Bryan Caplan  

What Happens When Signaling Gets Cheaper?

PRINT
Economists' Self-Conception... Is Iran a Threat?...
Arnold:

My understanding of the signaling model is that it depends crucially on the relative cost of signaling to people with and without the desired trait. You want the cost to be high for someone without the trait and low for someone with the trait.

With that in mind, I do not see how lowering the cost of signaling for people with the trait does anything other than cause people with the trait to choose the low-cost signal. The problem with a low-cost substitute for a diamond is that it lowers the cost of signaling for people without the desired trait (which is a willingness to buy an expensive gift).

The problem is that a low-cost substitute lowers the cost of signaling for everyone.  So if the cost per signal falls by 50%, you have to do twice as much signaling to separate yourself from the pack.

Simple example: Suppose that (a) good students are $20,000 more productive than bad students; (b) good students endure $5000 of suffering  per year of school; (c) bad students endure $10,000 of suffering per year of school.  Then in equilibrium, good students need at least two extra years of schooling to distinguish themselves from bad students.  Good students will be happy to do so, because it nets them $20,000-2*$5,000=$10,000.  Bad students won't bother, because imitating good students nets them $20,000-2*$10,000=$0.

Now what happens if the cost of education falls by 50% for both groups?  A two-year education gap is no longer stable!  Bad students will suddenly find two years of education profitable: $20,000-2*$5000=$10,000.  Now the good students need four years of schooling to distinguish themselves.  As a result, the total value of resources devoted to signaling remains unchanged.

If I come up with a low-cost way to earn a badge that signals intelligence, conscientiousness, and conformity, and that badge can only be earned by people with those traits, then my badge should find a market.

If you devise a low-cost signal that only high-ability people can earn, you're right.  But that's tautological.  In the real world, low-ability people can always try to imitate high-ability people.  If the signal everyone used to send gets 50% cheaper for everyone, the quantity of signaling has to double to preserve separation. 

One challenge is that when few people use the badge, it seems to signal non-conformity. Thus, the early adopters of my cheaper badge do not do as well as they should. But over time, there are two possibilities. One is that the conformity hurdle cannot be overcome, so that the incumbent signaling mechanism remains dominant forever. The other possibility is that eventually a tipping point is reached, and enough people use the new badge so that it no longer signals nonconformity. At that point, the market position of the old badge rapidly deteriorates.

I think that we will arrive at the second equilibrium at some point. However, predicting when it will occur is difficult.

What's your best guess, Arnold?  Now you barely sound more sanguine than I do.



COMMENTS (9 to date)
blink writes:

To the extent that "ability" refers to some fixed trait, then it seems reasonable that we will find a cheap and reliable way to measure it. We have (moderately reliable) IQ tests now, perhaps brain scans will be the future.

To the extent that "ability" simply means "productive" and encompasses traits like conformity, patience, and conscientiousness, it seems that at some point the best imitation strategy is simply to *be* those things rather than rely on prestigious schools, etc. That system is appropriate when an initial hire implies a long term relationship, so perhaps we can expect reductions in signaling through this margin, perhaps due to technological progress.

Mike writes:

I'm in Mauritius trying to create a new badge.

It's hard.

Keep in mind that there is a range of signals. Over here there is a huge population that wants to send a quality signal, but there is limited access to quality signalling devices. The school has a chance to develop a lower cost, but quality signal. The market here requires a lower cost signal. Income levels won't support the cost of a quality U.S. style signal in sufficient numbers. The market here is the perfect place to try.

I believe I can get employers to eventually accept my new badge. The plan is to have employers become part of the effort, the school will then train students specifically for the employer's needs as a start. A tough nut to crack will be getting institutions such as GMU to accept the badge as a qualification for further study.

The school is also here in part because the accreditation cartel in the U.S. will not accept a new badge easily. It will have to be forced upon them in some manner.

One more thing: the school will have to be profitable to make this happen. So it will be a few years before there is anything distinctly different about the school's badge, but we'll get there.

I like it when Bryan and Arnold have these discussions. It both affirms my efforts and causes me to think more deeply about how to make the school a success.

