Buying Equity in College Students

Oregon legislators are proposing to let students in the state attend college for free, in exchange for a share of their future income.  Longtime twitter correspondent Professor Tony Lima explains why this won’t work:

Following Prof. Akerlof’s lead, assume there are two types of students.  One type (S) majors in art, English composition, history, and ethnic studies.  The other (H) majors in mathematics, hard science, engineering, or even economics.  While type S individuals may not know it, their major will, on average, result in lower lifetime income than those in group H.  Type S individuals will happily accept Oregon’s offer since three percent of their income over 20 years is a good deal.  Type H individuals, however, are likely to think that three percent of their income is a high price to pay.  They will seek alternative methods of financing their education, paying the standard tuition and fees.  (This proposal could, in fact, revive the private student loan market — but without government intervention.)

Result: Oregon will collect far less than they are predicting.  The percentage of income will rise and the duration of the loan will also increase — to 25 years, then 30 years.  And, just as Prof. Akerlof predicted decades ago, the scheme will eventually collapse.

There’s a related problem: once people have gotten the education, they have a strong incentive to structure their lives around lower compensation (which the state gets a cut of) and more of other enjoyable things.  Take the job that pays less, but offers fewer hours, longer vacations, or more attractive office amenities.  Do something personally rewarding but financially unremunerative.  Note: there’s nothing wrong with those choices; I myself chose to use my MBA to become a journalist.  But I still repaid my (gigantic) student loans.  You don’t want to give people incentives to follow their bliss on the taxpayer dime.

Overall, this scheme seems very unlikely to work, unless the annual payment is so low that it won’t affect student behavior–in which case, it will also be so low that it won’t cover the cost of the education.  In today’s fiscal environment, I don’t see that this has great prospects.

61 thoughts on “Buying Equity in College Students

  1. I think this assumes a far more rational being than most 18 year olds are. If x% is low enough, everybody will do it, if it is too high, no one will do it, and I’m guessing the range of x is pretty tight, regardless of the major. 3 to 5 % sounds trivial to most people. 10% does not.

    But what do I know, I got a Master’s degree in Physics, then ended up writing software, which pays pretty well. I didn’t really plan any of this.

    • One must wonder, too, about the tax implications. My bet is that this is after-tax money. So the 3-5% might become 6-10% if the family is doing well. Lest we forget, this is just the tuition component. Most people who leave school with $100K student loans have paid for a four-year lifestyle, not just tuition. The fancy apartments, dining out like a retiree, and lavish vacations will continue to load up the student-loan profile.

  2. Another potential outcome I could see: taking a career path where financial rewards are heavily weighted towards people who are later in their careers (e.g. jobs with low starting salaries, but built-in raises and generous pensions). There’s a lot of earning potential past age 42

    • You write “a career path where financial rewards are heavily weighted towards people who are later in their careers”.

      Public school teaching is much like that. I have a suspcion that civil service jobs have that feature also.

    • I suspect that the 20 year cutoff was decided upon by careful examination of income levels of union workers.

  3. Akerlof showed that, in some cases, you get such an extreme separating equilibrium that the market collapses, but that’s not always the result. He was explaining why a relatively new used car sells at such a large discount – it’s unusually likely to be a lemon – but there are partial pooling equilibria where those with good used cars are willing to take the pooled, or average, price that incorporates the probability that the car is a lemon. Those with the highest expected earnings will separate out and borrow in another way if the pooled price is too far off, but that doesn’t necessarily mean that the scheme will collapse.

    I’m not saying, though, that this is a good idea, only that Akerlof’s adverse selection model doesn’t automatically predict a death spiral. One huge difference between a market and this proposal is that, as far as I can tell, the “price” (percentage of income) will be set by the government, which means that Oregon could screw it up in either direction. A market pricing scheme for student loans would give students good feedback on the general cost-benefit tradeoff of their majors, perhaps inducing some of those that are capable of it to switch into a major that will pay off better, later on. But both our current student loan system and this proposal are set up to shelter students from market information and pressure.

