Arnold Kling  


Some Motives Revealed... More on the Anti-Commons...

Salman Khan has interesting thoughts on education.

One root of the problem is the fact that the college degree is issued by the same institution that is in charge of setting, and enforcing, the standards of that credential, says Khan, who holds four degrees himself. This is tantamount to investment banks rating their own securities, he says. Meanwhile, the accrediting agencies that are in charge of making sure those "ratings" are legitimate do not currently focus on what students coming out of those institutions measurably know.

...What he does want to do is advocate for the creation and mainstreaming of credential-granting institutions that exist wholly separate ("decoupled," in Khan-speak) from the institutions (including his) that do the teaching.

As you know, I have advocated the same thing. This would be a fix for the evaluation problem that Bryan wrote about.

Comments and Sharing

COMMENTS (11 to date)
RPLong writes:

That is very definitely not the Salman Khan I was thinking of when I started reading this post... I did a double-take and then laughed.

Anonymous writes:

Katja Grace says the same thing here.

Publius The Lesser writes:

Arnold, Bloomberg has recently created a battery of tests to measure what new graduates in the financial industry actually know. You can find details here and searching for "bloomberg assessment test" or "bloomberg aptitutde test" will turn up some reviews.

il falcone writes:

Let's get this guy on Econtalk!

Ironman writes:


Well, as long as we're mistaking identities, this is how Arnold should have spelled his name in the post title....

Mike Rulle writes:

Some credentialism is very accurate, but most come from graduate education and specialties. There are also some credentials in the private sector that also indicate true expertise.

If we really are in a sort of bloated education bubble world (I have heard it so many times, I am beginning to doubt its veracity-----even as I believe it; think Kahneman), then one would think there is a ready arbitrage to be had. Companies can capture younger people by educating and training them with their own "private universities" specific to their own general skill requirements.

Certainly young employees would benefit----maybe less obvious companies would----but what do people learn as undergraduates applicable to jobs?

Having said this, I am still a believer in non technical education, such as history, literature, philosophy and the arts---would not like a world where this is systematically excluded.

Roger Sweeny writes:

If we really are in a sort of bloated education bubble world ..., then one would think there is a ready arbitrage to be had. Companies can capture younger people by educating and training them with their own "private universities" specific to their own general skill requirements.

I suspect there aren't enough facts and skills that are specific to that one company. Lots of what the younger people learn--both formally and "by doing"--will be useful elsewhere.

And Anglo-American law doesn't allow a company to force someone to work for them, even if the person has signed a contract to do so. It is thought of roughly as paid slavery.

One thing a company can do is pay for a worker's tuition, on the condition that the worker will pay it back if (s)he leaves in less than some period of time, say two years. Ironically, this feeds the education bubble.

Glen S. McGhee, FHEAP writes:

Khan is talking about the gatekeeping triad put in place by Congress with the 1992 "Program Integrity" amendments to the HEA. Without accreditation through an accreditor that is officially recognized by the Sec of Ed, schools do not have access to federal subsidies (i.e., student loan guarantees, grants, etc.) under Title IV.

The 1992 regulations can be found at 34 CFR 602, and there were then a dozen "minimum standards" that accreditors require their member schools to meet (see 602.16(a)(1)(i) to (x) for current standards).

Unfortunately, no objective standards were ever put in place, and the student loan debt problems of the late 1980s were never dealt with by the agencies whose members set up and enforce standards they approve. No surprise there.

As Khan says, this is just like investment banks rating their own securities.

The Spellings Commission (2005-2006) attempted to reform accreditation, but with no success. Colleges and universities closed ranks, and pushed back against what they saw as a massive threat of federal intrusion into their affairs.

Much of the problem goes back even earlier, to the 1979 Department of Education Organization Act, which included contradictory missions -- foremost was the expansion of access to higher ed and the new department was also given oversight of federal expeditures (quality control through NACIQI). Federal agencies have been split in two by conflicting interests, and this is no exception.

