A D+ For Dismal Scientists?Gene Koretz
By all counts, the forecasts presented to Congress each February and July by the Federal Reserve Board's Federal Open Market Committee (FOMC) should be pretty darn good. After all, as economist Paul L. Kasriel of Northern Trust Co. observes: "The FOMC is theoretically in the unique position of being able to use monetary policy to influence the events it is projecting."
In the Federal Reserve Bank of Boston's New England Economic Review, economist Stephen K. McNees assesses the accuracy of FOMC forecasts from 1980 through 1994. The forecasts of key economic variables are presented in two forms: the range of projections issued by individual FOMC members, and the "central tendency" of such projections, which is the range without its high and low extremes.
What's the score? From 1980 through 1994, the final readings for real gross domestic product ended up within the growth ranges projected in February of those years only a third of the time. The success rates of the July GDP projections were just 19% for the full range of member forecasts and a fat zero for the "central tendency" numbers. That is, in no year did real GDP end up within the central tendency range projected by the FOMC in July of the same year.
To be sure, the study found that the FOMC did no worse than leading private forecasters as a group and actually did better in its inflation forecasts. Still, the results suggest that anyone who thinks the Fed has a firm grasp of the economy's current direction when it sets monetary policy had better think twice.