## _060214

## Source details

**Canonical URL:** [_060214](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_060214.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_060214.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_060214.pdf.json)

---

### Executive summary — overview and priority
- Developing an effective framework for cross-border resolution is a key priority in international regulatory reform; large bank failures in the global financial crisis exposed lack of adequate tools for resolving “too-big-to-fail” institutions and absence of robust cross-border cooperation mechanisms.
- The Financial Stability Board’s Key Attributes of Effective Resolution Regimes for Financial Institutions establishes principles calling for:
  - comprehensive resolution powers for authorities; and
  - effective mechanisms for cross-border cooperation and allocation of losses to private stakeholders.
- Consensus exists on a framework, but important details remain to be worked out; as yet, orderly resolution of systemic cross-border banks is not a feasible option.

### Cross-border resolution in practice — concentration, complexity, incentives (key facts)
- Concentration and cross-border footprint:
  - In 2008, 145 banks globally accounted for 85 percent of the assets of the world’s top 1,000 banks.
  - The top five largest cross-border banking groups have $10 trillion of total assets and account for about 10 percent of global banking industry assets.
  - These five groups, all headquartered in Europe, have over 50 percent of their credit risk exposures and employ roughly 60 percent of their staff outside their “home” country, with subsidiaries in more than 60 countries.
- Structural and operational complexity:
  - Large cross-border groups often consist of hundreds of entities in complex webs of financial and operational interdependency, complicating coordinated resolution.
- Incentives and cooperation:
  - Resolution authorities answer to domestic constituencies; incentives to protect domestic stability and public funds can cause unilateral actions and “ring-fencing.”
  - Ex ante MoUs and supervisory college arrangements were often nonbinding, lacked escalation procedures, and were abandoned under domestic pressure.
  - Formal ex ante burden-sharing agreements have had little practical effect and can increase moral hazard.

### Key gaps and areas needing attention (national regimes, resolvability, loss absorbency, creditor hierarchy)
- National resolution regimes:
  - Several jurisdictions adopted far-reaching legal reforms, but progress is mixed; many countries still lack comprehensive resolution powers (including the power to “bail in” creditors) and effective mechanisms for recognition of foreign resolution measures.
- Firm-specific operational resolution strategies:
  - Systemic banks and supervisors are investing heavily in improving resolvability; reaching agreement between home and host supervisors is difficult due to legal impediments and firms’ complex structures.
- Loss absorbency:
  - The Key Attributes call for bank failures’ burden to fall on private creditors while preserving financial stability.
  - Credibility depends on ensuring banks have sufficient liabilities that can absorb losses without destabilizing the system — requires agreement on definition and size of buffers, where they should be located, and intra-group loss-transfer mechanisms.
- Creditor hierarchy and depositor preference:
  - Lack of harmonized creditor hierarchy across jurisdictions obstructs cooperative resolution; clearer guidance is needed on depositor preference.
- Use of public funds:
  - Residual risk that public funds will be needed creates incentives for unilateral action; agreement on buffer location, loss allocation, resolution strategies, and group structure is critical.
  - Structural changes (self-sufficient subsidiaries, restrictions on intra-group flows) should be considered if agreement is not feasible.
  - Cross-border simulation exercises can provide a forum to discuss alternatives to unilateral actions.
  - Efforts needed to ensure ex-post recovery of any potential use of public funds.
- Smaller jurisdictions/entities:
  - Many cross-border banks are not globally systemic but their disorderly resolution could undermine financial stability in home and host countries; reforms must be adaptable to different complexity levels and account for impacts on smaller hosts.

### Policy priorities and recommended actions
- Complete implementation of the Key Attributes by jurisdictions; all FSB members committed to implement by the end of 2015.
- Strengthen national legal powers and recognition mechanisms:
  - establish comprehensive resolution powers (including bail-in) and mechanisms for recognition of foreign resolution measures.
- Develop and agree operational, firm-specific resolution strategies:
  - overcome legal impediments and address complex operational and financial structures through joint home–host planning.
- Clarify loss-absorbing capacity and creditor hierarchy:
  - agree on definitions and sizing of buffers, placement within group structures, intra-group loss transfer mechanisms, and harmonization of creditor hierarchies (including depositor preference).
- Align incentives to reduce unilateral actions:
  - work toward burden-sharing agreements, consider subsidiarization or restrictions on intra-group flows where cooperation is not feasible, and conduct cross-border simulation exercises.
- Address needs of smaller hosts:
  - ensure proportional application of the Key Attributes and adapt frameworks to domestic and regional systemic banks’ risks.
- IMF role:
  - the Fund will continue to work with members via surveillance, technical assistance, program work, and the FSAP; advise members to strengthen resolution and crisis management frameworks in line with the Key Attributes.

### Implementation status and institutional arrangements
- 2013 FSB peer review: implementation within FSB membership at an early stage; regimes for banks more developed than for other financial institutions.
- Recent developments:
  - European Union adopted the Bank Recovery and Resolution Directive in May 2014.
  - Firm-specific CMGs set up for all G-SIFIs; preferred resolution strategies (SPE, MPE, or hybrid) being discussed and operationalized.
  - Firm-specific Cooperation Agreements remain limited and focus largely on procedural aspects.
- Fund engagement:
  - IMF/Fund to assess members’ implementation under the ROSC program once methodology is finalized; Fund provides advice on resolution authorities, early intervention, resolution powers and tools, safeguards, legal protections, resolution funding, judicial review limits, and information sharing.

### Institution-specific resolution strategies: SPE, MPE, Hybrid (features and implications)
- Single Point of Entry (SPE):
  - resolution at apex holding company; shareholders and apex creditors absorb group-wide losses by write-down/restructuring; capital freed at apex “passed down” to recapitalize subsidiaries.
  - Requires ability to downstream loss-absorbing capacity and willingness/capacity to provide liquidity support — dependent on high cooperation and trust among authorities.
- Multiple Point of Entry (MPE):
  - operating affiliates resolved separately in different jurisdictions; losses dealt with at subsidiary level.
  - Requires extensive preparation and coordination to preserve essential services and avoid disruption.
- Hybrid strategies:
  - combinations possible (e.g., core operations via SPE, stand-alone operations via MPE); RRPs help map losses/liquidity needs and inform burden-sharing.

### GLAC (Gone-Concern Loss-Absorbing Capacity): objectives and design issues
- Purpose:
  - ensure a credible buffer of liabilities beyond equity that can be mandatorily restructured in resolution and promote a uniform international approach.
- Definition of eligible claims:
  - likely eligible liabilities: subordinated debt, senior unsecured bonds, and intra-group loans; uninsured deposits, interbank loans, or other liabilities present risks.
- Amount of GLAC:
  - sufficient “bail-inable” liabilities needed to resolve distressed institutions; fiscal costs of past crises averaged 13 percent of financial system assets.
- Location of GLAC:
  - differences between where losses are generated and where GLAC is located complicate resolution; market depth and tax considerations influence issuance location.
  - If GLAC is not located where losses arise, mechanisms are needed to move losses to GLAC or GLAC to losses.
- Appropriate investor base and market effects:
  - GLAC should be held by creditors capable of absorbing losses without socially undesirable consequences; limit cross-holdings of senior unsecured debt by other financial institutions.
  - Credible bail-in perceptions may raise funding costs for targeted instruments; conversion into equity may create new controlling shareholders with regulatory suitability implications.
- Uniformity and disclosure:
  - uniform approach across jurisdictions improves level playing field; GLAC triggering should not discriminate on investor nationality.
  - banks’ publication of GLAC amounts and intended use could enhance market pricing and creditor discipline.
- FSB proposal:
  - a set of principles on GLAC was to be presented at the G20 summit in November 2014 (as noted in the source).

### GLAC and intra-group loss allocation (amount, location, mechanisms)
- The amount and location of GLAC determine ex ante how losses accrue in resolution and align home and host authorities’ incentives for private burden sharing.
- Intra-group GLAC (e.g., subsidiary debt held by parent) must be understood to know how losses will be absorbed and transmitted.
- MPE: each point of entry must issue sufficient GLAC for orderly resolution.
- SPE: clarity required on how losses will be imposed on GLAC-issuing entities; intra-group loss-transfer agreements necessary.
- Identified intra-group loss-transfer mechanisms:
  - conversion of parent-to-subsidiary loans into equity;
  - collateralized guarantees provided by the parent on subsidiaries’ borrowing.
- Key implementation question: how, when, and by whom such mechanisms will be activated must be addressed in loss-transfer agreements.

