EZ Scenario Analysis Update: The Next Storm
By Elisa Parisi-Capone, Nouriel Roubini and David Nowakowski
Aug 8, 2011 5:45:00 PM | Last Updated
Aug 8, 2011 5:45:00 PM | Last Updated
The ECB’s extension of the Securities Markets Program (SMP) to include Italian and Spanish securities—backed by a front-loaded fiscal and structural reform commitment from those countries, as well as a renewed pledge from Franco-German and G7 policy makers to safeguard the stability of the eurozone (EZ) and U.S. financial systems—was indispensable to prevent Italy and Spain from eventually losing market access.
However, ECB bond purchases, even if done on a large scale, provide but short-term relief.
In view of controversial fiscal integration choices with respect to the long-term stabilization of the EMU framework, a series of sequential public debt reschedulings with bail-ins, potentially even in large countries, is the most likely outcome over the next couple of years. This may be followed by selective exits from the EMU, with the departures of Greece and Portugal most likely over the next five years.
Large-scale asset purchases could amount to around 10% of GDP in both Italy and Spain, or €155 billion and €106 billion, respectively, based on the current SMP composition of Irish and Portuguese debt (Greek bonds amount to 20% of GDP).
Given the comparatively better fundamentals of Italy and Spain, such liquidity extension could have a better chance of success than in other recipient countries. Alternative benchmarks include the flow of maturing sovereign bonds to refinance over the remainder of the year (about €200 billion for both Italy and Spain).
However, the lack of unanimous support from ECB governing council members for SMP extension has undermined its effectiveness from the start. Large-scale intervention or quantitative easing (QE)—even on an unsterilized basis—is not only desirable for financial stability, but also in view of the rapidly deteriorating growth outlook in the EZ and globally.
At the same time, the credible ring-fencing of Spain and Italy via the European Financial Stability Facility (EFSF) requires a multiple of the agreed (although not yet ratified) €440 billion lending capacity. Even if policy makers were to follow the ECB and European Commission recommendations to double or treble the size of the EFSF, governance and size limitations make it an imperfect buyer-of-last-resort substitute to the ECB.
In addition, such a size extension would overwhelm the capacity of triple-A countries in terms of contingent liabilities if large countries were to step out of their commitments. This may be a sacrifice that creditor countries need to make, but we don’t assign the highest probability to this outcome. We see an important role for the EFSF in forcing bank recapitalizations à la troubled asset relief program (TARP), to sever the bank-sovereign feedback loop.
Other policy alternatives to sequential bailouts or bail-ins—e.g. “eurotaxes,” Eurobonds and temporary devaluations—are politically more difficult to implement since they lack binding sanction and incentive mechanisms from the side of creditor countries, as opposed to EFSF loans, which are subject to an IMF/EU surveillance program or other conditionality. Policy makers are aware of the need to move toward a closer political union, but that remains a long-term project.
Crucially, the restoration of fiscal sustainability via austerity or restructuring is no substitute for the restoration of growth and competitiveness, which requires internal devaluations, structural reforms, a deflationary spell and/or a much weaker common currency.
All of these will take a long time to implement and/or become effective and are fraught with short-term adverse side effects. Hence, we see a significant chance of Greece and Portugal exiting the EZ over a five-year horizon.
Illiquidity Turns Into Insolvency if Unaddressed—Italy, Spain Were on the Way to Losing Market Access Fast
Decision time for deeper integration versus disintegration has arrived earlier than we envisaged in our “EZ Does It! Part I: Eurozone Endgame Scenarios” paper in September 2010.
While, at the time, we viewed Spain as the main short-term systemic risk for the EZ up to 2013, the mix of idiosyncratic and systemic disruptions engulfing Italy since July has accelerated events dramatically. The main trigger points investors are watching are a rise in 10y bond yields above the 7% threshold that in the past has signaled loss of market access; a rise in the spread to triple-A securities above 450 bps that would increase by 15% margin calls on sovereign securities by LCH.Clearnet; and the refinancing calendar of sovereigns and banks at risk (Figures 1-3). These adverse sudden stop dynamics were triggered by the lowering of rating outlooks, collapsing GDP growth indicators and insufficient backup to credibly shield large countries from contagion.
The recourse to private-sector involvement in the case of Greece has also set a precedent that has led to a general repricing of risk among EZ sovereigns and banks.
