Abstract

The existence of a stable demand for money function and control of the monetary aggregates, are the necessary conditions for the conduct of an effective monetary policy. However, if the monetary aggregates are not defined and measured properly, they cannot be used effectively as intermediate monetary targets. The concepts of the domestic money supply could be defined either theoretically or empirically. Changes in the domestic payments mechanism and increased financial interdependence among economies, through the emergence of the Eurocurrency market, led to a sharp expansion of residents’ foreign currency deposits at domestic banks and of foreign non-bank holdings of domestic deposits denominated in both domestic and foreign currencies, of which some are Eurocurrency components. Although there is no precise definition of money for an open economy, the exclusion of these components from the domestic money concepts could reduce their usefulness for monetary policy, thereby making it more difficult for the authorities to implement effectively their policies. This paper examines the hypothesis whether the current definitions of the domestic monetary aggregate are appropriate for an open economy or whether they should be redefined to include foreign owned deposits and Eurocurrency deposits. This is given both theoretical and empirical considerations. By identifying similar characteristics between these components and domestic money, a case can be made for their inclusion in the domestic stock. To examine the hypothesis empirically, two methods are applied to the case of West Germany, whereby demand for money functions are estimated for alternative definitions of the money stock, based on quarterly data from 1968 IV to 1976 IV, and the most appropriate concept is selected. The first method applies Friedmann-Meiselman’s dual criteria based on money-income correlation. Accordingly, the best definition of money must have the highest correlation with income than any of the components separately. Four definitions of money supply are tested, of which the first two are the common concepts of the money stock currently used in Germany, and the latter two were constructed for testing purposes and they include residents’ foreign currency deposits in domestic banks and non residents’ domestic deposits denominated in both domestic and foreign currency. The second method estimates demand functions for alternative definitions of the money stock, then simulated them for the in-sample and out-of-sample periods, and compares their predictive performance, based on the size of the root-meansquare error (RMSE). The evidence in the case of Germany supports our hypothesis that a redefinition of the money supply is warranted in an open economy. Hence, as the eocnomy becomes more open, the significance of foreign monetary components increases and their inclusion in the domestic monetary aggregates may be warranted. The evidence shows that the Euro-currency components are related to the West German economic activity in the same manner as other components of the broadly defined money supply. Additional findings suggest that the narrow definitions of money seems to be marginally superior to or as good as an alternative definition of it: the redefinition of the broader money supply, by including foreign non-bank holdings of deposits in domestic banks and non-bank domestic deposits, is superior to the current definition. of Me. In summary, it seems that the increase in international financial interdependence through capital flows affects the broader definition of the money supply and provides a significant channel for domestic money and credit substitution.