Making Economics Fun: Part I

Every year the people who grade the economics AP exam get together for a week to do it. In 2015 they invited me to be their guest speaker. Since the audience consisted almost entirely of people who taught economics at either the high school or college level, I gave a talk on how to make economics fun. This post is and the next will be based on it.

Teaching Economics: The Problem

Economics is about the world the students are living in, things they are familiar with, uses words such as “efficiency” and “competition” whose meaning they think they know. That makes it tempting to listen with half an ear in the belief that the professor is just droning on in unnecessary detail about things you already understand and only discover that you were wrong when you get back your first exam. One solution to this problem is to find things where the economics gives the opposite of the answer your students expect.

Consider, for example, the economics of polygyny, one man marrying two or more women. The natural assumption is that that is good for men, since they get to have multiple wives, bad for the women who have to share a husband. But if you think about it a little you realize that that doesn’t work because wanting two wives doesn’t getting you two wives; would-be husbands are competing on the marriage market for a limited number of women who want to be wives. If polygyny is illegal one man can only bid for one wife, offering whatever characteristics make him a desirable husband and the explicit or implicit terms on which he is offering to marry her. Legalize it and now everybody can bid for one wife and some men for two. The demand curve for wives shifts out and their price goes up. In order for a man to get a second wife he has to offer her terms attractive enough so she prefers being his second wife to being the first wife of an alternative suitor. Some men who want two wives may be able to get them but on terms biased more in the wife’s favor than in the past, so may or may not benefit by the change. Men who end up with only one wife are worse off because they have to offer her more favorable terms due to the competition of the polygynists. Some men may be better off, some worse off. All women are better off.

That is not a perfect model of the marriage market — it has problems less common on ordinary markets, such as the difficulty of enforcing the implicit terms of a marriage contract — but it is enough to show why economics suggests the opposite of the result people expect.¹

We have no experimental evidence on the effect of polygyny (or polyandry) on the terms of marriage in modern societies but we have evidence of the same logic in a related context.² In colleges where women heavily outnumber men the men get what most of them want, a culture of hookups and casual promiscuity. In colleges with more men, dating patterns tend towards the long-term relationships that women, on average, prefer. A student, male or female, who wants to find a spouse in college should choose a school such as Georgia Tech, 66% male.

On the other hand, a student who …

For another example, suppose you discover that in the slums of your city some apartments lack hot water. You want to help poor people so persuade the city to make it illegal to rent out an apartment that does not have hot water.³

If the tenant was willing to pay more for hot water than it costs the landlord to provide it, it would be in their mutual interest to agree to include hot water in exchange for a higher rent. If hot water is not included that is evidence that the cost of providing it to the landlord is greater than the value to the tenant. If you work out how the demand and supply curves for a good shift when you change its characteristics in a way that makes it both more valuable and more costly to produce you can show that the shift will raise rent by more than the value of hot water to the tenant but less than the cost of providing it to the landlord.⁴ What looks like a legal change that benefits one group at the expense of another actually makes both worse off.

For a final example, consider the question of who pays for Social Security. What we are told, what many people believe, is that it is paid half by the employer, half by the employee. Every economist knows that this is window dressing. It does not matter whether some of the dollar bills that my employer pays me are handed over to the IRS by him just before I get them or by me just after; in either case the tax is the difference between what he pays and what I receive. How much worse off each of us is as a result depends on the elasticity of supply and demand for labor. The formal division of the tax between employer and employee is irrelevant.

Presenting students with situations where the economic analysis implies the opposite of what they expect may not always convince them but at least it may make them think, whether in trying to understand why you are right or to find a reason why you are wrong. If you tell them something they already believe they do not have to think about it and very likely won’t.

Another way of getting the attention of students is by offering them ideas that they might find useful. For example:

Leveraging Your Enemy’s Rationality

I have a very small collection of jokes that teach economics. Here is one that I got from Dennis Hanseman, the editor of my Price Theory text.

