IN SEARCH OF THE LIQUIDITY EFFECT

A short-run negative relationship between monetary aggregates and interest rates - the `liquidity effect' - is central to discussions of monetary policy. This paper searches for this empirical relationship. We investigate whether the characterization of the liquidity effect is sensitive to: (i) changes in sample period, (ii) conditioning the correlations on past information, (iii) assuming money growth is exogenous, and (iv) treating monetary changes as anticipated or unanticipated. The correlations change substantially in each case. We conclude that the traditional analysis and modern models, which rely completely on demand-side behavior, cannot explain the observed correlations. A successful characterization of the liquidity effect requires identification of both private and policy behavior.