Arnold Kling  

Should College Attendance be Subsidized?

PRINT
Israel's Nasty Minimum Wage... How Singapore's #1...

Guess who said :


Going to college is a lot like standing up at a concert to see better. Selfishly speaking, it works, but from a social point of view, we shouldn't encourage it.

Answer here.


Comments and Sharing





COMMENTS (24 to date)
Arthur_500 writes:

I'm in with Bryan on this one for a variety of reasons.
If you want it you will get it. If we subsidize it it loses its value so why bother working hard?
Education can be an important stepping stone in life and those who choose to pursue an educational base for their life's toolbox will probably do better than those who do not.
Subsidies typically devalue the thing being subsidized and can't be considered useful to a valuable education as a generality.

thrill writes:

From the link that Arnold provided:

Marty Nemko:All high-school students should receive a cost-benefit analysis of the various options suitable to their situations: four-year college, two-year degree program, short-term career-prep program, apprenticeship program, on-the-job training, self-employment, the military.
Marty's mostly right - except that these students shouldn't receive this analysis - they should have to generate and defend it as part of their graduation requirements.

wintercow20 writes:

I'm not sure that college should not be ... taxed. Sure college may reduce the amount of criminals - but it might create better criminals too. Sure, college might make you a better informed voter - but that is a negative externality as far as I am concerned. And given what is being taught at much of the colleges in the U.S. - it would seem to me that much of it offends near half the population.

I'd separate the discussion of imperfect capital markets from the externalities from higher education attainment. I am not sure the net social external effects are positive.

Les writes:

It seems to me that college attendance is like any other product or service.

Subsidize it and we'll get much more of it than can be economically justified (but much less of other things).

Or tax it and we'll get much less of it than can be economically justified (but much more of other things).

With either a subsidy or a tax, we are reducing economic efficiency and creating needless surpluses or shortages.

fundamentalist writes:

Colleges have become nothing but trade schools and serve two purposes. 1) A college degree cuts down on the competition for jobs in law, medicine, accounting, real estate, journalism, etc., so that those who can get through college can earn higher salaries. 2) It saves companies the expense of basic training for engineers and other professions that used to be learned in an apprenticeship. Because we allow state licensing of professionals, we pay far more for education than we would in a free market. That means we have less money to spend on really important things, like football tickets. And as Adam Smith wrote, state licensing provides cover for charlatans.

Ryan Vann writes:

"Marty's mostly right - except that these students shouldn't receive this analysis - they should have to generate and defend it as part of their graduation requirements."

That's assumes they have access to the data necessary to perform said analysis, which would assume that college's don't have an incentive not to make such data easily accessible. I do agree with the spirit of your argument though. There needs to be way more finance/econ taught k-12.

Greg Ransom writes:

It's worse than that.

College churns out people who are a negative drain on wealth production and happiness and a healthy society -- e.g. lobbyists, lawyers, English professors, undergraduate drunks, leftist school teachers, bad poets, socially delayed permanent adolecents, unreadable novelists, unhappy grad students, PC script writers, idiotic social critics, leftist journalists, Congressional aids, etc.

For a parallel think of the socially pathological output unintentionally produced by the "reform schools" -- e.g. teaching children who've done something wrong how to
be professional criminals or true sociopaths.


L

Ryan Vann writes:

@Greg,

Did you jokingly throw in PC script writers? Regardless if it was intended to be funny or not, I was laughing pretty hard at that one.

Thomas Leahey writes:

I would like to link this post to the earlier one on the fall of the Berlin Wall, because it seems to me that there is a certain degree of socialism in Arnold's comment about college.

Compare this (from Richter)
"Whatever the individual man may be, the Community has made him what he is."

to this (from Kling)
Selfishly speaking, it works, but from a social point of view, we shouldn't encourage it.

It seems to me that in both cases the interests of the Whole are being placed above the interests of the individual. My view is reinforced by Ransom's comments about "negative drain on wealth production" which implies that every citizen's positive duty is to contribute to the wealth of the Whole.

This looks to me like socialism via conformist pressure rather then Richter's socialism by force. In both systems, the individual is valued only to the degree that they contribute to some greater good.

Robinson writes:

And the prize for the world's easiest guessing game goes to...

Ryan Vann writes:

Thomas,

I'm not sure Kling's quote should be taken to mean that education is everywhere and always beneficial to the individual, and always a detriment or cost to society. I think his view is more nuanced, in that certain individuals absolutely benefit themselves and society by being educated, but most would not. The net effect of promoting education is negative because it ends up costing individuals more than it benefits on the whole.

