Financial liberalisation, monetary aggregates and monetary policy in the Seacen countries : an empirical investigation

Habibullah, Muzafar Shah (1998) Financial liberalisation, monetary aggregates and monetary policy in the Seacen countries : an empirical investigation. University of Southampton, Doctoral Thesis.

Abstract

The general thrust of the thesis is that, "money does matter for monetary policy purposes" for ten SEACEN countries investigated. Despite the financial liberalisation that is underway, we find evidence that the monetary aggregates - both the Simple-sum and Divisia aggregates - have close relationships with both the price level and nominal income. This suggests that they are appropriate intermediate indicators.

The thesis consists of three essays. The first essay is concerned with the effect of financial liberalisation on liquidity constraints. It is assumed that in a deregulated economy, liquidity constraints will be relaxed, and the relationship between consumption and current income will be unstable. Consequently, financial liberations would undermine the role of monetary aggregates as indicators and/or intermediate targets for monetary policy. The results suggest that the fraction of consumers who are liquidity constrained is substantial in these countries, and that financial liberalisation has no bearing in reducing liquidity constraints.

In the second essay, we formulate a P-Star (p*) model to link between money and the price level using the cointegration approach. For each country, four sets of P-Star models were tested based on four monetary aggregates - Simple-sum M1 and M2, and Divisia M1 and M2, used to construct p*. Generally, our findings suggest that p* ties together the level of money and prices in these economies. Divisia money has been found to be useful as intermediate indicator for the purpose of monetary policy action.

The third essay examines the long-run relationship between Divisia monetary aggregates (both M1 and M2) and nominal income. The series were tested for seasonal unit roots and seasonal cointegration. The results of the Engle-Granger two-step procedure for testing seasonal cointegration were compared with the estimation of the seasonal error-correction models (a direct test for seasonal cointegration). The results suggest that Divisia money and income exhibit long-run relationships not only at the long-run frequency but also at the biannual and annual frequencies. Our findings also suggest that the seasonal error-correction model for testing seasonal cointegration is more robust than the Engle-Granger two-step procedure.

This record has no associated files available for download.

Contributors

Download statistics