### Creditor hierarchy, depositor preference, and legal risks
- Creditor hierarchy principles:
  - losses first against shareholders, then subordinated debt holders and other creditors in ascending order; pari passu treatment generally applies within a class.
  - Going-concern resolutions may require departures from liquidation hierarchy (e.g., exempting certain claims ex ante or treating equal-ranking claims unequally) — such departures raise litigation risk mitigated by safeguards like “no creditor worse off.”
- Depositor preference (DP) forms:
  - Insured depositor preference: insured deposits preferred (DIS by subrogation); uninsured deposits pari passu with senior unsecured creditors.
  - General depositor preference: all deposits preferred over senior unsecured creditors; DIS subrogated for insured deposits and pari passu with uninsured deposits.
  - Tiered depositor preference: insured deposits (and DIS via subrogation) preferred over uninsured deposits, both preferred over senior unsecured creditors.
- DP trade-offs:
  - DP can mitigate contagion and reduce risk to DIS; general or tiered DP can clarify legal grounds for protecting deposits over senior unsecured creditors, potentially reducing legal challenges.
  - Potential cost: DP can affect bank funding costs or availability of unsecured wholesale funding.
- Jurisdictions with some form of DP (examples listed): Argentina, Australia, China, India, Hong Kong SAR, Indonesia, Mexico, Russia, Singapore, Switzerland, Turkey, the United States, and several European countries; recent reforms in the United Kingdom and the European Union.
- International guidance gaps:
  - Key Attribute 5.1 and 7.4 provide guidance but leave questions unanswered, particularly treatment of insured and uninsured deposits — further international guidance on creditor hierarchies and deposits is needed.

### Minimizing public sector burden and residual risks
- Regulatory objective: reduce probability of bank failure and fiscal costs if failures occur.
- Role of buffers:
  - increased capital and GLAC requirements impose market discipline and provide time for remedial action; GLAC can absorb losses before public funds are needed for solvency support.
- Residual public funding risk:
  - emergency liquidity assistance can leave central bank exposed; state guarantees or capping asset portfolio losses may be needed to ensure market access.
  - if GLAC cannot cover losses (or would cause contagion), solvency support for systemic institutions may be needed—but efforts should aim for ex-post recovery from the industry.
- Cross-border complications:
  - central banks can provide unlimited support only in their own currency; swap arrangements are required for other currencies, exposing counterpart central banks to foreign exchange risk and foreign collateral exposures.
  - realistic public-funding risk incentivizes ring-fencing.

### Ex-ante burden-sharing, planning, structural assessment, and tools to reduce imbalances
- Ex-ante public support agreements are difficult; Nordic-Baltic ex-ante burden-sharing framework cited as a non-binding example.
- Pre-crisis measures:
  - RRPs and resolvability assessments identify structural imbalances (net A/L position by country; asset encumbrance; asset quality; intra-group liquidity by currency and country; intra-group capital and GLAC adequacy; DIS/RF support needed; public costs needed; FX liquidity and foreign collateral needed).
  - Information exchange alone is insufficient; knowledge of asset concentration and GLAC locations helps generate common understanding and may prompt structural adjustments.
- Possible ex-ante measures:
  - reducing upfront exposure by limiting cross-border risk transfer (structural adjustments, regulated placement of assets/liabilities, prudential limits);
  - increasing acceptability of imbalances through agreement on emergency liquidity provision or residual solvency support at group level.
- Three guiding policy questions for bank configuration and resolution strategy:
  - To what extent can the burden of a bank’s failure be contained within the private sector?
  - What burden could remain with the public sector and how can this be minimized or eliminated?
  - What coordinated policies can minimize ex-post ring-fencing or unilateral decisions?
- System-wide perspective is needed where multiple contagion channels exist; example: U.K. and U.S. joint support for SPE strategies predicated on high cross-system costs of non-cooperation.

### Cross-border simulation exercises, subsidiarization, and cooperation trade-offs
- Cross-border simulation exercises:
  - provide forum to discuss alternatives to unilateral actions and address core questions:
    - (i) Would an ailing bank be allowed to keep up cross-border capital and funding flows?
    - (ii) Which sources of funding could be made available in crisis? Under what conditions?
    - (iii) Which claims would a country accept to be bailed in?
    - (iv) What size of imbalances would a country be comfortable accepting, and under what circumstances?
- Cooperation outcomes range from full ex ante ring-fencing and maximum resolvability to arrangements preserving shared critical functions.
- Subsidiarization vs. ring-fencing:
  - Subsidiarization ex ante may be superior to ring-fencing ex post when cross-border cooperation is unlikely; hosts may require local capital and liquidity ex ante.
  - Stand-alone subsidiarization can reduce hosts’ incentives to ring-fence by increasing local control and understanding of assets/liabilities.
  - Costs of isolating capital/liquidity include efficiency losses and impeded cross-border intermediation; costs are lower when operations are locally funded and assets are straightforward loans.
- For heavily interconnected markets, formal cooperation may be preferable; more effective dialogue, information sharing, and cooperation agreements reduce authorities’ concern about bank structure choices.

### Applicability of Key Attributes beyond G-SIFIs and small-host perspective
- Key Attributes primarily designed for G-SIFIs and advanced systems; many cross-border banks are not globally systemic.
- Implications for regional/domestic SIFIs:
  - Tailor reforms to system complexity; RRPs and resolvability assessments should be required for domestic SIFIs.
  - Crisis Management Groups or comparable arrangements should be required where firms are systemic in more than one jurisdiction.
  - Cooperation Agreements should be required where CMGs or comparable arrangements are needed.
- Loss-absorbing capacity in deposit-funded systems:
  - Regional/domestic systemic banks should have adequate LAC to reduce risk to DIS/public sector.
  - In less developed capital markets, substitution of extra capital for uninsured deposit loss-absorbency may be appropriate; retail deposits are unsuitable for absorbing losses.
- Small-host priorities:
  - small hosts may face negotiation asymmetries and limited participation in CMGs; resolution strategies should prioritize domestic financial stability: contain contagion, maintain critical services, minimize economic costs.
  - If ex ante cooperation is not credible, structural requirements (subsidiarization, local capital and liquidity, asset maintenance for branches) are appropriate and align with MPE or hybrid strategies.
  - Cooperation with home jurisdiction remains important for GLAC, critical services, and building trust.

### Conclusions, remaining gaps, and international work program
- Progress:
  - international reforms moved toward an effective cross-border resolution framework; full implementation of Key Attributes will provide authorities with tools to execute resolution strategies and place costs on shareholders and private creditors.
- Remaining gaps and priorities:
  - Define GLAC specifics (amount, eligible claims, location, investor base).
  - Better align creditor hierarchies and depositor preference between countries.
  - Minimize residual public-fund risks; credible bail-in options reduce but do not eliminate public-support risk.
  - Move decisions and actions to pre-crisis phase: identify structural imbalances, assess costs/benefits of bank structures and resolution strategies, and monitor potential cross-border sovereign risk transfer or public costs.
- Fund engagement:
  - the Fund will promote Key Attributes adoption, participate in FSB work streams to address gaps, and keep the Executive Board apprised.

### Annex I — selected public interventions in cross-border banks (summarized cases and key numbers)
- Citigroup:
  - assets around $2 trillion; 200 million customer accounts; by 2008 about 70 percent of revenue from foreign assets.
  - 2008 losses about $28 billion.
  - U.S. Treasury investment: $25 billion (TARP) on October 28, 2008; additional support announced November 23, 2008 including guarantee covering potential losses on more than $300 billion of assets and $20 billion TARP investment; Treasury sold investment by December 2010 at a profit of $12 billion.
- Dexia:
  - total assets about €650 billion pre-crisis; balance sheet growth about 150 percent during 2000–08.
  - short-term liquidity gap estimated at €260 billion in October 2008.
  - 2008 recapitalization by Belgium, France, Luxembourg: €3 billion plus private shareholder contribution; joint guarantees up to €150 billion.
  - Second intervention October 2011: orderly runoff strategy; Belgian acquisition of Belgian bank for €4 billion; new guarantee ceiling €85 billion; new burden-sharing key: Belgium 51.4 percent, France 45.6 percent, Luxembourg 3 percent (noting shifts from previous shares).
  - 2012 public capital injections in Dexia SA: €5.5 billion (Belgium €2.9 billion; France €2.5 billion).
- Fortis:
  - end-2007 total assets approximately €870 billion.
  - initial capital injection September 28, 2008: €11.2 billion (Belgium €4.7 billion; Netherlands €4 billion; Luxembourg €2.5 billion convertible loan).
  - Netherlands nationalized Dutch operations at cost €16.8 billion (October 3, 2008); total Dutch cost currently valued at €32 billion.
- ING:
  - total assets some €1.3 trillion end-2007; U.S. Alt-A exposure €27.5 billion at end-2007.
  - Dutch authorities provided €10 billion in Core Tier-1 Securities in 2008.
  - Illiquid Asset Back-Up Facility unwound early-2014 with cash profit €1.4 billion for government; >90 percent of €10 billion capital support repaid, remainder expected 2015.
- Icelandic banks:
  - combined balance sheet of Kaupthing, Glitnir, Landsbanki: €16 billion in 2003; by crisis banks’ assets were 10 times Icelandic GDP.
  - immediate funding needs €5 billion vs. reserves €2.6 billion; international deposits at risk €15 billion.
  - emergency legislation (Act 125/2008) and capital controls in October 2008.
- KBC:
  - 3Q 2008 losses €0.9 billion; €7 billion capital injection in 2008; asset relief guarantees €20 billion in 2009; subsequent repayments and private capital raisings listed (reimbursements and instruments with exact amounts).
- Hypo Alpe Adria (HAA):
  - liabilities under Carinthia guarantee about €20 billion; total assets €38 billion in 2007.
  - multiple rounds of capital and guarantee support from BayernLB and Austria (exact amounts listed in the source).
- Laiki Bank (Cyprus Popular Bank):
  - total assets end-2010 €43 billion (about 250 percent of Cypriot GDP); Greek-related losses €2.3 billion in 2011.
  - capital need €1.8 billion to reach EBA Core Tier-1 9 percent by end-June 2012; nationalization resulted with government acquiring 84 percent stake.
  - March 2013 resolution involved transfer of Greek-related assets to Piraeus Bank and recapitalization of Bank of Cyprus with shareholders, bank debt holders, and partial conversion of uninsured deposits into equity.
- RBS:
  - October 2008–December 2009: U.K. Treasury purchased RBS shares totaling £45.5 billion.
  - Asset protection scheme covered £282 billion; RBS posted £34 billion loss at end-2008; U.K. government retained 79 percent stake as of February 2014.
- UBS:
  - prior to crisis total assets almost $2 trillion; mid-2008 write-downs €43 billion.
  - Swiss recapitalization: mandatory convertible notes SwF 6 billion; StabFund to take up to $60 billion troubled assets with SNB financing 90 percent and UBS providing 10 percent capital.
  - CPFF borrowing cumulative $75 billion; bad-bank transfers $39 billion; wind-up completed late 2013 with proceeds distribution amounts given.

