The Eurozone crisis: A consensus view of the causes and a few possible solutions
This chapter introduces the eBook "The Eurozone Crisis: A Consensus View of the Causes" and extracts a consensus narrative of the EZ Crisis.
Search the site
The Eurozone crisis which broke out in May 2010 is a long way from finished. Charles Wyplosz puts it bluntly in his chapter – “Five years later, growth is miserable and is forecasted to remain miserable as far as the forecasters’ can see. Governance is in disarray as the tragic Summit of July 13 – the last of an incredible series of official meetings – showed”. Worse yet, there is widespread belief that the fragilities and imbalances that primed the monetary union for this crisis are still present.
As a first step to finding a broad consensus on what more needs to be done, this ebook gathers the views of 18 world-renown economists on a simple question – what caused the Eurozone Crisis? The focus was to be on thinking about the causes as a prelude to developing remedies.
Although the essays were largely uncoordinated – and the authors hark from diverse backgrounds – a remarkably coherent message emerges from this collection. In this introduction, we have two goals:
In the world of policymaking, narratives are incredibly important since if we cannot agree on what happened – or more precisely, on what were the most important things that happened – then we cannot agree on how to remedy the situation.
Again, we cannot know what needs fixing until we agree broadly on what was broken.
While the eBook’s focus is on the causes of the crisis, some of the authors also discussed possible remedies. These ideas will lead us in the next step of this project: an ebook on ways forward for the Eurozone.
The core reality behind virtual every crisis is the rapid unwinding of economic imbalances. The size and duration of the crisis typically depends upon the size of the initial imbalances, how the initial shock gets magnified by a variety of ‘amplifiers’, and how rapidly and effectively policy responds.
In the case of the Eurozone crisis, the imbalances were extremely unoriginal. They were the standard culprits that have been responsible for economic crises since time immemorial – namely, too much public and private debt borrowed from abroad. Too much, that is to say, in relation to the productive investment financed through the borrowing.
From the euro’s launch and up until the crisis, there were big capital flows from Eurozone core nations like Germany, France, and the Netherland to Eurozone periphery nations like Ireland, Portugal, Spain and Greece. A major slice of these were invested in non-traded sectors – housing and government services/consumption. This meant assets were not being created to help pay off in the investment. It also tended to drive up wages and costs in a way that harmed the competitiveness of the receivers’ export earnings, thus encouraging further worsening of their current accounts.
When the Eurozone crisis began – triggered ultimately by the Global Crisis – cross-border capital inflows stopped. This ‘sudden stop’ in investment financing raised concerns about the viability of banks and, in the case of Greece, even governments themselves. The close links between Eurozone banks and national governments provided the multiplier that made the crisis systemic.
Importantly, the Eurozone crisis should not be thought of as a sovereign debt crisis. The nations that ended up with bailouts were not those with the highest debt-to-GDP ratios. Belgium and Italy sailed into the crisis with public debts of about 100% of GDP and yet did not end up with IMF programmes, while Ireland and Spain, with ratios of just 40%, (admittedly kept artificially low by large tax revenues associated with the real estate bubble) needed bailouts. The key was foreign borrowing. Many of the nations that ran current account deficits – and thus were relying of foreign lending – suffered; none of those running current account surpluses were hit.
The initial shock – the rapid loss of Eurozone investors’ trust in the deficit nations– was amplified in several ways. Given the Eurozone design, governments who got in trouble had no lender of last resort. Which meant their euro denominated borrowing was akin to foreign currency debt in traditional sudden stop crises. The natural lender of last resort, the ECB, was explicitly forbidden from playing the role. This ruled out one of the classic ways out of avoiding government default – having the central bank print the money needed to service the debt.
The predominance of bank financing was another amplifier of problems. European banks were thinly capitalised and extremely large relative to the countries’ GDP. They were so large that they had to be saved, but their size also created a ‘double drowning’ scenario. This is exactly what happened in Ireland. In what might be called a tragic double-drowning scenario, Ireland’s banking system went down first, and the government of Ireland went down trying to save it. Spain and Belgium flirted with, but ultimately avoided the same fate.
A third amplifier is the so-called doom-loop – the potential for a vicious feedback cycle between banks and their government. Fear about the solvency of a nation’s government fans fears about the solvency the nation’s banks, which in turn weakens the economy, thus worsening the sustainability outlook for the nation. A deadly helix of rising risk premiums and deteriorating budget deficits can suck nations into a debt default vortex. This happened to Portugal and came close to happening to Italy, Spain and Belgium. Even France and Austria flirted with the shadow of doom-loops. The key element in this mechanism is that Eurozone banks are heavily invested in the debt of their own government.
The final amplifier was the rigidity of factor and product markets in nations that could not restore competitiveness via a currency devaluation. Indeed, five years down the road, few of the Eurozone nations have recovered their pre-crisis growth or employment rates, although Spain surprises observers with the strength of its recovery.
The third determinant of crisis severity – appropriateness of the policy response – was clearly a big problem in the Eurozone. Nothing in the Eurozone institutional infrastructure was prepared for a crisis on this scale. The possibility had simply not been considered when setting up the euro’s architecture. Many mistakes were made. Indeed, judging from market reactions, each policy intervention made things worse. The corner was only turned in the summer of 2012 with the ‘whatever it takes’ assertion by ECB President Mario Draghi.
The next section presents what we believe is a consensus narrative of the Eurozone crisis. It has not been approved by all the contributors to this ebook, but we believe it is in line with the essence of their analyses.
The subsequent section gathers what we view as a consensus on the causes of the crisis – both the proximate causes and, if you will, the ‘causes of the causes’.
While the ebook’s focus is on the causes of the crisis, some of the authors presented remedies and the next to last section summarises the most specific of these.
The final section presents a summary and our concluding remarks.
The first step to repairing the Eurozone is to answer the questions, ‘what is broken’ and ‘what broke it’? This section presents what we believe is a consensus view on the main elements of the crisis’s evolution as a prelude to answering the two questions.
Before detailing the damage, ‘it is worth recalling that the Eurozone did not fare badly in the first years of the Global Crisis”, as Giancarlo Corsetti writes in his chapter. “Participating in the European monetary union appeared to shelter countries from the early difficulties experienced by countries with a large financial sector relative to their tax base, such as the UK and Switzerland”. That was not to last.
