On Some Predictions of the Quantity Theory of Money

Abstract

Recently, efforts have been made to test some of the long-term propositions implied by the quantity theory of money. Of these, two of the more interesting studies are those of Lucas [11] and Lothian [10]. Lucas, using time series data for the United States and applying Fourier transforms detrends the series and then using scatter diagrams of these detrended series finds considerable support for the proposition of one to one correspondence between rate of growth of money and the rate of inflation, and for the Fisher hypothesis. Lothian [10], following upon a suggestion of Lucas [11], uses cross-country averages of time series data for twenty O.E.C.D. countries, to test the classical neutrality hypothesis, the monetary approach to the exchange rate determination and the Fisher hypothesis. His approach consists of treating the cross-country averages to represent longterm steady state points and once again uses scatter diagrams of the appropriate variables. He reports complete support for the first two hypotheses and partial support for the Fisher hypothesis. The aim of this paper is to reexamine two of the propositions implied by the quantity theory of money, namely, the neutrality hypothesis and the Fisher hypothesis for six developed countries, by using a methodology recently proposed by Wickens and Breusch [12] which is an elaboration of an earlier work by Bewley [2]. Their methodology starts with the well known autoregressive distributed lag model which can be estimated to extract long-term parameters of interest. Wickens and Breusch's point of departure is that unlike this traditional procedure, this model can be transformed in such a way that direct estimates of the long-term parameters along with their standard errors can be obtained provided the model is estimated appropriately as explained below. Clearly, this methodology is then ideally suited for application to the issue under consideration. The outline of the paper is as follows. The hypotheses and the models are explained in section II. The estimation issues are discussed in section III. The results are presented in section IV. The paper concludes with a brief summary of the main findings.