International political risk and government bond pricing

Abstract

This paper investigates the impact of international political risk on government bond yields in 34 debtor countries using a comprehensive database of 109 international political crises from 1988 through 2007. After employing the total number of international political crises as a proxy for political risk and controlling for country-specific economic conditions, we establish a positive and significant link between international political risk and government bond yields. This is consistent with global bond investors demanding higher returns at times of high political uncertainty. In addition, we show that international political risk has a reduced adverse effect on bond prices when the debtor country has a stable political system and strong investor protection.

Introduction

Political risk refers to the risk that arises as a result of the potential actions of governments and other political forces within and across nations; this type of risk implies uncertainty about potential changes in government policies and the impact of such policies on the future economic environment. Such uncertainty inevitably affects a country’s borrowing costs. Ample evidence connects political risk to a country’s debt pricing, with some studies even inferring political risk directly from government bond yields (see Bekaert et al. (2012) for an excellent review).⁴ However, because political reforms in a country are frequently accompanied by economic downturns, a common challenge for these studies is that it is extremely difficult empirically to disentangle changes in macroeconomic fundamentals from those of political risk (Kramer, 1971, Hibbs, 1977). In this study, we overcome this endogeneity issue of political risk by using international political crises that occur outside a country as our proxy for political risk. Specifically, we evaluate the relationship between political risk and government debt pricing in 34 countries from 1988 through 2007 and explore the marginal effects of political and legal institutions on the impact of international political risk.

A priori, we expect that political risk has an adverse impact on government bond prices. The rationale is that international political risk increases the uncertainty and instability that government bond investors face. As a result, investors in government bonds will require a higher return to compensate for bearing such uncertainty. Empirically, one would observe government bond yields to rise as international political uncertainty intensifies. To test this conjecture, it is important to measure international political risk appropriately. Unlike country-specific political uncertainty, political risk that arises from disputes among multiple countries and that bears an international influence is rare and difficult to quantify.⁵ In this study, we employ a comprehensive database, known as the International Crisis Behavior (ICB) database, of all the international political crises that have occurred since 1918. The database not only includes the number of international political crises but also documents important events in such crises in great detail (such as their beginnings and conclusions) and classifies the crises based on various characteristics, including superpower involvement, duration and gravity. Another distinctive feature of this database is that “crisis” is defined in a way that is aligned with the news events that attract investors’ interest. Specifically, a political event is deemed a crisis if it is a perceived change in the probability of a threat that results in the beginning or end of an international political crisis. In parallel, we measure a country’s borrowing costs using the end-of-month yields on local government bond indexes at 3-year, 5-year, 7-year, and 10-year maturities.

Using this detailed database of international political crises for the period from 1988 to 2007, we seek to answer two research questions empirically: (1) How and to what extent does international political risk relate to government bond yields? (2) How does the quality of political and legal institutions affect the impact of international political risk?

After controlling for country-specific macroeconomic fundamentals, our primary results reveal that the total number of international political crises in a month is significantly and positively related to end-of-month local government bond yields at various maturities. From an economic perspective, the impact of political risk is sizeable. The occurrence of one additional international political crisis leads to a 24.8–35.6 basis point increase in government bond yields. In addition, subsample analyses demonstrate that the total number of international political crises is positively associated with the yields on non-U.S. government bonds but insignificantly correlated with U.S. Treasury bonds. This finding might be consistent with the “flight-to-quality” phenomenon, whereby investors divert their investments into less risky assets (such as U.S. Treasury bonds) as political instability rises. Our primary results continue to hold when political risk is measured by the severity of political crises. Furthermore, an international political crisis exerts a significant and positive impact on government bond yields at any stage of crisis development. Overall, this evidence supports the notion that global political instability increases the investment uncertainty perceived by bond investors and leads to increased government bond yields.

Having established a positive link between international political risk and government bond yields, we continue to explore the impact of political stability and the strength of investor protection offered in debtor countries on the effects of international political risk. We expect that countries with more stable political systems are less likely to experience political changes despite a deteriorating international political environment. Furthermore, countries with strong investor protection are expected to honor sovereign debt obligations. As shown in Gennaioli et al. (2014), countries with investor-friendly institutions incur substantial declines in their private credit sectors in the case of default. As a result, we predict that countries with strong legal investor protection suffer less from the uncertainty and instability arising from international political risk and experience less adverse impact from international political risk on government bond yields. We find support for these arguments and document that the positive association between the occurrence of international political crises and government bond yields weakens significantly for countries with high-quality and stable governments and for countries that have a common law origin and provide strong legal protection for investors and creditors.

Our research relates to a recent focus in the literature that ties political uncertainty to asset prices (Berkman et al., 2011, Pastor and Veronesi, 2012, Bekaert et al., 2012, Gao and Qi, 2013). We extend this line of research and use government bond pricing as our empirical setting to evaluate the asset-pricing implications of political risk. Although research on the link between regular political changes in host countries and government bond yields is abundant (e.g., Pantzalis et al., 2000, Stein and Streb, 2004, Moser, 2007), evidence on the role of dramatic political risk in government debt pricing is much rarer. A notable exception is Baldacci et al. (2011), who emphasize the importance of domestic political violence and expropriation in emerging market credit prices.

Our study emphasizes international political risk and is distinct from prior research in two ways. First, political changes that occur in debtor countries are frequently triggered by economic downturns. Our measure of international political risk, the total number of international political crises in a given month, is largely exogenous to debtor countries’ economic circumstances. Second, we study the impact of more dramatic international political crises (rather than normal political events) on government bond pricing in a wide range of both developed and emerging economies over an extended period, whereas prior research has only focused on a small selection of countries over a short time span. Our findings add to the evidence that global factors play a significant role in determining sovereign bond yields (Mauro et al., 2002, Martell, 2008, Remolana et al., 2007, Longstaff et al., 2011).⁶ We not only establish the significant role of international political risk but also identify country characteristics that can alter the effects of international political risk.

The paper is organized as follows. Section 2 describes the data sources and variable construction. In Section 3, we perform an empirical analysis to study the impact of international political risk on government bond yields. In Section 4, we explore the marginal effects of political and legal institutions on the role of international political risk. We conduct multiple robustness tests in Section 5 and draw conclusions in Section 6.