Bryan Caplan  

Should Economists Go to Finishing School?

Siegel for the Long Run... Rent-Seeking in Sin City

People don't like economists, and they don't listen to us. Economists who want to influence public opinion often connect the two problems. Maybe if we had tact, manners, and warmth, we wouldn't turn off our audience. And if we didn't turn off our audience, it would listen to us. All this time, we thought economics was a tough sell. But the only thing we really need to make the world listen is a diploma from finishing school.

There is something here. Few economists radiate the compassion of the Dalai Lama. Those who do get a better reaction from the general public. By speaking with humility, and showing that they care, they reach their audience in ways that uncouth economists like me never do.

This is an interesting hypothesis, but it's probably false where it counts. The problem I've noticed with the well-groomed approach to economic education is that it makes the audience like you more, but understand you less. If you refrain from bluntly stating economic truisms, you make more friends. But they are your friends because they don't grasp that you are defending socially unacceptable conclusions.

There is no nice way to tell people they talk too much. Similarly, there is no nice way to tell people that the minimum wage causes unemployment. If a normal person is still smiling after you make your point, he probably didn't get it, because the conclusion is inherently unsettling.

One of my favorite examples: I gave a ninety minute lecture on the minimum wage to my intro econ class. It seemed to go over well, but afterwards, I felt a little guilty. "Big man, brow-beating a captive audience of freshmen," I thought. Then I gave the midterm, and found that only 2 out of 25 understood the basic point of my lecture. Barely five even realized that my lecture on the minimum wage had been critical!

How could I have communicated more effectively? Should I have made my initial lecture more genteel? Negative - then no one would have understood me. The solution, rather, was to be more blunt. I was, and by the final exam only two or three still didn't get it.

But what good does it do if the public understands you without liking you? Only a little. But if economic education is your goal, it is far worse if the public likes you without understanding you. Once people know your position, there is at least a chance that they will start wondering if you're right.

The people who want to send economists to finishing school have their hearts in the right place. Economists do need to communicate more effectively with broader audiences. But I think their diagnosis is wrong. The main reason the public doesn't listen to us is not that we are boorish, but that we are BORING.

Think about the typical economist on TV. Why do your parents change the channel? It is not because he offends them, but because he puts them to sleep.

And this is good news, because it means that economists are currently below their communication possibilities frontier. By being blunt, economists can bore less and teach more. In other words: We can increase both the amount that people listen to us and their likelihood of grasping our point.

Better still, it's a fine line between blunt and funny. If you speak your mind plainly, people often laugh. That means:

1. They aren't bored.
2. They know what you mean.

and, best of all:

3. They start to lose a great impediment to understanding economics and the world: The gut-level feeling that popular opinion is sacred.

Comments and Sharing

TRACKBACKS (6 to date)
TrackBack URL:
The author at Jim's blog in a related article titled Telling the truth about economics. writes:
    Economists are boring [...]I am unaware of any clear, straightforward explanation of why the free market works, of why, free market prices and free market production are correct [Tracked on May 8, 2005 9:50 PM]
The author at Bluegrass Policy Blog in a related article titled People don't like economists. writes:
    Bryan Caplan is a brilliant and well published economist, but my girlfriend could have told him this one. As usual he's right and he comes to the right conclusions after doing what good economists do best: delving deeper into the problem and seeing the... [Tracked on May 12, 2005 10:11 AM]
COMMENTS (34 to date)
Carodozo Bozo writes:

There's another problem with connecting on economics. For the most part, knowing economics doesn't really help people. Other than be an economist, you can't do anything with it. Think about it: what does knowing that the minimum wage causes unemployment get you? Nothing. Economics surround you, but you can't "work it." Economics is like any other advanced science: only the people who "do it" as a job need to know how it works.

The following people need to know economics for their job: Legislators, Judges, Regulators, Federal Reserve Chairman, and Investment Managers at Banks and Mutual Funds.

