Asymmetrical Information - Megan McArdle

09.10.12

Are We Paying Too Much for Higher Education?

College costs are rising. What are we getting for the money?

My newest piece for the magazine, on whether college remains a good investment, is now up on our website.

The short answer to that question is: yes, but . . . 

The average return to education remains very high--higher, indeed, than it was a couple of decades ago.  But that average includes a lot of variation.  Marginal students may not benefit as much--while the returns seem to be very positive, they can be very, very low.  For a student who spends a couple of semesters in school, runs up debt, and then drops out, they can be negative. And they’re low in part because of the rapid run-up in the cost of tuition that we’ve seen in recent decades.  The more you increase the up-front cost, the lower the net present value of your investment.

The other question is how much we as a society spend on college.  The cost goes well beyond tuition, to research subsidies (someone else will have to evaluate whether these are getting value for money), alumni donations, private scholarships, and of course, the subsidies that we give student loans.  Those subsidies come in a few ways:

1.  Federal guarantees of student loans  It’s not clear that the 10 year student loan for any major you choose is a product that would exist in the wild; certainly it wouldn’t exist at these interest rates.

2.  Federal subsidies of student loan interest payments  For some amount of student loans (it varies by need and where you are in school), the federal government pays the interest while you are in school.

3. Income-based repayment program: not clear what this is ultimately going to end up costing, but when the original law passed in 2007, the cost was estimated at about $3 billion a year: not huge in federal terms, but not nothing, either.  The program has since been amended to make it more generous, with a lower income cap and longer repayment term.

4.  Bankruptcy exemption for student loans.  It is theoretically possible to bankrupt your student loans, but darned if I’ve ever talked to anyone who managed.  In general, you have to be in pretty bad shape, because you have to convince a judge that you are not only unable to pay the thing right now, but also, unlikely to ever have any excess income that could be used to pay off your student loan.  This is a pretty high hurdle.

The taxpayers pay the first three subsidies; the last is paid by those who are unfortunate enough to default.  Unlike other forms of unsecured debt, it is very, very difficult to settle your student loans for much less than the principal plus interest.  If you made a bad educational bet, as you can imagine, this is problematic.

Overall, our investment in higher education has risen at a much faster clip than inflation.  As I note in the piece, some of this is undoubtedly Baumol’s Cost Disease:  NYU economist William Baumol has pointed out that while we’ve automated many professions, it still takes dozens of people to perform Beethoven’s Fifth: productivity in that sector is nearly stagnant.  But as other sectors get more productive, they pay higher wages, and the non-productivity-enhanced sectors also have to pay higher wages to keep up.  A factory that’s doing more with fewer workers can absorb the higher labor costs; a symphony that’s doing exactly the same amount of work with the same number of workers cannot.  So the price of concert tickets rises.  Professors, too, are not particularly scalable (though with MOOCs, maybe it’s just a matter of time.)

On the other hand, as economist Richard Vedder pointed out to me, most of the employees of a university are not professors: they’re secretaries and janitors and IT staff and administrators.  Why isn’t their productivity rising?

Vedder’s theory is that, as he put it, universities are raising tuition “because they can”.  With the decline of well-remunerated low-wage work, they have a valuable product to sell.  This is a bit recursive, of course, because as more people go to college, it’s easier for employers to demand diplomas for jobs that didn’t used to require one.  Which further bifurcates the market, and makes a diploma even more valuable.

It’s not necessarily a problem that we’re spending so much on education--and putting more people through school--if we’re actually adding value to the workforce.  In fact, if education is adding value to workers, then maybe it’s great: we’re upskilling our workforce, preparing them and our country for the 21st century.  That’s basically the thinking behind the president’s goal--now, I believe, part of the Democratic Party platform--of having the highest percentage of college graduates in the world in 2020.

But there’s a fly in the ointment, which is that higher education doesn’t only provide education: it also provides a credential.  This is known in economics as the “signalling model”: it’s hard to prove to employers that you’re intelligent, conscientious and persistent, and so you use a diploma to demonstrate it.

If you think signalling is a small component of degrees, then the extra investment in education is probably a good thing (although we might look askance at athletics and other non-curricular amenities, which have clearly been getting nicer and nicer over the years).  If you think that signalling is a large component of the value of a diploma, then the extra spending is worrying: we’re mostly just bidding up the cost of the credential, not investing in greater economic productivity.

