Nouriel Roubini's Global EconoMonitor

Much Ado About Relatively Little: Contagion Risks From an Orderly Greek Debt Restructuring Are Modest, Contained and Manageable


The ECB and other commentators have raised the specter that a debt restructuring in Greece would lead to massive and destructive contagion to financial markets, banks and financial institutions and other EZ sovereigns that would in turn lead to financial disaster. Those fears have now led to increasingly shrill warnings, despite their empirical basis being extremely weak.

The ECB’s dogmatic, dogged dismissal of debt relief and hyperbolic use of the D words of disorderly default, propagates rather than mitigates instability. Contagion, contamination and moral hazard are unfolding before our eyes already, since Greece’s debt is unsustainable and its program off-track; Ireland’s bank rescues have destroyed sovereign creditworthiness, compromising its ability to support those banks; Portugal has lost market access; and official discussion and ECB rejection of debt relief has re-ignited financial instability.

An objective analysis and pre-emptive, orderly debt rescheduling would actually mitigate rather than exacerbate the risks of contagion, contamination and moral hazard. Quid-pro-quos among creditors and debtors, public and private, bank creditors and shareholders, would make for a more politically viable adjustment. It would also avoid the moral hazard of indefinite bailouts and a fiscal transfer union by stealth that would result from the ECB’s approach of “kicking the can down the road,” a can that is already too heavy to kick or carry given the political realities of limited EZ fiscal integration and federalization.

This paper suggests that an orderly restructuring of debt is not likely to lead to meaningful distress and contagion for banks and sovereigns within and outside the EZ. Any possible form of contagion would be modest and/or managed, contained and/or ring-fenced with the appropriate policies. A rational analysis suggests that the increasingly excitable contagion concerns voiced by the ECB and others are vastly exaggerated and that the actual risks are manageable and preventable.

Introduction: A Serious Discussion About Contagion Risks

Those who are vehemently opposed to any orderly restructuring of Greece’s public debt—even a modest re-profiling of it—start with the ECB, especially president Jean-Claude Trichet and executive board member Lorenzo Bini Smaghi. They—and others within the ECB and outside of it—have been presenting the view that a debt restructuring in Greece—they incorrectly keep on referring to it as a default, which is a very different event from an orderly debt restructuring—would lead to financial Armageddon, with massive contagion to banks and sovereigns all over the EZ, destroying the capital of Greek and creditor banks in the bloc’s core, triggering widespread bank and sovereign debt runs and leading to a Lehman-style EZ and global financial meltdown. The ECB—or the vocal part of its executive board—is so upset at the thought of even a modest re-profiling—or “soft restructuring”—that Trichet stormed out of the May 6 “secret” meeting of EU finance ministers when the option of an orderly debt restructuring was discussed.

The reality is that these comments about massive contagion and financial disaster, akin to fear-mongering, have little conceptual or empirical basis. Indeed a rational—as opposed to an unsubstantiated—analysis of the risks of contagion deriving from an orderly restructuring of Greece’s debt suggests that those risks are modest, contained and manageable. The ECB’s ideological blindfolds—that have led to it not being able to seriously consider the possibility of an orderly restructuring and the ways of avoiding contagion—have left the institution in an awkward spot. Its asset ledger is now piled high with hundreds of billions of euros of claims against PIIGS banks that are relatively poor collateral—mostly public debt or bank debt guaranteed by governments—for its lending to PIIGS banks. And, on top of that, there is another pile of PIIGS public debt that is the result of its outright limited purchases of PIIGS government debt.

Thus, in part, the ECB’s opposition to a debt restructuring may now be coming more from concerns about the effects of such a restructuring on its balance sheet than a rational consideration of the pros and cons. Also, the ECB, by cornering itself in opposition to any mention of a restructuring, most likely did not do the necessary homework and scenario analysis of what an orderly restructuring might look like: Indeed, its most senior members do not even seem to know—or pretend in public not to know— the difference between a default, an orderly restructuring and a soft restructuring, or re-profiling. This is an institutional failure that borders on recklessness; any policy institution doing its job should properly consider and analyze—in confidentiality—a “Plan B,” in case “Plan A” fails. This time, ideological bias has likely meant the ECB did not do the necessary homework on Plan B that would have led to it realizing there are ways to accomplish an orderly debt restructuring and limit its contagious effects. At the very least, the top-down views of some of its executive board members appear to have suppressed such analysis, so one can have concerns about how completely or thoroughly it was undertaken and considered. And ECB members have barely gone beyond extremely generic claims of massive contagion and financial meltdown (which have not in public been supported by any evidence), thereby obstructing serious discussion of the pros and cons of debt restructuring and the specific risks—and likelihood—of specific contagion channels.

Therefore, let us have a serious discussion of the alleged contagion risks from debt restructuring and the ways to avoid and/or limit such risks. These risks can be split into the following categories:

  • Risk of contagion to Greek banks;
  • Risk of contagion to creditor banks in the core of the EZ (Germany, France, etc.);
  • Risk of contagion to the sovereign and/or banks in the already distressed periphery (Portugal, Ireland);
  • Moral hazard of a Greek restructuring (relenting on the austerity/reform effort if there is restructuring);
  • Moral hazard of a restructuring in Greece for Ireland and Portugal (i.e. the latter relenting on their fiscal/reform efforts and expecting debt relief if Greece receives it);
  • Risk of contagion to the sovereign and/or banks in the not-yet-distressed periphery and core (Spain, Italy, Belgium); and
  • Risk of global contagion—outside the EZ—to global credit and equity markets.

