Title: Dynamics of the Demand for Money and Uncertainty: The US Demand for Money Revisited
Abstract: It is now commonly agreed that the demand for money function suffers from several problems. In particular, this seems to be the case if it is fitted to US data. At the empirical level these problems boil down to parameter instability of the standard demand for money function (see e.g. Judd and Scadding, 1982 and Roley, 1985 for surveys). In the search to account for parameter instabilities in an empirically satisfactory way one should be able to find a satisfactory solution to many problems like how to measure the relevant concept of money in the light of developments in the financial markets, how to distinguish between exogenous and endogenous variables in the demand for money function and how to specify the functional form and dynamics. While redefining the demand for money concept may to some extent alleviate instability problems, it is hard to argue that the problem lies only in the measurement of money concept. Anyway in what follows we ignore measurement aspects and concentrate on some modelling issues associated with the demand for money. We use US quarterly data over the period 1951:1 – 1983:4.