Grexit news and stock returns

ABSTRACT

During the first 8 months of 2015, there was an ongoing debate about whether or not Greece should remain in the euro area. Using an event study approach, we quantify the effects of Grexit-related statements made by six important euro area politicians (Merkel, Schaeuble, Tsipras, Varoufakis, Juncker, and Schulz) on intraday stock returns in Germany, Greece, and the euro area during the period of 1 January 2015–19 August 2015. We show that positive statements indicating that a Grexit is less likely lead to higher returns, and negative statements to lower returns. The overall impact of negative statements is more pronounced. The cumulative absolute effects on stock returns are sizeable as the statements contribute to a variation of up to 58 percentage points in the ATHEX. These large effects are of particular relevance as our study only captures an 8-month snapshot of the Greek Government debt crisis.

KEYWORDS:

• Event study
• Grexit
• intraday data
• political statements
• stock returns

JEL CLASSIFICATION:

• G01
• G12
• G14

Acknowledgements

Thanks to Tobias Kranz, Florian Neumeier, and seminar participants at the University of Muenster for their helpful comments on earlier versions of the article. The usual disclaimer applies.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplemental Material

Supplemental data for this article can be accessed at here.

Notes

¹ A more established branch of the literature focuses on the determinants and consequences of central bank communication. See Blinder et al. (2008) for an extensive literature overview.

² This corresponds to 29 June 2015, 9:30 in case of the DAX and EUROSTOXX and 3 August 2015, 10:00 in case of the ATHEX. The inclusion of this impulse dummy is required to overcome heteroscedasticity in the residuals. See also Table 2.

³ All statements alongside their coding can be found in the Online Appendix to this article.

⁴ Figure A1 in the Supplementary Appendix provides an overview of the distribution of positive and negative statements by politician over time.

⁵ The large coefficient for negative statements by Schulz should not be over interpreted as the estimate is based on two observations only (see Table 1). In addition, the statement on 4 February 2015, at 19:57, coincides with the ECB’s decision to no longer accept Greek bonds as collateral in its monetary policy operations, which is also clearly very negative news for the Greek stock market.

⁶ Note that the coefficients on ${\gamma }_1$ indicate a significant leverage effect in the euro area stock market as negative standardized residuals exert a stronger effect on the conditional variance (${\alpha }_1-{\gamma }_1$) than positive ones (${\alpha }_1+{\gamma }_1$).