The detailed analysis below will show that each one of these “contagion” risks is limited and/or manageable and/or mostly independent of whether Greece restructures its debt or not. Before we start the discussion it would be worth clarifying what is meant by contagion as there is a conceptual difference between contagion and economic and financial spillovers that are driven by fundamental factors. Usually the term “contagion” in the markets is understood to refer to “correlations between prima facie unrelated assets.” Here, we will use it mostly to designate economic and financial transmission from one sector or economy to another, regardless of whether these transmissions are driven by fundamental economic and financial spillovers or rather from more pure forms of contagion. Indeed, what is often referred to as contagion could—at a deeper level—be triggered by some more complex set of economic and financial spillovers. So, in this paper, the term contagion is used somewhat loosely.

Contagion to Greek Banks and Other Financial Institutions

First, here is a critical point: As long as a restructuring of Greek debt takes the form of a debt exchange with a par bond (as in the Brady plan and in the debt exchanges for Pakistan, Ukraine, Uruguay, the Dominican Republic and other recent cases)—i.e. no face-value reduction but rather the extension of maturities and capping of the interest rate on the new bonds below currently unsustainable rates and below the old bond coupon—the country gets significant debt relief, while banks, insurance companies and pension funds that are holding the debt to maturity (or in the banking books) would be allowed— subject to some regulatory rules such as IAS39 that we have discussed —to hold the new debt at face value rather than being forced to write it down (see specifically “A How-to Manual for Plan B: Options for Restructuring Greek Public Debt”). Thus, one of the most important sources of contagion for banks and other financial institutions holding Greek public debt would be significantly contained, as they would not be forced to write down the value of their claims in a debt exchange with a par bond. And indeed, the par bond option was the route successfully taken in the Brady plan and in most recent sovereign debt restructurings—Pakistan, Ukraine, Uruguay, the Dominican Republic, etc.—to address the concerns and the contagion risks for banks and other hold-to-maturity investors.

These banks and financial institutions would be playing a game of regulatory fudge that is currently allowed and fully abetted by the ECB and other regulators. Indeed, this regulatory fudge is occurring anyhow today as these institutions are allowed to hold the assets at face value—if they are in their banking or hold-to-maturity books—even if their market value is now one-third or so lower than face. So, the ECB and other opponents of debt restructuring cannot have it both ways: Either they admit now—even in the absence of a debt restructuring—that the debt is worth below par and force financial institutions to write down the debt (even if there is not a formal debt restructuring) and recapitalize themselves as needed (and as they should anyhow regardless of a debt restructuring); or, if they maintain the current regulatory fudge of pretending that the debt is worth face value (when the market value is much lower), they cannot then argue that a debt restructuring—that would not reduce the face value and that would have a market value no lower than the current market value of the old debt—should force a write down of the claims and make them ineligible for ECB repo financings.

The ECB has taken the incorrect view that a debt restructuring should force a write down even if it is a par bond with no face-value reduction and no additional net present value loss compared with current market value. That is an utterly flawed approach as, if the ECB takes that view, it should then force a debt write down for banks today, even in the absence of a debt restructuring. It cannot have it both ways and argue that a debt restructuring would force a debt write down.

Also, the ECB’s related argument and threat that, after a debt restructuring, the new debt would not be eligible for ECB repo financing, is similarly flawed. First, for CDS-triggering purposes, a debt exchange that doesn’t use collective action clauses (CACs) or domestic legislation for changing the terms of the debt is not considered a credit event or a restructuring event; thus, it does not trigger the CDS protection. Thus, this is not a reason for making new debt ineligible for ECB repo financing. It is true that rating agencies would consider a debt exchange—even a voluntary one that occurs under the implicit threat of default—a credit event and downgrade the debt to selective default (SD) rating accordingly. But again, reality is more complex as the Uruguay debt exchange episode with a par bond re-profiling shows. First, in the Uruguay case, the de-rating to SD level did not occur at the time of the announcement of a debt exchange, but only when the debt exchange actually occurred, with the old debt being tendered at the time of the exchange. Second, the de-rating to SD lasted only 17 days and, after the successful debt exchange (and very few holdouts), Uruguay’s rating was upgraded again as the country regained greater debt sustainability. Third, since the orderly debt exchange made Uruguay’s debt more sustainable, the country regained international capital market access one month after the exchange offer occurred. Of course, a par bond restructuring may not achieve debt sustainability for Greece and a more radical debt reduction may become necessary down the line; but the early orderly restructuring option would allow the introduction of CACs in all of the public debt and thus allow an orderly debt reduction in the event of one becoming necessary a few years down the line. This was indeed the argument for introducing CACs in Uruguay and other par bond debt restructuring cases.

This means that the ECB should stop threatening Greece with halting its banks’ access to the ECB’s repo financing in the event of a restructuring or re-profiling. Even if Greek debt were downgraded to SD for a couple of weeks, the ECB could patiently wait and then restart new repo operations until the rating upgrade has occurred. Secondly, instead of threatening to disrupt an orderly exchange offer, the ECB should support it—and prevent holdout problems—by accepting the new debt that has been restructured as good collateral for repo, while refusing to accepted the untendered old bonds. Rather than using exit consents or domestic legislation to deal with potential holdouts, the ECB could contribute best to an orderly restructuring by accepting the new bonds for repo and refusing to accept the old debt for such liquidity operations (rather than using destructive threats to disrupt an orderly debt exchange). The new debt would have, at worst, an SD rating for only a few weeks; thus, either the ECB could use its discretionary power to repo debt that is rated SD; or, as a fig leaf, it could not perform additional repos on Greek debt until the rating upgrade has occurred a few weeks after a successful debt exchange.