    Megan is pointing out that there’s also a principal-agent problem with the percentage approach to paying back student loans, similar to the landowner/farmer problem. The best way for the landowner to induce hard work from the farmer is to rent the land out at a fixed price, so that the farmer keeps all of the excess. But from what I know about Oregon’s politics, they probably like the idea of certain types of choices being subsidized by others, even if it’s less efficient.

  4. McMegan said a while ago:

    It’s hardly surprising that colleges began to claim more and more of the surplus created by their college degree. Think about it this way: if colleges create an extra million in lifetime salary, you’re theoretically better off if you pay them the discounted present value of $999,999 in order to earn that extra million.

    College will love this new opportunity at price discrimination. Sure, right now it’s only 3% of gross wages. But eventually it can be tuned for the STEM graduates to be worth 90% of how much “extra” they are unfairly earning over high-school graduates.

    • OTOH, such price discrimination is reduced if there is easy entry for new colleges, or if there is a sudden undersupply of students. (Given the upcoming decline in the 18 y.o. cohort and the possibility of a B.A. (college) bubble, this may happen.)

  5. If it’s such a good idea, then make it a market. Bundle students together into “funds”, kind of like mortgage backed securities or CDO’s. Investors buy these securities, the state gets the money up front, then pays the investors back over time from the “tax” they collect on the students.

    This way, the state is more likely to get the “price” (the percentage of income and duration) right, and students would be exposed prior to beginning their studies to the relative expected value of a humanities degree vs. a STEM degree.

    • What a great idea! My lovely (and very smart) wife comments that “there will be tranches of students.” The only problem is that states have a nasty habit of defaulting, violating contracts, and so on.

      But I’m ready to set up the derivatives markets (options, futures, etc.) when you get the CDO packages ready!

    • It’s so ridiculous an idea that I’m afraid it would actually be implemented, along with insurance plans (underwriten by AIG of course), wage discrimination, etc. Rule of Unintended Consequences, here we come.

    • Yeah, Mike, I forgot the first rule of dealing with politicians: never mention a bad idea in public because one of them will think it’s terrific.

      BTW, this is the most fun I’ve had on a blog in quite a while. Thanks!

    • Tony — What a great idea! My lovely (and very smart) wife comments that “there will be tranches of students.”

      Too late. There are already tranches of students in full default every Thursday, Friday and Saturday evening all around the country.

      Oh, wait…

  6. The old Daedalus column in Nature once proposed something like this. IIRC, the idea was that students would sell a fraction of their future income in exchange for a lump sum to go towards their tutition. Thus, the higher your expected future income, the smaller the fraction of it you would have to sell to pay for $x worth of degree.

    Question: How far would this tweak go towards fixing all these problems? Far enough to make it workable?

  7. Why can’t they base it on expected future earnings? I assume Oregon has an income tax and the SS numbers and majors of students over the past x number of years. Run a report comparing the two and arrive and the percentage required per major.

    • Ah, but if they price it this way, it wouldn’t be a subsidy for certain majors (art history, how-to-be-a-victim studies, etc.) at the expense of others.

  8. I would add there is nothing to stop Oregon for requiring this form of financing, as I believe Australia has done. I’m sure you’ll agree that it’s difficult to have an adverse selection problem if only one option is provided.

    • Tough to enforce a law like that because you’d have to outlaw private lending. I was talking about loans other than “student loans.” In the U.S. “student loans” are government sponsored and controlled at this point. But somebody getting an MBA from Stanford might well be able to get a loan in the regular loan market.

    • Also there are 49 other states, plus a district and several territories, offering comparable education to (expected) future high earners without the 3% string. I’m sure you’ll agree that it’s difficult to behave like a monopolist when you don’t have monopoly power.

    • There are also 49 other states, plus a district and several territories and Canada, providing comparable education that can be financed in other ways. There’s nothing to keep Oregon students who expect to be high earners in Oregon colleges if the terms turn out too onerous. I’m sure you’ll agree that it’s difficult to behave like a monopolist when you don’t have a monopoly.

    • Harris,

      I think it would all hinge on the risk aversion of the median STEM graduate. In my experience, they tend to be quite risk averse.

