Abstract

The theory of real business cycles (RBC) interprets the often found link between money and output as one of reversed causality: A broad monetary aggregate such as M1 Granger-causes output only because of the response of endogenous money to changes in the production possibilities in the economy while exogenous money is neutral. Quarterly data for the period 1964 - 1989 are used to investigate the validity of the RBC hypothesis for Germany. On the basis of variance decomposition results for various vector autoregressive systems, it is shown that first, the monetary base and even currency have an impact on output, i.e. their innovations reduce the forecast error variance of output. This effect is at least as strong as the one stemming from the real shock, i.e. changes in the terms of trade. Secondly, while the variables of endogenous money, such as the money multiplier and demand deposits, seem to have the strongest effect on output, they are themselves not dominantly influenced by real factors. In fact, monetary policy, e.g. a change in the monetary base, has a substantial impact, so the movements of endogenous money cannot be viewed as being solely a response of the public to real shocks. These findings are at variance with the RBC interpretation.