Bryan Caplan  

Stably Wasteful: Why New Tech Won't Gut Higher Education

PRINT
Will Higher Education Tip?... Another Post on EFC...
If the human capital model of education were true, educators should be worried.  Modern information technology makes it possible to teach skills for a fraction of the traditional cost.  If imparting skills were the main function of schooling, higher education will soon, as Arnold suggests, go the way of Borders.

Unfortunately for the world, but fortunately for me, the human capital model is greatly overrated.  Education is not primarily about teaching concrete skills.  It's a stably wasteful way to sort people according to their intelligence, conscientiousness, conformity, etc. 

So what happens when an innovator claims to have a cheaper, easier substitute for traditional education?  The lazy and the weird gravitate to Cheap Easy U like moths to the flame.  As a result, employers correctly infer that graduates of Cheap Easy U are sub-par - and Cheap Easy U captures, at best, a niche market.  A sustainable business model, perhaps - but no real threat to the Expensive Painful Universities that blanket the land.

If our education system is going to improve, our salvation won't be low-cost alternatives to what we've got.  Our salvation will be education budgets so austere that middle class kids can no longer afford to finish four-year degrees - and therefore no longer need four-year degrees to convince employers to give them a chance.

P.S. Two years ago Alex Tabarrok expressed similar hopes/fears about the ability of technology to gut higher education.  I replied:
If [you] were right, then videotape would have put college professors out of business thirty years ago!
Then I offered Alex a bet:
I bet at even odds that 10 years from now, the fraction of American 18-24 year-olds enrolled in traditional four-year colleges will be no more than 10% (not 10 percentage-points!) lower than it is today.
As far as I remember, Alex didn't take the bait.  How about you, Arnold?


Comments and Sharing





COMMENTS (26 to date)
joeftansey writes:

What about Cheap Difficult U?

I don't see why "ability to pay" has to be part of the signalling.

Joel writes:

But modern technology also enables people to signal their intelligence (and/or conscientiousness, etc...) at a fraction of the cost (e.g. by keeping a blog, contributing to open-source projects, publishing an e-book, and so on).

Videotaped lectures did no such thing.

blink writes:

I agree that signaling via "Expensive Painful Universities" is part of the story, but traditional universities have even more going for them. Perhaps most importantly, they bring young people together, something video lectures cannot do. Then there is signaling via affiliation with prestigious academics (ala Robin Hanson) which works much better in person. Also, for many, college is more like an extended vacation than some "painful" ordeal and even first-rate video lectures cannot compete with that.

Bryan Willman writes:

Sadly, I think Bryan C. is right.

First, colleges still offer special physical resources in the forms of science labs, actual experience with biology and patients, and so on. No online course will ever completely replace that.

Second, a key function of college is to give smart young people a respectable and accepted place to become 4 to 6 years older, so as to be practically hireable by most of corporate america.

(Think about it - your wonder student finishes high school at 15 and web college at 18, and is as mature as many 30 year olds. But he still can't rent a car some places because he's less than 25. She can't buy a drink, or work in a bar, regardless of how suitable she is, because she's not 21. Can they sign enforceable contracts in every state? And many positions in business in america require renting cars, socializing with customers in places that serve alcohol, etc.)

One of the most profound "credentials" in our credentialist society is Age.

Third, while colleges may be "filtering and signalling" for some majors (perhaps economics?) that doesn't apply nearly so much to engineering, pre-med, and so on. It doesn't even apply to computer science - which one supposes could be well taught online, but historically has not been.

This also means, by the way, that young peole won't ever "not need to go to college", though they may call the transition something else....

andy writes:

It doesn't even apply to computer science - which one supposes could be well taught online, but historically has not been.