  9. Tony,

    Presumably the Oregon legislature can require its public universities to only accept students who opt for the % of future income option.

    As you know this proposal only applies to public colleges and universities in Oregon.

    • Well, given the goofiness of laws passed all over the country, this would not surprise me. In which case, I predict an outflow of students in the STEM and other highly paid professions. Naturally, that would have less of an impact — it’s much more costly to move than to find a lender — but some students will vote with their feet.

  10. I predict an outflow of students in the STEM and other highly paid professions

    That assumes Oregon doesn’t base repayment rate on eventual major.

    Also, to what degree are STEM majors risk averse? 3% of what ever you end up making is a lot less risky than $500/month that doesn’t care that you graduated into a recession.

    • Risk aversion or not, those expecting the highest earnings would be the most likely to leave. Future lawyers with any reason to believe they’ll work in a top 100 firm would leave first. Future doctors, almost certainly. Anyone in the energy engineering fields. Whatever the effect is (an empirical question that Oregon can help resolve), it’ll start at the top with the highest (expected) earners.

  11. Future doctors, almost certainly

    Half of med students are female, you still willing to make the same bet?

    • Women would tend to value the ability to ratchet down their careers when their children are younger. If you have $1,500/month in student loans that option is far less viable.

      As for people in tech, having $100k in loans is going to push you towards a high paying job in a large company vs. the freedom to join a startup that pays you mostly in stock (that may or may not pay off).

    • bronxcobra: Startups (at least in the Valley these days) pay competitively in addition to stock… they just expect you to work *much* more.

  12. Tony,

    In a sense you need to think of these students being entrepreneurs. Do they want to fund their business by taking on debt, or do they want to offer an equity stake? What are the trade-offs?

    • Absolutely terrific analogy. That means, of course, that the state of Oregon is going into the venture capital business. With your permission I’d like to add your comment (and my reply, of course) to my blog. I’ll probably tweet them, too. If you have a twitter handle, facebook page, or anyplace else you’d like to be credited, let me know.

    • the state of Oregon is going into the venture capital business.

      Sounds better than the current system of government loans being used to finance so much of higher education.

  13. In surprised that Megan’s libertarian leaning commentariat isn’t more in favor of a transition away from debt financing of higher education toward equity based financing.

    It would certainly do a lot to align the incentives on the universities with those of the students.

    Note: my proposal would have have government offer bridge financing to allow schools to transition toward a system in which they must rely on x amount of their students’ future earnings rather than tuition.

    • I’m not opposed to the idea – I’d definitely permit it. I’m just skeptical that an arrangement of that kind can work in the long run, without some rather draconian government regulation (like the Australian rule mandating equity-only financing, plus a prohibition on differential pricing among majors). As an investor (like the state/school, which invests the student’s upfront cost to earn future equity payments), I’d be terrified of selection effects: the people who are most likely to choose low-paying jobs with good lifestyle & pension benefits (say teachers) are those most likely to take equity financing. Oil industry engineers, seismologists, and chemists, who earn six digits early and for a while, on the other hand, would prefer a fixed price that lets them keep all the excess earnings. (If, contra my previous assumption, you allow differential pricing, you remove some of this risk, with the side effect of guiding people toward higher paying jobs.)

    • contra my previous assumption, you allow differential pricing, you remove some of this risk, with the side effect of guiding people toward higher paying jobs.)

      I totally agree. A UT chemical engineer (with a minir in geology) might have to pay 0.5% and a theater major at NYU might have to pay 25% – of future earnings.

    • with the side effect of guiding people toward higher paying jobs

      Or, alternatively, NYU would be pressured to find a way to deliver a theater education for somewhat less than the current ~$60k a year.

    • “Or, alternatively, NYU would be pressured to find a way to deliver a theater education for somewhat less than the current ~$60k a year.”

      If there were a like button here, I would use it on this comment.