(For those of you wondering why a small country like Mauritius, instead of, say, India, we recruit heavily in India. We are not located there because their regulatory and accrediting system is arguable worse than the U.S. Also, it is pretty easy to get Indian and African students to come to a tropical island paradise that is relatively close to home.)

Essen writes:

Sorry that I sound such a dumbo. But what is signaling as is referenced here in this thread?

Krishnan writes:

The "signal" can be more that just "cost"/expense - it can be "ease of getting that degree". Universities remain under pressure to "graduate" as many students as they can - even as the problems with the entering classes multiply, the pressure to graduate them as quickly as possible multiplies - credit requirements are being changed (lowered) and armies are being assembled to "help" with the graduation rates - the schools at the top are not immune either ... so, it will indeed get more difficult for the good students to differentiate themselves from the bad - using paper/grades alone.

Mike writes:

Essen and Krishnan,

In this context, the signal of higher education is that you possess certain qualities: conformity, drive, ambition, ability to follow instructions, ability to complete assignments, etc. School has nothing to do with learning anything.

PrometheeFeu writes:

What if instead the cost drops by $5,000 for everybody? (Which seems to better model a fall in tuition for instance) So now the equilibria is 4 years of schooling for good students netting them $20,000.

Ghost of Christmas Past writes:

A very accurate (much less cheap) signal of ability becomes less of a signal and more of a simple measurement.

As others have pointed out we have some tools along those lines, such as IQ tests. We could probably improve our tools to obtain measures of ability which were both fairly accurate and fairly cheap.

One big reason we don't is that too many people don't want accurate, cheap measures of ability. Notably, today's rich parents of tomorrow's successful kids want to be able to buy their progeny a place at the table, because genetic recombination and reversion toward the mean supply even successful parents with kids of varying and often less (even if only moderately less) ability-- yet parents want all their children to succeed.

Ambiguous, costly signals are made to order for rich parents-- they can buy those signals for their less brilliant kids. An accurate, cheap test of ability would simply reveal that their less brilliant kids are (drumroll, please) less brilliant-- than some upstarts from the middle or even lower classes.

Ambiguous signals can also be passed out as political favors. For example, a certain fraction of top college admission slots can be gifted to young people from a disgruntled class with lots of melanin and rather less ability. Substituting a cheap valid measurement for a costly ambiguous signal would be bad for those people, though obviously good for everyone else.

Finally, a good cheap measurement would tend to diminish the status and income of the people running the old signalling apparatus. But those are some of the most influential people in the country! Naturally they oppose any change and they have the power to delay and minimize change so that it will take a long time coming. (Our only hope lies in the arrogance of those people-- they have used their power for short-term gain, and have run up the cost of their apparatus to the point where it may just collapse soon, at least the weaker parts of it (say, Podunk College versus Harvard).)

Until you can figure out how to purge the rent-seekers from of the ability-signalling game, I doubt you can make fundamental reforms.

Chris Koresko writes:

Arnold Kling: My understanding of the signaling model is that it depends crucially on the relative cost of signaling to people with and without the desired trait. You want the cost to be high for someone without the trait and low for someone with the trait.

Bryan Caplan: The problem is that a low-cost substitute lowers the cost of signaling for everyone. So if the cost per signal falls by 50%, you have to do twice as much signaling to separate yourself from the pack.

I think you guys are talking past each other.

Arnold's idea of "signaling" is that each individual can carry a "badge" which indicates that he has some desirable trait. Reducing the cost of the badge encourages its use. The critical point is that in Kling's conception the willingness to pay the cost of the badge is not itself the signal. His badge is constructed such that only the people who have the desired trait are able to obtain it, for reasons independent of the cost.

For example, imagine that health insurers relied heavily on a test of some particular gene to determine who they did business with. Then a reduction in the cost of testing for that gene ought to make the use of it more widespread and influential.

Mike writes:

Ghost, nice post.

I have to deal with my fair share of rent seekers here.

I think the gate keepers and rent seekers at the Harvards of the world will long keep there privileged positions. I'm aiming right at the Podunks.

Comments for this entry have been closed
Return to top