How Game Theory Made Me a Better Businessman

Anyone who has taught microeconomics has, at one time, had to deal with the question, That's all very nice, professor, but how am I supposed to use any of this in the real world? If I'm running a business, I'm not going to know the demand curve for the product and I'm not going to be in a perfectly competitive or a monopoly market. What's the point?

From my own personal experiences as an economist and a business person, there are a number of useful lessons the business world can learn from microeconomics. The first is from the world of game theory.

Quick warning: I'll try not to let this get too 'preachy', but it is Sunday and my full first name is Michael Patrick so I hope you'll forgive any accidental sermonizing.

Most second year microeconomics courses introduce game theory and in particular, the Prisoner's Dilemma game:

Tanya and Cinque have been arrested for robbing the Hibernia Savings Bank and placed in separate isolation cells. Both care much more about their personal freedom than about the welfare of their accomplice. A clever prosecutor makes the following offer to each. “You may choose to confess or remain silent. If you confess and your accomplice remains silent I will drop all charges against you and use your testimony to ensure that your accomplice does serious time. Likewise, if your accomplice confesses while you remain silent, they will go free while you do the time. If you both confess I get two convictions, but I'll see to it that you both get early parole. If you both remain silent, I'll have to settle for token sentences on firearms possession charges. If you wish to confess, you must leave a note with the jailer before my return tomorrow morning.”

The “dilemma” faced by the prisoners here is that, whatever the other does, each is better off confessing than remaining silent. But the outcome obtained when both confess is worse for each than the outcome they would have obtained had both remained silent. A common view is that the puzzle illustrates a conflict between individual and group rationality. A group whose members pursue rational self-interest may all end up worse off than a group whose members act contrary to rational self-interest.

Students then go on to learn that in a situation where two players play the repeated game a number of times (but the number of games that will be played is unknown to the two players) that a much larger set of strategy options are available. By playing the games a repeated number of times, players can 'punish' the other for defecting, in a way that isn't possible in the single-shot version of the game.

A difficulty emerges, however, in the unknown-endpoint, repeated version of this game: there is no known dominant strategy that a player should use. An attempt to determine what the dominant strategy might look like is detailed in the Axelrod and Hamilton paper The Evolution of Cooperation (1981) and a follow-up 1984 book by Axelrod of the same name. These are absolute "must reads" for an undergraduate in economics - I recommend them to my students. The article and book describe a tournament by the authors. Wikipedia has an excellent summary:

Axelrod initially solicited strategies from other game theorists to compete in the first tournament. Each strategy was paired with each other strategy for 200 iterations of a Prisoner's Dilemma game, and scored on the total points accumulated through the tournament. The winner was a very simple strategy submitted by Anatol Rapoport called "TIT FOR TAT" (TFT) that cooperates on the first move, and subsequently echoes (reciprocates) what the other player did on the previous move. The results of the first tournament were analyzed and published, and a second tournament held to see if anyone could find a better strategy. TIT FOR TAT won again. Axelrod analyzed the results, and made some interesting discoveries about the nature of cooperation, which he describes in his book.

Axelrod concludes that TIT FOR TAT is successful because it contains four basic characteristics (again, summarized by Wikipedia):

1. Be nice: cooperate, never be the first to defect.
2. Be provocable: return defection for defection, cooperation for cooperation.
3. Don't be envious: be fair with your partner.
4. Don't be too clever: or, don't try to be tricky.

In the business context, I would condense this list of four characteristics to two basic principles:

1. Focus on creating value not on how big your share of the pie is. (A combination of characteristics 1, 3 and 4).


2. Don't do business with people you don't trust (or, alternatively, don't pet a dog that has already bitten you. Characteristic 2).
   