I am encouraged by the fact that Khan has targeted accreditation as problematic. You cannot have the same institutions involved with enforcing standards that those very same institutions have approved through their accrediting agencies. What we now have is regulation by accrediting guild, invariably in their favor. This is as it has always been (see Adam Smith, Thomas Hobbes, etc. on guilds).

Accreditation currently protects mediocre schools, and elite schools (such as Princeton) see it as irrelevant and meaningless. Robo-signing in the mortgage industry has certain similarities with accreditation, which is once every 10 years (!). I call it robo-accreditation.

NACIQI is meeting in Washington, DC, about this time, if you are interested in the review of accrediting agencies that is conducted by the Department of Ed every five years. The largest regional accreditor is up for review, and Congress has re-staffed NACIQI members, replacing long standing members. Take a look, and watch your tax dollars at work.

Floccina writes:

Yes a separation of education and grading might be beneficial. Like Deming has said cross missions are bad for organization. I think that to a large extent grading is squeezing out education. IMO a teachers should be 100% on their students' sides trying to help them learn as much as possible as efficiently as possible so they can get the credential that hey seek. That is at cross purpose with teachers' other job, that of preventing less qualified students for getting the credential.

Ghost of Christmas Past writes:

I agree that in theory separation of training and certification might be very useful. I'm not sure it would be so good in practice.

If a university is a kind of factory which produces graduates instead of simpler goods, then analyses which apply to ordinary commerce should offer some lessons for analyzing universities.

In the world of inanimate goods, there are sometimes independent testers and certifiers for products. However, most of the time customers rely on each supplier's own assurances of quality, weighted by the supplier's reputation--which can vary as its assurances prove more or less trustworthy. Only suppliers with good reputations can sustain high prices.

The reputation-based approach works for much commerce, why not for evaluating educational suppliers? Perhaps it does and that explains why Harvard graduates command higher starting salaries than Cal State Los Angeles graduates-- perhaps their different reputations are earned and deserved.

While independent testers seem to get employed for a few goods where the risk of failure is so extreme that customers cannot accept the way a supplier's reputation may lag behind its performance (e.g., independent labs testing very expensive avionics), previous attempts to expand that model very far have all ended in debacles: I give you S&P, Moody's, and Fitch.

The situation is filled with challenges. First is the problem of competition. If testers compete then consumers must judge their reputations, which is hardly easier than judging suppliers' reputations. On the other hand, given a monopoly tester (or an government-protected oligopoly like the NRSRO's), all the usual troubles occur and the testers will become untrustworthy.

(Competition to test university graduates would produce another problem: graduates would not like to sit many different tough tests on the same subjects. Employers might have to pay people to take tests even before hiring them. In theory employers could form the sort of consortium that universities have, to standardize tests in the same manner as the GRE, but then racial differences in performance would become apparent and the employers would all be charged with the Ultimate Crime.)

The second is the problem of corruption. Testers are in a position to take bribes from suppliers and experience shows they usually will. The only thing protecting standardized admissions tests like SAT/ACT/MCAT/GRE/LSAT/etc from corruption is that the "suppliers" (individual students supplying themselves) are too fragmented to offer convincing bribes. (Although they still cheat, as by employing shills.)

Lastly there is the problem of clannishness. One of the most important things any university offers is emotional affiliation. If we propose to reduce universities to interchangeable cram schools for independent exams, then all the millions of people who claim kinship with the fellow children of their Alma Mater will get angry and try to thwart us. Many people in positions of power are much more interested in loyalty than performance. They would much rather hire someone from their old school, or their old athletic conference, than someone with better test scores. Competition between employers might or might not punish such impulses-- in big firms, loyalty really might be worth more than raw ability since most employees never do anything special anyway.

Troy Camplin writes:

I'll be presenting on the decoupling of crediting from the colleges at an upcoming Fund for the Study of Spontaneous Orders colloquium on higher education as a spontaneous order. It is what we need to do if we want it to become a full-fledged spontaneous order.

Comments for this entry have been closed
Return to top