Bryan Caplan  

A Puzzle for Human Capital Extremists

Gun Control: Things I Didn't K... Long-term Care Insurance...
Suppose you believe whole-heartedly in the human capital theory of schooling.  Appearances notwithstanding, everything that students study gives them additional marketable skills - at least probabilistically.  I have a question for adherents of this position: Why do students rejoice whenever a teacher cancels class?

From a human capital standpoint, students' attitude is baffling.  They've paid good money to acquire additional skills.   Employers will judge them by the skills their teachers impart.  But when the students' agent, their teacher, unilaterally decides to teach them less without the slightest prospect of a refund, the students cheer.  How bizarre.  Would a contractor jump for joy when his roofers tell him they're taking short cuts on the shingles in order to go drinking?

A behavioral economist could say that the students are myopic; they're overly focused on short-run fun rather than long-run success.  But much of the appeal of human capital extremism is that it's a fully rational story of educational choice.  Once you admit that students are myopic, who knows what else you'll have to admit?

The signaling model of education has an even easier story: Students want not knowledge, but certification.  Future employers only see your grade and diploma - and the less a professor teaches, the less students have to learn to get the grade and diploma they want. 

So, human capital extremists, what gives?  Do you deny that canceling class makes students happy?  If not, how can you reconcile your theory with the facts?

Comments and Sharing

COMMENTS (32 to date)
Deman writes:

Holy Strawman - going by personal experience, students rejoice because they are not made to sit for another hour stuck to their desks rather than having fun, talking to their friends or looking at the opposite sex. No need to intellectualise this simple human emotion to fit ideological/competing theories of economic/human behaviour.

Robinson writes:

I agree with you wholeheartedly about the signalling model of education, but I think your post is a strawman.

You could ask a simpler but equivalent question- why do K-12 schools mandate attendance? An extremist who believed in full rationality would say that students would attend if school were optional, as they would recognize the benefits of free education. As nearly everyone along the educational belief spectrum (perhaps especially those who believe in the human capital model) says that such education should be mandatory, everyone has already conceded student irrationality pretty thoroughly.

Furthermore, the "happy when classes are cancelled" attitude does diminish with age, which is what you might expect if it were caused by student irrationality. K-12 students will almost universally rejoice when class is cancelled. College students (in my experience) celebrate less. Most of my fellow graduate students are disappointed when a class is cancelled, and I'm guessing those in night school or adult education typically appreciate every class they get.

The appeal of the human capital model isn't that it explains student's decisions as entirely rational- it's that it a policy prescription for paternalistic education. Put another way, they're not telling students "Why wouldn't you want to go to school and raise your capital?" They're saying "You'll thank me when you're older."

Blake Johnson writes:

"College students (in my experience) celebrate less. Most of my fellow graduate students are disappointed when a class is cancelled, and I'm guessing those in night school or adult education typically appreciate every class they get."

I hate to answer an anecdote with another anecdote, but this has not been my experience, or that of most of the people I know. Especially those in night school or adult education, for whom commuting to school is an added inconvenience on top of their regular work schedules.

Even Tyler Cowen, who is a strong proponent of the human capital theory, starts his I/O class by saying something along the lines of "You should be angry that you have to spend time in this class." He says this is because time spent in class is time you can't be working on your own papers, doing your own research etc. I've always meant to ask him how he squares that statement with his belief in the human capital theory. If we are really learning on the job skills in class, then we should not be unhappy to have time with a knowledgeable professor helping us to attain those skills.

Vipul Naik writes:

Are you confusing averages with margins? Students may well want to attend class on average, but think that the marginal gain from attending one more class is low. Or rather, that if a particular class is canceled, the marginal material that doesn't get covered as a result of fewer classes isn't worth the effort.

That may also explain why they may not choose to miss class _themselves_ but like it when the instructor cancels class because they know that with a canceled class, they only "miss" the least important material which the instructor isn't covering because of the canceled class.

In other words, students may think that all the *important* stuff will probably get covered even with a few classes canceled here and there.

Just an alternative theory. May be it's not true.

Bryan Willman writes:

Why do students celebrate snow days, even though everybody tells them this means they will be in school during nice weather in June rather than during crappy weather in winter?

Disgard this strawman and ask a different question:

Why do students take on debt to attend college, and then skip classes in their majors to go drinking?

