The pricing of sovereign risk and contagion during the European sovereign debt crisis

Abstract

The paper analyses the drivers of sovereign risk for 31 advanced and emerging economies during the European sovereign debt crisis. It shows that a deterioration in countries' fundamentals and fundamentals contagion – a sharp rise in the sensitivity of financial markets to fundamentals – are the main explanations for the rise in sovereign yield spreads and CDS spreads during the crisis, not only for euro area countries but globally. By contrast, regional spillovers and contagion have been less important, including for euro area countries. The paper also finds evidence for herding contagion – sharp, simultaneous increases in sovereign yields across countries – but this contagion has been concentrated in time and among a few markets. Finally, empirical models with economic fundamentals generally do a poor job in explaining sovereign risk in the pre-crisis period for European economies, suggesting that the market pricing of sovereign risk may not have been fully reflecting fundamentals prior to the crisis.

Introduction

The European sovereign debt crisis initially came as a surprise to most observers and policy-makers. Economic growth was generally strong, fiscal deficits limited and public debt levels were rising only modestly in most of Europe prior to the 2007–2008 global financial crisis, in particular among those euro area countries that are now engulfed most intensely in the subsequent debt crisis. This has spurred some observers and policy-makers to argue that financial markets have been overreacting and overpricing sovereign risk since the 2007–2008 crisis, and that this overreaction is due to contagion, in particular from the most affected countries, such as Greece, to other more innocent or prudent bystanders.

The question about the drivers of sovereign risk is important also from a longer-term policy perspective in order to understand how policy can react to the challenges of the sovereign debt crisis and the great global recession in the next decade. As Reinhart and Rogoff (2011, p. 3) argue: “The combination of high and climbing public debts (a rising share of which is held by major central banks) and the protracted process of private deleveraging makes it likely that the ten years from 2008 to 2017 will be aptly described as a decade of debt.”

To what extent have financial markets been overpricing sovereign risk in the euro area during the European sovereign debt? And what has been the role of contagion for sovereign risk? The paper critically examines these questions for a broad set of 31 advanced economies (AEs) and emerging market economies (EMEs) by empirically modeling the link between three measures of sovereign risk (long-term government spreads, CDS spreads and ratings of sovereigns) and economic fundamentals over the period 2000–2011. As is common in the literature, contagion is defined as the change in the way countries' own fundamentals or other factors are priced during a crisis period, i.e. a change in the reaction of financial markets either in response to observable factors, such as changes in sovereign risk among neighboring countries, or due to unobservables, such as herding behavior of market participants.

We motivate the empirical analysis for the determinants of sovereign yields through a standard definition of sovereign risk as reflecting credit risk, liquidity risk and risk appetite. Based on this conceptual framework, the first part of the analysis highlights that if one takes the relationship between fundamentals and sovereign risk during the pre-crisis period 2000–2007 as the true relationship, then sovereign risk is indeed substantially overpriced in many Europeans economies, and in particular among the euro area periphery (Greece, Ireland, Portugal, Spain and Italy – GIPSI), but not for many EMEs, especially outside Europe. However, it is striking that those fundamentals that one would expect to be the most important determinants for the price of sovereign risk – the public debt level, fiscal deficit, growth and the current account – explain very little of the pricing of risk in GIPSI countries before the crisis, but have much more explanatory power for sovereign risk in other AEs and EMEs. In fact, the most important determinant for the price of sovereign debt in GIPSI countries in the pre-crisis period was the price of public debt among other European countries, such as that of Germany. And indeed, the small spreads and very high comovements of sovereign yields within the euro area suggest that other factors than fundamentals may have been the prime determinants of sovereign debt in Europe before the crisis.¹

This finding thus suggests that country-specific fundamentals had less importance for the pricing of sovereign risk in the euro area during the pre-crisis period compared to other economies. The empirical analysis of the paper shows that the price of sovereign risk has been much more sensitive to fundamentals and that fundamentals explain a substantially higher share of the movements and cross-country differences in sovereign risk during the 2008–2011 crisis than in the pre-crisis period. Applying this counterfactual analysis for the crisis period shows that sovereign yields and CDS spreads would have been much more dispersed before 2007, in particular among euro area countries, if markets had priced fundamentals in the pre-crisis period in the same way that they did in 2008–2011. In fact, there is a negative correlation between the “mispricing” of sovereign risk – i.e. the deviation of actual market prices of risk from those implied by empirical models based on fundamentals – during the crisis and in the pre-crisis period. In other words, those countries for which sovereign risk was “underpriced” in the pre-crisis period were also those that became “overpriced” relative to economic fundamentals during the crisis.

