Abstract
This study examines the Granger-causal relationships between oil price movements and global stock returns by using time-varying Granger-causality tests in mean and in variance. We use the daily returns from Morgan Stanley Capital International (MSCI) G7 and the MSCI Emerging Stock Market Indexes to distinguish between the effects of daily oil price movements on G7 countries’ and emerging market countries’ stock markets. We further divide the emerging markets into two groups as oil-exporting and oil-importing countries. For the oil market, we use both the West Texas Intermediate (WTI) and Brent oil daily price movements. While the Granger-causality-in-mean tests indicate a causal link from WTI oil prices and G7 countries’ stock returns to MSCI emerging countries’ stock returns, the Granger-causality-in-variance tests suggest no causal link from global oil market prices to stock market returns. Nonetheless, a causal link from the G7 countries’ stock returns to the MSCI emerging countries’ stock returns is detected. In addition, G7 countries’ stock market volatility is found to Granger-cause Brent oil price volatility. The time-varying Granger-causality-in-mean and Granger-causality-in-variance tests present new and further insights. A causal relationship between oil price changes and G7 countries’ stock returns is found for some periods during and after the global financial crisis. Time-varying Granger-causality-in-variance test results indicate evidence of causal linkages among oil prices and global stock market returns that are specific only to certain time periods. We also find that there might be a difference between the movements in Brent and WTI oil prices with respect to their Granger-causal effects on oil-importing emerging markets’ stock returns—especially after the global financial crisis. Our results provide further evidence that the effects of oil price movements on stock returns might be different depending on the volatility in the stock markets.
Highlights
In the face of large volatility in crude oil prices, the effects of oil price movements on economic performance have led to a large amount of literature
A likelihood ratio (LR) test is employed to determine which model is more suitable in modeling the volatility of the return series, and the test results are presented in Table 3 (the LR test can be calculated by using the formula, LR = 2[L(Md)−L(M)], where L(Md) and L(M) are the maximum log likelihood values derived from the exponential GARCH (EGARCH) models with and without dummy variables, respectively)
Time-varying causality-in-variance test results indicate the presence of bidirectional causal links among the crude oil price changes and global stock market returns at specific time periods
Summary
In the face of large volatility in crude oil prices, the effects of oil price movements on economic performance have led to a large amount of literature. Creti et al [7] employed spectral analysis to examine the presence of time-varying dynamic relationships between stock market indices and oil prices separately for oil-importing and oil-exporting countries. Bastianin et al [12] examined the effect of oil supply and demand shocks on the stock market volatility for G7 countries by using the structural Vector Autoregression (VAR) model for the period, 1973–2015. We consider the effects of possible breaks in the series by using Inclan and Tiao’s [17] and Sanso et al.’s [18] procedures as the failure to do so would bias the causality test results As it will be discussed in detail, the stock return and crude oil price series are filtered using an EGARCH (1,1) (version 10, Eviews, IHS Inc., London, UK) specification with Generalized Error Distribution (GED) errors.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
