Arnold Kling

Education Loans

PRINT
Social Security Ad Nauseum... Anglosphere Challenge...

When I taught "Economics for the Citizen" this past fall at George Mason, I included a unit on education. In the latest issue of the Becker-Posner blog, Richard Posner brings up a point that I wrestled with in class.


But the "externalities" argument for subsidizing college education depends not only on how many kids would not attend college without the federal subsidy, but also on the cost of the subsidy to the taxpayer. I have no strong sense that the net external benefits are positive. If they are positive, it is very unlikely that they are large, considering the indirectness of this method of subsidizing education.

There is tremendous emotional support for public subsidies to enable young people to go to college. However, on closer examination, most of the benefits accrue to the individual rather than to other members of society. If your income increases, then that is a private benefit. If you enjoy the nice gym and the great basketball games at college, then that is a private benefit. Only some nebulous "better citizenship" benefits are public.

For Discussion. Are there estimates of the benefits of college that attempt to distinguish private benefits from public benefits?


Comments and Sharing





TRACKBACKS (1 to date)
TrackBack URL: http://econlog.econlib.org/mt/mt-tb.cgi/179
The author at Catallaxy in a related article titled Kling on Education writes:
    Whatever the strong sense may be, I had always believed (based in large part on my professors in college) that the overwhelming benefits of college accrue to the individual who gets the education. Social benefits, I don't believe, are easily... [Tracked on January 12, 2005 6:43 AM]
COMMENTS (27 to date)
Jonathan Brown writes:

The analysis of the benefits of student support for higher education is much too narrowly based. The California example is one proof of the benefits of this kind of subsidy and its public nature. A good deal of California's economic growth in the last couple of decades have come from two areas - first, the state had a relatively well educated workforce. Second, it could boast a relatively high number of research universities that helped to contribute to the growth in those industries. Is there any doubt that Berkeley, Stanford, USC, Calltech and UCLA and UCSD contributed to California's positionn in computers and biotechnology?

One could also look to the development of research-education related areas like Massachusetts and North Carolina for further confirmation.

spencer writes:

I would disagree with your comment that if your income increases it is a private benefit. Yes, it is a private benefit but it is also a public benefit. If the objective of public policy is to contribute to higher incomes it is also a public good. If you have a higher income because of your education presumably it is because you are more productive. But this means your employeer also benefits from your education -- under normal conditons the benefits of higher productivity are shared between both. If because of higher income from the subsidied education you pay higher taxes the govt or taxpayer also benefits from the education subsidy. If higher education leads to higher income and higher consumption everyone that sells you something you would not have bought without the higher education benefits. Consequently, I would argue that the
benefits to society as a whole of public subsidies of education are massive.

Mr. Econotarian writes:

I have heard of studies that show things along the lines of "enhanced tax revenues from college educated workers makes up for subsidized educations."

But the question is whether that enhanced revenue makes up for deadweight losses due to the taxes for the subsidiers, deadweight losses due to the subsidies, as well as forcing-out of lower cost private institutions by subsidized public ones.

Altogether, we have to admit that the US college system does a pretty good job...if we could apply the economic model of colleges (mix of private/public, price discriminatory mix of tuition and government funding) to all public schooling, it probably would be an improvement over a zero-price goverment monopoly.

Erik Sargent writes:

Just thinking that since I give >40% of my income to taxes, if my college education causes me to make $20-$50K more per year, that would provide quite a bit of return back into the public coffers from which the subsidy came.

Bob writes:

Spencer, while your point that massive benefits for all are created when people are educated is obviously correct, you can't ignore the opportuntiy cost if there is a budget constraint. That is, if you subsidize education for, say, ten million college students at $10k each, you lose that $100b. It is fair to ask whether the opportunity cost is too high, particularly when the private benefits are also huge so maybe 9.5m of those students would have gone anyway (taking into account the possibility that colleges capture much of the subsidy).

Bob writes:

"Just thinking that since I give >40% of my income to taxes, if my college education causes me to make $20-$50K more per year, that would provide quite a bit of return back into the public coffers from which the subsidy came."