    • This is a bit curious, since at the “baby-Ivy” I attended, there were many, many art & literature majors who went into bank management training programs, advertising, real estate development and media and who are now making very high salaries. They aren’t sitting around writing poetry in garrets. I also know a substantial amount of non-business majors who do analysis/marketing for private equity and hedge funds.

      I’m wondering how badly a literature degree from a top Ivy League school will actually hurt you; I think its when the lit degree is from Cal State Humbolt County that you’re taking a big risk.

    • The University I’m at has a fairly well-ranked theater school, and theater is one of the most expensive undergraduate majors to deliver, presumably because of small classes and lots of personal attention from faculty. Theater majors are heavily subsidized by various other areas – the tuition is an even worse measure of actual cost in this case than in most cases.

  14. One thing I almost never see come up in these discussions is the role of bankruptcy law. To my knowledge, federal bankruptcy law does not envision a debt wherein the debt constitutes a percentage of a person’s income. One can own a percent of the income produced by a business, but that is an ownership interest in the business. Taking an ownership interest in human beings, unlike businesses, is unlawful under the 13th amendment.

    The arrangement is different from Income Based Repayment programs in that with IBR, there is still a loan balance and interest rate, and if the balance is paid in full then the loan is satisfied permanently. With this, you will pay the percentage for the given period of time, no matter how much or how little you pay.

    Given that, I have a few questions for someone with better bankruptcy law knowledge than me:

    1. Is it possible for such an arrangement to be considered a student loan under the bankruptcy act, since it lacks key features of a loan including a principal balance or interest?

    2a. Is such an arrangement a lawful contract at all under the 13th amendment or other relevant contract law.

    2b. Does it matter for 2a this that the arrangement is being made by a state government with plenary taxation power over its citizens?

    2c. Does a student moving out of state (and thus ceasing to be subject to the tax power of the state where they went to school) change 2b?

    • In order:

      1. Not as loans are defined currently.

      2a. It’s lawful, as it doesn’t imply ownership of a person. It’s more akin to tithing or support arrangements, which are perfectly legal.

      2b. Not at all. It’s a contract, not a tax per se. Taxes have to clear different constitutional hurdles than contracts.

      2c. No, because it’s a contractual arrangement rather than an outright tax.

    • I was hoping someone would mention the 13th amendment. My question is simple: when does a long-term contract become long enough that it is equivalent to slavery? Remember, you’re not allowed to sell yourself into slavery (creating a major defect in the market for human capital). Any constitutional scholars out there that can cite relevant case law?

  15. OTOH, this scheme also places an incentive on Oregon public universities to increase the fraction of student places that are in majors with high income potential.

  16. From what I’ve gathered in reading the articles (including links) and comments that the way this works is:
    1. The plan is independent on any prior or subsequent schools you go to. If you go to an Oregon school for any time under this program, you are subject to it based upon what tuition the State paid.
    2. The percentage is straight and not discrimatory, i.e. not tied to any particular public school, major, etc. in Oregon.
    3. The loans CAN be discharged in bankruptcy court unlike the current student loan program. If that’s the case, you run into the classic free rider problem. If it’s not the case, I don’t see much of a difference between the current program and this one other than fact that loans do not consider your employment status for scheduled payments while the new plan simply takes a percentage only when you are employed.

    Fundamentally, this program, as well intended as it is, suffers from being too susceptible to gaming the system. Other than what Megan mentioned there is also incentive to simply not report your earnings, or underreport your true earnings esp if you move out of State to work. There’s just too much temptation to shirk your payments or not pay your full amount. Unlike a fixed loan amount where you can show someone isn’t making payments, you have to prove that someone is not making as much as they report. In order to track people down or audit people to make sure they are being honest, this would require a lot of effort and expenditure esp since it’s not a nationwide system but just a one for a single State. And then the system really isn’t paying for itself, and punishes those who are honest.

    • In order to track people down or audit people to make sure they are being honest

      As the IRS computers are already sending all W-2, 1099 etc and tax return information to the State of Oregon, what extra do you expect has to be done?

