Major central banks stress the importance of money growth analysis
and of a stable money demand function for monetary policy
purposes. The Deutsche Bundesbank, for example, has followed an
explicit monetary targeting strategy from to
, and
the analysis of monetary aggregates is one of the two pillars of
the European Central Bank's (ECB) monetary policy strategy.
Details about these central banks' monetary policy strategies, a
comparison and further references can be found in
Holtemöller (2002). The research on the existence and stability
of a money demand function is motivated inter alia by the
following two observations: (i) Money growth is highly correlated
with inflation, see McCandless and Weber (1995) for international
empirical evidence. Therefore, monetary policy makers use money
growth as one indicator for future risks to price stability. The
information content of monetary aggregates for future inflation
assessment is based on a stable relationship between money, prices
and other observable macroeconomic variables. This relationship is
usually analyzed in a money demand framework. (ii) The monetary policy
transmission process is still a ``black box'', see
Mishkin (1995) and Bernanke and Gertler (1995). If we are able to
specify a stable money demand function, an important element of
the monetary transmission mechanism is revealed, which may help to
learn more about monetary policy transmission.
There is a huge amount of literature about money demand. The majority of the studies is concerned with industrial countries. Examples are Hafer and Jansen (1991), Miller (1991), McNown and Wallace (1992) and Mehra (1993) for the USA; Lütkepohl and Wolters (1999), Coenen and Vega (1999), Brand and Cassola (2000) and Holtemöller (2004b) for the Euro area; Arize and Shwiff (1993), Miyao (1996) and Bahmani-Oskooee (2001) for Japan; Drake and Chrystal (1994) for the UK; Haug and Lucas (1996) for Canada; Lim (1993) for Australia and Orden and Fisher (1993) for New Zealand.
There is also a growing number of studies analyzing money demand in developing and emerging countries, primarily triggered by the concern among central bankers and researchers around the world about the impact of moving toward flexible exchange rate regimes, globalization of capital markets, ongoing financial liberalization, innovation in domestic markets, and the country-specific events on the demand for money (Sriram; 1999). Examples are Hafer and Kutan (1994) and Tseng (1994) for China; Moosa (1992) for India; Arize (1994) for Singapore and Deckle and Pradhan (1997) for ASEAN countries.
For Indonesia, a couple of studies have applied the cointegration and error-correction framework to money demand. Price and Insukindro (1994) use quarterly data from the period 1969:1 to 1987:4. Their results are based on different methods of testing for cointegration. The two-step Engle and Granger (1987) procedure delivers weak evidence for one cointegration relationship, while the Johansen likelihood ratio statistic supports up to two cointegrating vectors. In contrast, Deckle and Pradhan (1997), who use annual data, do not find any cointegrating relationship that can be interpreted as a money demand function.
The starting point of empirical money demand analysis is the choice of variables to be included in the money demand function. It is common practice to assume that the desired level of nominal money demand depends on the price level, a transaction (or scaling) variable, and a vector of opportunity costs (e.g., Goldfeld and Sichel; 1990; Ericsson; 1999):
![]() |
(11.1) |
![]() |
(11.2) |
![]() |
(11.3) |
![]() |
(11.5) |
Rearranging (11.4) yields:
Accordingly, the PAM can also be represented by an error-correction model like (11.6).