questioning the higher education bubble

I was recently quoted in the Huffington Post in a story about Thiel’s claim that higher education is a bubble. I was quoted as one of the skeptics. Later that day, Thomas Gorkey, who teaches at Syracuse, read the article and wrote, I said that higher education wasn’t a bubble in the normal sense of the word because education is not a commodity that you can trade:

This doesn’t seem right to me, but I’d love for you to convince me that I’m wrong! To me it seems that students (and mainly parents) do in fact treat college as a tradable commodity. They trade thousands of dollars (that they don’t have) for an education that is supposed to land them a job that pays even more than what their education cost.

I think this deserves a response. In my response, I relied on a distinction made by the economist Gary Becker: physical capital vs. human capital. Physical capital is some physical object that allows you to produce more goods and services for sale on the market. Human capital is the skill that someone has that allows them to gain income from the market. Becker’s argument was that education resulted in higher income because schools trained people in various skills and thus create human capital.

Now let’s get back to the higher education bubble. When people say “bubble,” they usually mean “an unsustainable rise in prices,” which is often, but not always, due to speculation. The reason that prices have to eventually go down is that the underlying value hasn’t risen. The gap, as I mentioned, is often attributed to speculation. People consistently buy something hoping to quickly sell it off.

Notice that in this description of a bubble, some commodity is circulated and there’s a consistent error that pushes the price up and up each time the commodity changes hands. This describes the housing bubble well. People (including banks) started believing that housing prices would just go up and up, which then justified all kinds of crazy loans. However, college education is a form of human capital. The stuff you learned can’t be transferred to another person, no matter how much they pay. So it can’t be a bubble, at least they way most people talk about bubbles.

Thus, something else is happening in higher education. For a certain segment of the market, there is a huge over investment in human capital. As in some bubbles, people are making some sort of gross over estimation of how much some economic resource is worth. But the difference is that there’s no Ponzi scheme since you can’t sell or transfer your college degree to someone else for an even higher price. Human capital simply can’t be transferred. I agree with Thiel that something is very much wrong in higher education, but it’s not quite analogous to the housing or dot-com bubbles.

Let me end with a few hypotheses about why some people – mainly middle class people sending their kids to private colleges – are paying way to much for college. (a) Can’t say no. (b) They simply don’t understand that their kids won’t make the income to cover the loan. (c) Emulation of the wealthy. (d) Consumption good – they simply value the four years a lot. (e) Myopic preferences – people seriously underestimate long term costs.

Written by fabiorojas

May 15, 2011 at 12:57 am

Posted in economics, education, fabio

23 Responses

Subscribe to comments with RSS.

  1. @Fabio

    That’s a really fascinating way to approach this. I think you’re definitely right that what’s going on with higher education is not directly analogous to the housing or dot-com bubbles because of lack of trading.

    That said, I don’t think that makes it potentially any less of a bubble. People can still over-estimate the returns to higher education and end up burned (i.e. through student loan debt on degrees of limited value). This is a case of “a consistent error that pushes the price up and up” *without* commodities having to change hands.

    The bubble could also still “pop” if a large number of current or future students saw their older peers fail to recuperate investment, and thus decided against higher education. Shocks from decreased demand for higher education then might also reverberate through the economy just as in these other bubbles.

    It’s a good point that there isn’t a true Ponzi scheme here. However assuming this really is a bubble, those how got burned the most would still be those who got in the latest: those who paid for higher education exactly at the time when prices were highest and there were the most people obtaining such degrees to compete with them.

    To summarize, I don’t see why trading is:
    1) a necessary part of the definition of “bubble”
    2) a necessary condition for unsustainable increases in prices (bubbles)

    Not trying to sound too critical here, though I realize it may have come off that way. It’s a really interesting question what differences in price dynamics there are due to lack of trading.


