Bryan Caplan  

Tyler's Embarrassing Question and Major Concession

A Shortage of Safe Assets?... Henderson on Stossel...
Tyler's has renewed our debate about signaling (see here for earlier rounds):
It is an embarrassing question for signaling models to ask: with what lag do employers get a good estimate of a worker's marginal product?  If you say "it takes 37 years" it is hard to account for all the recent changes in wage rates in response to technology, as discussed above.
Not embarrassing at all, I'd say.  There's a world of difference between knowing that skilled workers are more valuable than they used to be and knowing how skilled any particular worker is.  If I'm running a business and technology becomes more complex, I know I need more skilled workers.  But that doesn't make it any easier for me to measure applicants' skill.  A natural response to my imperfect information is to hire more college grads, knowing that they're more skilled on average.  If a lot of employers do the same, college grads' incomes skyrocket - whether or not college causes their skill to rise.

Tyler goes on to bring a tear to my eye with this major concession:

There is a way to nest signaling models within human capital models, rather than viewing them as competing hypotheses.  Using matching theories, let's say employers learn the quality of workers they have, but find it hard to estimate the quality of workers they don't have.  IV-elevated workers can't fool the market/the employer for very long, and so their high pecuniary returns from education really do measure productivity gains.  Nonetheless there can be undervalued "diamonds in the rough."  [Link mine. - B.C.]  Think of them as geniuses, or at least good workers, who hate getting the education.

From the point of view of these students (or dropouts, as the case may be), the signaling model will appear to be true.  They will resent the education and they won't need the education.  If it is costly enough to sample worker quality from the "outsiders bin," it will remain an equilibrium that a degree is required to get the job, at least provided workers of this kind are not too numerous.  If there were "lots and lots" of such workers, more employers would scrounge around in the outsider's bin.  In other words, the anecdotal evidence for signaling fits into a broader model precisely because such cases aren't too common.

My only disagreements with this passage:

1. "IV-elevated workers can't fool the market/the employer for very long, and so their high pecuniary returns from education really do measure productivity gains."  There's no need to say "productivity gains."  It could just as easily be "pre-existing higher productivity."

2. "From the point of view of these students (or dropouts, as the case may be), the signaling model will appear to be true."  Will "appear to be true" from "the point of view of these students"?!  No, is true about these students from anyone's point of view.

Tyler might retort that my failure to disagree with his last sentence is a major concession from me.  But it's been my position all along.  The reason why you have to go to college to get most good jobs is because a supermajority of qualified applicants for good jobs go to college.  In my dream world, the subsidies for college would dry up, most of the students would drop out, and employers would happily start "scrounging around in the outsider's bin."

COMMENTS (13 to date)
Brian Moore writes:

Regarding item #2, is there any evidence that high profile dropouts like Bill Gates caused any "scrounging around in the outsider's bin."? (not rhetorical)

Brian Moore writes:

Also, not to be annoying, and you can delete this comment after, but your first link does not work for me.

Tyler Cowen writes:

I would ask you to tell us "with what lag can employers detect workers' marginal products?" The point of my initial argument is that no answer to that question can solve the basic puzzles of the data. You're selecting one piece of the data (and not addressing how to tie the two major pieces together), and also not answering the question. I also still don't understand in your account how individual workers ever get raises or pay cuts.

J Storrs Hall writes:

It seems to me that nobody can say much with certainty about the issue until they've quantified the hysteresis and positive feedback of success in the job market. If you divided the applicants at random and gave half of them degrees and sent them all into the market, how much better would the degreed half be doing 10 years later, assuming everything else, including ability and actual education, was equal?

andy writes:

I just wonder how much is the education worth outside of government-licensed professions. Classical music..nobody asks you for a degree; however it's very easy to asses quality of job applicant, not exactly easy to asses quality of employee.

IT - very hard to asses quality of a job applicant, not easy to asses the quality of an employee. Although a programmer, it seems to me that you can get plenty experience to signal otherwise; my guess is that unless HR department is trying to screen individuals based on certifications, you will not meet many obstacles.

There is an immense range of positions ranging from manager positions, to 'do/manage some work', where it's very hard to conclude that the education has anything to do with what the person is going to do. Yet these positions are often screened by HR departments that often do require degrees.

It seems to me that the majority of positions has absolutely nothing to do with what the person did study. Academic carreer. Statisticians doing the actual math. Not medical doctors, not programmers, not accountants, I mean, they won't let you build a bridge without a lot of experience and oversight. It is very likely that if they let you do some actual work after finishing college, it is most likely because you were already doing something like it before.

