بحران بدهی های مستقل منطقه یورو : شناسایی گریز به نقدینگی و مکانیزم های سرریز

Looking at the daily period between January 2006 and December 2012, besides the traditional credit and liquidity risks, which explain the developments of sovereign yields relative to the Bund for Greece, Ireland, Portugal, Spain and Italy, two additional factors have played a key role in the developments of euro area sovereign yield spreads: flight to liquidity benefiting the German Bund and the spillover effect from Greece. The flight to liquidity premium, which is estimated by constructing the spread between a German state guaranteed agency bond and the Bund, is behind the pricing of all euro area spreads and, specifically, is the only factor explaining the sovereign spreads for Finland and the Netherlands. The spillover effect from Greece, which is identified using complementary approaches, has contributed to developments in spreads of countries with weaker fiscal fundamentals, a lower degree of competitiveness and a higher need of foreign financing. However, a large fraction of the spillover across countries remains unexplained.

With the intensification of the financial crisis in September 2008 after the collapse of Lehman, government bond yields relative to the German Bund have been rising after ten years of stability at very low levels (see Fig. 1). The situation started to improve in the course of the spring 2009 as global uncertainty receded. On 16 October 2009, the Greek Prime Minister George Papandreou in his first parliamentary speech disclosed the country's severe fiscal problems and immediately after on 5 November 2009 the Greek government revealed a revised budget deficit of 12.7% of GDP for 2009, which was double of the previous estimate. Since then, the sovereign spreads rose sharply for most of the euro area countries.The developments in 2010 and 2011 were remarkable with the Greek, Irish and Portuguese 10-year sovereign spreads hitting in July 2011 respectively 1600, 1200 and 1100 basis points. Similarly, Spanish and Italian spreads reached 400 basis points, Belgium hit 200 basis points and France hit 90 basis points. The market players have singled out the debt sustainability issue and have argued that the solvency risk for these countries deteriorated in the course of 2010 and 2011. After July 2011, the Irish spreads started a steady decline, while Austrian, Belgian, French, Greek, Italian, Portuguese and Spanish spreads recorded in the course of the year new high levels. The tension is further illustrated by the reaction in more highly rated sovereign papers. Benchmark French, Dutch, Austrian, Finnish and German yields have turned round to fluctuate with a declining trend over the 2008–2012 period. However, the spreads vis-à-vis the German Bund have been rising also on bonds issued by countries with solid fiscal fundamentals, such as Austria, Finland and the Netherlands. Given such developments in spreads, this study aims at answering the following four main questions: Why did the spreads of countries with solid fiscal fundamentals rise with the financial crisis? Why did the spreads of countries with weak fiscal fundamentals escalate with the financial crisis? Is there evidence of spillover effects from Greece, Ireland and Portugal on other euro area countries' spreads? What are the mechanisms that transmit financial shocks across economies? The underlying common question is identifying the factors that can explain such a large variety in sovereign spreads' developments over time and across euro area countries. An important issue of any paper studying the determinants of bond yields is the identification of the variables to be used as a proxy. The key contribution of this paper is the identification of the flight to liquidity premium benefiting the German Bund and the identification of the spillover effect. We identify a euro area common risk factor, which we argue captures the portfolio shift due to a higher appetite for the German Bund. The 10-year KfW (‘Kreditanstalt für Wiederaufbau’) bond and the German sovereign bond (i.e. Bund) are both guaranteed by the German government and, therefore, carry the same default risk (Ejsing et al., 2012, Longstaff, 2004, Monfort and Renne, 2011 and Schwarz, 2010). Any differences between agency and government bond yields should reflect international investors' preference for assets with the lowest liquidity risk. Beber et al. (2009) make explicit distinction credit risk measured by CDS and liquidity risk measured among others by bid–ask spreads, and find relative importance of liquidity over credit quality rising during times of heightened market uncertainty. We show that, after controlling for bond-specific liquidity risk (proxied by bid–ask spreads and/or volumes) and credit risk, flight to liquidity benefiting the German Bund measured by the KfW-Bund spread is an important additional reason explaining developments in spreads for all euro area countries, and in particular for Finland and the Netherlands. With regard to the spillover analysis, the literature has investigated the issue in depth adopting different definitions, but also a variety of econometric approaches. Forbes (2012); (ii) the analysis of cross-country market correlations; (iii) VAR analysis by means of impulse response functions; (iv) latent factors and GARCH models to exploit the fact that the variance can differ across regimes; (v) extreme value analysis to test whether tail observations in returns are correlated across countries. Each of these approaches for measuring contagion and spillover has its advantages and disadvantages and Forbes (2012) discussed them in depth. In this study, we investigate the spillover effect employing three complementary approaches.1 First, following Gande and Parsley (2005), we assess whether a credit event (explicit credit ratings as well as credit outlook and credit watch—as reported by Standard and Poor's (S&P), Moody's Investors Service (Moody's) and Fitch) in Greece, Ireland and Portugal has a significant effect on sovereign credit spreads of other countries. Second, we make use of impulse response functions to assess the impact of unexpected changes in credit events. Third, the spillover effect is identified by studying the impact of key economic news, such as the downgrade of the sovereign debt in Greece and Portugal to non-investment grade and the announcement of the magnitude of financial assistance to Greece, Ireland and Portugal, controlling carefully for contemporaneous credit events in all countries and for other news articles and press releases related to the financial assistance program such as those released by the Troika (European Commission, IMF and ECB) at the end of their mission in the country, the European Commission, the European Council and the Eurogroup or those related to the private sector involvement.2 Part of the analysis is carried out using credit rating reviews, because they have brought the contagion risk to the fore. For example, when Moody's downgraded the long-term government bond ratings of Portugal by four notches on 5 July 2011 from Baa1 to Ba2 and assigned a negative outlook, it argued that a voluntary rollover of Greek debt would imply a rising risk that private sector participation could become a precondition for additional rounds of official lending to Portugal as well.3 The debate about the voluntary debt rollover for Greece has contributed to a major re-assessment of the Portuguese outlook and triggered a significant increase in yield spreads for Portugal. In other words, Portugal was not primarily judged on the basis of its own fundamentals. We find clear evidence of spillover effects, particularly from Greece. Regardless of whether using S&P, Moody's or Fitch credit ratings, the empirical analysis indicates that rating events concerning Greek sovereign bonds lead to strong increases of sovereign yields in Ireland, Portugal, Italy, Spain, Belgium, France and Austria. The spillover effect from Ireland is also estimated to be sizeable, although the effect from Greece is larger and predominant. Moreover, by using impulse response functions, we can show that such effect only partly abated after the summer of 2011. The spillover effects identified by studying the impact of key economic news indicate that news associated to the financial support granted to Greece generate larger spillover effects to sovereign spreads of other countries and that the Greece's and Portuguese's downgrade to non-investment grade generated an increase in sovereign spreads in all euro area countries. What are the channels which can help explain the spillover across countries: the weakness of fiscal fundamentals and/or the lower degree of competitiveness, and/or the higher need of foreign financing? We can show that the larger the adverse spillover effect from Greece, the weaker the fiscal situation of the country, the less competitive or the higher the need for foreign financing. This result is obtained estimating the spillover effect from Greece on the entire yield curve of each country and running a cross-sectional study of the estimated spillover on the fiscal situation, the current account balance and the degree of competitiveness existing across countries in 2009, before the euro area sovereign debt crisis started. The mechanism we have disclosed is closely related to the “wake-up calls”; namely, when additional information or a reappraisal of one country's fundamentals leads to a reassessment of the risks in other countries (Goldstein, 1998). Only after the crisis erupted, countries perceived to have weaker economic fundamentals tended to be more vulnerable to spillover. However, a large fraction of the spillover across countries remains unexplained. From an asset-pricing perspective, of course, changes in default probabilities have an impact on the equilibrium required premium for bearing credit risk. Therefore, the econometric analysis is carried out controlling for the government budget balance (backward looking component of sovereign solvency risk)4 and credit rating reviews,5 which is expected to be more forward looking given that rating reviews are carried out based on expected economic developments; but also for credit default swaps (CDS), as a robustness check. We find that also the deterioration in credit risk plays a key role in the crisis. The remaining sections of the paper are structured as follows: Section 2 presents the underlying econometric specifications. Section 3 discusses the main risk factors. Section 3.4 estimates the model and investigates the spillover mechanisms. Section 4.5 discusses the robustness checks. Section 6 concludes.