Ouch. Now gov-subsidized student loans become a justification for high tax rates. Which finance increasing subsidies, which finance...what a racket!

Boonton writes:

In estimating the return on a gov't policy the benefit should be the total increase in income...not just the increase in tax revenue. If college increases yearly income by $5000 per year over a 35 year working life that totals $175,000. (I'll let someone else discount that using a rate they think is proper). Since student loan subsidies really only subsidize the interest the cost to the taxpayer is a lot less than the cost of college (or secondary education) itself. Plus there are also network effects to consider, increase the income of 25% of the population by $5,000 per year and the other 75% may increase as well.

I'm not sure we really know the deadweight loss of the tax since the gov't has multiple taxes & it can raise money by borrowing. Perhaps someone has constructed an 'average cost of revenue' for the gov't in the same manner that cost of capital can be constructed for a business?

An interesting question is why is the private sector unable to realize the profit in student loans?

One possibility may be the collecting them is too difficult. We have not yet arrived at the point where you can not be invisible. Even if you get a judgement on someone it's pretty hard to follow them around when they get a new job.

The gov't, though, is in the position of a fisherman on a lake. Anything that increases the number of fish on the lake is a benefit to him...even if he doesn't catch the extra fish. Even a person who doesn't pay back his student loan benefits society from the increased income.

Let's also not forget that the gov't has increased collection ability with student loans due to the fact that they can quickly attach income tax refunds.

An interesting question is why would the private market fail for providing student loans.

HispanicPundit writes:

I have a question.

Becker states, Put differently, lenders cannot take ownership of the human capital they finance since that means taking ownership of the individuals receiving the education, and no modern country allows people or institutions to own other individuals.

That’s odd. One of my favorite Economist Milton Friedman, in his book Capitalism and Freedom, argues for precisely that. Saying that student loan lenders should be allowed to take ownership in part of the salary of their customers, provided the customer accepts those terms. Am I missing something? Did something change since the 60’s and 70’s when Friedman wrote his book to now or might this just be a difference in opinion?

Eli writes:

Absent from this discussion has been the question of the effect of tuition subsidies on tuition itself. Tuition goes up in large increments because schools see little decrease in demand when tuition goes up a little. If the government got out of the education subsidy business altogether, it's likely that tuition would go down. One can gather, therefore, that many of the benefits of tuition subsidies accrue to academics, administrators, and university staff.

Jim Glass writes:

"Absent from this discussion has been the question of the effect of tuition subsidies on tuition itself. Tuition goes up in large increments ..."

You beat me to it.

Subsidies benefit those college and university staff who collect them far more than anyone else. Just look at how much more they collect as subsidies rise.

Colleges and universities are for-profit institutions just like any other, despite the fact that they are tax-exempt so the profits are divvied up among stakeholders rather than shareholders.

Harvard has a >$20 billion endowment, and instead of using it to give all its students free tuition it charges then near $40,000 and runs at an operating profit that contributes to the endowment.

Does anyone suppose that it and other colleges are any less "taking" of subsidies from the government?

dsquared writes:

I've always been surprised how many clever people trip up on this one.

The argument's quite simple; there is a financing constraint which prevents many people from buying as much education as they would want if that constraint did not exist.

The financing constraint exists because there is a missing market; it is not possible to pledge your own future income as collateral for a loan. This market is missing because of information asymmetries, and because the legal structure of a liberal democracy does not allow debt peonage.

Since this missing market is a "public bad" (as in, it is nonrival and nonexcludable), its alleviation can be considered a public good. So the choice is whether you think it makes more sense to alleviate it by providing the loans on a non-market basis, or by allowing debt peonage.

dsquared writes:

(further to the above, note that my proposition that "the alleviation of a public bad is a public good" is not rigorous, because there could be public bads that could be alleviated privately; I think I can defend it in most real-world cases though)

Daryl Gorman writes:

One good of subsidized education is that it counteracts classism - meaning it helps ensure that there is a route to wealth for the next generation other than being the lucky offspring of this generation's wealthy.

The idea that society is "fair" is external to the individual beneficiaries. I'm not sure how much that is worth, but I'm also not sure there is a cheaper way to buy it than subsidizing education.

