Inside the Crystal Ball: How to Make and Use Forecasts

Chapter 1What Makes a Successful Forecaster?

It's tough to make predictions, especially about the future.

—Yogi Berra

It was an embarrassing day for the forecasting profession: Wall Street's “crystal balls” were on display, and almost all of them were busted. A front-page article in the Wall Street Journal on January 22, 1993, told the story. It reported that during the previous decade, only 5 of 34 frequent forecasters had been right more than half of the time in predicting the direction of long-term bond yields over the next six months.¹ I was among those five seers who were the exception to the article's smug conclusion that a simple flip of the coin would have outperformed the interest-rate forecasts of Wall Street's best-known economists. Portfolio manager Robert Beckwitt of Fidelity Investments, who compiled and evaluated the data for the Wall Street Journal, had this to say about rate forecasters: “I wouldn't want to have that job—and I'm glad I don't have it.”

Were the industry's top economists poor practitioners of the art and science of economic forecasting? Or were their disappointing performances simply indicative of how hard it is for anyone to forecast interest rates? I would argue the latter. Indeed, in a nationally televised 2012 ad campaign for Ally Bank, the Nobel Prize winning economist Thomas Sargent was asked if he could tell what certificate of deposit (CD) rates would be two years hence. His simple response was “no.”²

Economists' forecasting lapses are ...