## Sources of Banking Stress July 2011–June 2012 (_cr12182)

## Source details

**Canonical URL:** [Sources of Banking Stress July 2011–June 2012 (_cr12182)](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2012/_cr12182.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2012/_cr12182.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2012/_cr12182.pdf.json)

---

### A. Overview and IMF position
- Immediate priority: creation of a common Eurozone financial stability architecture (“Banking Union”) to restore viability of the Economic and Monetary Union.
- Banking Union three main elements:
  - a eurozone deposit guarantee scheme,
  - a eurozone resolution authority,
  - a eurozone supervisory authority.
- June 29 EU Summit step: proposals for a single supervisory mechanism and, contingent on implementation, direct recapitalization of banks by the ESM.
- IMF summarized position:
  - Centralized financial stability architecture needed to match rapid growth and integration of European banks.
  - Centralized framework envisaged: common resolution authority with a common backstop; common deposit guarantee scheme; common supervisory authority.

### B. Stages and transmission channels of the July 2011–June 2012 crisis
- Initial shock (Summer 2011):
  - Wholesale funding markets began to freeze; sovereign borrowing costs of Italy and Spain rose significantly.
  - Euribor-OIS spread “reached about half the level following the collapse of Lehman Brothers.”
  - From June 2011, US investors reassessed exposures to large euro area banks → sharp increase in cost of US$-euro basis swaps.
  - French banks lost about US$ 100 billion of funding from the 10 largest US MMF between June and December of 2011.
- Contagion across markets:
  - Good collateral became scarce as demand moved to secured markets; banks substituted repos for unsecured borrowing, raising demand for good collateral.
  - Worsening sovereign crisis eroded collateral value: government bond yields rose and margin requirements increased after downgrades.
  - ECB unconventional liquidity alleviated secured-market funding stress but “collateral used for repo financing still remains significantly more expensive in June 2012 than it was in July of 2011.”
- Deleveraging:
  - EBA data: large European banks sold off more than €80 billion of periphery sovereign bonds during January–September 2011.
  - Core euro area banks were the largest sellers of Italian and Spanish sovereign bonds in the EBA sample.
- ECB intervention:
  - LTROs and US$ swap lines stabilized funding markets and temporarily reduced sovereign stress.
  - Stabilization partly achieved by increasing sovereign exposures of Spanish and Italian banks, strengthening adverse sovereign-bank loops.
- Sovereign-bank nexus in Spain:
  - Government announced on March 02 that previously agreed SGP fiscal targets were not attainable.
  - Sovereign downgrades on April 23, June 07 and June 13.
  - Take-over and bailout of Bankia raised uncertainties about real estate losses and sovereign balance sheet impacts; group of banks were downgraded.
- Determinants of banking stress:
  - Before first LTRO: stress mainly explained by sovereign exposures to Italy or Spain; dependence on wholesale funding; large derivatives activity.
  - After first LTRO: Spanish and Italian banks with larger domestic sovereign exposures underperformed peers independently of sovereign bond yields or CDS spread movements; wholesale funding and derivatives reliance became less significant.
- Key referenced market indicators/events:
  - Dec 21st LTRO allotment; EURibor-OIS spread; 5-Year Bank CDS spreads; difference (3 months euro repo - Euribor); 5-Year Sovereign CDS spreads.
  - Data sources: Datastream; Bloomberg L.P.; IMF staff calculations; EBA July 2011 stress tests; EBA recapitalization exercise.

### C. Downward spirals at the national level and fiscal costs
- Main mechanism:
  - Large cross-border banks remain responsibility of home authority → contingent liabilities can jeopardize sovereign balance sheets and amplify sovereign-bank-real economy spirals.
- Fiscal cost examples (financial support impact on public debt since 2008):
  - Ireland: more than 40 percent of GDP.
  - Spain: about 20 percent of GDP.
- Empirical links:
  - Acharya et al. (2012): higher bank CDS spreads before a bailout resulted in higher government CDS spreads after bailouts.
  - Event study of Bankia bailout links takeover/recapitalization announcements to increases in sovereign borrowing costs.
- Regression findings on bank features amplifying stress:
  - Before LTRO (July–December 2011): larger weekly stock price declines associated with higher derivatives assets & liabilities ratio and higher net exposures to Italy and Spain.
  - After LTRO (January–June 2012): net exposures to Italy and Spain remained significant predictors of underperformance at the 75th and 90th percentiles; reliance on wholesale funding and derivatives activity were less significant.
- Additional statistics and indicators:
  - Large European banks sold more than €80 billion of periphery sovereign bonds (January–September 2011).
  - French banks lost about US$ 100 billion of funding from the 10 largest US MMF between June and December of 2011.
  - Financial sector support 2008–11 as percent of 2011 GDP (selected): Ireland 41.2; Spain 19.5 (includes actual use of debt guarantees, asset purchases and capital support from the FROB as of March 2012 and the ESM/EFSF loan announced on June 9th); Netherlands 14.1; Germany 12.2; United Kingdom 6.8; United States 5.3; Greece 6.1; Belgium 7.0.
- Conclusion:
  - National-based responsibility for financial stability amid large cross-border banking groups contributed to financial de-integration and stronger adverse sovereign-bank feedback loops → supports case for Banking Union with common backstops.

### D. Channels through which weak sovereigns damage domestic banks
- Direct channels:
  - Home bias → large exposures to domestic sovereign debt → sovereign risks directly reduce market value of bank assets.
- Indirect channels:
  - (i) loss of implicit/explicit sovereign guarantee;
  - (ii) reduced deposit confidence as national guarantee schemes (usually not pre-funded) lose credibility.
- Collateral and liquidity:
  - Decline in collateral value (yields up, higher margin requirements after downgrades) reduces ability of domestic banks to access secured funding or euro-system liquidity.

### E. Banking stress, lending rates, and macro effects
- Banking stress increases cost of credit by curtailing supply and raising lending rates to rebuild capital.
- Weak economy and fragile non-bank balance sheets: lower loan demand, higher non-performing loans, slower deposit growth → reduced bank profitability.
- Empirical estimate (panel regression, four largest euro area countries, 2010 to March 2012):
  - A 150 bps increase in bank CDS spreads → lending rates to non-financial corporations up by 0.6 percentage points.
  - A 150 bps increase in bank CDS spreads → mortgage lending rates up by 0.4 percentage points.
- Comparative 2010 observation:
  - Bank CDS spreads in Spain and Italy rose by between 150 and 200 bps while lending rates to non-financial corporations increased by respectively 0.5 and 0.8 percentage points.

### F. Financial fragmentation and market reversals
- Pre-crisis high integration: example Greek bund spreads between 01/2002 and 12/2007 oscillated between 8 and 37 bps.
- Withdrawal of intra-euro area capital during crisis:
  - Bond markets: withdrawal of intra-euro area cross-border capital of €240 billion in 2011 and Q1 of 2012 (a 24 percent reduction since end 2010).
  - Interbank markets: reduction of cross-border claims of €190 billion in Q4 2011 and Q1 2012.
  - Example reversal in non-resident holdings: end-2011 non-residents held 34 percent and 33 percent of Italy and Spain sovereign debt, down from 44 percent and 48 percent at end-2009.
- Fragmentation effects on lending and interbank markets:
  - Lending rate dispersion: spread to non-financial corporations between Spain and Germany rose from 6 bps in August 2008 to 43 bps in January 2011 and to 187 bps in April 2012.
  - EONIA overnight average daily volumes: declined to €20 billion at end-May 2012 from €40 billion at end-2010.
  - Overnight bank deposits at the ECB: rose to about €750 billion from less than €200 billion in June 2011.
  - Net borrowing from the eurosystem remained high for the 3 program countries and sharply increased for Italian and Spanish banks; Dutch and German banks accumulated large net claims on the eurosystem.
  - Insured deposits decline: cumulative decline since end-2009 reached €86 billion in the three program countries; in Spain, July 2011–April 2012 insured deposits (households and non-financial corporations) declined by €55 billion.