Figure 1: 10y Sovereign Bond Yields (%)
Source: Bloomberg
Figure 2: Italy and Spain Bank and Sovereign Maturities (€, billions)
Source: Bloomberg
Figure 3: Total Maturing Bank and Sovereign Bonds (principal only; €, billions)
Source: Bloomberg. Note: Bank data includes a subset of the four to five largest banks by assets.
In Figure 4, we look at the available policy options in the short to medium term and assign probabilities to each.
RGE Base-Case Scenario: Sequential Reschedulings With Private-Sector Bail-Ins as Default Option
In RGE’s view, the ECB’s extension of the SMP to include Italian and Spanish securities—backed by a front-loaded fiscal and structural reform commitment from those countries, as well as a renewed pledge from Franco-German and G7 policy makers to safeguard the stability of the financial system in the eurozone and across the Atlantic—is an indispensable measure to prevent Italy and Spain from losing market access.
Figure 1: 10y Sovereign Bond Yields (%)
Source: Bloomberg
Figure 2: Italy and Spain Bank and Sovereign Maturities (€, billions)
Source: Bloomberg
Figure 3: Total Maturing Bank and Sovereign Bonds (principal only; €, billions)
Source: Bloomberg. Note: Bank data includes a subset of the four to five largest banks by assets.
In Figure 4, we look at the available policy options in the short to medium term and assign probabilities to each.
RGE Base-Case Scenario: Sequential Reschedulings With Private-Sector Bail-Ins as Default Option
In RGE’s view, the ECB’s extension of the SMP to include Italian and Spanish securities—backed by a front-loaded fiscal and structural reform commitment from those countries, as well as a renewed pledge from Franco-German and G7 policy makers to safeguard the stability of the financial system in the eurozone and across the Atlantic—is an indispensable measure to prevent Italy and Spain from losing market access.
However, ECB bond purchases even if done on a large scale provide but short-term relief. In view of difficult policy choices with respect to the long-term stabilization of the EMU framework, a series of sequential public debt reschedulings with bail-ins, potentially even in large countries, is the most likely outcome over the next couple of years. This may be followed by selective exits from the EMU, with the departures of Greece and Portugal most likely over the next five years.
Figure 4: Scenario Analysis With RGE Probabilities
Note: These scenarios are not mutually exclusive, so the sum of the probabilities exceeds 100%. Consistent with our previous scenario analysis, green indicates deeper integration, yellow indicates muddle-through and red indicates breakdown; orange indicates that "More PSI in Greece and Beyond" could lead to either deeper integration or breakdown.
Large-Scale ECB Intervention—Only a Short-Term Solution to Prevent Loss of Market Access
Taking current SMP purchases as a percentage of GDP for Ireland and Portugal as a benchmark, large-scale asset purchases could amount to around 10% of GDP in both Italy and Spain, or €155 billion and €106 billion, respectively (asset purchases of Greek debt amount to about 20% of GDP).
Figure 4: Scenario Analysis With RGE Probabilities
Note: These scenarios are not mutually exclusive, so the sum of the probabilities exceeds 100%. Consistent with our previous scenario analysis, green indicates deeper integration, yellow indicates muddle-through and red indicates breakdown; orange indicates that "More PSI in Greece and Beyond" could lead to either deeper integration or breakdown.
Large-Scale ECB Intervention—Only a Short-Term Solution to Prevent Loss of Market Access
Taking current SMP purchases as a percentage of GDP for Ireland and Portugal as a benchmark, large-scale asset purchases could amount to around 10% of GDP in both Italy and Spain, or €155 billion and €106 billion, respectively (asset purchases of Greek debt amount to about 20% of GDP).
A case could be made that given the comparatively better fundamentals of Italy and Spain, such liquidity extension could have a better chance of success than in other recipient countries. Alternative benchmarks include a fraction of the flow of maturing sovereign bonds to refinance over the remainder of the year (about €200 billion for both Italy and Spain).
However, ECB governing council members apparently have not unanimously supported the extension of the SMP, thus undermining its effectiveness from the start. In our view, large-scale intervention or QE—even on an unsterilized basis—is not only desirable from the perspective of financial stability, but also from the perspective of monetary policy, given the rapidly deteriorating growth outlook in the EZ and globally.
But the ECB is likely to resist such “large” purchases as opposed to a more limited bond-purchase program: It has internal—German—resistance to large-scale action; it has argued that fiscal problems need to be addressed with fiscal tools (the EFSF); it cannot really sterilize the potentially hundreds of billions of euro of bond purchases necessary to backstop Italy and Spain; it is not willing to monetize deficits via large unsterilized purchases; it sees its role as a temporary bridge to the time when EFSF reform is approved; and it is concerned about moral hazard and about losing its monetary independence.