The first principle, taken to a logical extreme, seems pretty crazy. If you only focus on making the pie larger, aren't you leaving yourself wide open to be exploited? Yes and no. If someone is obviously trying to rip you off or exploit you, you shouldn't deal with them (principle two). People who you would want to associate with will offer you something on the spectrum of a 'fair deal'. It is true, however, that you probably leave a fair bit of money on the table if you don't fight hard for a larger share of the pie. I often hear from friends (and my wife) things like "you know, you probably could have gotten more on that deal" or "you probably could have made more money for that job had you held out longer". I have no doubt that they're correct. But what you do get is a reputation for someone to do business with. While you'll get less on any one deal, you'll be presented with a lot more opportunities, many of them quite lucrative. Having more opportunities allows you to be pickier with who you deal with and what you will accept in exchange for your most precious commodity - the 24 hours that make up a day. Plus people will pick up the phone when you call them with a proposal - they know you have something good to offer.

Contrast this with the kind of guy who focuses on his share of the pie. Anyone who has played fantasy baseball probably knows this guy. He's always looking to trick people into making bad deals and occasionally is successful and takes someone to the cleaners. But we all know what ends up happening to that guy - after awhile, nobody returns his calls or e-mails. He loses opportunities to make his team better, because we assume anything he has for us is a bad deal and is not (in MBA speak) win-win.

I believe that all of us fall somewhere on the 'make the pie larger' and 'how big is my piece' spectrum. Life is a repeated game and in the long-run, you'll do much better being closer to the 'pie larger' side of the spectrum. Robert Axelrod and microeconomics taught me that.

Comments

[Greg Ransom]

YOU are the one who needs to learn from your student. Many of the central sientific lessons of economic science are contained in this fact which your student is bringing to your attention. And the Econ profession shows it's scientific incompetence by not "getting it" about the lesson for scion e of these facts.

I've been the student, and I've seen the incompetence of the professor on the other side of the question who remains oblivious to the signal import of the fact articulated by the student.

Mike wrote.

"Anyone who has taught microeconomics has, at one time, had to deal with the question, That's all very nice, professor, but how am I supposed to use any of this in the real world? If I'm running a business, I'm not going to know the demand curve for the product and I'm not going to be in a perfectly competitive or a monopoly market. What's the point?

From my own personal experiences as an economist and a business person, there are a number of useful lessons the business world can learn from microeconomics."

[Tim Worstall]

Not quite sure this is game theory but at least vaguely on subject.

I run a little company (three people) which acts as a wholesaler and expert in a couple of the very weird and wonderful metals. Very small markets (a few tonnes a year globally) and we're the largest company in these markets. No, we don't make a fortune but a living.

There's essentially two ways you can be a wholesaler.

1) Have a pot of $ to use as capital. Purchase materials and stock and or refine them using said capital.

2) Use "trust capital" to do the same thing. Sure, it's not easy getting 45 day credit terms out of a Chinese or Russian supplier. But it can be done.

A transaction might involve purchasing material on credit in one country, refining it in another and then shipping to a customer in a third. No money is flowing until the final customer gets it, approves it and pays for it.

That trust capital needs to be nurtured of course: there have been times when we've delivered under quality goods. The reaction has to be, Ooops, yes, agreed, we'll get you some more, no worries, please send back the bad stuff at your leisure.

There was one time when a customer went into Chapter 11 before paying us....we had to turn to the supplier and tell them this and then say, well, give us an extra 6 weeks will you and we'll get your money from somewhere.

After 15 years of this we still run on no $ capital (we've a few thousand $ in the bank right now) and yet there's $550,000 of material in shipment to us right now, supported essentially by our word that payment will be forthcoming: trust capital.

As an example of quite how far such trust can go. A long term customer, I just had to impose a 140% price rise on him. Tim, you sure the global price has gone up that much? Yes, sorry, it really has. OK then.

If I ever tried to be tricky with any of the people in this little network, suppliers or customers, the entire edifice would fall over immediately.

Funny how that archaic English phrase "my word is my bond" still seems to have some relevance really.

[hosertohoosier]

"Life is a repeated game and in the long-run, you'll do much better being closer to the 'pie larger' side of the spectrum. Robert Axelrod and microeconomics taught me that."