### G. Main elements and design features of a Eurozone Banking Union (policy recommendations)
- Purpose and benefits:
  - Delink financial intermediation from sovereign strength; end financial fragmentation; sever sovereign-bank feedback loops; restore financial integration; restart growth by ensuring healthy eurozone firms can obtain credit regardless of sovereign strength; stabilize financial systems and monetary policy transmission.
- Common deposit insurance scheme:
  - Collective and credible backing by all member countries.
  - Design suggestions:
    - Clear timetable for establishment.
    - Apply to all financial institutions to avoid deposit shifts.
    - Ultimately partially pre-funded by a levy on the industry.
    - Immediate effectiveness via access to additional funding: credit line from the euro-system or backstopped by common pool (e.g., ESM/EFSF) or limited joint and several guaranty bills.
    - Consistent with 2010 EU Commission proposal and Directive for recovery and resolution.
- Centralized bank resolution agency with common financing:
  - Centralized restructuring reduces sovereign exposure to bank failures.
  - Backed by a common fund financed by an industry levy and recourse to common government-provided resources or ECB liquidity backstop.
  - EFSF/ESM should be flexible to directly recapitalize banks.
  - Resolution powers consistent with EC harmonized resolution regime, including bail-in of unsecured creditors.
- Supra-national supervisory regime for large banks:
  - Align financing and monitoring responsibilities.
  - Several models:
    - ECB as eurozone supervisory authority with delegation to national entities, balancing LOLR role and monetary independence.
    - Two-tier system: ECB or other body supervises large/systemic banks; smaller local banks remain national.
- Scope and urgency:
  - Urgent for the euro area given capital flight consequences in a monetary union.
  - Deposit Guarantee Scheme covering all euro area depository institutions would reassure retail depositors.
  - Euro area bank resolution authority would help prevent bank-sovereign feedback loops.
  - Consider giving ECB explicit responsibility for financial stability and full lender of last resort functions to align roles and eliminate bank-sovereign linkages present in current ELA scheme.
  - Framework should allow opt-in by other EU countries and be consistent with EU-level frameworks and regulations.

*Prepared by Thierry Tressel (EURAE).*

---

### SMP interventions and subordination risk

### Background and key issues
- SMP effectiveness may have been reduced after the Greek PSI exemption and due to subordination risk from ECB treatment.
- ECB non-participation in Greek PSI (statement on July 22, 2011) reshaped seniority structure of official holdings and may have mispriced many bond markets.
- Dates/events:
  - Euro area summit 21 July 2011 (PSI for Greece first announced).
  - ECB President Trichet statement on July 22, 2011 that the Eurosystem would not participate.
  - ECB debt swap on February 16, 2012 examined as an event.

### SMP timing and scale (empirical observations)
- SMP purchases:
  - €36bn and €21bn in May-June 2010, then dropped.
  - Scaled up in November-December 2010.
  - Reactivated Aug–Sep 2011 with purchases of €51bn and €37bn.
  - Increased to €40bn in November 2011 before ECB adopted other measures.
- After PSI exemption, purchases largely put on hold:
  - Only one SMP intervention post-exemption in week of March 5-9, 2012 for €27 million (the fifth-smallest since May 2010).
- Conclusion: SMP helped temporarily reduce stress but limited scale/time horizon prevented full effectiveness.

### Empirical event-study findings
- Event windows around July 22, 2011 and Feb 16, 2012.
- Results:
  - Little impact of the ECB debt swap on periphery yields when evaluated around Feb 16, 2012.
  - The July 22, 2011 announcement that Eurosystem would not participate in Greek PSI had substantial short-term negative effects:
    - Longer-dated periphery bond yields and CDS premia rose unexpectedly after Trichet’s statement.
    - Yields and CDS premia had come down before July 21, 2011 summit but reversed after the statement.
  - Insufficient quantitative evidence to fully establish relation between subordination from ECB purchases and sovereign yields because ECB purchases were largely put on hold after the PSI exemption.

### Theoretical quantification — closed-form bond model (mechanism and examples)
- Key determinants of subordination impact: Probability of default (PD); Loss given default (LGD) = 1 − recovery rate; share of ECB bond holdings (b).
- Illustrative example:
  - Country debt trades at 60 cents to par before ECB senior creditor status.
  - Case interpretations:
    - Case I: 40 percent PD with 100 percent LGD (zero recovery) → seniority does not matter.
    - Case II: 80 percent PD with 50 percent recovery → seniority matters; if ECB share = 50 percent, private holders may be left with ex post price of 20 cents.
- Net impact depends on:
  - PD: subordination effect increases linearly with PD; larger b lowers ex post market price for given PD.
  - LGD: if large, seniority gives limited advantage; for small LGD, senior status less valuable; effect not monotonic.
  - b: when b exceeds original recovery rate a kink appears; effect monotonic but not linear; expectations about future b matter.

### Theoretical quantification — CDS reduced-form model
- CDS pricing: CDS = LGD x (1 – (1- PD)^{1/n}), n = number of periods.
- SMP liquidity effect reduces PD and can lower CDS spreads; subordination raises private-sector LGD and can offset PD decline.
- Model for LGD with restructuring needs and ECB share:
  - LGD = (ΔD/GDP)/(1- b), where ΔD/GDP = debt/GDP reduction needed.
  - Substitution yields: CDS = (ΔD/GDP)/(1- b) x (1 – (1- PD)^{1/n}).
- Implications:
  - Increasing b increases CDS spreads, more strongly when CDS are already high (PD elevated).
  - SMP credibility matters: subordination may lead to self-fulfilling default dynamics if central bank purchases cannot offset increased private LGD.

### Net effects and ambiguity
- SMP purchases have two opposing effects:
  - Increase subordination risk → expected transfer of value from private holders to ECB → may raise issuance costs for bonds not purchased by ECB.
  - Improve liquidity and reduce PD → increase value of residual bonds if intervention strategy is credible and well-communicated.
- Net impact is ex ante ambiguous and depends on PD, LGD, b, and credibility of interventions.

### Policy-relevant insight and empirical assessment
- Main finding: impact of ECB seniority primarily related to PD and ECB’s debt share (b); credibility of SMP interventions matters.
- Empirical assessment: SMP subordination “will be important only at the margin—as was the case in Greece—and depends on the program’s credibility.”
- SMP subordination “currently does not seem to play a large role in pricing and markets.”
- Policy option to attenuate fears: transfer some benefits for the ECB back to private sector bond holders.

---

### Fiscal integration, limited common debt, and macro simulations

### Rationale for fiscal integration and intermediate steps
- More fiscal integration with stronger governance and risk sharing reduces threat that shocks in one country endanger entire euro area.
- Limited, scalable common debt with governance safeguards could support Banking Union and signal commitment to fuller fiscal union.
- Adjustment in a currency union relies on factor mobility, price flexibility, and supporting financial/fiscal policies; euro area has low labor mobility and limited wage/price flexibility.

### Quantification approach and simulation methodology (baseline and steps)
- Baseline: April 2012 WEO.
- Three-step framework:
  - Step 1: Baseline scenario = April 2012 WEO.
  - Step 2: Quantify gap between fiscal plans under baseline and SGP targets (structural terms) holding GDP at WEO levels.
  - Step 3: Shock IMF’s GIMF and quantify output decline when multipliers operate.
- Fiscal shock defined as change in structural balance (Sb) to GDP ratio relative to WEO consolidation path for given GDP (WEO values).
- Simulations account for fiscal plans adopted or specified at time of April 2012 WEO forecasts.

### SGP rules and transition mechanics used
- SGP benchmarks applied:
  - Debt/GDP reduced below 60 percent.
  - Deficit/GDP below 3 percent.
  - Structural deficit/GDP below 1 percent (Maastricht) and subsequently below 0.5 percent.
  - Primary expenditure growth less than medium-term GDP growth.
- Transition mechanics (three-stage compliance over WEO horizon):
  - EDP phase: countries under Excessive Deficit Procedures expected to meet 3 percent deficit target by requested deadlines (between 2012 and 2015).
  - Grace period: exemption from 1/20th debt rule over three-year period after EDP closure; improve structural balances by at least 0.5 percent of GDP each year until MTOs.
  - 1/20th debt benchmark: three years after exiting EDP, improve structural balances by 0.5 percent of GDP per year or more to meet 1/20th rule (annual pace of debt reduction no less than 5 percent of gap between observed debt and 60 percent target).
- Implementation timing (figure note):
  - Expected entry into force: autumn 2012 for the two-pack; January 2013 for Directive on National Fiscal Frameworks; January 2014 for automatic correction mechanisms mandated by the Fiscal Compact.

### Country-specific MTOs (as presented)
- MTOs:
  - 0.5 for Belgium, Finland, and Luxembourg;
  - 0 for Austria, Cyprus, Estonia, Greece, France, Italy, Malta, and Spain;
  - -0.5 for Germany, Ireland, Netherlands, Portugal, and Slovak Republic;
  - -1 for Slovenia.

### Simulation scenarios, assumptions, and compositions
- Scenario 1: Myopia and growth-friendly consolidation
  - Consolidation tilted to measures with strong effects on household disposable income but little negative impact on factor supply and potential output.
  - Agents myopic: do not perceive permanence; beliefs change ex post.
  - Zero-interest floor (ZIF) assumed binding over 2012-17.
  - Variants: joint consolidation (HD and LD together) and stand-alone consolidation.
- Scenario 2: Credibility and growth-friendly consolidation
  - Same composition as scenario 1 but agents not myopic: changes perceived as permanent (partial Ricardian).
- Scenario 3: Credibility and growth-unfriendly consolidation
  - Composition switched toward high-multiplier measures:
    - (i) from consumption to corporate taxes;
    - (ii) from government consumption to public investment;
    - (iii) from general transfers to transfers targeted to households with high marginal propensity to consume.
- Composition fractions (Table 3):
  - Growth-friendly: 1/4 transfers, 1/3 government consumption, 1/4 labor income tax, 1/6 consumption tax.
  - Growth-unfriendly: 1/4 targeted transfers, 1/3 government investment, 1/4 labor income, 1/6 corporate tax.
- Monetary policy reaction:
  - ZIF: nominal interest rates unchanged over first 5 years.
  - Unconstrained: unconstrained reaction of nominal interest rates.
- Credibility assumptions:
  - Myopia vs partial Ricardian behavior as defined above.