### Annex II — European Banking Union and BRRD (scope, elements, and key numeric provisions)
- Banking Union elements:
  - single supervisory mechanism (SSM);
  - single resolution mechanism (SRM), including a single resolution fund;
  - harmonized legal framework for deposit insurance.
- Gap: lack of a common fiscal backstop for additional funds needed to facilitate resolution of systemic institutions.
- European Stability Mechanism (ESM):
  - can be used, in theory, to recapitalize systemic banks by direct equity injections jointly with applying member state, subject to conditions.
  - ESM direct recapitalization conditions include prior bail-in, infeasibility of private recapitalization, bank meeting minimum capitalization (common equity tier 1 equal to 4.5 percent of risk-weighted assets), and total ESM funds available for direct recapitalization capped at €60 billion.
- Bank Recovery and Resolution Directive (BRRD) — approved April 2014:
  - scope: credit institutions and certain financial holding companies established in EU member states and EEA members.
  - resolution powers include sale of business, transfer to bridge institution, asset separation, bail-in (with exceptions for insured deposits and liabilities with maturity less than seven days), management replacement, and temporary stays on payment of financial claims.
  - Recovery and Resolution Plans: institutions and authorities required to prepare RRPs; authorities can require restructurings to facilitate resolvability.
  - Creditor hierarchy under BRRD (waterfall):
    (i) covered deposits, and the deposit insurance scheme by subrogation to covered depositors;
    (ii) eligible deposits of natural persons and micro and SMEs exceeding deposit insurance coverage;
    (iii) deposits in foreign branches of EU banks;
    (iv) claims of ordinary unsecured, non-preferred creditors; and (iv) shareholders.
  - Recognition of third-country resolution proceedings:
    - in absence of agreement with a non-EU authority, recognition may be taken jointly by the European resolution college (if established) or by each resolution authority, granting more flexibility than the Winding-Up Directive.
  - Funding for resolution:
    - additional funding needs after bail-in generally provided through national resolution funds with bank contributions.
    - for euro area member states, national funds will be pooled into the proposed €55 billion Single Resolution Fund to be operational by 2016; contributions expected over an eight-year period.

*Source: INTERNATIONAL MONETARY FUND — CROSS-BORDER RESOLUTION (excerpt from _060214).*

### EXECUTIVE SUMMARY

### EXECUTIVE SUMMARY

### Overview
- Developing an effective framework for cross-border resolution is a key priority in international regulatory reform. Large bank failures during the global financial crisis exposed the lack of adequate tools for resolving “too-big-to-fail” institutions and the absence of robust cross-border cooperation mechanisms.
- The Financial Stability Board’s Key Attributes of Effective Resolution Regimes for Financial Institutions has established an agreed set of principles and best practices, calling for resolution regimes that give authorities comprehensive resolution powers and establish effective mechanisms for cross-border cooperation and allocation of losses to private stakeholders.
- Consensus on a framework exists, but important details remain to be worked out; as yet, orderly resolution of systemic cross-border banks is not a feasible option.

### Cross-border resolution in practice — key facts and challenges
- Concentration and cross-border footprint:
  - In 2008, 145 banks globally accounted for 85 percent of the assets of the world’s top 1,000 banks.
  - The top five largest cross-border banking groups have $10 trillion of total assets and account for about 10 percent of global banking industry assets.
  - These five groups, all headquartered in Europe, have over 50 percent of their credit risk exposures and employ roughly 60 percent of their staff outside their “home” country, with subsidiaries in more than 60 countries.
- Structural and operational complexity:
  - Large cross-border financial groups often consist of hundreds of entities in complex webs of financial and operational interdependency, complicating coordinated resolution.
- Incentives and cooperation:
  - Resolution authorities are accountable to domestic constituencies; incentives to protect domestic financial stability and public funds can lead to unilateral actions and “ring-fencing.”
  - Ex ante MoUs and supervisory college arrangements proved insufficient during the crisis: MoUs were nonbinding, lacked escalation procedures, and were often abandoned under domestic pressure (see Box 1).
  - Formal ex ante burden-sharing agreements have had little practical effect; they can be difficult to credibly implement and may increase moral hazard.

### Key gaps and areas needing attention (preserving source language and structure)
- National resolution regimes:
  - Several jurisdictions have adopted far-reaching legal reforms, but the reforms are complex and progress has been mixed overall.
  - Many countries still lack comprehensive resolution powers for banks and other financial institutions (including the power to “bail in” creditors), and effective mechanisms for the recognition of foreign resolution measures.
  - The absence of these pre-requisites makes effective resolution and cross-border cooperation difficult.
- Firm-specific operational resolution strategies:
  - Systemic banks and their supervisors are investing heavily in improving resolvability.
  - Reaching agreement between home and host supervisors on how such entities would be dealt with in a crisis has proven difficult, in particular because of legal impediments to cross-border cooperation and the complexities of these firms’ operational and financial structures.
- Loss absorbency:
  - The Key Attributes call for the burden of bank failure to fall firmly on the private creditors of banks while preserving financial stability.
  - The credibility of this commitment depends on ensuring that banks have sufficient liabilities that can absorb losses without destabilizing the financial system.
  - This requires agreement on the definition and size of buffers, where they should be located in a bank’s structure, and intra-group mechanisms for transferring losses.
- Greater harmonization of creditor hierarchies:
  - The lack of a harmonized approach to creditor hierarchy across jurisdictions is an important impediment to cooperative approaches to resolving a cross-border bank.
  - Clearer guidance is particularly needed on depositor preference—i.e., the ranking to be followed with respect to bank deposits.
- The use of public funds:
  - There will always be the risk that public funds will be needed, even if only temporarily, to preserve financial stability.
  - This risk creates powerful incentives for unilateral action and can undermine cooperative approaches.
  - Achieving agreement on the location of buffers and loss allocation during resolution, resolution strategies, and group structure will be critical in ensuring that incentives for cooperation are well aligned.
  - If agreement proves not feasible, consideration should be given to structural changes, such as establishment of self-sufficient subsidiaries and/or restrictions on intra-group flows to ensure entities can operate—and be resolved—on a stand-alone basis.
  - Cross-border simulation exercises could provide a forum for discussing alternatives to unilateral actions.
  - Efforts are needed to ensure ex-post recovery of any potential use of public funds.
- Smaller jurisdictions/entities:
  - Many cross-border banks are not globally systemic, but their disorderly resolution could undermine financial stability in home and host countries.
  - Reforms to resolution frameworks will need to be adaptable to different degrees of complexity and account for potential impacts on smaller host jurisdictions.

### Policy priorities and recommended actions
- Complete implementation of the Key Attributes by jurisdictions:
  - All FSB members have committed to implement the Key Attributes by the end of 2015—achieving this will require substantial additional work at the country level.
- Strengthen national legal powers and recognition mechanisms:
  - Ensure countries establish comprehensive resolution powers (including bail-in) and effective mechanisms for the recognition of foreign resolution measures.
- Develop and agree operational, firm-specific resolution strategies:
  - Overcome legal impediments to cross-border cooperation and address complex operational and financial structures through joint planning between home and host supervisors.
- Clarify loss-absorbing capacity and creditor hierarchy:
  - Reach agreement on definitions and sizing of buffers, their placement within group structures, intra-group loss transfer mechanisms, and harmonization of creditor hierarchies (including depositor preference).
- Align incentives to reduce unilateral actions:
  - Work toward agreements on burden sharing, consider structural measures (subsidiarization, restrictions on intra-group flows) where cooperation is not feasible, and conduct cross-border simulation exercises.
- Address needs of smaller hosts:
  - Ensure the proportional application of Key Attributes and adapt frameworks to the specific risks facing smaller jurisdictions and non-global-systemic banks.
- IMF role:
  - The Fund will continue to work with members on strengthening resolution regimes through surveillance, technical assistance, and program work, advising members to strengthen resolution and crisis management frameworks in line with the Key Attributes, and supporting the international reform agenda.