The 1990s was a time of exchange rate turbulence in Europe with major crises in 1992 and 1993. In the face of large differences in inflation rates and the occasional devaluations, markets demanded very large risk premiums. Nations like Italy, Spain and Portugal paid far higher rates on their debt than Germany and other countries such as the Netherlands, Austria, France, and Belgium. That all changed with the move towards monetary union.
Starting from 1995, Eurozone interest rates converged in anticipation of the single currency.
As Figure 1 shows, government bond yields for all prospective Eurozone members converged to the German rate which itself fell substantially and remained low until the start of the financial crisis. As Feld et al. write in their chapter, “Despite substantial differences in macroeconomic and fiscal policies, member countries were able to access financial markets at almost identical yields between 2001 and 2007”.
However, the convergence was, as Jeff Frankel writes in has chapter, “viewed as a good thing rather than a bad thing.” A sign of real convergence between North and South Eurozone countries.
Figure 1. Risk premiums disappeared in the run up to the crisis.
Source: OECD online database with authors’ elaboration.
The magnitude of the changes were astounding. Italy saw the nominal cost of borrowing fall from 13% to 3% in a decade. The cheapening of credit was not quite as stark for other Eurozone members, but even Germany’s government bond yields dropped from 7% to 3%. Indeed all around the world, the cost of capital fell steadily in what Ben Bernanke has been called a ‘global savings glut’.
This had consequences:
Each type of borrowing played a role in setting up the pre-crisis imbalances.
The Eurozone crisis was not, at its roots, a sovereign debt crisis. The culprit was the large intra-Eurozone capital flows that emerged before the crisis.
Daniel Gros puts it succinctly, “The euro crisis started as a classic ‘sudden stop’ to cross border capital inflows. As boom turned into bust, government lost their tax base and had to assume private debt, thus creating a public debt crisis. The highly leveraged banking system of the Eurozone, tightly linked to national governments, provided a multiplier, which made the crisis systemic”.
The problem, as Paul de Grauwe notes in his chapter was that “the European monetary union lacked a mechanism that could stop divergent economic developments… which were crystallized in the fact that some countries built up external deficits and other external surpluses”.
These imbalances baked problems into the Eurozone’s ‘cake’ since, as Guido Tabellini writes in his chapter, “if a sudden stop occurs, the sovereign most likely will lack the fiscal resources to cope with it. The size of the financial sector has grown just too large”.
Recalling that a nation’s current account is its net borrowing from abroad, large increases in foreign indebtedness shows up as a negative current account. A positive current account indicates that the nation is, on net, lending to foreigner nations.
To interpret the individual current accounts, we must depart from an essential fact:
Thus there was very little net lending from the rest of the world to Eurozone countries. Unlike in the US and UK, the global savings glut was not the main source of foreign borrowing – it was lending and borrowing among members of the Eurozone. For example, Germany’s large current account surpluses and the crisis countries deficits mean that German investors were, on net, lending to the crisis-hit nations – Greece, Ireland, Portugal and Spain.
Nevertheless, as Philip Lane points out, even if the savings wasn’t coming from China, the atmosphere mattered. “The evidence indicates that the 2003-2007 period can be characterized as a ‘credit supply’ shock,” he writes. “The global financial system was more willing to tolerate large net debt flows to these advanced economies”. He points to the deep causes as: “low policy rates adopted by advanced countries’ central banks, financial innovations (such as new types of securitisation) and shifting beliefs about risk levels and risk absorption capacity combined to foster an extraordinary boom in international capital flows“.
Figure 2. Current accounts: The core lent to the GIIPS from 2000 to 2007.
Source: WEO online database with authors’ elaboration. Note: Current account deficit (-) and surplus (+).
With these points in mind, Figure 2 shows the evolution of the current account surplus for the turnkey nations. The charts above shows the figures for the pre-crisis years as a share of each nation’s own GDP. This gives an idea of how important the flows were to the macroeconomy of each nation. As the top left panel of Figure 2 shows:
Italy, which almost got in trouble, was also increasingly relying, though to a much lesser extent than, say Spain, on foreign lenders during the decade preceding the crisis.
The top right panel shows that:
France is the exception that tests the rule. It started out with a positive current account but saw its position deteriorate in the late 2000s. Unlike the other core nations, France started to have problems in the height of the crisis (its debt rating was cut in 2012).
These balance-of-payment figures as a share of GDP illustrate the importance of foreign capital flows from the perspective of the individual nations. But given the gigantic size differences among Eurozone members, the ratios hide important information. The second row of charts shows the current account figures in billions of dollars, not as a proportion of national GDP. Here we see that there were really two outliers:
The figure for the next biggest net lending, Netherlands, was $50 billion or less.
No other nation was even close in terms on the lending or borrowing sides.
The current account is, by construction, the difference between the amount a nation invests and saves. Looking at savings and investment thus provides hints as to the drivers of current account imbalances.
As Figure 3 makes it clear:
The Dutch, Belgian and (at the end) German cases stand out in particular. Italy, Spain and Portugal, by contrast, stand out on the low side.
Spain and Ireland are the outliers here; they invested far more than the Eurozone average during the pre-crisis years.
At the time, this basic pattern was not viewed as a source of Eurozone problems, but rather as a badge of success. It was widely believed that these sorts of private capital flows were part of the natural real convergence within a monetary union. The poorer nations, which had abundant investment opportunities, were attracting investors from richer nations were capital faced diminishing returns. Or at least that was the contemporaneous thinking (see for instance Blanchard and Giavazzi in The Brookings Papers on Economic Activity in 2002).
A big problem was that much of the investment headed towards non-traded sectors like government consumption and housing.
Figure 3. Core and GIIPS savings and investment patterns
Source: WEO online database with authors’ elaboration. Note: Current account deficit (-) and surplus (+).
As Agnès Bénassy-Quéré put it, “capital flows tended to feed non-tradable sectors in the periphery of the Eurozone. In receiving countries, the increase in liabilities was not sustainable since it did not correspond to the building up of export capacities. Worse, if contributed to house price bubbles that would inevitably burst at some point”.
The inflows also tended to drive up wages and costs that resulted in competitiveness losses that validated the current account deficits. All four nations that eventually signed bailout packages – Greece, Ireland, Portugal and Spain – had inflation well in excess of the average.
As Feld et al. write, these nations “suffered a considerable loss in price competitiveness during the debt-financed booms, due to major wage increases and high inflation. Consequently, domestic export companies were put at a competitive disadvantage and lost trade shares. The loss of price competitiveness combined with the debt-financed increase in domestic demand and the associated imports resulted in high current account deficits”.