There is only one time when Joe American needs to know economics: when he votes. This is really, really important because (in this country anyway) we have to ask Joe's permission to elect economically literate officials or implement economically rational policies. Unfortunately, it's only one day every other year or so, so the stuff Joe needs to know to be an engineer/ actor/ carpenter/ computer programmer, etc. take precedence.

No immediate application = No interest in learning.

Carodozo Bozo writes:

I realized the voters have a collective action / free-rider problem when it comes to getting an education in economics.

If 51% of the population gets a thorough education in economics, we'd all be much wealthier.

Think about it.

Mark writes:

Well, Bryan, one reason that your students return so many "wrong" answers may be that you are teaching the subject imcompletely. I looked over those lecture notes of yours and found them inadequate. Every single item in your analysis makes the assumption that all employers are wage-takers--i.e., that pure competition and all the conditions it entails exists in labor markets.

As you surely must have learned at some point in your economic education, in the absence of pure competition, the standard economists' conclusions regarding price ceilings and floors--that ceilings cause shortages and floors cause surpluses--do not necessarily hold. In particular, in a labor market where employers have market power, a minimum wage law may not reduce employment.

Moreover, I am shocked that a recent Princeton Ph.D. appears unaware of the empirical work on the minimum wage issue by your former professors David Card and Alan Krueger. They have repeatedly found that minimum wage increases result in no discernible reduction in employment of affected workers. Even economists such as Tyler Cowen and Steven Landsburg--both of whom are as strong in their ideological devotion to laissez-faire as you are--have now acknowledged that on this issue, the evidence is with Card and Krueger.

Bryan, what I see you doing here is blaming your students for intuitively sensing what I, as a fellow Ph.D of yours, can see clearly--that you are giving them a one-sided, heavily biased treatment of the subject.

Walker writes:

Most people don't like listening to a chemistry lecture either, but that doesn't stop chemists from making their influence felt. If economics was also a science that produced useful results, economists wouldn't need public relations training.

James writes:


In your eagerness to advocate your own ideology you must have missed the point. Bryan's story about his students wasn't intended as a complaint that they didn't agree with his view. The point was that they didn't even know what his view was.

Regarding Card and Krueger, the standard economic theory is that a price floor increases the difference between quantity demanded and quantity supplied. People who are voluntarily unemployed (because they are unwilling to work at the present wage) become involuntarily unemployed. In order to get the conclusion you claim, C & K would have had to measure the increase in the number of sellers of labor. They did not do so.

You're correct to point out the possible problems arising from incorrect assumptions about markets. In this case, the assumption of efficiency is wrong not because employers comprise some sort of cartel, but because the product, labor, is not homogenous. There will always be workers whose marginal value product is less than the minimum wage. These are the workers hurt by the minimum wage. Including workers with MRPs above the minimum wage dilutes the effectiveness of the study. If these particularly unproductive workers are a very small part of the total labor force, empirical analysis will not show the effects that minimum wage laws have on this group unless the study targets this group specifically.

Dewey Munson writes:
How could I have communicated more effectively? Should I have made my initial lecture more genteel? Negative - then no one would have understood me. The solution, rather, was to be more blunt. I was, and by the final exam only two or three still didn't get it.

Maybe you are wrong!

An economy needs everyone. All CEO's (or economists) would be a disaster.

All farmers, ditto, except we would eat.

There should be a reward for playing a necessary part in a system without being penalized by the fact that many people can do it.

Support for some level of minimum wage is likely an intuitive response to the belief that the lower group of producers are more valuable than their reward.

After all, if the CEO's (or economists) all stayed home for a month no one would notice; on the otherhand if all the lower group stayed home, things would quickly grind to a halt.

It is quite possible to hold this opinion while also understanding the real world in which an excess wage can reduce employment.

"great impediment to understanding economics" - The gut-level feeling that (popular) ECONOMIST opinion is sacred.

