Stephen McNees and John Ries, both economists at the Federal Reserve Bank of Boston, showed that the most accurate forecaster in any one year had little hope of repeating his success the next. But, based on a study of 44 "blue-chip" forecasters' performances from 1977 through 1983, Bern- stein and Silbert argue that an average of many auguries may be useful.
The group went wrong more than once. In 1978, for example, the "consensus" forecasts for both inflation and change in...