*Source: INTERNATIONAL MONETARY FUND — EXECUTIVE SUMMARY.*

### 10.      In sum, the crisis exposed three major gaps in the international framework for

### _060214 - 10.      In sum, the crisis exposed three major gaps in the international framework for

### Major gaps exposed by the crisis
- Three principal gaps in the international framework for resolution were identified:
  - Inadequate resolution powers and tools to resolve a failing financial institution or group in an orderly and least-cost fashion.
  - Inadequate frameworks for enforcement of resolution measures on a cross-border basis and for coordination between authorities in different countries in the event of a cross-border failure.
  - Inadequate mechanisms for loss allocation, including the ability to credibly subject bank creditors to loss in a bank failure without endangering financial stability.

### Purpose and scope of the Key Attributes
- Objective:
  - Enable orderly resolution while ending distortions from “too-big-to-fail” perceptions and contain moral hazard by making orderly resolution credible for large financial groups.
- Core features:
  - Establish a new international standard for the design of resolution regimes for systemically important financial institutions and groups.
  - Call for resolution authorities to have a comprehensive toolkit of powers, including:
    - Power to take control of a failing financial institution.
    - Power to merge ailing banks with stronger institutions.
    - Power to transfer assets and liabilities to a healthy institution or a “bridge bank.”
    - Power to engage in mandatory debt restructuring or “bail-in” (writing down unsecured debt and/or converting it to equity while the institution remains open).
  - Powers to be exercisable without the need to obtain the consent of shareholders or creditors, subject to appropriate safeguards.
- Cross-border cooperation requirements:
  - Empower and strongly encourage resolution authorities to achieve cooperative solutions with foreign counterparts.
  - Empower resolution authorities to share information with foreign counterparts with protections for confidential information.
  - Provide transparent and expedited processes to give effect to foreign resolution measures.
- Recovery and Resolution Plans (RRPs) and Crisis Management:
  - Jurisdictions are required to put in place RRPs covering, at minimum, domestically incorporated firms that could be systemically significant or critical.
  - For G-SIFIs: home and key host authorities must maintain Crisis Management Groups (CMGs) and institution-specific Cooperation Agreements to enable execution of agreed resolution strategies and information exchange.

### Implementation status and challenges
- Implementation snapshot:
  - A 2013 FSB peer review found implementation within FSB membership at an early stage.
  - Resolution regimes were generally more developed for banks than for other financial institutions, but even for banks many lacked powers such as bail-in or the ability to temporarily suspend early termination rights under financial contracts.
  - Non-bank resolution powers (e.g., insurance) were, in many cases, very limited.
  - Few jurisdictions had robust mechanisms to give effect to foreign resolution measures or clear statutory mandates to cooperate/share information with foreign authorities.
- Recent developments:
  - The European Union adopted the Bank Recovery and Resolution Directive in May 2014, explicitly intended to align national frameworks with the Key Attributes.
  - Firm-specific CMGs have been set up for all global systemically important banks designated by the FSB; preferred resolution strategies (SPE, MPE, or hybrid) are being discussed and operationalized.
  - Firm-specific Cooperation Agreements remain limited in coverage and focus largely on procedural aspects (e.g., information exchange), not detailed resolution strategies.
- Political and supervisory actions:
  - All FSB members have committed to fully implement the Key Attributes by the end of 2015.
  - The FSB will support implementation through guidance, regular monitoring, and peer reviews.
  - The IMF/Fund engages members through program support, surveillance, technical assistance, and the FSAP, and will assess members’ implementation of the Key Attributes under the ROSC program once the assessment methodology is finalized.

### Institution-specific resolution strategies (SPE, MPE, Hybrid)
- Single Point of Entry (SPE):
  - Resolution occurs at the top of the financial group; operating parts are preserved.
  - Home resolution authority restructures the apex holding company; shareholders and creditors of the apex absorb group-wide losses through write-down or restructuring of equity and/or debt.
  - Capital freed at the apex is “passed down” to recapitalize and provide liquidity to loss-making subsidiaries.
  - Requires ability to downstream loss-absorbing capacity and willingness/capacity to provide liquidity support — dependent on high cooperation and trust among authorities.
- Multiple Point of Entry (MPE):
  - Operating affiliates are resolved separately in different jurisdictions by respective resolution authorities.
  - Losses are dealt with at the subsidiary level.
  - Requires extensive preparation and coordination to avoid disruption and preserve essential services and functions.
- Hybrid strategies:
  - Possible combinations (e.g., core operations resolved under SPE while stand-alone operations follow MPE).
  - Choice of strategy must be consistent with group structure; RRPs assist in mapping where losses/ liquidity needs may arise and inform burden-sharing discussions.

### Fund advice to members on resolution frameworks
- Typical areas of Fund advice include:
  - Resolution Authority: Establishing a dedicated and autonomous resolution authority with mandate to preserve financial stability and continuity of systemically important services.
  - Early Intervention: Strengthening powers for intervention before balance-sheet insolvency.
  - Resolution Powers and Tools: Broad restructuring powers, including transfers without shareholder or creditor approvals.
  - Safeguards: Creditor safeguards to respect claim hierarchy and ensure creditors are no worse off than under liquidation.
  - Legal Protection: Legal protection for officers, staff, and agents of resolution authorities for good-faith actions.
  - Resolution Funding: Explicit provision for emergency liquidity assistance with safeguards and use of deposit insurance funds to facilitate resolution.
  - Judicial Review: Legal frameworks that do not endanger effective implementation or legal certainty through appeals.
  - Information Sharing and Cooperation: Legal authority for information sharing with foreign authorities and clear mandates for cross-border cooperation.

### Enhancing mechanisms for loss allocation and making bail-in credible
- Key concerns:
  - Need to design how losses are allocated to private stakeholders and manage residual risks that public funds will be needed—failure to do so could incentivize unilateral actions by resolution authorities in crises.
- Making bail-in credible:
  - Holders of claims targeted for bail-in must be able to absorb losses without generating systemic risk themselves.
  - Financial groups need sufficient buffers of bail-inable claims so bail-in is both legally feasible and credible.
  - Doubts about absorptive capacity of claims could increase incentives to ring-fence, especially under SPE strategies.
- FSB proposal: Gone-Concern Loss-Absorbing Capacity (GLAC)
  - Purpose: Ensure a credible buffer of liabilities beyond equity that can be mandatorily restructured in resolution and promote a uniform international approach.
  - A set of principles on GLAC to be presented at the G20 summit in November 2014, expected to cover:
    - Definition of claims regarded as GLAC:
      - Although any unsecured liability could theoretically be bailed in, the most likely eligible liabilities are subordinated debt, senior unsecured bonds, and intra-group loans, given risks related to uninsured deposits, interbank loans, or other liabilities.
      - Perception of credible bail-in for such claims may raise funding costs for these instruments; funding costs could fall for liabilities less likely to suffer loss.
    - Amount of GLAC:
      - Sufficient “bail-inable” liabilities needed to ensure mandatory debt restructuring can resolve distressed institutions in an orderly manner and potentially forestall runs by short-term creditors.
      - This depends partly on past crisis experience: fiscal costs of crises averaged 13 percent of financial system assets.
    - Location of GLAC:
      - Differences between where losses are generated and where loss-absorbing liabilities are located can complicate cross-border resolution.
      - Market depth and tax considerations influence where marketable bank liabilities are issued.
      - If GLAC is not located where losses are most likely to arise, mechanisms are needed to move losses to GLAC or GLAC to losses; choice depends on bank structure and resolution strategy.
    - Appropriate investor base:
      - GLAC should be held by creditors capable of absorbing losses without socially undesirable consequences.
      - Mitigating contagion risk may require limiting cross-holdings of senior unsecured debt by other financial institutions.
      - Secondary-market trading may cause claims to migrate to investors perceived as needing protection from bail-in.
      - Conversion of GLAC into equity may create new controlling shareholders who may or may not meet regulatory suitability requirements.
    - A uniform approach:
      - Some jurisdictions limit mandatory debt restructuring/GLAC to systemic banks (e.g., the United States) while others plan broader application (e.g., the European Union); a common approach improves level playing field.
      - Any GLAC framework should not discriminate on the basis of investor nationality when triggering bail-in.
    - Disclosure:
      - Publication by banks of how much GLAC they have issued and how it would be used could enhance fair market pricing and creditor discipline.

*Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_060214.pdf*

### 25.      The amount and location of GLAC determine ex ante how losses accrue in resolution,

### _060214 - 25.      The amount and location of GLAC determine ex ante how losses accrue in resolution,

### GLAC (gone-concern loss-absorbing capacity) and intra-group loss allocation
- The amount and location of GLAC determine ex ante how losses accrue in resolution, aligning incentives of home and host authorities with respect to private burden sharing.
- Intra-group GLAC (such as debt of subsidiaries held by the parent) must be understood by resolution authorities to know how losses will be absorbed and transmitted within financial groups.
- MPE approaches: each point of entry will need to issue sufficient GLAC to allow for orderly resolution.
- SPE approaches: clarity is required on how losses will be imposed on the entities within the group that issue GLAC, which will require intra-group loss-transfer agreements.
- Examples of intra-group loss-transfer mechanisms identified:
  - conversion of loans from the parent company to the subsidiary into equity;
  - collateralized guarantees provided by the parent on borrowing by the subsidiaries.
- Key implementation question: how, when, and by whom such mechanisms will be activated must be addressed in loss-transfer agreements.

### Creditor hierarchy: role, differences, and legal risks
- Creditor hierarchies determine the ranking/order in which losses will be allocated against the claims of various stakeholders in the liquidation of a firm.
- Loss allocation order described: losses are first allocated against shareholders, before subordinated debt holders and other types of creditors in ascending order.
- Claims in the same class rank “pari passu” and are treated the same in liquidation (generally not possible for the liquidator to pay one but not the other).
- The creditor hierarchy for liquidation of a bank generally resembles ordinary corporate insolvency but must address bank-specific creditors: small depositors, the deposit insurance scheme, and the public sector providing liquidity or solvency support.
- In going-concern restructuring, priorities generally take into account liquidation order and treat claims in the same class equally, but resolution authorities may need to depart from general principles by:
  - exempting certain claims ex ante from loss imposition in a going-concern resolution even if they would suffer losses under liquidation hierarchy; or
  - empowering resolution authorities to treat equally ranking claims in an unequal manner in a specific institution’s resolution.
- Legal risk: unequal treatment or deviation from established creditor hierarchy raises risk of litigation by affected creditors; adequate safeguards such as a “no creditor worse off” requirement can mitigate this risk.
- Significant differences in creditor hierarchies between jurisdictions can obstruct cross-border resolution of branches:
  - If home and host rankings differ considerably, host authorities may decline to cooperate and may ring-fence rather than contribute local assets/liabilities to resolution per home law.
  - Incentives for cross-border cooperation are undermined if home country creditor hierarchy discriminates against foreign or nonresident creditors or creditors of foreign branches.
- International guidance gaps: Key Attributes provide guidance but leave important questions unanswered, particularly on the treatment of insured and uninsured deposits:
  - Key Attribute 5.1: “resolution powers should be exercised in a way that respects the hierarchy of claims,” and that “equity should absorb losses first, and no loss should be imposed on senior debt holders until subordinated debt has been written off entirely.” Key Attribute 5.1 allows departures from pari passu treatment only if necessary to contain systemic impact or maximize value for all creditors, advocating transparency.
  - Key Attribute 7.4: national resolution frameworks should not discriminate on the basis of the nationality of creditors, the location of their claim, or the jurisdiction in which it is payable.
- Conclusion: further guidance at the international level on creditor hierarchies—particularly treatment of deposits—is needed to facilitate bank resolution, especially cross-border.

### Depositor preference (DP): forms, effects, and trade-offs (Box 4)
- Definition: Depositor preference gives depositors a preferential claim on the assets of a failed bank; reflected in liquidation hierarchy and relevant in resolution.
- Common DP approaches described:
  - Insured depositor preference: preferential treatment for insured deposits (and the DIS through subrogation); uninsured deposits rank pari passu with senior unsecured creditors.
  - General depositor preference: preference to all deposits, including balances above the deposit insurance limit, over senior unsecured creditors; DIS is subrogated for insured deposits and ranks pari passu with uninsured deposits.
  - Tiered depositor preference: prefers insured deposits (and the DIS through subrogation) over uninsured deposits, and prefers both over senior unsecured creditors.
- DP benefits and implications:
  - DP can mitigate contagion by increasing depositors’ probability of recovering claims in a bank failure.
  - For countries with a DIS subrogated for insured deposits, insured or tiered DP mitigates risk of the DIS suffering losses, contributing to shifting resolution burden to private sector.
  - General or tiered DP can create clear legal grounds for preferential treatment of all deposits over other unsecured senior creditors, signaling which liabilities will be protected in resolution and potentially reducing legal challenges when transferring deposits or bail-in of non-deposit creditors.
  - Potential cost: changing creditor hierarchy under DP must be weighed against potential impact on bank funding costs or availability of unsecured wholesale funding.
- Jurisdictions with some form of DP (examples listed): Argentina, Australia, China, India, Hong Kong SAR, Indonesia, Mexico, Russia, Singapore, Switzerland, Turkey, the United States, and several European countries; similar reforms recently introduced in the United Kingdom and the European Union.

### Minimizing the potential burden for the public sector
- Regulatory reform objectives: reduce probability of bank failure and lower fiscal costs if failures occur.
- Post-crisis constraints: less capacity and little appetite to put future fiscal resources at risk to stem failure consequences of large financial institutions.
- Intended role of buffers:
  - Increased capital and GLAC requirements impose market discipline and better incentives for managers.
  - Thicker equity buffers bolster confidence and give more time for remedial action.
  - GLAC can absorb losses before public funds may be needed, at least regarding solvency support.
- Residual public funding risk:
  - Emergency liquidity assistance may be offered on terms or against collateral that leave the central bank with some risk of loss.
  - State guarantees or capping losses on asset portfolios may be needed to ensure continued market access.
  - If GLAC cannot cover potential losses (or would cause fresh contagion), solvency support for systemic institutions may be needed—even if to be recovered later from the industry.
- Mobilizing private-sector funding is unlikely once GLAC is exhausted due to short timeframes and legal frameworks that often do not enable private-sector “debtor in possession” financing.
- Liquidity and solvency support complicate cross-border resolution:
  - Liquidity support: central banks can provide unlimited support only in their own currency; swap arrangements are needed for other currencies, exposing counterpart central banks to foreign exchange risk. Central banks may need to accept foreign collateral, leaving them exposed if the foreign jurisdiction ring-fences or takes unilateral action.
  - Solvency support: realistic possibility of public funds being at risk may incentivize ring-fencing to minimize uncertainty and fiscal risks.

### Ex-ante burden-sharing, planning, and structural assessment
- Ex-ante public support agreements are difficult in practice; the Nordic-Baltic ex-ante burden sharing framework is cited as a non-binding example allocating public costs by share of the bank’s assets in each country and supervisory responsibility, with recognition of additional modifying factors.
- Regulatory focus has shifted toward avoiding public solvency support.
- Pre-crisis measures to reduce risk of public funding:
  - Recovery and resolution planning can identify structural imbalances in distribution of assets and liabilities in a financial group from national perspectives.
  - Information exchange alone is insufficient; information is incomplete and exposures can change quickly. However, knowledge about asset concentration, internal guarantees, and where losses can be absorbed with GLAC helps generate common understanding of group failure implications.
  - This knowledge can be used incrementally to require structural adjustments, regulate placement of assets and liabilities (as with GLAC), or impose prudential limits on specific exposures.
- Elements of an effective framework (as described):
  - Assessing structural financial imbalances between countries embedded in cross-border groups’ balance sheets (net A/L position by country; asset encumbrance by country; asset quality by country; intra-group liquidity by currency and country; intra-group capital; intra-group GLAC; whether intra-group GLAC is adequate; DIS/RF support needed; public costs needed; FX liquidity needed; foreign collateral needed; liability guarantees needed).
  - Determining if structural changes to cross-border groups are needed so each country is satisfied with trade-offs between hosting benefits and potential spillover risks.
  - Assessing potential for cross-border sovereign risk transfer or outright public costs arising from policy responses in the event of failure.
- Possible ex-ante measures to reduce unacceptable imbalances:
  - Reducing upfront exposure by limiting cross-border risk transfer (require banks to adjust structure; regulate placement of assets and liabilities; prudential limits on exposures such as foreign exchange risk, liquidity).
  - Increasing acceptability of imbalances through agreement (agreement on provision of emergency liquidity at group level or any residual solvency support).
- Three guiding policy questions for assessing bank configuration and resolution strategy:
  - To what extent can the burden of a bank’s failure be contained within the private sector?
  - What burden could remain with the public sector and how can this be minimized or eliminated?
  - What type of coordinated policies can be put in place to minimize ex-post ring-fencing or unilateral decisions?
- System-wide perspective: assessing beyond individual banks is important where multiple channels of contagion exist; example given of U.K. and U.S. joint support for SPE-type strategies predicated on high costs across both financial systems of failing to cooperate.

*Source: _060214 - 25.      The amount and location of GLAC determine ex ante how losses accrue in resolution,*

### 39.      Cross-border simulation exercises can provide a forum for discussing alternatives to

### _060214 - 39.      Cross-border simulation exercises can provide a forum for discussing alternatives to

### Simulation exercises and core cooperation questions (paragraph 39)
- Cross-border simulation exercises provide a forum to discuss alternatives to unilateral actions.
- Important elements for discussion:
  - (i) Would an ailing bank be allowed to keep up cross-border capital and funding flows?
  - (ii) Which sources of funding could be made available in crisis? Under what conditions?
  - (iii) Which claims would a country accept to be bailed in?
  - (iv) What size of imbalances would a country be comfortable accepting, and under what circumstances?

