Monetary Policy, Exchange Rate and Inflation Inertia in Chile: a Structural Approach

This paper estimates a DSGE model for Chile during the IT period using Bayesian techniques. In this setup, we investigate the way in which monetary policy has been designed. We also assess the performance of simple optimal rules under alternative preferences that the central bank may have, and we investigate whether the Phillips curve can be represented by a hybrid specification which considers inflation persistence. We conclude that the conduct of monetary policy in Chile during the last fifteen years can be characterized by a feedback rule in which the interest rate reacts to contemporaneous inflation misalignment from the target and to output deviations from its natural level. This policy presents an important degree of persistence, that has, however, declined in the past five years. Furthermore, fluctuations in the nominal exchange rate are not offset by the monetary authority: the interest rate response to movements in this variable is not different from zero. A second set of results indicates that inflation persistence, which is usually absent from the standard neo-Keynesian models, is a feature of the Chilean economy. In particular, a model in which the Phillips equation contains a lagged inflation term is preferred to an alternative one which does not consider inflation persistence. This inflation persistence has change in recent years, becoming less important than in the early nineties. On the other hand, since 1999 prices are adjusted less frequently: the probability of resetting prices has fallen. Finally, optimal simple rules for alternative preferences of the central bank show that, besides reacting to inflation and output, there are no welfare gains from reacting to exchange rate movements.