SFB 303 Discussion Paper No. B - 11


Author: Weissenberger E., and G. Müller-Brockhausen
Title: Exchange Rates, Interest Rates and Money: An Empirical Investigation
Abstract: In this paper we adopt an econometric-statistical perspective, although the specific methodology followed here differs from approaches in recent studies. Given that we have only vague a priori beliefs of which particular structure determines the stochastic interaction between interest rates, money, prices, income and exchange rates, we take a fairly eclectic approach. The techniques advocated by Sims (1980), (1981) seem ideal for our purpose.
This paper consists of two main parts. In the first part, an unrestricted five. Variable vector autoregression (VAR) is estimated and analysed applying Sims's innovation accounting technique. The three bilateral cases of the USA and the Federal Republic of Germany (FRG). US and Switzerland (CH) and the Federal Republic of Germany and Switzerland are studied using monthly data for the period from 1974 to 1984. The five variables in the system, each of which may depend on the history of all the other variables are: money stock (M1), industrial production (IP), consumer price index (CPI), interest rate (R) and exchange rate (X). Employing the VAR we shall examine the Granger causal orderings to be found in the data. Using this representation it is straight-forward to test the often imposed hypothesis of exogenous (with respect to the exchange rate) money and income. Next the moving average representation (see Sims (1980)) of the system is used to trace the temporal impact of the innovation in a particular variable on all the variables in the system. We find that some of these responses are quite the opposite to the predictions of the standard type models of the exchange rate.
Structural models of the exchange rate imply restrictions on the VAR. In the second part of the paper we look more specifically at some of these restrictions. For example the efficient asset market hypothesis implies that new information is instantaneously reflected in the asset price. The causal connection between the forcing variable (e.g. the money supply) and the exchange rate is then reflected in the contemporaneous correlation between the forcing variable and the innovation in the exchange rate. This implication of the efficient market hypothesis will be tested by analysing the structure of the variance-covariance matrix of the innovations.
Keywords:
JEL-Classification-Number:
Creation-Date: January 1985
Unfortunately this paper is not available online. Please contact us to order a hardcopy.

SFB 303 Homepage

30.09.1999, Webmaster