, and it is only now working its way into IEC-facing coverage as this fall’s list-building season gets underway, which is the reason it belongs in this issue now rather than 6 months ago.

The headline finding: a student from a top 1% family is more than 2 times as likely to attend an Ivy-Plus school than a middle class student with a comparable SAT or ACT score. Chetty and Friedman wrote in Time in January that if admissions offices treated top 1% applicants exactly the same way they treat everyone else with the same scores, the share of top 1% students at America’s most selective colleges would fall by close to half.

Here is where the paper gets genuinely useful rather than just confirming a suspicion. The researchers split that gap into 2 pieces. About two-thirds of it comes from top 1% applicants being admitted at a higher rate for the same test score. The remaining third comes from something else entirely: differences in how often equally qualified students at different income levels apply in the first place, and how often they enroll once admitted. Hold onto that second piece. It’s the part of this study that has nothing to do with what any admissions committee decides, and it’s the part your practice can actually move.

For the two-thirds of the gap that comes from differential admission rates, the paper names the specific channels and puts a number on each one, which is rarer in this kind of research than you’d expect.

1. Legacy preference is the largest single factor, responsible for 46% of the admission-rate advantage. An applicant from a top 1% family whose parent attended the same Ivy-Plus school is 5 times more likely to be admitted than an applicant with a comparable test score, demographic profile, and admissions committee rating whose parent did not attend. The researchers tested whether this is really legacy status doing the work, or some unmeasured family quality riding along with it, by checking whether a legacy applicant at one Ivy-Plus school gets any advantage at a different Ivy-Plus school their parent did not attend. They don’t. The advantage is specific to the exact institution the parent attended, which is about as clean a signal as this kind of research produces.

2. Non-academic ratings account for 30%. This is the admissions committee’s assessment of extracurricular activity, leadership, and personal qualities, the part of the file that is, in practice, easier to build when a family has the time and money to fund it. A research placement, a specialized instrument, a summer spent founding a nonprofit rather than working a retail shift: these read as stronger non-academic credentials, and they correlate with income for reasons that have very little to do with a 17-year-old’s actual character.

3. Recruited athletics explains the remaining 24%. The sports that carry weight at these particular schools skew toward ones that require private coaching, club travel, and specialized equipment, which is the same underlying pattern showing up in a different part of the application. Recruited athletes make up 10% to 15% of the incoming class at these schools, according to Chetty and his co-author John Friedman.

Now the finding that should change how you talk about all 3 of these with a client. None of them predict what happens after college. In the paper’s data, legacy status, non-academic ratings, and athletic recruitment are unrelated to a graduate’s later earnings, unrelated to admission into an elite graduate program, and unrelated to landing a job at a top firm. Academic credentials, meaning test scores and grades, are the one part of the file that reliably does predict those outcomes. The researchers ran the counterfactual directly: remove legacy preference, non-academic weighting, and athletic recruiting from the admissions process, admit the same distribution of test scores the current class already has, and the share of Ivy-Plus students coming from the bottom 95% of family income rises by 8.8%. Nothing about academic rigor has to give for that shift to happen. The bar just moves to something that isn’t for sale.

None of this shows up at flagship public universities. The paper is explicit: children from high-income families have no admissions advantage at flagship publics once you control for test scores. Not a smaller advantage. Whatever is driving the tilt at Ivy-Plus schools, whether it’s legacy pipelines, an admissions committee’s weighting of the kind of resume only money can build, or the pull of expensive youth sports, it belongs to a small number of private institutions. It is not a general feature of how American higher education sorts students by wealth.

That distinction is worth sitting with before you build a fall list. The stakes are real. In the same piece, the authors point out that fewer than 1% of Americans attend one of the 12 Ivy-Plus schools, yet those schools produce over 13% of the top 0.1% of earners, a quarter of US Senators, half of all Rhodes Scholars, and three-quarters of the Supreme Court justices appointed in the last 50 years. Families are not wrong to care about this gap. Where most of them go wrong is assuming the whole gap runs through channels their money can influence.

Everything above this line is the finding. What you actually say to a client sitting across from you this week, the one who has spent 4 years and a great deal of money building exactly the kind of resume this paper says doesn’t predict anything, is a harder conversation. That’s what the rest of this issue is for.

Higher Ed Insights is reader-supported. If you are reading this as a free subscriber and you work in college counseling, financial aid, enrollment management, or high school education, the paid section is written directly for you.

Upgrade to Paid