Economic policy uncertainty index and economic activity: what causes what?

This paper is a follow-up on the Economic Policy Uncertainty (EPU) index, developed in 2011 by Baker, Bloom, and Davis. The principal idea of the EPU index is to quantify the level of uncertainty in an economic system, based on three separate pillars: news media, number of federal tax code provisions expiring in the following years, and disagreement amongst professional forecasters on future tendencies of relevant macroeconomic variables. Although the original EPU index was designed and published for the US economy, it had instantly caught the attention of numerous academics and was rapidly introduced in 15 countries worldwide. Extensive academic debate has been triggered on the importance of economic uncertainty relating to the intensity and persistence of the recent crisis. Despite the intensive (mostly politically-motivated) debate, formal scientific confirmation of causality running from the EPU index to economic activity has not followed. Moreover, empirical literature has completely failed to conduct formal econometric testing of the Granger causality between the two mentioned phenomena. This paper provides an estimation of the Toda-Yamamoto causality test between the EPU index and economic activity in the USA and several European countries. The results do not provide a general conclusion: causality seems to run in both directions only for the USA, while only in one direction for France and Germany. Having taken into account the Great Recession of 2008, the main result does not change, therefore casting doubt on the index methodology and overall media bias.