Estimating U.S. Monetary Policy Rule Using a Markov Switching DSGE Model

Estimating U.S. Monetary Policy Rule Using a Markov Switching DSGE Model

This paper sets up a Markov switching DSGE model, where the parameters in the monetary policy rule are shifting according to unobservable regimes, and utilizes Bayesian estimation methods to find out whether there has been regime shifts in monetary policy in U.S. since 1960. The estimation results are mostly consistent with those of subsample analysis in the previous literature: the monetary policy has become more responsive to inflation since the early 1980s than before. Two novel findings in this paper are that the nonborrowed reserve targeting regime, which spans from 1979 to 1982, is estimated to be not so aggressive to inflation and that the 1960s is characterized as a stabilizing monetary policy regime.