K. Wojcik, Bail in in the banking union

Common Market Law Review 53: 91–138, 2016. © 2016 Kluwer Law International. Printed in the United Kingdom.

BAIL-IN IN THE BANKING UNION KARL-PHILIPP WOJCIK*

Abstract On 1 January 2016, the Single Resolution Mechanism as the second pillar of the EU’s Banking Union became fully operational. At the same moment, bail-in, i.e. the statutory power of resolution authorities to cancel shares and to write down or to convert liabilities of a bank which is failing or likely to fail, became mandatory in the EU. This paper sets out the newly created regulatory framework on the recovery and resolution of banks in the EU, focusing on the main features of the EU’s rules on bail-in in the Banking Union. It examines the legal and economic impact of bail-in in general and highlights the various challenges for the application of bail-in. Finally, this paper assesses whether bail-in can attain the goals which it is meant to achieve and what it will take to make it effective. 1.

Introduction

The latest financial crisis was a reminder to the world that banks, even large and cross border ones, can fail.1 The most likely measure to deal with a bank threatened with insolvency during that financial crisis was the use of taxpayers’ money to bail it out. According to data published by the European Commission, State aid for the financial sector in the EU amounting to in total ¤5.76 trillion was approved by the Commission between 2008 and 1 October 2014.2 * Member of the Legal Service, European Commission. In this capacity, the author advises the Commision inter alia on all legal aspects pertaining to the “Banking Union”, namely the creation and implementation of the Single Supervisory Mechanism (SSM), the Single Resolution Mechanism (SRM) and the European Deposit Insurance Scheme (EDIS). The views expressed by the author are strictly personal and do not engage the Commission. 1. For an account of earlier bank crises, see Haentjens, “Bank recovery and resolution: An overview of international initiatives”, 3 International Insolvency Law Review (2014), 255–270, at 255–257. 2. Aid to be used for guarantees, asset relief interventions, capital injections and liquidity measures. See (last visited 14 Dec. 2015). Not all of these approved amounts were actually used, but the figures demonstrate the economic and social significance and the extent to which governments were called to rescue financial institutions with State funds.

92

Wojcik

CML Rev. 2016

There are reasons why governments “bailed out”. Banks nowadays are increasingly complex, both in their corporate and in their financial structure. They are necessary to provide credit to the “real economy” and they are more and more interconnected. If they fail, the impact on the financial system and on the real economy is hard to predict or to control. There is a risk of further destabilization if depositors lose confidence in a bank and if deposit guarantee schemes are not sufficiently funded or their coverage not high enough to absorb the shock of a failing bank. Bank runs can be the result. Since depositors are voters and tend to assign responsiblity to politicians if they cannot withdraw their deposits from a bank, this puts pressure on decision-makers to act quickly and conclusively. Finally, as the Lehman bankruptcy of September 2008 illustrated,3 general corporate insolvency proceedings, even where they might allow the restructuring of the bank, are in most cases not appropriate to deal with unsound or failing banks.4 Such proceedings are usually not speedy enough; they usually cannot ensure the continuation of the critical functions of a financial institution; they lead to the destruction of value of assets and may thus impact on the real economy. In short, their application to a credit institution will most often not sufficiently preserve financial stability.5 However, when governments were looking for alternatives to ordinary insolvency proceedings or to bail-out, they realized that in most cases neither at national nor at EU level did a special framework for the recovery and resolution of banks exist which would have avoided the disadvantages of ordinary insolvency proceedings.6 Hence, they took the route of bail-out. That 3. Lehman was not bailed out; the complexity and knock-on effects of its bankruptcy made visible the risks such failures entail for the financial system. 4. See Huertas, “The case for bail-ins” in Dombret and Kenadjian (Eds.), The Bank Recovery and Resolution Directive (Institute for Law and Finance Series, 2013), pp. 167–188, at 167–168: “… bankruptcy for a bank is tantamount to liquidation”. 5. See Sommer, “Why bail-in? And how?”, 20 Economic Policy Review (2014), 207–228, at 208 et seq. Available at . For an analysis, see Binder, “Komplexitätsbewältigung durch Verwaltungsverfahren?: Krisenbewältigung und Krisenprävention nach der EU-Bankensanierungs- und –abwicklungsrichtlinie”, 179 ZHR (2015), 83–133, at 84–85. 6. The only legislative act at EU level, Directive 2001/24/EC of 4 April 2001 on the reorganization and winding up of credit institutions, O.J. 2001, L 125/15, does not contain substantive rules or powers to deal with insolvent credit institutions. According to Case C-85/12, LBI hf v. Kepler Capital Markets SA and Frédéric Giraux, ECLI:EU:C:2013:697, para 22, “Directive 2001/24 seeks to establish mutual recognition by the Member States of the measures taken by each of them to restore to viability the credit institutions which it has authorized. That objective, and that of guaranteeing equal treatment of creditors, laid down in recital 16 to that directive, require that the reorganization and winding-up measures taken by the authorities of the home Member State have, in all the other Member States, the effects which the law of the home Member State confers on them”. Directive 2001/24 presupposes national

