Monday, March 31, 2008

Wine, Anti-Foreign Bias, and Make-Work Bias

Last night before bed my wife and I watched "Legislative Report" (that's the WMP verison; Quicktime version here) on Nashville Public Television (if that link goes dead, you should be able to access the episode, from March 30, in this archive). The first issue covered was the recent debate over whether or not to allow sales of wine in grocery stores. Currently wine can only be sold in specialty retail stores.

Chip Christensen, president of the state retail liquor lobby spoke. Here's my transcript of what he had to say:
"This is a very simple issue between big and small. Um, every retailer in this room, every retailer in this state is a Tennessee resident. Few, if any, of the grocers we're talking about are Tennessee residents. We're talking about big, big businesses. The public doesn't realize the cost, or the consequences, that go along with this legislation. Kroger, Wal-Mart, Walgreens will not hire one additional employee to sell wine in their stores. I will absolutely guarantee you we will lose employees. There will be a net loss of employment if this bill passes. Citizens didn't start this movement; very high-paid public relations firms by very large grocery chains started this movement. Lastly, it's a regulated business, it needs to be regulated, and I think that going forward with this legislation is going to make that extremely difficult, um, and it's going to hurt 525 tennessee residents who are small-business-people."
I have posted before about Bryan Caplan's book, The Myth of the Rational Voter, and his concepts of make-work bias and anti-foreign bias. I think both are contained in the quote above.

Make-work bias is the belief that the purpose of production is to create jobs, rather than to create output. Let me paraphrase an example from the book: A businessman (or maybe an American or a diplomat; I can't remember all the details of the story, and my copy of the book is loaned out at the moment) visits a Chinese dam under construction. There are hundreds of Chinese workers digging with shovels. He asks the supervisor why they are using shovels, rather than powered equipment (that is, why labor rather than capital). The supervisor answers that the shovels are superior because they result in the employment of more people. The businessman responds that if their goal is to maximize employment, they should have them dig with spoons.

The point is that the goal of production is to produce stuff. Every worker that is devoted to producing a dam is unable to work on something else. By making them dig with shovels, or with spoons, it is true that the workers are all employed, but much of this employment is wasteful. That is, the supervisor is ignoring the opportunity costs of employing all these workers, because if the workers were given one steam-shovel many of them could be employed in more productive activities. It is unlikely that this bill would result in a net loss of jobs; jobs would simply shift from one industry to another as the workers became available.

The same goes for wine retailers. Yes, allowing grocery stores to sell wine will destroy some liquor retailing jobs--but this is a benefit, not a cost. All those workers are now free to do something more important and valuable to society. Also, consumers get to enjoy more wine. This is a net gain.

Some of you won't buy this argument, so I present you with a counter-proposal, along the lines of giving construction workers spoons. Suppose that we pass the "Tennessee Liquor Retailer Full Employment Act of 2008". This law will continue the requirement that win be sold only in retail stores, but will add another measure intended to create jobs: You must find your retail store through a liquor retailer retailer. That is, you go to the liquor retailer retailer, and he gives you a selection of liquor retailers from which to choose. Isn't this a wonderful idea? Now we've created another set of jobs, those of the liquor retailer retailers! But why stop there? Why not have liquor retailer retailer retailers? And we can stack on even more middlemen! Let's require that you contract with yet another firm to get the wine from the store to your car. And maybe we can have yet another business open your wine for you. Think of all the jobs that will be created!

Enough about the make-work bias. The other interesting bias revealed in the quote is a strange sort of anti-foreign bias. Technically, the lobbyist isn't really talking about foreigners, i.e., non-Americans. Apparently he just dislikes non-Tennesseans. It's not clear to me why I should want to preserve the job of some random Tennessean over that of some random (possibly) non-Tennessean. It seems to me to be some kind of strange bigotry. I can see how it would pay for a Tennessee politician to hold such bigoted beliefs, however, since it may help them get elected. Apparently a bill that would benefit consumers and possibly some non-Tennessee producers is a bad idea because it would remove some unnecessary Tennessee middlemen.

And finally, I found the quote about regulation interesting. One should be wary when an industry cries out for the government to regulate it, because the odds are good that the industry doesn't want the regulation for the sake of the consumer. Such regulation is often intended to reduce competition.

UPDATE:
The claim that grocery stores would not hire a single person to help them sell wine is a bit silly anyway, but Art Carden (who blogs over at Division of Labor) points out that the cheaper wine should create an income effect that results in more employment somewhere. That is, the cheaper wine leaves consumers with more money to spend on goods and services. This expenditure creates jobs in some other part of the economy.

Thursday, March 13, 2008

More Fun and Games from Econ 307

Here are some other fun games from Econ 307 (officially entitled Managerial Economics, although it is really a course in Industrial Organization).

Rock, Paper, Scissors as a Simultaneous Game:
Find the equilibria of the following game.