    • But how would the employer know to even send a W-2 (or similar) to Oregon unless the employer somehows knows this? What if the employee doesn’t tell them, or what if they tell them they went there but lie and say they did not accept the full ride program? And how is Oregon/the school going to know that the person is working and deserves part of their cut? They certainly aren’t going to follow every single graduate throughout their working lives.

      I simply see this as something which is either too easy to evade, or too cost-effective to enforce.

    • You write, “Other than what Megan mentioned there is also incentive to simply not report your earnings, or underreport your true earnings esp if you move out of State to work. There’s just too much temptation to shirk your payments or not pay your full amount.”

      There is a data-sharing deal between the Oregon tax authorities and the IRS. I assume that Oregon can just request the needed information about the SSNs of the students who owe under these arrangements. It wouldn’t be much different from whatever system is used to enforce child-support payments against people who leave the state.

  17. Re slavery, a good resource seems to be . On the one hand there are several legal-style definitions:

    ‘The Slavery Convention (article 1.1) in 1926 defined slavery as
    “…the status or condition of a person over whom any or all of the powers attaching to the right of ownership are exercised….” ‘
    ‘The 1926 Convention’s definition of slavery was broadened to include forced or compulsory labor in 1930 in the ILO Convention (No. 29) concerning Forced or Compulsory Labour (article 2.1):
    “…all work or service which is exacted from any person under the menace of any penalty and for which the said person has not offered himself voluntarily.” ‘

    But in the discussion that follows, slavery includes, “1) The practices and institutions of debt bondage: the status or condition arising from a pledge by a debtor of his personal services or of those of a person under his control as security for a debt, if the value of those services as reasonably assessed is not applied towards the liquidation of the debt or the length and nature of those services are not respectively limited and defined.’

    Might Oregon’s proposed law amount to debt bondage?

  18. I don’t think that a 3% threshold is going to be much of a motivating factor to lead people to the underground economy to lower their taxable income or for whole careers sectors to leave the state. Under this theory they would be far more motivated to avoid their actual taxes, Social Security or even Medicare. Since the money would likely be removed through payroll deduction, the average person would just adjust and continue to live their lives. I suspect that most people would not think about it very much.

    In my opinion, it is a far more rational way to fund the system. It allows for people to get an education with far less risk. This will provide for overall increased economic growth by essentially giving a fixed cost for education rather than the variable system we have know. By spreading the cost more evenly and fixing it at an affordable level, nearly every other sector of the economy will be helped. A 25 year old making 50K/year will pay $1500 dollars per year. That leaves income to buy cars, houses, washing machines, clothes etc. Whereas paying $500 to $600/month for student loans leads them back home to live with their parents. We need education to be inexpensive because it is a driver for the economy at large.

    • You are ignoring my original point. People don’t have to go into the underground economy to avoid this tax. Instead those who believe — correctly or incorrectly — that their lifetime earnings will be high have incentives to take actions to avoid the 3 percent tax. While possible actions include the underground economy, there are less drastic ways of avoiding the tax. I suggested private lending (say from a bank). Others have offered various alternatives. And I continue to predict that when the proposal becomes law it will raise less revenue than predicted. (Economists have a nasty habit of being correct about predictions like this.)

    • Again you are assuming that someone starting college is sitting down and making longterm projections about their future income and doing a cost benefit analysis. That very rarely happens . On top of that, most projections would be far from reality. People are not logic machines acting in their self-interest. Decision making is driven by logic, emotions, bad/good assumptions, impulse and other factors.

      I also reject that economists are good at predicting much of anything. Economics as a social science is better than nothing, but not by much. Case in point, very few economists could even predict the housing crash. While common sense tells you that prices cannot never rise forever and that they cannot long out strip income (since eventually fewer and fewer people can afford to buy).

      The idea of super rational markets and market actors exists only among the Randian market utopians. It is neither reflective of reality nor a good way to manage an economy. I will now step down off of my soap box.

