Tuesday, July 17, 2012

More on the 'Equity College'

Yesterday's post, "The Human Startup," posited that debt financing faces some real problems as a model for making investments in human capital -- namely, college. It suggested that an equity-financing model which replaces tuition with an equity stake in an individual's future earnings could be at least a viable competitor, if not better suited to the "startup" type of investment made through college education.

The post received a considerable amount of support as well as a number of critical questions. I appreciate both, and as I make the case for equity financing of human capital, I'd like to remind both sides to keep an open mind and to put forward the best version of their arguments. That's what I'm trying to do; I'm not wedded to this particular idea -- or any idea, for that matter -- but I want to give it a full hearing. If you were 18 and were watching as your generation was being saddled with mountains of debt and given college educations which in a surprising percent of cases provided a rather poor return on investment -- see this story in The New York Times -- you'd be trying to fix the system too.

Mike Konczal asks, "Isn't equity ultimately about control/ownership? What controls would financiers have over their students to handle agency problems?"

Some forms of equity are definitely about control. When some venture capital firms or private equity firms buy into startups or foundering firms, they are looking to replace management, restructure the corporation, layoff workers, cut costs, etc. On a less active level than that, pension funds and other institutional investors often get to select members on boards of directors of major corporations by virtue of the voting rights vested in their large equity stakes. But I don't see control as the defining feature of an equity position. When someone buys shares of an exchange-traded fund of small-capitalization stocks, they are also taking an equity position. To what extent are they in control of those companies? Instead, I see the defining features of an equity position as (1) being the junior asset class, as debt gets repaid first and (2) the upside potential as compared to the defined return of debt.

This matters because I don't want administrators assigning kids fields of study, having them sign away rights, having them work in that field for the rest of their lives, and lording over them to make sure they make every life decision in line with profit maximization. The degree of "active management" I thought  an Equity College should take is not of the venture capital/private equity sort; in fact, it looks much more like that of institutional investor. If colleges had an incentive to maximize the lifetime externality-adjusted income produced by their graduates, I think they would offer much more intensive advising and mentorship programs than they generally do now when students are in college, as they graduate and enter labor markets, in career transitions, and generally well into their adult lives.

To get right to the heart of it, I would think it a bad idea to empower the equity stakeholder to have a binding vote over such decisions. The most limited scenario I can think of is where students might have to pledge that they will work for XYZ Corporation for several years -- perhaps five -- after graduation in return for the payment of their tuition. (In fact, many minority-recruitment programs already do this in the United States.)

Konczal also asks about principal-agent problems -- he is worried that the incentives of the graduate are not aligned with the equity stakeholder. What matters in aligning incentives, obviously, is the payment structure: if Equity College tries to take half of graduates' income, they might choose alternatives to income, such as leisure or in-kind payment. Similarly, if the payment plan kicks in at too high of an income level, graduates might decide that they'd rather make just under the limit. (It is worth pointing out to Konczal here, with a little bit of humor, that if he sees the principal-agent problem as damning at 5 percent of income above $50k in constant dollars, well then, just wait until he sees the government's marginal tax rates on income!) If an Equity College grad keeps 95 cents on the after-tax dollar, is he or she really going to behave any differently than if they had that extra 5 percent? Especially given that college graduates feel a lot more loyalty to their alma mater than to the IRS -- I encourage you to compare charitable giving to colleges versus to the Treasury -- I would venture to guess not. Nor are their improper incentives for short-term gain on the part of the agent of the sort which plague executive compensation; I am stuck with myself for life, as far as I know.

Timothy B. Lee writes, "Also there are huge adverse selection problems. If I expect above-average lifetime income I will finance with debt."

If all bids are the same regardless of expected income, then individuals who can reasonably expect to earn more or less than average will choose the lower lifetime cost option, that being debt or equity financing respectively. I didn't see that before, and Lee has a good point there.

This can be rectified. The equilibrium price of debt and equity financing should be the same for any given student in a situation where there is no information asymmetry. To do this in practice, an Equity College may have to tailor its percent-of-income and income-treshold bids according to the expected annual income path of the applicant. With the introduction of asymmetric information and tailored bidding, applicants could try to mislead an Equity College. Lee might argue that those who have concealable information about their future earnings will try to game the system and generate adverse selection. Everybody will argue that their earning potential is far higher than it actually is -- in short, duds will try to pass themselves off as studs.

Does this remind you of anything? This is how equity financing works. No venture capitalist knows exactly the future returns their investments will see, and every company which seeks funding from Sand Hill Road promises to be "revolutionary" and to bring "paradigmatic, disruptive change" to their industry. Hyper-optimism oozes from the sales pitch. But it's the job of the venture capitalists, of the equity buyers, to accurately assess the value of the company from the information they have. And I don't think adverse selection via information asymmetry causes startups with bright futures to shun equity in favor of debt. Given a highly rigorous admissions process, I think equity financing can have sufficiently symmetric information to work. As a side note, Olin College, which "pays" half the tuition of its students, does this already in recruiting its small classes. (The interested reader may also want to see these two papers -- here and here -- on venture capital investment under information asymmetry. The former finds that mitigation of information asymmetry tends to result in better returns for the equity investors, and obviously that some firms succeed relatively in this endeavor; the latter finds that information asymmetry may actually help the case for equity financing if there is default risk or collection costs.)

