Affirmative Action, Legacy & the Power of Elite Colleges

CHRISTIANE AMANPOUR, CHIEF INTERNATIONAL ANCHOR: So, since the U.S. Supreme Court’s controversial decision to end affirmative action, American colleges are now being pressured to stop legacy admissions. It’s a process that historically benefits white wealthy applicants. Harvard economic professor, Raj Chetty, recently investigated this practice and the consequences of such highly selective admissions. And he is joining Hari Sreenivasan to discuss those findings.

(BEGIN VIDEO CLIP)

HARI SREENIVASAN, INTERNATIONAL CORRESPONDENT: Christiane, thanks. Raj Chetty, thanks so much for joining us. You run Opportunity Insights in Harvard, which looks at the kind of different social and structural forces that are giving kids chances or not giving them chances at success. So, you just cranked out a report here about highly selective colleges, what we think of as Ivy League schools, plus the kind of top tier schools. First, why did you do that and what did you find?

RAJ CHETTY, PROFESSOR OF ECONOMICS, HARVARD UNIVERSITY AND DIRECTOR, OPPORTUNITY INSIGHTS: Yes. Thanks, Hari. So, what we’re doing with this new study it’s looking at the role of a small number of colleges, highly selective colleges and what impact they have on economic mobility in the U.S. Now, to start, you might say, now these colleges they educate only something like a half a percent of Americans. So, in the grand scheme of things, they can’t be very important in driving inequality and social mobility in the U.S. as a whole. However, they do play a very outsized role in shaping America’s leaders. So, if you look at measures like what fraction of leading politicians, what fraction of inventors, scientists, Supreme Court justices, leading artists, et cetera, went to one of these Ivy League or other similar colleges, those numbers are often very large, often 30 percent, 50 percent of folks in these positions of influence have gone to one of these colleges.

SREENIVASAN: Yes. I’m looking at the report here and it says, there are 15 — so, if you graduated from one of these 12 colleges, 15 percent of the top 0.1 percent of the United States of income distribution, a quarter of all U.S. senators, half of all the road scholars and three quarters of Supreme Court justices appointed in the last 50 years went to one of these 12 schools. I mean, that’s an incredibly small group of schools that produces this group of leadership.

CHETTY: That’s exactly right. And that’s why I think it’s important to pay attention to what’s happening at these schools in particular. Because even if they educate a small number of Americans on the pool, the people they’re channeling into these positions have a great deal of influence on society in general, the CEOs of companies, making laws, inventing new things and so on. And so, motivated by that, what we’re focused on is asking, who’s getting into these colleges, right? Because it seems to matter in terms of, you know, who might be in these positions of influence. So, who is getting in and how might we make colleges more equitable in terms of who they let in.? And so, what we started out finding is that suppose, Hari, you take a set of kids who have the same SAT scores as a simple measure of pre-college qualifications. And you ask, suppose you’re a kid from a high-income family, so, the top 1 percent of the income distribution, and I’m a kid from a middle-class family and we both have the same SAT score, say we both got 1,500 on the SAT, which puts us at the 99th percentile. What are the odds that you go to one these colleges and what are the odds that I go to one of these college? It turns out that you’re about two and a half times as likely to go to an Ivy plus college, an Ivy League college, relative to me, even though we have the exact SAT scores if you come from a family in the top 1 percent, relative to the middle-class. So, these colleges tend enroll a disproportionate share of kids from high- income families, even controlling for a simple measure of their precollege qualification as measured by standardized tests.

SREENIVASAN: So, what about if these high-income students were to apply to different schools? How do you know that they might not get in there? Maybe because one of the things that you point out is the influence of legacy admissions, meaning if my dad went there or I have family members that went there, how much of a factor is that?

CHETTY: If your parents went to one of these college, you typically have a quite significant advantage in terms of your odds of getting ten. We actually estimate that you’re about five times as likely to get into one of these colleges as a candidate with comparable credentials whose parents did not happen to go to that college. Now, why does that amplify the high-income admissions advantage? Because a lot of folks who went to places like Harvard and Yale and Princeton have very high incomes. And so, naturally, the kids of those parents who are coming from high-income families, on average, and that’s contributing to the high-income admissions advantage that we started out talking about. So, that’s one important factor. The other important factors are what we call nonacademic readings. So, these are school practice, as you might know, holistic admissions. They don’t just look at your test course, they look at many other things. What kinds of extracurricular activities where you involved in, what is your overall portfolio look like that you’re applying with from your college and so on. And what we find is that kids from high-income families, the top 1 percent in particular, are much more likely to get high nonacademic ratings from admissions committees, and that’s coming entirely from the fact that they attend certain high schools, typically elite private high schools, very expensive schools, which tend to produce very high nonacademic rating for their students relative to public schools. So, why might that be? You know, you get involved in more extracurricular activities, you have greater support from your teachers and guidance counselors or people go to a small school where people are really able to invest a lot in developing your college application and building your profile. Whereas in a public school, you know, no one guidance counselor for 600 kids on average, that’s just not as feasible. And so, that creates another big advantage for kids from high-income families. And then, the third factor, Hari, is athletic recruitment. So, you might have the intuition that athletes, you know, come uniformly across the income distribution, that’s not the case. And many of these colleges, about 10 percent of students who are attending are recruited athletes, and those athletes tend to come predominantly from high-income families, actually.