### Cooperation outcomes, incentives, and trade-offs (paragraph 40 and surrounding text)
- Possible cooperation outcomes range from full ex ante ring-fencing and maximum resolvability to arrangements that preserve shared critical functions.
- Full ex ante ring-fencing and maximum resolvability are equilibrium strategies when asset and liability positions are considered in isolation, but they ignore efficiency gains from shared functions.
- Trade-offs to consider:
  - Where locally funded operations undertake mainly basic banking functions, the ease of resolution may favor ring-fencing/subsidiarization.
  - Where loss of shared platforms or global operations has higher costs, prior commitments to burden-sharing could be explored (Box 5).
  - Example: country A may accept a liability overhang in local operations if country B provides counterbalancing resolution funding.
- Local capital or liquidity requirements have a multilateral dimension: they can impede free movement of funds, but ring-fencing in crisis is costly since consolidated group capital is not the relevant national buffer.

### Box 5 — Incentives to Cooperate and Subsidiarization (text from Box 5)
- Group structure matters: subsidiaries versus branches affect who bears responsibility and the ease of isolating losses.
- Subsidiarization ex ante may be superior to ring-fencing ex post when cross-border cooperation is unlikely.
  - Ring-fencing during a crisis carries operational, financial, and legal risks and can worsen the situation.
  - If cooperative strategies are not credible, hosts may require cross-border banks to hold local capital and liquidity ex ante.
- Stand-alone subsidiarization can reduce hosts’ incentives to ring-fence in crisis because hosts have greater understanding and control of local assets/liabilities.
- Isolating pools of capital and liquidity could have efficiency costs and impede cross-border intermediation; the extent of these costs vs. resolvability benefits is not established.
  - Costs may be low when operations are locally funded and assets are straightforward loans.
  - Costs rise if branch structures are highly integrated or foreign banks provide significant long-term funding to a host economy.
- For heavily interconnected markets with systemic players, formal cooperation may be preferable.
  - Example: In December 2012 the United States and the United Kingdom stated their preference for a cooperative SPE strategy for their systemic banks.
- Greater cooperation (effective dialogue, information sharing, cooperation agreements) would make authorities indifferent about specific bank structures, allowing firms to select structures that best fit business models without limiting crisis risk management.

### Applicability of Key Attributes beyond G-SIFIs (paragraphs 41–43; Box 6)
- Many cross-border banks are not globally systemically important; the Key Attributes were designed primarily with G-SIFIs and advanced systems in mind.
- Key considerations for non-G-SIFIs, regionally or domestically systemic firms:
  - Tailor reforms to financial system complexity; some elements of the Key Attributes may not be relevant in less sophisticated systems (paragraph 42).
  - RRPs and resolvability assessments (Key Attributes 11 and 10) should, at minimum, be required for firms designated as domestic SIFIs, using relevant sector-specific methodologies (Box 6).
  - Crisis management groups (Key Attribute 8) or comparable arrangements should be required for firms systemic in more than one jurisdiction or where multi-jurisdictional entity resolution is material.
  - Cooperation Agreements (Key Attribute 9) should be required where a crisis management group or comparable arrangement is needed.
- Loss-absorbing capacity (LAC) issues in primarily deposit-funded systems (paragraph 43):
  - Regional/domestic systemic banks should have adequate LAC to reduce risks to deposit insurance schemes and/or the public sector.
  - In jurisdictions with less developed capital markets, authorities may be reluctant to allow uninsured deposits to incur losses; substitution of extra capital may be appropriate, recognizing differing cost and loss-absorbing characteristics of liabilities.
  - Equity absorbs losses under a wider set of scenarios than bond debt; retail deposits are unsuitable for absorbing losses.

### The "Small Host" perspective and resolution priorities (paragraphs 44–45)
- Small host countries may face significant asymmetries in negotiations with G-SIFI home jurisdictions; they may lack participation in crisis management groups or supervisory colleges and may not have timely access to critical information (paragraph 44).
- Resolution strategies for small hosts may need to prioritize domestic financial stability (paragraph 45):
  - First priorities in a bank failure: contain contagion, maintain critical financial services, and minimize economic costs.
  - If ex ante cooperative agreements on resolution and burden sharing are not credible, structural requirements (e.g., subsidiarization, requiring local capital and liquidity, asset maintenance requirements for branches) may be appropriate.
  - This approach aligns with MPE (or hybrid) resolution strategies and should be coordinated with home jurisdictions to fit into comprehensive resolution plans.
  - Potential global costs (inefficiencies from local capital/funding pools, limits on international bank financing) may be low where operations are “plain vanilla” (local deposits funding local lending) and bring benefits in reducing prudential risks and increasing resolvability.
  - Cooperation with home jurisdiction remains important for GLAC decisions, critical services maintenance, and building trust for smooth resolution.

### Conclusions, remaining gaps, and policy priorities (paragraphs 46–49)
- Progress:
  - Recent international regulatory reforms have moved toward an effective cross-border resolution framework; widespread implementation of the Key Attributes will provide resolution authorities with tools to execute resolution strategies and place costs on shareholders and private creditors (paragraph 46).
- Persistent risks and gaps:
  - Orderly cross-border resolution is still far from assured; pitfalls and misaligned incentives from the global financial crisis could recur (paragraph 47).
  - Mixed progress in implementing national legal and institutional reforms consistent with the Key Attributes.
  - The Key Attributes only partially align divergent national interests and incentives that work against cooperative resolution.
- Identified remaining gaps and areas for international work (paragraph 47 and 48):
  - Defining specific proposals for GLAC (amount, eligible claims, location, and investor base).
  - Better aligning creditor hierarchies and depositor preference between countries.
  - Minimizing residual risks that public funds will be needed to preserve financial stability; credible bail-in options reduce but do not eliminate residual public-support risk.
  - Moving decisions and actions to the pre-crisis phase to align incentives and enable credible commitments. Possible pre-crisis arrangements could cover:
    - (i) Identifying structural imbalances on banks’ balance sheets in various countries;
    - (ii) Assessing costs and benefits of different bank structures and resolution strategies, and taking action to make these consistent with country authorities’ objectives;
    - (iii) Assessing and monitoring potential cross-border sovereign risk transfer or outright public costs arising from policy responses in the event of failure, and updating resolution strategies accordingly.
- Fund engagement:
  - The Fund will continue to promote adoption of the Key Attributes in engagements with members, participate in FSB work streams to address gaps, and keep the Executive Board apprised of developments (paragraph 49).

### Discussion points raised for Directors (end of document)
- Do Directors agree that progress has been made toward a framework for an effective cross-border resolution regime?
- Do Directors agree with the gaps identified in the paper and the key areas that need to be worked on to avoid repeats of misaligned incentives and unilateral actions from the global financial crisis?
- Do Directors agree that greater harmonization of creditor hierarchies, including depositor preference, is desirable/necessary?
- Do Directors agree that progress in agreeing on standards for GLAC would help remove important obstacles to effective cross-border resolution? What are the challenges to reaching agreement?
- Do Directors agree that moving decisions to the pre-crisis phase, including possible ex-ante changes to bank business models and structure (paragraphs 35–40), would help support resolvability and minimize non-cooperative outcomes?
- How relevant are recent advances for "small host" economies? Would tailoring the Key Attributes to the complexity of financial systems in different jurisdictions benefit broader Fund membership?

*Source: IMF chapter section numbered paragraphs 39–49 on cross-border resolution and subsidiarization (text supplied).*

### Annex I. Public Interventions in Cross-Border Banks

### Annex I. Public Interventions in Cross-Border Banks

### A. Citigroup
- Background
  - Leading U.S. global bank with assets of around $2 trillion.
  - Some 200 million customer accounts.
  - Businesses in more than 160 countries and jurisdictions.
  - By 2008, about 70 percent of the group’s revenue was derived from foreign assets.
  - Citigroup was the second largest bank holding company in the U.S., the largest consumer finance lender in the world, the third largest mortgage servicer, the fourth largest student lender, and the world’s largest credit card lender.
- Sources of distress
  - In 2008, Citigroup suffered losses of about $28 billion and its stock price dropped precipitously.
  - Heavy exposure to troubled subprime mortgages and related credit instruments led to a severe deterioration in the group’s condition.
- Stabilization strategy and interventions
  - On October 28, 2008, the U.S. Treasury invested $25 billion in Citigroup by purchasing senior preferred stock with warrants through the Troubled Assets Relief Program (TARP).
  - The support failed to prevent a significant outflow of deposits.
  - On November 23, 2008, Treasury, the Board of Governors of the Federal Reserve System, and the Federal Deposit Insurance Corporation announced a broader support package for Citigroup, including:
    - A guarantee covering potential losses on more than $300 billion of assets in exchange for $7 billion of preferred stock.
    - An additional $20 billion investment of TARP funds in exchange for preferred stock.
  - By December 2010, Treasury had sold its investment in Citigroup at a profit of $12 billion.
- Restructuring and outcomes
  - At the peak of the crisis in 2009, Citi was reorganized into two main segments—Citicorp (global consumer banking and an institutional clients group) and Citi Holdings (non-core businesses: brokerage and asset management, global consumer finance, and Citi's special asset portfolios).
  - Risk management practices were overhauled.
  - The rescue of Citibank did not have significant destabilizing effects internationally.

