Abstract

The aim of this study is to test at world level for the period 1957 to 1971 the contention that, in accordance with the monetarist inflation thesis (i. e. the money growth-inflation relationship), world inflation can be controlled by controlling the world money supply. In this connection the present paper investigates (a) the hypothesis that under a fixed exchange rate regime, national inflation rates converge, moving in a systematic and predictable fashion consistent with purchasing power parity, and (b) the demand-for-money function at the level of the aggregate of fourteen advanced countries (as a proxy for the world). The work presented here is an extension and integration of previous research by the authors published in earlier issues of Kredit und Kapital (1977 and 1979). The results indicate that the world demand-for-money function significantly depends on both income and interest rates but that its form changed between the mid-1960s and 1971. Thus the elasticities determining this function are unstable and pedictions based on them would therefore be unreliable. One might argue, of course, that there are a number of additional determinants that ought to be included in any world demand-for-money function relating to the period close to the 1970s. Hence is these influences are not included in the function, we are bound to get unstable elasticities. The resulting policy implication of this analysis for the design of an anti-inflation policy is that it is highly doubtful whether world inflation can in fact be controlled in a predictable fashion by simply controlling the world money supply. (This is not to say that a reduction in the rate of increase of the world money supply would have no beneficial effect on world inflation.) The paper further challenges the interpretation of the function in the late 1960s und early 1970s. The fact that a simple autoregressive scheme can explain so much implies that it is not economic criteria which govern the function. If a lagged endogenous variable ends up by absorbing the greatest portion of the dependent variable’s variation, the whole function boils down to being determined by an autonomous growth. Through this process, the world demand-for-money function evaporates.