John writes:

I wonder why there isn't more competition in education. With the advent of the internet we now have things like "distance learning" and the MIT open course programs where classes are taught online at very low marginal cost. Where are the e-colleges that will compete with the existing college structure and exert some downward pressure on tuition costs. It seems that most of these types of programs I read about are of the disreputable sort (diploma mills). If the education market in its current form is suceptible to Baumol's disease, then some entrepreneur should be able to make a fortune by opening a chain of McCollege franchises.

Boonton writes:

"Harvard has a >$20 billion endowment, and instead of using it to give all its students free tuition it charges then near $40,000 and runs at an operating profit that contributes to the endowment. "

Like an exclusive club, Harvard has more customers than seats. Hence they can charge a price higher enough to almost equalize demand with supply. A corporation in this happy situation would expand supply. However Harvard isn't run by those seeking to profit from its operating profit but rather alumni seeking to hold down the supply of Havard degrees issued each year.

Lancelot Finn writes:

That education is publicly funded yet provides private benefits applies not just to tertiary but also to primary and secondary education. To my mind, the entire notion of education as a public good is misconceived.

Public provision of education is justified, but not because it is a public good. At least, not in the economic sense. Spencer writes that: "[Higher income due to education] is a private benefit but it is also a public benefit. If the objective of public policy is to contribute to higher incomes it is also a public good." By this argument, every private good is a public good. We might as well say that food is a public good because society is better off when people are better nourished. Public goods must be non-excludable, like national defense or law and order. Education is excludable.

Rather, the reason to provide public education is the one Daryl Gorman points out: "subsidized education ... counteracts classism - meaning it helps ensure that there is a route to wealth for the next generation other than being the lucky offspring of this generation's wealthy." Without public education, the children of well-to-do parents would do just fine, perhaps even better than at present, in the private education market. But children of poor or shiftless parents would not get an education and would see their earning power permanently reduced. Public education is a means of intergenerational redistribution, of equalizing the initial human capital endowment of the citizenry.

dsquared's claim is a variation of the same point: "The argument's quite simple; there is a financing constraint which prevents many people from buying as much education as they would want if that constraint did not exist." Whether this point applies to college education is debatable, but it surely applies to primary and secondary education, or it would in the absence of public education, since children are not considered legally capable of binding themselves into long-term credit arrangements. Moreover, the financing constraint would affect children differently: poor children would face it, richer children would not.

The weakness of dsquared's point is that 1) the market failure he assumes, namely that private markets could not enforce long-term debt contracts, is probably non-existent (I for one have borrowed a huge amount of money from private lenders for educational purposes) and 2) outside the market mechanism, we have no good way of knowing whether social investment in college education is paying off.

I do not find the claim that college education is an efficient investment in human capital for the purposes of raising long-term income plausible, either from the empirical evidence or from experience. One stylized fact figures against it: the rise of mass college education in the 1960s was followed, not by rising, but rather by a notable fall in productivity growth. In my experience, very little of what I learned in college has had much applicability to life in the real world, and I work at a think tank; for most of my friends, the usefulness would be much less.

I would interpret college differently. First, it's a form of consumption. For a lot of people, college is one of the best times of their lives. Second, it's a mark of class or rank, like medieval coats of arms: "Where did you go to college?" is a question that often arises early in a conversation with someone you've just met, and it's good to be able to say "Yale" or at least to have graduated, rather than to say, "One year at..." or "Nowhere." Another part of the class/rank aspect of college is that you make friends, and likely marry, with people as smart as you. Third, it is an investment in the capacity to enjoy. My Romantic literature class never put a dime in my pocket, directly or indirectly, but I can still read William Blake with pleasure. Perhaps more relevantly, I can participate in much more sophisticated and interesting conversations than I could have without college.