### B. Dexia
- Background
  - One of the largest European financial conglomerates before the crisis, with total assets of about €650 billion.
  - Created in 1996 through the merger of Credit Communal de Belgique and Credit Local de France.
  - Balance sheet growth of about 150 percent during 2000–08.
  - Business model relied on synergies between deposit-taking activities (mainly in Belgium and Luxembourg) and municipal funding (through Dexia Credit Local—DCL, which accounted for two-thirds of group assets).
- Sources of distress
  - Substantial investments in structured securities (including through the U.S. monoline insurance subsidiary of DCL).
  - Strong reliance on short-term wholesale funding to fuel pre-crisis expansion into largely long-term assets.
  - In the third quarter of 2008, the group suffered over €2 billion in losses and lost access to wholesale funding markets.
  - Short-term liquidity gap estimated at €260 billion in October 2008.
  - Substantial liquidity assistance was provided by various central banks in countries where the group operated.
- Initial stabilization strategy (late 2008)
  - States of Belgium, France, and Luxembourg jointly recapitalized the group for a total amount of €3 billion, with an equivalent contribution by private shareholders.
  - The three governments provided funding guarantees (of up to €150 billion).
  - Belgium and France guaranteed an asset portfolio of $12.5 billion in DCL’s U.S. subsidiary FSA (sold in 2011).
  - Burden sharing key for the three governments’ capital and guarantees was based on the proportion of the national ownership into the non-floating capital of the group before the capital injection.
  - The group regained market access and the stock price rose; the group engaged in a major program of divestments of international operations approved by the European Commission.
- Interim actions and legal measures
  - Belgian and French authorities discussed potential split of the bank into a ‘bad bank’ and a ‘good bank’ following the 2008 rescue; various proposals were rejected by either Belgian or French authorities.
  - Belgium passed in 2010 a law allowing the emergency expropriation of shareholders of financial institutions when necessary to preserve financial stability.
- Challenges and change in strategy (2011 intervention and after)
  - The Euro Area sovereign crisis caused further losses and loss of market access; renewed market turmoil in 2011 prompted renewed discussions on possible solutions.
  - Orderly dismantling with gradual decoupling of national operations and a runoff of the legacy portfolio emerged as an acceptable strategy.
  - The second intervention in October 2011 followed a complex long-term restructuring plan including:
    - The orderly runoff of the group’s assets supported by sovereign backing and national central banks funding of the holding company (Dexia SA) and DCL.
    - The gradual sale of international operations.
    - The acquisition of the Belgian bank by the Belgian state for €4 billion, rebranded in 2012 as Belfius.
  - Financial ties (secured and unsecured lending) between the Belgian bank and the rest of the group are being gradually extinguished, with the three governments continuing to guarantee obligations of Dexia SA and DCL to enable their long-term financing.
  - The new burden sharing key for each government’s contribution to the guarantee (of up to €90 billion) was 51.4 percent for Belgium (down from 60.5 percent).

*Annex I. Public Interventions in Cross-Border Banks*

### 45.6 percent for France (up from 36.5 percent), reflecting the nationalization of the Belgian

### _060214 - 45.6 percent for France (up from 36.5 percent), reflecting the nationalization of the Belgian

### Dexia — nationalization, capital injections, and guarantees
- New public capital injections in 2012: €5.5 billion in Dexia SA (Belgium €2.9 billion; France €2.5 billion).
- Luxembourg subsidiary (DBL) sold in October 2012; no Luxembourg capital injection required; Luxembourg continued to contribute to the guarantee (3 percent).
- European Commission decision (December 2012) led to creation of Société de Financement Local (Sfil) through purchase of capital and assets of Dexia Municipal Agency.
- New ceiling for the States’ guarantee set at €85 billion.
- Supervisory arrangements: Belgium was the consolidated supervisor; core supervisory college included France and Luxembourg and was backed by an MoU. Parliamentary report noted weak supervisory reach of the then Belgian supervisor (CBFA) on certain foreign non-regulated entities (FSA). Prudential supervision transferred to the National Bank of Belgium (NBB) in early 2011.

### Fortis — expansion, exposures, and dismantling
- Background: By end-2007, total assets approximately €870 billion.
- Initial public stabilization (September 28, 2008): total capital injection of €11.2 billion (Belgium €4.7 billion for 49 percent of Fortis Bank Belgium; Netherlands €4 billion for 49 percent of Fortis Bank Nederland; Luxembourg €2.5 billion loan convertible to 49 percent of Fortis Banque Luxembourg).
- Netherlands fully nationalized Dutch operations at a cost of €16.8 billion (October 3, 2008).
- Belgium acquired full ownership of Fortis Bank Belgium for €4.7 billion and sold 75 percent to BNP Paribas immediately.
- Structured credit portfolio valued at €11.8 billion transferred to a ‘bad bank’ with ownership: Ageas 45 percent, Belgian government 43 percent, BNP Paribas 12 percent.
- Dutch government took over short-term funding (€34 billion) and long-term subordinated debt (€16 billion) of the Dutch subsidiary from the holding company.
- Netherlands provided additional capital relief: write-down of long-term debt €6.5 billion and new capital €5 billion in following year.
- Total cost for Netherlands of nationalization of Dutch parts of Fortis currently valued at €32 billion.
- Belgium later sold the ‘bad bank’ for €6.7 billion to Credit Suisse and Lone Star, and sold remaining 25 percent stake in BNP Paribas Fortis for €3.25 billion.

### ING — Alt-A exposures, government support, and restructuring
- Background: ING total assets some €1.3 trillion as of end-2007; ING Direct client deposits almost €200 billion by end-2007.
- U.S. Alt-A mortgage-backed securities exposure: €27.5 billion at end-2007.
- Initial stabilization: Dutch authorities provided €10 billion in Core Tier-1 Securities (mix of equity and subordinated debt) in 2008.
- Illiquid Asset Back-Up Facility transferred 80 percent of economic ownership of ING’s Alt-A mortgage-backed securities portfolio to government in exchange for a fee; legal ownership remained with ING.
- Unwinding of the Illiquid Asset Back-Up Facility completed in early-2014, resulting in a cash profit for the government of €1.4 billion.
- Repayment of the €10 billion in capital support completed for more than 90 percent; remainder expected in 2015.
- European Commission required restructuring: full separation of banking and insurance activities, sale of a domestic banking unit, divestment of U.S. subsidiary ING Direct; temporary bans on acquisitions and price leadership.

### Icelandic banks — rapid expansion, runs, and emergency measures
- Combined balance sheet of Kaupthing, Glitnir, and Landsbanki: €16 billion in 2003.
- By onset of crisis, banks’ assets were 10 times Icelandic GDP.
- Immediate funding needs (Kaupthing, Glitnir, Landsbanki): €5 billion versus international reserves at Central Bank of Iceland of €2.6 billion.
- International deposits at risk: €15 billion.
- Emergency legislation (Act 125/2008) enabled supervisory takeovers, gave priority to depositors over bondholders, and declared banks would not be supported by government.
- Banks separated into “new banks” (domestic operations) and “old banks” (runoff international operations). Landsbanki closed; New Landsbanki created; “Icesave” outflows large.
- U.K. authorities seized Landsbanki London entities and issued freezing orders; Denmark extended government guarantees to Kaupthing’s Danish subsidiary (FIH) but Kaupthing could not meet obligations toward FIH (about €2 billion).
- October 2008: Iceland imposed capital controls to halt outflows.

### KBC — CDO losses, government support, and restructuring
- Reported losses in 3Q 2008: €0.9 billion (mainly CDO revaluation).
- Stabilization: capital injection of €7 billion in 2008 from Belgian federal government and Flemish government (completed in 2009); asset relief guarantees on a CDO portfolio of €20 billion provided in 2009.
- Repayments of federal capital injections: €0.5 billion in 2011 and €3 billion in 2012.
- Reimbursements to Flemish government: €1.166 billion in 2013 and €333 million in 2014.
- Private capital raised: new equity €1.25 billion in December 2012; contingent capital instruments €0.75 billion in January 2013; Additional Tier 1 instruments €1.4 billion in March 2014.
- Restructuring plan submitted September 30, 2009: retain integrated banking and insurance model; divest noncore markets in Central and Eastern Europe and noncore activities in Belgium; CDO portfolio in runoff.
- Asset disposals 2010–2013 included sale of KBL, Polish subsidiaries (Kredyt Bank, Zagiel, and Warta), Russian (Absolut Bank) and Slovenian (NLB) operations.

