Asymmetric effects of oil price shocks on stock returns: evidence from a two-stage Markov regime-switching approach

ABSTRACT

The asymmetric effects of oil price shocks on stock returns have attracted the attention of many researchers in the past several decades. Most of these researchers’ studies, however, do not separate out the sources of oil price shocks when examining the asymmetric effects. In this article, we address this limitation using a two-stage Markov regime-switching approach. Our results indicate that oil supply and demand shocks have a null or minimal impact on stock returns in a low-volatility regime and a statistically significant impact in a high-volatility regime. We observe that oil demand shocks affect stock returns significantly more than oil supply shocks. A positive aggregate demand shock significantly increases stock returns, whereas a positive oil-specific demand shock markedly decreases stock returns. These results have important implications for policymakers and investors.

KEYWORDS:

• Asymmetric effects
• stock returns
• oil price shocks
• Markov regime-switching

JEL CLASSIFICATION:

• G10
• G15
• C13

Acknowledgements

This research is supported by the National Natural Science Foundation of China under grants No. 71671062, No. 71521061 and No. 71431008.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ See, for example, Asea and Blomberg (1998), Cermeno (2002), Mizrach (2005), Chen (2007), Aloui and Jammazi (2009), Reboredo (2010), Balcilar and Ozdemir (2013), Balcilar, Gupta, and Miller (2015), Zhang and Zhang (2015).

² For example, the First Gulf Crisis in 1991, the Asian financial crisis in 1997–1998, the terrorist attack on the World Trade Center in 2001, the second Gulf War in 2003, the global financial crisis in 2007–2008, the European sovereign debt crisis starting in 2009, the Europe and US sanction against Russia in 2014, among others.

³ There are two points need to note. First, the lags of the oil structural shocks do not be included as explanatory variables in Eq. (3). That is because that the stock prices are the present discounted value of the future net earnings of firms, both the current and the expected future impacts of an oil price shock should be absorbed fairly quickly into stock prices and returns, without having to wait for those impacts to actually occur (Jones, Leiby, and Paik 2004). Second, we include one period lag of the stock market return as an explanatory variable because this specification provided better regression fit and residual diagnostics than a model without the lagged stock market return variable.

⁴ In fact, we also try to consider the three-state model, and the simulation indicates that the estimated results for one of the states are not significant. Comparatively, the two-regime model appeared superior. Therefore, here we estimate only the two-state model.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China: [Grant Numbers 71671062, 71521061 and 71431008].