David Thomson writes:

“For the most part, knowing economics doesn't really help people. “

The effects of the minimum wage on low earning individuals and society? As a cynic once remarked to me, this sort of intellectual interest might assist you in becoming a better person---but it will likely do little to fatten your wallet. Most people perceive such a discussion as similar to how many angels can sit on the head of a pin. The majority of Bryan Caplan’s students would instantaneously quit his class if a high grade was guaranteed. He might wear a clown costume and juggle bowling pins in the air, and it wouldn’t make any difference whatsoever. They simply fail to see how the issue directly impacts on their lives. Is there any hope? Oh yes, Professor Caplan merely needs to explain to them how a greater understanding of the minimum wage will jazz up their sex life!

Bob Knaus writes:

Bryan - at the risk of proposing a null hypothesis like "marginal utility of happiness" I'd say your problem is not blunt vs. genteel but persuasive vs. unpersuasive.

You have the privilege of teaching America's brightest young people. They may be opinionated, but surely they are not pig-headed. If the only way you can get them to learn is to browbeat them, you might want to take a serious look at how you teach, and how you lead.

It is possible to persuade (which combines teaching and leading) without being a tyrant. Your choices are not merely Mr. Boring or Mr. Blunt.

I used to be a management consultant. Nobody ever called us in when things were hunky-dory. We had to quickly sift through a lot of facts and opinions, and come up with a persuasive report that was bound to upset some apple carts. Making these reports "genteel" rather than "blunt" was the key to client acceptance, and to getting positive results. Often, the changes between draft and final were merely rewording to make the recommendations more palatable. The facts -- and the results -- didn't change.

If your students are finding your courses hard to swallow, you should take a similar approach.

spencer writes:

If this is what you are doing in the first class no wonder you are failing at your objective.

You are jumping into the middle of a difficult and complex subject tossing around terms that most of the students do not understand in the technical way you are using them.

I'll make you a challenge -- after you have made your point c about marginal product ---stop.
Ask some students to tell you what marginal product means. I doubt if 10% of the class will give you a satisfactory answer.

Your job in an intro class is to teach the student what marginal product means and why it is important. That is the reason they take the course, not to get a biased lecture on the minimum wage.

Who knows, maybe this excercise might even teach you something.

Bernard Yomtov writes:

Well said, Mark. I hope we won't be reading a lot of complaints from Bryan about ideological bias in the classroom.

The comments about the labor market are incomplete, as Mark points out. The comments about discrimination are laughable.

"[Anti-discrimination laws] reduce 'protected' groups' marginal productivity."

Sure they do.

David Thomson writes:

“You have the privilege of teaching America's brightest young people. “

Yup, and most of them care only about the money they will earn after college. They usually don’t give a damn about anything else. Does anybody actually think that these young people are sitting around after class vigorously debating the minimum wage issue? If you are---you need to immediately cease smoking that illegal stuff. It is damaging your brain cells.

Matt McIntosh writes:

Anyone who says that the minimum wage doesn't cause uenployment should be willing to perform the following experiment: double it and see what happens. If you're not willing to try that experiment, then maybe you might want to rethink your position.

Oh wait, some countries have tried that experiment before. Didn't work too well...

spencer writes:

Matt McIntosh -- In the United States the real minimum wage rose from $2.80 in 1956 to $4.67 in 1968-- sorry not quite a double, but close enough
for the experiment you suggested.

I think for one that experiment produced some very good results as that era was one of extremely strong employment growth, profits growth , etc., .

Im am not going to go through a long list of economic data, but that decade seemed to be generally recognized as having a very good
economy with especially very good results for

But I think the conclusion suggest very strongly you should reconsider your position.

Have you got a counter experiment?

Mark writes:

Anyone who says that the minimum wage doesn't cause uenployment should be willing to perform the following experiment: double it and see what happens. If you're not willing to try that experiment, then maybe you might want to rethink your position.

Matt, neither the theoretical analysis of the effect of minimum wages on non-competitive markets, nor the empirical analysis of Card and Krueger, suggest that the minimum wage can be increased without bound; therefore, your point is irrelevant.

