Policy measurement for the dynamic linear model with expectations variables: A multiplier approach

Abstract

To estimate the economic policy effects of per unit policy change, the conventional policy multipliers, as a measure of policy effects, can be easily calculated from the traditional dynamic econometric model without expectations variables. However, the past decade has witnessed much research and debate on the rational expectations hypothesis. In a model with expectations variables, the complexity of measuring policy effects arises not only from its dynamic properties, but also from its treatment of expectations variables. In this paper, we present a method of deriving the policy multipliers for the dynamic linear model with expectations variables and a backward recursive substitution algorithm to calculate these multipliers. The development of our methodology is basically along the traditional theory of the policy multiplier, with a substantial modification to distinguish unanticipated from anticipated policy effects.