SREENIVASAN: We’re having this conversation in the wake of an important Supreme Court ruling that said that using race as one of the data points to consider a student’s admission into a college was basically unconstitutional. So, what you’re pointing out here is interesting because it’s not race but there are certainly factors that tip the scales in favor of high-income students in a way that, well, the rest of us do not even have an opportunity to compete on because we’re just not wealthy, we didn’t go to those perfect schools that created the opportunities for us and we might not have had parents that already went to these schools.

CHETTY: I think that’s right, Hari. And so, you know, following the Supreme Court decision, which is focused on race, whereas we are focused on class here, those two things are related but different. A lot of the conversation that has ensued has focused on possible class-based affirmative action policies. So, if we can’t look directly at race, maybe we can give a hand to kids from lower-income families, maybe that would make sense as an alternative. What we’re finding here is actually, before you even think about that, you know, thinking about, in a sense, putting your thumb on the scale for kids from lower-income families, the first thing we could do is just take the thumb off the scale that we currently in favor of kids from high-income families. You know, just make it more neutral by income before thinking about giving an advantage even to kids from lower-income families. That itself would have quite a significant impact on socioeconomic diversity at these colleges that are shaping society.

SREENIVASAN: You know, just recently Wesleyan became — the Wesleyan University became the first one to come out and say, we’re not going to use legacy, we’re not going to use the fact your parents or grandparents went to this college as one of those decision criteria. Do you think that will change things?

CHETTY: I think it’s quite possible. I think there are a lot of conversations happening now about whether these kinds of admissions and practices makes sense, you know, prompted partly by the Supreme Court decision. I think lots of politicians are going to have to revisit how they do admissions. And what we’re seeing with this new study is it’s important to take, you know, broad look at whether what we’re doing makes sense. And one thing I’d emphasize is when you look at factors like legacy, nonacademic ratings and athletics and so forth, you know, you might make the argument that maybe it makes sense to put some weight on those factors because maybe those kids are genuinely more qualified, maybe they’re going to have better outcomes in the long run. They’re going to reach those positions of leadership that we started out talking about. But actually, Hari, you know, with this data, we’re able to follow hundreds of thousands of kids overtime and look at how they ended up doing 10 years after college. And what we’re finding is that there’s actually no evidence whatsoever that the kids who are getting these admissions advantages, the legacy students, the ones with the high nonacademic ratings, the recruited athletes, have any better outcomes. In fact, they actually have somewhat worse outcomes than the kids who don’t have those advantages, which in my view, makes it quite a bit harder to justify why we’d have those preferences.

SREENIVASAN: Where is the data? What are you looking at?

CHETTY: Right. So, in our team, Hari, you know, we study issues of economic mobility and our approach is a big data approach to tackling these questions. So, what we’re doing is linking data from several different sources, from federal income tax records covering students and their parents, to Department of Education data on college attendance to internal data from many colleges across America on who applied, who got in, who ended up coming. So, internal admissions records. All of that data is linked, anonymized, and then analyzed. So, that’s what allows us to understand people’s parental income backgrounds, follow their trajectory’s overtime, look in detail at how admission office rations are affecting outcomes and so forth. That’s the power big data, being able to study these questions.

SREENIVASAN: Was there anything that surprised you when you saw this?

CHETTY: Yes. So, you know, I think two things surprised us. So, first, if you are actually going to look at prior work on these issues, it suggests that going to one of these colleges might not actually matter so much based on this logic that, you know, think about the very selected set of kids who are getting into a place like Harvard or Yale at this point where have admission rates of less than 5 percent, you’re of course picking a very small set of very talented kids who presumably could’ve done well had — even had they not gone to these colleges. So, there’s a big debate about whether we see great outcomes from these schools because they’re actually doing something that’s adding value or it’s just that they’re selecting a set of kids who would’ve done well anyway, right? And if anything, there was some prior work that’s been picked up in popular discussions which suggested that maybe it doesn’t matter so much, maybe it’s more who’s getting as opposed to the effects of these colleges. Now, what we’re doing with this new data is using an approach of comparing kids who barely get in off the waitlist versus kids who don’t. So, look at a set of waitlisted applicants, they’re very close to the margin of getting into these colleges, but it turns out, by chance, you know, someone happens to play the right musical instrument that’s needed to fill a college orchestra in the year, and they get in. And somebody else, you know, you didn’t play that instrument, that you ended up not getting in. So, this gives us kind of an experiment where we can now compare our outcomes, look at what happens over time and we see that if you got lucky and got in, you have tremendously better outcomes, in particular, tremendously better chances of reaching the top of society defined in various ways. Having earnings in the top 1 percent, going to a top graduate school, working at a very prestigious firm that is often a pathway to positions of influence, like we started talking about. And so, I think that the magnitude of that, you know, doubling or tripling your odds of getting into these — some of these positions really surprised us.