James writes:


The window between 1956 and 1968 is wide enough to allow for all kinds of other variables to change. In order for the experiment to work, you'd have to make the change quickly enough that (among other things) entrepreneurs wouldn't be able to contaminate your sample by increasing workers' marginal product through capital formation. In other words, double the minimum wage effective tomorrow. Would you be willing to do this?

spencer writes:

No, I would not double the minium wage tomorrow. But my remarks were in response to a statement that did not say what the time frame had to be.

The right and left have been arguing over the minimum wage as long as I can remember -- and I went to grad school in the 1960s. The right use to claim that a higher minimum wage would cause higher inflation. Now they claim it will cause unemployment. When did the argument change and why?

Some of the best minds have tried all types of analytical techniques over this period to both prove and disprove every thesis and no one has been able to make their case in a way that the opposition has not been able to shoot down.

But reasonable increases in the minimum wage has never been shown to cause either rising inflation or rising unemployment.

I repeat my earlier challenge, show me a counter example.

spencer writes:

Bozo -- I would add to your list of people who need to use economics every business executive, planner and budgeter.

awptimus writes:

If what you were asserting were true, we could raise the minimum wage to $100 with no consequence. But of course, what you're saying is absurd; there would definately be consequences with such a change.

Even if we take your labor demand inelasticity assertion at its face value, you must also realise that substitution effects occur when the minimum wage is raised; minority (for lack of a better term) unemployment rises when the minimum wage is increased, plain and simple. This is the same law of Supply and Demand, just more specific. When you're teaching an intro class, it is definately not important to vet all the nuances, because then you'd be clouding the central point that the laws of Supply and Demand (and the affect ceilings and floors have) apply in labor markets just as they do everywhere else.

But again, try upping the minimum wage to $100 to see if your labor demand inelasticity assertion holds.

Bernard Yomtov writes:

If what you were asserting were true, we could raise the minimum wage to $100 with no consequence.

This is exactly what Mark did not assert. In fact, he took pains to make it clear that he was not asserting that.

In case you missed it, he wrote;

neither the theoretical analysis of the effect of minimum wages on non-competitive markets, nor the empirical analysis of Card and Krueger, suggest that the minimum wage can be increased without bound; therefore, your point is irrelevant.

Further his point was not about the elasticity of demand but rather the consequences of having a market for labor that is not purely competitive.

For the most part, knowing economics doesn't really help people. Other than be an economist, you can't do anything with it.

Of course you can. Economics is just evaluating different ways of getting something done. All businesses do it all the time. How would a construction company (big or small) know how much to bid on a project, without economics?

Or a body-fender shop know how much to ask for repairing accident damaged cars? Or a gardener how much to charge for pruning trees? Or an employee know if it's worth working overtime?

Even the decision of when to get out of bed is economics.

Also, the 'natural experiment' of nearly doubling the real minimum wage from 1956 to 1968, shows just how deadly it can be. It was in the sixties that the Fed lost control over inflation, and if one reads, say, Arthur Burns congressional testimony, it's clear that it was just what Hayek predicted in, 'A Tiger by the Tail'.

James writes:


I'd imagine that people thought the minimum wage would result in inflation because they thought that increased prices really could be legislated without changing the quantity to clear the market. I've never believed that so I have no interest in defending it. Similarly, I won't ask you to defend the view that the minimum wage reduces poverty.

I'm not sure what you mean by counterexample. If you mean a datum contrary to a null hypothesis, you might want to think about what the null hypothesis ought to be here. I'd let the null hypothesis be that the supply and demand curves for unskilled labor intersect. Yours seems to be that they are parallel, although your focus on reductions in the existing labor force would be a more appropriate test of the hypothesis that the demand curve is vertical.

Or, by counterexample, do you mean something to counter what you gave as an exmple? I certainly hope that you don't consider the minimum wage increases from 56-68 to be an example of the minimum wage increasing employment. There are too many other variables that you have no way of controlling for. Also, an empirical test has to actually test the theory in question. The mechanism by which the minimum wage is expected to create unemployment is that the quantity supplied for unskilled labor will increase and the quantity demanded will decrease. Your "example" doesn't isolate workers of different skill levels or measure changes in quantity supplied.