In other words, I think he is aggregating and not invoking some nebulous conception of society. I took Greg's post as more of a facetious rant.

Justin Martyr writes:

Let's suppose that college really is just signaling. Does that mean we shouldn't subsidize it? College doesn't signal wealth, but rather basic skills, internalized work ethic, and time preference. Given that employers do seem to value this signal, then it may still be efficient to subsidize college.

Moreover, call me old-fashioned but I think that college does provide externalities in terms of providing a civic education. It doesn't do that as well as it did in the past because colleges are full of stupid electives. Students should be required to take western philosophy, western history, a stats course, and a few social introductory social science courses like econ and psych 101. But even so, I think college provides benefits in that way.

Ryan Vann writes:

Justin,

I think there is some merit to the signaling assumption, but I'm more inclined to think it signals an acceptance of norms and compliance to expectations (which are also beneficial to a company, as renegade workers can be very problematic).

The problem is, when these norms becomes normal (see some of the bar graphs posted here about education percentages) that signal becomes dilluted.

Dan Weber writes:

One thing about signaling is that it's not just signaling that you are competent enough, you are signaling that you are more competent than everyone else. Hence Bryan's use of the concert metaphor.

It's like trying to help two people escape the bear by getting them to each run faster. The bear (brown bears, at least) can run over 40mph, so there is no level of improvement or training or running shoes you can give so they both escape. One of them is still going to get eaten.

If we stick with the college metaphor, it worked very well for me as a signaler. I came from a middle-class background but went to an elite school and that opened a lot of doors that I would not have had otherwise.

Using college as a signal sure can be expensive, though. I did learn a lot, though, and I think that was very good in and of itself. I probably could have learned it at a less elite institution, though.

One proposal is to cut out the signaling aspect: make educational achievements illegal to ask about on a resume, like you can't ask about race or family status. I'm sure some will still leak through, but, for signaling, there ought to be a more efficient way to tell employers I'm a hard worker than my willingness to sacrifice several years of income (assuming tuition is zero).

Justin Martyr writes:

One thing about signaling is that it's not just signaling that you are competent enough, you are signaling that you are more competent than everyone else. Hence Bryan's use of the concert metaphor.

I don't agree with that metaphor. We are not in a zero sum game in terms of the number of quality workers. (1) developing the signal makes it easier to create a signaling equilibrium. (2) creating an educational arms race helps establish social norms of self-discipline and deferred gratification. We end out with more disciplined and quality workers. It is win/win.

Dan Weber writes:

Key to the concept of college as "just signaling" is that it is not changing its inputs. The individuals that graduate come out about the same as they went in, but they are now have a stamp of approval.

If you accept that "college is just signaling" for the purpose of this argument (as you said), then there is minimal benefit to trying to shove more people into the contest. Trying to achieve the signal and failing is of no benefit.

There may be some societal benefit to putting as many 18-year-olds as possible through the four-year-college wringer. Even though we will only extract the best (say) 100 thousand a year, if we shove another half-million into the grinder, we will get out a slightly better 100 thousand at the end.

Ignoring the costs to the individuals, though, I'm not sure that for society it's worth the cost to slightly improve the quality of those who achieve the signal, given that we subsidize their education financially and are denied the labor of those who would have fine jobs as plumbers or electricians for a handful of years.

We are graduating plenty of B.A.s now that do not use their college education in their jobs. If we try to send another, say, 20% of the population to college, we will pile up more people at the bottom of the class, if they even manage to graduate.

We can eliminate most of the college-as-signal effect by masking it from employers.

ajb writes:

Too many defenders of subsidizing college education ignore the difference between the average college or college grad and the marginal school or student.

Are we really saying that the typical person who isn't able to make it to community college and then transfer to some four year college today should be subsidized? Do we really think that most of those who drop out of college should be subsidized till they finish college?

You don't have to fully agree with Caplan to appreciate that a large fraction of graduates of the typical four year college learn little of value or learn it inefficiently (We're not talking UCLA or Swarthmore here. Even GMU is well above the mean). My guess is that the typical graduate with a non-technical degree would learn just as much from a focused one year program that stressed writing skills, analytic reading, remedial algebra, and some basic computer usage. Guess what, the better high school grads already have those skills. And anyone who can't acquire those skills after high school in one year, shouldn't go to college at all unless they pay for it themselves.