Brian writes:

My experience in the sort of class where adults are trying to learn real skills is very much the opposite of what Bryan describes.

I have attended, and paid for along with my fellow students, classes in statistics, foreign languages, dancing, and nonprofit administration. We all expected to have to demonstrate real skills in the real world in statistics for jobs functions, fluency in our languages, competence to impress our dance partners, and some kind of plausible paperwork for the IRS.

Whenever any of those classes were delayed, cancelled, or otherwise cut off, we were plenty agitated. Nobody was happy about it.

Of course, back in college when all we wanted was a degree and a steady enough average to keep the loans and beer flowing, the attitude was different.

Brian writes:

Another one: law students seem as happy as other college kids when their classes are cancelled. How do they react when their bar review classes are cancelled? They're angry, aren't they?

Phil writes:

What if you cancelled ALL their classes, and told them that they'd just be given an A? Would they be happy? I bet they wouldn't.

My guess is they're happy when class is cancelled because they realize you'll just squeeze the material into fewer classroom sessions, and they realize that in that light, the marginal benefit of that one extra class is small.

That's something that wouldn't be true if you cancelled 20 classes, or all classes.

Blake Johnson writes:

"What if you cancelled ALL their classes, and told them that they'd just be given an A? Would they be happy? I bet they wouldn't."

I don't know that I think this is true. If you spend time with the average college student, most of them spend a lot of effort via websites like ratemyprofessor to find the absolute easiest professors and/or classes possible so they can spend minimal effort to get a good grade. If you offered these students a free A for some class where they were intending to spend a minimal amount of effort anyway, I think a lot of them would jump at the chance.

I think this would be especially true for any required courses outside of their major. If students could get free A's instead of having to take low level math, history, fine arts classes etc., I think a majority of them would do it.

Robert Kwasny writes:

Students may rejoice when a single class is canceled. But the real question is: would they rejoice if a whole course was cancelled and yet they would still be given credit for it?

My guess is, on average, not.

akasavani writes:

Sorry but this post is completely wrong. Say you are a student in a really I mean really bad school/college, and the teacher cancels the class, you don't rejoice. You know why, because cancelling a class, having easy classes, having less than qualified professors, easy exams etc are the things you expect from a bad school. Students may rejoice if their class is cancelled only when they know in their gut that it's an exception to the rule. May be you haven't seen bad schools. I'm from India and I know how bad schools can be. Most of this analysis is based on the fact that colleges are reasonably good.

Liam McDonald writes:

Because they are able to slack off without it being "their fault". Regardless if they've paid good money to attend. The money they have invested can not be seen or used for anything else and so to them it's spent and gone. Getting some free time is always enjoyable when you're young

Eelco Hoogendoorn writes:

Pretty much fully agreed with Bryan.

Not sure what school up until university is about; perhaps some signalling, but mostly daycare I suppose. Either way, few things could make me happier than it cancellation.

During my bachelors I was still rather ambivalent, since the choice between classes appeared to be between various forms of skulldrudgery. But during my masters program, where every course was an individual choice, not aimed at getting a grade, but learning something, cancelled classes were no longer the holy grail. Not that I would have cared much either considering my individual learning style; I was mostly interested in the professors stamp of approval of my skill rather than his attempts at conveying it. The cancellation of an exam or its postponement to next year would certainly have annoyed me a great deal, as it sometimes did for classes that managed to attract only two or three students that year.

In my experience, at any level of education, sitting in class is largely an excercise in appeasing your educator. Either to avoid direct punishment as in highschool, but also in university professors do not like the idea of students not needing their direct instruction. Ive sometimes been sent away by professors when walking in the first time for the exam; they would protest that I couldnt possibly be making a serious attempt without their instruction, and that I would waste their time having them correct my half baked attempt. Yeah right. Or theyd allow you to make the exam but mark you down for not solving the problem by their preferred route; even if yours was arguably more elegant.

In those situations, cancellations were still celebrated, despite a strong interest in the human capital afforded by the course. But those are the exceptions, I think overall, Bryans model explains 95% of the observed variance.