And the ECB is on the hook for the repo operations that it has already done—€90 billion of liquidity to Greek banks in exchange for a substantial €140 billion of collateral. The ECB is stuck with those existing repos for as far as the eye can see, regardless of whether Greece restructures its debt or not. It is indeed delusional to think that Greek banks will repay their loans to the ECB and get their collateral back any time soon; and the ECB knows and accepts this as it has implicitly endorsed the view that most Greek debt should eventually end up in the hands of the official sector (including the ECB) if a restructuring is to be avoided. The ECB could thus play a constructive role in an orderly debt restructuring of Greek debt, but instead it is now playing hardball by sending empty threats to pull the plug on old repos (empty as the ECB is on the hook and stuck for a very long time on the financing that it has already given to Greek banks) and destructive threats to stop new repos for Greek banks (as the new debt should be accepted as collateral for repos once an orderly restructuring occurs, while it is only the untendered debt that should not be accepted).

The discussion above suggests that the risk of contagion to Greek banks is very limited. They lost market access (in terms of interbank and cross-border loans) a long time ago and have relied on ECB liquidity and government guarantees of their unsecured debt to avoid financial distress. So a debt restructuring changes nothing for Greek banks: They don’t have market access now and they will not have market access after an exchange. And a par bond will allow them to keep pretending—as they have been allowed to do so far—that the public debt they hold is worth 100 cents on the dollar. Also, between the ECB’s repo operations and its outright purchases of Greek public debt (€45 billion), the Greek banks have already dumped most of their holdings of Greek public debt on the central bank. Thus, the additional exposure is modest and controllable with a par bond. It is true that Greek banks—like many other EZ banks—may eventually need to be recapitalized given their exposure to the sovereign and other bad claims against the private sector. But that recapitalization should occur—preferably sooner rather than later—regardless of whether a debt restructuring occurs or not. The ECB has made the unsubstantiated argument that a debt restructuring will destroy Greek banks’ entire capital: We believe this is incorrect as either these banks need to be recapitalized now (even without a debt restructuring) or because a restructuring with a par bond would lead to no material change in their capital needs.

The additional risk or contagion bugaboo that a debt restructuring would trigger a bank run on insured deposits is a red herring as those deposits are insured today by an already distressed sovereign. A debt restructuring would make that sovereign more solvent and more credibly able to backstop deposits than a clearly insolvent sovereign. So, if a bank run against insured deposits has not occurred so far, it will not occur after an orderly debt exchange that makes the sovereign more sovereign than it is now. It is true that a silent run on uninsured deposits—those above the deposit insurance limit—has already mostly occurred in Greece as wealthy nationals and firms holding large deposits are uncertain as to whether their large deposits are safe or not, but this run has already gone about as far as it will and Greece can always extend its deposit insurance—as many countries did during the peak of the global financial crisis post-Lehman—if it takes the view that such deposits should never be treated.

Finally, the position of the ECB is not only seriously obstructing the way to—and damaging the chances of—an orderly debt restructuring; it is also destructive as it could trigger the Greek bank run that it is claiming to want to avoid. Repeated statements by ECB members arguing that restructured debt would not be accepted as collateral for repo operations are already having a seriously disruptive effect on Greek banks; in a situation where banks have already lost market access for their unsecured debt and where a silent run on many large uninsured deposits has already occurred, such statements could trigger the run of insured and uninsured deposits, as well as destroy the chance of banks issuing debt guaranteed by the government. The ECB is thus playing with reckless fire, risking being the cause—with its statements—of the exact bank run that it is claiming to want to avoid.

Contagion to Creditor Banks in the EZ Core

The arguments made to explain why contagion to Greek banks would be limited apply a fortiori to other EZ banks, especially those in the core of the EZ (in Germany, France, etc.) that are holding Greek debt.

First, and most important, a par bond option would allow such banks and other financial institutions to continue—as with the Greek banks—the regulatory fudge that allows them now to hold the claims at face value as long as they are in the banking book (i.e. held to maturity) rather than in the trading book.

Second, a modest amount of moral suasion or the option of new bonds with the same face value and market value no lower than the current market value of the old bonds would induce these banks to accept the exchange offer (as they did in any other par bond exchange).

Third, German banks that hold Greek debt are mostly public banks (either because they were public in the first place like the Landesbanken or because they were taken over during the global financial crisis when they went bust like Hypo Real Estate or West LB). These banks would of course accept a debt exchange, and the risk of a run on their secured claims (bank deposits) or unsecured claims (senior debt) is zero as they are fully and credibly backstopped by a solvent sovereign (the German government). Some of these German banks may be treated under the new German insolvency regime and some of the bank debt may be at risk, especially if a resolution regime hits subordinated debt. But the issue of how to resolve insolvent German banks and their unsecured creditors is orthogonal to a Greek debt restructuring.

Fourth, the exposure of French and other core EZ banks doesn’t imply any serious contagion risk: Even if some of those banks are private rather than public, their deposits are guaranteed by a solvent sovereign and their senior bank debt was guaranteed after Lehman by their government.

Moreover, even if any of such banks were to take a capital loss in the unlikely event that a true haircut (say 30-40%) were to be imposed on their holdings of Greek debt, the amount of such holdings are modest enough that the banks could take such a hit without any disorderly effects or bank-run risk. So, either core EZ banks have enough capital to absorb a very unlikely hit—unlikely as a par bond doesn’t force them to take such a hit—or, alternatively, if such banks need to be recapitalized because they have poor assets against private and public sector debtors, they should eventually be recapitalized regardless of whether Greece restructures or not. Thus, the risk of massive contagion to core EZ banks is another red herring. It is a risk that is fully manageable for banks that have the effective formal and informal backstop of solvent sovereigns.

Finally, some may argue that a Greek debt restructuring may lead to a re-pricing of bank debt in EZ banks—whether in the periphery or core—as a debt restructuring in Greece might lead to a higher probability that senior unsecured debt and sub debt of banks would be treated and subject to haircuts down the line. But there is no relationship between a Greek public debt restructuring and the issue of unsecured bank debt. The Greek restructuring would apply to its public debt, not to its bank debt. And thus, there is no reason to believe that such a restructuring would have a material effect on the probability that any EZ bank debt—senior or sub—would be restructured down the line if an individual bank is in trouble. That contagion risk is—possibly but not surely—higher if Ireland were to restructure its senior secured or unsecured debt, but this is not what is on the table here and now. We are discussing the contagious effects of a restructuring of Greek public debt, not of its bank debt.

