Mechanism design theory may sound like an ironically named rock band from the 1970s. In fact, it is a development of the economic discipline of game theory. And for three US economists—Leonid Hurwicz, Eric Maskin and Roger Myerson—these intellectual games have merited the 2007 Nobel Memorial Prize in economics.
The Royal Swedish Academy of Sciences, which awards the prize, has given the prize to game theoreticians four times since 1994. The latest is for the study of how best to make decisions in institutions. The prize winners have addressed such questions as setting rules in regulated industries and arranging voting systems.
Game theory is appealing because it is supposed to be useful, unlike most micro-economics, which is often mired in hard-to-interpret equations.
Game theory promises help on concrete problems such as tax rates and price-setting.
It sounds promising. For example, Myerson is closely connected with the discovery that there is often no better way to decide on the price of something than a double-auction, in which both buyer and seller submit bids.
That information might be helpful to labour negotiators.
Or it might not. The practical result comes at the end of a long series of equations, which all rest on some simple assumptions about human nature.
The most notable is that each individual strives after only one selfish goal—“maximizing his or her expected payoff”, as the Nobel citation’s background document explains.
But people aren’t really such utility optimizers.
They are often willing to make economic or personal sacrifices for the sake of other goods. They are often not good judges of what is good for them.
In any successful institution, trust prevails over selfishness. It is not surprising that game theory has had little effect on actual economic practices.
Where it has, for example in utility regulation and spectrum auctions, it’s hard to see that the theory has made economic life better.
The Swedish academy is right to reward economists, who try to approach real—rather than purely theoretical—problems. But until the profession is willing to deal with the way real people behave, the problems are going to remain unsolved.