Fourth, for many people it's just a waste and a mistakeA lot of people I knew in college were confused and directionless. Around senior year an anxiety set in, a fear, terror even, of the Real World. Afterwards, they are deep in debt. The fun's over, and yet they didn't even have any fun. Related to this (fifth) it's a postponement of adulthood. It's sometimes very strange to note the basic life skills, such as cooking and paying bills, that college students don't have. On the other hand (sixth) it may have a high moral motivation, a quest for truth. That's what it was for me, and I was annoyed that so many people were just there to drink beer or make more money. Seventh, the increase in college education is partly a process of substitution for the failure of primary and secondary education to improve. While almost every economic sector is becoming more productive, the public education system is stagnant or deteriorating because it is a government-run monopoly. People go to college to learn what they could have learned before if the public education system were half-decent.

So while the financing-constraint argument makes sense, I disagree with it. I think it would be better if non-trust-fund babies had to work and save for a couple of years, and build up a credit rating, before they could go to college. If we took Mr. Econotarian's advice to "apply the economic model of colleges (mix of private/public, price discriminatory mix of tuition and government funding) to all public schooling," better primary and secondary education would probably reduce the need for mass tertiary education anyway. Those who did make it to college after a couple years of work would be more mature, more serious, less angst-ridden, with a valuable exposure to the Real World, and with more of a notion of what they wanted to do.

Meanwhile, there's another reason to publicly subsidize universities that has nothing to do with the benefit to students: knowledge creation. This is the reason that Jonathan Brown's comment-- "Is there any doubt that Berkeley, Stanford, USC, Calltech and UCLA and UCSD contributed to California's positionn in computers and biotechnology?"-- is cogent. So while we should probably stop subsidizing college education, we might want to look for other ways to continue, or even increase, subsidizing the universities' knowledge creation role. In the past, the British Crown would regularly shower rewards and money on favored inventors, writers, and so on. Maybe we should revive this function today: governors, legislatures, presidents and so on should create large cash awards for anyone whose work proved pleasing, or valuable: bloggers, book authors, scientists, etc. The Nobel Prize model of funding knowledge creation. If publicly funded scholars were more directly answerable to the public, it would be more difficult for academia to turn into a museum of discredited ideologies.

JT writes:

Lancelot Finn: what a great post! Thanks for taking the time.

I also find dsquared's argument unconvincing but, even if you accept it, why not tinker with debt & bankruptcy laws to facilitate the development of a private student loan market? I'll have to be cynical: the objection to such an approach would be that it doesn't create jobs for bureaucrats in government and education.

Deb Frisch writes:

AK: Only some nebulous "better citizenship" benefits are public.

Some percentage of my quality of life depends on how RATIONAL the government's policies are. If the government banned the trade of all goods and services except for heroin and sex, almost everyone's q of l would be much worse than it is today. If the government decided to tax 100% of people's incomes or ban usury, I'd be bummed.

Some of my q of l depends on whether my government is BIASED against my race, sex, sexual orientation, religion, physical disability, mental disability, etc. Is "my kind" discriminated against economically, socially, politically, physically, etc?

For example, the quality of life for an African-American or a female in the US varies more as a function of time than the q of l for white men.

Since this is a democracy and the laws affect me, my utility depends on the quality of other people's decisions, the accuracy of their beliefs, etc.

My mind is truly boggled that you view the benefits to self of others being educated as "nebulous."

Lancett writes:
"The financing constraint exists because there is a missing market; it is not possible to pledge your own future income as collateral for a loan. This market is missing because of information asymmetries, and because the legal structure of a liberal democracy does not allow debt peonage."

Actually, some universities offer exactly this financing option. The common model is to offer the student to pay back the tuition via a fixed percentage, say 8%, of his or her future income for some years. Of course, there is an absolute cieling to the rates.

"Fourth, for many people [college education is] just a waste and a mistake"

I don't think that applies. A key function of formal education in college or anywhere else is to train in thinking and applying aquired knowledge to problems - or learning to learn. Someone trained in a college will in most cases have a higher initial productivity when it comes to solving specific problems. Thus, the average educated person has bigger (potential) value in this regard and is also more flexible in dynamic job markets.

Lawrance George Lux writes:

Are there estimates of the benefits of college that attempt to distinguish private benefits from public benefits?

