DO INFLATION EXPECTATIONS IMPROVE MODEL-BASED INFLATION FORECASTS IN RUSSIAN ECONOMY?

The paper shows that the use of inflation expectations from surveys of Russian companies, households and professional forecasters does not lead to a significant improvement in the accuracy of the inflation forecast in factor models based on the Phillips curve over the past 5 years. For some forecasting horizons expectations of certain economic agents can lead to an increase in the accuracy of the forecast, but not by more than 10%. There is an improvement in the inflation forecast in the medium-term horizon from 1 to 2 years, with the addition of inflation expectations of analysts or companies. Iterative forecasting of inflation within the Phillips curve turns out to be more accurate than constructing a direct forecast or forecast based on a vector autoregression model, as well as a random walk model. Based on the estimated VAR model, the impulse response function of inflation to the shocks of inflation expectations indicates that short-term expectations of companies and households have the greatest impact on the dynamics of inflation in Russia on the horizon for at least 2 quarters. The Bank of Russia should conduct its communication policy to stabilize the expectations of companies and households. Also we find that the inflation forecasts of the Bank of Russia affect the inflation expectations of professional forecasters on the horizon up to 2 years. The use of long-term expectations of the consensus forecast of analysts as trend inflation leads to an increase in the accuracy of the inflation forecast in factor models.