IN GAME THEORY, WINNERS SEND, RECEIVE BETTER SIGNALS

Date: Saturday, October 22, 2005

Knowing basic economics has a good deal in common with knowing basic probability if you're playing poker. It can't guarantee you'll win, but you have much better odds of avoiding big losses. Thomas Schelling explained basic economics to a lot of people.

Schelling, professor emeritus at the University of Maryland, shared this year's Nobel Prize in economics with Robert Aumann, of the Hebrew University in Jerusalem, "for having enhanced our understanding of conflict and cooperation through game-theory analysis," according to the announcement by the Royal Swedish Academy of Sciences.

Game theory is sometimes called "interactive decision making," and it's not necessary to get into the mathematical analysis to recognize that many of the decisions we make are premised on our understanding of what the other guy is thinking, or even what we think he understands about what we're thinking, and so on. And part of the game -- think bluffing in poker -- is in signaling, sometimes deceitfully, about what we're thinking.

I'm not familiar with Aumann's work, but I was a delighted reader of Schelling's 1984 book, Choice and Consequence: Perspectives of an Errant Economist. I confess I'd forgotten the subtitle, but it captures what I remember of the book's spirit, which is to look at what real people do in real situations, as opposed to what bloodless abstractions of people do in non-errant economists' mathematical models.

Here's one example I remember; how does it happen that neighborhoods can become increasingly racially segregated even though most of the people in them would prefer to live in integrated neighborhoods?

One possibility is that different populations may prefer integration but have varying ideas about how much of it is comfortable. If whites on average feel that a neighborhood with up to 20 percent black residents is comfortably integrated, while blacks on average feel that a neighborhood with up to 50 percent whites is comfortably integrated -- feelings that need not derive from any greater or lesser tolerance on either side, merely from expectations based on the relative sizes of the populations -- then there's an unstable zone in the middle where whites will tend to move out and blacks will tend to move in.

Tyler Cowen, a professor of economics at George Mason University, claims Schelling as his mentor at Harvard. On his blog MarginalRevolution.com (a post dated Oct. 10, when the award was announced), he briefly surveys other issues that have figured in Schelling's work (and links to interviews and articles accessible to non-technical readers).

The logic of nuclear deterrence, which was explored in Schelling's 1960 book, The Strategy of Conflict, was specifically cited by the prize committee. It said he showed that "a party can strengthen its position by overtly worsening its own options, that the capability to retaliate can be more useful than the ability to resist an attack, and that uncertain retaliation is more credible and more efficient than certain retaliation."

Does that remind you the least little bit of Dr. Strangelove? "I recall Tom telling me he was briefly an adviser to Kubrick," Cowen says.

Precommitment: You can make yourself better off by deliberately limiting your choices in advance. You may be playing a game with your later self -- you know you'll want to drink at the party that night, but in the sober light of morning you'd like to rule out that option in a way that ensures you can't change your mind later. Or you may be facing an opponent whose best option is unacceptable to you, so you take irrevocable legal steps to ensure that option is no longer available.

Of course, you may discover that all the remaining options are, in fact, worse. Even the best strategy doesn't win every time.

Focal points: Some games are an attempt to arrive at a cooperative strategy in circumstances of imperfect communication. If you are to meet a friend in a certain city on a certain date, but you don't know the time or the location because messages went astray, where and when do you turn up? You're not only trying to figure out his strategy; you know he's trying to figure out yours, too.

For New York City, the canonical example of a focal point was "under the clock at Grand Central Terminal" but even that required a certain commonality of experience. I asked a longtime Denverite what it would be here and he said, "the lobby of the Brown Palace" hotel -- but I doubt that would have occurred to me if I were trying to find him. Not that I can think of anything better.

If you like off-beat books about economics, Choice and Consequence is one not to miss. It's high on the list of economics books recommended by Steven D. Levitt and Stephen J. Dubner, authors of Freakonomics: A Rogue Economist Explores the Hidden Side of Everything. Kindred subtitles, at least. Maybe kindred spirits as well.