RPLong writes:

Three words for you, Caplan:

Disutility of labor.

tribsantos writes:

Now I'm no human capital extremist, but I don't think this particular thought experiment helps in undermining that theory.
People could rejoice because they are myopic in the short term, rational in the long term, and so make commitments to make them more rational.
Going to college instead of studying on your own is a form of commitment. Of course I can learn from textbooks, but most people just don't have the self-control to wake up everyday and do math exercises. They need tests, social pressure, that kind of stuff. It's the "lock your own fridge" kind of thought.
I think you would agree that there's not much signaling in going to the gym. People workout because they actually want to look good and, if anything, they would be embarrassed if they had a diploma for going to the gym and were really fat. Still, I think that many people who workout would be happy if their gym was closed for one day

Ak Mike writes:

Why do people eat candy? It makes them fat, gives them diabetes, and rots their teeth. But they eat it anyway.

Roger Sweeny writes:

Why do students take on debt to attend college, and then skip classes in their majors to go drinking?

Because, says the signalling model, they figure they're still going to pass, and one more night of drinking isn't going to lower their grade.

BTW: I'm a high school teacher and I've heard some students say they don't want snow days because they don't want extra days in June. Admittedly, they're a minority.

Clay writes:

At a voluntary gym workout, people are happy when there is less minutes/miles left to go. They value the exercise, but usually don't like the physical pain and discomfort associated with that. Students often value the course material, but celebrate some relief from the high stress studying/learning activities that go along with it. Bryan is still mostly right, even good students are wired to value the grade and the class credit and degree more than the hard to measure abstract nature of actual learning. At a gym workout, there is no meaningful credentials and your health and physique is more directly tied to your exercise efforts.

Rachel writes:

[Comment removed for supplying false email address. Email the to request restoring this comment and your comment privileges. This is your final warning. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Steve Roth writes:

"Do you deny that canceling class makes students happy?"

It didn't make me happy. It pissed me off.

But then I wasn't like most students. I figured I was hiring the professors to deliver, not working for the professors and being paid in grades.

The latter was certainly the attitude of many students in the second-tier public university where I got my BA.

But I'm sure there are many in many schools who have attitudes at least similar to mine.

The "paid in grades" attitude not really true in my experience at the graduate level in a second-tier private university, or certainly not so much.

Monica writes:

I think even most believers in the signaling model would agree that working in a job relevant to one's long term career goals increases human capital because they learn skills far more relevant than they do in school. Still, most adults are happy to hear the news of a snow day (especially if they still get paid for it) even though it may very slightly decreases their long-term earnings potential. I think the combination of the magnitude of the decrease and the valuation of the present over the future makes both kids and adults happy to have an extra day off.

Glen S. McGhee writes:

Not strawman argument. Just heard that with dual European apprenticeship programs, kids will literally bail out school windows to get to the shop. At 15, they can actually be earning money in their trade. They are on the way.

The refusal of human capitalists to engage in comparative analysis undercuts their credibility. Hal Hansen's work distinguishes between US-credentialism and Germany's more human capital-based education systems.

JLA writes:

I have a challenge for you, Bryan:

This medical school teacher never showed up to class and gave all her kids an 'A'. Students complained that she never showed up to class, and she eventually resigned.

If education is mostly about signalling, why would the students complain about getting a free 'A'?

Robert Kwasny writes:


Though I disagree with Bryan's post, I assume that he meant college or high school education. Medical School also has a signaling component (Harvard Med will open many more doors than GWU Med), but med students do acquire many important skills that they will use regularly in their practice. And you just cannot say the same about English literature college major.

Deman writes:

The comments seem to have broadened to include everything from med school to graduate students etc. I assume the implicit assumption of the post was limited to schools. (Words: schooling/teacher/schools). The difference is important as post school educational activities require significant self-investments (loans etc) from the student relative to school.

Luke G. writes:

Some of you are arguing that students rejoice because it’s on the margin—just one day won’t hurt the class overall, but students would be concerned if they missed too many days.

So what about a similar puzzler?

A tenured college professor is in his last semester before retiring. He tells his students on the first day that because he’s tired of fighting with students over grades he’s giving everyone an A. A student could skip every single test, never show up to class, not do any work, and still get an A. Their grade would be unchallenged by the administration; it would be a perfectly valid course in every way.

Now, he’s still going to teach his class as he’s always taught it—lectures, tests, essays. He’ll follow the syllabus exactly. It’s just that grades won’t matter.

What would attendance be like for students over the course of the semester? How many would voluntarily take the tests and write the essays? Would it matter if it is a general ed requirement or a specialized course for a major?

