Money demand is an important element of monetary policy analysis. Inflation is supposed to be a monetary phenomenon in the long run, and the empirical relation between money and prices is usually discussed in a money demand framework. The main purpose of money demand studies is to analyze if a stable money demand function exists in a specific country, especially when a major structural change has taken place. Examples for such structural changes are the monetary union of West Germany and the former German Democratic Republic in 1990 and the introduction of the Euro in 1999. There is broad evidence that money demand has been quite stable both in Germany and in the Euro area, see for example Wolters, Teräsvirta and Lütkepohl (1998) and Holtemöller (2004a).
In this chapter, we explore the M2 money demand function for
Indonesia in the period 1990:1-2002:3. This period
is dominated by the Asian crises, which started in . In the
aftermath of the crisis, a number of immense financial and
economic problems have emerged in Indonesia. The price level increased by
about
percent in
compared to the previous year. In the
same period, the call money rate increased temporarily from
percent to
percent and the money stock increased
by about
percent. Additionally, Indonesia has faced a sharp
decrease in real economic activity: GNP decreased by about 11 percent.
Given these
extraordinary economic developments, it may not be expected that a
stable money demand function existed during that period.
The main contribution of this chapter is twofold. Firstly, we provide a careful analysis of Indonesian money demand, an emerging market economy for which only very few money demand studies exist. Secondly, we do not only apply the standard econometric methods but also the fuzzy Takagi-Sugeno model which allows for locally different functional relationships, for example during the Asian crisis. This is interesting and important because the assessment of monetary policy stance as well as monetary policy decisions depend on the relationship between money and other macroeconomic variables. Hence, a stable money demand function should be supported by various empirical methodologies.
In Section 11.1 we discuss money demand specification generally and
in Section 11.2 we
estimate a money demand function and the corresponding error-correction model
for Indonesia using standard regression techniques. In Section 11.3,
we exploit the fuzzy approach and its application to money demand.
Section 11.4 presents conclusions and a comparison of the two approaches.