The demand for money and its stability in Australia has received a great deal of attention in the past and has resulted in its own literature. Depending upon estimation method and period of analysis, previous research has provided mixed findings. By including a measure of economic uncertainty and a measure of monetary uncertainty (both GARCH-based) in the long-run money demand for M3, and by using the bounds testing approach under which variables could be stationary or non-stationary, we provide strong evidence that the M3 money demand in Australia is stable. Both uncertainty measures do have short-run as well as long-run effects on the demand for M3 in Australia, factors that previous research did not consider.
I . I n t r o d u c t i o nDue to advances in econometrics literature, it pays to revisit some of the old and traditional literature and the demand for money is no exception. This part of the literature in economics has received, perhaps, the most attention. Indeed, so much attention that most countries do have their own literature. A common theme in almost every study is to test for stability of the money demand. If money demand is found to be unstable, researchers point out omitted variables from their specifications and show that by including those omitted variables, stability could be achieved. For example, in testing for stability of the money demand in the US, McNown and Wallace (1992) concluded that in order for the demand for M1 monetary aggregate to be stable, one must include the nominal effective exchange rate in the specification that accounts for currency substitution between US and rest of the world, though the original idea goes back to Nobel laureate Mundell (1963, p. 454) who conjectured that 'The demand for money is likely to depend upon the exchange rate in addition to the interest rate and the level of income'.Since this paper deals with the demand for money and its stability in Australia, a brief review of recent studies in the literature will help us to highlight the contribution of this paper. Karfakis and Parikh (1993) considered M3 monetary aggregate and showed that using Johansen's comintegration method, since M3 is cointegrated with income, interest rate and the exchange rate, it is the appropriate aggregate for monetary policy. Although 1984IV was identified as a breaking point in the relation, application of the CUSUM and CUSUMSQ tests revealed stable M3 money demand function. Hoque and Al-Mutairi (1996) considered the demand for narrow money M1, and tried to determine whether deregulation of financial markets in Australia that began in 1976 introduced any instability into their specification of the money demand. Application of Engle