### Hypo Alpe Adria (HAA) — regional expansion, guarantees, and state support
- Liabilities under Carinthia guarantee reached about €20 billion; total assets €38 billion in 2007.
- BayernLB acquired majority stake in 2007; BayernLB total assets €416 billion in 2007.
- December 2008: HAA received €700 million capital from BayernLB; Austria provided €900 million capital support and guarantees for €1.35 billion of bond issues. BayernLB received €10 billion public capital injection and guarantee of up to €4.8 billion on ABS portfolio.
- By end-2011, non-performing loans had risen to 27 percent of total loans.
- Additional capital support: Austria provided €450 million in capital in 2010 and €200 million additional guarantee in January 2011.
- End-2012: additional capital support €500 million and guarantees on Tier-2 instruments €1 billion.
- December 2009: BayernLB transferred all its shares in HAA to the Austrian government without compensation; took a loss on Tier-2 capital and loans to HAA, effectively recapitalizing the bank by almost €1 billion; committed to keep remaining loans until 2013–15, callable and repayable by Austrian state in case of significant ownership changes.
- Restructuring plan: significant reduction of leasing activities, divestment of Austrian and Italian subsidiaries, focus on CEE bank network with aim to re-privatize; CEE subsidiaries to be sold by mid-2015; remainder to be transferred into a defeasance structure.

### Laiki Bank (Cyprus Popular Bank) — Greek exposures, nationalization, and resolution
- Total assets at end-2010: €43 billion (about 250 percent of Cypriot GDP).
- Losses due to Greek sovereign restructuring: €2.3 billion in 2011.
- Capital need to reach EBA Core Tier-1 Ratio of 9 percent by end-June 2012: €1.8 billion; capital provided in 2012 by issuance of an unfunded bond in exchange for equity, resulting in nationalization with government acquiring an 84 percent stake.
- Over 2012: lost a third of domestically raised deposits and €2 billion of deposits in its Greek branch.
- Independent due diligence in February 2013 identified additional significant capital shortfalls.
- March 2013: proposals for a one-off tax on deposits rejected on March 19, 2013; resolution under financial assistance program a week later:
  - Greek-related assets and liabilities transferred to Piraeus Bank in Greece.
  - Insured deposits, central bank funding, and almost all assets taken over by Bank of Cyprus via purchase and assumption.
  - Remaining liabilities (uninsured deposits, unsecured claims) stayed in legacy entity (insolvency estate) with a compensating 18 percent equity stake in Bank of Cyprus.
  - Bank of Cyprus recapitalized with participation of shareholders, bank debt holders, and partial conversion of uninsured deposits into equity.
- Deposits within U.K. branch of Laiki were transferred to U.K. subsidiary of Bank of Cyprus, avoiding loss on uninsured deposits.
- Resolution measures enabled by adoption of new resolution powers and depositor preference shortly before resolution.

### RBS — funding run, government recapitalization, and restructuring
- October 2008–December 2009: U.K. Treasury bought RBS shares in three rounds for a total of £45.5 billion.
- Asset protection scheme covering £282 billion of risky assets (benefited until 2012).
- RBS posted a £34 billion loss at end-2008.
- Objective to reduce assets by £251 billion from end-2008 levels; activities separated into core and non-core, with non-core placed under a new division for divestment/runoff.
- As of February 2014: U.K. government retained a 79 percent stake through U.K. Financial Investments, with a plan to return RBS to private ownership.
- Supervisory findings (FSA): pre-crisis supervisory approach inadequate; overly reactive; insufficient data and peer analysis on capital returns; supervisor did not identify full scale of RBS vulnerability pre-Lehman.

### UBS — subprime losses, StabFund bad bank, and government support
- Prior to crisis total assets almost $2 trillion.
- By mid-2008, UBS write downs totaled €43 billion; 22 percent attributable to DRCM investments.
- Late 2007 capital raising: $11.5 billion from institutional investors; April 2008 additional $15 billion capital raising announced and completed over following weeks.
- Swiss recapitalization: mandatory convertible notes SwF 6 billion.
- Swiss National Bank created StabFund to take over up to $60 billion of troubled asset classes; UBS provided capital equal to 10 percent of asset value; SNB financed 90 percent via secured loan with maximum duration 12 years.
- U.S. support: SNB used currency swap line with U.S. Federal Reserve; UBS benefited from Commercial Paper Funding Facility (CPFF).
- UBS transferred $39 billion in assets to bad bank based on market value as at September 30, 2008, in three tranches between December 2008 and April 2009.
- UBS was the biggest borrower under CPFF with total cumulative amount of $75 billion.
- Wind-up completed late 2013; proceeds shared: $3.8 billion for SNB and $2.8 billion for UBS.
- UBS announced in 2012 to scale down investment banking, refocus on global wealth management and universal banking services in Switzerland, and continue deleveraging.

*International Monetary Fund — CROSS-BORDER RESOLUTION (excerpt)*

### Annex II. European Banking Union and the Bank Recovery

### Annex II. European Banking Union and the Bank Recovery and Resolution Directive

### Banking Union: scope, elements, and gaps
- Provides a consistent framework for the sound, efficient provision of international financial services in a highly integrated region, including a monetary union.
- Underpinned by an EU-wide legal framework, including the recently-approved Bank Recovery and Resolution Directive (BRRD).
- Key elements of reform:
  - single supervisory mechanism (SSM)
  - single resolution mechanism (SRM), including a single resolution fund
  - harmonized legal framework for deposit insurance
- The earlier EU Credit Institutions (Reorganisation and Winding Up) Directive 2001 (“Winding-Up Directive”) provided principles and procedures for cooperation among national authorities but lacked effective resolution authority and tools.
- Expected benefits: strengthened coordination, information sharing, and decision-making for cross-border banks.
- Implementation status and gap:
  - Implementation is at an early stage.
  - Lack of a common fiscal backstop in case additional funds are needed to facilitate resolution of systemic financial institutions.

### The European Stability Mechanism (ESM)
- Provides a formalized inter-governmental framework for helping to strengthen Euro area financial stability.
- Functions:
  - provides official loans to Euro area member states in financial difficulty
  - can, theoretically, be used to recapitalize systemic banks in resolution through direct equity injections jointly with the applying member state
- Conditions and limits on direct recapitalization:
  - Can only be considered after a bank’s balance sheet has been restructured through bail-in.
  - Requires that recapitalization from private sources is not feasible.
  - Requires the bank already meets a minimum capitalization requirement (common equity tier 1 equal to 4.5 percent of risk-weighted assets).
  - The total amount of ESM funds available for direct recapitalization is capped at €60 billion.

### Bank Recovery and Resolution Directive (BRRD): scope and main provisions
- Approved by the European Parliament in April 2014.
- Establishes requirements for national resolution frameworks for all EU member states and members of the European Free Trade Area.
- Objective: ensure institutions can be resolved speedily, with minimal risk to financial stability, and with losses borne by shareholders and creditors.

- Entities covered:
  - credit institutions established in EU member states
  - parent companies, (mixed), financial holding companies, investment firms, or other financial institutions established in the EU and subject to consolidated banking supervision

- Resolution powers:
  - sale of an institution’s business
  - transfer of assets and liabilities to a bridge institution
  - separation of assets for transfer to an asset management vehicle
  - bail-in, which can be used to recapitalize a failing institution, convert debt into equity, or to reduce principal amounts on claims or debt instruments with the exception of specific such as insured deposits and liabilities with a maturity of less than seven days
  - powers to replace senior management
  - powers to impose temporary stays on payment of financial claims

- Recovery and Resolution Plans:
  - Institutions are required to prepare plans for recovery from financial distress.
  - Authorities will take the lead in preparing plans setting out modalities for resolving failed banks in a way that preserves their most critical functions and avoids bail out by taxpayers.
  - Resolution authorities will have powers to require an organization to restructure in order to facilitate its resolvability.

- Creditor Hierarchy:
  - The BRRD introduces tiered depositor preference.
  - The waterfall of creditor claims established under the BRRD is as follows:
    (i) covered deposits, and the deposit insurance scheme by virtue of subrogation to the rights of covered depositors;
    (ii) eligible deposits of natural persons and micro and SMEs exceeding the deposit insurance coverage level;
    (iii) deposits in foreign branches of EU banks;
    (iv) claims of ordinary unsecured, non-preferred creditors; and (iv) shareholders.

- Recognition of third country resolution proceedings:
  - In the absence of agreement between an EU member state and a non-EU resolution authority, decisions regarding the recognition of the non-EU authority’s resolution proceedings may be taken either:
    (i) jointly by the European resolution college (if established); or, in its absence,
    (ii) by each resolution authority.
  - This provides more flexibility to EU authorities than under the Winding-Up Directive under which EU home authorities have exclusive competence in reorganization or winding-up measures relating to EU banks and their EU branches, with full recognition and legal effectiveness of such actions within the European Union.

- Funding for resolution:
  - Additional funding needs for resolution after bail-in will generally be provided through national resolution funds with contributions from banks.
  - For euro area member states, national resolution funds will be pooled into the proposed €55 billion Single Resolution Fund to be operational by 2016.
  - Contributions from banks are expected to be made over an eight-year period.

*Annex II. European Banking Union and the Bank Recovery and Resolution Directive*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_060214.pdf_