Stability of the money demand function in Asian developing countries

Abstract

Previous studies that estimated the money demand function in Asian developing countries either employed traditional estimation techniques or recently popularized cointegration technique. While the first group suffers from ‘spurious regression’ problems, the second group interpreted their finding of cointegration as a sign of stability of estimated parameters. This study, after incorporating the CUSUM and CUSUMSQ tests into cointegration analysis, shows that in some Asian countries even though real M1 or M2 monetary aggregates are cointegrated with their determinants, the estimated parameters are unstable.

Notes

It should be mentioned that Hoffman et al. (1995) and Bahmani-Oskooee et al. (1998) who applied the Hansen-Johansen (1993) test to determine whether the number of cointegrating vectors and their associated eigenvalues changed over time. In the absence of any significant change in eigenvalues, they interpreted their finding as a sign of stability. However, when the short-run dynamics was incorporated in the testing procedure, Bahmani-Oskooee and Bohl (2000) and Bahmani-Oskooee and Barry (2000) found instability in the demand for money in Germany and Russia respectively.

It should also be indicated that Bahmani-Oskooee and Malixi (1991), Arize (1994), and Khalid (1999) included a few Asian countries in their analysis of the demand for money.

The finding of an unstable demand for money for Korea was similar to the results obtained for Germany by Bahmani-Oskooee and Bohl (2000) and for Russia by Bahmani-Oskooee and Barry (2000).

For more on the expected sign of d see Arango and Nadiri (1981) and Bahmani-Oskooee and Pourheydarian (1990).