You might also want to reconsider your comment about "reasonable" changes. If, by reasonable, you mean that a change has results that can be obscured in an empirical study, you wind up justifying policies that you probably wouldn't support, provided that they are adopted incrementally enough and while enough other events are going on to make it impossible to attribute causation.

spencer writes:

The suggestion was made-- try doubling the minumum wage. So I just very quickly with essentially no effort showed an example of the proposed experiment that did not show the implied results.

Everyone is jumping on me, but lots of arguments, but no one is citing a single data series.

Economic theory is great, but it has to be tested agaisnt reality.

My simple statement is that there is no "good" evidence that raising the minimum wage by a moderate causes unemployment.Show me the data that shows me to be wrong.

That is what I love about this blog, it is never confused by the facts.

spencer writes:

I just did a little exercise. I tested the correlation of the minimum wage against the unemployment rate, the unemployment rate lagged 6 months, 12 months, 18 months and 24 months.

The correlation results were plus 0.2 , not very significant.

So I looked at the correlation between the change in the mininmum wage, the real minimum wage and the change in the real minimum wage over the period 1954 to 2004 against the unemployment rate and lagged unemployment.

Guess what, the correlation are about the same orders of magnitute but the sign is negative, implying that raising the minimum wage leads to a drop in the unemployment rate.

Gee, I wonder what I did wrong that the facts do not support the theory.

Anyone got any ideas.

metis314 writes:


You could be right. It seems to reasonable to me that if you increased the wage of a hamburger flipper to $10 all that would result is hamburgers costing more money. It's not like McDonald's can get by without them, and it seems like it would be easier to just raise prices than fire workers - especially if the raise was incremental and not all at once. This suggests a correlation with inflation (as others here stated).

On the other hand, there are certainly jobs out there that would be done away with - for example a parking garage attendent could just be replaced with an automated garage gate.

So it doesn't seem to be clear that raising the minimum wage slowly would cause unemployment on a large scale; however it does seem that some inflation and/or unemployment must occur b/c the extra money doesn't grow on trees and the CEO's aren't going to volunteer to give up part of their paycheck to offset the extra costs.

spencer writes:

Actually, I agree that over time a gradual rise in the minimum wage probably would add additional pressure, along with a host of other factors that contribute to the long run trend of replacing labor with capital and causing the long run growth in living standards to increase.

But, that is what we want in this economy, is to eliminate low productivity jobs. So in this context you can look at a rising minimum wage as just another factor among all the other factors contributing to rising living standards.

What I would argue, just to take the example of a fast food resaurant is that raising the minimum wage leads to the owner doing things that raise the productivity of fast food restaurant employees and that the impact of this through rising incomes more then offset the negative impact on jobs.

That would be the real problem that I had with the minimum wage lecture. It assumes fixed capital and does not allow for the possibility that the owners adjust by adding capital and
improving the productivity of his employees.

Maybe he has fewer employees after the adjustment, but the total income of his employees should be higher then before the change in the minimum wage and this should lead to more jobs.

We are in a very dynamic economy where the number of job deaths and births is very, very high and this type of dynamic analysis is what is missing in his lecture notes and causes the lecture to be a very inadequate presentation of the real world in my mind. That is why I called it a biased lecture. It was focused on teaching the students his belief, and I emphasize the word belief, that a rise in the minimum wage causes unemployment when the reality is very different
and just some small changes in his lecture to consider other possibilites would make it a much better lecture and make it much more likely that the students would get it.

It is also why no one has ever been able to show
that minimum wage changes have big impacts, they are lost in the host of dynamics in the overall economy and maybe the second round impacts are what is the most important.

James writes:


There are plenty of empirical studies showing the harmful effects of the minimum wage. See this for a brief list. I'm just not itching to cite them because I'm skeptical of most of the empirical work on the minimum wage coming from both sides.

Gee, I wonder what I did wrong that the facts do not support the theory.

Anyone got any ideas.