CS is quite interesting. First, CS is not the same as "programming"; and aditionally, there are many types of 'programming' (if you do computer graphics/phyics models/etc., you definitely need some background in math. some background in logic (math) is required for some algorithms. 90% of programming does not need CS nor math background). From my experience, the good programmers did not learn programming on university. They already have had several years of experience when they entered the university. I know several people who didn't go to university/didn't finish, yet they are perfectly well established in the profession. And yes, you can learn even computer graphics by doing open-source projects. In my opinion university didn't teach that much these people. I skipped first 3 years of CS classes and in terms of practical knowledge, vast majority of students were significantly worse than me (the only thing I did was to read a book issued 20 years ago along with some practical experience).

Computer-science has not been 'taught online'; people massively learn computer science online (if by 'computer science' you mean 'practically useful' - Turing machines/lambda calculus/computability are not quite applicable to practice, yet this is mostly what is considered 'computer science').

I would say the opposite - IT has a great history of successful people who did not learn it at the university. And I don't mean only Steve Jobs and Bill Gates.

Randy writes:

Bryan,

I expect that the Reformation will turn out to be the applicable historical reference. Ala Martin Luther, the change will arise from within.

Joe Cushing writes:

As a frugal person I find signaling costs irritating. I find the story of how diamond rings because a part of our lives disgusting. I feel like millions of people were duped into handing over billions of dollars to a diamond monopoly. Many men probably feel like I feel but also feel like they are forced to comply with this cultural norm. Women of course enforce it. I'm not sure why either. I think this topic on signaling is a possible explanation. Maybe if people were aware of it, signaling would have less of a power over us.

I'm agree with the other joe. What about cheap difficult U? When I add up the cost of my MS degree; multiply it by other classmates in the class room; and subtract a guess for building costs, I get a lot more money than the professor gets paid. That makes me wonder where the money is going and why they are still calling me today, asking me for more money.

Chris Koresko writes:

Bryan Caplan: I bet at even odds that 10 years from now, the fraction of American 18-24 year-olds enrolled in traditional four-year colleges will be no more than 10% (not 10 percentage-points!) lower than it is today.

I'd take that bet. The rapid rise of education costs will soon price it out of reach of both individual students and the governments that subsidize it. Many of those costs are mandated "diversity coordinators" and such, so it's not likely they'll go away. Professors will fight tooth and nail to prevent "austerity" measures which include laying off the superfluous staff. My guess is that the cost issue will force the student population to shrink at least 10% in 10 years.

lewis writes:

A question for bryan:
why do big companies pay for existing employees to get MBA's? Employers are actually making it easier for employees to leave for higher paying positions elsewhere, by giving them the signal for free; and according to the signaling model, employers are not actually making their employees more productive.

I believe in signaling but this is puzzling.

Blake writes:

MIT, Harvard, Stanford, Oxford. They all give away free online classes in a several disciplines. These are real courses, the same ones all students take. Universities are happy to give away their supposed product (the teaching) and hold onto the credential.

Although lots of schools (Harvard & LSE included) try to capitalize both ways by offering lesser degrees through online and extension schools. They cost less in tuition because you get less of a degree (BLS instead of a BA or BS) not because the teaching is any different.

David R. Henderson writes:

@Bryan,
I might take that bet. But I don’t believe on betting on a change in a number without having a good idea of what the number is now. Please tell me the number and link to your source. That way we can agree on the source. Then I’ll tell you if I’ll take the bet.

PrometheeFeu writes:

@Bryan Williams:

If by computer science you mean complexity and information theory, advanced mathematics etc, then yes. You are correct. If you mean programming of the type that is done by the vase majority of programmers, it has been taught online for a very long time. I am a software engineer at a top silicon valley firm and while I did minor in CS, I very rarely use anything I ever learned in a classroom. On the other hand, I use every day lots of skills that I learned through reading online documentation, books, practice and tutorials. Even when it comes to topics where I did take a class, most of what I use comes from independent research when I needed to solve a specific problem. Here is a great illustration of this: http://xkcd.com/519/.

