RDP 1977-05: Modelling Monetary Disequilibrium 5. Conclusions

This paper tests the role of the excess demand for money in the transmission process. The related question of how to estimate the demand function for money is examined, with particular reference to whether the common assumption that the money stock is demand determined in the short run is appropriate for recent Australian experience.

Several conclusions emerge, subject to the limitations of the model building framework used in the current study, and of empirical macroeconometrics more generally. Applying the simplest test, based on the size and significance of the parameters, it is clear that the channels for monetary disequilibrium represented by γ1, γ3, γ4, γ6 and γ7 in model 1 are not rejected. In simulation tests, the version of the RBA76 model with several influences for monetary disequilibrium appears to fit the data about as well as one in which there is only the expenditure channel for excess money balances, and considerably better than a conventional version of the model with no direct channels for monetary disequilibrium. The dynamic simulation properties of each of these models in response to a sustained increase in real government spending are different, and the model with several channels for monetary disequilibrium appears to have more stable cyclical responses than the other two.

With respect to the demand function for money, versions of the model in which this function is estimated directly in the equation determining the quantity of money, as in several recent Australian studies, perform quite badly in several respects.

If it is accepted that of the models considered in this paper the basic model (model 1) provides the best representation of the Australian economy, several general conclusions can be drawn. The basic model suggests that there are important direct effects of excess real money balances on household expenditure, and thus on output; on prices, wages and perhaps on factor demands; and on the capital account of the balance of payments. The effects of excess money on domestic variables have indirect effects on all components of the balance of payments, effects which reinforce the direct effects on capital flows. The basic model can thus be said to provide an explicit structural view of the adjustment processes emphasised in the literature on the monetary approach to the balance of payments. The relative importance of the effects of excess money on domestic variables, and the evidence that adjustment is not instantaneous, indicates that very, simple models of the interaction of money and other variables are not of direct relevance to the Australian economy. This emphasises the need for macroeconomic models to integrate the effects which are emphasised in both closed and open-economy monetary theory.