    Charles Seguin

    May 15, 2011 at 3:30 am

  2. It seems like one parallel worth exploring, for its similarities and differences, is the role of government guarantees. As much as housing finance has been backed by the government, the programs and regulations on student loan debt are perhaps even crazier (from my limited understanding).

    Also, for a funny but sad take on the same topic, see this.


    Dan Hirschman

    May 15, 2011 at 4:52 am

  3. I disagree about the lack of trading in human capital. Obviously, one doesn’t sell human capital as it cannot be disassociated from the human being. But we trade the services of our human capital all the time — for salaries, wages, and other payments. I made investments of time, dollars, and foregone earnings to obtain a PhD. I trade access to the services of that human capital. I am now working at the fourth university that was willing to pay more than the incumbent “renter” of those services. But I am fortunate that I made these investments years ago. I have paid off all school debt, recouped the lost earnings, and earn significantly more than the second-best use of my time: driving a taxi. I am not sure a PhD student in her eighth year of a history PhD will ever be able to say this.

    I attended commencement today at my daughter’s alma mater. It was a B-school graduation ceremony. Many of the graduates have already agreed to sell the services from their human capital investment: hundreds of finance nerds who will do better than the last generation (or so their department head said) and hundreds of accountants who will skew the mean starting salary upward. Probably not much bubble effect in this group. But I am am less sanguine about the education price bubble in some other commencement ceremonies around campus.



    May 15, 2011 at 5:24 am

  4. “Let me end with a few hypotheses about why some people – mainly middle class people sending their kids to private colleges – are paying way to much for college. (a) Can’t say no. (b) They simply don’t understand that their kids won’t make the income to cover the loan. (c) Emulation of the wealthy. (d) Consumption good – they simply value the four years a lot. (e) Myopic preferences – people seriously underestimate long term costs.”

    I believe you are right. I also think being in some of the high schools has a life of it’s own. Suppose I went to nearby high schools to advise students about things I’ve done or studied. I suspect much of it would be a waste of time. I’d be saying that even top students are usually best off going to community colleges and then state colleges. This would require zero applications and zero standardized admissions tests during high school. This would be seen as shameful embarrassing loser advice. Or, somehow sad, like giving up, or revealing an insecurity or other psychological problem. I suspect many of them would not take seriously that it is the best combination of education and costs, including for the top students.



    May 15, 2011 at 7:47 am

  5. […] Questioning the higher ed bubble. […]


  6. “Human capital simply can’t be transferred.”

    Don’t we transfer human capitol every time we teach or train our competition?



    May 15, 2011 at 4:22 pm

  7. As a parent, I can see why some would consider the cost of not investing into their kids education. I am not an economist and I do not know how we ought to price regret, but saving in education costs can be a huge source of regret.
    Moreover, in the U.S. education seems also a lot like a luxury good (with the caveat of not having really lived there). Overpaying (supposedly) for your kids to go to Yale might be irrational in the same way it is irrational to overpay Bayerische Motoren Werke to experience the joy of driving when a Toyota would provide the same mobility. Education is not merely status for the kids but also status for the parents.



    May 15, 2011 at 5:01 pm

  8. It might be worth framing in terms of a counterfactual. If, as is occasionally suggested, we had equity stake student loans (i.e., I pay for your education in exchange for X% of your future earnings) then we would have fully tradable human capital. Such a market could lead to greater speculation.
    Currently if I’m a higher ed bull all I can do is go to college, which is rather time consuming and I can only really do once. However if I could buy an equity stake in your education then I could much more rapidly place much bigger bets. This would especially be the case if I could not only originate student loans but trade them or their derivatives on the secondary market.
    If it helps, imagine the difference between a housing market where the only way to own a house is to build it and you can only own one house, as compared to the housing market we actually had where you can buy multiple existing houses.



    May 15, 2011 at 5:52 pm

  9. “Human capital simply can’t be transferred.”

    “Don’t we transfer human capitol every time we teach or train our competition?”