By contrast, all the core nations except Netherlands and Luxembourg had inflation below the norm, especially Germany.
It was not supposed to work this way. In their chapter, Thorsten Beck and José-Luis Peydró note, “in fact, a key argument in favour of peripheral countries to adopt the euro was that the only way to go out of a crisis would be with a more flexible, competitive economy through structural reforms. Sadly, the substantial lower risk premiums… that came with the euro implied strong booms but too little economic reforms”.
In her chapter, Beatrice Weder di Mauro crystallises the thinking of all authors in writing – “there can hardly be any disagreement on the diagnosis that there was too much debt accumulation over the course of the first eight years of the existence of the euro”.
Figure 4 shows that the evolution of public debt does not line up well with the nations that subsequently got in trouble (Greece, Ireland, Portugal and Spain, and later Cyprus). The Eurozone as a whole lowered its debt-to-GDP ratio from 72% in 1999 to 66% in 2007. Of course, some nations did see higher public debt ratios but others fell substantially.
These two were paragons of fiscal rectitude, dramatically lowering their public debt burden to far, far below the Maastricht limit of 60%. In 2007, Ireland’s and Spain’s ratios were, respectively, 24% and 36% of GDP. It should be noted, however, that both countries’ government revenues were kept artificially high by tax revenues associated with a real estate boom.
Belgium, Netherlands and Finland likewise slashed their public debt.
Figure 4. Government debt improved for most Eurozone nations (1999 = 100).
Source: IMF WEO online database with authors’ elaboration.
Public debt did become an issue for the other two crisis-hit nations.
Greece’s burden reached 103% but Portugal went into the crisis with a modest debt ratio of 68%. But this was not a North versus South development.
This was despite the sharp decline in the budgetary burden of interest payments that came with lower borrowing costs compared to the 1990s.
Private debt accumulation – which became a huge issue during the crisis – was run up during the heydays of the Eurozone’s first decade in some EZ members.
Figure 5 shows the evolution. Of course, this was not a Eurozone specific issue. It occurred in the US, the UK and other OECD nations as well.
Figure 5. Rapid accumulation of bank debt was a problem
Source: OECD online database with authors’ elaboration.
Much of the bank lending in Spain and Ireland was going to the housing sector. As Figure 6 shows, prices were rising steadily. The prices in Germany, by contrast, were falling thus creating what might have been a very temping opportunity for German banks flush with more loanable funds than local investment opportunities.
Figure 6. House prices rose in the GIIPS and fell in Germany, 2000-2007.
Source: OECD. House price indices in real terms. Index value in first quarter of 1999 equals 100. Adapted from Baldwin, Gros and Laeven (2010).
We can zoom in more precisely for one particularly important form of cross-border private lending/borrowing – that of banks. Table 1 shows that banks from the ‘core’ (Germany, France, Austria, Belgium and the Netherlands) bought very large amounts of debt from the nations that would eventually get in trouble.
Table 1. Total lending from core countries’ banks into the periphery (billion euros)
Note: Eurozone core is Germany, France, Austria, Belgium and Netherlands.
Source: Adapted from Baldwin et al. (2010), which draws on BIS Consolidated Banking Statistics, June 2010.
This inter-linkage among core-nation banks and periphery-nations came to be a critical piece of the puzzle as the crisis unfolded.
In particular, it meant that the obvious solution of writing down Greek debt would have forced the problem back onto the core-nations leading the bailout. Solving Greece’s problem in the time-honoured way might well have created classic bank crises in France and Germany (much as the eventual Greek write-down sparked a banking crisis in Cyprus).
Importantly, nothing about this was unique to the Eurozone’s debt accumulation.
The sum of public and private indebtedness skyrocketed in nations ranging from the US and Japan to emerging markets. This has been called the global savings glut, since the reward to holding debt fell even as its level rose. As Figure 7 shows, borrowing costs plummeted even before central banks undertook quantitative easing and other non-standard monetary measures.
Figure 7. German long-run interest rates fell in line with global trends.
Source: OECD online database with authors’ elaboration. 10 year government bond yields.
The linchpin of the Eurozone vulnerabilities stem from the build-up of large current account imbalances. There is nothing intrinsically wrong with such flows. If nations borrow from foreigners to invest in capacity that helps them pay off the loan, everyone can be better off. To a large extent, this was not the case. In all the GIPS, the funds ended up in various non-traded sectors.
The first column of Table 1 shows the cumulated imbalance from the euro’s inception till the last year before Lehman Brothers went down in flames. The numbers for Greece, Cyprus, Portugal and Spain are enormously negative. This meant that these nations were investing far, far more than they were saving and implicitly financing it via foreign borrowing. Of course, all of this was the outcome of open markets. None of the governments (with the exception of Portugal and Greece) were involved systematically in the foreign borrowing or lending.
On the creditor side, the figures are high but not quite as high for the large core nations – Germany, France, and Netherlands. Italy, interestingly, is only modestly negative during these years at -8%.
The second column shows that for some of these nations, the inflow of foreign capital was implicitly financing budget deficits. In Greece and Portugal the large negative numbers in the first column (foreign borrowing) are matched by large negative ones in the second column (government borrowing). The large cumulative current account deficits also stand out for Spain, but it is not matched by corresponding government deficits.
Even Germany and France had cumulative public borrowing of 20 percentage points of GDP over this period. Italy’s was on a similar scale, although a bit higher. None of these countries however had large current account imbalances. On the surplus side, Finland and Luxembourg have unusually large numbers.
Table 2. Summary of pre-crisis imbalances.
Source: IMF and European Banking Association online data with authors’ elaboration.
Plainly, a great deal of public debt was being created during the Eurozone’s peaceful years. But as Figure 4 showed, the good growth during this period resulted in falling debt burdens in most euro nations. For reference, the endpoint government debt-to-GDP ratio is shown in the final column.
The third and fourth columns of Table 1 show the increase from 2000 to 2008 in bank assets as a fraction of GDP, and, respectively, the asset-to-GDP ratio on the cusp of the crisis. The numbers are remarkable.
By 2007, many banks were not only too big to fail, they were too big to save (Gros and Micossi 2008). Ireland’s banks had assets (and thus loans) worth seven times Irish GDP. The core economies were not much better with their banks holding more than twice their nation’s GDP. The figures were over three times for Germany, France and the Netherlands. Luxembourg’s number was astronomic.