An economist and a businessman were walking in the woods when they encountered a large and hungry bear. The economist turned to run.

Businessman: "Don’t be stupid, you can’t outrun a bear."

Economist: "But I might be able to outrun you."

What the bear wants is dinner. If you run fast enough, the easiest way for him to get dinner is to eat the businessman instead. Generalizing the principle, to keep someone from doing something you don’t want done you do not have to make it impossible, just unprofitable. You can leverage his rationality.

For another example of the same principle, consider the question of what happens if private citizens are allowed to carry handguns.  An argument against is that since criminals are professionals in the violence business a fight between an armed mugger and an armed victim the mugger will usually be won by the mugger.

Imagine that one little old lady out of ten carries a pistol in her purse. When a mugger tries to mug a little old lady, nine times out of ten she drops the pistol or shoots herself in the foot. One time out of ten she kills the mugger; on average, she is losing. But little old ladies rarely carry enough money to be worth one chance in a hundred of getting killed, so muggers switch to doing something safer. Mugging is still possible but no longer profitable.

My favorite demonstration of the principle is a science fiction story, Margin of Profit by Poul Anderson. The protagonist is Nicholas Van Rijn, head of Solar Spice & Liquors, a big interstellar trading firm. A small and rather unpleasant empire, Borthu, sitting on a trade route between two clusters of stars, has been expanding its navy by seizing passing merchant ships and brainwashing their crews into its service.  Van Rijn consults with his peers, the heads of the other major trading firms, about what they can do. The trading firms are rich and powerful, could fight a war with Borthu and win, but the war would cost more than the profits on the trade route would be worth so they are not going to do it. They could arm their ships. But a warship cannot carry enough cargo to pay for itself, so they are not going to do that either.

Van Rijn gets the others to agree that if one of them solves the problem the others will pay him a fraction of their profits on the route as a reward. He then arms one of his ships out of five. Merchant ships have small crews, Borthu warships have large crews. Four times out of five, Borthu captures a merchant ship and impresses its crew of four. One time out of five the merchant ship seizes the Borthu ship and its crew of a hundred. On average, Borthu is losing.

As Von Rijn explains to his fellows, “not being stupid heads, they will realize this and stop attacking us, and then maybe we can do business.”

The same point is made three times, twice in what the merchants could do but don’t, once in what Van Rijn can do and does. The objective is not to make what the other side is doing impossible, just to make it no longer in their interest to do it.

Market Failure

It is a thousand years ago somewhere in Europe; you are one of a line of ten thousand men with spears. Coming at you are another ten thousand men with spears, on horseback. You do a very fast cost-benefit calculation.

If all of us plant our spears and hold them steady, with luck we can break their charge; some of us will die but most of us will live. If we run, horses run faster than we do. I should stand.

Oops.

I made a mistake; I said “we.” I don't control the other men. If everybody else stands and I run, I will not be the one of the ones who gets killed; with 10,000 men in the line, whether I run has very little effect on whether we stop their charge. If everybody else runs I had better run too, since otherwise I'm dead.

Everybody makes the same calculation. We all run, most of us die.

Welcome to the dark side of rationality.

This is one example of what economists call market failure — a situation where individual rationality does not lead to group rationality. Each person correctly calculates how it is in his interest to act and everyone is worse off as a result.

For a modern version of the same logic, fast forward to Korea. You are a soldier in a foxhole. You are supposed to watch for enemies and, if you see one, shoot him. Sticking your head up to do that might get you killed so instead you keep your head down and fire blindly from time to time in the direction of the enemy. With luck that persuades them to keep their heads down too while persuading your officer that you are doing your job.

There is real world evidence for this story. For the US Army in World War II, about 25,000 bullets were fired for each enemy killed. Many of those killed were by bombs or artillery, so the ratio of bullets fired to enemies killed by bullets was close to a hundred thousand.

