In March 2009, the Fund established a new Framework Administered Account to administer external financial resources for selected Fund activities (the “SFA Instrument”). The financing of activities under the terms of the SFA Instrument is implemented through the establishment and operation of a subaccount within the SFA. This paper requests Executive Board approval to establish the AFRITAC Central subaccount under the terms of the SFA Instrument.
In March 2009, the Fund established a new Framework Administered Account to administer external financial resources for selected Fund activities (the “SFA Instrument”). The financing of activities under the terms of the SFA Instrument is implemented through the establishment and operation of a subaccount within the SFA. This paper requests Executive Board approval to establish the AFRITAC Central subaccount under the terms of the SFA Instrument.
1. This paper responds to calls from the G-20 leaders who, at their London Summit in April 2009, agreed “to support continued efforts by the IMF, FSB, World Bank, and BCBS to develop an international framework for cross-border bank resolution arrangements.” At their summit in Pittsburgh in October 2009, the G-20 leaders called for the development of “resolution tools and frameworks for the effective resolution of financial groups to help mitigate the disruption of financial institution failures and reduce moral hazard in future.”3 The paper builds on the work of the Basel Committee’s Cross-Border Bank Resolution Group.4
The call for work on this issue arises from two related considerations.
2. First, the establishment of an effective framework for the resolution of financial institutions is essential to any strategy that seeks to both secure financial stability and limit moral hazard. The recent crisis demonstrates the extent to which the existing system may force national authorities to choose between two equally unattractive options: (a) a bail-out that does not fully allocate losses to shareholders and creditors; or (b) reliance on an insolvency regime that is ill-equipped to restructure financial institutions in a manner that both preserves value and safeguards financial stability. Accordingly, a key objective is to establish a resolution mechanism that will facilitate rapid and preemptive action by the authorities to preserve business continuity while restructuring an institution in a manner that allocates all losses to shareholders and creditors as promptly as possible, consistent with financial stability objectives.
3. Second, a resolution framework will be ineffective unless it is accompanied by a robust cross-border coordination mechanism. Although large, complex financial institutions operate globally, their resolution is subject to national legal frameworks. One solution to this problem would be the conclusion of a multilateral treaty that would obligate countries to defer to the resolution decisions of the jurisdiction where the financial institution or group has its main activities. There are examples in other areas of international relations where treaty frameworks have been put in place (e.g., regulation of shipping accidents) but the adoption of such an approach in the area of financial regulation would not appear to be feasible in the foreseeable future. Given their concerns over financial stability and the potential fiscal costs of bank failure, the authorities of many countries have been unwilling to surrender control over these issues. In these circumstances, the most realistic approach—at least in the medium term—is one that focuses on enhancing co-ordination among national authorities—something that has generally been lacking. Indeed, unless such coordination is achieved, it may be argued that financial stability concerns may require a “de-globalization” of financial institutions so that they fit within the existing local resolution frameworks.
4. Recognizing the benefits of globalized financial institutions and the difficulty of establishing an international treaty that would be signed by a broad range of countries, this paper discusses key elements of a pragmatic framework for enhanced coordination. Although the implementation of such an approach is likely to require modifications to the domestic laws of some countries so as to give national authorities the mandate to coordinate their resolution actions with other jurisdictions, national authorities would only be required to do so to the extent that, in their judgment, coordination is consistent with the interests of creditors and financial stability.
5. This paper uses the term “resolution” broadly and generically to refer to the full range of recovery and resolution activities that involve public intervention (whether privately or publicly funded) including, for example, mergers and acquisitions, equity recapitalization, debt for equity conversions, transfers of assets and liabilities, temporary administration, reorganization, and liquidation.
6. The issues addressed in this paper apply to the resolution of international financial groups. For some international financial groups, a banking business will be their main activity. However, many cross-border banks exist within financial groups whose activities extend far beyond simple deposit-taking and lending to cover a full range of non-bank financial activities. Moreover, some of the most systemically-risky international financial groups are, at their core, investment banks and broker-dealers that conduct little or no deposit-taking activity.5 While the substantive elements of resolution regimes for banks and non-bank financial institutions of course differ, there are a number of aspects of the mechanisms for coordinating resolution action in respect of banks and non-banks that will be similar. While not all of the entities within a group will be regulated, it is assumed that many of them will be, given their systemic importance.
7. While effective supervision—whether nationally or in a cross-border context—is an essential component of any effective crisis prevention framework, it is not the focus of this paper. No matter how effective supervision is, failures of financial institutions will continue to occur and, for this reason, it is not a substitute for credible resolution mechanisms. Nevertheless, it would be appropriate for a country to require the existence of a robust supervision framework in other countries to be a condition for the establishment of a cross-border resolution coordination framework with these countries. For this reason, effective supervisory coordination is a key element of the enhanced resolution coordination framework that is discussed in this paper. In that context, it should be noted that the existence of an effective resolution framework will likely enhance supervision and reduce the risk of ‘regulatory forbearance’ by giving national authorities credible resolution options.
8. The paper is divided into two parts. Part I examines the growth of cross-border financial services and the challenges involved in effectively supervising and resolving international financial groups. Part II identifies a possible way forward, setting forth out the essential features of an international framework for cross-border resolution.
Part I: The Status Quo—and its Costs
A. The Globalization of Financial Institutions
9. Financial globalization has led to the emergence of a large number of international financial groups. Cross-border banking has expanded rapidly over the last decade. Many large banks now rely upon a global network of branches and subsidiaries, with centralized funding that is distributed within the financial group under a global strategic plan. The activities of these groups have expanded beyond traditional deposit-taking and lending to include a range of non-bank financial activities, such as securities and insurance brokerage and fund and asset management. In addition to these ‘universal’ banks,6 the international space is now dominated by several large financial institutions that operate across borders, in multiple currencies and time zones, and act as systemically-important nodes within a globalized market for capital.
10. Several factors drive the globalization of financial services:
Financial liberalization. In recent years, many countries have eliminated barriers to the entry of foreign financial institutions.
Risk diversification. The opportunity for financial institutions to expand abroad allows them to diversify their risk, reduce reliance on their home markets and seek new business opportunities in overseas markets.
Servicing key corporate clients. As corporations have expanded abroad, large banks have followed them to support and profit from their expansion plans.