Risk of Contagion to the Sovereign and/or Banks in the Distressed Periphery

The risk of serious contagion to the sovereigns and banks of the already distressed periphery—specifically Ireland and Portugal—is another red herring. First, banks and other financial institutions in Ireland and Portugal have little or no exposure to Greek debt. Second, the sovereigns of these two countries have already lost market access and rely for the next few years on the IMF-EFSF-EU lifeline—conditionality loans—to finance themselves. So, the risk that they will lose market access if Greece restructures is zero. Third, the banks of Portugal and Ireland have also already lost market access and rely for the next few years on ECB support—liquidity in exchange for good or dubious collateral—and on their governments’ guarantees on their senior secured or unsecured bank debt to finance themselves. So, the risk that they will lose market access if Greece restructures is also zero as they have already lost market access and/or any market access they do have is dependent on government guarantees. And the banks’ senior secured claims—insured deposits—are insured by a sovereign that, however distressed, is still credibly committed to backstop such deposits. Thus, if a poor government guarantee of senior bank debt is good enough, a government guarantee of insured deposits is certainly much better. Therefore, the risk of a run against deposits is far-fetched. There is no way that a Greek debt restructuring would have any meaningful material effect on the Portuguese or Irish sovereign and/or the countries’ banks.

Leaving aside for now Ireland’s flawed decision not to treat its banks’ senior unsecured and guaranteed debt (which is putting so many private losses on the government’s balance sheet that it will eventually make the sovereign likely insolvent in a matter of a few years), the ECB and official community approach to Ireland—under the mistaken view that any loss to senior creditors of Irish banks should be avoided at any cost to prevent “contagion”—will have the likely consequence of bankrupting the Irish sovereign and eventually undermining its ability to backstop the bank debt that it is supposed to guarantee. So, to prevent short-term “contagion,” the ECB will likely cause a sovereign debt crisis down the line that will lead to much greater and more destructive contagion.

Moral Hazard of a Restructuring for Greece (Risk of Greece Relenting on the Austerity/Reform Effort if There Is Debt Relief via Restructuring)

A variant of the contagion argument is the alleged moral hazard problem: If debt relief is provided to Greece, the incentive to implement tough fiscal austerity and structural reform will fizzle out. This is a weak argument for many reasons. First, debt relief is not a substitute for austerity/reform, but a complement to it. Second, if a country has a debt overhang, maintaining that overhang will ensure that austerity/reforms will fail, while providing some complementary debt relief will increase the probability of success. Third, the official community still has a significant amount of leverage over Greece—as the recent push for additional austerity/reforms/privatization shows—as Greece is continues to run a primary deficit; i.e. even after an extreme debt restructuring that would stop all interest payments on its debt for a few years (quite unlikely), it would still need ongoing official financing for many years to fill its financing gap. Thus, the official community can effectively and credibly force Greece to do more even after a debt restructuring, as the threat of pulling the plug can be made; i.e. the stick of pulling the financing can be complementary to the carrot of financing as a way to induce the appropriate amount of ongoing austerity/reforms/privatization, thus significantly limiting the moral hazard effects of a restructuring on policy efforts. Fourth, academic research—both theoretical and empirical—shows that, given an unsustainable debt overhang, debt relief increases the incentive to achieve austerity and reform rather than reducing it. Indeed, a debt overhang means significant reform/austerity cannot restore sustainability, so the incentive to adjust diminishes without debt relief (a reverse form of moral hazard, as the absence of debt relief carrots makes the sticks of reform/austerity less likely to be credibly implemented).

Moral Hazard of a Greek Restructuring for Ireland and Portugal (i.e. the Latter Countries Relenting/Relapsing on Their Fiscal/Reform Efforts and Expecting Debt Relief if Greece Receives It)

A second variant of the “moral hazard contagion” argument is that, as soon as Greece gets some debt relief, Portugal and Ireland would say “me too, me too!” and thus stop their own policy efforts to reduce their deficits and implement structural reforms. This form of alleged contagion is yet another red herring, for number of reasons.

First, as in the case of Greece, the official community has a significant amount of leverage over Portugal and Ireland, as these countries are still running primary and overall fiscal deficits; i.e. they need ongoing official financing for many years to fill their financing needs, while their banks need ECB financing as they have lost market access. Thus, the official community can induce these countries to continue austerity and reform under the threat of pulling the plug.

Second, in each case, the official approach is to try Plan A—i.e. assume that the country is illiquid but solvent given official financing cum austerity/reform. Only if Plan A fails—as it already did in Greece after a year of discussions—will Plan B start. Ireland has recently started to implement Plan A, while Portugal has not even started its version, which requires a lot of policy effort. So neither one can yet credibly demand to move to Plan B (debt relief with continued policy effort), when Plan A has not been tried and found wanting, as with Greece.

Third, some sovereigns (like Ireland) might actually want to differentiate themselves from Greece and show that their policy efforts will restore debt sustainability and growth as they have shown past willingness to make sacrifices to restore sustainability. Thus, a Greek debt restructuring would not automatically lead them to demand relief; and anyhow, relief would be received if Plan A has not yet been implemented in good faith and effort, and is shown to be failing.