### Models and regions used
- GIMF (six-region version): euro area split into HD and LD blocs; also US, Japan, emerging Asia, rest of world.
- G35: estimated structural macroeconometric model disaggregated into 35 national economies including 11 euro area countries.
- Spillovers transmitted via trade, financial, and commodity-price linkages.

### Simulation results — output effects (preserved numeric outcomes)
- Scenario 1 (Myopia and growth-friendly; ZIF binds):
  - Euro area: output is 1 percent lower than baseline by 2017.
  - Cumulative output loss: 3½ percent throughout 2012-17.
  - Loss decomposition:
    - Negative spillovers: around 40 percent of the loss.
    - ZIF (inability of monetary authorities to ease policy rate): 20 percent of the loss.
  - HD bloc: 1.4 percent of GDP loss by 2017; cumulative 5 percent over 2012-17.
  - LD bloc: 0.8 percent of GDP loss by 2017; cumulative 3 percent over 2012-17.
  - Cross-country cumulative output losses (2012-17, selected): Spain at around 10 percent; Portugal at almost 8 percent; Belgium, Finland, Ireland sizable negative spillovers; Greece positive net spillover from joint consolidation.
- Scenario 2 (Credibility and growth-friendly):
  - Euro area 2017 GDP loss falls from 1 percent (scenario 1) to 0.3 percent.
  - Cumulative losses 2012-17: -1.5 percent.
- Scenario 3 (Credibility and growth-unfriendly):
  - Euro area 2017 GDP loss: 0.8 percent (versus 0.3 percent in credible growth-friendly).
  - Cumulative losses 2012-17: -3.1percent.
- Sensitivity note: multipliers may be larger when excess capacity exists; recent empirical work suggests larger fiscal multipliers in such conditions.

### Policy perspectives and recommendations on consolidation
- Apply SGP rules flexibly; pace of consolidation should depend on state of public finances, growth, and monetary policy stance.
- Consolidation strategies that adjust to new information can be welfare improving; shift focus towards structural targets under SGP.
- Where financing conditions permit, pace of consolidation should account for adverse conditions (ZIF, negative output gaps, joint consolidation → larger multipliers).
- Avoid heavy front-loading; define steady pace in cyclically-adjusted terms.
- Composition: tilt adjustment toward growth-friendly measures; focus on spending side and inefficient spending.
- Credibility-enhancing reforms essential:
  - Anchor adjustment in well-specified medium-term plans.
  - Implement automatic correction mechanisms under the Fiscal Compact.
- Monetary policy should accommodate consolidation:
  - Significant output losses if monetary policy does not provide support.
  - When zero bound binds, unconventional monetary stimulus may be needed.

---

*Source: IMF staff report (content unit _cr12182).*

### 1. Sources of Banking Stress July 2011–June 2012 ..........................................................17

### 1. Sources of Banking Stress July 2011–June 2012

### A. Introduction
- The paper argues that the creation of a common Eurozone financial stability architecture (a “Banking Union”) is an immediate priority to restore the viability of the Economic and Monetary Union.
- The Banking Union should have three main elements: a eurozone deposit guarantee scheme, a eurozone resolution authority and a eurozone supervisory authority.
- The June 29 agreement of the EU Summit to present proposals for a single supervisory mechanism and, contingent on its implementation, allow direct recapitalization of banks by the ESM is described as an important step forward.
- IMF position summarized:
  - A centralized financial stability architecture is needed to keep pace with the rapid increase in the size of European banks and deeper economic and financial integration in the euro area.
  - The centralized framework envisioned: a common resolution authority with a common backstop, a common deposit guarantee scheme and a common supervisory authority.

### B. The Eurozone Crisis (July 2011–June 2012) — stages and transmission channels
- Initial shock (Summer 2011):
  - Wholesale funding markets for European banks began to freeze and sovereign borrowing costs of Italy and Spain began to rise significantly.
  - The Euribor-OIS spread rose sharply in August 2011 and “reached about half the level following the collapse of Lehman Brothers.”
  - From June 2011 onwards, US investors reassessed exposures to large euro area banks, causing a sharp increase in the cost of US$-euro basis swaps.
  - The US$ funding shock was particularly large for the French banks which lost about US$ 100 billion of funding from the 10 largest US MMF between June and December of 2011.
- Contagion across markets:
  - Good collateral became scarce as funding demand focused on secured markets; large European banks substituted short-term secured borrowing (repos) for unsecured borrowing, raising demand for good quality collateral.
  - Worsening sovereign crisis eroded collateral value as government bond yields rose and margin requirements increased following ratings downgrades.
  - ECB unconventional liquidity provision alleviated funding stress in secured markets but “collateral used for repo financing still remains significantly more expensive in June 2012 than it was in July of 2011.”
- Deleveraging:
  - According to EBA data, large European banks sold off more than €80 billion of bonds of periphery sovereigns during January–September 2011.
  - In the EBA sample, core euro area banks were the largest sellers of Italian and Spanish sovereign bonds while Italian and Spanish banks were the largest sellers of sovereign bonds of the program countries.
- ECB intervention:
  - The long-term refinancing operations (LTROs) of the ECB and the US$ swap lines stabilized funding markets and temporarily reduced sovereign stress.
  - Stabilization of bond yields was achieved in part by increasing sovereign exposures of Spanish and Italian banks, setting the stage for stronger adverse loops between weak sovereigns and domestic banks.
- Sovereign-bank nexus intensifying in Spain:
  - Government announcement on March 02 that previously agreed SGP fiscal targets were not attainable; downgrades of the sovereign occurred on April 23, June 07 and June 13.
  - The take-over and bailout of Bankia raised uncertainties about real estate losses and impacts on the sovereign balance sheet; a group of banks were downgraded.
- Determinants of banking stress before and after LTROs:
  - Before the first LTRO, bank-specific stress was mainly explained by:
    - sovereign exposures to Italy or Spain, and
    - dependence on wholesale funding and large activity in derivatives markets.
  - After the first LTRO, Spanish and Italian banks with larger exposures to their sovereign underperformed peers independently of sovereign bond yields or CDS spreads movements; reliance on wholesale funding or derivatives activity were no longer significant drivers of banking stress.
- Key market indicators and events referenced:
  - Dec 21st LTRO allotment.
  - EURibor-OIS spread, 5-Year Bank CDS spreads, difference (3 months euro repo - Euribor), and 5-Year Sovereign CDS spreads shown in text figures.
  - Data sources cited: Datastream; Bloomberg L.P.; IMF staff calculations; EBA July 2011 stress tests; EBA recapitalization exercise.

### C. Downward Spirals at the National Level
- Main mechanism:
  - Many European banks are large but remain the responsibility of the home authority; this contingent liability can jeopardize the sovereign balance sheet and amplify downward spirals between sovereigns, banks and the real economy.
- Fiscal cost examples:
  - Ireland: financial support impact on public debt has amounted to more than 40 percent of GDP since 2008.
  - Spain: financial support impact on public debt has amounted to about 20 percent of GDP since 2008.
- Empirical links between bank stress and sovereign costs:
  - Acharya et al. (2012) show that higher bank CDS spreads before a bailout resulted in higher government CDS spreads after bailouts.
  - An event study of the Bankia bailout suggests a likely link between announcements related to the take-over and recapitalization of Bankia and increases in sovereign borrowing costs around each event.
- Bank features that amplified stress (panel regression findings summarized):
  - Before LTRO (July–December 2011), larger weekly stock price declines (75th and 90th percentiles) were associated with higher derivatives assets & liabilities ratio and higher net exposures to Italy and Spain.
  - After LTRO (January–June 2012), net exposures to Italy and Spain remained important predictors of bank underperformance at the 75th and 90th percentiles; reliance on wholesale funding and derivatives activity were less significant.
- Additional statistics and indicators mentioned:
  - Large European banks sold more than €80 billion of periphery sovereign bonds (January–September 2011).
  - French banks lost about US$ 100 billion of funding from the 10 largest US MMF between June and December of 2011.
  - Financial sector support 2008–11 as percent of 2011 GDP for selected countries: Ireland 41.2; Spain 19.5 (includes actual use of debt guarantees, asset purchases and capital support from the FROB as of March 2012 and the ESM/EFSF loan announced on June 9th); Netherlands 14.1; Germany 12.2; United Kingdom 6.8; United States 5.3; Greece 6.1; Belgium 7.0.
- Conclusion drawn:
  - National-based responsibility for financial stability amid large, cross-border banking groups contributed to financial de-integration and stronger adverse sovereign-bank feedback loops, reinforcing the paper’s case for a Banking Union with common backstops.