RPLong writes:

The more I think about this, the more ridiculous the "debate" about these two theories seems to be.

Do any of us really think we live in a world where education is EITHER an investment in human capital OR an investment in market signalling?

It seems to me that education is clearly both. You could model it something like a classic Cobb-Douglas function: Y = [S^f] * [K^h]

Y = expected payoff of education
S = signalling returns
K = human capital returns
f and h = the extent to which each is important for a given course of study

Now obviously for people getting a certification in coding Javascript at a local vocational school f is nearly zero and h is very high.

For people enrolled at Harvard, f is extremely high and if h is greater than zero, well that's just gravy.

Clearly, every degree comes with some combination of market signalling and skills. An ideal education would be one for which f and h are both large positive numbers, but the reality of the situation is that for most of us with a generic business degree from a generic state college, our education consisted more of a dull signal than a large amount of human capital.

The question isn't "is education human capital or signalling," the question is, "for any given degree at any given university, to what extent is that an investment in a signal and to what extent is it an investment in human capital?"

The answer could vary between schools, degree programs, and potential employers. There's no reason to believe we have to choose between paradigms. They are not mutually exclusive.

Mark Brophy writes:

I learned to speak Spanish in San Francisco 20 years ago by taking a couple of courses at night. Many of the students were cooks, doctors, lawyers and others who needed to speak Spanish to improve their job performance, while the others were similarly motivated to learn the subject. The professors considered it a joy to work with us, and their day job a pain because they were teaching college students who were solely interested in earning a credential, not in learning Spanish. The government ought to stop subsidizing mediocrity while pretending to increase human capital.

The day students would rejoice if the teacher cancelled class, but the night students would resent being ripped off.

roystgnr writes:

Let me advance a hypothesis by answering that question with a question: isn't it a fortuitous coincidence that every subject of human knowledge can best be taught in precisely an integral number of semesters of instructional time? Algebra takes four semesters to teach to high schoolers, Statics takes one semester to teach to undergraduates, etc.? Isn't it convenient that there aren't any spring classes that finish up at the end of April because they only have .8 semesters worth of material to cover, or any year-long classes that have to stretch into July to cover 2.4 semesters worth?

Those 2.4 semester classes might be divided into three 0.8 semester classes, naturally, but that just takes care of the too-long subjects. What about the too-short ones? Do they not exist?

Clearly they must exist, so they must just get padded out somehow.

For instance, you may spend 80% of your time teaching the core curriculum, 10% of your time teaching interesting or useful topics that could have been skipped if necessary, and 10% of your time teaching the least-useless material you can find to fill the remaining space. In that case, if you have to cancel a class you're going to cut out the least worthwhile part of the syllabus you can, and it's a net win for the students.

Alternatively, you may just teach that .8 semesters of critical material, but teach them more slowly, to give more of your slower students an easier time keeping up. In that case, cancelling a class means you just have to go faster in the remaining classes; a bad thing for the 5th percentile kids who were benefiting from the extra time, but a win for everyone else.

This hypothesis may be completely wrong, and in fact I'd say the evidence points to "students are irrational" (ever seen a class canceled and it's the *slowest* students who are complaining rather than vice-versa?) but it's at least possible that students sometimes might rationally prefer cancelled classes too.

PL Kester writes:

An interesting discussion with a flawed central assumption: Most college students do not, themselves, pay for their education. Therefore, their reaction to receiving "less" education is not grounded in economics.

I put myself through college and always felt that I was not getting my money's worth when a class was cancelled. I gave my children the opportunity to put themselves at least partway through college for this very reason: to have some "skin in the game." Consequently, they also made economics part of their consideration when choosing a major and won't be showing up at any Occupy [your city's name here] rallies.

Glen S. McGhee writes:

Let me add another wrinkle -- signaling theory has limited explanatory utility outside corporate culture, that is, mass HR.

The hiring practices for small and medium businesses that do not rely heavily upon human resources procedures and the HR organizational culture itself (dating from the 1920s) do not demonstrate an appreciable sensitivity to signaling.

So does rural hiring, as opposed to hiring in urban centers.

Rather, in these cases, social networking topologies and SES give a clearer description of what is going on around the country, and around the world.

Bob Murphy writes:

How many kids would rejoice if their scheduled dental surgery got postponed? Does Bryan Caplan believe in a signalling model of root canals?

Comments for this entry have been closed
Return to top