| || || Kumar, Saten.|
| || || Demand for money in developing countries : alternative estimates and policy implications|
Institution: University of the South Pacific.
Call No.: pac In Process
Copyright:Under 10% of this thesis may be copied without the authors written permission
Abstract: This thesis applies alternative time series techniques viz., the general to specific approach (GETS) of the London School of Economics and Hendry1, the fully modified ordinary least squares (FMOLS) of Phillips and Hansen (1990) and the Johansen (1988) maximum likelihood (JML) approach to estimate demand for money functions for fifteen developing countries. The selected countries are Fiji, Vanuatu, Samoa, Solomon Islands, Tonga, India, Indonesia, Philippines, Thailand, Bangladesh, Kenya, Malawi, Jamaica, Rwanda and South Africa. Our results show that they are consistent with prior expectations and give similar summaries of the observed facts. Our results are also consistent across the three methods of estimation and imply that they may differ in precision but only marginally. We have also used the Gregory and Hansen (1992) structural break procedure to test the stability of demand for money but only for Fiji. Our estimates of the income elasticities are close to unity for all countries in our sample, except for the Philippines where it is higher at around 1.25, and the semi-interest rate elasticities have the expected negative signs, well determined and significant. Further, the dynamics are adequately captured with the GETS procedures. Dummies for political events and financial reforms together with the lagged inflation rates seems to have affected money demand in some countries. Stability tests show that demand for money functions in all our sample countries are temporally stable. This implies that the respective monetary authorities should consider using money supply, instead of the rate of interest, as their monetary policy instrument. Therefore, the switch in these countries to using the bank rate as their monetary policy instrument is hard to justify. If these countries want monetary stability then they should increase the real money supply at the rate at which real GDP grows.