Selecting an intermediate target variable for monetary policy when the goal is price stability

Conventional investigations of the "best" intermediate target variable for monetary policy have used a single criterion: the best fit between the behavior of an aggregate and that of some goal variable such as nominal spending or the aggregate price level. Ignored in this type of study, however, is the ability of the central bank to control the behavior of the aggregate which has the best fit relative to the goal variable. This paper treats the issue of monetary control explicitly and selects an intermediate target variable on the basis of a joint criterion of monetary control and relationship with the aggregate price level. The results indicate that all of the traditional simple sum aggregates perform poorly relative to Divisia aggregates or the currency-equivalent (CE) measure recently proposed by Rotemberg, et al.