Brand value in emerging markets. An internationally-recognized “brand” with a local presence in foreign markets can rapidly gain market share abroad.
11. The legal form of a complex financial group may not always reflect the economic substance or operational functions of that group. Several different factors may influence its structure and organization that go beyond legal considerations.
Commercial factors/operational efficiency. Groups may choose to organize their operations according to business lines using matrix management structures that do not reflect the relationships between legal entities. Often a large group will organize itself with centralized functions for the entire group such as capital and liquidity management, risk management and IT and with subsidiaries that, whilst legally separate, may have no de facto independence.
Separability/location of assets. Activities carried out in a host jurisdiction may reflect decisions taken in a remote home state rather than locally. For the group, this may be an efficient allocation of resources. For example, domestic Swedish deposits supported the expansion of Swedish banks in the Baltic region, and similarly, much of Dexia’s lending to French regional governments was funded using Belgian deposits.
Regulatory factors. Formal requirements may be imposed by home or host national authorities for the establishment and development of cross-border financial activities.
Tax Treatment. The structure and organization of the group may be influenced by tax considerations.
12. In some circumstances, a financial group may effectively function as a single entity—in particular where a guarantee has been issued by the parent for the components of the group. As result of the interconnectedness of the financial group’s legal entities, weaknesses in one entity can adversely affect the entire group. In group structures where liquidity is centralized, any sudden and material downgrading of the central entity’s credit ratings or the opening of insolvency proceedings against it would lead to the immediate illiquidity of the other entities in the group.7 The triggering of cross default or cross guarantee arrangements for funding purposes as a result of rating downgrades or otherwise may also lead to financial distress in other parts of the group.
13. Moreover, the scale of activity or size of an international financial group may create systemic risks for either the home or the host jurisdiction when such groups enter into financial distress. Certain branches or subsidiaries may, in economic terms, be comparatively insignificant to a group yet be of critical importance to their host country’s financial system. In the case of a subsidiary in this position, its legal separateness may as a legal matter permit parent banks to simply ‘walk away’ should the subsidiary encounter difficulties, irrespective of the impact on the host country economy. However, ‘abandoning’ a subsidiary in such a manner would involve reputational risk and could be counterproductive for the stability of a financial group.
B. Localized Resolution Frameworks
14. While international financial groups operate globally, the frameworks for addressing their distress and failure are local and apply to distinct parts of the group rather than to the group as a whole. By allowing financial institutions under their supervision to establish presences in a range of jurisdictions, home authorities expose themselves to the reality that the legal frameworks for facilitating cross border finance in stable periods are typically more effective than the cross-border resolution arrangements that are available in times of distress.
15. While the existing fragmented approach is due to a number of factors, a fundamental reason is the fact that resolution frameworks are established by national law and, absent the cooperation of the national authorities of other jurisdictions, are only enforceable vis-a-vis those institutions—or branches of institutions—operating in their territory.8 In the absence of an international legal framework that empowers a supra-national entity to resolve global institutions, the resolution of such institutions are subject to different national frameworks and, accordingly, national authorities must proactively coordinate their actions to avoid the significant costs of an uncoordinated approach.
16. Moreover, the legal frameworks of many jurisdictions do not sufficiently facilitate coordination. National frameworks in some jurisdictions do not sufficiently empower their supervisors or the relevant resolution authorities to share information with their counterparts in other jurisdictions. In the context of an ailing bank, the ringfencing of assets by host jurisdictions may undermine an effective resolution. Home country official administrators may face difficulties in having certain recovery operations, such as “purchase and assumption” transactions, implemented in the host jurisdictions of bank branches.
17. Effective coordination is also hampered by the absence of a minimum level of harmonization. National legal and regulatory frameworks often differ in key areas. In the context of bank insolvency, there is no universally-agreed approach to such questions as the triggers for the commencement of insolvency proceedings or the powers available to the supervisors to deal with an insolvent bank.
18. Even where there is a minimum degree of harmonization, the multiplicity of regulatory actors may impede coordination. A financial group (whose activities might cover a range of separately regulated banking and non-banking activities) would potentially be subject to oversight from a number of different competent authorities, even at a purely domestic level. Not surprisingly, in the context of an international financial group, overlapping competencies and difficulties in discerning the scope of various national supervisors’ responsibilities are amplified.
19. Finally, and perhaps most importantly, when the regulatory authorities are faced with the distress or failure of a financial institution within their territory, they tend to give primary consideration to the potential impact on their own stakeholders: namely, creditors to branches or subsidiaries located within their jurisdiction, depositors and, in the final analysis, local taxpayers. In these circumstances, national priorities translate into a “territorial” approach that effectively precludes coordination: where in the event of failure of a domestic branch of a foreign bank, local assets are “ring-fenced” for the benefit of creditors to the branch. The practice of ring-fencing is geared to favoring the interests of depositors and creditors to a bank’s local presence to the detriment of stakeholders in other jurisdictions (see Box 2).9 In contrast, universality implies no ring fencing and instead would place all similarly ranked international creditors on an equal footing.
Territoriality and Universality
The approaches developed by countries for dealing with cross-border insolvencies (including of banks) fall by and large in one of the following two categories:
Universality—Under an ‘universal’ approach, the insolvency proceedings initiated against the debtor in its home country will purport to have ‘universal reach.’ This implies that the home country trustee will seek to gain control over all of the debtor’s assets and liabilities—including those located in other countries—to realize all assets and pay out the resulting proceeds to both domestic and foreign creditors according to their ranking. To be effective, ‘universality’ of the home country depends on different host countries recognizing this extra-territorial effect of the home country proceedings. Such recognition is, however, far from evident for the reasons set out below.
Territoriality—Many countries follow some form of ‘territorial’ approach, under which a host country will initiate separate insolvency proceedings against a foreign debtor, instead of participating in, or deferring to, the insolvency proceedings opened by the home country. Typically, ‘territorial’ jurisdictions will ‘ring-fence’ the assets and liabilities of foreign entities that are located in its territory in order to satisfy the claims of local creditors. To be effective, a ‘territorial’ approach requires a sufficient amount of assets (and liabilities) to be located within the country. In the case of the local branch of a foreign bank, the effectiveness of ‘ring-fencing’ is buttressed by supervisory rules requiring the branch to maintain sufficient local assets relative to their local liabilities.