    Human capital can be transferred, and done so quite easily in the digital age. It’s just not lost when we transfer it. Human capital is non-rivalrous. I don’t lose it when I give it to you, and vice versa. I sometimes wonder why as sociologists we don’t understand intellectual property better. It would seem right in our wheelhouse. (If that’s even an expression)



    May 15, 2011 at 11:00 pm

  10. I would like to thank professor Rojas for his thoughtful response to my e-mail. The response is not exactly re-assuring, but I find that I agree with the assessment of why middle class families still send their children to private schools even as those schools are really stealing their futures.

    Although the “education bubble” might not be a bubble in exactly the same way that the housing or bubbles were bubbles, I’m not sure that making the distinction between physical capital and human capital removes the way in which education IS similar to these other bubbles. Namely the price of education is rising in an unsustainable way, students are graduating with more and more debt, and the jobs just aren’t there or aren’t very good. I worry that soon (next 5-10 years) we will see massive default rates that bring the entire university system down.

    It also seems like we should be able to take proactive steps to avoid this doomsday scenario and that the only steps that should be deemed serious are the ones that dramatically bring the price of tuition down and dramatically reduce the amount of debt that students incur.

    Although it’s true that you can’t offload learned skills the way that you can offload a house, I’m still not sure there isn’t speculation of the normal sort going on here on the level of the debt itself. The banks bury us in debt and then they get rich trading our debt around (okay, now my lay-understanding is really showing. I’m an artist who is interested in sociology and economics but I have no formal training in either). Student debt is skyrocketing, and there is money to be made as the vile Todd Campbell writes:

    “Overall, student loan debt remains a growth business. This year, student loan debt surpassed credit card debt for the first time. As costs continue to rise and students and parents continue to look for payment solutions, SLM shareholders will be rewarded.”

    On this side of things it really does seem like a dumb investment, one that really is driven by speculation of the normal sort and really will create a bubble of the normal sort. Some people will make a killing if they get out on top, the rest of us as usual will get screwed.

    What is so frustrating is being caught in the middle of this process, watching half of my students get robbed blind and getting paid a pittance as a adjunct. And on top of that being constantly reminded that the university doesn’t actually value education anymore. One particularly cruel reminder was Syracuse University’s decision to invite JP Morgan CEO Jamie Dimon to give last year’s commencement address where he lectured a crop of newly indebted students and parents on–and I’m not making this up–the importance of “holding yourself accountable.”

    It just seems like the high price of tuition is unnecessary, only a fraction of it goes to pay for actual education. As someone who still believes in the value of education it’s very frustrating to see other values take precedence. I care about my students and I think they’re getting screwed. I advise them to transfer if they or their families are taking on more than $50,000 in debt to a school where they can learn what they want but take on something less that $50,000 of debt.

    But with the rising price of tuition our universities have ceased to be a place of economic mobility and instead have become a machine for sorting the population into even more rigid and unequal classes. My job turns my stomach.



    May 15, 2011 at 11:06 pm

  11. […] Have a Great Summer Fabio Rojas, a professor of sociology who knows quite a bit more than I do about how money and universities work, has made a generous response to an e-mail I sent him about the coming education bubble. You can read his response and the debate in the comments here. […]


  12. @Gabriel: Actually, turning “human capital” into tradeable securities along the lines you describe was imagined by Milton Friedman in Capitalism and Freedom, (1962) and more or less came true. The idea of income-linked loans makes perfect sense: an accounting or finance major is an AAA risk, because they are highly likely to make bank; a sociology major, a bit riskier. There was a website called which provided a way for investors to lend to college students and offered different rates depending on the expected value of their future earnings (and thus the prospects for repayment). It allegedly became the fourth-largest private provider of student loans in the US in 2007; it went bankrupt in February 2009. See
    But the most germane instances of education-as-bubble are not in sociology doctoral programs, or at Yale, but in for-profit institutions whose basic business model seems to entail tricking low-income students into enrolling in dubious education programs funded by federally-guaranteed student loans. You’d be surprised how many for-profit companies are providing online degrees in massage therapy and “medical technology,” and how many young people are becoming, in effect, indentured servants who will never repay their educational debt.