There’s no slam-dunk data on how much of higher education is signal value, but some facts to consider:

1.  Colleges are increasingly competing to admit students with better stats like GPAs and standardized tests--and the stats of the incoming students make up a lot of the status of a school.  That can’t possibly be because the education is so great; it’s a pure signalling story.

2.  As both Bryan Caplan and Richard Vedder pointed out to me, students cheer when class is cancelled.  This makes no sense if the goal is accumulation of human capital.  In no other business are customers excited to get less than they were promised.

3.  Elite universities do not put their focus on hiring great teachers, monitoring teaching to ensure quality, or improving teaching; what they mostly care about is research.  If your teaching is really, really terrible, it may make a difference to whether you get tenure--though even then, Caplan argues that this will only be true if your research is marginal.  Professors are frequently wooed by promising to keep them as far as possible from students.  

It’s different at colleges that are explicitly teaching focused (though even there, research increasingly matters).  But there is no wage premium that I am aware of for going to an explicitly teaching focused school.

4.  Why do we worry about, and punish, cheating?  If school were mostly human capital acquisition, then cheating would literally just be cheating yourself.  There would be no need to punish it.

5.  It’s very easy, as Caplan points out, to avail yourself of an Ivy League education for free.  Go to an Ivy League school.  Sit in classes.  It’s quite unlikely that anyone will kick you out.  So why does almost no one do this? Because the one thing that sitting in Princeton classes will not give you is a degree from Princeton.

6.  You have probably forgotten almost everything you learned in college.  And yet the effect of forgetting those things, and the effect of never having attended those classes, is very different.

That doesn’t mean that higher education is all signalling.  For starters, in technical fields like engineering, nursing, and so forth, there’s clearly domain knowledge being acquired--although, the success of co-op programs at places like Northeastern suggests that at least some of that domain knowledge might be better acquired in the workplace.  And the gen-ed requirements indicate that it’s certainly not all about getting work-specific schools, even for engineers.

It’s also worth noting that there’s a wage premium for “some college” (though even that may have signal value).  There are non-academic gains--socialization, relationships.  There is the hard-to-quantify, but nonetheless real, value of “learning to think” . . . I entered business school a fairly typical English major, but by the time I left, I had changed the way I thought about data.  Instead of accepting figures and arguments, I had started to think, “Okay, what would make this not be true.”  This has proven invaluable.

James Heckman, the Nobel-prizewinning economist, also emphasizes the role of self-learning--even for the dropouts.  If you go to school, and find out it’s not for you, that’s valuable information.  And if you couldn’t have learned that information any other way, then maybe it’s worth the price.

Part of the difficulty in discussing this is that the people who engage in these debates--journalists and professors--are among the few professions that actually use the specific stuff they study in school.  I use what I learned in business school about accounting, economics and so forth almost every day.  Professors, too, are among the few people for whom undergraduate studies translate directly into work skills.

The other difficulty is, of course, that our economy is an increasingly hard place for people who do not have a college degree--and that the people who struggle most to get college degrees now tend to be poorer and have less social capital than the people who sail through four years at Harvard.  To question the value of a college degree sounds in some way like questioning whether those people shouldn’t just resign themselves to spending the rest of their lives down at Popeye’s, putting biscuits under the warming lamp.  You can’t just say, spend less on college, without offering the people who don’t go a better alternative.

Unfortunately, I don’t have a lot of easy answers to that.  There are colleges, notably Arizona State University, that are looking to do a better job by those marginal kids: to help them get the value out of college, rather than racing to improve their stats by denying admission to as many kids as possible.  There are apprenticeship programs, like Germany’s; I wouldn’t want to see us emulate Germany, which can be rigid and worsen outcomes for late-life career switchers, but I would like to see us work more on providing a work-based learning environment for kids who aren’t college bound.  Denmark manages to combine apprenticeships with support for late-life switchers, and while that’s expensive, and doesn’t always work (who pays your mortgage while you’re in training?), the same could be said about our current system.

Which is ultimately the point of the piece.  We’re spending a lot of money on higher education, without putting all that much thought into what we’re getting out of it.  Over the years, I've talked to a fair number of first-generation college kids who thought that simply getting any degree was a ticket to a good job.  I've also known a whole lot of parents who put themselves in some financial jeopardy to stretch for a school they couldn't really afford, because after all, it's education.

For a lot of kids the system works fine, but it’s definitely not working for everyone, and with budgets tightening everywhere, we need to think hard about how, and why, we invest in college.