Again from Konczal, "Price discovery difficult too. Also curious on PV of equity college - not ton of money?"

I don't think price discovery would be unusually difficult -- perhaps I am missing something. The market for private student loans is large and liquid enough for there to be competition. Yes, if students get a debt offer and an equity offer, it may seem like a comparison between apples and oranges, but ultimately it's about family finances -- i.e., marginal utility and the discount rate. It is best that families have a choice as to which model of financing human capital acquisition is right for them.

I'm thinking that the present value of an Equity College education would be in the ballpark of traditional colleges' present value at first -- which this Bloomberg Businessweek article estimates as $1 million. In the long run, I would expect the superior incentives of an Equity College to yield higher productivity gains in the form of rising rates of return on investment, and therefore a higher present value relative to traditional college.

From Josh Barro, "[T]here's also a significant moral hazard problem."

What are risks Equity College graduates incentivized to take as a result of the 5 percent equity stake? Not clear on this point -- see my prior discussions of principal-agent problems and information asymmetry.

From Matt Bruenig, "As appealing as indentured servitude sounds, why not universal income-based repayment instead?"

Hundred-thousand dollar debts, which often take large percentages of annual income for new college graduates, strike me as servitude also. (I refer you to the Times story again.) The problem is in the fact that people need to make a huge investment in their own human capital up-front if they are to maximize total lifetime utility -- nearly everyone will need a means of financing this. The question is one of means.

Only having a basic knowledge of income-based repayment -- see here for an outline of the policy -- I think it is more meant as a means of debt relief which increases federal liabilities. But in the sense that borrowers pay a percentage of their income, my equity proposal strikes me as similar. The major difference is that I would not cap the maximum payment in the case that the borrower does well in life, because I think it produces valuable incentives for the college to increase the income of its graduates and it would drive down the cost of college for those who make less and find that they have benefited less from their college education. Instead of a ceiling, I would have a floor -- you've got to be decently successful before an Equity College would require repayment. Incidentally, that makes my proposal much more progressive with income than Bruenig's.

From Billaire, "Didn’t Yale try this already, and fail at it, because of foreseeable adverse selection problems?"

I mentioned the Yale experiment, which I think has the design flaws mentioned in the article you provided me:
TPO's major problem was that its participants were set up in groups called "cohorts," meaning that all members would keep paying until the entire group's debt was paid off. As inflation mounted, tax laws changed and some participants defaulted, the cohort's debt burden persisted...Making matters worse, the wealthier segment of student participants bought out of TPO early, paying 150 percent of what was borrowed plus interest. That left lower-income students with a greater burden to be split among fewer people.
The solution is to write the contract differently than did Yale: No cohorts. No buyouts. Use real income. Work with Congress to configure the tax code so that equity contracts can garnish wages just as the debt contracts, student loans, can. If Tobin thought it was a good idea, maybe it remains a good idea, just one which needs more work.

From Sam, "If lenders were able to discriminate among students, [highly-qualified students] would be able to get very, VERY cheap financing for education. But I think the distributional reality makes it a nonstarter."

I think student loans already have distributional realities which are profoundly unjust. Under such a debt system, college affordability is contingent upon the ability of the family to support the prospective student -- that is, it's a great deal for the wealthy, price discrimination and need-based lending notwithstanding. Under an equity system, it's about the ability of that student to succeed with a college education. Although that may still favor the well-prepared, as they may tend to be better off, I see it as a huge relative improvement over what we've got. It opens the door for hard-working, low-income students; I'd rather the distributional benefits accrue to the hard-working student than the well-off family.

3 comments:

  1. Speaking of endowment funds and the like, have you heard about the Chambers and Dimson working paper on J.M. Keynes's investment performance? It received coverage in The Financial Times and other media outlets. If not, here is the working paper, Evan.

    http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2023011

    ReplyDelete
  2. You mention a "floor;" which to me indicates that highly unsuccessful students would not repay for a relatively long time. However, this raises a concern: will colleges readily sign on to a system which promises extremely large returns on the tail end (the occasional high-performer who provides millions of dollars of returns), but which would delay repayments more generally?

    Depending on your threshold, unless the system is at least as aggressive as the student loan system presently is, colleges will have to defer income into the future. Can they be convinced to do this?

    Also, how would you handle defaults? If a company loses its value and defaults, shareholders just lose, but lenders can still expect some degree of repayment. For student loans, default is impossible.

    Would students be allowed to default out of equity? And, if so, and if it weren't debt, would they thus escape wage garnishing? Or would the same provisions exist for an equity agreement that exist for a debt agreement?

    Also, it is worth noting that schools do already attempt to extract equity-like returns through pursuit of alumni giving, the presumption being that wealthier alumni will usually give more.

    ReplyDelete
  3. Adding to the above comment, the ability to "extract equity-like returns" through more contractual means might cause norms of generous alumni contributions to decline, much like the famous example of how a daycare center's policy of fining late parents backfired because they somehow believed that it absolved them of the social obligation to be punctual.

    ReplyDelete