I think that outcome is rather contextual. A tit for tat strategy does well in tournament settings, but its success is based on a number of factors - namely the presence of repeated interactions and the possibility of mutual gains. However, there are many settings in life where there are no repeated interactions. If I were a pimply-faced food service worker in an airport, for instance, I would have little reason to develop rapport with customers.

Similarly, there are many occasions where interactions involve relative, not absolute gains. If you are playing chess there is no potential to make both players better off, since the object of the game is to destroy one another. In international relations (where security is a big deal - you can't repeat interactions if you are forced out of the system), for instance, the implications of game theory tend to have a darker tone than in economics/business (and even then, if I'm Blue Ray and you're HD DVD, you are going down, pal).

What I get from Axelrod isn't that it is a good idea to be a nice guy. Rather that it is a good idea to be a nice guy in the right institutional settings. If I am the one building institutions, of course, it might make sense to incorporate some of the insights of the reciprocity argument. For instance, Japanese Keiretsu tend to have long-term relationships with suppliers, rather than jumping from low cost supplier to low cost supplier. This has enabled them to perform well in industries characterized by economies of scale, but less well elsewhere.

PS: I think you may also be talking as much about social capital vs. physical capital here.

[Mike Sproul]

"That's all very nice, professor, but how am I supposed to use any of this in the real world?"

Give economics some credit! Businesses use isocosts/isoquants to optimize input use. When selling into two different markets, they set marginal revenues equal in both markets. When operating two factories, they set marginal costs equal in both factories. They set marginal costs equal to marginal revenues, they set transfer prices equal to marginal manufacturing costs. They choose investments to maximize net present value. Economics tells them what to expect when there are price ceilings, price floors, taxes, and subsidies.

With all that useful stuff to talk about, I hope you're not wasting students' time talking about quasi-concave utility functions, but if you are, students will naturally ask how this applies to the real world.

[RSJ]

Business don't set marginal costs equal to marginal revenues. Or equivalently, you need to define "marginal costs" to include a variable profit requirement that makes "marginal costs" mostly meaningless, as it is a function not only of the specific businesses actual costs, but of return expectations and growth expectations.

Businesses have a required internal return on capital, which is a function of how quickly the business expects to grow, and this is the dominant factor in setting required margins. As the business grows, it either continues to innovate and maintain high margins (high returns on capital and re-investment), or it slows down over time, eventually becoming a cash-flow pig. The typical return on capital is in the 8-10% range, and they set costs in order to meet that required return.

Good data on this is here:
http://pages.stern.nyu.edu/~adamodar/

[Panayotis]

Food for thought.

Imperfection is defined by the criteria of a)asymmetry, b)heterogeneity, c)disintegration and d)dispersion. Let us deal with asymmetry in this comment.
An asymmetry of advantages implies imbalanced scarcity of secure resources that bounds practices and unbalanced danger of estimates of value that bounds decisions.For example, a buyer of complex products of forward performance and delivery has an inferior ex ante estimate of performance (issue of diversity) and an inferior ex post resource constraint of delivery (issue of evasion) than the seller who knows more about performance and can control approximately forward delivery. Thus asymmetry implies that the price is set by the seller, not neccessarily equal to the notional value of the buyer and the quantity is set by the buyer, also not neccessarily equal to the notional level of resources and production of the seller. (This is ONE source of the Keynesian solution of equilibrium stasis). Games that are played must take asymmetry into consideration.

[GA]

"If I were a pimply-faced food service worker in an airport, for instance, I would have little reason to develop rapport with customers."

Direction of causality may be the other way, or at least be reinforcing: would those who have the right skills, including social skills/capital, be here in the first place? If the pimply-faced worker proves good at developing that rapport, will they stay in that job or be offered something better, where those skills have a higher value, by someone with a nose for talent?

I'm not (intentionally) being facetious here. The institutional setting is both a sorting system and a reinforcement mechanism (of what behaviour is rewarded in that institution); the exact mix and direction of change depends on a lot of other issues.

Or going back to the original post: there's a strategy proposed that tends to be rewarded highly over time. It may be poorly rewarded in particular contexts - and yet, over the repeated interactions of a lifetime/career still be the dominant strategy.