Monetary policy evaluation and inverse control with endogenous policy regimes

This study applies a unified approach of econometric policy evaluation to investigate "endogenous" evolution of U.S. monetary policy during the post-war period. A policy regime is defined as a set of preference parameters which the Federal Reserve chooses for minimizing the value of loss function. The evolution of monetary policy is assumed to depend on a stochastic switching process that evolves according to a Markov chain. Preference parameters in the Federal Reserve's policy objective function together with parameters in the structure of economy are estimated simultaneously. This study uses forward-looking New Keynesian models as the structure of economy in which equilibrium values of output and inflation depend on their own future expected values.;The results suggest that three different policy regimes can be better used to describe U.S. monetary policy than two policy regimes and that in all policy regimes the Federal Reserve consistently placed far greater weight on inflation stabilization than on output and interest rate stabilizations. Estimating a baseline model with data from 1965 through 2001 shows that policy regime one is a special regime that only prevailed between 1979 and 1984, which is commonly known as the Volcker disinflation period. Policy regime two is a regime which the Federal Reserve's monetary policy switched into during expansionary periods, and regime three is a regime which the monetary policy switched into during recessionary periods. The estimation with optimal policy restriction can also alter and sharpen the estimates of model parameters. Estimating the baseline model with extended data from 1965 through 2008 shows similar results to those obtained with the shorter sample. The results from the longer sample suggest that the Federal Reserve did follow an optimal policy during the post-war period. The Federal Reserve's monetary actions were very close to optimal during policy regimes two and three. A counterfactual analysis shows that the small value of preference parameter placed on output stabilization plays an important role in conducting monetary policy. Finally, estimating an augmented model with extended data from 1965 through 2008 suggest that the findings associated with the three-regime monetary policy are robust.