Player 2

Player 1


Rock

Paper

Scissors

Rock

0 , 0

-1 , 1

1 , -1

Paper

1 ,-1

0 , 0

-1 , 1

Scissors

-1 , 1

1 , -1

0 , 0

Answer: There is no pure-strategy equilibrium in this game. There is a mixed-strategy Nash equilibrium, however, in which each player plays each strategy with probability 1/3. If you model this as an extended form (or sequential) game, the second player always loses, for obvious reasons. (I'd post that model, too, but it would involve working up some images, and I don't have time for that at the moment).

Warfare as a Collective Action Problem:

A soldier considers marching into a hail of gunfire with his comrades, or running away. All his comrades face the same decision. Suppose we model this decision by considering a representative pair of soldiers:


Soldier 2

Soldier 1


Fight

Run Away

Fight

10,000 , 10,000

-10,000,000 , 10,000,000

Run Away

10,000,000 , -10,000,000

10,000,000 , 10,000,000

What is the equilibrium outcome? Can you tell a story that makes sense of the payoffs in the table? (That is, why does the fellow who fights when his buddies run earn -10,000,000? Why do they get smaller but positive payoffs if they all fight? Why do they get 10,000,000 if they all run away?)

Answer:
(Run Away, Run Away) is the dominant strategy equilibrium. Here is one story that makes sense of the payoffs:

If all the soldiers fight, they stand a fair chance of surviving, as they might win. Therefore the payoff is positive. But if only one soldier tries to fight, he is certain to die, while those who run away are happy to be alive. Finally, if all the soldiers run away, they’ll all be happy to survive, although the war will not go well if this persists.

Suppose we alter the story so that an officer stands behind the soldiers with a machine gun. The machine gun is pointed at these (i.e., his own) soldiers, not at the enemy. Why might the officer do this, and how does it alter the payoffs?

Answer:

The officer exists to change the positive payoffs from running away into negative payoffs. By doing so he makes (Fight, Fight) the dominant strategy equilibrium. This example is not so far-fetched; the Soviet Army did this during the battle for Stalingrad during World War II (and probably on many other occasions). It has long been the case that soldiers who desert their post can be court-martialed (The excellent Kirk Douglas movie Paths of Glory deals with the case of soldiers in WWI who refused to march to certain death in the no-mans-land between the trenches, and were court-martialed and executed as a result). This should serve as a reminder that there can be ways around the Prisoner’s Dilemma. Using a third party as an enforcer is a powerful option.

Helmetless Hockey Players

(This problem was inspired by Robert Frank's The Economic Naturalist.) Consider the following game, in which hockey players choose whether or not to wear a helmet. Consider players 1 and 2 to be representative of any two players in a hockey game.


Player 2

Player 1


Wear a helmet

No helmet

Wear a helmet

0 , 0

-2 , 3

No helmet

3 , -2

-1 , -1

What is the equilibrium of this game? Is this the socially optimal outcome? Of what famous game does this remind you?

Answer: (No helmet, No helmet) is a dominant strategy equilibrium. This is similar to the prisoner’s dilemma.

Can you tell a story that makes sense of these payoffs and this equilibrium? To put it another way, why would hockey players support a rule that all players have to wear helmets, and why would such a rule be necessary?

Answer: If one player goes without a helmet, he has an advantage over the other players, in terms of visibility, mobility, and reduced weight. But each player reasons this way; “if other players are going without helmets, I, too, must forego my helmet, so that I will not be at a disadvantage.” As a result, they all end up going without helmets, eliminating the advantage of not having them while simultaneously putting themselves at greater risk of being injured. They would all favor a rule because it would prevent coordination failure--they all end up safer, and no one has an unfair advantage.

Hamsters and Rangers Everywhere, Rejoice!

Today I put a reference to Baldur's Gate 2 on a game theory quiz, and one of my students actually caught it (that is, he actually recognized that it was from Baldur's Gate 2). This was the first time that a student has acknowledged playing a computer game (versus a console game) and the highlight of my day. Here was the question:

An evil wizard captures you and your significant other and imprisons you each in cages in separate rooms. You cannot communicate with each other. The wizard comes to you and says “There is a button in your cage. Your significant other also has a button in his/her cage. If you press the button, but your loved one does not, you will die, but I will spare the life of your loved one. If you do not press the button, but your loved one does, you will live and he/she will die. If you both press your buttons, you will both die. If neither of you press your buttons, you will both die.” What will you do? That is, what kind of equilibria does the following game have, and what are they?


Significant Other

You


Press Button

Don’t Press Button

Press Button

-1 , -1

-1 , -2

Don’t Press Button

-2 , -1

-1 , -1

Answer: The game has two Nash equilibria, (Press Button, Press Button), and (Don’t Press Button, Don’t Press Button).

Extra Credit (2 points): Can you come up with an explanation for why the payoff is -2 if you don’t press the button, but your significant other does?

Answer: The only answer that makes sense to me is that you suffer shame and sadness, since you survived and your loved one sacrificed his or her life for you, but perhaps you can come up with a better explanation.