These categories are not absolute, and several countries have insolvency regimes with mixed features. For instance, as regards cross-border banks, the USA is ‘universal’ for locally domiciled banks, but ‘territorial’ with respect to branches of foreign banks. In a similar vein, the EU’s so-called “Winding Up Directive” follows an EU-wide ‘universal approach’ for EU banks, but member states are free to maintain a ‘territorial’ approach to branches of extra-EU banks.1/
1/ The “Winding Up Directive” (Directive 2001/24/EC of the European Parliament and of the Council of 4 April 2001 on the Reorganization and Winding up of credit Institutions) provides a harmonized legal framework for the reorganization and winding up of EU banks under which the home state authorities are exclusively responsible for the opening of insolvency procedures against the head office and all EU branches of an EU bank. Specifically, home state authorities will steer, and home state law will govern (with some exceptions), the insolvency procedures for all EU wide assets and liabilities of the EU bank. The Directive does not, however, establish a common framework for the insolvency treatment of EU branches of extra-EU banks. While thus keeping national rules largely intact, the Directive merely requires the various host state authorities of branches located in EU member states to “endeavor to coordinate their actions.”
20. Although the national focus of resolution frameworks appears at odds with internationally coordinated supervisory frameworks, a closer examination reveals that even these supervisory frameworks are shaped by national concerns (see Box 3). Moreover, the implementation of such frameworks by some countries anticipates the ring fencing approach they rely on during the resolution phase. For example, although licensed branches of foreign banks in the United States (and in some other jurisdictions) do not, as legal extensions of a foreign entity, have separate capital of their own, they are nevertheless required to deposit cash or eligible securities at approved depository banks to satisfy a “capital equivalency requirement” established by applicable law.10
International Coordination in Banking Supervision
Effective supervision at the international level has been promoted through the development of international standards and best practices that national authorities voluntarily implement through the enactment of legislation or the conclusion of memoranda of understanding with supervisors in other jurisdictions. The goal of these initiatives is consolidated supervision that aims at empowering bank supervisors with the tools necessary to understand, monitor, and, when appropriate, minimize the risks associated with an organization’s consolidated or group-wide activities.
Internationally agreed principles on the supervision of cross border banking groups have been in place for several decades. The Basel Committee on Banking Supervision (BCBS) issued its first statement of principles or “Concordat” regarding the supervision of banks’ foreign establishments in 1975. These basic principles have been underpinned by further statements from the Committee addressing cross border supervision and home-host supervisory relationships. Since then, it has consistently called for international cooperation to ensure that no foreign bank operation evades proper supervision—including through the issuance of principles on cross border supervision and home-host supervisory relationships.
These efforts have facilitated cooperation but have not been entirely successful in facilitating effective supervision at the international level. Despite progress made worldwide in the adoption of international standards on capital, risk management, accounting rules, and other prudential matters, national authorities are not yet able to construct a complete map of the key risks affecting financial firms on a consolidated basis. Problems include:
Legal constraints and regulatory perimeter. Supervisors lack, in some cases, the legal authority to share information with foreign counterparts.
Divergences between supervisory approaches. While there may be agreement on the regulatory standards to be applied, they may be applied differently by different national supervisors.
Diverse reporting systems. Different supervisory models lead to different reporting systems that hinder timely data compilation.
21. This focus on national interest is also reflected in the mandates of many financial supervisors. With important exceptions (such as in the EU framework), these mandates typically emphasize the need to protect financial stability at the national—and not the international—level. Hence, when a group becomes distressed the national supervisory authorities are likely to focus on domestic interests.
C. The Costs of the Existing Approach
22. The costs of the application of local resolution frameworks to global institutions may be distilled as follows:
First, the absence of an effective cross-border resolution framework undermines financial stability in a number of different respects.
Uncoordinated actions by national authorities may hasten the failure of a financial institution in a manner that destroys value. This could occur, for example, if, during a period of stress, the host jurisdiction required a transfer of assets to cover the liabilities of the branch, and destabilized the bank in the home jurisdiction.11
Moreover, recourse to uncoordinated local liquidation proceedings may prevent a recovery effort that seeks to preserve the continuity of critical functions, thereby giving rise to contagion. For example, the efforts of the national authorities to preserve continuity through a purchase and assumption transaction may be stymied if the national authority that has jurisdiction over the branch is unwilling to allow for the necessary transfer of assets and liabilities and focuses exclusively on a liquidation designed to satisfy stakeholders.
Finally, in circumstances where a financial institution or group operates in numerous jurisdictions, the uncertainty as to how the various national authorities will coordinate their actions makes it very difficult for effective action to be taken quickly—which is essential to any strategy that seeks to both preserve value and limit contagion.
23. Second, the existing framework exacerbates moral hazard. Given financial stability problems that arise from uncoordinated national approaches, as described above, it is not surprising that a more tempting approach is to provide public bail-outs without any effort to ensure that action is taken to ensure that shareholders and unsecured creditors assume the necessary losses before public funds are committed. Moreover, even if national resolution frameworks are relied upon, an uncoordinated approach may not maximize the value of the institution or the group, and, therefore, may increase the amount of financing that will have to be provided by a state. For example, a financial group operating in numerous jurisdictions may lose a significant portion of its franchise value—and therefore its attractiveness to potential private investors—if it is broken up along national rather than business lines. This is also true where the liquidation of a cross-border institution is implemented in a purely piecemeal manner.
24. Indeed, recent experience demonstrates that the more interconnected and integrated international financial institutions and groups become, the more disruptive and value-destroying uncoordinated local resolution actions are likely to be. The cases of Fortis and Lehman (Box 4) demonstrate how the existing approach may fail to realize coordination benefits both in the context of a restructuring or a liquidation of an integrated cross-border institution.
Fortis and Lehman
Recovery of Fortis
After the Benelux financial conglomerate Fortis Group fell into crisis in late 2008 the group was resolved along national lines in a protracted process that failed to preserve franchise value.
In the Netherlands, the Dutch state bought Fortis’ Dutch bank, its insurance arm as well as parts of ABN Amro that Fortis had recently acquired. In Belgium, the Belgian government bought Fortis’ Belgian bank (the largest component of the overall Fortis Group) and agreed to sell a 75% stake in it to BNP Paribas. BNP also bought Fortis’ Belgian insurance operations and acquired a majority stake in Fortis’ Luxembourg subsidiary.
