I’m an optimist about higher education

I’m an optimist about higher education

Sam Altman from OpenAI recently did a thread about how “US college education is nearer to collapsing than it appears.”

I have no idea how close to collapsing people think American higher education is, so it’s difficult for me to offer a clear view on that. But I think the whole thread is emblematic of a tendency among successful technology executives to understate the resilience of the elite aspects of American college education — the sort of schools that they dropped out of, in other words.

Altman dropped out of Stanford. Patrick Collison dropped out of MIT. Mark Zuckerberg dropped out of Harvard. I’m very bullish on all those places. But by definition, most of American higher education is not elite. Even though those schools and others like them occupy a huge share of the discourse, they constitute a very small part of the overall system.

I do think it’s true that the larger, less elite parts of the system are going to go through a rough time over the next few years — fundamentally not for the kind of big-think political or ideological reasons that people like to yell about, but because of demographics and economics. The United States has a shrinking supply of teenagers and a growing demand for labor, so the least elite schools are going to be competing not only with each other for a dwindling supply of customers, but with employers who are increasingly willing to offer what will sound like a very good deal to people who are not that into school.

So expect tumult. But the image you should have in your head is an 18-year-old who’s not great at math or particularly into books going to work at Burger King rather than attending community college, not a high school valedictorian choosing to launch a new startup rather than go to Princeton.

We’re running out of teens

When I first went to Camp Winnebago in the summer of 1991, there were three ghost bunks — functional sleeping accommodations that were simply unoccupied due to low enrollment. By the summer of 1996, they were all full. And when I returned in the summer of 1999 as a counselor, they’d added bunks.

This was pretty typical of the trajectory of the United States of America. My birth year of 1981 was a bit of a demographic dip: there are more people a bit older than me (hence the empty bunks) and also more people a bit younger than me (hence the re-opened bunks). If you combine that growing population with a secular trend toward a larger share of teenagers enrolling in college, we ended up with a huge college enrollment boom in the mid-to-late aughts right after I graduated. Then we got the Great Recession and a sluggish recovery that created a situation in 2011 or 2012 where lots of young people saw higher education as a refuge from the bleak labor market.

But the number of people in the 18-24 youth bracket peaked back in 2013 and has been falling since then (the 2020 numbers go up a bit because of Census statistical revisions).

This has resulted in a fall in the number of new first-year students enrolling in college. As with so many other things, this data reports with an annoying lag, but as of 2019 there was no decline in the rate at which young Americans enroll in college. But the actual number of new first-year students has fallen considerably despite that steady rate because the population of young people is dropping.

This, much more than anyone’s hot takes, is the basic business model issue for higher education going forward. Even if it doesn’t lose any appeal at all, it is running out of customers. And that now intersects with a labor market that is much stronger at the lower end.

Youth employment is booming

One of the biggest economic changes between how things are today and how things were 10 years ago is that right now it’s really easy to get a job.

It’s almost never the case that a person without relevant skills or experience can get a really lucrative, desirable, or glamorous job. But today you could definitely get a job. Somebody will hire you, teach you what they want you to do, and offer you hours as long as you are willing to show up on time and do what you’re supposed to do. That’s not really asking so much of the labor market. But it’s not something that the labor market was providing in 2012. And while it doesn’t necessarily make a huge difference to the median worker who is older and has some experience in a specific field, this is transformative from the perspective of people who are at the margins of the labor market.

That includes teenagers, where — as you can see — employment levels have continued rising from the recession low even though the number of teens has dropped.

The Atlanta Fed’s wage growth tracker also shows that nominal wages have been rising especially quickly for young people and low-wage workers.

So in addition to facing a dwindling pool of customers, higher education faces increasingly robust competition from the alternative of just getting paid to work. But while the discourse tends to care a lot about things like kids with a 1600 on their SATs doing the Thiel Fellowship instead of getting woke brainwashing at Yale, the typical case is probably more like dropping out of Generic Cardinal Direction State in order to get a bonus and shift from part-time to full-time hours at Walmart.

And of course, as the tech people themselves could tell you when they have their heads on straight, that’s how disruption works — it starts at the bottom end of the market with the customers who are worst served by the product. The biggest student loan issues aren’t with people who attended the most expensive schools; it’s with people who didn’t finish their degrees or who enrolled in programs that don’t garner much respect and thus end up not enjoying much of a college earnings premium. The fact that it’s now pretty easy for just about anyone to get an entry-level job while more companies are dropping formal degree requirements means those marginal students now have more options. This is driving some schools out of business or into mergers, and I think constitutes a real source of pressure on the institution. But rather than causing college to collapse, I think American higher education as a social practice basically emerges stronger.

A more competitive higher ed sector

For the first time in a long time, college tuition is growing more slowly than average inflation.

That’s largely because average inflation is growing so fast. But nominal wages are also growing fast, and that’s the important thing here. The ratio of gasoline prices to nominal wages is getting worse, but tuition costs are growing more slowly than wages. Aggregate student debt is shrinking as enrollment falls, tuition cost growth slows, and inflation rises.

You could imagine a situation where this causes a downward spiral for colleges as a whole. But I don’t think that’s what we’re looking at here.

Because the thing about college is a lot of people like it! Not just because college is fun and has parties and stuff, but because a lot of people learn a lot of interesting things at college. I think I’m pretty decent at learning things on my own. But I took plenty of classes in college that involved close readings of dense texts or detailed technical matters, and I’m very bad at studying that kind of thing independently. Without the formal structure of a technical logic class, I’d never have been able to get through the proof of the diagonal lemma or all the technical aspects of Tarski’s undefinability theorem or Gödel’s incompleteness theorem. Some people absolutely could do that. But I couldn’t.

The discipline of formal learning helped a lot with that. And with reading “Anna Karenina.” And with understanding David Lewis’ arguments about causation and possibility.

And even though I generally don’t use this information in my day-to-day work, it actually does come up from time to time. But more to the point, doing the work of struggling through it taught me things about how to learn and how to know what I don’t know.

The nice thing about a world in which colleges need to compete harder for the patronage of a dwindling population of 18-year-olds is that the schools themselves need to try to raise their game. And with the labor market options for the less academically inclined improving, that means schools have an opportunity to focus on providing a good value to people who really do want a college education. That’s what college should be, not a rent-seeking opportunity exploiting economic fears. Rather than collapsing, I think good programs are going to shine while weak ones struggle.

The student visa opportunity

With domestic demand for college declining, it’s as good a time as ever to rethink the extent to which the United States has squandered foreign demand for American college education.

Student visas got less popular under the Trump administration, which kept looking for ways to make them more burdensome and harder to get. The smart strategy should be just the reverse. If you can pull an 1100 SAT score and are willing to pay tuition, it should be incredibly easy to get a visa to study in the United States. And if you complete a reasonable program and someone wants to hire you, it should be easy to stay and work.

American colleges and universities, despite the haters, dominate global rankings and are a strong American export industry. Some secular trends in demographics and the labor market are going to hurt the worst actors in higher education, but that’s only going to make things better. We should let the customers buy the product!