Who cares about the Ivy League? (repost)

Too much of our education discourse focuses on a handful of tiny elite schools.

I wrote a post about the recent Supreme Court ruling on affirmative action, but I focused on the broader social and cultural implications. What I didn’t cover — but which the lawsuit itself was actually about — was the impact on elite education in America.

The reason I didn’t write about this is that I just don’t think it matters all that much. Americans make an absolutely huge amount of fuss over marginal changes in the student bodies of a few tiny overpriced private schools. My view is that this obsession is extremely misplaced. Instead of spending our political capital and our attention on trying to change the allocation of spots at Harvard, we should be working to expand access to quality education. The number of spots at elite schools should be radically increased — Japan, China, and Korea, whose top schools have student bodies more than twice the size of Harvard’s, provide examples to follow here. But an even more important task is to expand and support the low-priced state schools that provide a life-changing education to the children of the American working class. The Cal State and CUNY and SUNY systems are ultimately more important than Harvard and Princeton.

Anyway, I made this case in a post back in early 2021, before this blog had a large audience. I don’t think I have much to add to what I wrote then, so I thought I’d just repost it. Enjoy!

---

There’s a hell of a lot that’s broken and messed up about our education system here in the United States. But far too much of our discourse about education focuses on a handful of tiny elite private universities, mostly located in the Northeast.

Everyone seems to care a whole lot about the Ivy League. When a bunch of Ivies (and a few other schools) were found to have sold spots to a few rich kids back in 2019, it caused an unholy shitstorm of rage. A vast amount of ink has been spilled over that lawsuit alleging anti-Asian discrimination at Harvard. When Cornell changes the name of its English department to the Department of Literatures in English (apparently because the latter is less colonialist or something), it draws national commentary.

Over at Slow Boring, Matt Yglesias writes that if Ivies really wanted to promote social justice, they would let in more poor kids instead of fiddling with the name of the English department. Of course, he’s right. Elite schools let in mostly rich kids, because they have every incentive to do so. These schools all give out need-based financial aid, which means that rich kids are a profit center (they pay full price), while poor kids are a cost center (they get a free ride). Even a nonprofit business likes to maximize profit centers and minimize cost centers, so of course the Ivies try their hardest to let in rich kids. Also, given America’s low economic mobility, rich kids are highly likely to become rich adults, and rich adults give big gifts to their alma maters — another important source of income for top schools. So of course these schools aren’t trying to educate the poor. What incentive do they have to do so?

But on a more fundamental level, how much does any of this really matter? How central are the Ivies and other elite private schools to our educational system in the U.S.? And how much would it change our country if they changed their admissions policies?

My answer: Not very.

Elite private schools schools are tiny

The total undergraduate enrollment at the eight Ivy League schools for the Class of 2024, according to Statista, is 16,648. Multiplying that by four and then multiplying by 1.5 to add in Stanford, Duke, MIT and Caltech gives us a rough estimate of about 100,000 undergraduates at top private schools in the U.S. That might sound like a lot, but it’s compared to 22 million total undergraduates in the country. In other words, top private schools are educating less than half a percent of Americans. The entire Harvard undergrad student body could fit into the University of Michigan football stadium more than 15 times over.

Source: Harvard University via Bloomberg

Now, this may be a very elite and important half a percent, but it’s still just a small sliver of each generation. Ask yourself if changing the composition of that half percent is going to change educational outcomes for the vast majority of Americans. The answer is “no”. And now realize that all of the compositional changes that people are calling for — more poor kids, fewer legacy admits, more diversity, less discrimination against Asians, and so on — are changes on the margin. Even if implemented, they would affect only a modest fraction of half a percent.

There is no chance that changing the composition of the Ivy League student body will effect anything remotely resembling a broad-based change in American educational inequality, opportunity, or aggregate outcomes. Zero. None.

Now, I do think that Ivies and pseudo-Ivies should let in a lot more students than they do. I also think they’re unlikely to do that, because the selectivity of their admissions allows them to pump up their prestige, allowing them to charge higher headline tuition to rich kids, to hire top administrators, and generally to think of themselves as elite fancy people. But even if Ivies did double their class sizes, it would mean they’d be educating 0.9% of Americans instead of 0.45%. It would still not represent a broad-based expansion of educational opportunity.

Elite private schools are unlikely to change your destiny

As Matt notes, students at Ivy League schools don’t obviously display a lot of social mobility. This is because most of them are rich kids, and most rich kids become rich adults. The Equality of Opportunity Project calculates social mobility scores for colleges, based on how many low-income kids they admit (“access” and how likely low-income kids are to become high-income adults after graduating from that school (“success rate”). The winners are mostly a bunch of state schools that educate working-class locals. No Ivy or pseudo-Ivy appears on the list.