Completion of the resolution of Fortis was delayed for nearly six months between December 2008 and May 2009 after Belgian shareholders in Fortis succeeded in challenging the deal to sell most of the Belgian bank to BNP Paribas. The Belgian Court of Appeal found that shareholders were entitled to vote on the transaction in order for it to be valid under Belgian law. Shareholders subsequently voted against the transaction and subsequently approved it after agreeing to modifications.
The example of Fortis brings into sharp relief the problem of balancing private shareholder rights with the public interest in systemic stability through swift and decisive bank resolution.
The case also illustrates the tendency for national interests to come to the fore in a crisis and the difficulty in such circumstances of achieving a cross-border consensus, even between jurisdictions whose financial regulators have a long tradition of co-operation and whose legal frameworks are considerably harmonized.
Liquidation of Lehman
Lehman Brothers provides an example of the potential for competing proceedings in cross border liquidation of a financial group.
At the time Lehman Brothers filed for bankruptcy protection in the United States in September 2008, the firm had operations around the globe involving dozens of different group entities (both branches and subsidiaries).
With main proceedings in the United States and the United Kingdom, insolvency officials in numerous other jurisdictions are also engaged in winding down the various international components of Lehman Brothers with little or no coordination. Complex intra-group arrangements have also impeded the return of client property, for example with large amounts of client money that was segregated by Lehman’s UK broker-dealer having been deposited at a German affiliate which itself entered an insolvency proceeding and moratorium.
Part II: Possible Elements of Enhanced Coordination Framework
25. While the inadequacies of the existing framework are manifest, several options for improving the framework for cross-border resolution are available, each with its own advantages and disadvantages. Moreover, regardless of which steps are taken in relation to resolution, measures that address prevention and preparedness, including simplifying financial group structures where necessary to facilitate resolution, will also be of critical importance in the future (see Box 5).
Initiatives Directed Towards Crisis Prevention
Since the start of the financial crisis, policy makers have developed several proposals to strengthen cross-border supervision and to reduce the likelihood of a large cross-border financial group falling into difficulty. Some of these proposals have already been implemented while others are under discussion.
Colleges of banking supervisors have been expanded now to almost forty financial groups. Although colleges are not a new initiative, it is intended that home and host supervisors will have enhanced direct and frequent liaison between each other and with the banks on key issues such as risk management, capital and liquidity, which in turn will enhance mutual trust among national authorities. To make these colleges more effective, however, amendments to national legal frameworks will, in some cases, be necessary—in particular to authorize the sharing of critical information between supervisors when the financial conditions of banks are deteriorating.
The Financial Stability Board (FSB) will identify jurisdictions that fail to implement internationally agreed standards concerning international cooperation and information exchange. The FSB will engage with such jurisdictions in order to bring them toward full compliance and, in some cases, may impose counter-measures.
Proposals are under consideration to: (a) discourage banks from engaging in activities that give rise to systemic risk through a systemic risk charge on “systemically important” institutions; (b) make large complex financial institutions more resilient to shocks by increasing capital levels and buffers; and (c) reduce the complexity of large financial groups (i.e., “de-risking” of cross border firms and “subsidiarization”).
Crisis management groups (CMG) have been established under the auspices of the Cross Border Crisis Management Working Group of the FSB for the major international financial firms. Formed of supervisors, central banks and resolution authorities from the key home and host jurisdictions of the major international firms, they are tasked with developing recovery and resolution plans (RRPs) for these firms. RRPs can be useful tools for ensuring the preparedness of firms and authorities if they are used to identify measures which the firm and/or authorities can undertake prior to a shock to facilitate more effective and coordinated recovery or resolution. This might include measures the firm should undertake to strengthen their capital position or liquidity buffer or to improve their ability to provide the detailed information needed quickly in a resolution. RRPs may also help identify measures the authorities should undertake to strengthen their resolution powers or incentivize structural changes in the firm.
A. The Cross-Border Resolution Group
26. Of the several international initiatives on cross-border resolution, the most important contribution to date has been that of the Cross Border Bank Resolution Group (CBRG) of the Basel Committee on Banking Supervision (BCBS).12 The group published its final Report and Recommendations on cross-border bank resolution in March 2010. Other important regional initiatives are also underway on cross-border resolution, including the European Commission’s consultations directed towards improving the EU framework for cross-border bank crisis management.
27. In its Report, the CBRG observed that a number of alternative approaches to cross-border resolution are available:
Full ‘universality’ via a binding legal instrument, such as an international treaty. To be fully effective, the CBRG recognized that such a treaty would need to include substantive obligations related to key issues such as selection of lead authority and burden sharing.
De-globalization of financial institutions. At the other extreme from a pure ‘universal’ solution would be a uniformly ‘territorial’ approach in which institutions would be separately structured for capital, liquidity, assets and operations within each jurisdiction. By promoting the separate functionality of financial organizations through stand-alone subsidiaries, such an approach could contribute to the resilience of host country operations.
A ‘middle ground’ approach. The CBRG recognized that enhanced coordination among resolution authorities might provide a solution that steers a path between territoriality and universality. The CBRG recommended that national authorities develop procedures to facilitate the mutual recognition of crisis management and resolution proceedings and/or measures.
