A bivariate Granger-causality test on money and output finds statistically significant causality when data are measured in log levels, but not when they are measured in first differences of the logs. Bootstrap simulation experiments indicate that, most probably, the first difference results reflect lack of power, whereas the level results reflect Granger-causality that is actually in the data The reason for the lack of power in the first difference F-statistic is that first differencing the data appears to entail a specification error. By showing that money does Granger-cause output in the bivariate relation, we remove a potential embarrassment for models that assign an important role to money in business fluctuations.
ASJC Scopus subject areas
- Economics and Econometrics