Finally, Portugal and Ireland may eventually need some debt relief for their sovereigns and other restructuring of senior bank debt if, as is likely, Plan A fails. But this debt relief would not occur as an automatic result of a Greek debt restructuring; it would occur only later, and if and when other alternatives have failed. Conceptually, to fully minimize the “contagion” risk, a restructuring of Greek debt should occur at the same time as the restructuring of other insolvent sovereigns and banks. But the sequential nature of Plan As—which have been implemented at different times—will imply a staggered sequence for Plan Bs (debt restructurings) if and only if Plan As fail on their own merit. So, an automatic domino effect from a Greek restructuring to an Irish or Portuguese restructuring is nonsense and a logical non-sequitur.

A restructuring of senior Irish bank debt should be considered on its own merits regardless of what Greece does. And certainly one can make a strong argument that not treating senior bank debt will lead to worse contagion when the Irish sovereign becomes insolvent because of this stubborn and ill-conceived ECB and EU-IMF policy attitude. The political system and the voters in Ireland will realize that not bailing in the senior creditors of banks is bad policy regardless of whether Greece restructures its debt or not. So, there is no need to blame a Greek restructuring for any decision Ireland makes on treating its senior bank debt, which should be anyhow restructured on its own merits.

Risk of Contagion to the Sovereign and/or Banks in the Not-Yet-Distressed Periphery and/or EZ Core

The next contagion risk to be considered is that a debt restructuring in Greece could lead to contagion to sovereigns and/or their banks/financial institutions in some of the EZ economies in the core and/or periphery that (unlike Ireland or Portugal) have not yet lost market access. The concern here is that even an orderly restructuring of Greek debt would cause sovereign spreads to spike in these other countries and lead them to lose market access and, additionally, lead their banks to lose market access, as Greek, Irish and Portuguese banks have. These fears of contagion to the rest of the EZ have been enhanced by a rise in sovereign spreads over the past week, especially in Spain and Italy. Note, however, that this alleged “contagion” is more related to idiosyncratic factors specific to Spain and Italy (in Spain, the rout of the Socialist Party in regional elections and concerns about hidden deficits and the debt of regional governments; in Italy, regional elections and S&P’s negative watch on Italian debt) rather than direct contagion from the noise about a Greek debt restructuring.

In addressing the concerns about contagion risk from Greece to other weak EZ economies, it is important to consider the following critical point: The best way for any country to decouple from trouble from Greece and avoid contagion is to follow sound macro, fiscal and financial policies. So, if policies are credible and sound, the risk that an economy like Spain (just to consider a case)—or, better, its sovereign and/or banks—would lose market access is extremely low. Conversely, as the cases of Ireland and Portugal prove, if the economic and financial policies are not sound and credible, then the economy would lose market access even if Greece does not restructure its debt. So, at the first approximation, the risk of contagion is very limited as long as Spain, Italy and Greece follow sound policies.

Second, there is always a risk of a self-fulfilling run on a sovereign and/or its banks that do follow sound policies if there is residual uncertainty about the credibility of those policies; i.e. the commitment of the sovereign to pursue sound policies, especially if that sovereign does not have enough liquid resources to prevent a self-fulfilling run on its debt and/or the debt of its banks. Indeed, the risk of a pure self-fulfilling run on the Spanish sovereign and its banks was higher when the size of the European Financial Stability Facility (EFSF) had not been yet ramped up to ensure that, if the Spanish sovereign and/or banks were to lose market access, there were enough resources to backstop them, at least for a few years; i.e. Spain was too big to fail, but also too big to be saved or to be bailed out. But with the recent EU decision to flexibilize the EFSF so that the full €440 billion of resources would be available for lending to economies under stress, there is now enough official money—at least in principle—between the IMF, the EFSF and the EU to backstop the Spanish sovereign and its banks for three years, if they were to lose market access. So, the risk of a pure self-fulfilling run on Spain—even if the country were to follow sound policies—because of pure illiquidity triggering a bad equilibrium with a self-fulfilling run, is now significantly reduced. Also, in the event of fundamental factors such as policy slippages leading to a loss of market access and the need for official support, Spain would still be too big to fail, but no longer too big to be saved; official resources would be sufficient in principle to backstop the government and its banks.

Third, one could argue that, while it is true that a country with poor policies would lose market access even without a Greek restructuring, the probability that a country with sounder policies could experience contagion, see its spreads rise and eventually lose market access would be significantly higher in the event of a restructuring of Greek debt. And since Spain is still at risk of losing market access, it would not be wise to increase such a risk of contagion by embarking on a restructuring of Greek debt. This argument should be addressed by noting that, in the past couple of years, surges of Spanish spreads (and Italian ones for that matter) have occurred mostly when policy slippages have led to a rise in the risk of a loss of market access. Pure contagion, from concerns about Greece (or Ireland or Portugal) losing market access, to Spain does not seem to be consistent with the data. Indeed, official rhetoric up until recent weeks had been that Spain—by pursuing fiscal austerity and reforms—had successfully decoupled from the troubles of Greece, Ireland and Portugal as its sovereign spreads had not widened and as its banks had not lost market access and thus were less reliant on ECB financing. So, there is a contradiction between the arguments of those who until recently argued that Spain (or Italy for that matter)—by following “sound and credible” policies—had successfully decoupled from Greece, Portugal and Ireland, and the recently heard arguments that an orderly restructuring of Greek debt would cause massive contagion to Spain.

Fourth, even if the probability of a loss of market access for Spain were to somehow rise in the event of a Greek debt restructuring—and that is only a guess rather than a certainty—the official community can significantly reduce such a risk of contagion by properly ring-fencing Spain, Italy or other economies at risk. As noted above, larger official resources now prevent the risk of a pure self-fulfilling run. Second, the official sector could provide Spain or Italy with a shield similar to the IMF’s FCL line: A line of credit that is not used until necessary and that allows a sound sovereign to borrow significant amounts from the official sector without excessive conditionality in the event of pure market panic leading to the risk of a loss of market access. And if that line of defense is not sufficient if policy slippages increase the risk of a fundamentals-driven loss of market access, large official resources conditional on austerity and reform could be used to backstop the Spanish government and its banks for up to three years.

