Forecast errors before and during the Great Moderation

This paper investigates the change in private sector and Federal Reserve forecasts before and during the Great Moderation. We view the Great Moderation as a natural experiment. Using forecasts produced by the Survey of Professional Forecasters and the Federal Reserve (Greenbook forecasts) we investigate three questions: (1) How large was the decline in forecast errors? (2) Did forecast accuracy improve relative to the decline in volatility of growth and inflation? (3) Did forecasters respond to the Great Moderation? We find that the absolute median error as well as the cross-sectional volatility of forecast errors decreased significantly. Forecasters appeared to have narrowed the dispersion of their forecasts in response to the Great Moderation. Forecast accuracy did not improve relative to the reduction in the volatility of the economy.