An Empirical Investigation of the Lucas Hypothesis: the Yield Curve and Non Linearity in the Money-Output Relationship

Existing evidence about the effectiveness of money growth to stimulate economic activity has been criticized from different perspectives. In addition, high correlation between money and output is not helpful to detect the direction of causality. From a policy perspective, in fact, positive correlation may arise from two opposite policy conducts: either the monetary authority sets the supply of money to influence future output fluctuations, or the central bank controls money growth as a reaction to the recent evolution of macro variables. In this work the relationship between money and output is analysed within a non linear framework that ascribes a primary role to expectations. In particular, we find evidence that the Lucas (1973) hypothesis, that exists an inverse correlation between the variance of nominal shocks and the magnitude of output response to nominal shocks, is supported by data evidence when the yield curve is either flat or downward sloping. We also provide evidence suggesting that the Friedman (1977) hypothesis, that the variability of inflation exerts a negative effect on the natural level of output, holds when a positive risk premium is incorporated in an upward sloping term structure of interest rates.