Asymmetric Effects of Economic Policy Uncertainty  on Chinese Stock-Market Returns: Evidence from Advanced Quantile-Based Approaches

Abstract

This paper investigates the asymmetric impact of economic policy uncertainty (EPU) on stock returns in China from January 3, 2017, to December 30, 2022.  To detect the asymmetric and heterogeneous relations between variables, we employ a series of quantile-based techniques, including quantile-on-quantile regression (QQR), nonparametric causality in quantiles, and cross-quantile (CQ) dependence.  Indicatively, the empirical findings are in folds.  First, from the QQR, EPU almost exerts a negative effect on stock returns in the mid-low quantile of stock returns (0.05–0.55).  This negative association is stronger in some tail quantiles of EPU.  In contrast, there are overall positive linkages between EPU and Chinese stock returns in the mid-high quantile of the stock market (0.6–0.95), and this positive effect is more pronounced at high quantiles of EPU and stock returns. Second, we find a Granger causality from EPU to stock returns at quantiles 0.15–0.85.  These causal effects are nonlinear and asymmetrical across quantiles. Third, the CQ result reveals that dependencies between EPU and stock returns are asymmetric at lag 1.  The EPU has negative directional predictability for stock returns when the EPU is in the low-mid quantiles (< 0.6 quantile).  In contrast, when the EPU is in high quantiles (>0.8 quantile), it is more likely to have a significant positive directional predictability. Therefore, this paper provides a comprehensive analysis of the interdependence between EPU and stock returns.  The findings contribute significantly to existing research on EPU and its correlation with the stock market. These results can help governmental authorities and investors formulate dynamic economic policies and investment strategies.

Keywords: quantile-on-quantile, Causality in quantiles, Cross-quantile dependence, Economic Policy Uncertainty, stock returns