*Prepared by Thierry Tressel (EURAE).*

### 14.      Weak sovereigns also damage the balance sheet of domestic banks. Three main

### 14.      Weak sovereigns also damage the balance sheet of domestic banks. Three main

### Channels through which weak sovereigns affect domestic banks
- Domestic banks usually have large exposures to domestic sovereign debt due to home bias in portfolio decisions; sovereign risks directly impact the market value of bank assets.
- Weak sovereigns can lose their ability to honor their financial safety net obligations, adversely impacting domestic banks through:
  - (i) the loss of the implicit (or explicit) sovereign guarantee;
  - (ii) reduced deposit confidence as domestic guarantee schemes (which are usually not pre-funded)) become less credible.
- The decline in the collateral value of sovereign bonds (as yields increase or as higher margin requirements are applied after ratings downgrades) reduces the ability of domestic banks to access secured funding or euro-system liquidity.

### Banking stress, lending rates, and the real economy
- Banking stress raises the cost of credit by causing banks to curtail credit supply and raise lending rates to strengthen capital buffers and increase internally generated funds.
- Weak economic performance and fragile non-bank balance sheets reduce loan demand, raise non-performing loans, and slow deposit growth, thereby affecting bank profitability.
- Empirical estimate (panel regression of monthly lending rates in the four largest euro area countries, 2010 to March 2012) findings:
  - A 150 bps increase in bank CDS spreads is associated with an increase in lending rates of 0.6 percentage points for loans to non-financial corporations.
  - A 150 bps increase in bank CDS spreads is associated with an increase in lending rates of 0.4 percentage points for mortgage loans.
- Comparative observation for 2010:
  - Bank CDS spreads in Spain and Italy rose by between 150 and 200 bps while lending rates to non-financial corporations increased by respectively 0.5 and 0.8 percentage points.

### Interaction between sovereign weakness and macroeconomic outlook
- Weak sovereigns and a deteriorating macroeconomic outlook tend to reinforce each other.
- In absence of fiscal risk sharing, a weak sovereign cannot implement countercyclical fiscal policies when the economy is slowing down and may have to consolidate in a downturn to restore confidence and lower debt levels.
- Fiscal consolidation in a downturn can further weaken economic growth through compression of domestic demand.
- Ambiguity of financial investors’ beliefs about the link between growth and fiscal consolidation could make adjustment more arduous.

### Adverse feedback loops in a monetary union
- Feedback loops are stronger in a monetary union because:
  - Absence of a domestic exchange rate removes a buffer against intra-euro area sudden stops on sovereign borrowing costs.
  - Lack of exchange rate depreciation prevents export stimulation that could compensate fiscal-driven demand compression.
- Sovereign borrowing costs can rapidly spiral if market anticipations turn pessimistic, making fiscal adjustment more difficult unless the monetary authority signals the possibility of future loosening.

### Financial fragmentation of the EMU: pre-crisis integration and crisis-driven reversal
- Before the crisis, the eurozone had a very high degree of financial integration; sovereign bond yield spreads compressed (example: between 01/2002 and 12/2007, Greek sovereign–German bund spreads oscillated between 8 and 37 bps).
- Prior to the crisis, current account deficits of peripheral countries were almost entirely financed by private capital flows originating from within the eurozone, allowing net foreign liabilities to accumulate.
- Sovereign risks in periphery countries were de facto shared between core euro area and domestic investors (example: end-2010, more than 75 percent of foreign-held Spanish sovereign debt was held by euro area residents).
- During the crisis, withdrawals of intra-euro area cross-border capital flows sharply reduced sovereign cross-border risk sharing and caused divergence of sovereign yields, impeding the interest rate channel of monetary policy.
  - In 2011 and Q1 of 2012, bond markets experienced a withdrawal of intra-euro area cross-border capital of €240 billion (a 24 percent reduction since end 2010).
  - Intra-euro area interbank markets experienced a sharp reduction of cross-border claims of €190 billion in Q4 of 2011 and Q1 of 2012.
  - Example of reversal in non-resident holdings: at end-2011 non-residents held 34 percent and 33 percent of the sovereign debt of Italy and Spain, down from 44 percent and 48 percent at end-2009.

### Fragmentation effects on lending, interbank markets, and deposits
- Increased cross-country dispersion of lending rates has weakened capital allocation at the eurozone level and disrupted the bank lending channel of monetary policy.
  - Example: spread on lending rates to non-financial corporations between Spain and Germany increased from 6 bps in August 2008 to 43 bps in January 2011 and to 187 bps in April 2012.
  - Firms face increasingly different credit supply conditions across countries irrespective of their own profitability.
  - The bank lending channel operates less well in countries under stress: lowest lending rates responded to policy rate changes while highest lending rates moved in the opposite direction and tracked banking stress.
- De-integration of interbank markets disrupts monetary policy transmission:
  - Overnight EONIA average daily volumes declined to €20 billion at the end of May 2012 from €40 billion at the end of 2010.
  - Overnight bank deposits at the ECB rose to about €750 billion from less than €200 billion in June 2011.
  - Interbank markets have become increasingly segmented along national borders; ECB is intermediating capital across national borders.
  - Net borrowing from the eurosystem remained high for the 3 program countries and sharply increased for Italian and Spanish banks, while Dutch and German banks accumulated large net claims on the eurosystem; private markets were replaced by fixed rate full allotment of liquidity by the ECB.
- Insured deposits are declining in several periphery countries:
  - Since end-2009, cumulative decline in the stock of insured deposits reached €86 billion in the three program countries.
  - In Spain, between July 2011 and April 2012, insured deposits (households and non-financial corporations) declined by €55 billion.

### Main elements of a Eurozone Banking Union (policy recommendations)
- Purpose and expected benefits:
  - Establishing a banking union will delink financial intermediation from sovereign strength and end financial fragmentation.
  - It will help sever feedback loops between sovereigns and domestic banks, restore financial integration, restart growth by ensuring healthy eurozone firms can obtain credit regardless of sovereign strength, and stabilize financial systems to ensure smoother monetary policy transmission in weak countries.
- Common deposit insurance scheme:
  - A common deposit insurance scheme with collective and credible backing by all member countries should be established to delink banks and sovereigns where the latter cannot honor safety net obligations.
  - National DGS are not pre-funded and therefore weak sovereigns cannot credibly insure non-bank private deposits; national DGS, even if pre-funded, cannot reassure agents that their savings are protected in euros.
  - Design features suggested:
    - Announcement should include a clear timetable for establishment.
    - Deposit insurance should apply to all financial institutions to avoid accelerating withdrawals or deposit shifts.
    - Ultimately partially pre-funded by a levy on the industry.
    - To be effective immediately, should have access to additional funding such as a credit line from the euro-system or be backstopped by a common pool of government resources (e.g., ESM/EFSF) or the possibility to issue a limited amount of joint and several guaranty bills.
    - Scheme should be consistent with best practices in the 2010 EU Commission proposal to harmonize national deposit guarantee schemes and with the Directive for an EU framework for recovery and resolution of credit institutions.
- Centralized bank resolution agency with common financing:
  - A centralized resolution agency would enable more efficient bank restructuring without threatening sovereign balance sheets.
  - Centralized restructuring when common resources are relied upon ensures ex ante burden sharing and efficient cross-border cooperation when cross-border banks fail.
  - The resolution authority should be backed by a common fund financed by an industry levy and have recourse to a common pool of government-provided resources or a liquidity backstop from the ECB.
  - EFSF/ESM facilities should be empowered with flexibility to directly recapitalize banks of member states to delink sovereign debt from bank restructuring costs.
  - Resolution powers should be consistent with the EC framework for a harmonized resolution regime, including powers to bail-in unsecured creditors and to emphasize preparedness and prevention.
- Supra-national supervisory regime for large banks:
  - Financing and monitoring responsibilities should be aligned; a common supervisor would align incentives with the common good and reduce incentives to protect “national champions”.
  - A supra-national supervisor would help solve coordination problems and foster financial stability to prevent further fragmentation.
  - Several models conceivable:
    - The ECB could be the eurozone supervisory authority for all banks with delegation of tasks to national entities, balancing LOLR role and monetary independence.
    - A two-tier system where the ECB or another body supervises large/systemic banks while smaller local banks remain under national oversight.
- Scope and urgency:
  - While other EU countries should be able to join, the banking union is urgent for the euro area due to the dramatic consequences of capital flight in a monetary union.
  - A Deposit Guarantee Scheme covering all euro area depository institutions would reassure retail depositors across the EMU.
  - A euro area bank resolution authority would help prevent bank-sovereign feedback loops within the currency union.
  - The ECB could be given explicit responsibility for financial stability and full lender of last resort functions to align roles and eliminate bank-sovereign linkages present in the current ELA scheme.
  - The Banking Union framework should allow opt-in by other EU countries and be consistent with EU-level financial stability frameworks and regulations.

