Inflation expectations: Australian consumer survey data versus the bond market

Abstract

This paper analyzes the relationship between interest rates, inflation expectations and inflation rates in Australia using the approach suggested by Toda and Yamamoto (1995) to test for Granger causality. The empirical evidence reported here suggests that uni-directional Granger causality is running from medium- and long-term government bond yields to short-term inflation expectations as measured by a survey among consumers. Moreover, there is bidirectional Granger causality among short-term interest rates and short-term inflation expectations among consumers. Additionally, interest rates and the sentiment data measuring inflation expectations can help to predict inflation rates. These findings are supportive for the Fisher effect and also seem to indicate that the bond market is quite efficient predicting inflation rates. Furthermore, some problems of more traditional tests of the Fisher hypothesis are discussed. In this context the role of financial deregulation in Australia is examined. Thus, the Lucas critique seems to be of some relevance testing the Fisher effect (see Lucas, 1976).