Womanhood on Wall Street, Fame at the Top, Tennis for Economists, and Other Matters. Tennis at Harvard
By DANIEL SELIGMAN REPORTER ASSOCIATE Edward C. Baig

(FORTUNE Magazine) – You are playing tennis. Your opponent frequently challenges you by coming to the net. Assuming that you try to win every point, you basically have two choices when he does that: lob over his head or try a passing shot. Which do you try? Wait -- you obviously won't want to use the same shot every time. You should be trying some mixture of lobs and passing shots. Okay, but what's the optimal mix?

For the answer to this question, and also to the question of whether people in general are economically rational, we recommend an absorbing article in the November/December issue of Sciences. Entitled ''Making Up Our Minds'' and written by Harvard psychologists Richard J. Herrnstein and James E. Mazur, the article reports on a number of experimental findings suggesting that people are far less rational than assumed in economic theory. Instead of maximizing the bottom line, they have a powerful tendency to ''meliorate'' -- meaning that they try to maximize outcomes in each sector of the operation. Sometimes that works out fine. But sometimes the bottom line would be higher if they reduced returns in one sector (by, say, making it a loss leader). This brings us back to tennis and the choice of shots. We start out knowing that lobs are more effective if you have recently been using passing shots -- and vice versa. The Sciences article describes experiments in which testees at a desk played a simulated tennis game based on certain hypothetical figures. They were told that if they had been hitting nothing but lobs, then another lob would on average be worth only 0.1 point whereas a passing shot would be worth 0.4 point. And if they had been hitting nothing but passing shots, then another passing shot would be worth 0.3 point but a lob would be worth 0.9 point. It can be shown that on these assumptions, there is a point at which lobs and passing shots are equally effective. You reach that point when you lob two-thirds of the time and hit passing shots the other one-third. And, in fact, that is the ratio to which the experimental subjects keep gravitating. Using that ratio, they win the point precisely 36.6% of the time. But it's the wrong ratio. They would do better if they stopped treating the two shots as separate cases and trying to maximize each case -- but instead went for ''an overarching strategy'' (the authors' term) that accepted a slightly lower than possible return on passing shots in order to get distinctly superior return on lobs. You maximize the bottom line (and win 43.6% of the points) when you increase the lob ratio to 38.9% and reduce the passing shot ratio to 61.1%. Suppose it's true that individuals are more often irrational about economic issues than previously supposed. What follows from that? One conclusion that should not follow -- as Herrnstein and Mazur instantly agree -- is that the government needs to step in and tell people what to do with their money. There is no evidence that government decisions are more rational than private decisions. Or that the government knows how to play tennis better either.