Is the Response of Output to Monetary Policy Asymmetric? Evidence from a Regime-Switching Coefficients Model
FRB of St. Louis Working Paper No. 2001-022A
33 Pages Posted: 13 Mar 2002
Date Written: April 2002
This paper adopts a general approach to investigate asymmetry in the leading explanatory power of interest-rate-based indicators of monetary policy for U.S. output. The purpose is to provide robust results by utilizing a model that does not pre-specify a particular condition under which asymmetry arises. In particular, we augment an unobserved-components decomposition of real output into permanent and transitory components with monetary policy indicators. Asymmetry is captured by allowing the coefficients on the policy variables to undergo regime switching. This framework produces a latent state variable that documents when policy actions have led output. In a second stage, we use the estimated latent state variable to jointly evaluate the evidence for three specific types of asymmetry. We find that policy actions taken during periods of cyclical downturn in the economy are more likely to lead output than those taken during cyclical expansions. We also find that during periods of cyclical downturn, large policy actions and contractionary policy actions are more likely to lead output than those that are small or expansionary.
Keywords: Asymmetry, Business Cycles, Regime Switching, Monetary Policy
JEL Classification: C32, E32
Suggested Citation: Suggested Citation