Asymmetric preferences and monetary policy deviations

- An asymmetric preferences model is examined under both commitment and discretion.
- An asymmetric version of Walsh's (2003) speed limits to monetary policy is derived.
- Asymmetric commitment policy is shown to nest asymmetric discretionary policy.
- Commitment policy estimates are provided as well as nested restricted model tests.
- Asymmetric commitment policy produces smaller average deviations than discretion.

This paper examines the linkages between monetary policy equilibria and asymmetric preferences. Much has been written on the relationship between discretionary monetary policy and policy under commitment from a timeless perspective. Additionally, there is a relatively robust literature that suggests monetary policy in the U.S. responds asymmetrically to fluctuations in the output gap. Here, asymmetric optimal policy is analyzed under both commitment and discretion in a simple dynamic New-Keynesian model. The timeless perspective equilibrium leads to a policy rule with inertia (consistent with the literature), but in this case implies asymmetry over time. This result is not found under discretion. Both model variants and those imposing linearity are estimated for five developed economies. Policy deviations are simulated for the linear and nonlinear rules. Results imply that the asymmetric commitment policy produces the lowest average deviations from observed policy, and also the policy deviations with the smallest variance, for all five countries. A linear Taylor rule produces statistically larger average deviations than both asymmetric commitment and asymmetric discretionary policy.