28. When evaluating the various alternatives identified by the CBRG, it is important to bear in mind a number of considerations. First, the ongoing debate on the merits of universality versus territoriality is somewhat theoretical and, as recognized by the CBRG, is not entirely relevant to the existing problem. It is theoretical because, at least in the short-term, it is very unlikely that all key jurisdictions will agree to sacrifice the degree of national sovereignty necessary to implement full universality. It is not entirely relevant because the debate applies exclusively to single entities (i.e., a parent bank and its branches) and is not applicable to the resolution of inter-connected but separate legal entities within a group. Second, the de-globalization of financial groups and institutions is problematic on a number of different levels. It would both reduce efficiencies and could undermine access to credit to emerging market economies. While some reduction in the scope of the international activities of large international banks may contribute to financial stability, the presence of large international banks in emerging markets has, in some cases, strengthened the resilience of these markets. As has been demonstrated recently in Central and Eastern Europe, the financial support provided by parent banks to subsidiaries operating in member countries experiencing a financial crisis played an important role in crisis resolution.13
29. In light of the above, this paper proposes possible elements of a framework that would underpin the ‘middle ground’ approach, one that would facilitate coordination across borders without requiring a surrender of national sovereignty. This framework draws on many of the achievements of the United Nations Commission on International Trade Law (UNCITRAL) in the field of cross-border corporate insolvency (see Box 6).14 While recognizing that the specific features of corporate insolvency are not applicable to the financial services industry (in particular, a financial institution’s resolution will generally be led by specific resolution authorities rather than courts), the staff is of the view that two elements of the approach developed by UNCITRAL are of potential relevance. First, while a court is required under the UNCITRAL framework to “recognize” the existence of insolvency proceedings in other jurisdictions, it retains broad discretion as to the degree to which it will actually defer to the decisions and requests made by the courts and insolvency officials in such jurisdictions. Second, UNCITRAL addresses a number of the specific procedural issues that can hamper coordination as a matter of practice.
30. It is recognized that, in the context of financial institutions, the host authorities will only feel that they can cooperate with the home authorities if they have confidence that the home authorities are willing and able to take effective action. Indeed, the universalist framework that exists in the European Union is a product of a very high level of integration amongst the countries of the Union. (see Box 2). Although it is recognized that this level of integration would be difficult to replicate outside the EU, the proposed approach recognizes the need to have some minimum level of commonality of resolution and supervision systems in order for cross-border co-operation to be effective. Accordingly, and as a supplement to the two elements derived from the UNCITRAL framework described above, the approach proposed in the paper would identify certain “core coordination standards” that countries would need to have in place in order to be eligible to participate in the enhanced coordination framework.
31. Taking into account the above analysis, the proposed approach envisages the establishment of an enhanced coordination framework that would be put in place through a nonbinding multilateral understanding reached among those countries that are in position to adhere to its various elements.15 These elements would include the following:
UNCITRAL and Cross-border Corporate Insolvency
The United Nations Commission on International Trade Law (UNCITRAL) has been a driving force for progress in the development of an international framework for the coordination of cross-border corporate insolvency proceedings. An important achievement of UNCITRAL in this area is its Model Law on Cross-border Insolvency (the “Model Law”), adopted in 1997. The model law is not a treaty but, as its name suggests, a model that countries may voluntarily incorporate into their domestic legal frameworks.
The Model Law applies to the insolvency of a single firm with a presence in foreign jurisdictions. It does not apply to types of entities for which special insolvency regimes may exist in national law—in particular, banks and insurance companies. Moreover, it does not apply to corporate groups comprised of legally distinct subsidiaries or affiliates. The insolvency of corporate groups is currently the subject of a separate UNCITRAL project—the preparation of a legislative guide on the treatment of enterprise groups in insolvency.
The Model Law sets out a framework for managing the insolvency of a cross-border financial firm in a fair and orderly manner. A central feature of the model law is the principle under which the courts of one jurisdiction will “recognize” proceedings in another jurisdiction. Importantly—and of particular relevance to the issues discussed in this paper—recognition generally permits—but does not require—the court to grant relief to a foreign insolvency representative if it determines that the interests of the debtor and creditors would be protected. While some aspects of the framework contemplated in the Model Law may not be entirely appropriate for the insolvency of a cross-border financial group, other features are of great relevance, including the following.
Center of Main Interest. The Model Law distinguishes between the “main” and “nonmain” insolvency proceedings respecting an enterprise. In identifying the “main” proceeding, the Model Law looks to the jurisdiction in which the debtor has its “center of main interests” (COMI).
Cooperation. The Model Law provides legal authority for insolvency representatives in different jurisdictions to collaborate with each other (via direct communication and information sharing) and to coordinate concurrent insolvency proceedings.
Discretionary Relief. With one exception (i.e., an automatic stay on execution in connection with a foreign “main” proceeding), the granting of relief to a foreign representative is at the discretion of the court. Moreover, it is subject to conditions. In particular, the court must ensure that the debtor and its creditors are adequately protected. The Model Law forbids discrimination against foreign creditors.
Protocols. The framework for cooperation set out in the Model Law has been very effectively supplemented through the negotiation of protocols on cooperation between insolvency officials in individual cases. Protocols are formal agreements typically negotiated through professionals representing major interests involved in an insolvency. They are normally approved by relevant courts. Since the adoption of the Model Law in 1997, a huge body of protocols have been negotiated.
First, the modification of domestic laws that would require national authorities to coordinate with foreign jurisdictions—but only to the extent that, in the judgment of the national authority in question, such coordination would be consistent with the interests of creditors and domestic financial stability.16
Second, the identification of “core coordination standards” that would be used to identify those countries with whom a more coordinated cross-border resolution would be expected to take place.17
Third, recognizing that public funding in the resolution process may, on occasion, be needed, if only on a temporary basis, the establishment of principles that would set forth the criteria and parameters that would guide the burden sharing process among the members of the enhanced coordination framework.
Finally, the specification of coordination procedures to be relied upon by those countries that adhere to the enhanced coordination framework.
Each of these elements is discussed in greater detail below.
B. Facilitating Coordination
32. The authorities of a country should be required to coordinate with resolution authorities in other jurisdictions, but only to the extent that the authorities determine that such co-ordination is consistent with their own national interests. More specifically, members would ensure that their domestic legislation requires national authorities to coordinate their resolution efforts with their counterparts in other jurisdictions to the maximum extent consistent with the interests of creditors and domestic financial stability. In determining whether a coordinated approach is consistent with the interests of creditors, the national authorities of a host jurisdiction would assess whether, under a coordinated approach, creditors to branches or subsidiaries located on their territory are likely to receive at least what they would receive had the branch or entity been liquidated on a territorial basis by the host jurisdiction. Of course, a coordinated approach that is consistent with the interests of creditors may still involve the imposition of losses upon creditors. National authorities would continue to retain the discretion to act independently if, in their judgment, such action is more consistent with the interests of creditors and financial stability.18
33. At present, there are cases where a country’s framework does not sufficiently facilitate coordination.19 For example, in some jurisdictions, existing laws may effectively prevent competent authorities from sharing information with foreign competent authorities.20 Moreover, local law may encourage the ring-fencing of assets of the branch of a foreign bank for the benefit of the creditors of the branch. In addition, attempts by the national authorities of the home jurisdiction to continue critical operations of a bank through a purchase and assumption (P&A) transaction may be frustrated by the regulatory actions of the host authorities with respect to branches falling under their control.