The G.I. Bill was the first introduction in Government subsidy, and without the trained labor produced, the economy could not have expanded at the rate it did in the 1950s-60s. Libertarians have a habit of describing increasing Seniors/Worker over the coming decades, and then suggest We leave such Labor uneducated and untrained. lgl

dsquared writes:

Two points:

First, the constraint will always be there because there is moral hazard. As far as I am aware, banks usually only stump up loans for medical, legal or MBA education, where there is a clear path toward employment that will repay the loan. I don't see how you're going to finance scientific or general education in this way.

Second, it is not just a supply-side problem. Because there is no market in which an individual can trade his expected lifetime earnings for its certainty equivalent, people will in general spend less than the optimal amount on education simply because the investment bears undiversifiable risk.

(I'd also like to make a methodological quibble; there are no such things as "rich" and "poor" children; apart from a few orphans and other odd cases, children are all in general poor in both assets and income. This is an important distinction to make at the college stage, because it is the key argument against claims that the education subsidy is a regressive redistribution to "wealthy students".)

dsquared writes:

By the way, reading back my original comment, I pointed out that there was a choice of options between subsidising education and "allowing debt peonage". I've never argued that the law couldn't be changed so as to make it possible to have general use of privately provided loans for education; only that liberal democracies are probably better places to live in because of the legal arrangements which make this market impractical.

spencer writes:

Some very good points but the one I would take major exception to is Dsquared comment on no poor children. Now really, upper income individuals do pay for their childrens education so to say that their children have no assets or income is besides the point.

Another aside on Harvard and other elite schools tutition -- the actual people that pay full tutition is very narrow and is heavily concentrated among foreign student.

JT writes:

dsquared: Thanks for the response. It made me consider why the student loan market would be different from consumer loans for the purchase of things other than hard assets, an example being the credit card market. Credit card issuers are willing to lend money for the purchase of assets which cannot be repossessed, generally speaking. But they also charge high rates of interest, often in excess of 15%. I can't see students being able or willing to pay those sort of annual rates. Hope this translation into non-specialist terms is correct and helpful.

triticale writes:
...then some entrepreneur should be able to make a fortune by opening a chain of McCollege franchises.
Not a franchise, but isn't this about where U of Phoenix and Bryant & Stratton come in?
Jon writes:

Besides reducing "classism", public subsidies of education also give a larger number of people a stake in the system. This means that people are more willing to behave as good citizens, serve on juries, and serve in "service" type professions (such as the military). If the system were designed so only the few brightest or born wealthy could afford an education, then the rest would be less apt to support the system.

This is probably why it is the poor conservative areas in red states, not the educated liberal suburbs that generate the most ridiculous jury verdicts.

Furthermore, the argument that the less wealthy should work a few years before going to college, flies in the face of biological reality. Delaying the onset of a career this way makes it more difficult to raise a family. Furthermore, the claims of "private" financing ignore that much of it is through non-profit institutions (i.e. the schools) who don't really function as "market" institutions or eventually gets a government guarantee.

Finally a college education often involves factors that are not provided by websites -- mentoring, opportunities to do research, and discussions with faculty and other students. The claim that the internet can completely substitute for this has yet to be proven.

Dezakin writes:

Well, while I found Lancelott's post verbose and interesting, I'm not entirely sure I agree with its conclusions, aside from the necessity of universities as repositories of knowledge creation.

Rather I think the problem with modern universities has been the lack of emphasis of pragmatic education. Less that 5% of US graduates have engineering related degrees, where in India and China its closer to 50%. While Lancelott eluded to the cause being a monopoly of some sort or other, I think the real problem is much more heavily cultural. You might counter this by only subsidizing 'useful' programs, like engineering, medicine, maybe law; Or perhaps the size of grants is proportional to the average starting salery companies say they give to graduates. I believe the real problem is we subsidize the philosophy and art degrees nearly as much as the engineering degree.

Of course if we really want to not subsidize universities and still reap the benifits of subsidizing universities, we can do it for free today: Offer green cards to all who wish to come work in the US who have a college degree. (This would have an amazing impact on medical costs if we could hire foreign doctors here)

Comments for this entry have been closed
Return to top