Fifth, how could one deal with the risk of a partial bank run on insured deposits, in Spain for example? Indeed, while EFSF resources are now sufficient to backstop the Spanish sovereign and its banks for three years (for their unsecured debts), they would not be sufficient if a bank run in Spain were to lead to a fall in insured deposits. Even a modest 10% run on insured deposits in Spain—together with a loss of market access for the sovereign and its banks—would require official resources that would be several hundred billion euros larger than what the EFSF, the EU and the IMF could provide, given the envelope of official resources. But, as long as the Spanish sovereign is solvent or credibly committed to backstop insured deposits, the risk of a bank run would be very low. Also, official resources are sufficient to backstop unsecured debts of banks. And such unsecured debts can be further backstopped by ECB repo financing facilities and government guarantees of unsecured debt. And in the extreme (and highly unlikely) case that a partial run on insured deposits in Spain were to occur, the ECB could radically increase its lender-of-last-resort support activities to nip in the bud any run on insured deposits in Spain and/or in other EZ banks.

Sixth, among the EZ economies potentially at risk—Spain, Italy and Belgium—the one with a sovereign and banks under greatest stress is Spain, as the economy has seen a boom and bust in the housing sector that has led a severe economic downturn. While until recently Spain had decoupled from other EZ countries in distress, we remain more bearish than market spreads and market consensus about the risk that Spain will lose market access in the next year. But our bearishness is independent of what happens in Greece and whether Greece restructures or not; it is rather dependent on a more bearish assessment of Spain’s fiscal, economic and financial fundamentals. If we are right, then Spain is likely to lose market access regardless of a restructuring of Greek debt; if consensus is right on Spain having sounder fundamentals, a Greek restructuring would not tip Spain into losing market access.

Conversely, sovereign risk and “contagion” to sovereign spreads is the lowest for Italy among periphery EZ members—despite S&P’s recent decision to put Italy on negative watch—as the fiscal flow deficit is low and falling. Indeed, Italy allowed only automatic stabilizers to kick in during the global recession, thus limiting the increase in its deficit to 5% of GDP, and has now taken action to reduce it further; also, it reduced its long-term fiscal liabilities through pension reforms—such as automatic indexation of the retirement age to longevity—that significantly reduced long-term contingent public liabilities; while sound regulation and supervision by the Bank of Italy of the banking system prevented housing bubbles, the toxic underwriting of mortgages and loans and the purchase of toxic foreign assets; thus, the fiscal cost of “bailing out” Italian banks has been non-existent, apart from short-term liquidity support.

So, the decoupling of Italian sovereign spreads from the rest of the PIIGS and the ongoing market access for Italian banks is justified by Italy’s sounder fiscal and financial/banking fundamentals. Thus, as long as Italy maintains sound fiscal and financial policies, the risk of contagion from a Greek debt restructuring will remain low. If policies in Italy and Spain are sound, a Greek restructuring could possibly increase their sovereign spreads and lead some of their banks to a degree of greater reliance on ECB financing. But the risk that sovereign spreads increase so much that the countries end up losing market access and requiring an IMF-EU bailout are extremely low.

These arguments hold for Belgium as well: Whether Belgium and its banks lose market access depends more on factors specific to the country—its fiscal policies and the soundness of its banks—than on the consequences of a possible Greek debt restructuring. On the positive side, Belgium has relatively low fiscal deficits and high private savings; on the negative side, it has a large stock of public debt, possibly larger bank losses than has been recognized in public and a serious political gridlock problem that could (under an extreme scenario) even lead to its break-up. These factors matter much more than what happens in Greece in terms of determining whether Belgium is at risk of losing market access.

Risk of Global Contagion—Outside the EZ—to Global Credit and Equity Markets

The final contagion risks to be addressed are that a Greek debt restructuring could lead to contagion to global financial markets, either in the form a sharp rise in sovereign spreads in emerging markets (EMs) and advanced economies outside the EZ; or in the form of a large correction of global equity markets.

The first risk—contagion to sovereign spreads outside the EZ—is really far-fetched. Such spreads will depend mostly on fiscal conditions in these extra-EZ EMs and advanced economies. Spreads in EMs are low and have shown significant decoupling from trouble in the EZ periphery as most EMs have sound fiscal and financial conditions. And there is again no evidence of contagion from EZ sovereign spreads to sovereign yields in high-deficit advanced economies—such as the U.S., the UK or Japan—or to yields in advanced economies with sounder macro and fiscal fundamentals (the Nordics, Canada, Australia, etc.).

There have been episodes where debt restructurings and defaults have led to seizures in global credit markets, specifically after the Russian default and the Lehman collapse. But episodes where contagion was massive had two features: One, they were unexpected and surprising events when the market had underestimated the risk of default as there was a moral hazard play (expectations of bailout that were shattered); two, they resulted in actual disorderly default rather than orderly debt restructuring. In other episodes, where the default was highly predicted in advance by investors (Argentina and Ecuador) or where the debt restructuring was orderly via a par bond and prevented a formal default (Pakistan, Ukraine, Uruguay, etc.), the contagion was very limited, if non-existent. In Greece, the risk of a debt restructuring has been priced in by investors for a very long time and a moral hazard play is non-existent as spreads are already pricing in a very high likelihood of a debt restructuring; and whatever happens in Greece, we will not have a disorderly default; at worst, we may have an orderly and market-oriented debt exchange instead of a formal disruptive default. Thus, the risk of a Russian/Lehman-style contagion to either EZ or extra-EZ debt markets following a restructuring in Greece is extremely low.

