Monetary Policy Estimation in Real Time: Forward-Looking Taylor Rules without Forward-Looking Data

Abstract

I propose a methodology for estimating forward-looking Taylor rules in real time when forward-looking real-time central bank data is unavailable. The methodology consists of choosing appropriate models to closely replicate U.S. Greenbook forecasts, and then applying these models to Canada, Germany, and the U.K. The results show that the Greenbook output gap series is well described by rolling window linear detrending, while Greenbook inflation forecasts can be closely replicated using Bayesian model averaging over Autoregressive Distributed Lag models in inflation and the GDP growth rate. German and U.S. Taylor rules are characterized by inflation coefficients increasing with the forecast horizon and a positive output gap response. The U.K. and Canada interest rate reaction functions achieve maximum inflation response at middle-term horizons of about 1/2 years and the output gap coefficient is insignificant