Although the pandemic was an exogenous shock leading to real damages and a surge of sovereign debt everywhere (Baldwin and Weder di Mauro 2020, Benassy-Quéré et al. 2020), sovereign risk premia significantly diverged in the euro area in Spring 2020. There was a risk of “another euro crisis”, as Olivier Blanchard put it bluntly early into the pandemic, referring to the massive sell-off of peripheral bonds in 2011-2012 (Blanchard 2020). However, after a sharp increase at the beginning of the pandemic, peripheral spreads have been on a downward trend, with the result that 2020 levels are dwarfed by those of 2011-2012 (See Figure 1). First, the difference between the two episodes may be due to significant reforms over the last decade (Baldwin and Giavazzi 2015). Second, there were swift policy reactions in 2020 compared to the previous episode, by the ECB and by euro area leaders, all of whom announced major monetary and financial assistance packages as early as March and April 2020. What drove the euro area sovereign risk premium during the pandemic crisis? What explains heterogeneity across members? Did the stabilisation of spreads reflect investor confidence in the capacity of the euro area institutions to sustain the euro?
Figure 1 Sovereign bond spreads in the euro area
Panel a 2020
Panel b 2010-2020
Notes: This figure shows the 10 year bond yield spread against German bond during the pandemic crisis (panel a) and since 2010 (panel b). In panel b, we exclude Greece which spread level dwarfs the others. We observe that the increase in bond spreads during the pandemic are limited compared to 2010.
In Delatte and Guillaume (2020), we explore the dynamics of euro area members’ sovereign spreads between 2 January and 25 May 2020 by testing potential drivers of sovereign risk premia. On the one hand, the Covid-19 pandemic hit members very differently, from 0.5 to 81.1 deaths per 100,000 inhabitants. Countries were differently equipped to deal with this shock, with health expenditures as a percentage of GDP varying from 6.2% to 11.5% throughout the region. In order to mitigate casualty, countries have implemented heterogeneous lockdown policies as measured by the Stringency Index computed by Hale et al. (2020), whose period average by member varies between 34.6 to 55.2 (on a scale from 0 to 100). Public debt as a percentage of GDP varies between 8.4% and 179.2% implying potential amplification dynamics for largely indebted countries because the higher the burden of the debt of a country, the higher investors price the risk of repudiation (Calvo 1988). Countries have implemented different levels of fiscal stimulus to mitigate macroeconomic consequences of isolation (between 0.7% and 7.2% of GDP). Finally, the non-performing loans ratio in the domestic banking sectors, varies between 0.6% and 33.4% in the euro area as a whole.1
On the other hand, compared to 2011-2012 episode, members have benefited from swift and sizeable common policy responses. On the monetary side, the ECB announced the Pandemic Emergency Purchase Program (PEPP) on 18 March, a €750 billion program of private and public securities purchases with flexible capital keys as part of the Asset Purchase Program, under which €220 billion of securities were purchased within the first two months. In addition, the ECB eased further collateral eligibility rules to enable banks to mobilise more collateral. On the financial assistance side, a total of €580 billion of loans and €40 billion of budget line have been mobilised and discussions of an additional €750 billion are ongoing at the time of writing this column. In the meanwhile, the European Commission lifted the budgetary rules of the Stability and Growth Pact on 20 March 2020.
Figure 2 plots the contribution of different factors and events to the sovereign bond spreads of 14 euro area members in basis points. We plot the contributions separately for largely indebted countries (Italy, Spain, Portugal, and Greece) from the rest of the sample. The figure reads easily: for example, the 5 May ruling of the German Court (D0505) widened the spreads by +7.5 basis points on average.
Figure 2 Sovereign bond spreads in the euro area
Notes: This figure plots the average contribution by factor based on standardized coefficients estimated in Delatte and Guillaume (2020). The largely indebted countries include Greece, Italy, Spain, Portugal. Here we plot only significant coefficients.
Our results suggest that stock market volatility (RVol) has contributed to an increase in sovereign spreads overall, with highly indebted countries suffering more than the rest of the sample. (+7.1 versus +1.4 basis points); the first ECB announcement on 12 March (D0312) widened the spreads by 7.9 basis points. In turn, the 18 March PEEP announcement (D0318) was the most powerful to reduce spreads (-18.1 basis points), with even larger contribution to largely indebted countries (-53.4 basis points). Both securities purchase programs had an effective but more modest contribution on the spreads of largely indebted countries (-2.1 basis points for PSPP and -1.7 basis points for PEPP) and no significant contribution on the rest of the sample. The 9 April European Council coordination (D0409) made a large contribution on largely indebted countries (-25.2 basis points) and milder on the rest of the sample ( -6.1 basis points). However we find that financial assistance programs based on loans (LoansEU) have contributed to widen the spreads of largely indebted countries (+36 basis points), i.e. wiping out the positive 18 March effect. We come back to this finding below. The 18 May Franco-German proposal (D0518) contributed to reduce the spreads of largely indebted countries by -5.9 basis points. We find that a difference of one standard deviation of healthcare expenditures across countries increases the spread by 21.2 basis points; similarly a difference of one standard deviation in the level of non-performing loans and of public debt increases the spreads by +25.4 and +17.8 basis points, respectively. Lastly, Figure 3 breaks down the contribution of each factor for Italy, Spain, Portugal, and Greece, and suggests that the Italian spread is the one that has benefited most from the interventions of the European institutions with a total reduction of 99 basis points, i.e. an almost complete compensation for their unfavourable initial condition.