*Source: IMF staff report (content unit _cr12182).*

### 30.      The eurozone architecture must be completed by creating a Banking Union. A

### _cr12182 - 30.      The eurozone architecture must be completed by creating a Banking Union. A

### Banking Union: rationale and core elements
- A Banking Union would help delink sovereigns from their banks and provide incentives to various players to end the financial fragmentation of the eurozone.
- The Banking Union has three main elements:
  - a deposit guarantee scheme backstopped by common resources,
  - a resolution authority with access to a common fund,
  - a supervisory authority.
- While a Banking Union could operate without fiscal union, it will perform more effectively when elements of a fiscal union providing a clearer framework for fiscal transfers are also in place.
- The Banking Union is first and foremost a necessity for the euro area, but should be opened to other EU countries.

### Recent policy steps and proposals (June 29)
- On June 29, the European authorities agreed to pursue a key element of a Banking Union and changes in the use of the EFSF/ESM resources:
  - The European Commission will present proposals for the establishment of a single supervisory mechanism that would involve the ECB.
  - Contingent on the single supervisor being in place, the ESM would be allowed to directly recapitalize banks.
- The EU Council is asked only to consider the proposals of the EU Commission by the end of 2012; timing and speed of implementation remain to be clarified.

### Implementation challenges and roadmap requirements
- Key details and the path towards the new architecture remain to be clarified:
  - Pros and cons of various models of a Banking Union must be weighed carefully to ensure convergence on the most effective financial stability architecture.
  - A clear roadmap toward full institutional set-up is needed, with a clear understanding of transitional and sequencing issues.
  - Fast implementation of EU Directives and regulations (such as the bank resolution Directive and the CRDIV/CRR) would provide the initial building block of the new framework.
  - The ESM/EFSF should urgently be made flexible enough to allow direct equity injections into banks.
  - The path towards the Banking Union cannot be dissociated from the current crisis response; choices in terms of banking stability are not independent from other dimensions, including those related to a fiscal union.
  - It is essential to identify the aspects of the reform process that will have the largest impact on market sentiment in the short-term.
  - EU-level consistency: any euro-area steps must be consistent with EU directives and regulations and will require interacting with or reinforcing existing EU institutions (the ESAs, the ESRB).
  - Subsidiarity: while the Banking Union will centralize powers and financing responsibility, the subsidiarity principle will apply to determine the future role of national authorities and their interaction with supra-national institutions.

### Box 1 — Sources of Banking Stress (July 2011–June 2012): econometric findings
- Sample and period:
  - Weekly stock prices and CDS spreads of 58 banks between July 2011 and June 8th 2012.
  - Sample split into pre- and post-LTRO.
  - Banks included were listed European banks participating in the 2011 EBA stress tests; exposures from the EBA recapitalization exercise.
- Main specification (as described): ittiStiSiSitit qZpExpExpXy, where variables are defined in the text.
- Key findings before the LTRO:
  - CDS spreads and equity returns of European banks co-moved positively with stress experienced by the Italian and Spanish sovereigns, in proportion to their exposures.
  - Banks more exposed to the Italian and Spanish sovereigns experienced higher declines in stock prices (respectively higher increase in CDS spreads) when bond yields (respectively sovereign CDS spreads) soared.
  - The co-movement was economically significant mostly for domestic banks with very large exposures to their sovereign.
  - Weekly stock price declined by 2 percentage point between July and end December on average; the impact of the median weekly increase in Italian bond yields on weekly stock prices was about 1 percentage points weekly only for the top 10 percent banks with the largest exposures to the Italian sovereign.
  - Similar conclusions hold for exposures to the Spanish sovereign.
  - Banking stress also appeared higher for core EA banks very active in derivative markets: top 10 percent banks with the largest derivative activities experienced, at the mean change in the VIX, higher declines in weekly stock price of about 2 percent on average before the LTRO.
- Key findings after the first LTRO:
  - As funding and derivative markets stabilized, derivative activity and wholesale funding had insignificant impact on banking stress.
  - Large exposures to the Italian and Spanish sovereign became indicators of bank stress irrespective of yields or CDS movements for those sovereigns.
  - Banks with the top 10 percent largest exposures to the Italian and Spanish sovereigns underperformed their peers in equity markets by 3-4 percentage points weekly.
  - Portuguese bank equity prices also started to comove with domestic government bond yields.
  - Equity market stress of banks with larger exposures to the Spanish (or Greek) sovereign started to comove negatively with stress in government bond markets (possible reasons noted in the text).

### Box 2 — Event Study: Bankia and Spanish sovereign borrowing costs
- Context and timeline:
  - Bankia established in December 2010 by merger of seven savings banks; initially owned by holding company BFA.
  - July 2011: Bankia IPO of about €4 billion of shares.
  - As of end 2011, BFA-Bankia was the fourth largest bank in Spain with total assets of about 30 percent of GDP and the largest holder of real estate assets.
  - May 7th 2012: Bankia Chairman Rodrigo de Rato stepped down and government announced the take-over of Bankia.
  - May 10th 2012: Government converted its €4.5 billion of preferred shares in BFA into voting shares, obtaining a controlling stake.
  - May 25th 2012: Reported that Bankia had negotiated a further €19 billion injection.
  - June 9th: Eurogroup announced willingness to commit support to Spain through an EFSF/ESM loan of up to €100 billion for financial institution recapitalization.
- Event study results:
  - Focusing on windows of ±4 days around each of four events, sovereign borrowing costs (10 year government bonds) rose on average by 40 bps.
  - Spreads on 5 year sovereign CDS contracts rose by 37 bps on average.
  - No increase in CDS spreads followed the announcement of the Eurogroup loan (noted explanation).
  - Evidence suggests concerns about the fiscal cost of bailing out a large financial institution adversely affected Spanish government borrowing costs, possibly due to anticipated impact on debt dynamics.
  - Alternative hypothesis that deteriorating macro outlook explained the increases is not strongly supported by the data; supporting evidence cited in the text.

### Fiscal integration: case, constraints, and implications for Banking Union
- More fiscal integration with stronger governance and more risk sharing can reduce the threat that economic shocks in one country endanger the euro area as a whole.
- A limited and scalable introduction of common debt with appropriate governance safeguards could support the creation of a Banking Union and signal a strong intermediate commitment to a fuller fiscal union.
- Adjustment under a common currency:
  - In a common currency area, adjustment to idiosyncratic shocks falls on factor mobility, price flexibility, and supporting financial and fiscal policies.
  - Where labor mobility is low, intra-area capital flows are volatile, and structural rigidities impede price adjustment and reallocation, fiscal and financial policies must take on some adjustment burden.
- Where the euro area stands (selected findings):
  - Labor mobility:
    - Only about 1 percent of the working age population changes residence within their country in a given year, and even less move between euro area countries.
    - Comparisons: about 3 percent in the US, 2 percent in Australia, and slightly less than 2 percent in Canada (Figure 1 references).
    - Obstacles: cultural and language barriers, housing market distortions, limits to portability of pensions, absence of a cross country social safety net.
  - Capital flows:
    - Capital moves freely across the euro area but is susceptible to sudden swings that challenge financial stability.
    - Cross border credit increased rapidly pre-2008, then reversed; core countries became recipients of net private capital flows by 2008.
  - Lack of common backstop:
    - Absence of strong common regulation, supervision, resolution powers, and deposit insurance, all supported by sufficient supra-national backstops, generates adverse feedback loops between financial sector and sovereigns and fuels contagion.
    - Given the size of national banking systems, problems can overwhelm fiscal capacity of individual sovereigns; banks have significant exposures to their sovereigns.
  - Wage and price flexibility:
    - Wage and price flexibility in the euro area is limited, slowing correction of macroeconomic imbalances.
    - Prices and wages have not responded strongly to deteriorating macroeconomic conditions; with few exceptions (e.g., Ireland), contributing to lengthy and costly adjustment.
- Insufficient risk sharing and governance:
  - The EU budget is small and was not designed as a risk sharing tool; it provided little help to crisis-hit countries.
  - The Stability and Growth Pact (SGP) failed to encourage creation of sufficient fiscal space.
  - Estimates suggest strict adherence to structural targets during 2000-07 would have reduced debt by about 7 percent of GDP by 2007, all other things being equal (Figure 3).

