Bryan Caplan  

For Ye Have Signaling Always With You

My Reply to Bill - Digest Vers... Buying a House With 20/20 Hind...
Many economists assume that market forces will somehow figure out a way to make signaling costs disappear.  But as far as I can tell, they never explain why signaling costs would be easier to eliminate than any other costs.  And on reflection, the truth is precisely the reverse: Signaling costs are especially hard to eliminate.  Why?  Because when you make signaling cheaper, agents' natural response is to signal more intensely or on another dimension.

Let me illustrate my claim with a prediction: The typical engagement ring will always cost several weeks' income.  If industry figures out how to cheaply synthesize gold and diamonds, we'll start making engagement rings out of something else - platinum and rubies, or ivory and T-rex teeth.  Why?  Because one major function of engagement rings is to signal commitment with an expensive gift!  To separate the sheep from the goats, the signal has to be expensive enough to convince the goats to give up.

The same goes for signaling generally.  Once students can download perfect Latin from a computer with a single click, employers will no longer be impressed by students with good grades in Latin.  So how then will students manage to impress the employers in this brave new world?  By excelling in subjects that remain difficult to learn.  And if necessary, schools will respond by inventing challenging new hoops that make Latin look useful by comparison. 

Comments and Sharing

COMMENTS (4 to date)
agnostic writes:

How much of the biology stuff on signaling do economists read, even if in popularizations? There, the focus is mostly on honest vs. dishonest signaling, and it's known that one way for a signal to be honest is for it to be costly.

Economists, whose potential insights are often prevented by too high a level of cynicism, assume that people engage in costly signaling just to show off and lord it over the rest. In reality, they're doing it to give an honest signal, one that enhances welfare by improving the efficiency of communication. (God, economese sounds so terrible...)

Consider when two BAMFs are on the brink of fighting each other, possibly to the death, possibly to the deaths of those caught in the crossfire. If one can honestly signal superior quality, the lower-quality one may back down. Again one of the simplest ways is to display a costly signal of quality.

If we eliminate costly signals, we'll immediately raise the level of noise in displays, and we'll have much more and much greater conflict, like the two guys escalating to a brawl because neither can tell if they're better than the other beforehand.

Hyena writes:

I think the "signals should get cheaper" argument isn't that it will be cheaper to make engagement rings, but that some enterprising entrepreneur would develop the "engagement contract" with severe penalties for divorce. People who can afford very expensive engagement gifts already do this: they craft prenuptials which create a substantial loss for the proposing partner.

Agnostic is correct that honest signaling is the primary issue, but efficient signaling doesn't raise the cost of signals: it lowers them by making them harder to fake.

Which I suspect you realize; after all, you've spent a lot of time arguing that we should dump education for a more efficient, less error-prone signaling method.

Lars P writes:

According to David Friedman (here) expensive engagement rings came about as a way to facilitate premarital sex when US courts stopped awarding damages for breach of promise to marry in the 1930s.

If the man left, the woman could sell the expensive ring, and the world staid economically efficient!

Rebecca Burlingame writes:

What is so interesting about signaling, is that it tells just how much of the economic world of valuation exists primarily in our own heads. What would people think if they ever realized the processes of valuation and accreditation actually lie within each individual's mind, and the true economies of scale that could actually create?

Comments for this entry have been closed
Return to top