On a more fundamental level, the ECB’s lender-of-last-resort role (not only for banks, but also for sovereigns, without which members of a currency union are subject to sudden stops in a self-fulfilling manner), would in principle need to be legitimized in the Treaty on par with the price-stability mandate.
The primary focus on the latter, even in times of crisis, leads to a measurable tightening bias (Figure 5), which is detrimental from the perspective of financial stability and which opens the door to policy mistakes that subsequently need to be reversed.
That said, the current political constellation does not favor a permanent redefinition of the ECB’s role, which is why we assign a diminishing probability over time to the option of large-scale purchases of Spanish and Italian bonds.
Figure 5: Eurosystem Base Money (€, billions)
Source: ECB
Double/Treble EFSF Size/Scope
EU leaders want to fast-track ratification of the revamped EFSF legislation by the end of September (as announced by both Germany and France on August 7) and as quickly as possible in all EZ countries. This includes not only the effective capacity expansion to €440 billion agreed in June, but also the new tools approved on July 21 (pre-emptive loans, bank recapitalizations and secondary market purchases on the basis of ECB approval.)
As we have noted elsewhere, since the expansion of the size and scope of the EFSF requires the unanimous agreement of national governments, and in most cases also of national parliaments, the question arises: How difficult will the ratification process be in anti-bailout creditor countries such as Finland, Germany, the Netherlands and Slovakia? And what if a large guarantor opts out of its guarantee commitment?
Figure 5: Eurosystem Base Money (€, billions)
Source: ECB
Double/Treble EFSF Size/Scope
EU leaders want to fast-track ratification of the revamped EFSF legislation by the end of September (as announced by both Germany and France on August 7) and as quickly as possible in all EZ countries. This includes not only the effective capacity expansion to €440 billion agreed in June, but also the new tools approved on July 21 (pre-emptive loans, bank recapitalizations and secondary market purchases on the basis of ECB approval.)
As we have noted elsewhere, since the expansion of the size and scope of the EFSF requires the unanimous agreement of national governments, and in most cases also of national parliaments, the question arises: How difficult will the ratification process be in anti-bailout creditor countries such as Finland, Germany, the Netherlands and Slovakia? And what if a large guarantor opts out of its guarantee commitment?
Our short answer is that the potential opt-out of a small country like Slovakia is surmountable, while the backing of large and triple-A countries is indispensable.
In Germany, the latest signs indicate that there is enough support within the coalition to pass the package in September despite dissenting voices, especially among the junior coalition partner, the FDP. And even if there isn’t, the opposition has expressed support for the deal (although that would put the coalition at risk).
While the Bundestag opposed secondary market purchases by the EFSF, the private-sector involvement in Greece is regarded as an important achievement.
Similarly, the Netherlands has insisted on private-sector participation even if the consequence were to be a selective default—a demand that has been met.
And Article 9 of the July 21 statement meets Finland’s collateral and burden-sharing requirements for future EFSF participation.
In Slovakia, where the junior coalition partner opposes the deal, EFSF/ESM approval will rely on the pro-EZ opposition, which puts the government’s small majority at risk.
Meanwhile, rising borrowing costs in Spain and Italy put their effective participation in EFSF loan disbursements at risk; if that comes to pass, it would trigger large contingent liabilities in the books of the triple-A countries that would have to cover the Spanish and Italian shares and a renewed wave of contagion to the core, including potential downgrades (France is particularly vulnerable despite the rating agencies’ recent confirmation of its triple-A status with stable outlook).
Meanwhile, rising borrowing costs in Spain and Italy put their effective participation in EFSF loan disbursements at risk; if that comes to pass, it would trigger large contingent liabilities in the books of the triple-A countries that would have to cover the Spanish and Italian shares and a renewed wave of contagion to the core, including potential downgrades (France is particularly vulnerable despite the rating agencies’ recent confirmation of its triple-A status with stable outlook).
The nature of the EFSF as a collateralized debt obligation or special investment vehicle—where a group of many less-than-triple-A distressed sovereigns pool guarantees to obtain a triple-A rating for the debt issued—is increasingly becoming clear.
That is why Germany is sharply resisting an increase in the size of the EFSF: Eventually its own triple-A rating may be put at risk if the EFSF becomes too large and the German taxpayer ends up backstopping the debt of the entire periphery (as well as Germany’s own).