GU writes:
"A college degree cuts down on the competition for jobs in law, medicine ..."

Maybe the J.D. & M.D. cut down on competition for jobs in law & medicine, but a "college degree"? With all of the lawyer/doctor bashing that goes on in libertarian circles (much of it justified) many seem to forget that law and medicine are both for the most part really difficult professions to practice.

When only 10-20% of the population has the cognitive skills needed to perform reasonably in skilled professions like law and medicine—and with much fewer willing to put in the hard work learning the skills, and long work weeks associated with, law & medicine—the supply of competent lawyers and doctors is never going to be very large. Do current licensing regimes restrict supply? Yes, but only a small bit, especially if you restrict "supply" to competent practitioners (and not everyone who claims they want to be a doctor or lawyer).

My point: libertarians (of which I am one) should drop the doctors/lawyers are evil rent-seekers canard. There are much more important things to worry about/try to change. We are never going to have competent human-provided medical care or legal services for $20/hour. C'est la vie.

fundamentalist writes:

GU: "We are never going to have competent human-provided medical care or legal services for $20/hour."

Nurse practitioners are doing it, in many cases better than MD's. I can't tell you how hard the work in law or medicine is, but in my experience I haven't found doctors or lawyers to be particularly bright people, especially lawyers. The President is a prime example. They tend to have good memories so they can regurgitate well, but lack basic reasoning skills. BTW, research has shown that computer programs do a much better job of diagnosing diseases than do doctors.

Brian B writes:

Bryan's position seems strange coming from someone in higher education. Why is he in this profession if he thinks it only provides a signaling effect?

The reality is that a college education (especially the degree) is an example of signaling, but it's not "just signaling." College provides a benefit to anyone who attends, and, just as important, provides a net benefit to society. The more people who receive a college education, the better off society is, without limit. A plumber with a college education is a better plumber and a better citizen than one who doesn't, all other things being equal.

GU writes:

I am in favor of letting nurse practitioners do simple things like draw blood. But there are some problems that, at least in the near future, must be dealt with by highly skilled human beings.

I know more about law, so I will give an example from that field. You will not be able to (in the near future, and maybe never) type into a computer program your legal problem and have it litigate a case for you (or negotiate a complicated merger transaction).

Are there ways to make legal services cheaper? Sure, have paralegals do non-difficult administrative and research tasks for example. Pre-made legal forms for routine contracts is another example. But complex legal problems require real, live, skilled lawyers, and to make it worth their while to go through all the trouble of representing you, you have to pay them. You get what you pay for in many instances in life; it is disheartening that there is no magic fix, but there is little to be done about it at the moment.

That's why I say, focus on easy fixes or things that matter much more. On the list of evils in the world, gov't. licensing of doctors and lawyers must be pretty darn low on the list.

ps Obama is not lacking in reasoning skills. Like most politicians, he is an amoral power-monger and he has already proven to be extremely effective at increasing the size and scope of government.

Ryan Vann writes:

"A plumber with a college education is a better plumber and a better citizen than one who doesn't, all other things being equal."

I don't see how you would go about demonstrating any of that.

Brian B writes:

"'A plumber with a college education is a better plumber and a better citizen than one who doesn't, all other things being equal.'

I don't see how you would go about demonstrating any of that."


Ryan,

I don't have any proof that it's true--I'm just hypothesizing--but it wouldn't be particularly difficult to investigate. If we acknowledge that plumbing involves more than just knowing what pipe goes where, and that better plumbers likely have better customer relations, are more likely to supervise other workers, are more likely to make higher incomes, etc., it wouldn't be difficult to test the outcomes of college-educated plumbers versus those who are not. In terms of citizenship, you can bet that college-educated plumbers are more likely to be involved in their communities and take leadership roles. With a large enough sample size, it would be straightforward to adjust for things like IQ, family background, place of business, etc.

GU writes:
"you can bet that college-educated plumbers are more likely to be involved in their communities and take leadership roles.

1. First a minor quibble. In my experience,"getting involved in their communities" usually means (unjustifiably) getting involved in other people's business. Often those "involved in the community" are involved in lobbying the government to curtail the liberties of others in some manner.

Of course there are plenty of "good" ways to get involved in one's community—I'm not denying that.

2. There can only be so many leaders. Something like 60% of people are going to college these days; do you really think there's a huge untapped potential leaders contingent out there? Also, again, being a leader in today's world means getting what your special interest group wants at the expense of everyone else. More leaders? No thanks.

Comments for this entry have been closed
Return to top