Monetary Policy at the Zero Lower Bound: An Endogenous Switching Approach to Forward Guidance

The literature on the zero lower bound has grown rapidly since Eggertsson and Woodford (2003). But estimation of the effects of practical monetary policy at the zero lower bound (ZLB) remains a daunting task. In this paper, we propose a new approach to estimating a log-linearized DSGE model with monetary policy at the zero lower bound (ZLB). Our approach 1) allows rational-expectations solutions to take account of a probable recurrence of the ZLB regime and 2) provides a computationally feasible tool that can be utilized to estimate a large-scale monetary DSGE model with the ZLB. Our paper is a positive analysis of practical monetary policy, where forward guidance is an essential gradient and the zero lower bound is explicitly taken into account. We hope that our approach will lay ground work for more applied research on the effects of monetary and fiscal policy in the zero lower bound environment.