C. Coordination Standards
34. Even if domestic legal frameworks are modified to establish a coordination mandate, subject to the interests of creditors and financial stability, experience demonstrates that the national authorities will only be willing to coordinate their activities if they have adequate confidence in their counterparts. To that end, the objective would be to identify certain standards that countries would be expected to adhere to as a condition for cooperation. As a matter of practice, it would be presumed that all countries that meet these standards would be expected to coordinate their activities with each other in the context of a resolution, it being recognized that this presumption could always be rebutted by a national authority who had reached the judgment that, in a particular case, independent action is necessary to protect financial stability or the interests of creditors.21 The following standards would appear to be the most relevant.
1. Minimum level of harmonization of national resolution rules
35. Host country authorities will only be willing to cooperate with home country authorities if their national frameworks have a reasonable level of high quality convergence. In particular, the legal framework of the authorities involved in group-wide resolution will need to share certain key features:
Non-discrimination against foreign creditors. With respect to jurisdictions where branches of foreign banks are located, the authorities of the host countries will need to be satisfied that other countries’ resolution procedures will not discriminate against the creditors of the local branch including depositors and, by extension, deposit guarantee schemes (DGS), and governments. Domestic depositor preference in the home country, based upon the nationality or location of the depositor would be inconsistent with this principle.
Effective intervention tools. Many countries are recognizing the need for special bank resolution regimes and official administration procedures that allow competent authorities to intervene rapidly and in a manner that both preserves the critical functions of the institution and avoids contagion. Strengthening countries’ domestic legal frameworks for resolution would, in itself, represent an important step forward, and work on developing best practices in this critical area is being pursued in a number of different fora, including in the FSB. Among those intervention powers that are currently considered to be the most critical are the following:
∘Early intervention authority, i.e., the existence of common “triggers” that allow the authorities to take action well before balance sheet insolvency.
∘Powers that would enable the authorities to unilaterally restructure the various claims of an institution, e.g., debt-for-equity conversions or the reduction of the value of unsecured creditors.
∘The authority to conclude mergers and acquisitions without shareholder consent.
∘The unilateral power to transfer assets and liabilities to other institutions, including a bridge bank that would be established for this purpose, without the need to obtain the consent of third parties.
∘The authority to provide bridge financing to facilitate the transactions described above.
∘The ability to assume public ownership of the institution on a temporary basis, once the shareholders and unsecured creditors have absorbed the necessary losses.22
∘As a means of both limiting contagion and preserving critical operations, the temporary suspension of termination provisions contained in some financial contracts.
Appropriate creditor safeguards. The intervention powers described above are exercisable in pursuance of a public interest in financial stability. However, these powers potentially interfere with private contractual and property rights. Accordingly, rules on creditor safeguards and the judicial review of supervisory and recovery/resolution action to ensure the equitable treatment of creditors are essential features of resolution regimes (see Box 7). Where a bank is resolved under a special resolution framework, compensation ought to be available to creditors to ensure that they are left no worse off in the resolution than if the firm had been allowed to fail and lapsed into liquidation. Similarly, where resolution powers permit transfers of property, resolution regimes need to provide sufficient safeguards to stakeholders by protecting customer property rights, security interests and financial collateral arrangements in financial contracts (including netting rights).
At least for banks, sufficiently robust and harmonized rules on priority that recognize the interests of host country insured depositors and deposit guarantee schemes (DGS). If these rules in the home country do not ensure equal priority for the host country insured depositors and DGS, the latter’s resolution authorities will have a strong incentive to choose a domestic solution. Arguably, this may require a broader harmonization of DGS features, including the categories of insured depositors and the amounts of the protection.
2. Robust Supervision
36. For any host country authority to accept the leadership of home country authorities and to collaborate with other host authorities, the former will also need to be satisfied that the level of prudential supervision in the latter is of sufficient quality and that the relevant supervisors engage in consolidated supervision (e.g., including insurance firms, securities firms). It is true that for some types of financial institutions, there already exists a set of broadly accepted international standards (e.g., the Basel Core Principles on Effective Bank Supervision).23 Similarly, at least for banks, the Basel Concordat already includes the principle that host countries should not grant market access to foreign banks if the latter are not well supervised in their home jurisdiction (and vice versa). Nevertheless, in light of the crisis, it is felt by many supervisory authorities that these standards have not brought about a sufficient increase in the quality of prudential supervision and the willingness of supervisors to intervene in all relevant countries. To coordinate with foreign resolution authorities, they might thus require higher quality supervision and greater convergence on these points. The establishment of colleges of banking supervisors and the steps being taken by the FSB to promote global adherence to international cooperation and information sharing standards are measures in the right direction.
Resolution powers overrule ordinary private property and contractual rights in the interests of wider public interests such as financial stability. Accordingly, countries which adopt such resolution powers need to have strong safeguards mechanisms which ensure that powers are exercised appropriately. The effectiveness of international resolution action depends on all the involved countries having minimum safeguard protections that would be available to all creditors of the affected entity irrespective of nationality. These safeguards would, inter alia, ensure that:
-secured property is not transferred out of a failing bank without the benefit of security moving with it;
-other netting and financial collateral arrangements are respected (subject potentially to the temporary suspension of close out netting rights in respect of financial contracts transferred to a solvent third party);
-no creditor (domestic or foreign) of a resolved bank is left any worse off as a consequence of the resolution action than they would have been had the bank not been resolved but instead had failed and been liquidated; and
-foreign creditors are not discriminated against either based upon their nationality or location.
Safeguards are particularly important in ensuring that resolution actions do not infringe constitutional or human rights relating to property and ownership. For example, the safeguards adopted in various pieces of secondary legislation by the UK authorities in its special bank resolution regime are important not only for providing legal certainty as to the treatment and status of netting and collateral rights in resolution but also more generally in ensuring that the UK complies with applicable obligations under the European Convention on Human Rights.
