Knowledge that Transforms

CONSUMERS' expenditure forms a major component of the Gross National Expenditure and the parameters of this function play a crucial role in the mechanism of income generation in a macro-model. Because of their importance, a considerable amount of care and attention should be devoted to the estimation of these parameters and their probable movement. This consideration becomes especially important in mediumand long-range forecasting because over a long-run horizon the consumer expenditure parameters may be strongly influenced by socio-economic factors such as female participation in the labour force, education level of the population, urbanization, etc. In this paper we make a beginning in the direction of estimating parameters of the consumption function which are shifting over time due to these socio-economic factors. The importance of taking into account the shifting parameters in a macro-equation can also be emphasized from another point of view. It is becoming increasingly clear that macroeconomic analysis and policy will be seriously inadequate in a medium-run horizon unless attention is paid to changes in the composition of aggregative units. One line of attack is, of course, to have disaggregative models. However, quite often the detailed time-series data are not available on all the variables concerned. Even when the data are available, they are somewhat unreliable at a disaggregate level and in statistical estimation the ratio of noise to signal tends to be high in such disaggregate equations. Moreover, there are often some overall macro-constraints which are difficult to take into account in disaggregate models. It is, therefore, desirable as a supplement (if not as an alternative) to the disaggregated models, to have a macro-equation capture the compositional effects as far as possible. Our DaDer is one example of an attempt to capture the influence of changing population composition in a macro-equation, by introducing shifting parameters. For purposes of estimating these shifting parameters, we make combined use of cross-section and time-series data. The lagged reaction path of consumption due to income and price changes is also considered, and we use the Almon technique (1965) for this purpose.

T HIS study considers the market for Eurodollar deposits as a structural system with determinate demand and supply relationships. It is important at the outset to distinguish the stock demand for Euro-dollar deposits, which we shall be investigating here, from the flow demand for Euro-dollar credit. Machlup (1970) points out that the demand for Euro-dollar credit is a demand for a flow of funds to borrowers who, in turn, plan to pay out the funds they have borrowed. The demand for Euro-dollar deposit balances is, on the other hand, a demand for money, or near-money, to hold. We shall be concentrating upon the factors affecting the demand for a stock of Eurodollar deposits to hold.1 Since we shall not be taking the stock of Euro-dollar deposits in existence at a point in time as exogenously given, we shall also be identifying an equation to determine the stock of Euro-dollar balances supplied by Euro-dollar issuing institutions (hereafter Euro-banks). This will be based on an identifiable stock of reserves held by Euro-banks. Briefly, the procedure and results are as follows. Quarterly data from 1964-III through 1970-IV are employed to obtain parameter estimates for scale and substitution arguments in a stock demand function for Euro-dollar deposit balances as well as to obtain an estimated equation for the stock of dollar claims produced by Euro-banks. The empirical results suggest that a stable stock-demand function for Euro-dollar deposit balances exists along with a stable stock-supply function for Eurodollar deposits. The results also suggest that about 40 per cent of the growth of Euro-dollar deposits in the 1964-III 1970-IV period was due to the multiple deposit expansion process.

Paul R. Gregory, John M. Campbell, Benjamin S. Cheng, A Simultaneous Equation Model of Birth Rates in the United States, The Review of Economics and Statistics, Vol. 54, No. 4 (Nov., 1972), pp. 374-380

Yoel Haitovsky, George Treyz, Forecasts with Quarterly Macroeconometric Models, Equation Adjustments, and Benchmark Predictions: The U.S. Experience, The Review of Economics and Statistics, Vol. 54, No. 3 (Aug., 1972), pp. 317-325

Goldberger, A. S., Econometric Theory (New York: John Wiley and Sons, Inc., 1964). Johnston, J., Econometric Methods (New York: McGraw-Hill Book Company, 1963). Yotopoulos, P. A., and L. J. Lau, A Test for Balanced and Unbalanced Growth, this REVIEW, LII (Nov. 1970), pp. 376-383. Yule, G. U., and M. G. Kendall, An Introduction to the Theory of Statistics (14th edition) (London: Griffin and Company, Ltd. 1950).

"Inl statistical inference proper, the model is never questioned.... The methods of mathe-matical statistics do not provide us with a means of specifying the model. " 1