Source: Equality of Opportunity Project

More than half of Stony Brook students from the bottom 20% of the income distribution make it to the top 20%! That’s amazing!

Now, maybe you think this methodology is bad, and they should have looked at other stuff, like graduation rates, indebtedness, time to graduation, and so on. U.S. News has its own ranking that takes some of these things into account, and their top performers are a bunch of public colleges and HBCUs that tend to educate a bunch of Hispanic and Black kids. CollegeNET has yet another ranking, and it’s dominated by Cal State and CUNY. No elite private schools are ranked highly on any of these lists.

Of course, when you look at college outcomes, you always have to be careful to separate correlation from causation. College does have substantial causal effects on income overall, but how much does it matter whether you go to a fancy school? Does getting into Harvard actually change your destiny, relative to going to a less elite university?

Probably not a heck of a lot. Twenty years ago, economists Stacy Berg Dale and Alan B. Krueger compared students who go to elite schools to those who are admitted to elite schools but choose to go elsewhere. They found that on average, it doesn’t make a difference to future earnings! They do find an effect from students from the poorest family backgrounds, but the income boost is around 7%. In 2011 the authors followed up with a paper that tracked students over the long term, and found the same thing. In 2018, economists Suqin Ge, Elliott Isaac and Amalia Miller did a similar paper with better data, and found no income boost for men (but they did find a fairly substantial effect for married women).

A 7% income boost for poor kids isn’t nothing. But it drives home how very little about American equality or opportunity would be altered by a marginal change in the composition of these tiny elite schools’ student bodies. And it also explains why even if the Ivies aren’t going to quadruple their enrollment, which they’re not going to do, it wouldn’t give a huge boost to U.S. educational quality — these schools aren’t in the business of providing a leg up out of the working class, they’re in the business of being fancy, and they’re in that business for themselves.

Yes, it would be good if Harvard et al. let in a lot more poor kids. But if we really want to boost opportunity for the mass of working-class Americans, we should be worrying more about expanding access to places like UC Riverside, SUNY Stony Brook, Cal State Fresno, and so on.

What is the point of elite colleges?

If it’s not broad-based educational uplift, what is the economic rationale for having a bunch of educational resources and top students concentrated in a few small elite schools? The rationale is matching. Matching means the country’s most talented people into the country’s most important jobs, so that they’ll do those jobs better than less talented people would have.

But how powerful is this matching effect? A third of Harvard’s Class of 2015 went into finance or consulting. Do we really think that making sure the next generation of bankers and management consultants are composed of the country’s best and brightest is a major economic priority? OK, maybe that’s unfair — maybe they only work those jobs for a few years, or maybe the other 2/3 of Harvardians go to the places where we really need the very top talent, like the CDC or Texas utility regulators or vaccine research.

But still, we need to question the importance of matching in our society. Tyler Cowen makes a good case in his book The Complacent Class that America has over-indexed on sorting and stratification in our society. There’s probably diminishing returns to putting all the top people in one room or on one team. It means ideas spread around less, and the top people have less of a chance to educate others. It’s like a gifted-and-talented program for all of society, but instead of one hour a day it’s forever, and instead of just extracting the nerds it extracts the rich kids too.

And there’s a strong possibility that our obsession with matching has negative side effects — when all the talented and rich people befriend and marry all the other talented and rich people, the rest of society is left to sort of fend for itself. What’s more, even the commentariat doesn’t seem to realize what a problem this is, possibly because so many of us went to those top schools too.

A creeping, toxic elitism

I don’t want to veer too far into cultural hand-waving here, but it seems like obsession with the Ivy League is a symptom of a creeping, toxic elitism that has permeated American society over the last four decades. We’re obsessed with high-status winner-take-all jobs. Our economy is dominated by superstar companies. The cult of Hollywood celebrities may have given way to the cult of Instagram influencers and YouTube stars, but it’s still all about the glittering few. Even the people who spend all day yelling about Elon Musk on Twitter are still spending all day…thinking about Elon Musk. It’s as if the inequality of income and wealth is mirrored in a general inequality of status, where only a few people and institutions matter and everyone else is left to watch the glitterati from the cheap seats.

I don’t know how we reverse this trend. I don’t know how the everyperson becomes central to our culture and our policy and our visions of our own lives again. Maybe redistribution will do the trick. But I think one small piece of it is for the commentariat (and that includes me) to focus less on the Harvards and Stanfords of the world, and more on the Cal State Long Beaches and the SUNY Stony Brooks. Already I like what the Biden administration is doing, focusing more resources on HBCUs and community colleges. Perhaps the old man is onto something.

---