Bryan Caplan writes:

Thanks for pointing out the broken link, Brian.

joeftansey writes:

"with what lag can employers detect workers' marginal products?"

Depends what you mean by "detect".

The signaling model posits that employers can immediately and successfully "infer" workers' marginal products, and this is really all you need in order to explain wages/hiring practices.

As for how long it takes workers to make a physical proof of their marginal products? This will vary by industry.

"I also still don't understand in your account how individual workers ever get raises or pay cuts."

I don't see how the signaling model of *formal education* would rule out a productivity theory of workplace training and experience.

Glen S. McGhee FHEAP writes:

"The reason why you have to go to college to get most good jobs is because a supermajority of qualified applicants for good jobs go to college."

This is only a recent development, only as recent as HR departments in companies after WW I started using college degrees as screening devices for filling management positions. Pretty soon, you weren't a legitimate enterprise without your own personnel department, and guess who staffed the new departments? Yup -- recent college graduates, who then turned around and (you guessed it) hired more people just like themselves.

The end result has been rampant credentialism, credential inflation, grade inflation, and the other ills that afflict higher ed.

Nathan Smith writes:


(a) college is an investment in a lifetime of enjoying higher status and thinking better, that is, it is a kind of long-term consumption; and

(b) people who expect higher lifetime earnings will be more inclined to invest in these things


Signalling will emerge as a side-effect. College grads really will be more productive, not because the average college grad is signalling but because he is anticipating a higher lifetime income and consuming some of it in advance. Then employers will take college as a signal. Signalling will have a feedback effect: more people will go to college than would have found it worthwhile, in order to look smart.

That fits my experience.

TallChris writes:

I completed an internship for a bank and was offered a full time position contingent on me graduating. Why would they care if I graduate? They had 3/4 of my grades for my college career. They evaluated me for an entire summer and decided I was qualified and worth hiring. The role was in risk and not anywhere they could advertise my credentials. Not sure this is completely on topic but it's been puzzling me for a while.

ohwilleke writes:

Signalling operates overwhelmingly before the employer has any opportunity to observe an employee's performance in the hiring process. It is a way to reduce the risk of expensive on the job training being wasted. This can be economically rational for the individual employer even if it deprives the economic as a whole of valueable human capital in the form of smart people who may lack a degree because they couldn't afford to go to college since their family was poor.

Even if the lag time between hiring and determination of marginal product is just six months or a year, credentialism has denied people who lack the credential a foot in any desirable door, and an increased perception and reality of meritocracy in education opportunities (relative to the 1950s and earlier) has also diminished in both employer expectations and reality, the likelihood that a diamond in the rough without the credential is out there.

The point of signaling theory is not that educational credentials are not accurate signals, but that education is often not adding much value that wasn't there when the admissions office sent you the fat envelope, and so requiring the elaborate motions of education rather than simply taking it on faith that high school graduates with good grades and test scores are up to sophisticated work challenges, is wildly inefficient.

joeftansey writes:

"Why would they care if I graduate? They had 3/4 of my grades for my college career."

To preserve equality in the workplace? It might demoralize your coworkers if they feel like you're getting the same reward for jumping through fewer hoops.

It also proves that you can still conform and buckle down, even after you've gotten your carrot. Its more behavioural data for them.

Bryan Willman writes:

Real World Observations.

Good colleges (or rather, good use of time in college) is about building human capital AND buildng up certain "signals" (credentials.)

Those attributes serve the owner in different ways, depending on the number of applicants for various positions, etc.

I worked at a place that would interview and sometimes hire people hadn't finished *high school* in the early '80s, and due to the sheer volume of resumes had to require degrees by the late 80s - it was just a matter of "how do we sort through this" - the difficult interview process remained the whole time. And well into the credential era (and I imagine yet today) if somebody was known to be skilled, they could get an interview, regardless of what their resume did or didn't say.

So there are actually at least THREE different values to college.

1. Development of human capital. (Not everyone partakes of this at a level that justifies the costs.)

2. Acquisition of credentials and other tools for penetrating filters. (People sometimes fumble this too.)

3. Life discovery - "what do I want do? what am I good at?" - this process is required. Whether it has to happen at a publicly subsidized institution is a different matter. But it has to happen.

Comments for this entry have been closed
Return to top