How College Admissions Hurt Intergenerational Mobility

(Illustration by Adam McCauley) 

Raj Chetty, a professor of economics at Harvard University and the director of Opportunity Insights, a Harvard-based research and policy group that analyzes big data, has spent much of his career so far analyzing intergenerational mobility—the extent to which people’s economic outcomes are shaped by their parents’. In a 2011 paper in the Quarterly Journal of Economics, Chetty led a group of researchers that examined the effects of kindergarten quality on long-term student outcomes. His team pored over data from Project STAR, a study of 12,000 Tennessee kindergarteners conducted in the 1980s. Among several measures they used to determine success was whether and where students attended college.

By the end of the project, the team had unearthed new research questions—but about college, not kindergarten. Looking at outcomes they were seeing in their own data, they saw an opportunity to explore differences between colleges, especially around access and outcomes for students, two key elements of intergenerational mobility.

A new study by Chetty; John N. Friedman, a professor of economics and international and political affairs at Brown University; Nick Turner, a principal economist at the Federal Reserve; and Emmanuel Saez and Danny Yagan, both professors of economics at the University of California, Berkeley, has been in the making since the 2011 kindergarten study. Using data based on confidential access to anonymized tax records at the IRS, the group calculated the parental incomes of students at various colleges and followed student earnings after graduation.

They asked: What do the data say about intergenerational mobility? What percentage of students at the most selective colleges come from wealthy families? How do the children of low-income families fare in college admissions? What would it take to get them attending more selective colleges at higher rates? Rather than looking only at average incomes, they analyzed the entire distribution to obtain a full and nuanced picture of parental incomes and student earnings after college. The researchers found that even minor changes to the college admissions process would boost intergenerational mobility.

To investigate whether differences in preparation at the end of high school explain differences in the fraction of students who attend selective schools, Chetty and his colleagues asked whether two students from different family backgrounds but with the same SAT/ACT scores were as likely to attend a selective school. For example, looking at all students who scored a 1080 on the SAT, they found that 75 percent from the richest fifth of families attended a selective school, compared with only 51 percent from the poorest quintile of families. This discrepancy between high- and low-income students was not simply a reflection of differences in academic preparation and K-12 educational experiences, they concluded, but something about the processes through which students apply, get admitted, and choose a college.

“We should be able to fix this problem,” Friedman says. “This isn’t about not enough kids scoring a 1080, which is a separate problem that also needs to be addressed, but about how the admissions department handles those students who do have similar scores.” 

Looking at “Ivy-Plus” schools—the eight Ivy League colleges plus Duke University, MIT, Stanford University, and the University of Chicago—the researchers discovered that students who hailed from lower- and middle-class backgrounds were heavily underrepresented. More students came from families in the top 1 percent of earners than from families in the bottom half of the income distribution.

The mobility prospects for low-income students at Ivy-Plus schools are enormous, because graduates have high earning prospects. For low-income students, the team found that admissions departments could shrink intergenerational mobility gaps by more than 25 percent simply by providing need-affirmative preferences—the same types of advantages already afforded to legacy students and athletes. If colleges equalized attendance rates for middle-class students, conditional on test scores, the fraction of students from the so-called missing middle would grow from 28 percent to 38 percent.

“Every year is an opportunity to make an enormous difference,” Friedman says. “The data don’t really point to what people sometimes refer to as the magic age model,” he adds, referring to several studies that have focused on kindergarten or the senior year of high school as critical moments when educational gains are won or lost.

“This is the most important—and definitive—recent study about economic inequality in access to selective colleges,” says Sean Reardon, Professor of Poverty and Inequality in Education at the Stanford Graduate School of Education. “It reveals the important role that college admissions practices play in limiting economic mobility—and suggests that changing admissions policies at selective colleges might lead to greater educational and economic equality.”

^{Raj Chetty, John N. Friedman, Emmanuel Saez, Nicholas Turner, and Danny Yagan, “Income Segregation and Intergenerational Mobility Across Colleges in the United States,” Quarterly Journal of Economics, vol. 135, no. 3, 2020, pp. 1567-1633.}

Read more stories by Daniela Blei.