Bail-in

93

decision was not only expensive, creating difficult political issues and immense social costs. Bail-outs also highlighted the increasing divergence between Member States within the euro area in their ability to support failing banks domiciled in their respective territories. They were conducive to the creation of an unlevel playing field for credit institutions within the EU internal market for financial services and at the same time exacerbated the sovereign debt crisis, thus threatening the very existence of the single European currency, the euro.7 Ultimately, the bail-out of banks by governments illustrated the “too big to fail/too complex to fail” paradigm which, by the implicit State guarantee on the operations of a banks, gave such institutions an advantage over any other corporation, creating competition distortions and moral hazard.8 The EU’s response to the financial crisis resulted in sweeping and game changing regulatory reform and a complete overhaul of the EU financial system.9 Its main achievement was the creation of the “Banking Union”, primarily geared to the euro area, while remaining open to non-euro area Member States to participate.10 In the current shape of the Banking Union the measures, but does not create them. During the crisis and before the adoption of the BRRD some Member States introduced new legislation, e.g. Germany, adopting the Gesetz zur Restrukturierung und geordneten Abwicklung von Kreditinstituten, zur Errichtung eines Restrukturierungsfonds für Kreditinstitute und zur Verlängerung der Verjährungsfrist der aktienrechtlichen Organhaftung (Restrukturierungsgesetz), Bundesgesetzblatt I, No. 63, 14 Dec. 2010, at 1900–1932. 7. Regulation (EU) 806/2014 of 15 July 2014 establishing uniform rules and a uniform procedure for the resolution of credit institutions and certain investment firms in the framework of a Single Resolution Mechanism and a Single Resolution Fund and amending Regulation (EU) 1093/2010 (SRM Regulation), O.J. 2014, L 225/1, Recitals 1 et seq.; Moloney, “European Banking Union: Assessing its risks and resilience”, 51 CML Rev. (2014), 1609–1670, at 1622 et seq. 8. Adolff and Eschweg, “Lastenverteilung bei der Finanzmarktstabilisierung”, 177 ZHR (2013), 902–966, at 902–905; Huertas, op. cit. supra note 4, at 167. For an attempt to quantify the implicit State guarantee studies, see e.g. Ueda and Weder di Mauro, “Quantifying structural subsidy values for systemically important financial institutions”, 37 Journal of Banking & Finance (2013), 3830–3841. 9. Moloney, op. cit. supra note 7, at 1610. 10. On this term and concept, see Commission Communication, A blueprint for a deep and genuine economic and monetary union Launching a European Debate, COM(2012)777 final, at 3.3.1. See also Binder, “The European Banking Union: Rationale and key policy issues” in Binder and Gortsos (Eds.), Banking Union: A Compendium (Nomos, forthcoming 2015), available at ; Lastra, International Financial and Monetary Law, 2nd ed. (OUP, 2015), Section 10. The full Banking Union includes a European Deposit Insurance Scheme (EDIS); see Five Presidents’ Report, “Completing Europe’s Economic and Monetary Union”, 21 Oct. 2015, available at , at 11. To complete the European Banking Union, the Commission adopted a legislative proposal on 24 Nov. 2015, amending Regulation (EU) 806/2014 in order to establish a European Deposit Insurance Scheme, COM(2015)586 final,

