The term structure of interest rates in Australia: an application of long run structural modelling

Abstract

The term structure of interest rates in Australia, using data of different types as well as frequencies covering the period 1991(11) to 2000(9) is investigated using a relatively new modelling strategy, previously untested on Australian interest rate data. Developed by Pesaran & Shin (1997), this strategy incorporates long- run structural relationships in an otherwise unrestricted vector autoregression model (VAR). The long-run relationships included in the model are based on the Expectations Theory of the term structure. This model is then subjected to Granger-causality, generalised variance decompositions, generalised impulse response and persistence profile analysis, in the context of a vector error correction framework, in an attempt to gather knowledge regarding the nature and mechanism of the Term Structure in Australia. The econometric tests indicate that in Australia, contrary to popular belief, long- term interest rates more often than not lead shorter- term interest rates, at least for the interest rates and time period under investigation. While these findings are not conclusive, if they are an accurate representation of interest rate behaviour, this does pose a major challenge for the monetary policy in Australia. Australian monetary policy relies on the assertion that the shorter term rate leads the longer term rates, and that changes in the cash rate will reverberate through the yield curve to the longer term rates, which in turn affect aggregate demand and other economic indicators. The findings of this study, based on recent rigorous time - series techniques, tend to cast doubts on the efficiency and effectiveness of current policy in Australia.