Monetary policy and the exchange rate: Evaluation of VAR models

This paper examines the ability of vector autoregressive (VAR) models to properly identify the transmission of monetary policy in a controlled experiment. Simulating data from a small open economy DSGE model estimated for Australia, we find that sign-restricted VAR models do reasonably well at estimating the responses of macroeconomic variables to monetary policy shocks. This is in contrast to models that use recursive zero-type restrictions, for which inflation can rise following an unexpected interest rate increase while the exchange rate can appreciate or depreciate depending on the ordering of the variables. Sign-restricted VAR models seem to be able to overcome puzzles related to the real exchange rate, provided that a sufficient number of different types of shocks are identified. Despite delivering the correct sign of the impulse responses, central tendency measures of sign-restricted VAR models can, however, be misleading and hardly ever coincide with the true impulses. This finding casts doubt on the common notion that the median impulses are the ‘most probable’ description of the true data-generating process.