Dan Carroll writes:

In the private sector, disruptive technology doesn't usually arise from within established industry - the establishment first denies, then resists, then experiments, then panics (often through M&A). Companies that successfully navigate disruption often do so by diversifying into other businesses and leveraging a brand (i.e., IBM).

In the case of education, the establishment is partially owned/controlled by government. The motivated innovators are home schoolers, private schools, and online universities. The barrier to entry is certification and signaling. For secondary schools, the SAT/ACT act as a partial leveler because of the heterogenity of quality of high schools. For 4-year colleges, it is the brand identity that matters (Harvard versus po-dunk U), and there are no standardized exit exams.

The difference between the video and online is that online is instant, interactive, takes up no physical space (thus the library is effectively infinite), and provides almost zero cost distribution.

For disciplines that require labs and live experiments, obviously those can't be done in the living room. But one doesn't need to travel to the other side of the country, either.

The interactivity of online actually provides a better learning environment then the current 4-year vacation that we call the university. The interaction with other students can be a benefit, but for most offers more of a social benefit then a learning benefit.

For places like Harvard, the alumni network is very important. That is a lot less true for the rest of us. That doesn't mean that Harvard can't innovate and look for ways to leverage the brand. (It doesn't necessarily mean that they will, see paragraph 1 above.)

I won't take the bet because I can't predict the degree of resistence, legislative interference, and the rapidity of cultural change. It's not surprising, however, that the low end has adopted the technology first - they have nothing to lose and everything to gain by challenging the established signaling model of education.

I've long predicted that telecommuting will partially replace physical communiting. It is, slowly, but there is still significant cultural and physical adjustments (optimized houses) that need to take place.

Matt P writes:

Lewis, I can think of a few reasons:
1) Educational benefits provided to employees are not subject to income or payroll taxes, so the cost of paying for an MBA to the company is much less than the equivalent in salary.
2) Employer-provided education is often accompanied by a requirement to stay with the company for some number of years.
3) The credentials of a company's employees are also a signal to that company's customers. This is especially true for consulting and professional service firms.

Bryan Willman writes:

PrometheeFeu, andy,

Yes, of course, you can have a long and happy career as a programmer without needing to know the details of the Halting Theorem or the Lambda Calculus.

But at University (27+ years ago) I did learn about structure, about large systems that have to work for more than one weekend, and so on. Of course, I then got 2 decades of advanced on-the-job training in those topics that literally never ended.

But you could well counter that working on open source projects, or other collaborations on things that need to be ongoing. So maybe "industrial programming" will indeed cease to be much of university topic.

Dan Carroll - telecommuting has a lot of issues even in companies that are hyper technical. (Notably Microsoft) It turns out that being in the same physical office building means there are a certain number of random encounters among people, that turn out to be important. (Meeting the Senior VP in the elevator....) Telecommuting works great for the formally set up meetings, and for the side conversations that directly attach to them. But not for the "I saw Jim in the elevator and brought it up and he agreed..." things, which turn out to be a big deal in many orgs.

There's also the issue of the social purposes of work - some people need to get out of their house...

Evan writes:
Perhaps most importantly, they bring young people together, something video lectures cannot do. Then there is signaling via affiliation with prestigious academics (ala Robin Hanson) which works much better in person. Also, for many, college is more like an extended vacation than some "painful" ordeal and even first-rate video lectures cannot compete with that.
Agreed. I enjoyed going to college. I found attending college and having a part-time job to be far more enjoyable than having a full time job. And the social aspects of college are certainly important, it's a lot easier to find people you have something in common with when they're all concentrated like that, and when you're going to be forced to interact with them.

As with signalling, there is likely a more efficient way to obtain those goods that college provides. But I don't think college will go away until the "signalling," "socializing," and "vacation" aspects are all accounted for.

