Spectral Estimation of the Business Cycle Component if the Russian GDP under High Dependence on the Terms of Trade

The article proposes a new approach for decomposition of the real GDP of the Russian economy into trend and business cycle components based on the cointegrating regression model. The novelty of the approach is that when building the components of the long-term trend, firstly, we take into account the long-run dependence of the Russian real GDP on oil prices. Under this assumption the fall of oil prices leads to a decline in the permanent level of the real GDP. Secondly, we allow the presence of breaks in the longrun trend of non-oil component of the output. This allows us to take into account the changes in the long-term phases of Russia's economic development and distinguish periods of transformational decline and recovery growth. The business cycle component is estimated in two steps. At the first step, the non-stationary component consisting of a deterministic trend with structural breaks, and components characterizing the long-run impact of oil prices on the Russian economy are eliminated from the series of GDP (in logs). At the second step, the component of the business cycle with a periodicity of fluctuations from 6 to 32 quarters is extracted from the stationary residuals using spectral analysis methods. We find that the business cycle component for the period from 2014 to 2016 was zero while other methods give negative estimates. The new monetary policy regime of the floating exchange rate allowed the Russian real GDP to be adjusted quickly to its new lower potential level because of a drop in oil prices, while cyclical fluctuations were rather moderate.