    Jerry Davis

    May 15, 2011 at 11:48 pm

  13. Austen,

    You seem to be confusing intellectual property (information, as contained in media) and human capital (skill, as embodied in human beings). These are related but conceptually distinct and have different properties. IP can help create HC, but only with the admixture of labor.

    Here’s a concrete illustration of how IP can be easily transferable but HC is not. I have in my possession a lot of IP related to the scripting language Python (manuals, software, etc). As it happens, these materials were cheap and often free to acquire. Yet there is no paradox in saying that I don’t know Python because I have yet to devote the enormous amount of work it would take to read these books, practice writing Python code, etc. That is, there’s a huge difference between giving someone information (IP) and teaching someone a skill (HC). There will never be a Napster for college education (this is true even if there is a robust circulation of PDFs of textbooks, MP3s of lectures, etc.)

    Alternately, maybe you’re not confused at all, but you’re just talking about Robin Hanson’s speculative projections about how in the future our digitally uploaded cyborg descendants will blur the line between HC and IP.



    May 15, 2011 at 11:49 pm

  14. Jerry,

    I knew about Friedman’s pitch which is why I said “as is occasionally suggested,” but I wasn’t aware that this was the business model of MyRichUncle. Fascinating, thanks.

    Also, I completely agree with you that the really big problem in education is not at the elite institutions but on the margins, which in practice mostly means for-profit schools relying on subsidized (and non-dischargeable) loans. College education seems to be a reasonably good deal no matter what major or institution if you graduate, but obviously isn’t for dropouts which means we should not be happy about (and as a policy matter, should not be subsidizing) institutions with 90% dropout rates.



    May 15, 2011 at 11:59 pm

  15. A couple thoughts:
    1) These discussions are primarily focused on a very small segment of students: those who attend high-priced schools and come from families well off enough to not qualify for financial aid but not well off enough to avoid loans and do not choose remunerative majors. At my institution, in-state students pay about $7,000 a year; most students end up with under $20,000 debt. If there is nothing bubblish about taking an auto loan for that amount, given the immediate depreciation of car values, I can’t imagine why there is anything wrong with taking an education loan for that amount either.

    2) One of the major differences between education and other goods is the high likelihood–at least for middle and upper class families–that the person footing the bill is not the person making the primary decisions. Frankly, I think the only other economic sector that works like this is health care–which clearly has similar problems, but for similar reasons it would be hard to imagine how it is a bubble.



    May 16, 2011 at 12:05 am

  16. Gabriel,
    There are other instances of securitizing human capital–my favorites are the Bowie Bonds of the late 1990s (bonds backed by the expected future royalties of David Bowie’s album sales, a fine model disrupted by Napster). I’ve also considered the manifest benefits in securitizing kidneys, although I’m not sure if organ donations count as “human capital.”
    On an unrelated note, I’m in your town for a talk Friday, in case you’re around:
    May 20, 2011 Jerry Davis University of Michigan
    “Corporations and economic inequality around the world: The paradox of hierarchy”
    Cornell Hall, D-310
    2:00 to 3:30pm


    Jerry Davis

    May 16, 2011 at 12:33 am

  17. Apparently, while most Americans think college is too expensive, most college graduates think it was worth it.



    May 16, 2011 at 4:10 pm

  18. Mikaila,

    Ummm, yeah, and people who got drafted by the NFL probably think it was worth it to devote their youth to practicing football, people who won the lottery think it was worth it to buy a ticket. and entrepreneurs with a successful IPO think it was worth it to start a business.

