Tuesday, May 8, 2007

Signal to Noise Ratio

Will Wilkinson takes a walk on the dark side. He agrees with Bryan Caplan that education is mostly just a signal that you are already talented and suggests that there is “a huge entrepreneurial opportunity for whomever can come up with an alternative scheme of credible human capital certification”

My longstanding objection to the signaling hypothesis is two fold

1) The market abhors unclaimed profits. That is, it is hard to imagine that individuals in the economy are spending vast resources on something which is largely useless and that profit maximizing firms are rewarding them for doing so!

2) There is little to no evidence for the signaling hypothesis. The evidence shows that when we control for ability the return to college goes up, not down. For me this explains why nerds tend to find school useless. It was useless for them. That doesn’t mean that it’s useless for everyone.

My own theory is that the human capital formed in higher education is the ability to solve problems, manage time, work independently, seek out assistance, communicate your ideas, form and in some cases direct teams. These skills, not knowledge are what employers want.

Traditionally education focused on the classics for the same reason that athletes run hills. They’re hard. It’s not that your ability to run hills signals your ability as a football player or wrestler. It makes you a better football player or wrestler.

Now in athletics there was a revolution in “sport specific training.” That is, some people said “hey why don’t we have these kids train in ways that are actually related to the sport they are playing.”

It made a difference but not a huge one. The key factors of strength, speed, motivation, endurance, emotional self-discipline and tolerance for pain where sometimes better honed under conditions that had nothing to do with the actual sport.

Education, I believe, operates mostly the same way. The skills needed to direct an engineering project are sometimes best honed by slogging your way through purposefully near impossible homework assignments. The understanding of unintended consequences needed for white space thinking in big business may come best from analyzing why it is that none of your favorite economic policy proposals work out the way you thought. The communication skills needed to lead a company through restructuring my first come from writing and rewriting an essay that can persuade a stern professor that Hume is properly considered an atheist, not a deist.

In each case it is the skill set that is important. The knowledge is relatively unimportant.

hat tip to Kling


James Somers said...

I'm a nerd at the University of Michigan.

I agree: employers want you to be able to solve hard problems. Moreover, they want the mental acuity that comes out of the *process* of solving hard problems. So as a student, the substance of what you're doing, be it math or physics or physiology, is irrelevant -- the point is that it's *hard*.

So when you apply for a job, that's what you're signaling. A math degree from MIT will get you more cash out of the gate than a psych B.A. from UT, because the average employer will think you solved harder problems doing math at MIT than psych at UT.

It's almost self-evident to me that it's not *what* you did at college that matters, but how well the stuff you did translates into credentials/interview material/whatever. In other words, I think you've hit the nail on the head in terms of explicating the substance of a college education, but I think you've failed to prove that it's not a signaling game.

Keep up the good work.