@Joe Cushing

As a frugal person I find signaling costs irritating. I find the story of how diamond rings because a part of our lives disgusting. I feel like millions of people were duped into handing over billions of dollars to a diamond monopoly. Many men probably feel like I feel but also feel like they are forced to comply with this cultural norm. Women of course enforce it. I'm not sure why either. I think this topic on signaling is a possible explanation. Maybe if people were aware of it, signaling would have less of a power over us.
The reason women enforce it is because, like it or not, money is the unit by which we measure caring. Showing someone you are willing to spend money on them is a very efficient way to show that you care. If diamonds were not controlled by a cartel (and hence were cheap) women would probably switch to rubies or sapphires. And while I may wish women demanded men spend their money on something a little more useful than jewelry to show they care, I cannot in principle object to women wanting men to prove they care by spending money.

That being said, I admit that in principle it may sometimes be possible to "show you care" adequately enough in other areas that the person you're signalling to is less demanding of money-based signalling than normal. I believe this because that's what happened to me when I got in a relationship.

Arnold Kling writes:

David's point about agreeing on a data source is important.

I think it is unlikely that the situation will tip within the next 10 years. On the other hand, when the tipping point hits, my guess is that we will see a much larger loss in market share than 10 percent for traditional colleges.

So in terms of a 50-50 bet, I would prefer a longer time frame and be willing to accept a more stringent definition of tipping.

Luke writes:

I wish someone would convince the teachers' unions that online education is not a threat:

Union Protests UC Online Classes Due to Potential Job Cuts

Silas Barta writes:

I agree with joeftansey that the plausible alternative is a cheap difficult university that saves on monetary costs by making more hellish. I proposed a business model long these lines in a recent blog post (and a comment on this site).

One thing to remember is that the market for Cheap Difficult U is already mostly fulfilled by the military -- it has low (negative) monetary cost, but is much more draining than college in other ways, and not surprisingly, many employers view it as a substitute.

chandran writes:

http://www.telegraph.co.uk/education/universityeducation/8846236/Universities-see-40pc-fall-in-soft-subject-applications.html

Tom West writes:

Our salvation will be education budgets so austere that middle class kids can no longer afford to finish four-year degrees - and therefore no longer need four-year degrees to convince employers to give them a chance.

Except for the ten year period necessary for social expectations of employers to adjust to reality.

During that period, a generation of students will have their careers sidelined permanently by being unable to obtain the good jobs they may deserve because they lack the credentials of a university education.

Also, there's the possibility that employers will try to find more foreign workers to fulfill the needs for university graduates. Signals do not die easily - human beings love easily measured metrics.

Dan Carroll writes:

Bryan Willman - my telecommuting comment was a little more offhand then I intended. Social interaction greases the wheels of organizations more than I implied, although that is more true in some disciplines than others. Also, there are potential offsets with interactive technology and cultural changes already in the making, as the young are far more comfortable with social interaction using technology than the old (Facebook, etc.).

Telecommuting will reduce and redistribute physical commuting rather than replace it completely. Outlying suburban bedroom communities have sprung up to accomodate it.

Walter Sobchak writes:

Brian: to me the real question is whether it has to be this expensive.

I have long believed that the reason why tuition goes up is that there so much money to pay it.

I think that if we took the federal money, and much of the state money out of the system, tuitions would be forced down.

Of course universities would have to cut costs. College administrators spend could spend the time that they now spend dialing for dollars, figuring out how to cut costs. Maybe they should start with their own ranks and their own salaries. But, please don't stop there. Unproductive tenured faculty, unfunded research projects, departments with more faculty and staff than students, faddish programs that are no longer faddish, faddish programs that are still faddish, gourmet food in the dorms, gold plated gymnasiums, and a thousand other things could be slashed.

Walter Sobchak writes:

Bryan Willman wrote: "First, colleges still offer special physical resources in the forms ... actual experience with biology"

Mostly human reproductive biology by braille.

Colleges are also marriage markets. Parents want their kids to stop hanging out in biker bars and get with a better class of POTOS.

davis writes:

tough one. if one wants a divorce then it should be granted. if neither want a divorce then stick to those vows.

Comments for this entry have been closed
Return to top