This paper examines the use of econometric tests for causality that are based on the work of Granger and Sims. The main conclusion is that it is not possible to infer the direction of causation from observed correlations. There are two reasons why the causality tests cannot reveal the direction of causation from observed correlations. First, the null hypothesis tested is necessary but not sufficient to imply causality. Second, any specification error renders the causality test results uninterpretable.
No takes yet. Share an insight, caveat, or question.
Jacobs et al. (1979) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: