This paper is focused on the evolution of in‡ation and output dynamics over the last 50 years, the changes in the behavior of the Federal Reserve, and the role of agents’ beliefs. I consider a new Keynesian dynamic stochastic general equilibrium model with Markov-switching structural parameters and heteroskedastic shocks. Agents are aware of the possibility of regime changes and they form expectations accordingly. The results support the view that there were regime switches in the conduct of monetary policy. However, the idea that US monetary policy can be described in terms of pre- and post-Volcker proves to be misleading. The behavior of the Federal Reserve has instead repeatedly ‡uctuated between a Hawk- and a Dove-regime. Counterfactual simulations show that if agents had anticipated the appointment of an extremely conservative Chairman, in‡ation would not have reached the peaks of the late ‘70s and the inflation-output trade-off would have been less severe. This result suggests that in the ‘70s the Federal Reserve was facing a serious problem of credibility and that there are potentially important gains from committing to a regime of in‡ation targeting. Finally, I show that in the last year the Fed has systematically deviated from standard monetary practice. As a technical contribution, the paper provides a Bayesian algorithm to estimate a Markov-switching DSGE model.

This paper considers the problem of forecasting under continuous and discrete structural breaks and proposes weighting observations to obtain optimal forecasts in the MSFE sense. We derive optimal weights for continuous and discrete break processes. Under continuous breaks, our approach recovers exponential smoothing weights. Under discrete breaks, we provide analytical expressions for the weights in models with a single regressor and asympotically for larger models. It is shown that in these cases the value of the optimal weight is the same across observations within a given regime and differs only across regimes. In practice, where information on structural breaks is uncertain a forecasting procedure based on robust weights is proposed. Monte Carlo experiments and an empirical application to the predictive power of the yield curve analyze the performance of our approach relative to other forecasting methods.