As more jurisdictions move towards the adoption of special bank resolution regimes many may have to grapple with complex constitutional issues. For some jurisdictions, providing resolution authorities with these powers might–absent appropriate safeguards – be incompatible with constitutional limitations on the state’s powers.
3. Institutional Capacity to implement an International Solution
37. For host country authorities to accept the leadership of home country authorities and to collaborate with other host authorities, the former must feel comfortable that the latter can effectively implement an international solution. This will require an organizational structure and staff that is capable of acting swiftly across borders. Given that several of the largest financial groups are active in more than 30 countries, this might in and by itself constitute an enormous challenge to overcome. Obviously, supervisory colleges are a tool to build up such capacity, as well as the necessary contacts with host authorities so as to facilitate cross-border inter-institutional cooperation.24 The coordination criteria described in this section B could take the form of a set of international standards to which countries could choose to adhere. Adherence would indicate to other countries a capability to implement an international resolution.
D. The Funding of Cross-Border Resolution
38. Although one of the key objectives of any resolution is to minimize the need for public funding, such funding may, on occasion, be needed, if only on a temporary basis. Given this fact, the question arises as to whether—and how—agreements on financial burden sharing among national authorities should be an element of any cross-border coordination framework.
39. As noted above, one of the key objectives of the proposed framework is that the final cost of the resolution is borne by private stakeholders. Box 8 gives a brief overview of how the final costs are typically allocated in both the liquidation and recovery of banks, and illustrates that it is more straightforward to impose losses through liquidation than through recovery action. In recovery, the equity position of pre-insolvency shareholders may be significantly diluted or completely wiped out, but imposing losses on existing creditors may be more difficult. As discussed earlier, a key objective is to design recovery tools that also allow for the imposition of “haircuts” (including through the establishment of the necessary legal basis for such an approach).
40. However, even if losses are imposed on creditors at the time of recovery, temporary public funding may still be necessary for a number of reasons. First, most legal frameworks do not have the necessary underpinnings for private sector “debtor-in-possession”-type25 financing of bank resolution processes. Second, even if such underpinnings were available, private providers of funds would often have difficulty in organizing the funding and structuring the process within the urgent context that is typical to failures of large, systemic banks. This is particularly the case in the context of systemic turmoil when other financial institutions face generalized funding pressures. Faced with such market failure, up front public funding provided by, as appropriate, the Ministry of Finance or the central bank (with protection against future losses by the MOF), may be the only option. A (partially) pre-funded “orderly resolution fund” (or a deposit insurance fund) may contribute to such funding.26 To the extent that there is a risk that, at the end of the process, the recovery fails and the national authorities face the risk of a loss, this can be addressed through the establishment of a fund that would receive ex ante (or ex post) contributions from the private sector.
Cost Allocation in the Domestic and International Contexts
The insolvency liquidation of a bank typically imposes costs on pre-insolvency private stakeholders. The unencumbered assets of the insolvent bank are sold and the proceeds distributed to creditors according their ranking. Losses are attributed consecutively to shareholders, subordinated creditors, and unsecured creditors. In practice, however, secured deposits are often transferred with a corresponding amount of unencumbered assets through a “purchase and assumption transaction,” before the shell is put into liquidation. Such transactions typically hinge on a preference granted to insured depositors and the DGS. As a result, the DGS would only incur losses if the available assets are less than the insured deposits, while shareholders, subordinated creditors, and unsecured creditors are unlikely to recover the full amount of their claim since most of the assets will be transferred to the acquiring bank. Similarly, if central banks have provided collateralized liquidity assistance prior to the insolvency, much of the estate’s assets will serve with priority the repayment of the central bank.
The structure of a recovery operation will determine whether and how the cost is shared between pre-insolvency private stakeholders. Recoveries are typically organized through the combination of the following techniques:
Capital Increases—The (private and public) providers of new capital will normally make their investment conditional to write downs on the capital of pre-insolvency shareholders, thus significantly diluting the interest of the latter. Moreover, newly issued preferred shares will take priority over ordinary shares both regarding income and liquidation dividends. However, the pre-insolvency unsecured creditors will benefit from the capital increase as they are more likely to be repaid.
Issuance of New Debt— New borrowing will impose costs upon the existing shareholders to the extent that it reduces the bank’s net profits. (In that regard, premia paid for public guarantees are included in the cost.) The claims of pre-existing subordinated creditors will be inferior to new unsecured claims. In contrast, the pre-insolvency unsecured creditors will not be directly adversely affected, although indirectly their positions may suffer if the new borrowing is not coupled with a capital increase.
Reduction of Liabilities—The unsecured debt of a bank can be reduced by (i) court-imposed haircuts, (ii) voluntary or forced conversion into equity, or (iii) a “leave behind” through a “purchase and assumption” transaction. Such operations impose almost by definition losses on pre-insolvency unsecured creditors and shareholders.
41. In light of the potential for temporary financing needs, the question arises as to how such needs should be coordinated in the context of the resolution of an international financial group. Home countries are likely to be unwilling or incapable of delivering all the public funding necessary to stabilize a large international financial group. By consequence, host countries may need to contribute financing if they want to keep the international financial group intact. Moreover, a host country’s decision whether or not to financially contribute to a group-wide solution ought to be informed by the fact that funding from the host country is likely to be required even if a strictly national solution is pursued.
42. Some form of financial burden sharing might thus be necessary and there will be cases where reaching an agreement between national authorities after a crisis has occurred will facilitate the recovery of a troubled financial institution. Ideally, agreement on burden sharing should be reached by the authorities of the principal jurisdictions on an institution-specific basis before a crisis occurs, especially if such agreements were to be supported by institution-specific recovery and resolutions plans (RRPs) or “living wills”. However, regardless whether it is before or after the crisis has occurred, reaching agreement on these questions will never be an easy task. For this reason, it would be desirable for the enhanced coordination framework to set out the range of criteria and parameters that would guide the burden-sharing process, e.g., (a) the relative systemic importance of the group across jurisdictions, (b) the relative contribution from DGS and any other resolution funds (if available) from different countries, and (c) the relative distribution of losses across jurisdictions.
E. Establishment of Coordination Procedures
43. Even if there is a group of countries that have satisfied the above coordination standards, their ability to actually coordinate rapidly and effectively will be enhanced if there is an established set of procedures that will serve as a road map in the context of a crisis.
