Instability and Nonlinearity of Long-Run Money Demand: Econometric Theory and Empirical Analysis (OeNB-Nr. 18692)

Many central banks, including the ESCB and the FED, are currently reviewing their strategies. Despite a less prominent role of the monetary pillar due to (i) apparent money demand instability and thus presumably diminished predictive power on the one hand and (ii) a shift towards interest-rate rules centered monetary policy on the other, a detailed understanding of money demand - in particular its determinants, its stability and its functional form - remains important for assessing the transmission of monetary policy, in particular over the medium- to long-run. The project will identify sources of instabilities or nonlinearities of money demand and the determinants of money demand. The empirical analysis will be performed for a large number of countries over long periods, which implies that detailed understanding of the (dis-)similarities of money demand across countries and episodes is gained; providing a sound basis for reassessing the importance of money demand for monetary policy.The project has two intertwined components: Econometric theory development and encompassing empirical application. We tackle the problems plaguing money demand analysis by developing a smooth transition cointegration centered modelling cycle that: (i) allows to identify the relevant variables, (ii) allows to test for instabilities and nonlinearities, (iii) is flexible with respect to the types or shapes of instabilities and/or nonlinearities and (iv) identifies the drivers and shapes of the instabilities and nonlinearities.The developed methods will be used to analyze in great detail long-run money demand for as many countries as possible for as long periods as possible; including a large number of variables potentially relevant for money demand. This will lead to thorough understanding of the similarities and dissimilarities of money demand across countries and periods, including existence, stability and form of long-run money demand relations as well as the relevant variables.The applied research methods correspond to the intertwined components of the project - econometric theory development and empirical application. This means that we employ the following research methods: (i) development of econometric theory, (ii) a careful implementation and performance evaluation of the developed methodology to gauge its usefulness and applicability (as well as the relative performance, where applicable, to competing methods), (iv) careful construction of encompassing data sets, (iv) detailed and thorough applied econometric analysis using our new methodology as well as a battery of existing time-series econometric tools, mostly for preliminary analyses.

Weitere Informationen