Dynamic risk spillovers from oil to stock markets: Fresh evidence from GARCH copula quantile regression-based CoVaR model

This study proposes a GARCH copula quantile regression model to capture the downside and upside tail dependence between oil price change and stock market returns at different risk levels. In the model, ten copulas are provided to measure the nonlinearity of the tail dependence with the marginal distribution built on the GARCH family models. Using daily price data of stock markets in ten important countries and Brent oil market, we estimate the downward and upward risk spillovers from oil to stock markets. The empirical results suggest strong evidence of risk spillover effects from oil to stock markets. Furthermore, downside and upside risk spillovers from oil to the Italian and German stock markets, and the Brazilian and Russian stock markets are the largest for developed countries and emerging market countries, respectively. The US and Mexican stock markets display the smallest downside and upside risk spillovers for two types of countries. We also find evidence that the downside risk spillovers are larger than upside risk spillovers, a finding which is consistent with the flight-to-quality phenomenon. Finally, the dynamic risk spillover effects show heterogeneity over time and are compar-atively different for each country. Our results provide significant implications for portfolio managers and in-ternational regulators who want to optimize their investment portfolios and maintain stock market stability.

Automated summary

This study proposes a model to capture the tail dependence between oil price change and stock market returns, and finds strong evidence of risk spillover effects from oil to stock markets. The results also show heterogeneity in risk spillovers among different countries, with larger downside risk spillovers and consistent with the flight-to-quality phenomenon.