Abstract

The Existence of a World Demand-for-Money Function under Fixed and Flexible Exchange Rates: Preliminary Results

At a national level a demand-for-money function can be used as a guideline for the effectiveness of monetary policy. Once a quantitative relationship between money stock, income and the interest rate is established, then we can assess the contractionary effect of interest rate increases on money cash balances as well as the necessary expansion of the money stock to given GNP increases. At an international level the pertinence of such a function is relevant only under fixed exchange rates because the currencies are freely convertible and one can virtually talk about a common currency. Although the fixed exchange rate period now belongs to the past, our study is still not of historical importance alone, since it is not unlikely that this practice might reappear in the future. Besides, so long as there are groups of countries, such as E.E.C., which aim at a monetary union at a future date the findings of this paper could be of some importance. Contrary to past studies on the same subject, aggregation remains at the national level although the application of our model relates to a group of ten countries. The specification adopted allows for specific shift variables which represent, in the main, heterogeneities that exist between countries. Also, the sample size is increased to several hundreds of observations which enable us to claim asymptotic properties for our parameter estimates and easy sub-division of the sample in order to test the stability of the function. Our main findings are: firstly, interest rate plays an important role in the world demand-for-money function, especially when expressed by the eurodollar rate. Secondly, both a static and a dynamic formulation are significant, with the latter performing better. Thirdly, the stability of the world demand-for-money function over time and under fixed exchange rates is not to be taken for granted, while under flexible exchange rates it almost breaks down. This implies that world inflation cannot easily be regulated by controlling the world money supply. These results are somewhat in contrast to those obtained by similar studies on the world demand-£ormoney using merely time-series data. The significance of the interest-rate elasticity is obscured in these studies, especially in the simple static model, while the stability of the function cannot readily be tested due to lack of degrees of freedom.