Macroeconomic forecasting and macroeconomic forecasts

Abstract

The specificity of macroeconomic forecasts is determined not so much by the list of predicted indicators or by mathematical tools used, but by the unavoidable human factor, which often generates great difference between forecasts made by various professionals. In Russian-language literature, this psychological aspect of macroeconomic forecasting has not received any attention; our work is designed to fill this gap. As a source of statistical data we used the forecasts extracted from the quarterly Poll of Professional Forecasters (PPF), which began in the first quarter of 2000. An analysis of real GDP and CPI forecasts made it possible to identify optimists and pessimists among independent experts, and also to show that the official forecasts by the Russian Ministry of Economic Development and the Bank of Russia are often too optimistic (with the possible exception of forecasts for a current year). It was also confirmed that the accuracy of consensus-forecasts exceeds (in the long run) not only that of official forecasts, but also forecasts made by the vast majority of independent experts. This asserts consensus-forecasts as a benchmark against which macroeconomic forecasts of other experts and organizations should be compared. An analysis of errors for different forecast horizons showed that in the medium term, the most significant ones are associated with underestimation of the resilience of the Russian economy to external shocks. This aspect of macroeconomic forecasting is closely related to forecasting cyclical dynamics (in particular, recessions, their beginning, depth and duration). Currently, this is just what requires additional attention by macroeconomic forecasters.