In view of the ongoing contagion to larger countries like Spain and Italy, EU leaders could eventually decide to leave the size of the EFSF open to review (as was decided for its post-2013 successor, the European Stabilization Mechanism, ESM), although in our view a credible backstop of large countries like Italy would overwhelm the triple-A capacity of creditor countries.
Our calculations show that, while the revamped EFSF would be able to backstop Spanish financing needs for three years, propping up Italy would require resources of about €1.5 trillion.
In this regard, creditor countries like Germany and the Netherlands have been swift in voicing their opposition to the reopening of EFSF negotiations promoted by the ECB and the European Commission so soon after agreeing on the size and scope expansion in June and July.
This leaves either the ECB as a very reluctant lender of last resort or the introduction of a common Eurobond-like instrument, for which there are several proposals circulating (see section below).
This leaves either the ECB as a very reluctant lender of last resort or the introduction of a common Eurobond-like instrument, for which there are several proposals circulating (see section below).
We also note that governance and size limitations make the EFSF inherently a less flexible and therefore imperfect buyer-of-last-resort substitute for the ECB. One area where we see an important role for the EFSF is, for example, a TARP-like recapitalization of EZ banks in order to break the sovereign-banking feedback loop, including in creditor countries. The initiative to convert the EFSF and ESM into a prefunded “European Monetary Fund” or a fledgling fiscal authority is also positive in our view, although geared toward the longer term.
In the Greek bail-in, a small amount of EFSF resources will be used to help buy back bonds. To accomplish this, the EFSF will be lending long-term to Greece at around 3.5% for the government to buy back its bonds, probably at yields north of 12%. This is one of the only meaningful debt-reduction components in the program, but it is quite small, and there are the usual objections to buybacks being an effective way to generate debt relief (see also Bulow and Rogoff).
Still, a greatly enlarged EFSF could be used to effectively cap Italian or Spanish yields by credibly committing to buy any long-term bonds at, say, swaps plus 100 bps. This is probably superior to conducting mega-exchanges, but involves risk-pooling similar to the creation of Eurobonds.
Eurobonds
There are several proposals circulating (see, for example, Steinmeier and Steinbrück, Juncker and Tremonti, Amato and Verhofstadt, Lorenzo Bini Smaghi and the Levy Institute) for common Eurobonds, including those with or without fiscal transfers or joint and several liabilities.
In its basic form (Delpla and von Weizsaecker, Paul de Grauwe and Moesen), the idea is to transfer a share of national debt issuance, say 40-60% of GDP, to the EU level while the remainder remains at the national level, subject to idiosyncratic market and credit risk.
To minimize the negative spillover to fiscally sound countries cross-subsidizing the others, different haircuts or participation fees would apply according to a countries’ creditworthiness at the time of conversion. The main benefit is that, over time, a very liquid market akin to U.S. Treasurys could be created, which would also be easier for the ECB to deal with as an EZ-wide benchmark.
One drawback is the potentially higher borrowing costs for sound creditor countries.
One drawback is the potentially higher borrowing costs for sound creditor countries.
Another is that, unlike EFSF loans, once national bonds are converted into Eurobonds, creditor countries have no policy leverage over fiscally profligate countries to ensure responsible behavior unless the institution of Eurobonds comes with a significant loss of sovereignty for many countries.
The wider the gap between fiscally sound and challenged countries, the more difficult it is to reconcile the interests of these stakeholders.
The current constellation of policy makers is therefore not on board with the idea, but some opposition parties (including in creditor countries) have voiced support for this form of fiscal integration, if it comes with sufficient conditions and fiscal guarantees at the national level.
We assign a cautiously optimistic probability to this scenario materializing in the longer term, but not the highest odds in the short run.
Exit of Greece and, Eventually, Portugal
Exit of Greece and, Eventually, Portugal
RGE Chairman Nouriel Roubini cautions that, unless the EMU moves toward a broader fiscal, economic and political union that resolves the fundamental problems of divergence (economic, fiscal and in terms of competitiveness) within the union, the system will move first toward disorderly debt workouts and then, eventually, even break-up, with weaker members departing.
Over a five-year horizon, the odds of a break-up are at least one-third, and will rise rapidly if the underlying problems—not just the symptoms—are not addressed. Based on macroeconomic fundamentals, we see Greece and Portugal as the highest-probability candidates for an exit from the EMU within the next five years.
As we have discussed elsewhere, Mario Blejer and Eduardo Levy Yeyati point to the adverse consequences of exit and devaluation as a cure-all, based on Argentina’s experience.
