From Dornbusch to Murphy: Stylized monetary dynamics of a contemporary macroeconometric model

Newer applied macroeconometric models have adopted a mixture of paradigms t o capture at least the major differences between the behavior of financial markets and markets for commodities and for factors. Thus, in the Murphy Mo del of the Australian macroeconomy (MM), rational expectations and very rap id adjustment are assumed to apply in markets for financial assets, but fri ctions and departures from rationality manifest themselves in the short-ter m behavior of other markets. A popular story for the reaction of exchange a nd interest rates to monetary shocks is provided by Dornbusch's 1976 oversh ooting exchange rate model (DBM). The rational-expectations version of this archetypal model underpins MM, yet the dynamic adjustment paths of variabl es in MM differ markedly from those in DBM. A leading case in point is the exchange rate which in MM actually undershoots its new equilibrium value af ter the injection of a monetary shock. This paper gives a simplified accoun t of how the differences come about. The emphasis is not so much on theoret ical rigor as on providing a convincing practical demonstration. Using the simplest form of DBM as a starting point, it is shown how one can progressi vely develop a miniature model exhibiting an MM-like response to a monetary shock. The key idea in this development is that aggregate demand does not respond instantaneously to shocks in the macroeconomic environment. The veh icle used to implement the numerical miniature model is a computer spreadsh eet. Although the specifics relate to MM, the generics of this development seem likely to be applicable to most models adopting the mixture of paradig ms identified above. This conclusion survives the current proclivity to rep lace Dornbusch's money demand function in the larger models by a policy rea ction function in which interest rates are manipulated to control inflation . (C) 2000 Society for Policy Modeling. Published by Elsevier Science Inc.