After the disclosure of Greece's severe fiscal problems at the end of 2009, the sovereign spreads rose sharply for most of the euro area countries, causing the biggest challenge for the European monetary union since its creation. The factors affecting the sovereign bond yields are associated to aggregate risk, country-specific risk, spillover risk and contagion risk. The aggregate risk is driven by changes in monetary policy, global uncertainty and risk aversion, while the country-specific risk is related to changes in default probabilities on the sovereign debt, the ability to raise funds in the primary market and liquidity factors in the secondary market. Separating liquidity risk, spillover risk and contagion risk from aggregate risk and credit risk is very important from a policy making perspective, because an intervention by the central bank can be successful if financial markets face liquidity problems or subject to spillover effects or contagion. If, on the contrary, the rise in spreads is due to aggregate risk or sovereign risk then a central bank has only little room for maneuver. The euro area sovereign debt crisis brought forward in the debate the importance of the sovereign solvency risk, flight to liquidity and the transmission mechanism of the spillover effect. Looking at the daily period between January 2006 and December 2012, we can show that traditional country-specific credit and liquidity risks can explain the developments of sovereign yields relative to the Bund for Greece, Ireland, Portugal, Spain and Italy. However, two additional factors have played a key role in the developments of euro area sovereign yield spreads: flight to liquidity benefiting the Bund and the spillover effect from Greece. Flight to liquidity, which is estimated by constructing the spread between a German state guaranteed agency bond and the Bund, is behind the pricing of all euro area spreads and, specifically, is the only factor explaining the sovereign spreads for Finland and the Netherlands. The spillover effect from Greece, which is identified using various approaches (i.e. rating downgrade in Greece and the financial support to Greece), has also contributed to developments in sovereign spreads. The transmission mechanism we have disclosed is closely related to the “wake-up calls”. Only after the crisis erupted, countries perceived to have weaker economic fundamentals, and in particular weaker fiscal conditions, a lower degree of competitiveness and a higher need of foreign financing, tended to be more vulnerable to spillover; namely, Ireland, Portugal, Italy, Spain, but also Belgium, France and Austria. However, a large fraction of the spillover effect across countries remains unexplained. From a policy-making point of view, to safeguard the stability of the euro area financial system, the highest priority is to reduce the sovereign solvency risk and tackle contagion. Will policy-makers close the large gap between spending and revenues? Will they cut spending and/or raise taxes enough to make the debt sustainable? The challenge that the policy-makers face is to reduce inefficient spending and avoid a deflationary spell, through structural reforms. There are two research avenues that could be undertaken to further deepen the analysis: 1) assess whether there are regime switching in the determination of sovereign yield spreads, which might have been particularly relevant in the summers of 2011 and 2012; 2) quantify the redenomination risk that, according to the ECB president Mr. Draghi, is one of the key factors behind the pricing of sovereign spreads particularly of Italian and Spanish sovereigns in 2012.