Does Kim Jong Il Know the Brouwer Fixed Point Theorem?

Great column by David Brooks in the NYTimes:

When it comes to understanding the world's thugs and menaces, I'd trust the first 40 names in James Carville's P.D.A. faster than I'd trust a conference-load of game theorists or risk-assessment officers. I'd trust politicians, who, whatever their faults, have finely tuned antennae for the flow of events. I'd trust Mafia bosses, studio heads and anybody who has read a Dostoyevsky novel during the past five years.

David Brooks realizes that intuition matters. Knowing what motivates other people--knowing how they think, how real, flesh and blood people go about the business of chosing actions, knowing what they value and why--matters. And never more than when violence is involved. Theorists distance themselves from intuition, almost perversely. I would not want a theorist to command my army. It is good to remember this when listening to Krugman lecture us on Iraq and North Korea. (He has a lovely little model to back up everything he says, I have no doubt).

Dr. Edgeworth Boks, who is an economist specializing in game theory (for those new to this blog), would engage in armed conflict by choosing a best response to what he perceived would be his opponents strategy. The good doctor thnks all others would do likewise. I asked him for a direct quote on the subject. "I will choose an optimal stratgey," he said, "and my opponent, realizing that I will do so, will choose a best response to my best response, and I, having anticipated that he will have anticipated that I will do so, will have limited my choice to best reponses to his best response to my best response, and so on, so that eventually, in a way that is hard to make explicit, my opponent and I will find ourselves on the blessed equilibrium path, with each of us best-responding to the other, and from which there us no deviation by either of us that will yield a preferred outcome." So says Boks.

Picture now, if you will, Dr. Boks' opponent in this conflict. The enemy is a person ufamiliar with game theory. He is a person who believes passionately that his cause is just. He does not speak of equilibrium paths or best responses, but says : "I would die to protect my homeland."

Who would you rather have on your side?

P.S.Babcock

Comments

[Vinayak Nagaraj]

Hehe... just starting out on game theory myself at the LSE... but it seems the kind Mr. Il's best response function would make for an interesting thought! Which is the more credible threat... that he has the nukes... or that he'd be crazy enough to use them?

[P.S.Babcock]

Yes, Vinayak, I agree. The interesting aspect of the problem is to discern Kim's valuations of the various outcomes! The task of finding "equilibrium" strategies is secondary at best and provides little insight. The West tends to focus on elaborate strategies that involve bluffs or cheap talk. "Abide by this agreement or else!" But we don't back up those threats. So anyone playing against us knows not to believe what we *pretend* will be our valuation function or best response function. Economics and game theory offer almost nothing in this context. Insight about your opponent's beliefs and pysche, knowledge of how real human beings behave (the type of knowledge economists rarely possess) is decisive here.

Hope you enjoy game theory! It's a lot of fun.

[Vinayak Nagaraj]

I cannot agree more! Since Im in the summer school course, we have three hour lectures every day, so we're covering stuff at a really fast pace. But each time I stop to think, it just seems like game theory can be a completely useless tool if we leave our analysis of the opponent to chance. Everything from the prisoner's dilemma straight until perfect bayesian nash equilibriums... they all depend upon this one factor that we NEVER seem to think about enough.

[Simon]

Maybe he knows kakutanis fpt though