We examine the potential policy role of monetary aggregates by attempting to use them as effectively as possible in the analysis of empirical relationships. We focus on their role as information variables, since any more ambitious use, such as in a policy rule, would presuppose some information content in the aggregates. Our results show that in the United States since 1979, the monetary aggregates fall considerably short of those requirements, and results with German M3 are hardly more favorable. We also investigate whether empirical relationships are not reflective of information relationships because of the use of these variables in countercyclical policy. The results are not consistent with that interpretation.