It is, ex post, surprising that the building fragilities went unnoticed. In a sense, this was the counterpart of US authorities not realising the toxicity of the rising pile of subprime housing loans.
Till 2007, the Eurozone was widely judged as somewhere between a good thing and a great thing. The rose-garden feeling, however, started to disintegrate with the fall of Lehman Brothers in September 2008. Slowing growth and heightening fear of risk soon started to tell on the Eurozone economy as a whole, but especially for those that had built up large stocks of public and private debt, or run up large current account deficits.
Yet as it became clear in the Summer of 2009 that the Lehman shock would not create a second Great Depression, Eurozone spreads declined substantially. This was not to last.
Every crisis has a trigger. In Europe it was the revelation of the Greek deficit deceit.
The true deficit was twice as large as previously announced – a whopping 12.5%.
“The Greek fiscal crisis acted as a detonator in two ways,” writes Stefano Micossi. “It alerted the authorities and public opinions in Germany and the other ‘core’ countries to the possibility of large (and hidden) violations of the common fiscal rules; and it alerted financial markets to the risk of a sovereign default in a system where the provision of liquidity to ensure the orderly rollover of distressed sovereigns is not guaranteed”.
What followed was a six-month Greek effort to save itself. This failed. Greece was caught in a classic public debt vortex.
A nation’s debt is sustainable when the debt burden – commonly measured by the debt-to-GDP ratio – is not rising forever. Higher debt-service costs combined with a plummeting GDP made many investor suspect that Greek debt might be unsustainable. As markets raised their estimate of Greek default risk, they demanded higher interest rates to compensate for the extra risk. Higher rates, however, raised Greece’s debt-service payments, worsening the budget deficit.
In response, Greece slashed spending and raised taxes. But this backfired – it set off an austerity cycle. Since the austerity involved mostly tax rises, the belt tightening fanned a recession which reduce tax revenues and raised social spending, thus dragging the nation ever closer to the precipice of unsustainability. Credit agencies repeatedly downgraded Greek government debt and its borrowing cost rapidly rose from 1.5% to 5%. Note that all this happened before the first bailout.
Figure 8. Prelude and phase one of the crisis in Eurozone periphery
Source: OECD online database with authors’ elaboration. Note: The spreads are the difference between national 10-year government bond yields and those of Germany, in percentage points.
If a public debt vortex like this goes for long enough, there are only two ways to stop it – a default or a bailout by a ‘lender of last resort’.
Europe’s leaders decided that it was unthinkable for a Eurozone member to default, so Greece had to be bailed out. In the event, the ‘lenders of last resort’ were the Troika –Eurozone governments, the ECB and the IMF.
The rescue did not work and Greece’s package was too little too late. Markets did the math and realised that Greece was not on a clear path to debt sustainability. The rushed and politically charged way in which the package was put together did nothing to bolster confidence in Eurozone leaders’ ability to handle fast-moving crises.
But worse was to come.
From early 2010, markets wondered whether Greece’s inability to save itself might also apply to other nations. These doubts – combined with the remorseless logic of public debt vortexes – was enough to drive up the yields in other Eurozone nations.
Importantly, debt levels were not the determinant issue when it came to which nations got in trouble.
Given the worldwide recession, all Eurozone governments were running deficits and thus having to borrow more on the markets. Only nations that relied (implicitly) on foreigner leading (i.e. were running current account deficits) faced contagion. The borrowing costs of Portugal and Ireland rose briskly once the Greek bailout was announced (Figure 8).
This was the beginning of a ‘sudden stop’. Nations that relied on foreign capital to cover their savings-investment gap – Ireland, Portugal, Spain and Italy – all nations with substantial current account deficits were affected. As it turned out, Eurozone investors were far more wary about lending to other Eurozone governments than they were about lending to their own.
The rise in the risk premiums set in train debt vortexes that pulled down both Ireland and Portugal although via very different mechanisms. In Ireland’s case, the imbalance that mattered lay in the state of its banks.
Banks, like nations, can be subject to debt vortexes. Banks borrow money short-term to lend it out long term. For each euro borrowed short term, the bank makes long-term loans of a dozen or more euros – this is called leverage. It is as profitable in good times as it is dangerous in bad times.
What puts the ‘bank’ in bankruptcy is the fact that banks go broke any time their short-term funders refuse to rollover the short-term funding. Banks, in other words, operate with a business model that would look financially irresponsible in any other sector. The whole thing only works since people believe that the banks can be rescued by a ‘lender of last resort’ – typically the national government or national central bank.
Yet two critical differences make systemic banking crises extremely pernicious – leverage and maturities.
The average Eurozone nation had a debt stock of about 70% of GDP going into the Crisis. Eurozone banks were holding vastly larger debts. In 2007, Irish banks held debt equal to 700% of the entire country’s GDP. Plainly a systemic banking crisis in Ireland could – and in fact did – bring down the whole nation.
Moreover, the maturity of bank borrowing is typically much shorter than that of nations, so the need for new funding is radically more pressing. A typical Eurozone government may have to seek fresh loans to cover, say, 10% of its outstanding debt per year. A typical Eurozone bank has to seek fresh loans worth 10% or more of its total debt on a daily basis.
This daily need for billions means that the vortex – once it gets going – can accelerate at a frightening pace. During the Lehman Brothers debacle, one bit of bad news – Lehman’s default – brought the entire US credit market to a halt within hours; it spread to the rest of the world within days.
The ‘doom loop’ is closed by the fact that the banks – who view their national government as their lender-of-last resort – are also major lenders to the governments. The rescue, in essence, could require the rescuers to borrow from the rescued. This is how Ireland went down.
Irish banks got in trouble in 2010 and the Irish government bailed them out. In this way, private debt imbalances became a public debt imbalance.
Like a tragic double-drowning, Ireland’s banking system went down first, and the government of Ireland went down trying to save it. The Irish bailout was signed in November 2010. This was the Eurozone’s first example of the doom-loop linking bad bank debt to national solvency.
Ireland’s bailout package did little to calm investors’ fears. The borrowing costs of Greece, Portugal and Ireland continued to rise. As with the Greek bailout, the Irish bailout saved the day but worsened the crisis.
By the time of the Portuguese bailout in May 2011, markets were demanding 16% for holding Greek bonds – a ruinous level even for nations in good economic shape. The Greek economy, however, was collapsing. After contracting about 5% in 2009 and 2010, it crashed by almost 9% in 2011.
