Monetary policy reaction functions in Australia

Date

Authors

de Brouwer, Gordon
Gilbert, John

Journal Title

Journal ISSN

Volume Title

Publisher

Blackwell Publishing Ltd

Abstract

Introduction: Monetary policy is the central tool for maintaining price stability and it has become the key tool in managing the business cycle. Yet, while there is a large empirical literature which assesses the behaviour of monetary policy in major economies, there has been little systematic empirical analysis of how monetary policy has been set in Australia. This paper examines the reaction function(s) of the Reserve Bank of Australia in setting interest rates in the post-float period, with special emphasis on the inflation-targeting period since 1991. We examine calibrated and estimated backward-looking and forward-looking reaction functions to assess the stability and consistency of interest-rate setting in the post float period. We aim to answer the following questions: In setting interest rates, has the Reserve Bank acted as if it was following a ‘rule’? Has it done so consistently? Does it respond systematically to variables other than inflation and output? How has the introduction of inflation targeting affected monetary policy? What is the neutral rate of interest? We also use these reaction functions to enter into two ongoing debates about the measurement of the output gap. The first of these centres on the problem of revisions to national accounts data. Orphanides (1998, 2000) argues that reaction functions should be specified using the real-time data available to policymakers at the time they made their decisions. Others, like Clarida et al. (1998), Taylor (1999a,b; 2000) and Rudebusch (1999), use current or final release data. The appeal of current final data is that they are easy to use and that they are more likely to capture the ‘yes’ economic pulse of the time since initial estimates have considerable noise. The second centres on the statistical techniques used to estimate output gaps: are theoretically based approaches, like the Phillips-curve-based estimates of the output gap used by Orphanides and van Norden (2001) and Gruen, Robinson and Stone (2002), superior to those derived from simple time series trends, like the widely used Hodrick-Prescott filter? Our presumption is that the ‘best’ gap measure is the one which most accurately explains what the central bank does. The paper is structured as follows. We first set out some issues about reaction functions and outline the data. We then estimate backward-looking and forward-looking monetary policy reaction functions. The conclusion presents our assessment. It discusses two issues about the analytical tools used to assess monetary policy and four issues about what the study of simple interest rate reaction functions reveals about monetary policy in Australia.

Description

Keywords

monetary policy, Australia, reaction functions, interest rate, inflation

Citation

Source

The Economic Record

Type

Journal article

Book Title

Entity type

Access Statement

Open Access

License Rights

DOI

10.1111/j.1475-4932.2005.00238.x

Restricted until