I. A Short Introduction to Bank Resolution in a Global Context
Governments around the world were forced to step in and save banks in order to stem the effects of the global financial crisis starting in 2008. They were faced with the choice of either ‘bailing out’ financial institutions or risk the deepening of the financial crisis. The dilemma of financial institutions being ‘too big to fail’ created perverse incentives for bankers to engage in excessive risk taking and to ‘socialize’ their losses. Avoiding the recurrence of this situation has become a priority of governments ever since. At the same time, States must ensure the continuity of systemic banking functions, such as the running of payment systems, or else risk economic havoc and social upheaval.
In order to meet these goals, various tools have been devised which are usually summarized as ‘resolution measures’.Footnote 1 They are designed to ensure that critical banking functions will be fulfilled in times of crisis, to prevent contagion from affecting solvent institutions, and to avoid a general loss of trust in the banking system that might ultimately result in a bank run, without spending taxpayer money. The most important resolution measures are the bail-in tool and the transfer tool.Footnote 2 The bail-in tool aims at redressing the financial situation of the bank by reducing the equity of its shareholders and the liabilities held by its creditors. This is done by writing down equity and liabilities until as much as zero or by converting the liabilities into equity.Footnote 3 The transfer tool is designed to ensure the continuity of the bank's systemically important functions by transferring them, eg to a solvent institution or to a bridge bank. Both measures are comparable to an expropriation of an ailing bank's shareholders and creditors.Footnote 4 They are thought to stimulate them to exercise proper control over the bank's management and to punish them for their failure to do so, thus helping to maintain the stability of the financial system.
A crucial condition to make resolution tools work is their effectiveness across borders. In a world in which financial activities are no longer confined to States, banks frequently have assets abroad, for instance in a branch, a securities account or in parts of their payment system. If the resolution tools were restricted to the territory of the enacting State and spare foreign assets, an inefficient split-up of economic entities would follow, which might threaten the continuity of systemic banking functions. The measure taken by the enacting State must therefore encompass all of the ailing bank's assets, no matter where they are located. Similarly, debt instruments issued by banks are often governed not by the law of their home country, but by foreign law. A bail-in would be less effective if it did not cover this debt. It could even provide opportunities for arbitrage and trigger claims of creditor discrimination if certain bank debts were exempt from resolution.
But herein lies a problem: under traditional principles of conflict of laws or private international law, these assets and debt are governed by foreign private law.Footnote 5 They are therefore beyond the authority of the State that adopts the resolution measure (the ‘resolution State’). Rather, it is the law of the State governing the asset or debt (the ‘target State’) that will determine their ownership and content. Legislation or measures by the resolution State can have effect only to the extent permitted by the law of the target State.
Private international law on the one hand, and banking resolution on the other, are thus on a collision course. The problem is serious enough for the Financial Stability Board (FSB) to have recently published principles for the cross-border effectiveness of bank resolution measures.Footnote 6 It suggests different strategies which may be adopted by national legislators and supervisors in order to give resolution measures transnational force, independent of the law applicable under traditional rules of conflict of laws. From the supervisory perspective, the problem is particularly acute because of the legal requirement that banks must at all times have sufficient equity or debt that can be written down or (in the case of debt only) be converted into equity.Footnote 7 This threshold—called MRELFootnote 8 —must be distinguished from the ‘normal’ capital rules to which banks are subject under the Basel III standards and their transposition into EU law.Footnote 9 The requirement for the bank to have ‘enough debt’ may seem counterintuitive, yet this is necessary if the exercise of the bail-in tool by the resolution authority is to have some effect. Liabilities can only be counted towards MREL where the decision to bail them in is legally enforceable. With regard to bank liabilities governed by the law of a third country, the resolution authority may—and in certain circumstances must—require an institution to demonstrate that a bail-in decision would be effective with regard to liabilities governed by the law of a third country.Footnote 10 Should the resolution authority not be satisfied that this is the case, the liability will not be eligible for the calculation of MREL. The cross-border effectiveness of resolution therefore ought to be clarified in advance. The FSB recommends that States should identify those contracts and assets that cannot be transferred with legal certainty and assess the implications for the successful operation of the resolution tool.Footnote 11
But how can the effectiveness of resolution actions by a State be assessed where the law of that State does not govern the asset in question? And assuming that the resolution action is not effective, how can it be made effective? These are the questions which this essay examines. Special emphasis will be placed on the law of the European Union (EU) because the problems of cross-border effectiveness are most severe due to the integration of various States into a single market.
The structure of this article is as follows: Part I explains the background to the conflict between private international law on the one hand and bank resolution on the other, as well as strategies to circumvent it. Part II shows how EU law attempts to resolve the problem. Part III provides a brief overview of third country regimes and their interaction with the EU mechanism. Part IV suggests how the effectiveness of resolution worldwide might be improved. Part V provides a conclusion.
II. The Clash of Two Regimes
A. Private International Law vs Bank Resolution
Conflict of laws or private international law determines the law applicable to disputes and legal relations. Its modern form is a product of nineteenth-century legal thinking.Footnote 12 Although private international law derives from national law and differs from country to country, in most parts of the world similar or even identical rules are followed.Footnote 13 This means that the applicable law to a dispute is more or less determined in the same way no matter where a suit is brought. This so-called ‘decisional harmony’ essentially serves the interests of persons involved in private litigation. Ideally, they can foresee the applicable law regardless of the court in which a dispute is decided. Traditional private international law achieves this by classifying legal issues according to specific categories and using a specific connecting factor for each of them to identify the applicable law.Footnote 14 The aim is to find the country which has the closest or ‘most significant’ connection with the dispute or legal relationship.Footnote 15
This splitting up of legal issues becomes more noticeable when determining the law applicable to an asset or debt. Private international law distinguishes between different asset classes and determines the connecting factor for each of them differently. For tangible assets, whether movable or immovable, the connecting factor is their geographic location, or ‘situs’. They are therefore governed by the law of the State in which they are located. This is the so-called lex rei sitae rule,Footnote 16 which is known all over the world.Footnote 17 For securities embodied in physical documents such as share certificates, the so-called lex cartae sitae rule refers to the place where the shares are located.Footnote 18 Where shares are registered in electronic accounts, as is most often the case today, some countries use the place of the relevant intermediary which manages the securities account as the connecting factor (so-called ‘Place of the Relevant Intermediary’ or ‘PRIMA’ approach).Footnote 19 Different conflicts rules apply for unsecuritized bank debt, which is considered to be a contractual obligation. According to the principle of party autonomy, parties are free to choose the law governing their contract.Footnote 20 Where they fail to do so, a subsidiary rule is used: in many countries, the contract will be governed by the law in force at the habitual residence of the seller or service provider or, more generally, the party to the contract that must effect the characteristic performance.Footnote 21 For particular types of contracts, there are special connecting factors to be observed. For consumer contracts, for example, it is generally acknowledged that the law at the habitual residence or domicile of the consumer plays an important role.Footnote 22
In sum, private international law divides the assets and liabilities of the bank into specific categories and identifies the applicable law for each of them separately. The goal is to find the law that has the closest or most significant connection to the legal relationship in question. This serves the private interest of having a law applied to a legal relationship that is both foreseeable and close to the situation, and, most importantly, that does not depend on the court that is seized of a particular dispute.
Resolution measures interfere with this regime. They require obedience independent of the law that is applicable under the rules of private international law. To illustrate, an asset transfer purports to affect all property rights of the bank no matter where the assets are located. From the viewpoint of the target State, the measure seeks to have extraterritorial effect beyond the borders of the resolution State. Similarly, a bail-in seeks to write-down, cancel or convert debt independently of the law that applies to the debt in question. It even purports to affect debt that is governed by the law of a foreign country.
It follows that there exists a very deep conflict between private international law on the one hand and resolution on the other. At the heart of this clash seems to be a conflict between private and public interests. From the viewpoint of the private individuals involved, it would be preferable if the assets and liabilities remained subject to the law that is normally applicable to them.Footnote 23 Yet public interest requires deviation from such an approach. The immediate purpose of bank resolution measures is to maintain global financial stability, which is a common good. To achieve this end, it is necessary that the applicable law is that of the resolution State. The authorities of this State must have the power to bail in equity and debt and transfer assets which are governed by the law of another State. This may seem radical, but it is inevitable, given the need to address the critical situation of the ailing bank without creating additional risks for the stability of the global financial system. Ultimately, this stability also benefits private individuals because stability is the backdrop against which private transactions can flourish.
Presented in this way, it seems clear that resolution measures should be prioritized because they serve important public interests. Yet such a view would be too one-dimensional because behind the collision between private international law and bank resolution lurks a manifest conflict between sovereign States.Footnote 24 While the resolution State feels an obvious need to recapitalize its credit institution, the target State may have other goals. It may fear, in particular, that in the process of restructuring important business activities may be moved abroad. More palpable still is its desire to keep as many assets as possible for local creditors in the event of a looming bankruptcy.Footnote 25
Beyond such self-interest, there may also be other reasons to resist the effectiveness of foreign cross-border measures. A foreign government might engage in ‘abusive resolution’, for instance by discriminating against a target State's creditors or by using the resolution procedure as a pretext to cleanse its banks of debt.Footnote 26 It is also possible that a foreign resolution measure negatively impacts on the financial stability of domestic institutions whose rights as creditors are curtailed by a bail-in. In situations such as these, the governing law deserves respect.
The tension between private international law and resolution measures is therefore much more complex than the somewhat simplistic juxtaposition of private and public interests suggests. It must be borne in mind that private international law also safeguards important public interests, such as the rule of law, legal certainty and the protection of investments. Resolution measures should therefore not be enforced at any cost, but only where they serve overriding public interests and are in line with the principle of proportionality, ie do not go beyond what is necessary for the protection of public interests.
B. Contractual Bail-In Clauses as a Solution?
There are different strategies for avoiding the clash between private international law and bank resolution that has been described above. The most important of them is to include an explicit clause in bank debt that gives effect to resolution measures by the competent national authority. The FSB strongly recommends the inclusion of such contractual bail-in clauses in all debt instruments governed by the law of a third country.Footnote 27 The EU has turned this into a general obligation. According to the BRRD, Member States should require institutions to stipulate in debt instruments that the counterparty recognizes the write-down and conversion powers of the resolution authority and agrees to be bound by any of its decisions.Footnote 28 Through such a contractual provision, the counterparty autonomously accepts the resolution measures. Their basis is therefore no longer a command by the resolution authority, but the free will of the creditor, ie the counterparty. In other words, resolution is transformed from a statutory into a contractual mechanism.
Bail-in clauses will function as long as they are valid under the applicable foreign law governing the contract. Nevertheless, one must caution against an overly optimistic view of such clauses.Footnote 29 First of all, it will not always be easy to convince the bank's counterparty to accept them. Usually it will ask for an interest rate premium in return. It is true that such a premium will also be charged when the issuer is subject to a statutory bail-in regime, but a contractual bail-in regime is different because it will prove more costly than a statutory regime at the outset: the initial investors in the new instruments will require a hefty surcharge because their interests will be subordinated to almost all other creditors. Second, it is uncertain that bail-in clauses will be accepted by courts. Though they are supported by the principle of freedom of contract, there are important policy concerns that may be raised against private parties submitting to the decisions of foreign public authorities. At any rate, court judgments on this are lacking. Another drawback of contractual bail-in clauses is that it is difficult for the supervisor resolution authority to assess their effectiveness because their terms and their interpretation may vary from one instrument to another. A further disadvantage is that it takes time for the institution to build up a cushion of debt which can be contractually bailed in. Finally, contractual resolution clauses are of limited value for asset transfers. For instance, they do not work for transfers of immovable or intellectual property. The transfer of these assets often requires more than just an agreement by the parties, eg a record in a public register.Footnote 30
For all of these reasons, contractual bail-in alone is not sufficient, but merely a complementary piece in the puzzle. The FSB is right in recommending that States should continue to pursue the development and adoption of statutory frameworks, which may supplement and even supersede contractual approaches.Footnote 31 There is therefore no contractual panacea for the clash of private international and banking law.
C. The Role of Bank Group Structure and Structural Bail-in
The tension between the governing law and bank resolution may be somewhat reduced in the case of bank groups. It must be remembered that most credit institutions are not single institutions, but part of a larger group of companies with a presence in a number of States. The measures for their restructuring will be decided in the EU by a so-called resolution college.Footnote 32 This college brings together resolution authorities of the different countries in which parents, subsidiaries and branches of the bank group are located. Once the college decides upon a so-called ‘group resolution scheme’ all the authorities that have not explicitly disagreed must implement its decision.Footnote 33 The group resolution scheme therefore has an automatic cross-border effect. Provided that all of the assets and liabilities of the bank group are located in States bound by the resolution college's decision, there can be no conflict of laws. Unfortunately, however, the group resolution scheme is not always effective because every Member State reserves the right to disagree and depart from it.Footnote 34 Moreover, it does not work where some assets of the group are governed by the law of third States that are not represented in the resolution college.
Such deficiencies do not plague structural bail-in measures, which is a more elegant way of dealing with bank groups. They are mostly practised in the United States, where such groups usually have a holding structure. A structural bail-in is achieved through a write-down, cancellation or debt/equity swap at the level of the holding company. The measure is taken by the resolution authority in the home country of the holding company. Provided that the holding company has issued debt exclusively under the law of its home jurisdiction, no conflict with other countries can arise.Footnote 35 This is the so-called single point of entry (SPOE) strategy.Footnote 36
The SPOE strategy has various advantages compared to resolution measures that affect individual group members.Footnote 37 First, it functions even where the law of a third State governs the bank's assets since it does not purport to transfer those assets but simply redresses the financial situation of the holding company. Second, the SPOE strategy is less costly than a group resolution because only the holding company must be endowed with additional capital that can be bailed-in, which can be used to recapitalize any of the group's subsidiaries. Third, a structural bail-in is easier to execute and enforce than a bail-in at the level of the multiple entities of the group. Finally, this strategy avoids the risk of obstruction by the regulators of the various jurisdictions in which group members are established.
Yet the SPOE strategy also suffers from distinct disadvantages. First of all, it only works properly where the group has a holding structure.Footnote 38 This is not the case for many European groups.Footnote 39 It would therefore be necessary to introduce a statutory requirement for reorganization in order to provide the conditions for SPOE, which may be costly. Second, once the creditors of the holding company have been bailed-in, it will be necessary to transfer the capital from the holding to the subsidiaries in need. This can be done by writing down, cancelling or swapping debt that has previously been issued by the subsidiary to the parent. But where this debt is not governed by the law of the resolution State, the conflicts problem re-emerges. Third, the countries in which the group's subsidiaries are based may worry that the resolution authority responsible for the holding company will not act as vigorously to recapitalize a troubled subsidiary abroad as it would with regard to its domestic institution.Footnote 40 Fourth, if the home country does not act, the resolution authority of the subsidiary's host State may be tempted to adopt its own resolution measures and cancel or write down the debt issued to the parent in order to protect its national interest. It is true that it could adopt a ‘self-denying ordinance’ promising to refrain from any such measure,Footnote 41 but it would be impossible to legally enforce such a pledge. The possibility for regulatory conflict is therefore not completely avoided.
In spite of some undeniable advantages, the SPOE strategy does not eliminate all problems of cross-border resolution. This is why the BRRD presents SPOE as one among a range of options.Footnote 42 Equally possible is a multiple point of entry-strategy (MPOE). At least in Europe, resolution will most often be done by mandatory statutory rules at the level of the individual institution. This comprises bail-in, but also asset transfers. In these cases, conflicts with other States are inevitable.
D. Insolvency Law and the Principle of Universality
A way to resolve the clash between private international law and banking resolution is to characterize resolution measures as belonging to insolvency law. This is due to a basic tenet of insolvency law: the principle of universality. According to this principle, bankruptcy proceedings opened in one State aspire to have a worldwide effect.Footnote 43 Applying this principle to resolution could reconcile the latter with private international law. It would endow resolution measures with the necessary worldwide effect without putting the applicable law into question.
It is true that the principle of universality is not accepted by all States under all conditions, and is often abandoned in favour of the alternative approach: the principle of territoriality of insolvency. Yet universality prevails at least in certain areas and situations. A case in point is bank insolvencies in the EU. The so-called Winding-up Directive empowers the authorities of the Member State in which a credit institution is established to decide on the bank's reorganization.Footnote 44 The competent authority may rely on the law of its State (the ‘lex fori’) when making these decisions.Footnote 45 According to the Directive, the measures taken on the basis of the lex fori will be fully effective in other Member States.Footnote 46 The principle of universality thus governs bank insolvencies in the EU.
Crucially, the EU has decided to extend the Winding-up Directive's rules to resolution measures. It has introduced a new provision into the Directive according to which it also covers financial institutions or firms and their parents that are subject to measures under the BRRD.Footnote 47 More importantly, the term ‘reorganisation measure’ has been defined so as to include resolution tools and the exercise of resolution powers.Footnote 48 The BRRD and the Winding-up Directive are thus not separate regimes, but complement each other. The latter endows decisions taken under the former with universal effect throughout the EU.
Yet this technicality does not confer upon resolution measures transnational force in other Member States. The reason for this lies in the many exceptions that the Winding-up Directive contains. For instance, rights in rem in respect of tangible and intangible assets situated in other Member States are not affected by reorganization measures within the meaning of the Directive.Footnote 49 This exception is designed to protect the justified expectations of those creditors which have secured their rights by charges on the debtor's foreign assets and which rely on the exclusive applicability of the lex situs.Footnote 50 Reservations of title and set-off remain unaffected for similar reasons.Footnote 51 The same is true for proprietary rights registered abroad, which are subject only to the lex rei sitae.Footnote 52 The Winding-up Directive also spares transactions on regulated markets, such as the sale or acquisition of shares and bonds, in an attempt not to interfere with the functioning of exchanges.Footnote 53
The situation with netting agreements and repos is more complex. Netting agreements are contractual clauses that aim to reduce counterparty risk by providing for the acceleration, termination and set-off of contractual obligations in case of financial distress of the co-contractor.Footnote 54 Repos (repurchase agreements) are contracts under which a seller agrees to sell a security—a share or a bond—and to buy it back at a later date for a price fixed in advance, with the purpose of extending a collateralized loan to the seller and/or allowing the buyer to use the share or bond, eg to cover the obligation resulting from a short sale.Footnote 55 In principle, netting agreements and repos are solely governed by the contract law chosen by the parties.Footnote 56 However, the Winding-up Directive provides certain exceptions that help to ensure the effectiveness of resolution measures. These exceptions will later be studied in more detail.Footnote 57 It suffices to say here that the law normally applicable to termination clauses (which are part of netting agreements) and repos is not completely discarded in favour of that of the resolution authority.
In sum, although the Winding-up Directive gives universal effect to reorganization measures and has the effect of centralizing the applicable law, it has many exceptions. The Directive alone therefore does not overcome the clash between private international law and bank resolution. The transnational effect which it gives to resolution measures is not comprehensive.
It is not surprising that the Winding-up Directive does not provide the solution to the cross-border enforcement of resolution measures. The Directive principally pursues an insolvency approach, which is evidenced by the parallel structure to the European Insolvency Regulation.Footnote 58 Banking resolution is different. Its purpose is macroeconomic, not microeconomic. Contrary to insolvency, it does not merely seek to maintain the debtor's business or share its assets among the creditors. Rather, it aims at preserving the stability of the financial system, by avoiding bank runs and contagion to other institutions as well as by maintaining critical functions such as payment systems.Footnote 59 To put it simply, without effective resolution mechanisms citizens might not be able to draw money from ATMs in the event of a banking crisis. Making resolution effective is thus crucial to public welfare, and much more so than the reorganization of an ‘ordinary company’ like a manufacturer of automobiles or other products. The goal of resolution measures goes well beyond that of insolvency law, and is also alien to it. That is why the determination of jurisdiction does not follow the COMI (centre of main interest) principle that is found in insolvency law.Footnote 60 Instead, the competent resolution authority is that of the State in which the bank is supervised, independently of the seat of the creditors. The particular goal of resolution is also the reason why typical resolution measures such as debt cancellation or the transfer of systemically important functions to a bridge entity are unheard of in insolvency law. The latter is built on the principle of equal treatment of creditors (‘par conditio creditorum’). Though it provides for different ranks and privileges, it does not allow for the targeted reduction or even cancellation of particular debt. To be sure, the special resolution regime for credit institutions has nothing to do with extending special privileges to bankers. On the contrary, the goal is to make bank shareholders and creditors accountable without endangering the stability of the financial system.
III. EU Law and Cross-border Effectiveness of Resolution
A. A System of Differentiated Effectiveness
The key to the clash between the applicable law and bank resolution is to be found in the EU texts on resolution measures, the BRRDFootnote 61 and the SRM Regulation.Footnote 62 They organize a system of transnational cooperation designed to overcome differences in the applicable law. Yet they do so to a varying degree. With regard to the cross-border effects of resolution measures, EU law distinguishes between three different areas: the Eurozone, the Union at large and third countries.Footnote 63
The strongest effects of resolution measures will be felt within the Eurozone. Inside the 19 Member States sharing the common currency, resolution decisions will be adopted at the EU level by the Single Resolution Board (SRB) in Brussels. This is a European agency charged with taking resolution measures with regard to banks in the Eurozone.Footnote 64 The SRB's decisions must be transposed by the Member States.Footnote 65 In this way, they have transnational effect throughout all Euro-countries. One could call this the ‘centralized’Footnote 66 or ‘federal’ model.
Between the Eurozone and non-Eurozone countries and amongst the latter, a decentralized model is followed. The home country of the credit institution adopts the necessary measures, which the other Member States must recognize and support to become effective.Footnote 67 This is the country-of-origin principle which is found in other areas of EU law.Footnote 68
In relation to third countries, the EU lacks any prescriptive and enforcement powers. The competent European or national resolution authority therefore must rely on persuasion to make sure that its decisions are respected; where this is not the case, they are without effect. This could be called the ‘coordinative model’.
B. Solution of Intra-Eurozone Conflicts
The simple three-tier model that has been outlined above must be qualified. Though it is an essential element of the Banking Union, the SRB is not responsible for all banks in the Eurozone, but only for those that are supervised by the ECB. Broadly speaking, these are the systemically important institutions, or ‘SIFIs’. The SRB has powers also with regard to other banks where national authorities fail to respect its instructions.Footnote 69 Finally, it is competent whenever a resolution measure requires financing by the Single Resolution Fund (SRF).Footnote 70 In all other circumstances, Eurozone Member States remain in charge.
Even where the SRB is competent to decide, it usually does not have the power to take directly effective measures itself. Instead, it adopts a resolution scheme, which the national resolution authorities must implement.Footnote 71 For this purpose, the latter will use the powers conferred on them under the national law transposing the BRRD.Footnote 72 Only in the exceptional circumstance that a national authority fails to comply with a resolution scheme can the SRB exercise powers directly. In this case, it may itself order an asset transfer or a conversion of debt into equity.Footnote 73 As its measures are based on a regulation, they will have a direct effect in the Member States. They are supranational law, binding across the whole Eurozone.
One may wonder why the EU has not given the supranational powers to the SRB in all other cases as well, and provided its decisions with transnational effect in the Eurozone countries. The reason is the lack of a suitable competence in EU treaty law.Footnote 74 Under the so-called Meroni doctrine, the powers of the Commission cannot be delegated to new EU bodies that are not foreseen in primary law.Footnote 75 The European Parliament and the Council therefore chose to follow the model set by ESA Regulations which establish ‘supervisors of supervisors’ on the EU level;Footnote 76 a solution that has received the blessing of the CJEU.Footnote 77 Instead of a new agency with direct powers over market participants, it has created an EU resolution authority above national resolution authorities. The effectiveness of the latter's decisions is further restricted by the veto right that the Commission and the Council may exercise within 24 hours after the decision has been transmitted to them by the SRB.Footnote 78 This introduces a political element into the process, which creates uncertainty.
In sum, though the Eurozone has the most advanced system of integration with regard to resolution measures, the need for implementation by Member States’ authorities and the possibility of objections by the Commission and the Council may lead to delay and political wrangling. This can become a crucial obstacle given the severe time constraints under which restructurings must be carried out. Therefore, the transnational effectiveness of resolution measures is not even fully guaranteed in the Eurozone.
C. Solution of Intra-EU Conflicts
The most detailed rule on how to overcome the clash between private international law and bank resolution can be found in the BRRD. Its Article 66 addresses the problem with regard to all conflicts inside the EU. This includes relations within the Eurozone because Member States must use the powers conferred on them under their national law which transpose the BRRD to implement resolution measures adopted by the SRB, save for the few cases in which the SRB is allowed to implement its resolution decisions itself.Footnote 79
The relationship between Article 66 BRRD and the Winding-up Directive is not easy to fathom. It has already been shown that the Winding-up Directive gives automatic effect to resolution measures in other Member States, but contains a certain number of exceptions.Footnote 80 These exceptions are not repeated in Article 66 BRRD. The best way to interpret the relationship between the two texts is therefore to consider that the Winding-up Directive requires Member States to recognize foreign resolution measures, while Article 66 BRRD sets out the details of what this recognition entails, including those cases for which the Directive provides an exception.
1. Transfer measures
For the purpose of determining the cross-border effect of resolution measures, Article 66 BRRD distinguishes between two different groups of such measures. A first group concerns the ‘transfer of shares, other instruments of ownership or assets, rights or liabilities’.Footnote 81 Such a transfer can be effected by three resolution measures: the sale of business tool, the bridge institution tool and the asset separation tool. Under the Winding-up Directive, the effectiveness of these measures is limited, eg where they concern rights in rem related to assets located abroad. Article 66(1) BRRD therefore provides for an alternative mechanism. It obliges Member States to ensure that the resolution measures have effect ‘in or under the law’ of the other Member State in which assets are located, ie the target State.Footnote 82 The Member State which takes the resolution measure must comply with the conditions set by this law in order to effect the transfer.
This approach thus respects the rules of private international law. The major advantage of this is that it ensures legal certainty and protects the legitimate expectations of third parties. For instance, where the assets affected by a transfer measure are entered into a register, such as, for example, shares or real estate, the transfer must be made public in order to take effect. This guarantees certainty regarding legal ownership and the dissemination of information to the public. In addition, the EU might have had qualms about interfering directly with the Member States’ property law regime.
The drawback of this approach, however, is that it may undermine the resolution measure's efficiency. The authorities of the bank's home country must respect the law of the target State, but they will not necessarily be familiar with the law of the other 27 Member States in which the assets may be situated. Where they fail to comply with the prerequisites for a transfer under any of these laws, the measures they take will be without effect in such Member State.
The EU has attempted to address this information problem by providing an original cooperation mechanism. According to Article 66(2) BRRD, Member States shall provide the resolution authority with ‘all reasonable assistance’ to ensure that the transfer complies with any applicable requirements of national law. The resolution State and the target State whose law is applicable will thus have to work together. The Directive does not say what this implies exactly. At a minimum, the target State will have to provide the resolution State with information about its law in general. But that alone is not sufficient. One may assume that the target State is also required to inform the resolution State about the concrete steps necessary to transfer the particular assets in question. This is a far-reaching obligation. It is even more so because it applies not only to actual, but also to prospective transfers, because the addressee of the information is described as the resolution authority that ‘has made or intends to make the transfer’.Footnote 83 It follows that the information obligation already applies when a resolution is only at the planning stage. Member States must therefore cooperate early on to ensure that the measure will be effective under the applicable law.
2. Bail-In
With regard to write-down, cancellation or conversion of debt governed by foreign law or owed to foreign creditors, the BRRD follows another precept. Article 66(4) BRRD obliges the target State (called ‘Member State B’) to ensure that the transfer ordered by the resolution State (called ‘Member State A’) is effective. In other words, the burden is placed here on the target State. It must adopt measures in order for the debt of the ailing bank to be written down, cancelled or converted into equity under its law. One way to fulfil this obligation is to introduce a provision recognizing bail-in orders by other Member States. The result is a transnational effectiveness of the bail-in order without the need for any further steps on the part of the resolution authority.
At first sight, Article 66(4) BRRD seems difficult to square with the rules of private international law. The recovery measure directly affects debt governed by the law of another Member State. It is the target State's law which determines the content of the debt instrument and the person that is its creditor. On the other hand, the provision does not change the applicable law. It merely requires the Member State whose substantive law governs to change this law in view of the goal of the transfer. Applying the principle of equivalence,Footnote 84 the measure adopted by the recovery State must have the same force as measures taken by its own authorities. This is a truly transnational effect.
Member States will be able to comply with the obligation imposed on them where the debt in question is governed by their own law, ie the situation described in Article 66(4)(a) BRRD. Much more problematic is Article 66(4)(b), which obliges them to ensure the effectiveness of bail-in measures by other Member States where the creditor of a debt is located in their territory. In this case, the debt is not necessarily governed by their law under private international law rules. For instance, the parties may have chosen to apply the law of another State.Footnote 85 In such instances, it will be impossible for the State in which the creditor is located to write down, cancel or swap the claim because it is governed by another State's law.
The drafters of the BRRD seem to assume that a Member State could interfere with a claim under foreign law on the basis that the claim's creditor is domiciled on its soil. However, this is not the case. National law can introduce overriding mandatory or public policy rules, yet such provisions will only have limited effect before the courts of other States.Footnote 86 The legislator can also decide to expropriate the claim, but it appears that under general principles of public international law, an expropriation will only be valid where the debtor has its domicile or assets in the expropriating country.Footnote 87 The location of the creditor is currently not considered a sufficient connection justifying the taking away of a claim. Unless there is a change in the rules of public international law, which does not seem to be imminent, an expropriation by the country in which the creditor is domiciled is without effect. Therefore, the Member State in which the creditor is located cannot fulfil its task under Article 66(4)(b) BRRD. Its measures will, in all likelihood, not be enforced by the courts of other States.
3. Remedies
One possibly fatal obstacle for recovery and resolution measures could be legal actions seeking to enjoin them. If creditors, owners of shares or third parties could delay such measures by questioning their validity before the courts in different countries, then they are likely to fail given that resolution must be implemented in an extremely short time frame (often over a weekend) in order to avoid instilling a panic on the market. At the same time, the groups affected by resolution actions must have some access to justice. After all, measures such as a bail-in come close to expropriations.Footnote 88 The rule of law requires that courts control their legality.
The EU legislator has struck a balance. On the one hand, it has asked Member States to ensure that resolution measures cannot be challenged.Footnote 89 On the other hand, it obliges Member States that adopt such measures to grant the persons affected a right of appeal.Footnote 90 In this way, access to justice is preserved. It is simply monopolized in the resolution country. The BRRD adds that the appeal must be governed by the law of the Member State which adopts the measure.Footnote 91 But this is self-evident: since resolution measures are part of public law, they can only be controlled under the law of the State that enacts them and not under the law of another State.Footnote 92 The trickier issue which is not addressed by the legislator is the fact that resolution measures may also be the subject of civil proceedings. For instance, where the holder of a bond sues the issuer despite the fact that the bond has been written down by a bail-in measure, the civil court must incidentally rule on the validity of the bail-in measure in order to decide about the claim. Such disputes can be brought in many different courts for the same bond issue, and the law applied to them may vary. This creates substantial judicial uncertainty.Footnote 93
D. Solution of Conflicts with Third States
The EU lacks any prescriptive powers over non-Member States. It therefore cannot require them to change their law or to recognize orders issued by the SRB or a Member State authority. This has dramatic consequences for the efficiency of recovery and resolution. One need not look further than the many debt instruments issued under New York law. If the creditors of such instruments are European banks, the EU cannot change the identity of the creditor by transferring the instrument to a bridge bank or another institution. The reason is that, from the perspective of the state of New York, these instruments are beyond the remit of the EU.Footnote 94 The identity of the creditor is a question exclusively determined by New York law. Other States will take a similar position based on traditional principles of private international law. Absent a special agreement with the EU on this point, this may include British courts after Brexit, which would be particularly harmful given the high number of financial instruments subject to English law. In the converse situation, in which European banks are debtors of instruments issued under a third country's law, the EU also lacks influence. It does not have the power to directly write down these instruments or convert them into equity. The governing law shields the instruments against interference by foreign legislators. In a sense, that is exactly the purpose of a choice-of-law clause.
Yet one must bear in mind that bail-ins and resolution measures are necessary to safeguard global financial stability, an interest shared by all sovereigns and private individuals alike. Such measures would be incomplete if they spared assets and instruments governed by foreign law. From a macroeconomic perspective, it is therefore preferable that the resolution encompasses them as well, provided that the measure itself is legitimate and justified. The way in which the EU makes them effective must necessarily be indirect.
1. Cooperation agreements
The smoothest way to make resolution measures transnationally effective is to conclude an agreement with third States. The BRRD seeks to promote such agreements by authorizing them at different levels. It empowers the Council, on a proposal by the Commission, to conclude cooperation agreements.Footnote 95 Where the Council has not done so, each Member State may conclude a bilateral agreement.Footnote 96
The BRRD does not clearly state the subject of cooperation. Its Article 93(1) mentions that the agreements will concern information sharing, but also includes the important words ‘inter alia’. That recognition of EU resolution measures is the ultimate goal of cooperation can be gleaned from Article 94 BRRD. The provision allows Member States to unilaterally recognize and enforce third country resolution measures. It applies ‘unless and until an international agreement as referred to in Article 93(1) enters into force with the relevant third country’. This clause implies that where the EU enters into a cooperation agreement, it must ensure that it provides for the execution of resolution orders by third country measures. In exchange, the third State will demand a similar commitment with regard to EU measures.
Alternatively, EBA may also enter into a framework agreement.Footnote 97 Such an agreement may complement or substitute an agreement by the Council. It may also coexist with Member State agreements. As one of its subjects, the BRRD mentions ‘the application of resolution tools and the exercise of resolution powers’.Footnote 98 This may cover recognition of foreign measures. Yet such a framework agreement is by its very nature non-binding. It may thus not commit non-Member States to recognize EU resolution measures, and vice versa.
2. Unilateral recognition by third State
Another strategy to make EU resolution measures effective across borders is to consider whether the third country will recognize the EU measure of its own volition. This is by no means impossible. Many States around the world feel a similar need to restructure banks in financial distress.Footnote 99
The EU Member States’ power to unilaterally recognize resolution measures by third States is not contingent on reciprocity.Footnote 100 This is in line with the FSB recommendation that such a condition should be avoided.Footnote 101 It gives cooperation a head start and helps lift bureaucratic obstacles. Yet in recognizing foreign resolution actions, the EU certainly hopes that other States will return the favour. Article 94 BRRD therefore is a sort of advance compensation for future cooperation and the ‘bait’ that shall entice foreign countries to recognize the European measures.
3. Extraterritorial leverage
Where no international agreements exist and where third countries do not recognize European measures voluntarily, one must have resort to other methods. A well-known strategy of governments wanting to give extraterritorial effect to their actions is to exert pressure on those persons subject to their authority. Examples include embargo legislation or anti-suit injunctions.Footnote 102 Companies or persons present in the jurisdiction are forced to adopt a certain mode of behaviour with regard to foreign actions or relations.
Such extraterritorial leverage is also to be found in the BRRD. The Directive obliges the persons exercising control of the ailing bank as well as the transferee, ie an acquirer of the ailing institution or a bridge bank, to take all measures necessary to ensure that the transfer is effective.Footnote 103 These persons, the control person and the transferee, are subject to the EU Member States’ jurisdiction and can therefore be expected to fulfil their duties under EU law. They are obliged to take the necessary steps to ensure that the transfer is effective as a matter of foreign law.
It is not entirely clear which steps the person must take. In case of a transfer order, one can surmise that it must try to contractually effect the transfer from the bank to the buyer or bridge entity. In case of a write-down or conversion, it will be obligated to try to convince the counterparty either to renounce the debt or agree with its conversion into equity. Yet there will be obstacles to what the controlling person can achieve. The counterparty will normally not agree to bail-in measures without receiving substantial compensation. The obligation therefore only carries so far. What is factually impossible cannot be legally demanded, in line with the adage impossibilium nulla obligatio est.
4. Substitutes
For cases in which the above strategies are unsuccessful, the BRRD provides a default rule. The person controlling the ailing bank will have to hold the assets ‘on behalf’ of the recipient until the resolution measure becomes effective.Footnote 104 In other words, the EU obliges those under its remit—i.e. the persons controlling the ailing bank—to provide an interim solution. The formula gives room for interpretation, as can be witnessed by the different ways in which it has been transposed into national law.Footnote 105
5. Abstention and invalidity of measures
Despite the many provisions of EU law, there will be instances where a transfer or bail-in fails because of the applicability of foreign law and the absence of a contractual clause ensuring enforceability on the counterparty. The BRRD provides that where it is ‘highly unlikely’ that a recovery or resolution measure will become effective, the resolution authority shall not proceed with it.Footnote 106 The formula is conspicuously imprecise. It is also asking a lot to demand that the resolution authority, which must typically act under considerable time pressure, to analyse the chances of a transfer or bail-in of the many thousands of assets and obligations of a bank. Ideally, this question should therefore already be assessed at the resolution planning stage.
What happens if the resolution authority proceeds with the resolution despite it having little chance of success? The BRRD provides that in such cases the order should be ‘void’.Footnote 107 This sanction is quite rigid. It is also detrimental to legal certainty given that the conditions triggering the measure being void are so imprecise. According to the text of the Directive, the resolution measure is void when it is ‘highly unlikely’ that it will be effective. This is extremely vague. It is not even clear whether its assessment has to be made on an ex-ante or an ex-post basis. Given the many questions surrounding invalidity, it would have been better to avoid providing for such draconian consequences. It seems that the European rule did not go down well with the Member States, which is reflected in domestic law diverging from the EU position.Footnote 108
E. The Special Case of Termination and Netting Clauses
A very peculiar problem arises from termination clauses and their typical corollary provisions for close-out netting.Footnote 109 These clauses feature in complex debt instruments such as derivatives. They provide that in the event of an insolvency of one of the parties, the contract will end prematurely and all claims will be set off against each other.Footnote 110 If clauses like these were triggered by resolution measures, they could thwart any bail-in attempt because their operation reduces the number of claims and hence the opportunity for writing down, cancelling or converting debt into equity. The BRRD prevents this result by providing in Article 68(1) the fiction that resolution measures shall not be deemed to be an enforcement event for the purposes of the termination clauses and the European Directives that support their validity. In addition, its Article 71 requires Member States to grant their resolution authorities the power to suspend early termination rights that may arise for other reasons. This follows recommendations by the FSB that early termination rights and netting clauses under the relevant contracts should be temporarily stayed in order to prevent their interference with resolution measures.Footnote 111 A stay of a few days will generally suffice since the resolution is usually implemented very quickly.
For the temporary stay to work, it must also apply if the contract—including the netting clause—is governed by a law other than that of the resolution State. One will recall in this context that under the Winding-up Directive, netting clauses in general are not affected by resolution measures.Footnote 112 Yet the Directive now contains a special twist that allows the distinctive rules of Article 68 and 71 BRRD to work independently of the law applicable to the netting clause.Footnote 113 In addition, Article 68 is explicitly designated as an ‘overriding mandatory rule’ in the sense of Article 9 of the Rome I Regulation.Footnote 114 This is well meant, but it is insufficient. It works only inside the EU where the Rome I Regulation applies and where Union law thus has the power to define the notion ‘overriding mandatory rules’. It does not work before tribunals of third States which are not obliged to heed the temporary stay contained in the BRRD.
Again, having reached the limit of its prescriptive powers, the EU hopes for the cooperation of other States. The Directive extends the benefit of a temporary stay to resolution measures adopted by the authorities of third States.Footnote 115 They will not be considered as events triggering early termination rights. The only condition is that the foreign measure has been recognized by a Member State under Article 94 BRRD, but even this requirement can be avoided where an EU resolution authority so decides.Footnote 116 This is cross-border cooperation at its best. Crucially, reciprocity is not a requirement. But it is obvious that the EU hopes other States will return the favour of a temporary stay for European measures.
The most important initiative in the area of temporary stay, however, does not come from a national legislator, but from a private actor. The International Swaps and Derivatives Association (ISDA) published a ‘Resolution Stay Protocol’ in 2014, which was updated one year later by the ‘Universal Resolution Stay Protocol’, as well as a ‘Jurisdictional Modular Protocol’.Footnote 117 ISDA suggests that the users of its Master Agreement adhere to these protocols by signing them. Where both parties to a derivative have done so, an annex is automatically added to their contract, by which a special resolution regime regarding one of them will be enforceable upon the other. The various special resolution regimes are defined for different countries. The main effect of this annex is a temporary stay of the default rights, in particular the right to early termination. Thus through contract, the national resolution regimes are made binding upon parties outside the borders of the enacting State.
F. Inconsistencies in Member States’ Transposition
1. Legislative divergence
The BRRD is a significant step forward in the resolution of the conflict between bank resolution and private international law, but it has also become clear that its rules are complex and sometimes vague. It is therefore little wonder that the way in which they have been transposed into Member State law differs considerably. There is not enough space in this article to trace those differences exhaustively. Some examples might, however, give an idea of their type and degree.
As has been seen, the relationship between the BRRD and the Winding-up Directive creates some confusion.Footnote 118 This has led to divergent approaches in the Member States. The UK, for instance, has followed Article 117 BRRD's approach and treats resolution measures adopted by other Member States as reorganization measures under its Credit Institutions (Reorganisation and Winding up) Regulations 2004.Footnote 119 This means that such measures are recognized, but also that the exceptions found in that text apply. Immovable property, registrable rights, or reservation of title agreements, among others, will therefore be exempted from recognition.Footnote 120 The same approach is followed in France.Footnote 121 In contrast, the German approach simply provides that resolution measures by other Member States ‘have the same effect’ as those taken by the German resolution authority itself.Footnote 122 In other words, the German legislator puts foreign measures on an equal footing with its own. This far-reaching approach certainly helps the efficiency of foreign resolution measures, but contrasts with the more reserved attitude of other legislators.
Another example of inconsistent transposition relates to the situation in which Member State measures are not recognized in a third State. The BRRD prescribes that in this event, the person controlling the ailing bank will have to hold the assets ‘on behalf’ of the recipient until the resolution measure becomes effective.Footnote 123 This requirement has been transposed quite differently. French law obliges the controlling person to conserve the assets.Footnote 124 The UK Act states that the bank, and not the person controlling it, must hold the property for the benefit of the transferee.Footnote 125 Germany takes a middle way by obliging the controlling person to ensure that the bank will hold the property until the transfer becomes effective.Footnote 126 While the compatibility of the UK and German implementation of the Directive with the Directive itself is debatable, the differences between the legal systems are striking. The only point that all three countries agree upon is that payments to the transferee must be made on behalf of the bank.Footnote 127
A final example of inconsistency concerns the voidness of the resolution measure. The BRRD foresees that a resolution measure by a Member State is void in case it is ‘highly unlikely’ that it will become effective in a third country. This sanction has been criticized before as being overly rigid and impracticable.Footnote 128 It is interesting to note that the Member States have found ways to avoid it. The German legislator does not declare the transfer void, but instead obliges the resolution authority to revoke the resolution order.Footnote 129 The UK Banking Act provides for voidness, but conditions it upon a determination by the Bank of England that it is not possible to effectuate the transfer.Footnote 130 This determination necessarily comes ex post facto and is not based on the unlikeliness, but on the actual impossibility of the transfer taking effect. The French legislator has chosen to simply not address the issue and merely obliges the authority to abstain from the action.Footnote 131
In sum, national laws still diverge considerably regarding the recognition of resolution actions by other Member States and their relation to third States. Despite efforts at harmonization in the BRRD and the Winding-up Directive, their effectiveness is not guaranteed. This is partly due to conscious choices by Member States’ legislatures, but is also a result of misunderstandings caused by the complexity of the texts. With hindsight, it would have been preferable if the EU had proceeded by way of a regulation to avoid such differences. At least, it should have given the Member States a precise text on which to model their transpositions.
2. Judicial uncertainty
The legislative discrepancies are compounded by vagarities in the case law. The BRRD is still young, but it took off on the wrong foot. Two recent decisions illustrate the uncertainty surrounding it.
In May 2015, the Regional Court ‘Munich I’ had to decide on an aspect of the Hypo Alpe Adria saga that had clouded the relationship between Bavaria and Austria for a long time.Footnote 132 Hypo Alpe Adria was an Austrian bank that had lost its authorization as a credit institution and went into resolution. Its assets were transferred to a wind-down entity called HETA. Hypo Alpe Adria and HETA had both issued debt, with the Republic of Austria and the state of Carinthia acting as guarantors. The German public bank BayernLB had acquired part of this debt. A few days after the BRRD came into force in July 2014, the Austrian legislator adopted an act that cancelled or suspended some of these debt instruments and associated guarantees which were specifically mentioned.Footnote 133 Undeterred, BayernLB brought an action against Hypo Alpe Adria in the Munich court for the payment of the debt and guarantees.
One of the contentious points was whether the German court was obliged to heed the Austrian Act because of the BRRD. The claimant BayernLB alleged that the Austrian Act was not a transposition of the Directive, but merely a ‘fig leaf’ because the moratorium was not intended to restructure an active bank. Moreover, it argued that the BRRD would not apply to the case at hand since Hypo Alpe Adria only had subsidiaries and not branches in other Member States. The Regional Court clarified that the Directive covers credit institutions with subsidiaries in other Member States. Nevertheless, it came to the conclusion that the cancellation and moratorium were not covered by the BRRD, for two reasons. First, it agreed with the claimant that the Austrian Act was not intended to recapitalize an active bank, but rather resolved an ailing institution. Second, the court highlighted that the measure was not taken by an authority but by the legislature itself. It was therefore not a resolution measure within the meaning of the Directive, and a German court thus had no obligation to recognize it. While one can easily accept the first ground, the second one is more difficult to follow. It is hard to understand why the obligation to recognize a resolution measure by another Member State's administrative authority should not apply a fortiori to measures adopted by its legislature. Before these questions could be decided on appeal, the Austrian Constitutional Court put an end to the affair by invalidating the Act of 2014 on the ground that it discriminated between bondholders and unjustly repealed the guarantees taken out by the Austrian Republic and Carinthia.Footnote 134
Another case in which uncertainties over the BRRD emerged was decided by the High Court of Justice of England and Wales (Commercial Court) in 2015.Footnote 135 The claimant Goldman Sachs had acquired debt owed by the Portuguese Banco Espírito Santo. In August 2015, the Portuguese resolution authority decided to transfer all of the bank's debt to a bridge entity called Novo Banco. On the basis of this act, Goldman Sachs brought an action against Novo Banco in the English courts. The particularity of this action, when compared to that of BayernLB in the Hypo Alpe Adria case, was that Goldman Sachs wanted the resolution measure to be effective. In December of the same year, however, the Portuguese resolution authority had adopted another decision according to which the transfer to Novo Banco excluded the debt held by Goldman Sachs International on the ground that it had been a former shareholder of Banco Espírito Santo and therefore should be bailed in.
The question before Justice Hamblen of the Commercial Court was whether the court had jurisdiction over Novo Banco based on the forum selection clause in the debt instrument. Novo Banco opposed this on the basis of the December decision, which excluded its subrogation to the debt held by Goldman Sachs International. Novo Banco took the view that the Commercial Court was required by the BRRD to recognize this decision taken by the Portuguese resolution authority. It argued that Goldman Sachs International could not have the ‘plums’ without the ‘duff’, ie the August decision without the December decision. Justice Hamblen saw things differently, however. He ruled that the UK only needed to recognize the August decision which operated the transfer of all assets and debt, but not the December decision which excluded the debt held by the claimants from the transfer. He noted that at the time of the August decision there was no ‘duff’. Although he accepted that the BRRD allows for the transfer as well as the retransfer of debt, he refused to qualify the December decision in this way. In his mind, it did not fit any of the categories of resolution measures provided by the BRRD and was therefore outside of its scope. This manner of arguing seems highly legalistic and inadequate.Footnote 136 Because the resolution authority has the power to transfer and retransfer debt of its own volition, it must also be able to delineate the breadth of its own measures. In this context, the precise moment such a decision is taken can hardly matter.
Both of these rulings show a certain disquiet on the part of Member State courts to accept foreign resolution measures. This uneasiness is most probably due to the political nature of such measures and their interference with legal certainty and private property rights. The State which adopts them may be motivated by egoistic interests, such as cleansing its national banks from debt. Understandably, courts of other countries therefore consider it to be their prime role to protect their citizens and companies from foreign resolution measures. In doing so, however, they may undermine the goals of the BRRD.Footnote 137 One of these goals is to centralize remedies against resolution measures in the courts of the country of origin. Contrary to this intention, the validity and scope of such measures is made subject to scrutiny in civil proceedings concerning the enforcement of debt in other Member States. If one takes into account that such proceedings may be brought in many different civil courts all over the EU and even outside of it, it is easy to see that this may seriously compromise the efficiency of resolution measures.
3. Political risk
Apart from the legislative and the judicial branch, there is also the threat that governments may not heed the new resolution regime. They may be tempted to disregard its rules for political reasons. A recent illustration is the situation in Italy, where banks have a staggering 360 billion Euro of non-performing loans in their books.Footnote 138 The way to overcome this situation and recapitalize the banks under the BRRD is to slash their equity and debt or convert the debt into equity. Yet the Italian government is hesitant to do so because this would effect a large number of private investors, who were lured to lend money to banks with promises of high yields and safety. For fear of creating social unrest and political instability, the government therefore has looked into alternative options, including a taxpayer funded bail-out.Footnote 139 This would, however, directly contravene the prescriptions of the BRRD. Such political risk is difficult to avoid unless there is a supranational institution which has the power and willingness to enforce the resolution regime.Footnote 140 It is also not a typical cross-border problem, but one that surfaces in the domestic context as well.
IV. An Overview of Third Country Recognition Regimes
The efficiency of resolution measures under the BRRD in large part depends upon third States, which must have a regime in place that allows for the recognition of EU measures.
Switzerland is one of them. It has introduced a special resolution regime for credit institutions as early as 2003.Footnote 141 This was probably due to the fact that the country has huge banks in comparison to the size of its GDP, and therefore cannot bail them out, which forcefully demonstrates the need for a special resolution regime. Under the current law, the Swiss financial markets authority (FINMA) is empowered to recognize foreign insolvency decrees and other measures issued by the authorities at the real seat of a bank.Footnote 142 Although it does not directly follow from the text of the relevant provision of the Swiss Banking Act, the travaux préparatoires make it clear that this power also includes the recognition of third country recovery procedures.Footnote 143 The FSB estimates that the Swiss recognition procedure takes two months,Footnote 144 which is quite long considering the urgency with which resolution measures must be implemented. Alternatively, FINMA may allow access by the foreign resolution authority to assets located on the territory of Switzerland, the only condition being that Swiss creditors receive equal treatment in comparison to other creditors.Footnote 145 Following the FSB's recommendation, the Swiss rules do not contain any requirement of reciprocity. The Swiss authorities can therefore recognize foreign resolution actions even where the State from which they emanate would not or did not do the same with regard to Swiss resolution measures.
The United States is another country that establishes a comprehensive resolution regime.Footnote 146 The so-called ‘Orderly Liquidation Authority’ (OLA) was introduced shortly after the height of the financial crisis by the Dodd-Frank Act.Footnote 147 It comprises many tools that can also be found in the BRRD: asset and debt transfer, bail-in of unsecured creditors, suspension of termination clauses. However, it dedicates relatively few provisions to international cooperation.Footnote 148 The Federal Deposit Insurance Cooperation (FDIC) has drafted a joint paper with the Bank of England on a common SPOE approach, yet it does not contain any binding commitment.Footnote 149 In the absence of any specific text on recognition of foreign resolution actions, US courts will follow the same general principles as in bankruptcy cases. They may give effect to foreign resolution measures under the principle of comity, but will do so only on a case-by-case basis. In the past, pre-insolvency measures have been both accepted,Footnote 150 and rejected.Footnote 151 It is therefore impossible to determine in advance whether foreign resolution measures will enjoy recognition in the US.Footnote 152
Other major financial centres also have resolution mechanisms in place, such as JapanFootnote 153 and Singapore,Footnote 154 but it is also uncertain how they will deal with foreign measures. In other regions, the situation is even bleaker. To illustrate, special resolution regimes are still anathema in Latin America.Footnote 155 The IMF concluded in 2014 that
Many countries still lack comprehensive resolution powers for banks and other financial institutions (including the power to ‘bail in’ creditors) and effective mechanisms for the recognition of foreign resolution measures.Footnote 156
This fact makes international cooperation difficult and impedes the cross-border effectiveness of resolution actions.
V. Ideas on How to Improve the Cross-border Effectiveness of Resolution
The foregoing has shown that international cooperation is crucial to making resolution transnationally effective, but that it is uncertain whether other States recognize or support EU measures in this regard. It is particularly frustrating that this uncertainty will last until the very moment of crisis, when recognition and support is most needed. Moreover, the relative rarity of such events does not help cooperation. Game theory provides that the likelihood of cooperation increases with the frequency of encounters between the two sides.Footnote 157 States that are called upon to recognize foreign resolution measures of another country might assume—rightly or wrongly—that they will not find themselves in the same position at any point in the near future and, for this reason, they withhold cooperation. It might be feared, therefore, that cooperation will not emerge automatically over time. Even where States have cooperated in the drafting of a prior resolution plan, there is a time-inconsistency problem, as the precise conditions surrounding the resolution cannot be known in advance.
The conditions for mutual recognition and support are thus far from being ideal. International cooperation is not to be assumed, but needs to be fostered. What can be done? The FSB has taken an important step with the publication of ‘Key Attributes of Effective Resolution Regimes for Financial Institutions’ and the ‘Principles for the Cross-border Effectiveness of Resolution Actions’. Yet it seems unlikely that they will suffice. In addition to being non-binding, they are notoriously imprecise. For example, the texts are conspicuously vague about the situations that may allow resolution actions.Footnote 158 This may lead to abuse, which may in turn hurt recognition by other States. The conditions for such recognition are equally loosely formulated. The FSB in some instances simply relies on the ‘good faith’ of the resolution authorities.Footnote 159 This formula is too unclear for adjudication; States need more precise criteria. Another drawback of the ‘Key Attributes’ and the ‘Principles’ is that they are couched in regulatory terms. They mainly focus on cooperation between resolution authorities. Yet ultimately, the effectiveness of resolution actions will depend on courts in which they will be tested. If it will not be upheld in litigation, the extensive regulatory framework that has been developed over the past years may break down like a house of cards. It is therefore necessary to convince judges of the need to recognize foreign resolution measures as effective, even where they are at variance with the applicable law under classic conflicts principles. Many judges will be unfamiliar with international soft law texts such as the FSB recommendations, and might find it difficult to transpose them into a legal context. In order to spur worldwide convergence on resolution, one must resort to other methods of transnational law-making.
A. An International Treaty?
One idea would be to negotiate, sign and ratify an international treaty which might require States parties to respect and support each other's resolution measures.Footnote 160 It could stipulate conditions for such an obligation, for instance that the measure was adopted by the authorities at the seat of the bank. It could also place limitations upon resolution powers or restrict them to certain cases.
For all its possible benefits, it is unlikely that a treaty would work. First of all, States will be reluctant to assume any binding obligation under public international law in an area that is so crucial to their economic, political and social well-being as the protection of banks and bank creditors’ assets. And even if they were, one may reasonably doubt whether it would be of any use in a crisis.Footnote 161 Confronted with pressing realities, national authorities would probably prefer to protect short-term domestic interests, even at the cost of violating their treaty obligations. Public international law is inherently weak where economic and political stakes are high. Other governments know this and will anticipate non-compliance. States would therefore find themselves in the uncomfortable situation that they were unable to make a credible commitment before a crisis.
B. Uniform Resolution Law
A more promising suggestion would be to draft a uniform text on resolution, such as a legislative guide or—even better—a model law.Footnote 162 The idea would be to formulate a set of rules containing precise conditions for bail-ins and transfer orders, and for recognition and support of resolution measures by other States. The backbone of such a uniform text would be formed by the FSB ‘Key Attributes of Effective Resolution Regimes for Financial Institutions’ and the ‘Principles for Efficient Cross-border Resolution Actions’. They would, however, need to be more precise.
Such precise rules could be inspired by national legislation. The resolution regimes eg of Switzerland or of EU Member States could serve as a starting point. They would have to be stripped of their idiosyncrasies and transformed into a baseplate that is universally acceptable. This work could be done by one of the different organizations that have uniform law as their mission, such as UNIDROIT, UNCITRAL, or the Hague Conference on Private International Law.
A legislative guide or model law could bring many benefits. First, it would ensure that resolution actions are recognized by the courts and not just by administrative agencies. One could imagine, for instance, that courts could be obliged to respect such measures provided that they have been agreed upon or accepted by their national resolution agency. In this way, cross-border regulatory planning and cooperation becomes effective in courthouses. Second, a uniform text would guarantee a level playing field for resolution. A State recognizing the measures of another would have a reasonable chance that its own measures would receive the same treatment in similar circumstances. It would therefore not withhold its cooperation for fear of being duped. In addition, a legislative guide or model law could assist those States who were unsure which text to adopt. Many States lack the expertise and the experience in financial law that is necessary to draft their own resolution law. Being far from the financial centres of the world, they also have no particular interest or stake in these matters. This may be a major hindrance for any adoption of an appropriate law. An internationally backed model could act as a catalyst inspiring all States to take action on this matter.
An obvious counterargument against harmonization is sovereignty—an ambiguous word, which has numerous dimensions. First, sovereignty may mean that there is a need to adapt a State's resolution regime to the particularities of its existing legal system bearing on such questions. Yet resolution is a highly technical area, stands on its own and is not tightly woven into the rest of a country's legislation. It is also a new area and so it is unlikely that there will be much existing law which could stand in the way of the adoption of a new act. There will thus be not much need to adapt a common approach to local circumstances.
Second, sovereignty may express the need for policy space and regulatory diversity.Footnote 163 Yet a model law on resolution is not at variance with this aspect either. Resolution is not an area where creativity is needed, but rather strong and robust rules must be applied in the same way around the globe. While policy space and regulatory competition are important, they would not be beneficial in an area where the temptation to protect domestic interests is so strong as in resolution. Any benefits of new and innovative rules would be clearly outweighed by the damage of legislative divergence.
In addition, sovereignty is not directly compromised because the suggestion of a model law leaves room for national deviation. Being a model law, its rules can still be departed from. Whilst this limits its harmonizing effect, a model law may nevertheless bring some benefits. First, it will provide guidance for those legislators who do not know which rule to adopt. Second, it will provide transparency. Against its backdrop, any domestic deviation stands out. It will thus become clearer where States diverge and how. As a result, national legislators might feel pressure to at least justify any departures from the global model regime. Moreover, the experience of other model laws, such as the Uniform Commercial Code (UCC), suggests that deviations will be limited. Finally, a uniform text may lead to the emergence of global case law on resolution that can be catalogued and compared, such as that which exists in relation to the Convention on the International Sale of Goods (CISG) and other texts.Footnote 164
A model law, if well drafted, will therefore ultimately result in increased transparency and uniformity, while maintaining some policy space for States. It would greatly enhance the potential for cross-border effectiveness of resolution actions. It would also entice global cooperation and strengthen the foreseeability of such actions.
C. More Inclusive Resolution Colleges
A further and final proposal is to improve the cooperation through resolution colleges. So far, such colleges are set up when a bank has significant branches or subsidiaries in more than one country.Footnote 165 But other States will be affected by such a measure as well. This is true, in particular, for the target State whose law governs the bank's assets and debt. This State has an interest in protecting local creditors. It may therefore be tempted to resist any foreign resolution measure or to adopt its own one.
Including target States in the resolution decision could improve the effectiveness of the measure. The countervailing interests could be openly discussed and a mutually satisfying solution be found. It could also help to make clear from the beginning which actions have the chance of being recognized by the target State and which do not.
One should therefore seriously consider the possibility of including target States into resolution colleges. This, of course, need not apply to each and every State whose laws govern a miniscule amount of a bank's assets or debt. The proposal made here is restricted to those States in which a considerable part of the patrimony is located or whose law applies to an important number of debt instruments that are subject to resolution and recovery measures.
Arguably, the presence of target States complicates the decision-making process. Where they are in a minority, they may be wary of submitting to an institution in which they are outnumbered by EU Member States. Conversely, where assets are dispersed over many third countries, these target States could be in a majority, prompting concerns that Member States could be outvoted. Yet these concerns are largely unfounded because decision-making in the resolution college is not binding upon the participating States: The BRRD allows participating States to disagree with a group resolution scheme or depart from it provided it gives detailed reasons for its behaviour.Footnote 166 It is true that the discussions in the resolution college may be more partisan and fractioned where non-EU States are involved. Yet this only mirrors the practical problems that will be encountered anyway once steps are taken to put the bank into resolution. Dealing with third countries cannot be avoided at this stage, when they may be more inclined to reject decisions in which they have not been involved. It therefore makes sense to involve them from the beginning.
VI. Conclusion: Resolution and Law's Elasticity
Katharina Pistor has observed that financial law tends to be relatively elastic at the apex of the financial system and more binding on the periphery.Footnote 167 The foregoing analysis of the effectiveness of resolution measures provides a striking confirmation of this insight. Resolution measures are prime examples of law's elasticity. They may transfer rights in assets to third parties, and they may reduce or even cancel straightforward contractual claims (‘IOUs’). The more effective resolution measures are, the less binding property rights and obligations are. As has been shown, the effectiveness of resolution measures diminishes on a sliding scale from the Eurozone to third States. The measures are most effective in the Eurozone, which is the most deeply integrated part of the EU. That means that property rights and assets are most flexible there. A middle area is occupied by the Member States that are not part of the Eurozone, which recognize resolution measures taken by other Member States or by the SRB. The measures are least effective in relation to third countries, where property rights are seen as ‘hard law’.
This diminishing effectiveness of resolution, and the consequent hardness of law, does not come as a surprise. Its basic reason is the limitation of sovereign powers. The EU strives to make its resolution measures effective across borders. Nevertheless, it encounters natural limits where the law of a non-Member State is applicable according to the traditional rules of private international law. The EU cannot simply impose its measures on third States whose law governs debt or where assets are located. Rather, it must rely on transnational coordination. It is this coordination which makes resolution measures effective and thus increases the law's flexibility. Two suggestions on how such coordination might be enhanced have been made in this article: the elaboration of a model law and the inclusion of target States—ie States whose law applies to the bank's assets—into resolution colleges.