In July 2011, the second Greek package was agreed in principle, but one of its elements enflamed the overall situation. As part of the Eurozone leaders’ new view that the private sector should bear part of the cost of the bailout, private holders of Greek government debt would see about half the face value of investment disappear in what was called Private Sector Involvement. This was a wake-up call for investors who still believed the Maastricht Treaty’s no-default clause.
Seeing private investors explicitly having to write down Eurozone government debt, and seeing how Eurozone leaders seemed unable to put the crisis behind them, markets drew the natural conclusion that holders of the debt of other Eurozone nations might also be forced into a write down.
Markets, already wary of lending across borders, became even more reluctant. Portugal, who had borrowed (via the current account) 10% of its GDP abroad in 2009 and 2010, was the next to suffer a sudden stop. Its bailout was signed in November 2011.
Once again, the bailout saved the day but worsened the crisis. After a brief respite, Irish and Portuguese rates continued their ascent towards levels that would bankrupt almost any nation. But worse was still to come.
The three countries hereto caught in the crisis were small and their debts were insignificant compared to the overall Eurozone output. Worries started to mount when markets started demanding higher rates for the government bonds of Belgium, Spain and Italy. Italy in particular was a mortal threat to the Eurozone given the size of its economy and its massive debt.
‘Phase Two’ of the Eurozone crisis had started. As IMF Chief Christine Lagarde candidly put it: "Developments this summer have indicated we are in a dangerous new phase" (Lagarde 2011).
Repeated attempts at getting ahead of the curve failed, giving rise to a general suspicion that the Eurozone crisis might spiral out of control.
This self-feeding aspect is one feature of the crisis that is both essential and elusive as it rests on perceptions.
Figure 9. Phase two – Contagion spreads to the Eurozone core
Source: OECD online database with authors’ elaboration. Note: The spreads are the difference between national 10-year government bond yields and those of Germany, in percentage points.
When investors started to lose confidence in Italy – a trend which was not helped by its volatile political situation – they sold Italian government bonds in an effort to avoid future losses. These sales pushed interest rates up, making it harder for Italy to fund the rollover of its debt at reasonable rates. Seeing the funding difficulties, markets demanded higher interest rates and the spiral continued. In this way, a liquidity crisis (i.e. difficulty in rolling over debt) can – all on its own – become a solvency crisis. It’s a matter of expectations.
A good way to think of this is as there being two equilibrium situations. In the first, the ‘good equilibrium’, markets believe Italy is solvent and thus are willing to rollover Italy’s debt at reasonable rates, so Italy stays solvent. This was the situation up until the beginning of Phase Two. In the second, the ‘bad equilibrium’, markets suspect Italy is insolvent and thus demand interest rates that make Italian debt unsustainable – thus confirming their suspicions.
The botched Portuguese bailout and Eurozone leaders’ inability to get ahead of the curve seems to have switched Europe from the good equilibrium to the bad one. It is impossible to know why markets think what they do, but many point to the 50% ‘haircut’ that was a pre-condition for Greece’s second bailout.
By insisting that private holders of Greek debt lose money, Eurozone leaders transformed fears of losses into real losses. Any lingering belief that default was impossible inside the Eurozone were erased. The thought foundations of the pre-Crisis imbalances in public, private and cross-border debt were shattered. The consequences were not long in coming.
The massive pre-2008 lending across Eurozone borders had exposed banks in the core to debt in the periphery. In early October 2011, a Franco-Belgian bank, Dexia, was pushed into a bank-debt vortex by worries over its exposure to Greek government debt. It was nationalised by Belgium by the end of the month. Fearing an Irish-like end to the story, the sudden stop started to drive up Belgian interest rates. Belgium had, by this point of the recession, turned from a net creditor to foreigners into a net borrower (it was in current deficit from 2008).
Given Spain’s large bank debt and collapsing property markets, similar worries had begun to spread to Spain since the Portuguese bailout. Events in Belgium accelerated the process.
Italian yields also soared, but not due to bank problems. With a debt-to-GDP ratio over 100%, Italy needed both good growth and reasonable borrowing costs to stay afloat. It was vulnerable to a sudden stop since its implicit reliance on foreign investors rose steadily during the crisis with its rising current account deficit. Italy, in short, was coming down with the sudden-stop disease.
Sharp actions by national governments calmed the waters for a few months, but attempts to switch investors’ from the bad-equilibrium expectation to the good-equilibrium expectations failed. It all started up again after the second Greek bailout once again disappointed markets.
The Great Recession produced counter-cyclical fiscal policy via the usual automatic stabilisers – falling tax receipts and rising social spending. This surely dampened the shock and prevented the Great Recession from becoming the second Great Depression.
Figure 10. Fiscal policy turned pro-cyclical from 2010
Note: General government primary net lending/borrowing (billion EUR).
Source: IMF WEO online database with authors’ elaboration.
From 2010, however, the fiscal policy stance flipped from stimulus to contraction, as Figure 10 shows. The Eurozone as a whole saw its 2010 primary deficit move from about minus €350 billion in 2010 to €10 billion in 2014. This was a massive contractionary shock – equal to 4 percentage points of the monetary union’s economy.
The tightening by Greece, Ireland, Italy, Portugal and Spain was unavoidable as they were either in bailout packages that prescribed fiscal tightening, or they were doing the tightening themselves to evade the debt-vortex.
The effects on the economy were amplified by the fact that most countries achieved the tightening mostly by raising taxes – cutting public spending would have been less contractionary (Alesina et al. 2015).
Table 3. Pro-cyclical fiscal policy, 2010 to 2014.
Source: IMF WEO online database with authors’ elaboration.
The economic impact of the crisis was also made worse by tightening in the core nations who were not suffering from the sudden stop. As Guido Tabellini puts it: “When hit by a sudden stop, domestic fiscal policy has no option but to become more restrictive, and a credit squeeze cannot be avoided as domestic banks are forced to deleverage. To avoid a deep and prolonged recession, active aggregate demand management at the level of the Eurozone as a whole is required. But this did not happen”.
Things were plainly going from bad to worse. Each attempt to end the crisis seemed to make matters worse.
By this time, the contagion spread all the way to France. Its debt was downgraded and market yields rose substantially above those of other ‘core’ Eurozone nations like Germany and the Netherlands. British Prime Minister Gordon Brown unhelpfully suggested that Italy and France might need a bailout. The Belgian problem – domestic banks in trouble due to Greek lending – spread to Cyprus. Its banks were severely affected by the Greek debt write down, so the nation asked for a bailout in June 2012 (granted in March 2013).
