Monetary Policies in Interdependent Economies with Stochastic Disturbances: A Strategic Approach

Abstract

Western economies have become increasingly interdependent during recent years. An important consequence of this is that the effects of macroeconomic policies within these economies have become more closely related. A policy implemented in one country will generate effects abroad, while the impacts of this policy on the domestic economy are modified by the behaviour and policies of the foreign economies with which it is interacting. Thus policy making in a multicountry context necessarily involves strategic behaviour. Analysis of strategic behaviour within the context of international macroeconomic policy began with the pioneering work of Hamada (1976), who investigated this question under the assumptions of Cournot and Stackelberg behaviour. His analysis is based on a fixed exchange rate regime and the objective function involves the tradeoff between inflation and the balance of payments. More recently, Canzoneri and Gray (1 985) consider alternative strategic monetary policies within the context of two economies subject to a mutual oil disturbance.1 This paper continues the analysis of strategic monetary policy. The framework it employs is a two country stochastic macro model in which both economies are subjected to stochastic demand and supply shocks and expectations are rational.2 The policy makers in the two countries seek to optimise their respective objective functions, which are taken to be functions of unanticipated movements in output and in the consumer price index. The model begins by determining the usual Cournot and Stackelberg equilibria for this model. However, these represent just two possible equilibria, and a number of alternatives are also considered. In particular, in the derivation of the Cournot equilibrium, each agent takes the behaviour of his opponent as given, and therefore assumes that his rival does not react to his actions. On the other hand, each agent is shown to respond in accordance with a reaction function, so that ex post, the assumption of no response is incorrect. By contrast, we also consider a Consistent Conjectural Variations equilibrium (CCV) in which each policy maker, in determining his own actions, correctly conjectures the response of his opponent. The requirement of con