44. Drawing on the corporate insolvency experience and, more specifically, the coordination framework established by UNCITRAL, a framework for the resolution of international financial groups could be designed in a manner that ensures that, in particular, there is an understanding of (a) who will take leadership in the initiation and conduct of resolution proceedings and how such leadership will be exercised, and (b) the modalities of communication and consultation that will take place during the process. The framework could apply between jurisdictions that adhere to the elements identified above. Moreover, it would need to be designed in a manner that could provide guidance both with respect to the resolution of a parent bank with foreign branches, and an international financial group, involving bank and/or non-bank subsidiaries. Although a number of issues would need to be resolved, the following general points may be made.
45. Where a financial institution with branches in foreign jurisdictions falls into financial difficulty, it is important to have clear understandings as to who will play the lead role in the initiation and conduct of the resolution proceedings. It would appear appropriate for the lead role to be played by the home country authorities. This approach would be consistent with the Concordat, and would reflect the reality that the parent jurisdiction is likely to be principal source of public funds necessary to finance a restructuring. As noted earlier, the procedural framework would need to specifically acknowledge that, while it would be presumed that a host country would accept the leadership of the home jurisdiction that adheres to the coordination framework, the host jurisdiction would reserve the right to act independently if it formed the judgment that independent action is more consistent with domestic financial stability and the interests of creditors.27
46. The modalities of leadership would depend on the circumstances. In the case of court-based proceedings, the home authorities could be given standing to launch proceedings in the host jurisdiction’s courts directly or through the host authorities acting on the basis of the guidance of the home authorities.28 In the case of administrative proceedings, the host jurisdiction’s legal framework could either permit the host authorities to conduct such proceedings on the basis of guidance provided by the home authorities, or permit the home authorities to do so directly.29 It would be expected that the home authority would design the overall resolution strategy, decide on the type of proceeding (e.g., restructuring vs. liquidation) to be launched in the home and host jurisdictions, and would play the lead role in the conduct of resolution proceedings. This would be a substantial departure from current practice.
47. While the above procedures are most directly applicable to a financial institution and its branches, the framework could also identify the modalities of leadership and coordination that would be applicable to the resolution of financial groups. The framework would clarify who is responsible for the resolution of each entity within the financial group. Following the approach taken at the national level in some jurisdictions (e.g., Italy), each country could designate a lead authority to initiate and conduct all resolution proceedings with respect to all bank and non-bank subsidiaries (both regulated and unregulated) and branches located within its territory, and to serve as a point of contact with “lead” authorities in other jurisdictions. Moreover, the framework would require these authorities to coordinate their actions to the maximum extent possible. While separate insolvency proceedings would be conducted with respect to each legal entity within the group, a host “lead” authority would be required to consult with the home “lead” authority before initiating resolution proceedings against a local subsidiary. It is also possible that the framework could require authorities to coordinate their actions on a range of issues—for example, by consolidating court proceedings involving separate entities that form part of a financial group (i.e., where the proceedings would remain separate but would be adjudicated by a single court at the same time) wherever possible, coordinating actions taken to protect assets, and by cooperating on the resolution of intra-group claims, and with creditors.
48. The implementation of such a system, for example, would require a very high level of communication and sharing of information amongst the supervisors and the resolution authorities. In taking key decisions on the resolution, the home authorities would be required to consult with the host authorities, to consider the impact of the decision on host jurisdictions. The relevant authorities and, in some cases, the relevant courts would need to have in place arrangements for communication and consultation and would need to have the statutory authority necessary to share highly sensitive information.
49. The framework should also require the sharing of information at an early stage of a financial institution’s difficulties. Such a requirement is particularly important to address the information asymmetries that exist between home and host authorities. Home authorities invariably have more information on an institution than do their counterparts in host jurisdictions. Unless host authorities have a high level of trust in the home supervisor and are confident that they will be fully informed of developments in the institution’s financial position and of the possibility of action by the home authorities, they will have every incentive not to cooperate and to ring-fence assets.
50. As a means of facilitating communication and consultation, consideration could be given to the establishment of institution-specific agreements. In the context of cross-border insolvency, the cooperation framework established under the UNCITRAL Model Law has been supplemented by cooperation agreements that are reached in the context of specific cases. These agreements, referred to as “protocols”, are approved by the courts and address specific modalities for communication and consultation that are relevant to the case at hand. While such protocols would be useful in the context of the resolution of financial institutions, they would need to be reached in advance of a crisis, given the need for rapid action. Such standing protocols could form part of the recovery and resolution plans that large financial groups will be required to establish.30
51. While one can debate the precise contours of the solution to the problems described in this paper, it is clear that there is a need for urgent action. Countries need to strengthen their resolution frameworks at the national level to ensure that ailing financial institutions and groups can be dealt with promptly and in a manner that protects the stability of the financial system. But effective action at the national level is not enough. Given the global nature of the financial services industry and its dominant institutions, national resolution frameworks will only be effective if they facilitate effective cooperation between authorities at the international level.
52. The approach outlined in this paper seeks to facilitate such coordination in a manner that is achievable in the near future. It is recognized that a number of issues would need to be resolved before this approach is implemented—in particular, what mechanism would be used to determine whether a country met the “core coordination standards” or how to monitor their compliance with these standards over time. However, the approach described above would form the basis for incremental progress being made as more and more countries voluntarily adhere to the framework over time. The “carrot” that would encourage countries to do so would be the possibility of a more effective and value-preserving international resolution. In the near term, a limited group of countries that already meet the standards described above could begin to cooperate amongst themselves. To the extent that these countries include the world’s principal financial centers, such cooperation would represent a major step forward. As other countries (e.g., developing countries and emerging markets) adhere to the standards over time, the circle of cooperation would expand. It would therefore represent a pragmatic and achievable mechanism for the strengthening of international cooperation worldwide.
Do Directors agree with the elements of the approach outlined above?
Do directors agree that an agreement among the world’s principal financial centers on an enhanced coordination framework would represent a major step forward? Should this requirement be complemented with a time-bound specific action plan?
Do directors think that it is desirable/necessary for the enhanced co-ordination framework to identify in advance the criteria and range of parameters that would be used to important to guide the burden-sharing process