Needless to say, a crisis that threatened Italy and France was a crisis of global dimension. This was no longer an issue of Greece fiddling the books to pay for the Olympics – this had the potential of blowing up the Eurozone and possibly the EU itself. The world economy was looking at another Lehman-sized shock. With Eurozone leaders manifestly incapable of mastering events, something had to be done.
Figure 11. Yields converged again after Draghi’s intervention.
Source: OECD online database with authors’ elaboration.
That something was a forceful intervention by ECB President Mario Draghi in his famous July 2012 speech. He told markets that the ECB would do “whatever it takes” to keep the Eurozone together. That did the trick. It switched expectations from 2011 and 2012’s doom-is-inevitable back to the old we-will-get-through-this-thing expectations of 2009 and 2010. Borrow cost returned to pre-Crisis levels (Figure 11).
The basic switching mechanism that Draghi triggered is a direct corollary of the debt-vortex logic. The rush to unload debt is driven by fear. The fear is driven by the suspicion that everyone else will sell the nation’s debt, thus driving borrowing costs up to the point where the nation actually goes broke. But if there is a debt buyer-of-last-resort – someone who can buy unlimited amounts of debt – the suspicion dissolves and investors are happy to hold the debt. This is what Mario Draghi did in the Summer of 2012. So far it has worked.
Even in Greece, things seemed to be getting better. In April 2014, Greece successfully sold new debt on the open market at reasonable rates and the economy appeared to be recovering slowly. The election of the far left Syriza coalition threw things off track. Whether the Greek government’s stance was justified or not (our authors disagree), the outcome is indisputable. Borrowing costs soared back to levels that made Greek debt unsustainable.
The proximate cause of the Eurozone crisis was the rapid unwinding of intra-Eurozone lending/borrowing imbalances that built up in the 2000s. Some of this was to private borrowers (especially in Ireland and Spain) and some of it to public borrowers (especially in Greece and Portugal), but in every case the difficult debt mostly ended up in government hands. As Thorsten Beck and José-Luis Peydró put it, “[o]ften this private over-indebtedness ends up on governments’ balance sheets, so that the rise in public debt is more a consequence than a cause of a financial crisis”.
This ‘sudden stop’ was a crisis rather than a problem since Eurozone members could not devalue and their central banks could not bail out the government. As Paul de Grauwe writes, “[c]ountries that have their own currency and that are faced with such imbalances can devalue or revalue their currencies”. This was not an option for the indebted countries.
The causes of the crisis – imbalances and lack of crisis management mechanisms –tell us that there are really three sorts of underlying causes:
Some of these failures involved unanticipated events. Others were a failure to implement the provisions agreed in the Maastricht Treaty.
At one level, all these causes stem from a fundamental design flaw, emphasised by Feld et al. The key deficiency was the misalignment between accountability and authority, or as Feld et al. put it, the “divergence of liability and control”. If the control and liability had been supranational – as it is in US’s monetary union – the imbalances could surely have been handled without provoking a continent-wide economic crisis. It is much more likely that at least the public debt run up would have been not allowed to go so far in Greece. Likewise, if control and liability had been effectively unified at the national level, nations would have had to deal with their own debt problems, perhaps with the help of the IMF. This might also have prevented or reduced some of the pre-crisis build-ups, as happens among US states (most of whom have balanced budget clauses in their state constitutions).
Paul de Grauwe elaborates on the dire consequences of divorcing authority and accountability. “When the Eurozone was started, a fundamental stabilizing force that existed at the level of the member-states was taken away from these countries. This is the lender of last resort function of the central bank.” Eurozone governments “could no longer guarantee that the cash would always be available to roll over the government debt”. Unlike stand-alone nations, Eurozone members did not have “the power to force the central bank to provide liquidity in times of crisis”.
This created a fundamental fragility in the monetary union. Without a buyer-of-last-resort, shocks that provide re-funding difficulties in banks or nations can trigger self-fulfilling liquidity crises that degenerate into solvency problems.
At an even deeper level, Elias Papaioannou stresses the differences in national institutions as the bedrock source of problems. “Somewhat paradoxically, the criteria for joining the euro did not touch upon key institutional issues, related to state capacity (tax collection), property rights protection, investor rights, red tape, and administrative-bureaucratic quality. The high growth during the convergence period came mostly from increased investment and some limited reforms, mostly on banking and monetary policy stability”. The Maastricht criteria failed, he asserts since they focused on nominal targets rather than the fundamental institutions that are source of divergences in the first place.
The Maastricht Treaty assigned monetary policy to the ECB. The Bank was made politically independent and instructed to maintain price stability. This worked. Fiscal policy was a different matter.
Many warned that a monetary union without a fiscal union would be problematic as Eurozone members might run up unsustainable debts that would either lead to pressure on the ECB to inflate it away, or to pressure on members to bailout an insolvent member to avoid the fallout from a sovereign default.
Ultimately, fiscal union was viewed as politically unrealistic so it was left to member governments. Three safeguards were put in place to prevent problems:
During the pre-crisis years, the Stability and Growth Pact failed. As Feld et al. write, the Stability and Growth Pact “sanctioning mechanisms were barely employed; there were 34 breaches of the 3% threshold for the general government deficit between 1999 and 2007… The breaches of the pact by Germany and France set particularly detrimental precedents”.
Many Eurozone banks were dangerously overleveraged going into 2010 due to regulatory failures before 2007, and half-hearted bank clean-ups after the Global Crisis. The responsibility for bank regulation was, as with fiscal policy, left to Eurozone member states, but there was no Stability and Growth Pact for banking.
Banking supervision was not really a focus of the 1990s discussion leading up to the Maastricht Treaty. Coordination of banking rules was at best mentioned in passing in the 1990s and few thought of moving deposit insurance, supervision, or bank resolution to Eurozone levels.
National central banks in both the surplus and the deficit countries -- which at the time still in charge of local banking supervision -- failed to realize what the huge intra-Eurozone credit flows were financing.
The public and private debt problems interact in a deadly embrace called the ‘doom loop. This is a cause of the sudden stop and crisis contagion in the sense that it opens the door to self-fulfilling crises. It makes the monetary union vulnerable to shocks that get amplified all out of proportion even if the debt imbalances are not extreme.