*Source: IMF staff analysis as presented in the provided content.*

### 8.      In  sum,  the  euro  area  can  be  characterized as an area with incomplete internal

### In sum, the euro area can be characterized as an area with incomplete internal adjustment mechanisms and insufficient policy coordination

### What Makes Currency Areas Viable? — Fiscal risk sharing
- Existing currency areas feature a high degree of fiscal and financial integration, often with strong governance requirements and formalized mechanisms of ex-ante insurance against fiscal and financial risks.
- Common backstops for the financial system enhance financial stability and prevent negative links between banks and sovereigns; example: the Federal Deposit Insurance Corporation (FDIC) insures deposits regardless of state of registration and acts to resolve banks countrywide.
- Risk sharing mechanisms typically include tax sharing arrangements and transfers from the central to regional government and respond to cyclical developments at the regional level.
- Staff analysis: on average, a 1 percent increase in a region’s output gap can lead to an increase in central government transfers offsetting between 5 and 20 percent of the income shock.
- Past findings: risk sharing could smooth about 10 to 20 percent of regional income shocks (Melitz and Zumer, 2002; von Hagen, 2007).
- Central provision of public goods and services conveys fiscal risk sharing because during a downturn a region’s relative tax contribution falls while benefits remain unchanged; central government spending is a significant share of general government in a number of countries.

### Box 1 — Risk Sharing and Redistribution (empirical results)
- Sample: six federations — Australia, Brazil, Canada, China, Mexico and the United States.
- Method: panel regression of gross transfers from central government to sub-national budgets (ratio of state GDP) on real per capita GDP and the output gap, with regional fixed effects.
- Key empirical finding: in all countries, transfers respond significantly to a variation in the regional output gap, offsetting between 5 and 20 percent of cyclical fluctuations.
- Regression excerpt (selected coefficients reported exactly as in source):
  - Real per capita GDP coefficients: 0.18 (Australia), -2.04 (Brazil), -2.74 (Canada), 6.91*** (China), -0.02 (Mexico), 3.83*** (USA).
  - Output gap coefficients: -0.14*** (Australia), -0.13*** (Brazil), -0.12** (Canada), -0.18*** (China), -0.04** (Mexico), -0.09*** (USA).
  - Constants and significance: Constant values and significance reported in source table (e.g., 9.00** for Australia; -23.45 for Brazil; 4.00 for Canada; -51.32** for China; *7.13-7.85** for Mexico; details preserved in source table).
  - Observations and R-squared (as reported): Obs 98 39 01 92 41 92 24 19 24 86 72 (table formatting preserved in original); R-sq uared 0.34 0.24 0.02 0.44 0.02 0.21 (as shown in source).

### What Makes Currency Areas Viable? — Strong governance
- Fiscal integration often couples risk sharing with governance frameworks that limit regional fiscal sovereignty and encourage fiscal behavior aligned with common standards.
- Common elements:
  - A center with area-wide tax authority where public goods are provided centrally.
  - Where transfers are significant, the center often has the right to intervene more directly in regions’ public finances.
  - Legally enforceable fiscal rules are commonly used (examples in other areas: balanced budget rules for states in U.S., the debt brake rule for Germany’s Länder).
- Bailout arrangements vary: formalized, ad-hoc, or explicitly banned (no-bailout clauses). Where bailouts exist, deployment combines loss of regional fiscal sovereignty in return for central resources.
- Summary: unitary states (e.g., France, U.K.) tend to have high fiscal risk sharing and low regional fiscal sovereignty; federations (e.g., Germany, Switzerland, U.S.) have intermediate risk sharing and delegation; confederations (e.g., EU) have lower degrees of both.

### Options for the Euro Area — General considerations
- Designing fiscal integration can follow subsidiarity and allocate functions centrally where benefits outweigh costs; this requires adequate financing and democratic accountability.
- Functions: preserve economic stability after idiosyncratic shocks (insure regions against fiscal risks) and provide a common financial sector backstop.
- Financing: a sufficiently sized central budget with centrally controlled taxation or other revenue; extra-budgetary solutions for specific central functions (e.g., financial sector); common borrowing can support these functions and offer horizontal risk sharing.
- Accountability and governance: democratic accountability and strong governance safeguards at the regional level are necessary to limit moral hazard.

### Options for the Euro Area — Intermediate steps toward risk sharing
- A near-term effective step: provide a common fiscal backstop for a banking union (could take the form of common debt), helping break bank-sovereign adverse feedback loops and prevent market fragmentation; delegation of responsibilities must go hand in hand with delegation of oversight to common institutions.
- Common borrowing could ensure market access for stressed sovereigns and create safe assets for the banking sector.
- Two proposals highlighted:
  - Eurobills: joint and several liability instruments with maturity of less than one year, covering up to 10 percent of each country’s own GDP; participating countries unconstrained for long maturities; participation could be conditional on fiscal targets; scalable and maturity could be lengthened over time.
  - Debt Redemption Fund: gradual transfer of debt exceeding 60 percent of GDP into a fund for which EMU members would be jointly and severally liable; participating countries repay transferred debts within a total of 25 years; participation conditional on a debt reduction plan and structural reforms; countries required to deposit collateral and earmark part of tax revenues for payment obligations.
- Extra-budgetary and project-specific borrowing (e.g., via European Investment Bank or “project bonds”) are alternatives but offer limited benefits due to small size and low risk sharing.

### The need for more governance (design elements)
- Roadmap toward fiscal integration must define governance requirements; partial schemes of common borrowing and collateral mechanisms can mitigate moral hazard.
- Partial Eurobonds: tranching could lower average borrowing costs while raising marginal costs beyond common debt framework; restricting common debt to short maturities and conditioning participation on fiscal behavior strengthens incentives.
- Collateral mechanisms: in absence of a federal treasury, participants should commit collateral (present assets and/or future revenues) to guarantee future payments; possible collateral includes present assets, future revenues, surcharges on taxes, currency and gold reserves.
- Central governance enhancements (building on recent reforms) could include:
  - Time-bound commitments to improve fiscal transparency.
  - Obligation to periodically publish comprehensive fiscal risk assessments.
  - Time-bound commitments to improve budgetary practices (top-down budgeting, moving to accrual accounting, regular spending reviews).
  - Consent to annual auditing of public accounts conducted by independent parties.

### Conclusions
- A more viable EMU involves more fiscal and financial integration.
- Given low labor mobility and volatile capital flows, the euro area needs fiscal and financial policies that entail a higher level of risk sharing to respond more effectively to idiosyncratic regional shocks.
- Greater fiscal integration would centralize macroeconomic and financial stabilization, provide ex ante insurance against risks, and signal strong commitment to making EMU a closer union.

*IMF staff analysis and findings as presented in the source content.*

### 23.      Although getting to the endpoint of fiscal integration will take time, intermediate

### _cr12182 - 23.      Although getting to the endpoint of fiscal integration will take time, intermediate

### Intermediate step: limited but scalable common debt
- The limited but scalable introduction of common debt, with appropriate governance safeguards, would:
  - help break the adverse feedback loop between banks and sovereigns;
  - support the development of a banking union;
  - be a step towards a closer fiscal union.

### Quantification approach and simulation methodology
- Baseline: April 2012 WEO.
- Three-step conceptual framework:
  - Step 1: Choose as baseline scenario the April 2012 WEO.
  - Step 2: Quantify the gap between fiscal plans under this baseline and the SGP targets (in structural terms) keeping GDP at WEO levels (i.e., no multiplier effects are at play). The fiscal shock is defined as the change in the structural balance (Sb) to GDP ratio relative to the fiscal consolidation path projected in the WEO, for a given GDP (at WEO values).
  - Step 3: Shock the IMF’s dynamic stochastic general equilibrium model—the Global Integrated Monetary and Fiscal model (GIMF)—with the calculated shock and quantify the output decline when multiplier effects are at work.
- The shock and simulation results take into account fiscal plans adopted or specified in sufficient detail at the time of the elaboration of the April 2012 WEO forecasts.

### SGP rules, ordering assumptions, and interpretation
- Key SGP benchmarks and constraints (as applied in the analysis):
  - Debt/GDP must be reduced to below 60 percent.
  - Deficit/GDP below 3 percent at any t.
  - Structural deficit/GDP to remain below 1 percent (Maastricht) and subsequently below 0.5 percent (later reforms).
  - Primary expenditure (excluding unemployment benefits and tax discretionary increases) grows less than medium-term GDP growth.
- Order of prevalence assumption:
  - Where there is overlap between rules, the strictest criterion applies. This allows calculation of SGP consolidation paths unambiguously and rules out over-determination.
- The SGP occupies a central role in the EU fiscal framework and is being upgraded through multiple reforms aimed at tighter national enforcement, expanded surveillance, and enhanced EU oversight.

