In 2023, the U.S. Supreme Court’s decision in Students for Fair Admissions v. Harvard (SFFA) ended the longstanding practice of considering race in college admissions. A new study by Michael D. Bloem, Ashley Edwards, and six colleagues affiliated with the College Board finds that the ruling reduced both the likelihood that high-achieving underrepresented minority students enroll in highly selective colleges and overall campus diversity at those institutions.
Linking data from the College Board, National Student Clearinghouse, and the American Community Survey, the authors constructed a dataset of more than 12 million students from the 2021 to 2024 graduating cohorts at more than 2,600 institutions. The dataset includes SAT, PSAT, and Advanced Placement exam scores; college enrollment destinations; demographic information; and neighborhood income. To estimate the causal impact of the SFFA ruling, the authors compare enrollment trends at four-year public and private universities to a control group of federal military academies—where race can be considered for national security purposes—and public colleges in states with preexisting race-based affirmative action bans.
In fall 2024, high-achieving underrepresented minority (URM) students in the first class to enter college after the SFFA ruling—Native American, Hispanic, Black, or Pacific Islander students with SAT scores above 1300—were up to 10 percentage points less likely to enroll at highly selective public and private colleges (those with pre-pandemic acceptance rates below 25 percent). Across the roughly 74,000 students enrolled at those institutions, this translates to 2,800 fewer URM students and 2,150 more non-URM students. The impacts were more pronounced among Black students, whose rates of entry into highly selective colleges were between 2.7 and 3.6 percentage points lower than Hispanic students.
At the same time, high-achieving URM students became more likely to enroll in less-selective institutions. Enrollment increased by 4.4 percentage points at selective public colleges (acceptance rates of 25-60 percent) and by 2.5 percentage points at less selective public colleges (acceptance rates above 60 percent). These trends suggest a downward “cascade,” with URM students shifting into institutions that tend to have lower graduation rates and earning outcomes. As a result, the share of URM students in first-year classes at highly selective colleges fell by 4 to 5 percentage points in 2024, while selective public colleges experienced a smaller decline (1.4 percentage points) as enrollees shifted down the selectivity spectrum.
The study also finds limited evidence that institutions replaced race-based admissions with class-based preferences. The share of students from lower-income neighborhoods at highly selective colleges did increase by less than 1 percentage point in 2024, but URM enrollment declined across all socioeconomic groups. Racial contrasts were especially apparent between enrollees from higher-income neighborhoods: URM students were 3 to 3.7 percentage points less likely to attend highly selective colleges, while non-URM students were 2.8 to 3.5 percentage points more likely to do so.
The authors write that the post-SFFA displacement of high-achieving URM students from the nation’s top universities could harm their labor market outcomes, as highly selective colleges are often gateways to prestigious graduate programs, higher-income jobs, stronger networks, and leadership roles in society. Meanwhile, limiting non-URM students’ access to diverse campuses could hinder their ability to engage with alternate viewpoints, bridge cultural differences, and build stronger communities.