As we have discussed elsewhere, Mario Blejer and Eduardo Levy Yeyati point to the adverse consequences of exit and devaluation as a cure-all, based on Argentina’s experience.
Judging from that example, in which the national currency remained in circulation, exit costs include the following: first, the “drachmatization” of contracts with asymmetric balance-sheet effects; second, heavy restrictions on banks including deposit freezes; third, an external debt restructuring; fourth, capital and exchange controls; and fifth, inflationary currency collapse and partial debt monetization. These costs must be tackled jointly and up front or they are bound to increase.
Bear in mind also that capital and exchange controls are not compatible with EU rules and that, from a purely legal perspective, withdrawal from the EMU without a parallel withdrawal from the EU would therefore be legally inconceivable.
Bear in mind also that capital and exchange controls are not compatible with EU rules and that, from a purely legal perspective, withdrawal from the EMU without a parallel withdrawal from the EU would therefore be legally inconceivable.
However, since the Treaty does not foresee an exit from the EMU, on legal grounds only an agreed exit (including with the ECB) from the EZ is conceivable, although in practice there is no enforceable recourse against sovereign states that choose to follow their own course of action.
Other Options
Other very low-probability options not considered in our scenario analysis include eurotaxes and temporary devaluations:
1. Eurotaxes Instead of Transfers
One way to circumvent the moral hazard problem and the implicit transfers involved with joint guarantees such as the EFSF is the levying of a set of “eurotaxes,” as for example proposed by Zingales/Perotti. These include eurotaxes on bank debt, a sovereign debt tax or a value-added tax contribution commensurate with each country’s GDP to back Eurobonds. The main drawback of these provisions is the already high tax burden in Europe and the fact that such a procyclical tax puts additional strain on banks and credit provision.
One form of EU-wide countercyclical automatic stabilizer would be a common unemployment insurance scheme or an EZ-wide deposit protection guarantee, backed, for example, by a cross-border bank systemic risk levy. However, these aren’t currently on the agenda and we don’t expect them to be anytime soon.
One way to circumvent the moral hazard problem and the implicit transfers involved with joint guarantees such as the EFSF is the levying of a set of “eurotaxes,” as for example proposed by Zingales/Perotti. These include eurotaxes on bank debt, a sovereign debt tax or a value-added tax contribution commensurate with each country’s GDP to back Eurobonds. The main drawback of these provisions is the already high tax burden in Europe and the fact that such a procyclical tax puts additional strain on banks and credit provision.
One form of EU-wide countercyclical automatic stabilizer would be a common unemployment insurance scheme or an EZ-wide deposit protection guarantee, backed, for example, by a cross-border bank systemic risk levy. However, these aren’t currently on the agenda and we don’t expect them to be anytime soon.
2. Temporary Devaluation
As we noted in “Second Bailout Package for Greece and Stabilization of the EMU Framework—Options and Solutions,” one proposal (promoted by Charles Goodhart and Dimitrios Tsomocos and Martin Feldstein in early 2010) is for an “EZ holiday” where, according to Feldstein, “the rest of the eurozone could allow Greece to take a temporary leave of absence with the right and the obligation to return at a more competitive exchange rate”—say at 1.3 drachmas per euro. In practice, “Greece would shift its currency from the euro to the drachma, with an initial exchange rate of one euro to one drachma. Bank balances and obligations would remain in euros. Wages and prices would be set in drachma.” Feldstein argued that allowing Greece to repeg at a more competitive exchange rate is preferable than a decade of internal devaluation or permanent bailouts from creditor countries.
The counterarguments to this proposal (see, for example, Richard Baldwin and Charles Wyplosz) are that, first, it is not only impractical since the issuance of a new currency takes months, but, second, that it is also time inconsistent—who guarantees that parity will eventually be restored?
The counterarguments to this proposal (see, for example, Richard Baldwin and Charles Wyplosz) are that, first, it is not only impractical since the issuance of a new currency takes months, but, second, that it is also time inconsistent—who guarantees that parity will eventually be restored?
RGE’s view is in line with this counterargument, in that the introduction of parallel currencies or IOUs would serve to ultimately undermine the credibility of the entire EZ project and would result in immediate and severe contagion. We don’t view this as a viable or a high-probability outcome.
Financial Institutions Strategist Jennifer Kapila, Senior Research Analyst James Mason and Head of Research Christian Menegatti contributed to this analysis.
