Forecasts, Interest Rates Headline G-7 Summit

Article excerpt

Well what do you know? Just as the central bankers and finance ministers from the Group of Seven (G-7) industrialized nations were gathering in Washington, and half-ready to cry uncle, what does the U.S. economy do but abruptly look brighter. And all thanks to the big turn in the gross domestic product (GDP). No wonder they call economics the dismal science.

Let's be clear here: No one -- neither at the Treasury nor Commerce departments, at the Federal Reserve or at any major Wall Street investment house -- predicted that the U.S. economy would expand in the first quarter of 2001 at a heartening annual rate of 2 percent. And neither did the International Monetary Fund (IMF) in its bulky World Economic Outlook, which was released on April 26 to coincide with its spring meeting. Earlier in the year the fear was that there would be no growth at all.

Does that mean we are out of the woods and the bright blue yonder beckons? Probably not. There's still a sharp souring of consumer confidence. There is horrendous unused productive capacity to be reckoned with and an investment squeeze.

But the surprising performance of the U.S. economy in the first quarter of the year does suggest that forecasts should be looked at a lot more skeptically. How appropriate then that the IMF chose in this year's World Economic Outlook to place near the front of the 244-page tome an essay entitled, "How Well Do Forecasters Predict Turning Points?"

Not at all well, as it turns out. As the IMF notes, the consensus forecast for real GDP growth in 2001 fell from 3.7 percent in September 2000 to 1.7 percent in April 2001. That represented "the most rapid adjustment in expectations since the early 1990s, when the United States entered its last recession."

Nor was that the first time the dismal scientists got it wrong. The IMF cites a July 1999 study published in the International Journal of Forecasting that reviews a broad sample of predictions for advanced and emerging economies during the nineties. Yes, the consensus mean forecast always was out of sync with what happened. Forecasters initially were slow to adjust their projections and then went on to revise them in a steady but cautious manner, always lagging behind events. Another recent study -- this time out of the University of California-San Diego -- found that forecasters tend to be highly influenced by the consensus prediction of the previous month and seldom move from it very far.

In other words there's a herd instinct -- a feature noted back in the eighties when a researcher suggested that forecasters fear adjusting in jumpy or jagged ways because that may be treated as "signs of inconsistency."

Not that the IMF is risking anything in its report, either. …