For a more mundane example of market failure consider a restaurant meal with a group of friends. It was quiet when you first came in. More people came in, the noise of their conversation made it hard for you and your friends to hear each other, so you started talking a little louder. So did they. So you had to talk still louder. So did they. Twenty minutes into the meal everyone in the restaurant is talking as loudly as possible and nobody can be heard.

And another …

According to the civics class model of democracy, politicians do good things because if they don't people won’t vote for them. For that to work, voters have to know what the politician should be doing and what he is doing. Both are hard; politicians almost never campaign on the theme “I'm the bad guy.”  Nobody has ever introduced legislation to congress with the title “A Bill to Make Farmers Richer and City Folk Poorer,” although legislation designed to do that has been in effect every year of my lifetime.

In order for the civics class model to work the individual voter has to spend time and effort keeping himself well informed about what the politicians he will vote on are doing and what they should be doing, what policies are good or bad. In a U.S. presidential election the chance that one vote will be decisive is perhaps one in 10 million — more in some states, much less in others. If, improbably, your vote does result in getting the better candidate elected there is a large benefit but one that you share with three hundred and thirty million other Americans, just as the soldier in my first story would have shared the benefit of standing instead of running with ten thousand other soldiers.

If you are deciding whether to spend your time and energy doing your job better, educating your kid, reading a good book, or trying to understand what politicians are doing and what they should be doing, the answer for most voters is pretty clear. Better to vote for whichever candidate looks nicer, gives prettier sounding speeches, the one your friends all like, and spend your time and energy doing something that produces a benefit for you with odds of better than one in ten million.

Public choice economists, economists who use economics to try to understand politics, call the result rational ignorance, rational because it is rational not to acquire information if its cost is greater than its value.

Engineering Incentives

Sometimes you can engineer around market failure. My favorite example is a story and a puzzle.

Two Bedouins are riding their camels through the desert with the oasis two miles away. One of them starts complaining:

“This camel, this snail masquerading as a camel, has got to be the slowest beast in all the deserts of Arabia.”

“You think your camel is slow? This tortoise I am riding on …”

The argument about which camel is slower continues; they eventually agree on a bet. The owner of the camel that gets to the oasis last wins the bet and collects a golden dinar from the other.

One of them goes slowly, the other goes more slowly, and an hour later there are two Bedouins sitting their camels stock still in the blazing sun in the middle of the Arabian desert with the oasis still a mile away.

At this point a wise man comes walking along:

“Why are you two idiots sitting your camels stock still in the blazing sun in the middle of the desert with the oasis only a mile off?”

They get off their camels and explain the situation to him. He thinks a moment, then whispers two words to them. They leap back on the camels and ride off for the oasis as fast as they can.

Puzzle: What are the two words?

For the answer see the footnote.⁵

For another example of engineering incentives, consider the legal system of Periclean Athens;⁶ I like to describe it as the work of a mad economist, full of clever ideas that might or might not work. One was their solution to the problem of paying for public goods: If you were one of the richest Athenians you had to produce a public good every other year.

You have heard that we're sending a team to the Olympics this year? Congratulation, you’re the sponsor.”

Or

Look at that beautiful trireme down by the dock. Guess who is paying for it this year.⁷

To get out of the obligation you had to show that you had already produced a public good this year or last or that there was another Athenian who had not done it last year, was not doing it this year, and was richer than you — at which point he gets to do it.

How, in a world without accountants or the IRS, where wealth largely consists of land and slaves, neither of which has a publicly known value, do I prove that you are richer than I am?

Answer (see footnote):⁸

It is an economist’s answer, not an accountant’s.

For a modern example of the same approach, consider the problem of how to set up a horse race where all the horses are about equally good. The solution is a claims race. By entering your horse you are agreeing to sell it for a fixed price, how much depending on the race. You could enter a hundred thousand dollar horse in a ten thousand dollar claims race and be almost certain to win, but it would be the last race you entered it in.

My friend Ami Glaser, an economist at UC Irvine, has a strategy for buying a used car, always a risky proposition. When he finds one he likes he asks whether, if he pays a little more, the dealer will give him a one year warranty. The dealer declines the offer so Ami keeps looking.

