| Popis: |
Money demand is probably one of the most extensively studied economic relationship in applied economics. While useful surveys of existing literature are available, much of the attention has focused on the US. However, a considerable number of papers have recently been produced dealing with the situation in the EU, both at a country and at an area- wide level, with much of this research being carried out at EU central banks. Therefore, it appears useful to also examine this recent work, both in order to assess the current situation, and to guide future research. The first part of the paper covers a range of general issues arising such as the theoretical models, the specifications, the variables employed, and estimation techniques that have been used. It is emphasised that the basis of all these papers appears to be a benchmark model, in which money is a function of a scale variable, of interest rates and, when necessary, of variables accounting for financial innovation. The second part of the paper focuses on the estimated equations for the individual countries, paying particular attention to the case of Germany. A reasonable summary of the results obtained in general is that money demand equations perform fairly well in EU countries. Estimated parameters have the signs, if not always the magnitudes, predicted by economic theory. In most cases, the evidence points to the existence of the long-run equilibrium relation between money and a few determinant variables (real income, prices and interest rates) although the size of the adjustment coefficients indicate that deviations from the steady state may be of long duration. The next section provides a review of empirical evidence of aggregate money demand in grouping of EU countries taken together, an area in which interest is increasing as EMU approaches. In this strand of the literature, there seems to be a consensus that EU-wide equations yield satisfactory results, with area-wide equations often performing better than comparable national equations. Some reasons underlying this result are also examined. |
