DIY U: Is There a Bubble in the Higher Education Market?

Update: Coincidentally, Anya Kamenetz just posted an interview in which she briefly compares the student loan situation with the real estate bubble here.

When I opened my email the other day, I was greeted by the following blurb from Today’s Campus:

The U.S. college-buying public numbers in the tens of millions.  Many of them are first-time purchasers.  Most of them are not.   Among a marketplace of that size, one ‘buying cluster’ could supply enough students to fill several campuses.  One very interesting cluster we’ve identified is husbands purchasing higher education for their wives.  Click here to order a free sample report with our compliments.

We don’t often hear higher education talked about in such baldly commercial terms. Most faculty and campus administrators have visceral negative reactions to this kind of language. But that doesn’t mean that colleges—even not-for-profit colleges—aren’t essentially run as businesses. It just means that (a) they’re often run as businesses badly because it is considered distasteful to draw on the lessons and vocabulary of the business world, and (b) educators often fail to identify the tensions between their school’s mission and its operating model, thus making it less likely that they will be able to confront some of the system’s ethical challenges successfully. But since the heart of DIY U is an economic argument, it is worthwhile to try framing issues in the language of economics. As the title of this post suggests, the specific question I want to address is whether there is a bubble in the college education market.

An economic bubble is what happens when the price of an asset class rises dramatically and well out of proportion with the rise in its intrinsic value. So the first question you ask when looking for a bubble is whether there has been an outrageous price spike. Has there been an alarming spike in the price of a college degree?

Here’s a graph of the cost of a four-year degree at Rensselaer Polytechnic Institute (RPI) from 1950 to 2009:

Yikes. I would certainly call that alarming. But maybe RPI is an anomaly. A College Board report I found has a chart of the changes in cost of degrees at four-year colleges from 1976-77 to 2006-07. Let’s see what that looks like in the form of a graph:

If anything that looks even worse.

But is that rise out of proportion with the intrinsic value of a college education? That’s a harder question to answer. According to a 2007 study by the College Board, an undergraduate degree makes a pretty substantial difference in annual income:

College looks like a good deal in this graph. Here’s a graph of expected lifetime earnings from the same report:

Again, it looks pretty good so far.

But wait. What if we factor in college debt? Here’s what happens:

What you want to look at here is where the purple and blue lines (representing the net income after college debt of people with two-year and four-year degrees, respectively) cross the green line (representing the baseline income of people with high school educations). It turns out that the average Associates Degree—if you go to school immediately upon high school graduation and earn it in two years, which are both dubious assumptions—will start paying off net of debt (on average) when you turn 29, or after nine years of full-time work. If you get a four-year bachelor’s degree—again, assuming that you go right to college after high school and complete it in four years—it will take you on average until you are 34 years old, or 14 years of full-time work after graduation.

In assessing the degree to which there may be a gap between the cost of a college degree and its value, a lot depends on what you assume about the people doing the evaluation. If we were all strictly rational long-term-income-maximizing automata, a college degree would still look like a good deal on average. After all, you make substantially more by the end of your life. Let’s call this the efficient market theory of college pricing. But as we have recently discovered much to our pain, markets aren’t always efficient. A more reasonable theory might be that most college students and their parents simply don’t do this calculation. They take it on faith, with perhaps a sprinkling of anecdote and experience of generations past, that having a college degree is worth the investment prima facie. This evaluation would be helped, of course, by the fact that there can be very good reasons to go to college other than pure economic pay-back. But at the very least, I suspect that the number of people who would be willing to get a degree knowing that it may not pay off for quite some time would be significantly smaller than the number of people who are getting their degrees right now.

The other thing to consider here is that we’re talking about averages. As my father likes to say, if you stick your head in the oven and your feet in the freezer, on average you’ll be comfortable. I’d like to see what the payback period is for students going to the bottom two quintiles of colleges. How big is the payback gap between a Harvard graduate and a Whatsamatta U graduate? My guess is that it’s big enough to change the look of these graphs entirely. For a significant number of graduates from the bottom quintile of schools—especially those that started late and/or had to go to school part time while working—it’s possible that the degree may never pay for itself. Again, this is just a guess, but I would like to see that data.

This situation is likely to get worse rather than better. College costs continue to rise dramatically while wages stagnate. Take a look a this graph of education-based differences in income from 1971 to 2005:

Look at the graph for males, on the right side. Income for college graduates actually declined in real dollars from 1971 to 2005. The reason main reason why a college education “pays” is because income for non-college-educated males declined a lot more. Unless either real wages go up or college costs stop climbing through the roof, the College Board may soon have to publish a report entitled “College Doesn’t Pay.”

So here’s what we know:

• The cost of a college education has climbed at a rate that raises the question of whether there is an economic bubble.
• The cost of education, on average, is lower than its lifetime payback, but the gap is probably a lot less than many students and parents realize.
• It is entirely possible that, for some significant portion of the college-going population, costs have already exceeded pay-back.
• Costs continue to rise rapidly while payback is falling in absolute real dollar terms and is mainly considered payback at all because the income of the non-college-educated is falling in absolute real dollar terms at an even faster rate.

It sure feels like it could be a bubble. Let’s add the fact that these tuition rises are financed primarily through credit. Larger and larger portions of student education are financed on debt as easy credit feeds an arms race among colleges where one of the two primary markers of “quality” in the market is cost. I’m not going to lay out the details here; if you need to be convinced, then read chapter 3 of DIY U. Kamenetz does a much better job of making the case than I could in the space of a blog post. Suffice it to say that we have too much money chasing after too few assets, which is a classic cause of economic bubbles.

I believe thinking about the college situation in terms of an economic bubble can be useful because it gives us a model to think through implications such as the following:

• Bubbles burst. Always. If what we are experiencing is a bubble in the higher education market, then we can expect a dramatic collapse in the price of a degree at some point.
• The two most likely triggers of the bubble bursting would be either credit drying up or demand dropping off—and these tend to go together. Credit could dry up if the number of students defaulting on loans increases, which is inevitable if the cost of education relative to real incomes continues to rise as it has. Demand could drop off if a significant number of prospective students discover lower-cost paths to the same economic track, i.e., if significant numbers of students as rational investors in their own futures find alternative investments that have better cost/benefit ratios than college and the debt that it brings with it. This is exactly what Kamenetz is calling for.
• If there is a bubble and it bursts, it wouldn’t necessarily mean the end of higher education as we know it, but it would probably mean large and painful contractions in college budgets leading to layoffs, cuts in services and the closing of a significant number of colleges.
• Students in the “sub-prime” position who are taking on large amounts of high-interest debt to get educations from low-performing schools are likely to be hurt worst as those schools crumple under the hardships caused by the bubble bursting. And, unlike your home, your education is a fundamentally illiquid asset. You can’t sell off your diploma, even at a loss, to pay back your bank loans. These students will be screwed six ways from Sunday, which is even more ways than they’re getting screwed now.

Scary.