The problem, as Daniel Gros writes is that “[i]n Europe, the banks and the sovereign are usually so closely linked that one cannot survive without the other”. As discussed above in the context of the doom loop, Eurozone national governments are the ultimate guarantor of their banks, but the banks are key holders of public debt. As a result, “insolvency of a government would also wipe out the capital of the banks and bankrupt them as well. But an insolvent government would no longer be able to save its banks.”
The key element in this doom loop is that banks hold large amounts of the debt of their own government. If banks instead held more diversified portfolios of public debt, there would be much less doom in the doom loop.
The problem this creates, according to Agnes Bénassy-Quéré, is that it makes sovereign default very difficult. “The risk is that a sovereign default could trigger bank insolvency. The latter would then trigger a liquidity crisis since insolvent banks cannot get refinancing within the lender-of-last-resort procedures. To avoid a collapse of the national economy, there would be no other way than to re-introduce a national currency to refinance the banks”. This was a real possibility for Greece during its Summer 2015 crisis.
Beatrice Weder di Mauro illustrates the problems created by the inability to deal with insolvent Eurozone members. “Although Greece is extreme in many ways, it has repeatedly highlighted the European failure to establish a regime for dealing with cases of unsustainable debt. After the no-bail out clause failed to prevent excess debt accumulation, the Eurozone had to find a quick fix when Greece lost market access in 2010”.
The risks of credit imbalances can be diminished by surveillance and avoiding the accumulation of excessive imbalances. But it will never disappear. As Paul de Grauwe says, a fundamental feature of a capitalistic system “is that it is characterized by booms and busts; bubbles and crashes”.
If a sudden stop occurs, the sovereign most likely will lack the fiscal resources to cope with it. The size of the financial sector has grown just too large: at the end of 2007, bank assets were several multiples of GDP in most Eurozone countries (Table 1). Unlike a typical country hit by a sudden stop, an Eurozone member cannot devalue its currency to cope with the crisis. Yet, as became crystal clear in Greece but also in other countries of Southern Europe, currency risk is a major concern for market participants.
During a sudden stop, the bank-sovereign loop that we have seen at work during the crisis becomes inevitable. The home bias of bank portfolios aggravates the loop. But in the presence of currency risk, even a bank with well-diversified assets would not be able to withstand the flight to safety of its depositors. And the sovereign would typically not be in a position to help, given that it cannot devalue nor print money.
In countries with high public debt the sovereign itself could be the primary source of fragility, and its exposure to debt runs could activate the bank-sovereign loop. Any country with a large public debt, and with no access to monetary financing, could be subject to a run on its debt, even if it was solvent in the long run. In other words, a liquidity crisis triggered by lack of confidence could push into insolvency not only banks, but also sovereigns with high public debts and no access to the printing press.
By the end of the acute phase of the crisis, Spain was able to borrow from a Eurozone-level institution to fix its banks. This was a major improvement over the situation faced by Ireland at the start of the crisis. But the funds lent by the European Stability Mechanism did not go directly to Spanish banks, increasing their capital. They were borrowed by the Spanish government, which in turned used them to increase bank capital. The outcome was a further increase in sovereign debt.
The Eurozone crisis was mismanaged on many levels. When it came to the firefighting moment, it is puzzling that, as Beck and Peydro state, “European policy makers decided not to draw on the extensive crisis resolution experience in and outside Europe, a decision which has led to substantial number of policy mistakes”.
Charles Wyplosz provides the response, “[t]he simple answer to these questions is that the European treaties never anticipated that there could be such a crisis”. He continues, judging that “politicians have reluctantly been led to micromanage complex technical issues as the result of an amazing accumulation of economic mistakes, which they are unwilling to recognize”.
The lack of clear lines of liability and control caused problems for crisis management. As Giancarlo Corsetti writes, “[i]nstability grew out of a disruptive deadlock between national governments forced to address and correct fundamental weaknesses in their national economies on their own, and the Eurozone-level policymaking, which could have created the conditions for successful implementation of national policies, but did too little, too late (at best)”.
This lacuna was on fully display in Summer 2015. Despite repeated debt restructuring, the economic fallout from this year’s long-lasting negotiations between Greece and its creditors has probably made the nation insolvent. According to the IMF Greek debt is unsustainable without further restructuring.
More specifically, when Greece got into its sovereign debt crisis in 2010, the standard solution would have been for Greece to turn to the IMF for help. But, as Jeff Frankel writes, “the reaction of leaders in both Frankfurt and Brussels was that going to the IMF was unthinkable, that this was a problem to be settled within Europe. They chose to play for time”. This turned out to be a critical mistake.
When it became clear that the Irish banks were in trouble, no Eurozone structures existed to facilitate a collective rescue. In this sense, the policy mistake was not having anticipated the problems that would arise when the responsibility for banks was with national governments but the capacity to bailout them was only at the Eurozone level.
While there were important mistakes made in crafting the bailout packages, this is not the place for detailed critiques. Nevertheless, common problems came in the sequencing of reforms and a general lack of national ownership of the credit-for-reform deals. In retrospect, it is also clear that it was a mistake to not write down more of Greece’s debt early on when most of the debt was still held by private creditors. As Frankel writes, private debt holders, “could usefully have taken a ‘haircut,’ in a way that public sector creditors cannot … But again, leaders in both Frankfurt and Brussels insisted in 2010 and 2011 that writing down the debt was unthinkable”.
Some of the ‘mistakes’ were in fact linked to deeper conflicts. As Philip Lane stresses, the creditors and debtors in the bailout packages share a common currency and are deeply intertwined in terms of economic and political linkages. This inevitably creates conflicts of interest when it comes to the design of the bailout programmes and the potential role of debt restructuring mechanisms.
Eurozone governments systemically failed to understand the regime change implied by adoption of a common currency. A key aspect of this was the implication of financial integration. The thinking in the 1980s and 1990s, when financial integration was introduced before the common currency, was that it would contribute to both convergence and macroeconomic stability. Instead, capital flows tended to feed non-tradable sectors in the periphery of the Eurozone. Second, financial integration did not play as a smoothing device when the crisis hit. Quite the opposite, crisis countries suffered sudden stops.
In this same line is the idea that the Eurozone was as a whole a large but fairly closed economy, while Eurozone governments continued with the mind-sets of small open economies – each ignoring the impact of their own actions on the situation faced by the collective.
The goal of this ebook is to establish a consensus on the causes and a narrative for the Eurozone crisis. The idea is to agree what happen as a first step towards developing a consensus on what should be done to fix the current problems and to create mechanisms that will make the next crisis less damaging.
