Arnold Kling  

Why is College Expensive?

Thiel's Priceless Publicity... Time to Re-think Securitizatio...

Robert B. Archibald and David H. Feldman write,

First, higher education is a service industry. From 1947 to 2009 the average annual price increase for services was 4.0 percent, while for goods the average annual price increase was only 2.4 percent...

Higher education also shares with many other personal services a reliance on a highly educated workforce. In the late 1970s, the wages of highly educated workers began a sustained rise.

instead of reducing the number of labor hours it takes to produce a class, new technologies alter what we teach and how we teach it...Just like modern medicine, colleges and universities must meet a standard of care, and that standard is set in the labor market that will employ our graduates.

Do not post a comment until you follow the link and read the whole thing.

There is a chicken-and-egg problem with cost-disease stories. That is, do we subsidize health care and education because their costs are rising relative to other good? Or are their costs rising because we subsidize them?

As you know, I think that credentials bottlenecks are at the root of much evil here. What if it were easier for a non-accredited university to compete? What if it were easier for an accredited university to fail? Why are some careers (I'll use my favorite example, physical therapy) only available via a classroom path rather than an apprenticeship path?

I think that if the problem were just that college education uses a highly-educated labor force that is in limited supply, we would see different economic behavior. When you have a scarce resource, you try not to waste it. But it seems to me that professors waste most of their time engaged in status competition, meaning the attempt to publish.

Comments and Sharing

COMMENTS (13 to date)
fundamentalist writes:

Seems to me this would be a good issue on which to deploy econometrics. Yes, there will be quite a bit of correlation to contaminate the analysis, but using something like partial least squares would clear up a lot of that. I would say that all of the issues the authors raise are likely candidates, but which ones have larger effects can't be determined from anecdotes. You need some serious data and the appropriate statistical analyses. I don't think it could be done outside of PLS.

My gut feeling is that demand increased rapidly as more and more field began to require college degrees in an effort to limit the supply of workers in those field. For example, Oklahoma now requires a college degree to get a real estate license. Are you kidding me?!!!

And demand has increased through more state subsidies of education.

But the supply has not kept up with demand because the requirements for PhD's are pretty onerous. The authors mention that the average cpi increase for all services was 4% in the period 1947 to 2009. But since 1981 tuition has increased 7.1%. Room and board has increased 5.3% while cpi has increased just 3.2%. Increased demand probably explains the differences.

Dan Weber writes:

There is no evidence whatsoever that quality-adjusted room and board prices are going up any faster than their private sector counterparts.

Waitaminute -- the whole point was that college amenities were rising out-of-proportion with society.

I do wonder how college students who live fancy lives will manage to live with an entry level paycheck.

Fundamentally, the customers are not the ones who pay. I'm not talking about government subsidies, although that plays a role. But the student is the customer, and the parents are the ones who pay. There is a zero-sum race between colleges to be more attractive so the student likes it. The parents acquiesce, because the student is getting "a college education."

B.B. writes:

I don't buy it.

The cost disease argument does say that the real cost of tuitions will rise. That is not the issue.

The two professors may be economists but they are also professors with a stake in the status quo. Suppose the status quo is flawed and needs to change in ways that don't make life easier for professors?

In the past 30 years, tuition costs have increased 669%, in the CPI. Medical services have increased 444%. Personal care services costs rose 171%. The only component of the CPI that has increasd more than tuition is tobacco, because of surging excise taxes.

College is the new subprime. A bubble in costs, massive lending supported by the government, heavy consumer leverage, and a lack of realism on the ability of consumers to pay back the loans out of wages.

We need an objective study of college costs. What other sector of the private economy is essentially devoid of any gauge of productivity or quality? It is a nonprofit, cartelized system regulated and largely run by the government and captured by politically-connected special interests.

Anon writes:

If it helps with your story, I was "politely" told by some folks here at my university that by offering and teaching an extra upper level seminar to our students (without pay) was "setting a bad example."

And upon seeing what I try to do with our large intro level classes (such as figuring out a way to see and talk to students) I was sat down to make sure I was not "burning out."

Oh, and my 5 years in grad school did not one iota of good toward preparing me to teach what I teach - in fact I think it made it harder, aside from the credentialing issues.

Maybe I just like to whine, but in this case, I am not sure what advantage I would gain from it.

