Contents lists available at ScienceDirect International Review of Financial Analysis journal homepage: www.elsevier.com/locate/irfa Does political risk matter for sovereign wealth funds? International evidence ☆ Jocelyn Grira a, ⁎ , Chiraz Labidi a , Wael Rouatbi b a United Arab Emirates University, United Arab Emirates b Montpellier Business School, France ARTICLE INFO Keywords: Sovereign wealth funds Political risk Conflicts Quality of institutions Democratic tendencies ABSTRACT We investigate the impact of political risk on the investment decisions of sovereign wealth funds (SWFs). Using an international sample of 302 targets involved in 427 SWFs' deals, we find that political risk matters in de- termining SWFs' portfolio strategies. Among the four dimensions of political risk, we show that conflicts and democratic tendencies are the main components that explain variations in SWF behaviour, whereas the quality of institutions and government action matter less. Our results are robust to a battery of sensitivity tests, alter- native model specifications, subsample analysis, and cultural bias. 1. Introduction Asset pricing theory predicts higher yields on investments bearing higher non-diversifiable risk. As argued by Butler and Joaquin (1998), political risk, which has been increasing steadily in the last decade (Belkhir, Boubakri, & Grira, 2017), necessarily implies higher required returns on foreign direct investments if that risk cannot be diversified away. The last decade also witnessed the rise of sovereign wealth funds (SWFs) as major institutional investors on international financial mar- kets (Johnson, 2007). Their activities have created media and political controversy, specifically during the 2007–2008 global financial crisis (GFC), as some countries hosting SWF investments viewed these as the financial arms of the SWFs' own countries, hence creating concerns of potential political interference (Megginson & Fotak, 2015). SWFs pre- sent, indeed, some unique characteristics (i.e. state ownership and opacity) that distinguish them from other institutional investors and their alleged lack of proper governance and transparency might distort their objectives and priorities. For instance, as state-owned funds, SWFs might not strictly act within a wealth maximization and/or a risk-return trade-off framework and may promote broader national interests. Previous literature has analysed potential motives and factors ex- plaining SWFs' investment decisions and reports mixed results. While a few studies argue that SWFs present nonfinancial motives in their in- vestment decision-making process (e.g., Knill, Lee, & Mauck, 2012), many studies claim that SWFs are mostly rational investors (e.g., Bortolotti, Fotak, Megginson, & Miracky, 2015; Boubakri, Cosset, & Grira, 2016; Dewenter, Han, & Malatesta, 2010; Kotter & Lel, 2011). In this paper, we argue that rationality not only encompasses risk–return optimization but also includes cultural and political biases (Chhaochharia & Laeven, 2009). We build on the rationality hypothesis and argue that SWFs should necessarily account for political risk. As political risk is mainly non-diversifiable, it should be priced. If SWFs are rational investors, they will account for political risk in their investment decisions and price it accordingly. On the other hand, if SWFs decisions are driven by other agendas, the underlying motives behind their ca- pital allocation could deviate from asset pricing theory, thus not ne- cessarily pricing political risk. We use an international sample of 302 targeted firms involved in 427 SWFs' deals and show that political risk, proxied by the International Country Risk Guide (ICRG) index, matters in determining SWFs' portfolio strategies. Furthermore, we use the decomposition of Bekaert, Harvey, Lundblad, and Siegel (2014) of political risk into four categories—conflicts, quality of institutions, democratic tendencies, and government action—and show that conflicts and democratic ten- dencies are the main components that explain variations in SWF be- haviour, whereas the quality of institutions and government action matter less. Our findings support the rationality hypothesis, which is consistent with the findings of Bortolotti et al. (2015), Boubakri et al. (2016), Dewenter et al. (2010), and Kotter and Lel (2011). Our results are robust to a battery of sensitivity tests, alternative model https://doi.org/10.1016/j.irfa.2018.07.013 Received 1 March 2018; Received in revised form 28 June 2018; Accepted 23 July 2018 ☆ We gratefully acknowledge financial support from United Arab Emirates University's Program for Advanced Research, from United Arab Emirates University's Center for Public Policy and Leadership, and from Abu Dhabi Department of Education and Knowledge. ⁎ Corresponding author. E-mail addresses: jocelyn.grira@uaeu.ac.ae (J. Grira), labidi@uaeu.ac.ae (C. Labidi), w.rouatbi@montpellier-bs.com (W. Rouatbi). International Review of Financial Analysis xxx (xxxx) xxx–xxx 1057-5219/ © 2018 Published by Elsevier Inc. Please cite this article as: Grira, J., International Review of Financial Analysis (2018), https://doi.org/10.1016/j.irfa.2018.07.013