The Causal Relationship Between Money Base and Money Suppy in Australia: 1960–1979

Australian monetary economists have for some time been intermittently concerned with the relationship between the monetary base and the money stock. Controversy has centred around a number of issues, including: the exogeneity of the money base; the substitutability of cash and other liquid assets in the management of banks’ reserves; the causal significance of the budget deficit of the consolidated government sector relative to imported disturbances; and the appropriate definition of the money base in Australia. This paper provides some additional evidence relevant to this debate. Changes in the financial system during the last two decades, including partial deregulation of the banking system, and the growth of finance companies and permanent building societies, suggest the need for a careful examination of the money/money base relationship. These changes are likely to have altered both the time series properties of the monetary aggregates and may also have changed the observed relationship between various measures of money base and the money stock. The evidence presented in this paper suggests that this may have occurred. The analysis is directed towards three questions. Does the money multiplier vary significantly in terms of its stability with differing measures of money and money base? What is the behavioural significance of differences in stability? How does the causal relationship between money and money base vary over time and with different concepts of base and money? Graphical, numerical and time series analysis are used to shed light on these issues. The results are in some cases conflicting due to the strigent techniques for determining causality but we conclude: • there is evidence of considerable instability in the money multiplier for most concepts of money base during the 60s and 70s. Financial innovation may have been important in this instability. The timing of significant changes in the financial system is consistent with observed instabilities of the multipliers; • the evidence suggests that the M2 money multiplier for the extended base is the most stable for both the sixties and seventies; • there is evidence of contemporaneous correlation between money base measures and the money stock (for M2 and M3) but that the results for the seventies indicate the presence of both a positive causal effect on the money supply from innovations in the money base and feedback effects from the money stock to money base; • there is evidence of some structural change in the time series models for some aggregates between the late sixties and the late seventies; in addition, there is evidence of a structural change within the sixties relating primarily to the private sector’s demand for cash relative to other monetary aggregates and to the growth of the permanent building societies.