A bootstrap-corrected causality test: Another look at the money-income relationship

Previous studies of the causal relationship between money supply and real output are based on asymptotic distributions. If the assumption of normality is not fulfilled and if ARCH effects are present, asymptotic distributions perform inaccurately. In this paper, we reinvestigate the potential causal relationship between money and output by applying an alternative methodology based on the leveraged bootstrapped simulation techniques using data from Denmark, Japan, Sweden, and the US. We find unidirectional causality from money to output for the sample countries except for Sweden for which causality is bi-directional. This finding of unidirectional causality between money and output supports monetary business-cycle models and reveals one important policy implication-that is, in looking for the sources of output fluctuations, money might be a major factor.