What about the risk that a restructuring of Greek debt would lead to a sharp correction of global equity markets that would then trigger a sharp slowdown of global growth? Those concerned about this risk point out that in spring/summer 2010 Greece’s troubles led to a global equity market correction of almost 20% that enveloped not only Greece and the EZ, but also the U.S., other advanced economies and EMs. This comparison between equity markets in 2010 and what would happen to global equity markets following an orderly restructuring of Greek debt is inappropriate. The chance of a massive global equity market correction is low. First, what led to the contagion in 2010 was the five months that were lost—between January and May—to decide whether Greece needed help, whether the IMF should be involved in helping Greece or not and whether the Germans would accept a bailout of Greece given their own electoral constraints. The risk of a disorderly default by Greece and even of a disorderly break-up of the EZ reached a peak in May 2010 before the decision was finally taken to create the European Financial Stabilization Mechanism and the EFSF and the ECB was persuaded to provide massive liquidity support to banks and to purchase government debt. The further correction in global equity markets in the summer was associated with lingering doubts about the design of the EFSF, the saga of inter-ECB in-fighting about purchasing long-term government bonds and the sharp slowdown in U.S. and advanced economies—partly associated with the end of QE1 and market expectations of early exit from zero policy rates—that led first to worries of a soft patch, then to worries of a stall speed and finally to worries of a double-dip recession. These concerns were contained only once the Fed signaled QE2 in August, thus triggering another cycle of asset reflation.

So, Greece did not cause massive global contagion to global equity markets in 2010; much more was going on that triggered that equity market correction. And indeed, once the EFSF was in place, the loss of market access by Ireland and then Spain did not lead to a global equity market correction as—after the bailout of Greece—the risk of a disorderly EZ break-up was sharply reduced and the risk that a country losing market access would default was also reduced. Indeed, the Greek bailout showed global investors what would happen if Ireland and Portugal were to lose market access (as, of course, they eventually did): The provision of massive liquidity support to sovereigns and their banks conditional on fiscal austerity, structural reforms and, if necessary, privatizations. Indeed, global equity markets have mostly shrugged off the loss of market access of Ireland and Portugal and even the most recent correction of global equity markets appears to be associated more with renewed concerns about a slowdown of U.S./global growth together with sharp increases in oil prices than with the risks of a Greek debt restructuring. Further correction in global equity markets—so far modest—may or may not occur depending on the outlook for corporate earnings, profit margins and the growth outlook for the U.S., China and other advanced economies and EMs. But it is highly unlikely to be dependent—to an extent that is meaningful—on whether Greece restructures its debt or not.

It is indeed our view that each country/sector/asset will be judged on its own merits rather than its relationship (economic linkage) with Greece and thus be mostly insulated from the volatility surrounding a “treatment” of Greek debt. One can certainly argue that we should allow for several factors coming together rather than taking a ceteris paribus approach. Indeed, in 2010, a series of developments coincident with the unfolding of Greece concerns triggered a market sell-off; one cannot rule out a similar confluence of events in the event of a rescheduling of Greek debt, especially now that markets are moving from risk-on to risk-off with the clouds gathering—starting with evidence of a global economic slowdown—over the global economy and financial markets.

Subject to the above caveat, our analysis suggests that the risks of contagion to global credit and equity markets from a Greek debt restructuring are likely to be very small.

Conclusion: Much Ado About Almost Nothing

The ECB and other public and private observers in Europe have raised the specter that a debt restructuring in Greece—even an orderly and market-oriented one, or indeed a soft restructuring in the form of the re-profiling of the country’s debt—would lead to massive and destructive contagion to financial markets, banks and financial institutions and other EZ sovereigns that would in turn lead to financial Armageddon. Those fears have now led to hysterical warnings that have increased in volume, despite their empirical basis being extremely weak. The ECB and other private and public sector agents have not presented publicly any serious analysis of the possible channels of contagion (or of their probability and likelihood), historical examples of contagion (or examples of when contagion did not occur) or of ways to manage, control and prevent contagion. The official mantra has become the repeated and increasing shrill statements—without any factual or analytical basis—that any type of restructuring of Greek debt would lead to utter financial disaster: Bank runs all over Europe, bank collapses, sovereign contagion, domino effects and the loss of market access for many other EZ sovereigns. So, either the official sector can back up its claims with serious analysis; or it contributes to market turmoil and volatility and uncertainty rather than constructively helping to prevent it.

The serious and detailed analysis of the possible channels of contagion in this paper suggests the exact opposite: An orderly restructuring of debt—which is feasible as it has been implemented in countless other episodes of sovereign distress—is not likely to lead to meaningful distress and contagion for banks and sovereigns within and outside the EZ. And any possible form of contagion can be constructively managed, contained and/or ring-fenced with appropriate policies and liquidity support, which is feasible and available as needed. So, if the main argument against an orderly restructuring of Greek debt is the contagion risk, a calm analysis of the data and of the concepts suggests that such concerns are vastly exaggerated and that the risks are manageable and preventable.

ECB and other critics of an orderly debt restructuring are not only wrong; it seems that they have not even done their homework. And their policy threats are risky and dangerous: Rather than a reasoned discussion of the specific risks of contagion and how to manage them, they actually risk damaging the prospects of an orderly restructuring and causing panic and market turmoil. The policy of kicking the can down the road and throwing more good more after bad money—ramping up the amount of official support for bailouts that are clearly failing rather than implementing orderly restructuring—is bound to lead to disorderly restructuring and massive losses for private investors once most Greek debt ends up in official hands and a brutal haircut for the remaining private claims becomes necessary and unavoidable. So, the current blind policy that tries to prevent a restructuring at any cost under the excuse of a risk of short-term contagion risks, in turn, creating much larger contagion and a disorderly workout when the option of an orderly one has been lost.

