Is The Dismal Science Dazed And Confused?


A New General Theory of Social and Economic Behavior

By Paul Ormerod

Pantheon 217pp $24

By calling his first book The Death of Economics, Britain's Paul Ormerod didn't leave much room for a sequel. Economics: Still Dead would hardly have leapt off the bookshelves. Ormerod's solution is to announce that economics isn't dead after all--only sick. And he has some ideas for how to heal it. In Butterfly Economics, Ormerod expands on the main point of his well-received first book, namely that his fellow economists must shake off 19th-century determinism and accept that the world is chaotic, nonmechanistic, and unpredictable. The title comes from chaos theorists' "butterfly effect," which is the concept of snowballing: The swirls of air from a butterfly's wings in the Pacific could theoretically trigger a hurricane days later in the Atlantic.

This is a presumptuous book. Its subtitle, A New General Theory of Social and Economic Behavior, echoes the title that the great John Maynard Keynes gave to his masterwork, General Theory of Employment, Interest and Money. At bottom, though, Ormerod is arguing only that economists defy common sense by acting as if people's tastes and preferences were fixed. Conventional economics, says Ormerod, misses the obvious fact that "the behavior of an individual can be directly affected by the behavior of others. In other words, people see what others do, and may be influenced by it."

In a linear world, the size of an effect is proportional to its cause. Not so, Ormerod observes, in the nonlinear world we actually inhabit. Fads, stock-market crashes, and avalanches are all examples of huge events set in motion by tiny causes. The craziness of the Internet is all about nonlinearity: Entrepreneurs spill red ink so they can acquire a "first-mover advantage," attain "critical mass," and then enjoy "increasing returns." By ignoring nonlinearity, forecasters constantly get things wrong--missing, for example, the contagion of fear that infected Asia and the world after the fall of the Thai baht in 1997.

There are just two problems with this book. (Actually three, if you count repetition: Most of its big points appeared in The Death of Economics, which was published in Europe in 1994 and the U.S. in 1997.) The first problem is that Ormerod's argument isn't as revolutionary as his book's bold title implies. Many of the people he approvingly quotes number among the most revered members of the economics Establishment, including Keynes, Adam Smith, David Ricardo, and Joseph Schumpeter, as well as such noted modern figures as Paul Samuelson, Milton Friedman, Kenneth Arrow, Paul Krugman, and Joseph Stiglitz.

The second, related problem with Butterfly Economics is that it doesn't--in fact, it can't--offer a fully satisfying alternative to the dunderheaded, linear approach. Recognizing the existence of nonlinearity doesn't make your forecasts come out better. It just helps you understand why forecasting is so darn hard. If fluency in chaos theory were the secret formula for soothsaying, the chaos theorists at the Santa Fe Institute in New Mexico would be trillionaires by now.

That said, Ormerod is on to something. Too many economists still waste their time on tiny tweaks and elaborations of orthodox economic models that are simply wrong. Ormerod wants to unify microeconomics and macroeconomics by showing how the aggregate behavior of an economy grows out of the interactions of individual agents: people and companies. Ormerod shows off his own simple computer model of a hypothetical capitalist economy. It starts out with 225 tiny companies. Over a simulated 130 years, some stay small, and some get very big. The distribution of company size, and the average growth rate for the economy, turn out similar to those of Western economies over the past 130 years.

To be sure, Ormerod hasn't cracked the code of economic forecasting. But at least he and his fellow nonlinear thinkers are looking in the right place. Economists who stick to linear models because they're more tractable are like drunks who look for their car keys under the street lamp because the light is better there.

What's more, an appreciation of "butterfly economics" helps clarify what governments can and cannot do to manage their economies. Mostly what they should do, Ormerod says, is steer clear of fine-tuning, which is likely to have unintended consequences. Less is more, he says. For instance, he counsels governments against trying to manage the value of their currencies by punishing speculation. By encouraging speculation, he says, nations can get foreign-exchange dealers to dart in and out of a currency more often. That way, it will never get far out of line with fundamentals. The source of this insight? Computer models of ant colonies vacillating between two sources of food.

Governments could learn from business executives, who wisely refuse to be imprisoned by theory, Ormerod writes. And economists, he says, could learn from the great political economists of yore like Adam Smith. They saw their mission as understanding society as a whole, not tinkering with mathematical constructs. Ormerod isn't the first to make this point. And to be fair, most economists aren't as clueless as he makes them out to be. But despite this book's flaws, its main message is sound. The tools of economics are too useful to be squandered on shadow plays.

Questions? Comments? E-mail tech& or fax (202) 383-2125

Before it's here, it's on the Bloomberg Terminal. LEARN MORE