The findings raise the question of what constitutes a “fair” pricing of sovereign risk and an over-pricing or under-pricing of such risk.² A basic intertemporal budget constraint for a government highlights the importance of expectations for determining the sensitivity of the pricing of sovereign risk – market expectations about the future primary balance, debt level, inflation, as well as about a government's willingness and ability serve debt all influence how markets price existing fundamentals that are relevant for the sustainability of public debt. As such, the empirical findings of the paper suggest that there may be multiple equilibria between the market price of sovereign risk and underlying fundamentals, which depend on existing market expectations.

What explains these disparities and shift in the pricing of sovereign debt during the 2008–2011 period? There are three different conceptual reasons for such a change. First, market participants may come to price the same fundamentals in a different way over time. While they may have ignored cross-country differences or changes in country-specific fundamentals during some periods, they may react a lot more strongly during a crisis period. This is what the literature has referred to as “wake-up call” contagion or fundamentals contagion (Goldstein, 1998; Bekaert et al., 2010). In fact, the findings indicate that for some countries, such as the GIPSI countries, there is strong evidence in favor of this “wake-up call” contagion, though for other countries there is much less of such evidence.

Second, the pricing of sovereign risk may have been affected by cross-country contagion, i.e. the transmission of a negative sovereign shock in countries such as Greece may have raised the price of sovereign risk in other, related countries. We refer to this as “regional contagion” following the argument of some that such a transmission across countries was particularly important within the euro area in 2008–2011. The third conceptual reason for changes in the pricing of sovereign risk relates to herding behavior or panic among investors. The literature refers to this type of contagion often as “pure contagion” or herding contagion. It is the most difficult type of contagion to measure empirically, as it at least partly reflects factors that are unobservable to the economic modeler. Yet it may also be the most difficult one to address for policy-makers as using firewalls, financial support and improving fundamentals may be insufficient to fully address it.

We find evidence that regional contagion has not been important during the 2008–2011 sovereign debt crisis in Europe. Interestingly, the estimates indicate that the cross-country spillovers of sovereign risk were stronger prior to the crisis than during the crisis. In other words, while financial markets tended to price sovereign risk within a region, in particular within the euro area, in a similar way, irrespective of differences across countries' fundamentals, they started to discriminate on the basis of fundamentals more strongly during the crisis. Moreover, even after accounting for “fundamentals contagion” and “regional contagion”, there is a substantial part of the increase in the price of sovereign risk in 2008–2011 that remains unexplained and that points to the importance of “pure contagion”.

To get at the role of pure contagion, we analyze the comovements of that part of the price of sovereign risk that cannot be explained by either changes in fundamentals, by fundamentals contagion or by regional contagion. Following the approach of Boyson et al. (2010), we analyze the clustering in time of large unexplained changes in the pricing of sovereign risk. We find that there is indeed some evidence of such clustering among euro area countries, but that this occurred at the height of the global financial crisis in 2008 and mostly not during 2010 and 2011 with the exception of July–September 2011 when 70% of euro area countries experienced sharp increases in the pricing of their sovereign risk. However, this period was very short, indicating that herding contagion can help explain the overall dynamics of sovereign risk to a very limited extent.

For the last part of the analysis, we try to quantify the importance of each of the three types of contagion for the pricing of sovereign risk during the 2008–2011 sovereign debt crisis. A first important finding in this regard is that most of the increase in the price of sovereign risk during the 2008–2011 sovereign debt crisis among GIPSI and other euro area countries was due to a deterioration in countries' fundamentals and fundamentals contagion. By contrast, regional contagion and spillovers were relatively unimportant overall while also pure contagion played a small, but limited role. In fact, we find strong evidence for a decoupling among European sovereign debt markets during the crisis, with changes in one country's sovereign debt being transmitted to neighboring countries much less intensely during the crisis than compared to before the crisis. Overall, therefore, the findings suggest that the deterioration of fundamentals and fundamentals contagion are the prime explanations for the sharp rise in sovereign risk during the European sovereign debt crisis.

The paper is organized as follows. Section 2 discusses the related academic literature on modeling the pricing of sovereign risk. Section 3 describes the methodology to measure the impact of fundamentals, regional risk and contagion on the price of sovereign risk, while Section 4 describes the data and presents a number of stylized facts about the evolution of sovereign risk during the crisis. Section 5 outlines the main empirical results and various extensions. Finally, Section 6 summarizes the findings and discusses implications for the policy discussion.