    I’d be much more interested in how people with “some college” (read: dropouts) feel about this issue. Whether or not college is a good deal for people who graduate (and I agree that it probably is), it certainly is not for the 45% of matriculants who drop out. This is a very large number of people and more importantly, especially characterizes people making the marginal decision to attend college or not. (My major problem with Thiel is that he is focusing on the right tail of students and institutions whereas the real issue is with marginal students taking remedial education at public schools and with the 16% graduation rate at the University of Phoenix). The survey apparently didn’t ask the “worth it” question to dropouts but on most of the other questions they resemble HS grads much more than they resemble BA holders. On the other hand, the “some college” folks not infrequently have a lot of student loan debt.

    Whether college is a good value on average is basically the arithmetic mean of the outcomes for graduates and some college people. I expect this mean would be positive. Whether college is a good average on the margin would have to be weighted towards the “some college” outcomes by a ratio of about 4:1. Such a calculation at the margin is almost certainly negative.



    sounds good.



    May 16, 2011 at 5:01 pm

  19. gabrielrossman–

    Yes indeed, I am deliberately blurring the line between human capital and intellectual property. For better or worse, I’ve become convinced the real growth in economic wealth both nowadays and the ‘future’ will come from better accounting of human capital as bottom-line property.

    If human capital were simply skill, it would be called human skill. It is called ‘capital’ because the skill is economically valued or is vaguely thought to possibly be so. A company with superior human capital should be valued as such, and it should show up on balance sheets as intellectual property.

    It already does in many cases, in the form of database ownership. A company that owns data is more valuable than one that doesn’t, all things being equal. We already know that. What I anticipate is that accounting will better take stock of which company actually has the skill and know-how (that is, the human capital) to take advantage of that data. And so yes, I suspect the line will get blurrier still.



    May 16, 2011 at 8:11 pm

  20. Austen, in your example of a company holding human capital (i.e., employing highly-skilled workers), human capital may indeed be transferable, as workers move from company to company. (Although then companies don’t typically “buy” the workers from one another. Slavery and sports (soccer) are exceptions). I guess this could lead to a bubble in human capital with companies over-investing in skilled workers (again, see the outrageous transfer sums payed for soccer players), but that’s still different from students over-investing in *education*.



    May 16, 2011 at 8:48 pm

  21. Gabriel, if the problem is with dropouts, than it is not a problem of too much loan debt. It is, rather, a problem of admitting students who are not prepared to succeed, or of policies that don’t help them stay enrolled and complete, or of remedial education that does not work. There is actually a lot of research and curriculum reform going on right now around remedial education that is showing promise in decreasing the number of semesters students stay in remedial courses, thus decreasing their likelihood of dropping out.

    In addition, I would point out that a good number of those dropouts come back eventually–I had at least 7 such students this semester. If you asked them the year they dropped out whether college was a good idea, they probably would have said it was not. But if you asked them THIS year, they would all say they are glad they got at least some of it out of the way initially, even if they wish they had worked harder or taken it seriously back then. Which reminds me–the other way to reduce the dropout rate would be to expand programs like Americorps and encourage some of the marginal or less academically-able students to wait until they are older and more mature before coming back to school.



    May 17, 2011 at 4:39 pm

  22. @Mikaila: It IS also about cost and debt. As a college education gets more expensive, it gets harder to obtain–even if the material gets easier (grade inflation rant; Arum & Roska). A student who can succeed with 3 classes, gets squeezed when she has to work 20+, 30+ hrs. a week. She can take fewer courses, but then it gets more expensive to finish. Time and again a student who is getting by with work and school work gets derailed by the expense of a car repair or, even worse, a family illness. This doesn’t mean there’s a problem with admissions.


    David S. Meyer

    May 17, 2011 at 5:01 pm

  23. […] of Higher Education to the New York Times. I highly recommend OrgTheory’s discussion of these topics as one entry point. *** This statement may be more than a little optimistic. No program is ever […]


Comments are closed.

%d bloggers like this: