Abstract
The global financial crisis hit the Netherlands' financial sector hard. This note analyzes the Dutch framework for crisis management and bank resolution, and formulates recommendations to address observed weaknesses. The overall framework for official financial support to stem systemic crisis is appropriate. The current framework for resolving ailing banks in going could be strengthened considerably. The framework for the orderly liquidation of banks could be strengthened and fine-tuned. The deposit guarantee scheme (DGS) has a number of helpful characteristics, but could be significantly enhanced.
I. Introduction1
1. The global financial crisis hit the Netherlands financial sector hard. The Dutch authorities intervened decisively and were able to contain the crisis and protect financial stability, but this was achieved with unprecedented levels of governmental financial support. Dutch taxpayers ended up contributing disproportionally, in part because of the lack of burden sharing arrangements (both ex ante and ex post) between countries and the lack of an effective international resolution framework. In addition, as in many other crisis-hit countries, the crisis revealed weaknesses in the Dutch financial oversight and legal frameworks, both for crisis prevention and for the orderly resolution of globally active financial institutions. Thus, understanding and addressing the weaknesses of the resolution framework will be instrumental to strengthening financial stability more broadly.
2. As part of the Financial Sector Assessment Program for the Netherlands, this Note analyzes the Dutch framework for crisis management and bank resolution and, where appropriate, formulates recommendations to address observed weaknesses. The Note is based upon a detailed analysis of the relevant legal and policy documents, and was complemented with intensive discussions with the Ministry of Finance (MoF), the Ministry of Justice, De Nederlandsche Bank (DNB) and the Authority for Financial Markets (AFM), as well as with stakeholders (including banks, insolvency practitioners and academics), during a mission to the Netherlands that took place from November 28 to December 14, 2010.
3. The key findings and recommendations of this Note can be summarized as follows:
Institutional Framework for Crisis Management—The institutional set-up of the official crisis response is broadly appropriate. The main areas requiring attention going forward are: (i) a shift of decision making power from the Judiciary to DNB in the context of bank resolution, reflecting DNB’s professional competence in this area as well as its role in maintaining financial stability; (ii) enhancing the willingness of the State, DNB and the AFM to support forceful actions by making them liable only in cases of gross negligence or willful misconduct, and (iii) a fine calibration of the respective roles of the MoF and DNB in bank resolution.
Official Financial Support—The overall framework for official financial support to stem systemic crisis is appropriate. The only significant weakness is the absence of a standing budgetary authorization for the Government to fund solvency support. Any such authorization would need to be designed in a way that does not increase moral hazard.
Restructuring under Official Control—The current framework for resolving ailing banks in going concern could be strengthened considerably. This challenging task could best be achieved by establishing a single regime for resolving banks under official control; such regime should set adequate objectives (including financial stability), tasks and powers for the official administrators.
Orderly Resolution in Gone Concern—The framework for the orderly liquidation of banks could be strengthened and fine-tuned. Introducing a consultative, if not steering, role for DNB and the AFM throughout the liquidation process would be particularly helpful in ensuring that financial stability concerns receive appropriate attention. The introduction of mechanisms supporting the rapid transfer of deposits (where appropriate combined with simultaneous transfers of assets) and essential functions (e.g., IT and payments) would also be critical.
Deposit Guarantee Scheme—The deposit guarantee scheme (DGS) has a number of helpful characteristics, but could be significantly enhanced. To be able to play a meaningful role in crisis management, the scheme should be ex ante funded and be authorized to fund bank resolution operations (including by financing the transfer of deposit books and “purchase and assumption” transactions). In addition, a well-designed depositor preference would buttress the effectiveness of such operations and reduce the cost of resolution for the back-up financiers of the scheme (i.e., the other banks and, ultimately, the State).
4. This Note is structured as follows. Following some preliminary observations (Section II), it will assess the institutional aspects of crisis management and bank resolution (Section III). The next sections will discuss the preventive (Section IV), early intervention (Section V) and crisis management (Section VI) tools available to the authorities, with a focus on official financial support (both emergency liquidity assistance and solvency support), resolution in going and gone concern, and the deposit guarantee scheme. The final section(Section VII) will discuss the role of the liability threshold for the effectiveness of resolution action.
II. Preliminary Observations
5. Before entering into the detailed analysis of the crisis management framework, it is worthwhile to consider briefly whether it is possible for national authorities to effectively resolve large international financial groups whose operations, assets and risks spread well beyond the national borders. This question is particularly relevant for the Netherlands, which is home to several large international financial groups, and which—as a member of the European Union (EU)—participates in a Union-wide single capital and banking market that has in a number of ways removed the levers of a member state to prevent, contain and manage a crisis within its borders. Furthermore, the present system of “home country” control and resolution lacks strong incentives for effective cross border supervision and resolution, in part because of the absence of ex ante burden sharing arrangements, which could induce home authorities to neglect prudential developments in other member states.2 For their part, while host states have few powers over foreign EU bank branches located in their territory, they cannot politically afford to go along with inadequate crisis management by home states (as illustrated by the Icesave failure). In conjunction with the absence of harmonized resolution frameworks, this may induce countries to ring-fence in times of crisis. In that light, IMF management and staff have argued that the integrated banking system of the EU needs an integrated framework for crisis prevention, management and resolution.3
6. The lack of ex ante burden sharing arrangements with relevant host countries is particularly unhelpful for the Netherlands as the country is home to the headquarters of a number of large financial groups. Although one of the key objectives of resolution frameworks is to minimize moral hazard, there may, on occasion, some forms of public funding may be needed—if only on a temporary basis—to reduce systemic risks. While there may be cases where an agreement between national authorities after the crisis has occurred (“ex post”) will facilitate resolution, the likelihood of success substantially increases if such agreement can be reached before crisis (“ex ante”). Ex ante agreements help to set the right incentives for home and host countries, save precious time during crisis, and avoid protracted negotiations that undermine confidence. Parameters and criteria for reaching such burden sharing agreement could be usefully included in an international enhanced coordination framework for bank resolution,4 but it is questionable whether such framework will be developed.
7. In the absence of strong imminent institutional reform at the international and EU level, the Dutch authorities would be well advised to strengthen soon and forcefully their domestic crisis management and bank resolution frameworks. While the ongoing discussions and proposals of the EU and the Financial Stability Board offer a useful conceptual framework and guidance, the importance of strong national initiatives cannot be overestimated.
III. Institutional Aspects
A sound institutional framework for crisis management requires (i) strong and clear mandates for the relevant institutions, (ii) an adequate allocation of labor across the institutions, and (iii) explicit coordination mechanisms between the institutions, including solid underpinnings for the exchange of confidential information in times of distress.
8. The mandates and allocation of labor regarding crisis management and bank resolution are broadly adequate. Such mandates are sometimes laid down more or less explicitly in law, and sometimes they are a derivative from more general constitutional and institutional principles. The powers and responsibilities of the key institutions can be summarized as follows:
MoF—Financial stability is appropriately considered to be a political responsibility of the Government, as represented by the MoF. In addition, and in line with its general responsibility for the budget, the MoF is directly competent for any measure that involves directly the public purse.5 In line with this logic, the MoF is involved in crisis management and bank resolution whenever systemic stability is concerned or a measure requires a direct contribution by the public finances. This is illustrated by the fact that the MoF was in the lead when it came to acquiring the Dutch part of Fortis, including its share in ABN-Amro, and providing solvency support to ING and the other institutions.
DNB—Subject to the caveat set out below, DNB is the lead authority for all bank resolution measures that do not raise systemic concerns or public solvency support. For instance, DNB is competent for emergency liquidity assistance to illiquid but solvent banks, as well as for imposing corrective measures to problematic banks. Moreover, DNB may limit the powers of the decision-making bodies of an ailing bank (through curatorship: see below) and instigate bank restructuring or liquidation proceedings by seeking the triggering of the “emergency mechanism.”
AFM—Whereas the AFM has no specific competences regarding crisis management and bank resolution, it contributes to the effectiveness of such action by balancing the maintenance of market transparency rules and sound conduct-of -business practices with the discretion needed to ensure the effectiveness of official financial support. More specifically, the AFM supports discrete bank resolution measures while requiring transparency when and if necessary.
DGS—The DGS is responsible for compensating insured depositors for losses caused by bank failures. The DGS does not have separate legal personality, but is managed and operated by DNB. While its overall mandate should be strengthened (see below), its institutional set up is appropriate. Specifically, the management by the central bank supports recourse to financial stability considerations in designing pay-out strategies, and avoids governance problems whereby other commercial banks are too closely involved in the resolution of troubled banks.
Judiciary—The Judiciary plays a prominent role in the resolution of banks and other licensed financial firms. Specifically, the courts—and not DNB—control the imposition of a trustee in bank resolution through the “emergency mechanisms” and oversee this trustee whether he undertakes to resolve the bank through restructuring or liquidation. Less directly but equally relevant, through the “Vie d’Or” case law, the courts have established themselves as the ultimate quality control of financial institution resolution measures in the Netherlands.
9. The overall inter-institutional cooperation arrangements (including information sharing) for crisis management are also appropriate. Based upon their respective mandates, the MoF and DNB have codified their roles regarding crisis management in a well-designed memorandum of understanding (MoU). This MoU recognizes the role of the MoF in crisis management and its ultimate responsibility for financial stability, while confirming the task of DNB to contribute to financial stability through actions falling under its jurisdiction (such as emergency liquidity assistance or micro-prudential supervision). Another plus point is that the Act on Financial Supervision (AFS) explicitly organizes the coordination between DNB and the AFM in cases where a curator is appointed, a license is withdrawn, or a bank resolution measure is triggered (Article 1:47). Finally, Dutch laws allow for information sharing in the context of crisis management, both between DNB and the AFM, and between DNB and the MoF.6
10. In combination with the many features of the “Twin Peak” model, the MOU between DNB and the AFM facilitated the robust and rapid response of the authorities to the crisis. In addition to the clear institutional set up and the information sharing arrangements, the combination of prudential supervisor and central bank under one roof was particularly conducive to shaping the authorities’ policy response.7 This was exemplified by DNB’s decision to establish and expand a coordination group on financial stability: this group originally met 3-5 times a year, but began meeting weekly during the crisis. Also, the group was quickly expanded to include representatives from DNB’s financial market, payment, and legal divisions. The process was further expanded and formalized: a crisis management group named “Pecunia” was set up including also representatives from the MoF and the AFM, and met daily. The coordination between DNB and the AFM also went smoothly, based on specific arrangements under the AFS.
11. Nevertheless, some improvements to the inter-institutional allocation of labor would be useful. We specifically advocate reforms along the following lines:
DNB and MOF—In designing the new bank resolution framework, the authorities will need to be precise in allocating responsibilities between the MoF and DNB for those cases where crisis management measures require both solvency support and the restructuring of a bank’s balance sheet. In such case, the legal framework should leave no doubt as to who is responsible for what. Specifically, if the power to take technical resolution decisions is generally allocated to DNB, the legal framework should indicate how the involvement of the MOF due to systemic and public purse concerns alters this responsibility (if at all) and shed clarity on which of the two agencies is ultimately responsible for the resolution process. In any event, accountability arrangements should mirror the division of labor.
DNB and Judiciary—Due consideration should be given to fine-tuning the interaction between DNB and the Judiciary with respect to bank resolution measures. This would ensure that the DNB, throughout the resolution process, can leverage its mandate and professional expertise with respect to the banking sector and financial stability more broadly. In that regard, it is recommended to strengthen the decision-making and steering role of DNB while transforming the current ex ante involvement of the courts into a crisp but adequate ex post review.
Liability Threshold—As will be discussed below, a well designed enhanced liability threshold which provides the authorities with greater legal protection would be a meaningful complement to the more prominent role that DNB would assume.
IV. Prevention
12. Effective implementation and enhancement of the supervisory framework would be an important element of effective crisis management framework While a detailed assessment of the preventive framework for financial crisis falls outside the scope of this Note,8 it may be useful to mention here briefly some initiatives that could considerably enhance the capacity of banks to withstand distress.
13. Important work is under way to design “buffers” that could protect banks in the event of distress. Supervisory authorities are examining a variety of macro- and micro-prudential tools that could add to the resilience of banks and reducing the likelihood of a bank failure. Amongst these, higher bank equity requirements (in line with Basel III) and contingent capital instruments with automatic debt-equity conversion mechanisms could be useful additions to the crisis prevention toolkit.
14. Another strand consists of the so-called “recovery and resolution plans.” For the Dutch authorities, a fundamental question will arise as to whether those instruments will be used for mainly informative purposes, or whether they will be employed to force substantial organizational and legal changes upon financial institutions so as to facilitate their orderly resolution. While hesitations to micro-manage business structures of the banks under supervision are understandable, the transfer of essential functions of insolvent banks (see below) could be considerably facilitated if those functions would be well organized within the appropriate legal structure.
V. Supervisory Early Intervention of Problem Banks
The authorities should have explicit and strong powers to require the board and management of an ailing bank to undertake action aimed at strengthening the bank, or to refrain from action that would further weaken it. In case the board and/or management would be unhelpful in the stabilization effort, the authorities should have the powers to replace directors and management rapidly and effectively.
Corrective Action
15. DNB currently has specific powers to impose corrective action on weak banks. In addition to fines and the modification or withdrawal of the license, the AFS authorizes DNB to impose the following measures on banks that are not in compliance with their regulatory obligations:
Directions—DNB may impose a “direction” on a financial institution that is in breach of obligations imposed by, or pursuant to, the AFS or if DNB detect signs of a development in solvency, equity or liquidity that may endanger the financial institution (Article 1:75). The purpose of the direction is to obligate the institution to adopt within a reasonable time frame a determined behavior on the issues set out in the direction.
Cease and Desist Orders—DNB may impose cease and desist orders upon financial institutions in case of a breach of the obligations imposed by, or pursuant to, the AFS (Article 1:79). The MoF may establish detailed rules regarding this power, but has not yet done so.
Special Measures—DNB can under certain circumstances impose restrictive and other risk reduction measures, such as higher equity or a specific provisioning policy (Article 3:111a).
16. Action should be undertaken to strengthen the hands of DNB in utilizing those powers as an effective early intervention tool. DNB has already developed internal guidelines that will facilitate recourse to formal corrective instruments. This should be complemented by action along the following two strands. First, DNB should start using those powers actively vis-à-vis weak banks,9 as well as document and publish its administrative practice in that regard (but not its action vis-à-vis individually named banks). This would indicate to stakeholders when and how DNB is using those powers, which would in turn increase the credibility of the tool as well as DNB’s moral suasion more broadly. As a positive side-effect, such transparency may also enhance the position of DNB when and if its early intervention powers would be challenged in Court. Second, the MoF should develop detailed rules for DNB’s cease-and-desist orders so as to facilitate the use of those orders as an early intervention tool. In doing so, the focus should be on how DNB can use those tools to impose decisive action for strengthening a weak bank. For instance, the MoF could usefully clarify that DNB can enforce through a cease-and-desist order the implementation of a restructuring plan, which could include the divestment of business lines and subsidiaries.10 Finally, it might be useful to consider whether some elements of the framework need specific legislative amendments so as to remove uncertainties and other obstacles to their use.11
Change of management
17. The powers of DNB to change directors and management of banks could be made more explicit. Currently, the AFS only authorizes DNB to remove underperforming directors and managers through a “direction” (Article 1:47 jo. Article 1:75). In consequence, DNB still relies on action of the bank to effectively dismiss and remove those officials. It might be useful to strengthen the powers of DNB in that regard. Specifically, consideration should be given to granting DNB the power to remove, under certain circumstances, directors (including members of the supervisory councils) and managers directly, and where necessary to replace them by others. Such reform could be coupled with a broader enhancement of the fit and proper rules, both regarding their content and their continuing application. Under some circumstances, such powers could be used by DNB as an “ultimum remedium” before putting troubled banks under official control.
VI. Crisis Management Tools
The tools for crisis management and bank resolution should include solid but flexible arrangements for official (temporary) financial support of banks, robust resolution powers for banks as a going concern, a mechanism for orderly liquidation as a gone concern, and a well-designed deposit guarantee scheme that is allowed to participate in bank resolution in a manner that minimizes the negative impact on depositors, other banks and the public purse.
A. Official Financial Support
Emergency liquidity assistance
The framework for emergency liquidity assistance should allow the state or an official agency (in particulars a central bank) to provide rapidly and in a legally robust manner emergency liquidity to illiquid but solvent banks. The liquidity provider should have tools to manage credit risks, including collateral requirements.
18. The Netherlands has a robust framework for emergency liquidity assistance (ELA). Under the Statutes of the European System of Central Banks, the member states remain competent to establish the framework for providing ELA to local counterparties.12 Apart from certain coordination issues to ensure the effectiveness of monetary policy, the European Central Bank (ECB) has no jurisdiction over ELA, which in the EU and elsewhere is typically provided by, and at the risk of, the (national) central banks.13 The legal framework of DNB is considered to allow it to enter into ELA operations with banks.14 On that basis, the authorities have developed the policy framework for providing such assistance (Box 1). That framework offers considerable discretion to DNB on the “when” and “how” of ELA. Looking backwards, DNB has been able to resort to ELA when and if it was considered appropriate. However, ex post external reviews have raised some questions on the modalities of such operations, such as the difference between DNB’s own collateral framework and that of the ECB for its monetary policy operations. To address those questions, it might be judicious to clarify (to the benefit of financial institutions, the political authorities, and the public at large) the parameters that would guide DNB in providing ELA, while safeguarding the right of DNB to maintain the necessary constructive ambiguity concerning its interventions in specific cases.
Policy Framework for ELA
The Netherlands’s policy framework for ELA is to a large extent guided by the 2003 letter of the Miniter of Finance to the Lower House and the 2007 Crisis Management MoU between the MoF and DNB. This framework can be summarized as follows:
DNB is confirmed to be the lender of last resort of the Netherlands and is responsible for providing ELA;
DNB has discretion to decide when and if to provide ELA;
DNB may provide ELA to banks and banking parts of financial conglomerates;
The framework is silent on the interest that DNB would charge for ELA operations;
DNB has discretion as to the adequacy of collateral provided to secure ELA transactions, although there is an expectation that the MoF and DNB will consult on collateral requirements when the amounts of ELA would reach significant levels; and
In principle, the MoF does not provide a state guarantee to DNB for exposures flowing from ELA.
Solvency support
To limit contagion risks and safeguard financial system stability, the framework for (temporary) solvency support should allow the state (or an agency of the state) (i) to enter rapidly and in a legally robust manner into transactions that buttress the solvency of ailing banks (recapitalization, guarantees, etc.,) and (ii) to execute those transactions rapidly.15
19. The framework for official solvency support is broadly appropriate, albeit with some room for improvement. The Government has taken over a bank (the Dutch part of Fortis, including its share in ABN-Amro) and provided solvency support to ailing banks in the form of recapitalizations, guarantees, and balance sheet relief (see the “Illiquid Assets Back-Up Facility” designed for ING). These operations have been structured as direct contractual agreements between the Kingdom (as represented by the Minister of Finance) and the receiving banks (or holding companies). Whereas the legal framework includes an explicit authorization for the Minister to enter into such agreements, the legal underpinnings for the ensuing financial transfers are less explicit. In particular, the Public Accounts Law (“Comptabiliteitswet”) requires the Parliament to grant prior approval to all participations in companies by the State, which is not helpful when rapid and discrete action is required in crisis circumstances. While the authorities have effectively dealt with this uncertainty in practice, the crisis management framework would be considerably strengthened if the availability of financial resources for solvency support could be explicitly ensured ex ante while ensuring that the design of such arrangements is such as to avoid moral hazard.
20. To provide more robust assurances, we strongly recommend introducing a standing budgetary authorization for the MoF to make disbursements for (temporary) solvency support. Specifically, a legislative amendment could offer legal underpinnings to the performance of all financial operations and transactions that seek to contribute to the stability of the Dutch financial system by providing financial assistance on a firm-specific (provided the firm is considered to be systemically relevant) or system-wide basis.
21. To minimize moral hazard, such authorization should be coupled with a number of measures that contribute to the judicious use of public moneys in support of the financial sector.
The solvency support should be predicated upon close interaction between the MoF and DNB. Specifically, it is recommended that the MoF should consult beforehand with DNB on the use of public funds, and that DNB is required to issue a written opinion on the appropriateness of such fiscal support. Furthermore, consideration should be given to an approach where the law makes the solvency support conditional to a determination by DNB (i) that the Netherlands is experiencing a systemic crisis and (ii) that DNB will not be capable of addressing the systemic pressure through its own tools (ELA and bank resolution).
Resolution policies and instruments should attribute losses to pre-insolvency financial stakeholders. The proposed resolution framework (see Sections B and C hereafter) would allow the authorities to allocate losses, including to unsecured creditors, during bank resolution processes. When strengthening the above-mentioned budgetary framework, the authorities could make an as explicit as possible link between future (temporary) solvency support and the allocation of such losses.16 This could be done, for instance, in a Governmental policy statement, a motion of Parliament, or even in legislation.17
The solvency support should be subject to robust ex post transparency requirements towards the Parliament.
B. Orderly and Effective Resolution
Orderly and Effective Resolution: Restructuring Versus Liquidation
Like other firms, failing banks can basically be resolved in two ways: restructuring or liquidation. (as is explicitly recognized by the EU Winding Up Directives) Under restructuring, the bank is resolved as a going concern, thus keeping the bank as a legal and licensed entity alive by sanitizing the balance sheet. (This is typically achieved by increasing equity, reducing debt, or a combination of both, coupled with shrinking the balance sheet.) Under liquidation, the bank is resolved by selling entire businesses, parts of businesses and individual or combined assets, and the distribution of proceeds to the creditors, thus effectively removing the failed bank from the market. For this note, liquidation is discussed as orderly resolution of a gone concern.
However, because of the specific nature of banks, the boundary between the two approaches is sometimes blurred, as illustrated by the following two examples.
Restructuring and liquidation may make use of the same techniques, such as transfers of assets and liabilities. However, in the former case, such transfers (e.g., to shed part of the bank’s network) will only lead to shrinking the balance sheet, whereas in the latter case they may (and typically will) transfer viable parts of the banking business as part of the bank’s subsequent liquidation as a gone concern (thus possibly causing losses to creditors that were left behind).
Whereas the loss of private control over non-financial firms in favor of an official agent is normally only triggered in case of liquidation due to insolvency (“bankruptcy”), bank resolution laws typically provide for going concern restructuring under official control (“official administration”), mainly because of financial stability concerns. Still, differences between these two types of official intervention remain. For instance, “claw back” and contract repudiation rules will in most (but not all) cases only be triggered in case of bankruptcy, and not apply to official administration.
Notwithstanding those complexities, the primary concern should be to provide to the authorities a broad toolbox so as to be able to choose the best possible approach in function of the idiosyncrasies of each failing bank under the prevailing circumstances.
Restructuring under official control
In case a bank cannot be stabilized in private hands, it will be necessary to transfer control over the bank to an official agent (“administrator”) before the bank appears to be technically insolvent. The administrator should have strong powers to restructure the bank in going concern if possible, and these powers will need to be buttressed by adequate support mechanisms so as to make this restructuring successful.
22. The authorities currently have two mechanisms for restructuring troubled banks in going concern by transferring varying degrees of control to an official agent. First, DNB may appoint a “curator” over troubled banks: the curator’s main task is to control the existing decision-making bodies of the bank without actually totally removing the latter (Article 1:76). Second, a trustee may be imposed upon troubled banks by virtue of the “emergency mechanism” (Article 3:160-162). The “going concern” task of such trustee is basically to transfer all or part of the liabilities of the troubled bank to a third party (Article 3:163).
23. Each mechanism separately features some useful characteristics, but overall they do not offer effective restructuring options in going concern. The main strengths and weaknesses of each mechanism are the following:
Curatorship—On the plus side, the curator is appointed and dismissed by DNB, and the AFS indicates that DNB has some degree of control over the appointee (although the law does not explicitly mention that DNB may give instructions). The grounds for appointing the curator are broadly appropriate. On the down side, the curator exercises some form of control (mainly veto power) over the decision-making bodies, but does not replace them. Moreover, the AFS is silent on the objectives of curatorship, and does not grant the curator explicit and effective restructuring tools. Finally, in particular as regards banks, it is highly questionable whether the curator can be used as a “silent” form of restructuring.
EmergencyMechanism—The grounds for opening the Emergency Mechanism (“EM”) seem broadly appropriate, and the trustee effectively takes over full control over the ailing bank. Less useful is that the trustee is appointed by the Courts—the DSB case has illustrated the risk of such approach—and that the law imposes an almost total moratorium—quite unhelpful from a financial stability perspective—while limiting the restructuring option to the transfer of liabilities only. Equally problematic is that the sole objective of the EM is to serve the interest of creditors; there is no room for financial stability concerns.
24. A single, strengthened regime should be introduced for restructuring banks under official administration that includes the following features:
Triggers—It should be possible to impose official administration at a sufficiently early stage (i.e. before the bank is actually insolvent) but on grounds of objective, clear, and proportionate triggers. A combination of qualitative (similar to the current triggers) and quantitative (e.g., related to capital adequacy, leverage and liquidity ratios) triggers would be ideal. Appropriate quantitative triggers (e.g., CAR at 50 percent or 25 percent) could effectively set a floor for intervention. The qualitative triggers might include references to the adequacy of management and to a “loss of market confidence that endangers the survival of the bank.” The opening of official administration need not be limited to a public interest test, or to systemic institutions only.
Appointment of Official Administrator—The administrator(s) should be appointed and dismissed by DNB, and not by the courts. DNB should have the power to appoint one or more administrators—who would then jointly operate as a college—in function of the size and the problems of the bank.
Status of Official Administrator—The official administrators should be agents of DNB: they should be under an obligation to follow DNB’s instructions, and be accountable to DNB. The key decisions should be subject to prior approval of DNB. As agents, they would not incur own liability, and DNB (or the Kingdom) would be liable for their actions.
Objectives—The main objective of the official administrator should be to contribute to the overall soundness of the banking system by providing orderly resolution; the pursuit of creditor interests ought to come only in second place. Most importantly, such calibrated objectives would allow the official administrator to take action (e.g., bank debt restructuring: see Box 3) that is not directly in the interest of all equally ranked creditors.
Basic Tasks—The basic tasks of the official administrator should be to (i) assess the real financial situation of the bank, (ii) where necessary, to establish a new balance sheet reflecting a fair and true view of that financial situation, (iii) if appropriate, write down capital accordingly, and convert subordinated debt into capital as needed; (iv) where possible, design and implement a restructuring plan for the bank (after DNB approval), and (v) in case restructuring is not an option, to prepare the bank for orderly liquidation.
Control over Bank—The official administrator should have strong and explicit powers to enter into restructuring transactions. This requires, first and foremost, that the administrator fully takes over the powers of all decision-making bodies of the ailing bank: the management board, the supervisory board and the general assembly of shareholders. (This does not imply that the directors and senior managers are, or need to be, automatically dismissed.) In consequence, the administrator effectively takes over managerial control over the bank, and is competent to enter into any business transaction and take any decision that falls under the jurisdiction of the decision-making bodies of the bank.
Restructuring Powers—The administrator needs the following restructuring tools: (i) rapid recapitalization of the bank without pre-emptive rights of pre-existing shareholders, (ii) conversion of subordinate debt into equity, and (iii) transfer (e.g. sale) of assets (including businesses and subsidiaries) with assumption of liabilities (sometimes called “purchase and assumption” transactions) to third party acquirers.18 To facilitate those operations in going concern, it might be useful to impose a selective stay on certain liabilities. Last but not least, consideration should be given to introducing a bank debt restructuring mechanism to restore the bank to viability (see Box 3 below).
Supporting Legal Framework—A supportive legal framework is needed for the bank restructuring decisions and transactions. For instance, deposit liabilities should be transferable without the consent of the depositors, and it might be useful to consider the need for an explicit “bridge bank” regime. Given the concentration of the banking sector, it would be judicious to analyze whether “purchase and assumption” transactions with other domestic financial institutions might be disproportionately hindered by competition rules.
Modalities—First, the costs of the official administrator should be borne by the bank; these costs should have first priority in case the bank is subsequently declared insolvent.19 Second, the restructuring transactions need to be given “safe harbor” protection against claw-back rules in case the transferor bank would subsequently be put into liquidation because the restructuring has failed.
Orderly liquidation
Authorities should develop procedures that allow for liquidating banks in an orderly manner. This involves rapidly transferring insured deposits (possibly combined with assets and other relevant assets) and critical banking functions (payment services, trade finance) out of the insolvent estate before the remainder is liquidated in the traditional fashion. Thus, those critical elements continue to operate in going concern, while the remainder of the failed bank is liquidated and removed from the market.
25. The current framework comprises two mechanisms for the orderly liquidation of banks. First, it is possible to liquidate ailing banks under the EM (Article 3:163). Second, even if a bank is being liquidated under the EM, it is possible to declare the bank bankrupt in case of (i) balance sheet insolvency, and (ii) when the EM is no longer useful. The Bankruptcy Law includes a special chapter on the bankruptcy of banks (Articles 212g et seq.). Specific provisions of the AFS (Article 1: 104) govern the withdrawal of the license of institutions under liquidation so as to remove the failed firms from the market.
A Bank Debt Restructuring Framework
In some circumstances, restructuring the debt of an ailing bank could be the most effective and efficient manner of resolving the bank. This will be particularly the case when a bank has a high loan-to-deposit ratio due to significant bond financing. The design of an adequate bank debt restructuring framework is, however, a challenge. To ensure that the bank remains functional as a going concern, the restructuring must take place within a very short time frame, avoid a run of creditors, and provide for financing during the restructuring period. In that light, it is recommended to consider introducing a bank debt restructuring framework with the following key features:
Triggers—The debt restructuring should be triggered at a late stage, i.e. when the bank is balance sheet insolvent, or about to become thus insolvent. In addition, the restructuring should only be approved if it is necessary for, and will bring about, a return to viability of the ailing bank.
Control of Management—Given that the unviability of the bank suggests that the board and management has previously failed in its stabilization efforts, it would generally be advisable to restructure the bank debt as part of “official administration.” This does not exclude, however, that parts of management stay in place where this is appropriate and useful.
Scope of Restructuring—Only subordinated and senior debt should be restructured. Insured and inter-bank deposits, secured debt (covered bonds), trade finance, and repos should be excluded.
Waterfall of Restructuring— To avoid that pre-restructuring shareholders and junior creditors benefit from haircuts imposed upon senior creditors, the restructuring should reflect the order of priorities in liquidation. Thus, losses should first be attributed to pre-existing equity (including post-conversion of contingent capital) until it has been fully written down. Next, any subordinated debt outstanding at the time of restructuring should absorb losses. Senior creditors should only take losses after shareholders and junior creditors.
New Capital—The restructuring should provide sufficient new capital to make the bank viable. This would be combined with dilution of the pre-restructuring shareholders, either by conversion of part of the haircutted debt into equity or through the issuance of new equity from third parties.
Limits on Walk Away Clauses—The debt restructuring should not trigger in and of itself the termination of the bank’s transactions and agreements. This can be achieved by inserting an appropriately designed prohibition on “walk away” clauses in relevant legislation.
Stay and Moratorium on Selected Liabilities—It should be possible to stay creditor action and impose a moratorium on selected liabilities (in particular the debt under restructuring) while allowing the bank to remain current on non-stayed debt.
Liquidity Assistance and Official Guarantees—To fund unavoidable outflows of non-stayed debt, and given the probable temporary loss of market access, the restructuring will need to be coupled with adequate official liquidity assistance. Official guarantees for some non-stayed debt would stem outflows.
Creditor Approval—The need for quick and decisive action in the interest of financial stability pleads against incorporating a procedure for creditor approval, even though such approval is typical in corporate debt restructuring.
Decision-Making process—The restructuring decision should be initiated by administrative decision of DNB. Depending on legal traditions and policy preferences, consideration could be given to requiring judicial approval in lieu of optional judicial review. Judicial involvement in the bank debt restructuring decision should, however, be structured in such a manner that it ensures speedy action and gives appropriate weight to the specialized expertise of DNB.
Cross-Border Effectiveness—The restructuring framework must be designed such that it qualifies as a “reorganization measure” under the EU Winding Up Directive, which would ensure that the haircuts are recognized in EU jurisdictions whose law might be applicable to the restructured debt.
26. A single, robust and flexible mechanism for the orderly liquidation of failed banks should be established. Both the liquidation component of the EM and the bankruptcy procedure comprise features that are helpful in liquidating banks in an orderly fashion. This is, for instance, the case of the power to transfer entire businesses, or parts of businesses, instead of selling each asset on a piecemeal basis.20 In any event, liquidators prefer to liquidate banks through bankruptcy instead of through the EM, even though the latter was especially designed for that purpose. This is explained by the fact that the liquidators feel more acquainted, and comfortable, with bankruptcy, and also because several aspects (such as procedural elements) of the EM give rise to questions and uncertainty. A combination of streamlining and fine-tuning the two mechanisms may therefore facilitate orderly liquidation.
27. To enhance financial stability, the streamlined liquidation regime should include mechanisms supporting the rapid transfer of assets, liabilities (including deposits) and essential functions. The mechanism should include the following:
Expressed power in law to transfer to a third party the assets, liabilities, and combined portfolios of both (sometimes called “purchase and assumption” transactions) in the case of actual insolvency.
To the extent not authorized under current law, the power to establish a “bridge bank” to exist under government ownership to acquire in a temporary capacity the assets, liabilities and essential functions in the event that an appropriate third party acquirer is not available.21
The law should authorize the liquidator to organize rapid transfers of assets such as shares in single purpose vehicles of the failed bank charged with, for instance, IT and payment system services. In case those services would not be separately incorporated, the liquidator should have the power to transfer rapidly all relevant assets, contracts and staff from the insolvent estate to a solvent acquirer so as to ensure continuity of those functions. Such transfers will require close interaction between the liquidators and the authorities.
28. Furthermore, the role of DNB and the AFM in the liquidation procedure should be strengthened. Currently, once a liquidation procedure is opened, DNB and the AFM cease to be involved in the further steps taken under the proceedings. The two agencies would have insights into the strengths and weaknesses of the banks under liquidation, as well as the suitability of any third party acquirer, which would be particularly useful for the liquidators. As a minimum, the two agencies should be consulted on all major disposals, and should have the right to formulate recommendations to the Courts and liquidators as to how the liquidation should proceed. (A stronger alternative would be to grant DNB and AFM powers to direct the liquidators, but this might pose difficult coordination issues with the Courts.) This would allow DNB to feed financial stability concerns into the liquidation process, and the AFM to share consumer protection concerns that are relevant to the liquidation measures.22
Intra-group and cross-border coordination
29. Resolution is rendered more complex in the context of a banking group composed of multiple branches and subsidiaries established in various jurisdictions. In some cases, these groups consist of banks only, while in others they include all kinds of financial companies (insurance, leasing and asset management for example) and thus exist as a true financial conglomerate. Therefore, the failure of a Dutch banking group might engage several authorities both within and outside the EU and, conversely, the failure of a foreign banking group may cause a need for resolution action vis-à-vis branches or subsidiaries established in the Netherlands. While the situation regarding jurisdictional competency for the winding up of EU branches of banks and insurance firms is comprehensively dealt with by the EU Winding Up Directives, the question of coordinating concurrent regulatory interventions and resolution actions in the context of a failing banking group remains largely unaddressed. To remedy this weakness, frameworks for early intervention and resolution need to reflect the fact that banks are increasingly part of larger, international groups.23
Intra-group cooperation
30. While the current resolution framework is legal entity-specific, it includes some, albeit limited, intra-group coordination tools. The AFS authorizes the appointment of a curator with respect to “financial institutions;” but the curator has only powers over the specific institution for which he has been appointed, and not its subsidiaries or affiliates. This being said, the term “financial institution” is defined widely and includes not only banks but also insurance firms and clearing institutions. Consequently, DNB could in theory appoint the same curator(s) for affiliated components of a financial group, thus ensuring some degree of coordination. In addition, DNB is empowered to determine the “interests” that should guide the curators. This power might be used to clarify to the various curators of intra-groups components that due regard should be given to the interconnectedness of the entity for which they are responsible with other group components.
31. The limited intra-group coordination tools should be enhanced in the context of the general overhaul of the bank resolution framework. Collapsing the insolvency estates of group entities into one single estate with a single body of creditors would be unhelpful and should be avoided. Such collapsing would make ex ante due diligence of the counterparties all but impossible, and would rely too much on consolidated accounts for counterparty risk management purposes. A preferred alternative is that the authorities consider including an explicit mechanism in law whereby the resolution procedures of a bank and its subsidiaries and affiliates are closely coordinated by the courts and DNB. Ideally, such mechanism should also allow for coordination between the Dutch authorities and the foreign administrators or liquidators dealing with foreign affiliates or subsidiaries of the troubled bank.
32. Such enhanced coordination tools would naturally be buttressed by strong insolvency frameworks for nonbank financial institutions. While the assessment of such frameworks goes beyond the scope of this Note, the authorities would be well advised to review the resolution frameworks for insurance firms, investment firms and other businesses that habitually constitute financial conglomerates.
Cross-border cooperation
33. The resolution framework includes several underpinnings for intensive cross-border coordination. The implementation of the EU “Winding Up Directive” is particularly helpful in that regard. Equally relevant is that DNB may request the opening of insolvency procedures over Dutch branches of foreign banks, without being under an absolute obligation to do so (Article 3:206), thus giving DNB the discretion to cooperate with foreign resolution authorities when that would be appropriate. Finally, for banks established in the Netherlands, the AFS explicitly recognizes the role of the Dutch courts in coordinating their insolvency procedures with administrative and judiciary authorities in host countries.
34. Going forward, there is room for further strengthening this framework. It would be particularly important to ensure that official administrators are explicitly authorized to coordinate their actions with their foreign counterparts with jurisdiction over the bank’s branches, subsidiaries and affiliates. This could usefully be buttressed by a specific mandate to take into account other countries’ financial stability concerns in so far as that would not be detrimental to the interest of the Netherlands itself. Another useful tool in liquidation would be the fine-tuning of the AFS (Article 3:209-211) so as to make ring-fencing optional when the EM is opened against the Dutch branch of a foreign bank. This would allow a Dutch liquidator to cooperate with the home country liquidator if that would be reconcilable with the Dutch national interest.
C. Deposit Guarantee Scheme
Whereas deposit guarantee schemes were originally designed to pay out insured depositors for their losses within a reasonable time frame (which could easily take months), it is increasingly well understood that such schemes can play a more prominent role in bank resolution, thus minimizing the fiscal cost and distress for depositors while maximizing trust in the overall banking system. This is assisted by allowing an ex ante funded scheme to contribute to bank resolution transactions based upon well-designed depositor preference rules.
35. The DGS can be made more effective as a bank resolution tool. When a bank becomes insolvent, DNB pays out covered depositors for their insured deposits, and recuperates subsequently those amounts from the other banks.24 While such approach might have had some advantages (i.e., it is ex ante cheaper for banks due to absence of contributions) in the past, it fails to provide the authorities with a strong and effective resolution tool in times of crisis, mainly DGS funds cannot be used to support bank resolution transactions. Furthermore, it might have offered some banks an opportunity of regulatory arbitrage relative to other EU member states by not fully internalizing the costs of a safe system for collecting deposits from the public.
36. To remedy this weakness, the DGS architecture is in urgent need of reform. In particular, the DGS should be redesigned away from a “pay-box system” into an effective tool for resolving insolvent banks. (The mission does not, however, advocate changes to the institutional set up of the DGS, which we deem appropriate: see higher.) While the forthcoming reform at EU level will set the broad parameters, there is room for rapid and decisive action that is not inconsistent with these current and future EU rules. In that regard, the following three aspects would be particularly germane to strengthening the DGS as a resolution tool.
Ex-ante funding
37. The DGS should be funded ex ante by contributions paid by banks in proportion to their insured deposits. Contributions can be determined as a flat rate, or on a risk-based basis. The latter looks theoretically attractive, but poses practical challenges with regard to the adequate measurements of risk and the avoidance of regulatory arbitrage. Once they are paid in, the contributions should be invested with due regard to the liquidity of the investments so as to be readily available for pay-out.
38. In case the available funds would be insufficient for the full coverage of insured deposits, the authorities should be explicitly authorized to close the funding gap through loans or grants, or a combination of the two. (Loans being the first best option.) Such official support is preferable to credit lines from the other banks, which could act pro-cyclically and thus exacerbate systemic distress. This begs the question whether official financial support for DGS shortfalls should be provided by the MoF or DNB; to be optimal, any future reform of the DGS should shed clarity on this matter.
Use of DGS funds as bank restructuring tool
39. The DGS should be allowed to financially support certain bank restructuring transactions. Specifically, the DGS should be explicitly authorized to perform, at the request of DNB, its insurance obligations by financially supporting the transfer of the insured deposits to another bank. This transfer could be achieved either through the auction of the relevant component of the deposit book or through so called “purchase and assumption” transactions, whereby insured deposits and a corresponding amount of assets are transferred from the insolvent bank to a transferee bank. While such transfer would be organized by DNB with support of the official administrator and/or liquidator, the role of the DGS would be to top up the transaction in case of insufficient assets, so as to fully compensate the transferee for acquiring the liabilities. Those powers should, however, be well circumscribed: the DGS should only provide financial support up to the amount of insured deposits, and under the condition that the transferor bank is subsequently put into liquidation (thus avoiding open bank assistance inappropriately benefitting the pre-insolvency stakeholders). Finally, for avoidance of doubt, this enhanced role of the DGS should be strictly supportive of the overall resolution strategy designed and implemented by DNB; we do not advocate morphing the DGS into a full-fledged bank resolution authority in lieu and stead of DNB.
40. To be able to perform such transactions effectively and rapidly, the authorizations should be complemented by regulatory obligations regarding the availability of information regarding insured depositors, operational manuals, as well as contractual frameworks and due diligence tools to prepare and perfect the necessary transfers within a rapid timeframe. Finally, due consideration should be given to the potential use of “bridge banks” in case private acquirers of the deposits would be wanting.
Depositor preference
41. Insured depositors and/or the DGS (either directly or through subrogation) should be given a high ranking priority right over the estate of the failed bank. Once a DGS pays out insured depositors, it would be subrogated in the claims of the latter. Without priority right, the DGS would then merely be an unsecured creditor of the insolvent estate, and be likely to suffer severe financial losses alongside (“pari passu”) the other unsecured creditors. A priority right, in contrast, would grant the DGS a preferential right over the unencumbered assets of the insolvent bank’s estate, thus allowing the DGS to recover its claim by preference over the ordinary unsecured creditors. Consequently, the immediate effect of such a priority right would be to protect the DGS, the contributing banks and ultimately the public purse against the cost of deposit insurance. The soundness of the banking system and financial interests of the State are public goods that deserve to be supported by a priority right relative to the position of ordinary unsecured creditors (not unlike the general priority right of the Treasury). By providing a mechanism for recovering the pay-outs from the assets of the estate, the priority right minimizes the final cost of the insurance, the possibility of shortfalls, and ultimately need for official support and topping up by the other banks. In addition, it dramatically simplifies “purchase and assumption” transactions by creating an inequality between insured depositors and other senior creditors. This justifies splitting the bank into going or gone concern by transferring the insured deposits away while leaving the senior creditors behind.
VII. Closing the Circle
42. An enhanced liability threshold for the State, DNB, the AFM and the official administrators, in the sense that they are only liable in the event of gross negligence or willful misconduct, would be instrumental in supporting forceful official action in times of distress. A strong crisis management framework requires both robust institutional arrangements and forceful tools. The link between the two consists of the willingness of the authorities to use those tools. Under the proposed framework, they will have to take difficult decisions based upon imperfect information (including on valuation of assets), and with the aim to avoid possibly disastrous scenarios. In an increasingly litigious society, there is a risk that stakeholders, who were forced to contribute to the resolution effort, seek compensation. This risk will become more serious if the proposed resolution framework shifts the cost of resolution from the public purse to the private pre-resolution stakeholders. In that light, it should be stipulated that legal liability for the actions of the State (in particular the actions of MoF with respect to crisis management and bank resolution), DNB, the AFM, and the official administrators will only be triggered by gross negligence or willful misconduct, and by a burden on proof that lies firmly with the plaintiffs.25
This Technical Note has been prepared by Wouter Bossu (Legal Department) and Jianping Zhou (Monetary and Capital Markets Department).
This will particularly be the case if those authorities have a policy preference for only saving the domestic part of the international banking group.
See Strauss-Kahn, Crisis Management Arrangements for a European Banking System, keynote speech at the EC Conference, Brussels May 19, 2010, and Fonteyne, Bossu, Cortavarria et al., Crisis Management and Resolution for a European Banking System, IMF Working Paper No. 10/70, on www.imf.org.
See IMF, Resolution of Cross-Border Banks - A Proposed Framework for Enhanced Coordination, 2010, p. 23-25, on www.imf.org.
In that regard, emergency liquidity assistance provided by DNB is not considered as directly involving the public purse, because the requirement that such assistance should be adequately collateralized would normally prevent the budget from suffering negative consequences in the form of either a reduced DNB dividend or the need to recapitalize DNB in case of severe losses.
The authorities were satisfied that information sharing between MOF and DNB related to crisis management was not hindered by legal obstacles. The AFS includes a general principle that confidential supervisory information cannot be shared with third parties (Article 1:89), unless the Law provides for a specific exception. Exceptions include for other supervisory authorities, curators and trustees, criminal prosecution services, and central banks. The sharing of information between DNB and the MoF has three legal bases: Article 1:42 (accountability arrangement), Art: 1:89 (that allows DNB to share information if necessary for performing its tasks”), and lastly Article. 37 of the DNB Law.
See the Technical Note on the Twin Peaks Model.
For an analysis on how an enhanced framework for macro-prudential supervision could contribute to crisis prevention: see the Technical Note on Financial Sector Supervision: the Twin Peaks Model.
Until now, DNB has been using those instruments extensively, but mainly vis-a-vis insurance firms and pension funds. The hesitation to apply them to banks has been ascribed to concerns regarding confidentiality and trust in the banks.
Divestment measures are not currently included in the early intervention powers of Article 136 of the EU Banking Directive. In the Netherlands, restructuring plans could be imposed through directions, the breach of which should be addressed by cease-and-desist orders.
Enhancing the liability threshold as advocated further in this Note would also be helpful in that regard.
However, the EU Treaty prohibits “monetary financing,” which is held to prohibit national central banks from providing liquidity support to insolvent banks.
The European Commission also has some jurisdiction over ELA provided by national central banks. In particular, it has decided that “pure” ELA (i.e., liquidity assistance to illiquid but solvent) does not fall under state aid rules, but liquidity assistance to insolvent banks would be governed and constrained by state aid rules.
See the March 2003 letter of the Minister of Finance to the Lower House, wherein the Minister takes the position that Article 8.1 of the DNB Law constitutes the legal basis for ELA by DNB.
On this issue, see IMF, An Overview of the legal, Institutional and Regulatory Framework for Bank Insolvency, 2009.
On this issue: see IMF, Resolution of Cross-Border Banks—A Proposed Framework for Enhanced Coordination, 2010, p. 23-25.
Possible legislative vehicles for such safeguards may be a new bank resolution law, a law on a bank resolution fund, or possibly even the budget law itself. In any event, if the legislative route is chosen, care should be given to balance safeguards with flexibility so as not to tie excessively the hands of the authorities when the next crisis strikes.
This may cause the transferor bank to become unviable, which should addressed by the withdrawal of the license and the bank’s subsequent liquidation.
Such rule is currently included in the AFS for trustees under the “emergency mechanism” (Article 3: 175), and should be kept going forward.
This may be useful in preserving critical components of the ailing banks by transferring them rapidly to third parties upon opening of the liquidation procedure.
It appears that the general capacities of the authorities include the powers to establish such “bridge bank.” In case there would be doubts regarding such powers, a more explicit empowerment would be useful.
Thus, the AFM could contribute to the rapid transfer to a third party of client securities, thereby limiting the contagion effect of the bank failure.
See IMF, Resolution of Cross-Border Banks—A Proposed Framework for Enhanced Coordination, 2010, p. 25-27.
The legal framework for the DGS is laid down in the Articles 3:258 et seq. of the AFS and the Royal Decree of October 12, 2006.
The recent Belgian Bank Resolution Law (Law of June 2, 2010: see Articles 3.5 and 5.5) stipulates that this willful misconduct or gross negligence must be considered by the Courts in light of “the concrete circumstances of the case, and in particular the urgency with which the resolution decisions were taken, the practices on the financial markets, the complexity of the case, the menaces for the saving system and the danger of damage to the national economy due to the failure of the bank/insurance firm concerned.”