Indeed, the comparison between Argentina and Uruguay shows what happens when policy makers stick their heads in the sand and postpone orderly restructuring: The end result is disorderly and massively contagious defaults. Instead, when pre-emptive and orderly restructuring of unsustainable debt is implemented—as happened when the painful lesson of Argentina was learned in Uruguay—there was no contagion, no disorderly defaults, no bank run and no toxic fallout from an orderly and market-oriented restructuring of debts.

Hopefully, the ECB and other policy makers in Europe will start reading their history books rather than repeating unfounded statements and damaging the chances of an orderly outcome for Greece and other EZ economies in distress. A serious discussion about contagion risks, what kind of contagion is likely and whether it is preventable is perfectly legitimate, necessary and useful; while using the bugaboo of contagion for scare mongering is dangerous and eventually could seriously damage the chance of a constructive policy outcome to difficult economic and financial challenges for many distressed economies in the eurozone.

15 Responses to “Much Ado About Relatively Little: Contagion Risks From an Orderly Greek Debt Restructuring Are Modest, Contained and Manageable”

miquelroigMay 25th, 2011 at 9:26 pm

So you say that "if policies are credible and sound, the risk that an economy like Spain would lose market access is extremely low". But then, you add "If we are right, then Spain is likely to lose market access regardless of a restructuring of Greek debt". So, I assume that your source of pessimism about Spain is not just economic fundamentals, but the unability of its Government (current and future) to implement sound policies.

miquelroigMay 25th, 2011 at 9:57 pm

Thanks for your reply. But I still not get one thing: Can Spain avoid losing market access by executing sound economic policies or is it too late?

Nouriel_RoubiniMay 25th, 2011 at 10:09 pm

not too late but they will have to work hard to improve their fundamentals, fiscal austerity, structural reform, etc…

GobanianMay 26th, 2011 at 9:10 am

The real danger of the ECB approach is that it is increasing the danger of a chaotic default, which would have very serious consequences.
It is obvious that the ECB approach of denying solvency problems was flawed from the outset. It allowed the ECB to feel satisfied that it was not engaging in the sort of special measures done by the Fed, but it simply postponed the problems,
Now the ECB finds itself in a situation where very real questions should be asked about it prudential management of its own balance sheet. I think that avoiding those questions has, as you suggest, distorted the role it has played in the crisis. It has ceased to act as an honest broker.

OctavioRichettaMay 26th, 2011 at 10:48 pm

Professor, how does Europe´s Sovereign Debt crisis compare to the US so called sub-prime/securitization crisis – which turned out to be a lot more than just sub-prime?

It seems to me that even if an optimal approach to handling the European mess is implemented from now on, a huge destruction of sovereign debt principal will have to take place. There is no way of plugging the black whole left by bubble implosions.

OcjooMay 27th, 2011 at 1:38 am

Whether it’s much ado over nothing or something, letting the Greeks technically default on their debt and restructure it is the only realistic way to rescue Greece and stabilize the situation. Keep in mind that a decade ago, Euro was trading at par with USD. Recently, after some weakening, Euro is buying USD 1.4 – good luck to those who still believe the Greeks can pay their way out of debt.

SANTOSH PANDITMay 27th, 2011 at 9:13 am

Why was Greece not allowed to default right in the beginning? By providing half-hearted support to Greece, the ECB and IMF have wasted their money.

Even Grandma knows throwing good money after bad is not good. Unfortunately celebrity economists lack common wisdom.

GKrepsMay 27th, 2011 at 3:42 pm

As an early investor in Brady Bonds of near two decades ago I've had some experience with these restructurings or re-profilings (however you want to characterize them). Nouriel's analysis seems right on the mark. It should be pointed out that the success of Brady Bonds was very much tied to the principal and rolling interest collateral. This could obviously be replicated by the ECB or whatever facility the Europeans designate as a guarantor.

nasMay 28th, 2011 at 4:08 pm

In my humble opinion, and as I have written over 1 year ago, there must be meaniful structural changes to the ECB. I would think and hope that as the big banks in France and especially Germany slowly deleverage out of the Euro Zone and it's debt, and as China likely continues to implement a slow diversification of it's sovereign wealth from US dollar to Euros, necessary ECB reform will become more of a reality.

I think the biggest threat to the Euro is wether it can sufficiently implement enough debt restructuring, institutional changes to the EU governing body and especially structural changes to the ECB before the next global financial & economic crisis / recession hits ( later rather than sooner for all of our sakes. )

Rest assured that the next global recession is likely to hit us all sooner that the historical average for recessions. If the EU can not get it's house in order within the next 3 years or so, it will be tempting the possibility of the collapse of the Euro and the European Union itself. But I believe that a three year window of time is sufficient, as long as there are no large, global, unforseen financial & economic crises in the meantime. One of the first tests we will all face will be how the U.S. Fed will handle the challenge of deleveraging the toxic U.S. mortages off of its balance sheet.

guapsterMay 29th, 2011 at 1:24 pm

What is a "repo financing facility?"

Great article! In the past Prof. Roubini, I believe, has indicated that long term solutions require structural reforms, otherwise, a country may just be "kicking the can down the road." What structural reforms should Greece undertake? What needs to be done from the point of view of an ordinary Greek citizen?

Also, I recently had a conversation with Dutch tourists on the New York subway in which they objected to what they described as "handouts" for the Greeks because the Greeks enjoyed richer social services, better pensions and a much lower retirement age for civil service workers than they enjoyed. They felt that if they had to work much longer to enjoy a government provided retirement benefit, so should the Greeks before asking Dutch citizens to help them out. Was their complaint unfounded?

Health InsuranceJune 17th, 2011 at 5:05 am

Hiya, I am really glad I’ve found this information. Nowadays bloggers publish only about gossip and net stuff and this is really annoying. A good site with exciting content, that’s what I need. Thank you for making this web-site, and I’ll be visiting again. Do you do newsletters? I Cant find it.