RDP 9410: An Empirical Examination of the Fisher Effect in Australia 1. Introduction

Does a rise in short-term interest rates reflect a rise in real interest rates and, hence, a tightening of monetary policy or, alternatively, a rise in expected inflation? A standard view, commonly referred to as the Fisher effect, is that movements in short-term interest rates primarily reflect fluctuations in expected inflation and, as such, they have predictive ability for future inflation. If this is the case a rise in short-term interest rates does not indicate a tightening of the stance of monetary policy but rather a rise in expected inflation. This would suggest a need for caution in using the level of nominal interest rates as indicators of the tightness of monetary policy.

Not surprisingly, because of its importance to policy, the relationship between the level of interest rates and future inflation has been studied in many countries. The Fisher effect has been found to be strong in some countries over certain periods, for example in the United States, Canada and the United Kingdom in the post-war period until 1979. However, the correlation between interest rates and expected inflation has not been high post-1979 or in other countries.^[1]

These results raise the puzzle of why the Fisher effect is strong for certain countries and certain time periods but not for others. In a recent paper, Mishkin (1992) finds an answer to this puzzle for the United States. The paper finds support for a long-run Fisher relationship in which inflation and interest rates are cointegrated. The existence of a long-run Fisher effect implies that when inflation and interest rates exhibit trends, as occurred with the rise in rates in the 1970s, these two series will trend together and, thus, there will be a strong correlation between the two. As a result the Fisher effect appears to be strong in the periods when interest rates and inflation exhibit trends. On the other hand, the evidence does not support the existence of a short-run Fisher relationship in which a change in expected inflation is associated with a change in interest rates. So, when interest rates and expected inflation do not exhibit trends, as occurred though the 1980s when inflation and interest rates were relatively stable, a strong correlation between interest rates and inflation will not appear.

This paper uses the methodology outlined in Mishkin (1992) to analyse the strength of the Fisher effect in Australia. In particular we make use of Monte Carlo simulations to determine the true distributions of test statistics. We find no evidence for a short-run Fisher relationship although there is some evidence for a long-run Fisher relationship. This evidence suggests that changes in short-term interest rates in Australia indicate a tightening or loosening of monetary policy, whereas the longer-run level of short-term interest rates, instead of providing information on the tightness of monetary policy, reflect expectations of inflation.

Footnote

See, for example, Fama (1975), Nelson and Schwert (1977), Mishkin (1981, 1984, 1988), Fama and Gibbons (1982). [1]