Eventually he finds a car whose seller agrees, for an extra five hundred dollars, to provide a one year warranty. Ami buys the car — without the warranty.

Ami also had a proposal for how the antitrust division of the Justice Department should decide whether to allow two companies in the same industry to merge. The argument for a merger is that the two companies can do a better job of producing their products by working together, produce better products or produce them at lower cost. The argument against is that the two companies want to merge so that they can jointly hold production down and the price up.

Ami’s rule was simple:

If the other companies in the industry oppose the merger, you permit it. If the other companies support it, you don't. 

Applications of Economics to Teaching Economics

Engineering around market failure is a useful approach to many things, including teaching. For example…

Halfway through my lecture I pause to ask my students if everyone has followed me so far. Nobody replies. I keep going and discover my mistake when I grade the final.

This again is a conflict between individual and group rationality. The students as a group would learn more if someone had the courage to tell me that they are totally lost, that if I keep going I will be wasting both my time and theirs. But each individual student is afraid of making himself look stupid.

I have a simple solution. Put on the floor in front of each seat a button which a student can unobtrusively push with his foot, at the back of the classroom a large sign showing how many buttons are being pushed. When I notice the eyes of my audience beginning to glaze I pause and ask everyone who has followed me so far to push his button. The number two appears on the screen. I go back and start over.⁹

A student is taking an exam; the only question left is one he does not know the answer to. He spends the final ten minutes of the exam writing something sufficiently unclear so that, with luck, the professor will think he knows part of the answer and give him partial credit.  Doing that consumes his time, consumes the time of the professor grading the exam, and makes grading less accurate.

My solution to this problem starts from Greek antiquity.

The Delphic oracle had told one of Socrates’ friends that Socrates was the wisest man in Athens. Socrates responded that he couldn’t be, since he did not know anything. After interrogating other Athenians as to what they knew he concluded that they did not know anything either — but thought they did.

Which made him the wisest man in Athens.

On my exams, you got credit for knowing that you don’t know something. If you tried to answer a question and failed you got no credit for that question. If you left it blank or wrote “I don’t know,” you got 20%. Unless you knew enough of the answer to get more than that, you were better off leaving it blank. That saved me time, saved you time, made my grading a little more accurate.

Incentives are relevant to me as well as to my students. A very long time ago I wrote a price theory textbook. Many years later I decided to rewrite it as a book aimed at the intelligent layman. My model was The Selfish Gene, a book on evolutionary theory that I had read for the fun of it.

It occurred to me that this book, unlike a textbook, was one that nobody was going to be forced to read; if at any point the reader lost interest he would stop. To deal with that problem I started every chapter with a hook, a puzzle, that would get resolved by the end of the chapter.

Consider in contrast the incentives for a textbook author. His incentive is to write a book that professors will assign, whether or not their students will enjoy reading it.

1. The analysis is worked out, both for that model and for one where there is no way of offering a potential spouse anything beyond your characteristics, in Chapter 21 of my  Price Theory.

2. Jon Birger, "Unequal Gender Ratios at Colleges Are Driving Hookup Culture.”

3. A rule sometimes imposed by courts under the label of “implied warrantee of habitability.”

4. For the details of the argument, applied to a different restriction on rental terms, see Chapter 7 of my Price Theory.

5. “Switch camels.”

6. For details, see the chapter on Athens in my Legal Systems Very Different from Ours.

7. A trireme was a war ship, so called because it had three banks of rowers and oars, one above the other. The cost was actually divided between two wealthy men; part of their job was either captaining the ship or hiring someone else to do so.

8. I offer to trade everything I own for everything you own. If you turn me down you have admitted you are richer than I am.

9. I have never actually done it — I am by nature a theorist, not an experimentalist — but feedback systems along these lines have been set up by others using wireless devices, garage door openers or, more recently, cell phones.