The endogeneity of monetary policy in Australia, 1961-1974

The aim of this thesis is to analyse the setting of the individual monetary policy instruments by the Australian monetary authorities over the period 1961-1974. The motivation for the analysis of such an issue stems from the increasing interest in monetary economics and the growing sophistication of the Australian capital market. The rapid increase in the money stock from 1973 onward has emphasized the need for more direct controls on the growth of the money supply. Successful control depends on a working knowledge of the behaviour of the instruments required. the analysis of the setting o:f individual policy instruments is pursued in order to determine whether the monetary authorities set the instruments endogenously and in response to movements in policy targets, or whether they set the instruments independently. If the instrument setting is found to be endogenous, several related issues arise: the possible assignment of the instruments, the interdependence between the instruments and the temporal stability of the policy responses. Before attempting to accomplish this aim, two chapters are devoted to a review of monetary-fiscal literature and previous reaction function estimation. These chapters serve as a background to the problem at hand. The monetary-fiscal debate brings to light two related issues - reverse causation and the identification problem. The reverse causation argument concerns the issue of whether the money stock is the cause or effect of economic activity. If it is the cause, then the money supply can be treated as exogenous. Alternatively, if it is the effect, then the money supply should be treated endogenously. The identification problem relates to the issue of how the variables are defined and identified. The success of the empirical tests in the monetary-fiscal debate depends upon this issue. The review of the previous reaction function studies provides an insight into the identification problem. Each of these studies has attempted to define the variables in the most suitable way. Although none of these earlier formulations are exactly suited to the present problem, they serve to provide valuable background. In Chapter 4, a simple model of money supply determination is used to identify the appropriate monetary instruments for the Australian economy. The reaction functions are then formulated and the problems of data and estimation are discussed. The regression results are discussed in Chapter 5. Results are also presented for the time period split into the contractionary and expansionary phases of economic policy. The differing response of the instruments to each policy target is observed. This test is an important aspect of the more general problem of temporal stability of the reaction functions. The problem ts examined by applying the TIMVAR technique and identifying the various periods of instability. Much of the observed instability is obviously due to the changing weights on the targets during different phases of monetary policy. The SRD function is analysed in terms of the movement in Australia's economic cycle. It is observed that there is a close connection between the periods of instability in the function and movements in the cycle. A similar analysis is carried out for the securities function, this time in terms of the movement in the level of foreign reserves. In this case, however, the connection is no as obvious or specific. The results of this analysis are presented in Chapter 6. In Chapter· 7, the policy implications of the analysis are discussed. It is felt that more work in this area is warranted, especially in regard to the determination of response lags