### Fiscal regimes and transition mechanics used in the simulations
- Three-stage compliance process over the WEO horizon (all fiscal commitments translated into deviations from WEO in structural deficit to GDP ratio):
  - EDP phase:
    - Countries currently under Excessive Deficit Procedures (EDP) are expected to deliver structural adjustments needed to meet the 3 percent of GDP deficit target by the requested deadlines (between 2012 and 2015; see Table 2).
  - Grace period:
    - An exemption from the 1/20th debt reduction rule applies over the three-year period following the closure of the EDP.
    - During this period, countries are expected to improve structural balances by at least 0.5 percent of GDP each year until they reach their respective medium-term objectives (MTOs).
  - 1/20th debt benchmark:
    - Three years after exiting the EDP, structural balances will improve by 0.5 percent of GDP per year or more, if required by the 1/20th debt benchmark.
    - The benchmark ensures an annual pace of debt reduction no less than 5 percent of the gap between the observed debt level and the 60 percent of GDP target.
    - EU authorities will first verify compliance with the debt rule in a backward-looking manner and then in a forward-looking manner for countries breaching the first criterion.
- Implementation timing for EU reforms (Figure note):
  - Expected entry into force: autumn 2012 for the two-pack; January 2013 for the Directive on National Fiscal Frameworks; and January 2014 for the automatic correction mechanisms mandated by the Fiscal Compact.

### Country-specific MTOs (as presented)
- Current MTOs are:
  - 0.5 for Belgium, Finland, and Luxembourg;
  - 0 for Austria, Cyprus, Estonia, Greece, France, Italy, Malta, and Spain;
  - -0.5 for Germany, Ireland, Netherlands, Portugal, and Slovak Republic;
  - -1 for Slovenia.

### Modeling caveats and scope
- GDP is kept at WEO levels when computing the structural shock (OECD budgetary semi-elasticities used to break down overall deficit into structural and cyclical components).
- The magnitude and timing of output effects depend on composition and credibility of fiscal packages and the ability of monetary policy to cushion fiscal tightening.
- The paper presents simulations of output effects under different sets of assumptions and concludes with a number of policy recommendations (Section D).

_Italic: IMF staff paper drawing on the April 2012 WEO and SGP regulations_

### 8.      Overall, planned fiscal efforts in the EA fall significantly short of SGP

### 8.      Overall, planned fiscal efforts in the EA fall significantly short of SGP requirements

### Planned fiscal effort: magnitudes and cross-country split
- For the euro area as a whole, the additional consolidation amounts to 1 percent of GDP over 2012-17, nearly half of which would be frontloaded over 2012-13.
- Split of country blocs:
  - High-debt (HD) bloc: Greece, Ireland, Italy, Portugal, Spain and Belgium.
  - Low-debt (LD) bloc: the rest of the euro area.
- Additional consolidation needs:
  - HD bloc: around 2.2 percent of GDP over 2012-17.
  - LD bloc: 0.4 percent of GDP over 2012-17.
- Cross-country notes:
  - Additional fiscal effort is highest in Spain, mainly due to requirements under the EDP.
  - Germany, Cyprus and Estonia have no additional adjustment (WEO path more demanding than SGP).
  - Among larger euro area countries, additional consolidation is particularly frontloaded in Spain and the Netherlands.
- Footnote definition preserved: high-debt group includes countries with debt projected to be above 85 percent of GDP by 2017.

### Scenarios and key simulation assumptions (as specified)
- Scenario 1: Myopia and growth-friendly consolidation
  - Consolidation tilted toward measures with strong effects on households’ current disposable income but little negative impact on factor supply and potential output.
  - Agents are myopic: do not perceive government commitment as permanent; change beliefs ex post once they verify measures remain in place.
  - Zero-interest floor (ZIF) assumed to bind over the 2012-17 period.
  - Two variants to gauge spillovers: joint consolidation (HD and LD together) and stand-alone consolidation (HD or LD separately).
- Scenario 2: Credibility and growth-friendly consolidation
  - Same as scenario 1 except agents are not myopic: changes in the structural balance are perceived as permanent as of the year of implementation (partial Ricardian behavior). Fiscal changes are not anticipated.
- Scenario 3: Credibility and growth-unfriendly consolidation
  - Variant of scenario 2 with composition switched toward high-multiplier measures:
    - (i) from consumption to corporate taxes;
    - (ii) from government consumption to public investment;
    - (iii) from general transfers to transfers targeted to households with a high marginal propensity to consume.
- Composition fractions (Table 3, preserved verbatim):
  - Growth-friendly consolidation: 1/4 transfers, 1/3 government consumption, 1/4 labor income tax, 1/6 consumption tax.
  - Growth-unfriendly consolidation: 1/4 targeted transfers, 1/3 government investment, 1/4 labor income, 1/6 corporate tax.
- Monetary policy reaction:
  - ZIF: nominal interest rates unchanged over first 5 years (ZIF binds).
  - Unconstrained: unconstrained reaction of nominal interest rates.
- Credibility assumptions:
  - Myopia: agents do not perceive permanence; change beliefs ex post.
  - Partial Ricardian behavior: changes perceived as permanent as of year of implementation.

### Simulation tools and methodology
- Models used:
  - GIMF (six-region version): euro area split into HD and LD blocs; models liquidity-constrained and finite-planning-horizon households; used for main simulations.
  - G35: an estimated structural macroeconometric model disaggregated into 35 national economies, including 11 euro area countries (Austria, Belgium, Finland, France, Germany, Greece, Ireland, Italy, Netherlands, Portugal, Spain); used for country-specific effects.
- Regions in GIMF: HD bloc, LD bloc, US, Japan, emerging Asia, rest of world.
- Spillovers transmitted via trade, financial, and commodity price linkages.

### Simulation results: output effects (numeric outcomes preserved)
- Scenario 1 (Myopia and growth-friendly; ZIF binds):
  - Euro area: output is 1 percent lower than baseline by 2017.
  - Cumulative output loss: 3½ percent throughout 2012-17.
  - Loss decomposition:
    - Negative spillovers: around 40 percent of the loss.
    - Inability of monetary authorities to ease policy rate (ZIF): 20 percent of the loss.
  - HD bloc: 1.4 percent of GDP loss by 2017; cumulative 5 percent over 2012-17.
  - LD bloc: 0.8 percent of GDP loss by 2017; cumulative 3 percent over 2012-17.
- Cross-country cumulative output losses (2012-17, selected):
  - Spain: at around 10 percent.
  - Portugal: at almost 8 percent.
  - Belgium, Finland, Ireland: sizable negative spillovers (small open economies).
  - Greece: positive net spillover from joint consolidation due to improved terms of trade from reduced world commodity demand (Greece relatively closed).
- Scenario 2 (Credibility and growth-friendly):
  - Euro area 2017 GDP loss falls from 1 percent (scenario 1) to 0.3 percent of GDP.
  - Cumulative losses throughout 2012-17: -1.5 percent.
- Scenario 3 (Credibility and growth-unfriendly):
  - Euro area 2017 GDP loss: 0.8 percent of GDP (versus 0.3 percent in credible growth-friendly).
  - Cumulative losses throughout 2012-17: -3.1percent.
- Sensitivity note: Multipliers may be larger when there is excess capacity; recent empirical work suggests fiscal multipliers are larger in such conditions.

### Policy perspectives and recommendations (preserve original emphasis and wording)
- Apply SGP rules flexibly to accommodate unexpected events; appropriate pace of consolidation should depend on the state of public finances and growth, and the monetary policy stance.
- Consolidation strategies that adjust for new information can be welfare improving; shift of focus towards structural targets under the SGP is appropriate.
- Where financing conditions permit, the pace of fiscal consolidation should account for current adverse conditions:
  - With limited scope for monetary policy, negative output gaps, and joint consolidation, multipliers likely larger than normal.
  - Multipliers might increase with the size of consolidation.
  - Adjustment should occur at a steady pace defined in cyclically-adjusted terms and should avoid heavy front-loading.
- Composition of adjustment should be tilted towards growth-friendly measures:
  - Focus on the spending side, targeting areas where multipliers are low or spending is most inefficient.
- Reforms that underpin credibility are essential:
  - Credible policies raise agents’ expectations about positive future income effects of consolidation and can reduce multipliers in the short term, acting as a substitute for heavy frontloading.
  - Anchor adjustment in well-specified medium-term plans.
  - Responsible implementation of automatic correction mechanisms under the Fiscal Compact important to safeguard durable fiscal efforts.
- Monetary policy should accommodate consolidation:
  - Simulations suggest significant output losses if monetary policy does not provide support.
  - When the zero bound is binding or conventional interest rate cuts are less effective, unconventional monetary policy stimulus may be needed.

*Source: _cr12182 - 8.      Overall, planned fiscal efforts in the EA fall significantly short of SGP*

### 3.      Currently, a key question is whether the SMP has become less effective after the

### _cr12182 - 3.      Currently, a key question is whether the SMP has become less effective after the

### Background and key issue
- The SMP’s effectiveness may have been reduced after the Greek PSI exemption and possibly earlier due to subordination risk stemming from ECB treatment.
- The treatment of the ECB may have reshaped the seniority structure of all official holdings of sovereign bonds, implying possible mispricing of many euro area bond markets and leaving future SMP beneficiaries subject to rating downgrades as ECB interventions reduce the private investor base and increase losses in the event of restructuring.
- Anecdotal market evidence suggests controversy over the impact of SMP purchases; some of this may have been priced in prior to the Greek debt exchange in February 2012.
- Relevant dates and statements:
  - Euro area summit of heads of state and government on July 21, 2011 (PSI for Greece first announced at that summit).
  - ECB President Trichet statement on July 22, 2011 that the Eurosystem would not participate in the voluntary private sector involvement for Greece (footnote references and quote indicated).
  - ECB debt swap on February 16, 2012 examined as an event.

