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.”