    • Many economists did predict the housing crash just as we predicted that the stock market boom of the late 1990s was not sustainable. But those are macro issues which are always harder to predict. On the micro level (which is where this policy lies) we’re pretty good. I’ll offer myself as an example. Over three years ago I predicted that unemployment would remain high as long as President Obama was in office. You can read it here: You might also take a look at and That last one is pretty heavy going. You might skip the math and just jump to the conclusions. Let me leave you with a quotation from John Maynard Keynes: “The ideas of economists and political philosophers, both when they are right and when they are wrong are more powerful than is commonly understood. Indeed, the world is ruled by little else. Practical men, who believe themselves to be quite exempt from any intellectual influences, are usually slaves of some defunct economist.”

  19. Predicting unemployment would be high for several years after a profound economic disaster, is barely a prediction. I think there is a massive over emphasis on the effect the government has on the economy. There is little empirical evidence that government tax policies have huge effects on economic activity. Obviously, a 90% tax rate would create a lot of unintended consequences and be a drag, but when tax rates change a few percentage points it is so little money compared to the overall economy it is marginal.

    The cost of education and healthcare is much more of an economic drag than tax policy. High healthcare costs in particular limit the ability switch jobs, limits creation and expansion of businesses. The effect is less pronounced in low wage/low skill jobs (since they provide few benefits), but in high wage/high skill jobs providing healthcare is a must to be competitive. Speaking from experience, you must spend a lot of time negotiating healthcare costs instead of creating new products to bring to market. I would much rather have fixed healthcare and education costs and pay higher taxes. Tracking profitability and a break even point is exceedingly difficult with highly variable costs that are a huge factor in employee retention.

    BTW — After dealing with several years 10% increases in healthcare costs. I shut my business down rather than expanding it. I had more customers than I could service, but it wasn’t worth the hassle as a small business.

    • Apparently you only read my conclusions. Read the analysis again. “High unemployment after a profound economic disaster?” The Obama administration has implemented many policies that keep unemployment high. As I said at the end of one of those articles, “there’s more to fiscal policy than government spending and taxes.” Indeed, a recent survey of small businesses by Gallup shows that many are reluctant to hire because of the uncertainty created by the unknown regulations emanating from Washington, D.C. Regarding the issue of economic drag, that is an empirical subject. Show me the evidence from reputable economic research (using actual data from the real world) and I might believe you. But in economics assertions that are not supported by either good theory or empirical evidence are not worth the electrons used to produce them.

  20. Great discussion – here are my 2 cents…
    I believe that the core issue with the market for higher education is that prospective students have no way of knowing the true marginal utility of their college degree since no one can predict their individual career path and success in life. Accordingly, students run the risk that the cost of their education ends up being higher than the marginal utility they derive from it over the span of their careers. Historically, a college degree has provided a powerful boost to lifetime earnings so that risk has been small. However, this has also established the perception of high marginal utility, which has given tremendous pricing power to colleges and universities. Now, add widely available financing, which is guaranteed by the government thus relieving lenders from having to measure that risk as well, and you have an explosive combination. The perception of high utility and easily available financing conspire to produce an explosive mix of ever rising tuition costs.

    That is not to say that making financing available for higher education is a bad idea. On the contrary, such financing is an important driver of economic growth and upward mobility because it enables people to get educated based on their future income potential not their current financial circumstances. However, providing financing without measuring the risk of overpayment (degree cost is greater than its utility) has enabled colleges to raise tuition costs unchecked.

    How do you solve this problem? First and foremost, the risk of overpayment should be transferred to the colleges and universities (because they overcharged for the value of the degree in the first place). This can be accomplished through a system of income-based payments and limited repayment period (say 20 years). After the expiration of the repayment period, the respective college or university is held liable for the remaining debt.

    In practice, the risk of overpayment can be priced through an insurance program. Lenders will demand that colleges and universities buy an insurance policy to cover the potential liability for unpaid student debt. Premiums will be determined by student repayment records. Schools with a good repayment record will have low premiums. Schools with poor repayment record (meaning the lifetime value of the degree was lower than its cost) will have high premiums. By comparing the premiums upfront, prospective students can easily determine which schools offer the best value and in the process, begin to apply market pressures on colleges to become more efficient and provide greater value. Under the current system, colleges and universities are paid regardless of the value of their product.

Comments are closed.