SREENIVASAN: So, what should they do? I mean, that’s one of the things that the top 12 schools are going to look at your report and say, well, thank you very much for confirming what we thought, which is that we produce phenomenal outcomes and our alumni go on to great things, but, you know, Harvard’s going to say, legacies are just one part of this and, you know, whether you play the tuba or not is just one part of this, we look at the holistic students. So, what is your suggestion on what could improve, I guess, society and give people better opportunities? How should they think about their admissions policies?

CHETTY: We consider two different types of approaches in the study we’ve released. One is to directly address the three factors we’ve identified that explain the high-income admissions advantage. So, take a look at, you know, ending legacy admissions, putting less weight on these nonacademic factors that basically seem to favor high-income students but don’t predict future success, and recruit athletes differently. If you look at state flagship public colleges, take a place like U.C., Berkeley or University of Michigan, for example, Hari, there you find a very different pattern where there’s actually no difference in admissions rates conditional on SAT scores by parental income. And what are those colleges doing differently? They don’t have legacy admissions, they put very little, if any, weight on these nonacademic factors and they recruit athletes uniformly across the income distribution. They don’t just have sports teams and athletics that tilt towards very high-income families. If we move in that direction, we estimate that that would increase the number of low- and middle-income students at these colleges by about 10 percent. And just to put that number in context, in the context of current discussions and other domains, other studies have estimated that ending race-based affirmative action, if it didn’t result in any other changes in admissions practices at these colleges, would also change the number of — would reduce the number of black and Hispanic students by about 10 percent on this colleges. So, that’s just to say, you know, in terms of magnitudes, the issues we’re talking about here are of comparable magnitude and, here, focused on the class dimension, they’re focus on the race dimension. So, these are important policy changes that colleges could consider. Let me make one final point here. So, sometimes colleges will say, well, you know, sitting there in your office, like, theoretically, that seems like a good idea but we need to have a sports team and, you know, we care about alumni relations and that’s why we have to maintain legacies and so forth. So, I can understand, you know, it’s a complicated decision, there are many factors going into play. Another way you could look at it is what if we provided a little bit of support in the admissions process, a little bit of a preference for the kid who got a 1,500 on the SAT coming from a very low-income family. Just like we give a preference for legacy students or kids from high-income families, effectively, what if we essentially gave a preference for those kinds of kids? High academic achievement kids from lower-income families who are currently getting in at much lower rates, we provided some support there that could also have a very similar, I think, positive effect in the long run.

SREENIVASAN: If you had a magic wand and you could tell the president of these 12 colleges one thing that they can do what would you say that they ought to do and why would it be in their best interest? Because right now, the system that they’re in, whether they like it or not, gives them the prestige of being super exclusive, whether that’s right or wrong, and it also says, our alumni go on to do great things. So, how would you kind of think of that economically and say to this president, here’s a reason why it would be good for you to do what I’m talking about to change your admissions policies?

CHETTY: Yes. So, Hari, I’ve been talking with many college presidents and leaders of colleges in the context of doing this work. And the first thing I would notice, I think, many people have a sincere intention of trying to maximize the social impact of these kinds of institutions. They recognize they have a social purpose. I think that’s manifested. And the fact that many colleges are quite concerned about racial diversity, right? So, in light of the Supreme Court decision. So, there is an interest in diversity. And I think this is another important dimension of diversity that people are genuinely interested in. Now, I would say concretely, there are things one can do in admissions that will genuinely make a difference. We see that very clearly in the data now. Do they come at a cost? You know, that’s harder to say. Are they going to affect donations and fund-raising and other aspects of campus life that universities have to consider? We don’t directly speak to that in the study. What I will note though is it’s not totally obvious that there’s going to be a big cost there. You could imagine, for instance, a scenario where child from a lower-income family who gets into these — one of these colleges and goes on to become extremely successful may actually feel more compelled to donate back and support the institution because they genuinely feel that it made a difference as opposed to someone from a high-income family who kind of felt like they were going to make it anyway, this was kind of place to stop, you know, along that pathway. So, I think it’s not clear that there are trade-offs and we need to explore these issues more directly.

SREENIVASAN: Raj Chetty, thank you so much for joining us.

CHETTY: Thank you, Hari. My pleasure.

(END VIDEO CLIP)