Here are a few: Your correlation analysis fails to control for other explanatory variables. You neglect to measure the variables predicted to change by the theory you are claiming to test, quantity supplied and quantity demanded of unskilled labor. The variation in unemployment not explained by your model exceeds the fraction of the labor force earning minimum wage.

But let's forget all that. Your test was inconclusive and I'll suppose that all of the empirical studies in the link I provided are inconclusive as well. What null hypothesis should we revert to?

Lancelot Finn writes:

My essay "Work, Service and Worship" is an answer to this problem. My argument is that God made man subject to scarcity, He gave us Needs, so that fear of hunger and pain and death would force us to live in community; He thereby spared us from the hell of total alienation that would have been our lot alone. Ever since I wrote it, I've lost that sneaking feeling that as a free-market economist I was somehow heartless. Au contraire.

Mark writes:

There are plenty of empirical studies showing the harmful effects of the minimum wage

Most of the studies listed in the link you provide are over 20 years old; the empirical debate has moved far beyond them. As I noted in an earlier post, we are now to the point where even some pretty hard-core libertarian economists now acknowledge that Card and Krueger are correct and that the employment impact of reasonable minimum wage increases is minimal.

spencer writes:

Ok James -- I know -What I did is poor.

But, I was not the one that first made a bad statement. I just responded to the
statement about a county raising the mininum wage by 50%,
but I was the one that got all the static, not the person who made the stupid argument.

I just wanted to make the quick point that the 50% statement was stupid, sorry about that.

My basic belief that in the US the minimum wage doens not do much good or much harm.

But I still think the lecture is extremely biased and that is probably why the kids do not get it.
They know there is something wrong with it, but
they do not have enough experience to know what.

A good teacher would show the theory and economics behind several alternative views on the minimum wage. If his view is so correct the student should reach the desired conclusion
on their own.

dave writes:

Anyone genuinely interested in why the public at large tends to ignore economists MUST read Michael Bernstein's "A Perilous Progress: Economists and Public Purpose in Twentieth Century America".

Bernstein concludes: "At century's end, the discipline ... was confronted with a remarkable paradox. On the one side, the very prestige and influence of economics as a social science had been built upon statist agendas. Yet on the other, the contemporary profession had turned inward, spurning the fiscal activism and frank political engagement of its forebears, preferring a more withdrawn posture that ostensibly depoliticized its work while at the same time it made more and more practitioners mere shills for particular corporate elites eager to seize upon public assets now increasingly `privatized.'"
Paul N writes:

I personally think the minimum wage, and whether it increases or stays the same, is irrelevant as long as 98%+ of wage earners make more than it.

Randy writes:


Agreed. And further, if the minimum wage ever is set high enough to have an effect, the effect it will have will not be unemployment, but an increase in under the table employment. Has anybody ever asked themselves how all these illegal immigrants do it? How do they work illegally? Answer; Where there's a will, there's a way.

spencer writes:

After this great discussion, really, th only question remaining is if our host has though about changing his lecture on the minimum wage to include other possibilities.

Actually, I have a real question . If the thesis is right that an increase in the minimum wage leads to employeers cutting employment;does a 10% increase in the minimum wage lead to a lay off of 10% of the number of employees, a cut in every employees hours by 10%, or every employee being fired and the firm going out of business. From our host lecture notes it looks like each of these is a posibility and I would love to see his answer to this exam question.

Jacqueline Passey has a cleverly spot-on post on her website on the topic of incentives in economics classrooms:

How ironic but true it is that economics professors often miss demonstrating the basics of the very subject they teach by not instituting appropriate incentives for their very own course requirements. The explanation is surely that teachers' own incentives are determined by things other than internal consistency, from research to assigning grades that the school hopes to be comparable across different fields--fields where failing to apply incentive-based grading is not as embarrassing.

Economists in the classroom may indeed be below their possibilities frontier, as Caplan notes; but if so, one explanation to consider is whether the incentive structure in those classrooms is commensurate with achieving that frontier.

Comments for this entry have been closed
Return to top