Figure 3 Sovereign bond spreads in the euro area
Notes: This figure plots the average contribution by factor based on the standardised coefficients estimated in Delatte and Guillaume (2020). Here we plot only significant coefficients. Red squares represent the mean of the spreads over January 2-May 25 period and white diamonds represent the value of the fixed effect.
Three lessons can be drawn from our findings.
1. Financial assistance in the form of a loan does not work as an adjustment mechanism as do fiscal transfers. Indeed, we find that loan-based financial assistance programs have contributed to widening the spreads of heavily indebted countries. Fiscal transfers play an important role in most monetary unions in offsetting region-specific demand shocks (Kennen 1969). However, loans add debt on already largely indebted countries and thereby contribute to increase future government budget constraints. As a result, loan-based financial assistance increases the risk premium of the countries most in need of these transfers. The recovery plan based mainly on fiscal transfers proposed by the European Commission after a Franco-German initiative would most likely be a positive outcome if adopted.
2. Monetary policy speech is stronger than deeds during a sovereign debt crisis episode. Indeed, our findings suggest that the main spreads movers have been central bank speeches while securities purchase programs had a limited contribution. A likely mechanism suggested by Philip Lane is that heightened risk aversion triggered a global flight-to-safety episode implying large sell-off of largely indebted government bonds.2 By committing to purchase debt securities, the ECB ‘crowded in’ other investors by securing liquidity to all euro area governments, thereby protecting members of the euro area from a beliefs-driven self-fulfilling switch in equilibria. Our results suggest that the first ECB announcement on 12 March failed to convince investors, unlike the second announcement on 18 March which did shift investors’ beliefs. The 18 March speech associated with a limited contribution of actual securities purchase recalls the OMT-whatever-it-takes speech effect in 2012 which was enough to bring the spreads down without activating the purchase program (Altavilla et al. 2016, Delatte et al. 2017). Comparing both episodes suggests that the lender of last resort effect of speeches can work with different initial conditions and different modus operandi.
3. Exogenous shocks affect the sovereign risk premia of euro area members differently, making some euro area members more vulnerable to shocks. Countries that lose their safe asset status experience higher borrowing costs and find it more difficult to manage the crisis in the short run. Moreover, we find that differences in health capacity led to differences in individual sovereign risk premia during the pandemic. In other words, part of each country’s resilience to a crisis depends on national public policies specific to the nature of the shock. In the long run, it implies a vicious circle for countries with a ‘wobbly’ safe asset status because their spending capacity to build a strong public policy is more constrained than others. It is a serious challenge to strengthen resilience to shocks if the frequency of extreme events is to increase in the future. We show how central bank activism has contributed to reducing the cost of borrowing for the most affected countries. However, this solution is sustainable as long as the stabilisation of sovereign risk premia does not contradict the objectives of stabilising real prices. It is important to think about alternatives for building a safe common asset in the zone in order to mitigate ‘flight-to-safety’ episodes (Brunermeier et al. 2016). The current proposal to mutualise debt financed by the European Commission’s bond issue, which goes in this direction, would therefore be a positive outcome if adopted.
Aizenman, J, M Binici and M Hutchison (2013), “Credit ratings and the pricing of sovereign debt during the euro crisis”, Oxford Review of Economic Policy 29 (3): 582–609.
Altavilla, C, D Giannone and M Lenzaa (2016), “The Financial and Macroeconomic Effects of the OMT Announcements”, International Journal of Central Banking.
Baldwin, R E and F Giavazzi (2015), The Eurozone crisis: A consensus view of the causes and a few possible remedies, a VoxEU.org eBook, London: CEPR Press.
Baldwin, R and B Weder di Mauro (2020), Mitigating the COVID Economic Crisis: Act Fast and Do Whatever It Takes, a VoxEU.org eBook, London: CEPR Press.
Bénassy-Quéré, A, R Marimon, J Pisani-Ferry, L Reichlin, D Schoenmaker and B Weder di Mauro (2020), “COVID-19: Europe needs a catastrophe relief plan”, VoxEU.org, 11 March.
Blanchard, O (2020), “Italy, the ECB, and the need to avoid another euro crisis”, in R Baldwin and B Weder di Mauro (eds.), Mitigating the COVID Economic Crisis: Act Fast and Do Whatever It Takes, a VoxEU.org eBook, 18 March.
Brunnermeier, M K, L Garicano, P R Lane, M Pagano, R Reis, T Santos and D Vayanos (2016), “The sovereign-bank diabolic loop and ESBies”, American Economic Review 106(5): 508-12.
Delatte, A-L, J Fouquau and R Portes (2017), “Regime-dependent sovereign risk pricing during the euro crisis”, Review of Finance 21(1): 363–385.
Delatte, A L and A Guillaume (2020), “Covid 19: a new challenge for the EMU”, CEPR Discussion Paper 14848.
Hale, T, A Petherick, T Phillips and S Webster (2020), “Variation in government responses to COVID-19”, Blavatnik School of Government Working Paper 31.
Kenen, P (1969), “The theory of optimum currency areas: an eclectic view”, Monetary problems of the international economy 45(3): 41–60
1 Graphs and maps illustrating cross-country heterogeneity as well as a detailed timeline of the crisis can be found on the companion website of the paper.
2 See “The market stabilisation role of the pandemic emergency purchase programme” by Philip R. Lane on 22 June, 2020.