Estimating the Output Gap of the Russian Economy: A Multivariate Approach Based on BVAR and the Beveridge–Nelson Filter

Abstract

This study estimates Russia’s output gap using a multivariate Beveridge–Nelson filter and Bayeian vector autoregression according to the methodology of Morley et al. (2023), taking into account the pandemic distortions (Lenza and Primiceri, 2022). The model includes 14 macroeconomic indicators grouped into blocks reflecting the dynamics of internal and external demand. Estimates made on data from 2005Q1 to 2025Q1 show consistency with the results of univariate filters in identifying recession phases (2008–2009, 2020, 2022), but the multivariate approach allows for a deeper understanding of the dynamics of the business cycle through historical decomposition and the analysis of the contributions of individual factors. It is found that the external demand shocks, namely fluctuations in oil prices and changes in global demand, played a key role in previous crisis episodes. In 2022, on the contrary, the decisive factor was the decline in internal demand, while external conditions had a moderately supportive effect, reflecting a structural shift in the mechanisms of recession formation.