The Causal Relationship between Economic Policy Uncertainty and Stock Returns in China and India: Evidence from a Bootstrap Rolling-Window Approach

This paper examines the causal link between economic policy uncertainty and stock returns in China and India, using bootstrap Granger full-sample causality test and sub-sample rolling window estimation. We use monthly data covering from 1995:02 to 2013:02 for China and 2003:02-2013:02 for India. The bootstrap full-sample Granger causality test suggests no evidence of any causality between economic policy uncertainty and stock returns for the two countries. However, taking structural changes into account, we assess stability of parameters of the estimated vector autoregressive (VAR) models. We find both the short-run and long-run relationships between economic policy uncertainty and stock return estimated using full-sample data are unstable over the sample period. This suggests that full-sample causality tests cannot be relied upon. We turn to propose a time-varying (bootstrap) rolling window approach to revisit the dynamic causal relationship between the two variables. Using a rolling window of 24 months, we do find that there are bidirectional causal relationships between stock returns and EPU for several sub-periods in China and India. However, the association between EPU and stock returns is, in general, weak in these two emerging countries. These findings have important implications for policy makers as well as investors.