Although the main focus of this ebook is the analysis of what happened trying to see if there is a common narrative, some authors did not refrain from suggesting how the monetary union should be reformed. Here proposals vary and understandably is harder to find a common theme.
Guido Tabellini stresses the need to develop shock absorbers at the Eurozone level. Referring to the classic macroeconomic trilemma (the Eurozone cannot have full financial integration, financial stability and no common fiscal policy), he writes, “this trilemma implies that, in order to preserve financial integration and avoid future crisis, we need adequate common fiscal resources to cope with both systemic banking crisis and sovereign debt runs”.
He also stresses the need for a remedy to the tendency of Eurozone members to go for pro-cyclical fiscal policy. “This aggregate demand mismanagement was not just the result of human error. It reflects the institutional design of the Eurozone”. His solutions would surely attract disagreement from other authors. “These institutional features that led to this mismanagement ought to be corrected by changing the mandate of the ECB, by removing the constraints on monetary financing in order to facilitate a coordinated monetary and fiscal expansion, and by endowing the Eurozone with the possibility of issuing and servicing its own debt”.
A fresh solution to the difficulties of adjustment with a fixed nominal exchange rate is suggested by Pesenti. The idea is that structural reforms make it possible to substitute relative price changes with shifts in the composition of output. But setting up firms and new production lines is costly and typically requires financial resources. Structural reform cannot succeed without appropriate policies that address tight credit constraints on investment and firms' activity due to liquidity and balance sheet problems hitting banks. We no longer have a dichotomy between costly reforms and anti-recessionary monetary policy, but rather an integrated and perhaps coordinated vision of monetary and structural policies to restore growth. The alternative, unattractive, option is continuing reliance on deflationary adjustment in a currency union stuck at the zero lower bound.
Beatrice Weder di Mauro stresses the legacy debt as the core fragility that must be eliminated if any solution is to work. Referring to a report she helped write earlier this year that accepts the no-mutualisation red line as a given, she writes, “Corsetti et al. (2005) have proposed an alternative approach for a rapid and concerted debt reduction. The proposal involves an agreement by all Eurozone countries to commit future revenues for the sake of retiring debt. They would bring forward current and future income streams and commit their net present value to buy back the national debt now. Capitalising even small current and future income streams over a long horizon generates in net present value terms a large sum of money to buy back the debt. In addition, elements of solidarity and a debt equity swap could make the debt reduction deal viable and equitable”.
“The aim of the debt reduction deal is to eliminate legacy public debt, bringing debt in countries (expect Greece, which is a special case), below 95% and thus plausibly into the zone of solvency”.
“The second pillar of the proposal is a regime to deal with cases of unsustainable sovereign debt, which would help prevent countries from becoming too big to fail, again”.
When Europe’s Economic and Monetary Union (EMU) was being designed in the late 1980s, there was no clear vision on the standards of political and institutional cohesion among members that would be required to make the project viable. “The consensus view was that the member states of the union would be able to reach agreement and cooperate on how to create the common currency institutions over time”, as Giancarlo Corsetti put it. Euro nations would be able to muddle through any problems.
Although shocks would create fault lines and policy conflicts, it was widely expected that the cooperation would prevail – as it had always done before. The historical importance of tying together Europe would provide sufficient motivation to overcome obstacles, smooth differences over policy, and elicit solidarity.
And the need for further changes was quite clear. For example, CEPR’s Monitoring the ECB report (Begg et al. 1998) wrote, “The ECB suffers serious faults in its design that sooner or later will surface. This is likely to happen when large shocks, such as the world financial crisis, hit euroland” (the world crisis referred to here was the 1997 Asian Crisis). “The lack of centralized banking supervision, together with the absence of clear responsibilities in crisis management, risk making the financial system in euroland fragile. No secure mechanism exists for creating liquidity in a crisis, and there remain flaws in proposals for dealing with insolvency during a large banking collapse… These design faults are due to a failure to put sufficient decision making power at the centre of the system”.
Europe’s bad luck exposed the costs of relying on Monnet’s view. Shocks the size of the Global Crisis and Great Recession were not really what the Eurozone’s architects had in mind when they thought they could rely on muddling-through. From 2010 to 2012, Monnet’s logic was turned on its head. Discussion of how to complete the currency union proved divisive and destabilising. It made matters worse and a consensus had to be reached on institutional issues while knitting together agreements on immediate emergency measures.
The consequences were and still are dreadful. Europe’s lingering economic malaise is not just a slow recovery. Mainstream forecasts predict that hundreds of millions of Europeans will miss out on the opportunities that past generations took for granted. The crisis-burden falls hardest on Europe’s youth whose lifetime earning-profiles have already suffered.
Money, however, is not the main issue. This is no longer just an economic crisis. The economic hardship has fuelled populism and political extremism. In a setting that is more unstable than any time since the 1930s, nationalistic, anti-European rhetoric is becoming mainstream. Political parties argue for breaking up the Eurozone and the EU. It is not inconceivable that far-right or far-left populist parties could soon hold or share power in several EU nations.
Many influential observers recognise the bind in which Europe finds itself. A broad gamut of useful solutions have been suggested. Yet existing rules, institutions and political bargains prevent effective action. Policymakers seem to have painted themselves into a corner.
This ebook is a first step in a bigger project called “Rebooting Europe”. It seeks to marshal a critical mass of Europe’s best thinkers in developing ways to get Europe working again. To undertake a systematic rethink of today’s European socio-economic-political system. In short, to figure out a way to update Europe’s ‘operating system’ and reboot.
Alesina, A, O Barbiero, C Favero, F Giavazzi, and M Paradisi (2015) “Austerity in 2009–13”, Economic Policy 30 (83): 383-437.
Baldwin, R, D Gros and L Laeven (2010), “Introduction” in Baldwin, Gros and Laeven (eds.) Completing the Eurozone rescue: What more needs to be done?, CEPR Press.
Begg, D, F Giavazzi, P De Grauwe, H Uhlig and C Wyplosz (1998), The ECB: Safe at Any Speed?, monitoring the ECB, Vol. 1, CEPR Press.
Gros, D and S Micossi (2008), “The beginning of the end game…”, VoxEU.org, 20 September 2008.
Lagarde, C (2011), "Global Risks Are Rising, But There Is a Path to Recovery", remarks at Jackson Hole, 27 August.