Time-varying causality impact of economic policy uncertainty on stock market returns: Global evidence from developed and emerging countries

In this study, we examine the causal relationship between economic policy uncertainty (EPU) and stock indices in the Group of Seven and Emerging Market Seven nations from 1997 to 2022, using a novel time-varying Granger causality test based on the lag-augmented vector autoregressive model. The main findings are as follows. First, the results of the linear and non-linear Granger tests for the whole sample period show that either the global or national EPU has a causal impact on the stock market in less than half of the 14 countries, while the stock market is more likely to affect EPU. Second, the time-varying Granger causality tests show that the period in which the EPU affects the stock market is often accompanied by a major international crisis or national event. Global EPU has a relatively significant impact on developed market equity markets compared to domestic EPU. Resource exporting emerging nations tend to be significantly affected by global EPU, but developing countries with complete industrial systems or unstable political situations are more likely to be affected by local EPU. Finally, multi-scale, time-varying Granger causality tests show that both global and domestic EPUs have a significant causal impact on equity markets in the long run, although the impact of global EPUs is concentrated in the short term, whereas the impact of domestic EPUs appears to be longer-term. The new coronavirus (COVID-19) has dramatically amplified the short-term impact of global EPU. Our findings support political implications for the development of government policies and investors' global asset allocation.