Tuesday, June 07, 2005

Monkeys, risk aversion, game theory, inflation, and more

Steven Levitt on monkeys and investors:

Monkey Business - New York Times: "...economics is increasingly being recognized as a science whose statistical tools can be put to work on nearly any aspect of modern life. That's because economics is in essence the study of incentives, and how people -- perhaps even monkeys -- respond to those incentives. A quick scan of the current literature reveals that top economists are studying subjects like prostitution, rock 'n' roll, baseball cards and media bias."

So why not monkeys!?!? Sure enough, researchers are teaching mokeys to use money. Really! And much more. For example what happens when people (err, monkeys) don't cooperate? (they get punished), when they are too cooperative? (they get less food), and even when there are wealth effects or inflation (they are not always rational). Amazing stuff.

From the NY Times:
"...were fairly cooperative but still showed a healthy amount of self-interest: over repeated encounters with fellow monkeys, the typical tamarin pulled the lever about 40 percent of the time. Then Hauser and Chen heightened the drama. They conditioned one tamarin to always pull the lever (thus creating an altruistic stooge) and another to never pull the lever (thus creating a selfish jerk). The stooge and the jerk were then sent to play the game with the other tamarins. The stooge blithely pulled her lever over and over, never failing to dump a marshmallow into the other monkey's cage. Initially, the other monkeys responded in kind, pulling their own levers 50 percent of the time. But once they figured out that their partner was a pushover...their rate of reciprocation dropped to 30 percent -- lower than the original average rate. The selfish jerk, meanwhile, was punished even worse. Once her reputation was established, whenever she was led into the experimenting chamber, the other tamarins ''would just go nuts,'' Chen recalls. ''They'd throw their feces at the wall, walk into the corner and sit on their hands, kind of sulk.''

* or even better in discussing what happens when the reward is risky (i.e. the monkey may either win a grape or lose a grape. (expected payoff was identical). It turns out that monkeys favored gambling to win, rather than to not lose. This is similar to people.

"In similar experiments, it turns out that humans tend to make the same type of irrational decision at a nearly identical rate...The data generated by the capuchin monkeys, Chen says, ''make them statistically indistinguishable from most stock-market investors.'"

I do not know whether to laugh or to be worried that the monkeys are as intelligent as the average investor! ;)

As an aside, Steven Levitt is good. He is smart, interesting, and I would suppose (I do not know him) he may even be a nice guy. But his PR person must really be something! Not only does Levitt teach at Chicago, but he also has a best-selling book (Freakonomics), he has a blog, he was also on C-span this weekend, and he writes a column for the NY Times.

No comments: