Модель зависимости обменного курса рубля от цен на нефть с марковскими переключениями режимов

This paper examines the relationship between the Russian ruble/US dollar exchange rate and global oil prices using autoregressive model with Markovian regime shifts. Empirical analysis on daily data for 2009–2019 shows that exchange rate dynamics is best described by three regimes, characterized as follows: 1) weak exchange rate reaction to oil price shocks – low conditional volatility of exchange rate changes; 2) strong reaction – moderate volatility; 3) strong reaction – high volatility. Regime 3 covers crisis periods, when ruble depreciated substantially. Regime 1 prevailed during the period of managed exchange rate arrangement lasted until November 2014. After adoption of a floating exchange rate and inflation targeting policy, regime 1 became regularly identified since mid-2017. This result can be attributed to the introduction in 2017 of a new budget rule, aimed to reduce dependence of exchange rate on oil price fluctuations. Switches between regimes could also be due to fluctuations in the uncertainty measured by the indices of geopolitical risk and economic policy uncertainty for Russia. It is also shown that the model with three regimes outperforms the random walk and linear models of the ruble exchange rate in an out-of-sample fit exercise. The proposed model can be used for identifying the current exchange rate regime in real time, scenario analysis of the consequences for the ruble exchange rate under alternative oil price trajectories, as well as in developing strategies for hedging currency risks by the private sector.