Granger causality from exchange rates to fundamentals: What does the bootstrap test show us?

•The existing Granger causality evidence is not strong enough.•The new evidence of causality relation from exchange rates to fundamental is weak.•Findings in the previous study are caused by size-distortion.•The bootstrap test performs better than the asymptotic test in this application.

We use a residual-based bootstrap method to re-examine the finding of the Granger causality relation from exchange rates to fundamentals in Engel and West (2005), in which the relation is taken as evidence for their explanation for the present-value model for exchange rates. Our test results are against the previous findings. The Monte Carlo experiment results suggest that the previous evidence for the causality relation from exchange rates to fundamentals is very likely caused by the size distortion. Hence, the existing Granger causality evidence is not strong enough to validate the new explanation for the present-value model.