Does Only Unanticipated Mone¬tary Growth Matter? An Econometric Investigation of Ten Asian Countries

This paper is the first attempt to provide econometric evidence on the response of real output, and the price level to movements in unanticipated as wel1 as anticipated monetary growth for ten Asian developing countries in the context of New Classical rational expectations - natural rate theory. Since predictions from ARIMA model are non-stationary, unlike many other previous works in this field, ARMA modelling of the growth rate of money supply is employed here for each country; the residuals of the selected model are used as the unanticipated variable, whereas the predicted values of the model are used as the anticipated variable. Moreover, unlike most other works (in which inefficient two-step OLS estimation was used), in view of the existence of generated regressors in the structural equations, we used Systems Maximum Likelihood Method (in which all three equations - both the expectation and the other two structural equations for output and price - are estimated simultaneously) in order to render the estimates both consistent and efficient. The tentative results of our empirical tests are more in line with the non-neutrality postulate of the Non-Classical or New Keynesian rational expectations model. The findings appear to be in conflict with the New Classical neutrality hypothesis which maintains that only the unanticipated monetary growth matters for real economic activity. The evidence tends to suggest, rather strongly, that during the period under review (1955-1990), the anticipated monetary growth although significantly affected the rate of growth of prices, did also have significant impact on real output in all ten Asian countries under study. This finding will have distinct implications for macroeconomic stabilization policies in all these ten countries.