### SMP interventions, timing, and scale (empirical observations)
- SMP purchases and related market movements:
  - Purchases amounted to €36bn and €21bn in the first two months of the program (May-June 2010) but dropped to less than a few billion afterwards.
  - Purchases were scaled up again in November-December 2010.
  - Reactivation in August and September 2011 with purchases of €51bn and €37bn led to a reduction in broad market stresses.
  - As stress re-emerged, SMP purchases increased again to €40bn in November 2011 before the ECB adopted other measures.
- After the PSI exemption, SMP purchases under the SMP were largely put on hold:
  - Only one SMP intervention took place afterwards, in the week of March 5-9, 2012 for an amount of €27 million (the fifth-smallest intervention since May 2010), based on weekly Eurosystem financial statements.
- Conclusion from these patterns:
  - While the SMP helped temporarily reduce stress in government bond markets, due to the limited scale/time horizon of its effective use, it did not appear fully effective.

### Empirical event-study findings on subordination risk
- Data and methodology notes:
  - Event windows examined around July 22, 2011 (ECB statement) and February 16, 2012 (debt swap).
  - Standardized cumulative abnormal yields and CDS spreads estimated in first differences following MacKinlay (1997).
  - Models used 70 daily observations with data up until one month ahead of the event to avoid coefficient bias.
- Key empirical results:
  - Little impact of the ECB debt swap on periphery yields when evaluated around the time of the debt swap announcement (Figure 2).
  - The initial PSI announcement and the ECB’s non-participation (July 22, 2011) had substantial short-term negative effects:
    - Longer-dated periphery bond yields and CDS default risk premia rose unexpectedly following President Trichet’s statement that the ECB would not participate in the voluntary Greek PSI.
    - Bond yields and CDS premia had come down substantially in the days ahead of the July 21, 2011 summit, but this was reversed after the ECB’s statement.
  - Insufficient quantitative evidence to establish empirically the relation between subordination risk from ECB debt purchases and sovereign yields, because ECB debt purchases were largely put on hold after the PSI exemption.

### Theoretical quantification of subordination effects — closed-form bond model
- Fundamental factors determining subordination impact:
  - Probability of default (PD).
  - Loss given default (LGD) = 1 − recovery rate.
  - Share of ECB bond holdings (ECB share, b).
- Illustrative case (Figure 3 and discussion):
  - Example: a country’s debt trades at 60 cents to par before ECB senior creditor status.
    - Case I interpretation: 40 percent default probability with 100 percent loss given default (zero recovery) and 60 percent non-default probability → seniority does not matter.
    - Case II interpretation: 80 percent default probability with 50 percent recovery → seniority matters; if ECB share = 50 percent, ECB can recover full claims at face value while private holders may be left with ex post price of 20 cents given remaining 20 percent probability of non-default.
- Net impact depends on three major factors (Figure 4):
  - Probability of default (PD): increases linearly with the subordination effect; larger ECB market share b lowers ex post market price for any given PD.
  - Loss given default (LGD): if large, seniority provides limited advantage; for small LGD, senior status is less valuable; the overall effect is not monotonic.
  - ECB share (b): when the ECB’s share is larger than the original recovery rate, a kink appears—above which private bond holders receive nothing ex post in bad states. The effect is monotonic but not linear. Expectations about future ECB share matter for pricing; increases in expected ECB share (or related uncertainty) can create further negative effects.

### Theoretical quantification of subordination effects — CDS reduced-form model
- CDS pricing elements and SMP implications:
  - CDS formula used: CDS = LGD x (1 – (1- PD)^{1/n}), where n denotes the number of periods (years).
  - Liquidity effect: credible and sustained SMP interventions lower sovereign yields, reduce cumulative PD, and thus can reduce CDS spreads (non-linear effect).
  - Subordination risk effect: perceived ECB senior creditor status may increase private sector LGD and offset the lower PD.
- Modeling LGD with restructuring needs and ECB share:
  - LGD = (ΔD/GDP)/(1- b), where ΔD/GDP denotes debt/GDP reduction needed and b is ECB share.
  - Substitution yields: CDS = (ΔD/GDP)/(1- b) x (1 – (1- PD)^{1/n}).
  - Implications:
    - An increase in b increases CDS spreads, more strongly when CDS prices are already high (PD elevated); limited effect when CDS premia are low.
    - In dynamic context, SMP credibility matters: subordination may lead to self-fulfilling default dynamics if central bank purchases cannot offset increased private sector LGD caused by ECB seniority.

### Net effects and ambiguity
- SMP purchases have two opposing effects on bond prices and yields:
  - Raising subordination risk: announcement of debt swaps favoring the ECB reflects a net expected transfer of value from private bond holders to the ECB, undermining future SMP effectiveness by increasing issuance costs for bonds not purchased by the ECB.
  - Improving liquidity: SMP can reduce PD by improving liquidity conditions and lowering interest rates, increasing the value of residual bonds—especially if the intervention strategy is credible and clearly communicated.
- The net impact of subordination versus liquidity support on bond prices is ex ante ambiguous.

### Main finding and policy-relevant insight
- Main finding stated in the paper:
  - The impact of ECB seniority is primarily related to perceived probability of default and the proportion of outstanding debt already in the hands of the central bank.
  - Credibility of SMP interventions matters.

*Prepared by IMF staff (sections and figures as provided in the source content).*

### 15.      This SIP illustrates that the SMP may have a subordination effect, but this effect

### _cr12182 - 15.      This SIP illustrates that the SMP may have a subordination effect, but this effect

### Findings on SMP subordination effect
- The SMP may have a subordination effect, but this effect "will be important only at the margin-as was the case in Greece-and depends on the program’s credibility."
- The rise in yields for the most distressed sovereigns at the time of announcement of the ECB’s exemption from the Greek PSI "reflected a net expected transfer of value from private sovereign bond holders to the ECB."
- This de facto subordination "may undermine future SMP interventions because sovereigns may face higher issuance cost on any bonds not purchased by the ECB."
- Theoretical analysis identifies three factors on which the subordination effect depends: probability of default (PD), loss given default (LGD), and the share of ECB bond holdings (b).
- At low PD levels, when LGD is relatively low and the debt market share of SMP is not too high, "subordination risk plays a limited role."
- A CDS model analysis shows further SMP interventions have a negative effect "only when adjustment needs are very high or CDS spreads are already extremely high."
- The ECB’s credibility is critical: if credibility is low, SMP interventions "may be unable to stop self-fulfilling debt default dynamics" because the ECB cannot reduce sovereign spreads or yields, which would otherwise support debt sustainability and lower PD.

### Policy implications and accommodation options
- SMP subordination "currently does not seem to play a large role in pricing and markets."
- To attenuate market fears about SMP subordination (beyond what is captured in the stylized models), one option is "transferring some of the benefits for the ECB back to private sector bond holders."

### Key quantitative example and valuation relationships (Annex)
- Example market observation:
  - Country debt trading at 60 cents on the euro (q0).
  - Market estimates: 80 percent default probability (PD) with 50 percent loss given default (LGD).
  - Overall discount is 40 percent (beyond a risk-free discount assumed to be zero).
  - Total face value F = 100 million euro, so total market value V0 = 60 million euro.
- Core valuation equations (preserve notation):
  - V0 = (1 – PD*LGD)*F
  - q0 = V0 / F
- ECB and private sector initial holdings:
  - If ECB’s share is b percent of total outstanding, then:
    - VE0 = b*V0
    - VP0 = V0 – VE0
  - Example: if b = 20 percent, VE0 is 20 percent of V0; q0 is unchanged by b.
- Ex-post when ECB is senior creditor (effect depends on b and expectations about b):
  - Private sector residual claim: VP1 = V0 – VE1
  - ECB loss given default when senior (LGDE1):
    - LGDE1 = 0, if b (1 – LGD).
    - LGDE1 = 1 – (1 – LGD)/b, otherwise.
  - ECB holding valuation ex-post:
    - VE1 = (1 – PD*LGDE1)*b*F
    - Shadow price facing ECB: qE1 = 1 – PD*LGDE1
  - Private sector price ex-post:
    - qP1 = VP1 / (1 – b)F
- Mechanism summary:
  - When the ECB is shielded from losses (granted senior status), private claims become subordinated and their market value falls; the degree of decline varies with PD, LGD, and b.
  - Expectations and uncertainty about future b from SMP purchases can create additional market distress.

*Source: IMF staff position as presented in the provided content unit.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2012/_cr12182.pdf_