Hyena writes:

I think the article is correct: the various cost centers of the university system need to be picked apart and the statistical methodologies need to be examined to clear out artifacts. I'd like to see an actual number given to the tuition discounting programs. Rising college costs from price discrimination don't bother me as much, they don't imply the sort of apocalyptic scenario people fear.

I'd also like to see how this applies to the community college systems, which are generally far lower cost and rely heavily on less credentialed faculty. That would give us some indication, by comparing freshman and sophomore costs, of what (if any) inefficiencies lurk in the university system.

Douglass Holmes writes:

You are a hard man, Professor Kling, making us read the entire article. At least you didn't make us read the entire book!

Archibald & Feldman say, "the claim that government subsidy simply raises cost by pushing up demand is based on a misunderstanding of basic economics. The notion that rising demand leads to rising price relies on the assumption of an upward sloping supply curve. Yet there is a wealth of evidence that the long-run supply curve for a college or university is flat."

I'm glad I didn't take economics from them. It seems to me that if the demand rises and the supply doesn't, that the price will rise. Perhaps I need to go back over basic economics.

Floccina writes:

If we separated testing (credentialing/grading) and education what would happen to the price of an education. I would predict that it would plummet.

Floccina writes:

Below is from the state of Florida's web site:

2. What are the requirements to sit for the CPA Examination in Florida?

Current Rule Requirements:
Applicants sitting for the exam must have completed 120 semester or 160 quarter hours from an accredited college or university, unless one is using the provisions of the unaccredited institutions rule, as explained below. The accounting education program must include 24 semester or 36 quarter hours of upper division accounting to include coverage of taxation, auditing, cost/managerial, financial, and accounting information systems. The business education program must include 24 semester hours of upper division general business courses with some exceptions. One microeconomics, one macroeconomics, one statistics, one business law, and one introduction to computers course may be lower division. As part of the general business hours, applicants are required to have a total of six semester hours or eight quarter hours of business law courses, which must cover contracts, torts, and the Uniform Commercial Code. Excess upper division accounting courses may be used to meet the general business requirement

I ask why?

Chris Koresko writes:

@Floccina: That's a really good question!

I wonder what would happen if someone were to pretend to have the requisite coursework and take the test, pass it with flying colors, and then admit to never having taken any of the relevant classes.

Would the CPA credentials be revoked? It would be interesting to find out.

Not that I'm advocating the experiment...

infopractical writes:

Times are just beginning to change. I make much of my living writing math books. I don't even have a degree. Those that I write and other that my publisher publishes have become quite popular. And they cost substantially less than most textbooks.

Lower costs of textbooks won't solve the problem, but it's an example of one area of the problem that can be approached from outside of academia.

Jacob Oost writes:

It amazes me that some people actually manage to have a career teaching economics when their understanding of how an economy works is so poor. I consider them glorified statisticians, rather than real economic thinkers.

Government subsidies prevent the full cost of college from being passed on to the customer. As a result, people over-buy, which is to say too many people go to college. Colleges know they are guaranteed customers in one amount or another, so they are quite free to raise prices to what they think the "market" will bear, in this case what Uncle Sam will dole out.

And the accrediting bodies, while in a sense performing a useful role, by their monopoly nature inhibit competition.

Schepp writes:


Excellent perspective, I think that real competition on getting people passing the credentials with the least effort is central to reform. But I am surprised that this market force has not moved the ball toward the less expensive options.

Great blog and comments.

Walter Sobchak writes:

The cost push story simply does not make any sense.

If labor costs were going up that quickly, management would take steps to get them under control. Lots of steps are imaginable. To their credit, management has hired a lot of low priced temps to do the teaching that the tenured profs don't want to do. But lots more could be done.

Why in this day and age do professors write and deliver their own lectures? Why aren't the lectures recorded and replayed for the classes? Indeed, why have the large lecture halls when lectures could be watched on lap-top screens. Why are so many schools not utilizing their class rooms before 9:00 a.m. or on Fridays?

Another attack on the question. A full time student takes 4 courses per semester, two semesters per year, 16 weeks per semester, 3 hours per week. My guess is that 10 students per undergraduate course would be a low estimate of that number. Therefore each student costs about 40 hours of labor for classroom time. Let us double that for non-face time services (grading, paper marking, etc.) So 80 hours of service per student.

Remember that more than half of that time is provided by graduate students and adjuncts, who are very poorly paid. A $40,000 per yer tuition would seem to imply an average cost of $500/hr. Is that believable?

Comments for this entry have been closed
Return to top