94

Wojcik

CML Rev. 2016

supervision of the euro area’s credit institutions is centralized at the level of the European Central Bank within a Single Supervisory Mechanism (SSM),11 and the resolution of euro area credit institutions and some other financial institutions is centralized at the level of the Brussels based EU agency, the Single Resolution Board (SRB) within a Single Resolution Mechanism (SRM).12 The SSM aims at ensuring the implementation of the so-called “Single Rulebook” in the area of bank supervision. This contains, in different legal acts, based on Regulation 575/2013 (“CRR”) and Directive 2013/36(“CRD IV”),13 the substantive rules on the supervision of banks for all EU Member States in the internal market. Likewise, the SRM aims at ensuring the implementation of the newly adopted EU-wide rules on the recovery and

which will form the third pillar of the Banking Union, once adopted. More sceptical, see Gordon and Ringe, “Bank resolution in the European Banking Union: A Transatlantic perspective on what it would take”, 115 Columbia Law Review (2015), 1297–1369, at 1309. 11. Regulation (EU) 1024/2013 of 15 Oct. 2013 conferring specific tasks on the European Central Bank concerning policies relating to the prudential supervision of credit institutions (SSM Regulation), O.J. 2013, L 287/63. On the SSM in general, see Babis and Ferran, “The European Single Supervisory Mechanism”, University of Cambridge Legal Studies Research Paper No. 10/2013 (March, 2014), available at ; Ceyssens, “Teufelskreis zwischen Banken und Staatsfinanzen: Der neue Europäische Bankaufsichtsmechanismus”, 51 NJW (2013), 3704–3708.; Lastra, “Banking Union and single market: Conflict or companionship?”, 36 Fordham Int. L.J. (2013), 1189–1222, at 1201 et seq.; Moloney, op. cit. supra note 7, at 1630 et seq.; Ruthig, “Die EZB in der europäischen Bankenunion”, 178 ZHR (2014), 443–485, at 443 et seq.; Tröger, “The Single Supervisory Mechanism: Panacea or quack Banking Regulation? Preliminary assessment of the new regime for the prudential supervision of banks with ECB involvement”, 15 European Business Organization Law Review (2014), 449–497; Wojcik, “Sachlicher Anwendungsbereich der Kapital- und Zahlungsverkehrsfreiheit” in Von der Groeben, Schwarze and Hatje (Eds.), Europäisches Unionsrecht (Nomos, 2015), Art. 63 AEUV, at para 96. Specifically on accountability within the SSM, see Ter Kuile, Wissink and Bovenschen, “Tailor-made accountability within the Single Supervisory Mechanism”, 52 CML Rev. (2015), 155–189, at 155 et seq. 12. SRM Regulation, supra note 7; on the SRM (section 2.2. infra) see Louis, “La difficile naissance du mécanisme européen de résolution des banques”, 50 CDE (2014), 7–21; Moloney, op. cit. supra note 7, at 1638 et seq.; Wojcik and Ceyssens, “Der einheitliche EU-Bankenabwicklungsmechanismus: Vollendung der Bankenunion, Schutz des Steuerzahlers”, 23 EuZW (2014), 893–898, at 803 et seq.; Zavvos and Kaltsouni, “The Single Resolution Mechanism in the European Banking Union”, in Haentjens and Wessels (Eds.), Research Handbook On Crisis Management in the Banking Sector (Edward Elgar, 2015), 117–149. 13. Regulation (EU) 575/2013 of 26 June 2013 on prudential requirements for credit institutions and investment firms and amending Regulation (EU) 648/2012, O.J. 2013, L 176/1; Directive 2013/36/EU of 26 June 2013 on access to the activity of credit institutions and the prudential supervision of credit institutions and investment firms, amending Directive 2002/87/EC and repealing Directives 2006/48/EC and 2006/49/EC, O.J. 2013, L 176/338.

Bail-in

95

resolution of credit institutions, based on the Bank Recovery and Resolution Directive (“BRRD”), for all EU Member States in the internal market.14 In this regulatory overhaul, the EU legislature laid down the legal foundations for a new type of regulatory measure which regulators in the EU will be able to apply to failing credit institutions.15 “Bail-in” – terminologically16 opposed to “bail-out” – seems an innocent expression used by the EU legislature or, at worst, another bit of jargon so common in the world of financial services. On its substance, however, bail-in is a far-reaching instrument which describes a new power of public authorities to decide to impose losses of failing banks on shareholders and creditors through write down or conversion, carried out in an administrative procedure different from normal insolvency proceedings. It is time that even before the introduction of bail-in through the BRRD, banks and their creditors could already agree between themselves on contractual clauses whereby, upon the occurrence of agreed trigger events, a liability would be converted into equity.17 The novelty of the BRRD, however, is the introduction of a statutory bail-in mechanism in EU law exercised by public authorities. That marks a fundamental change in the regulatory approach to deal with failing banks. It is the innovative centerpiece of the relevant EU legislation.

14. Directive 2014/59/EU of the European Parliament and of the Council of 15 May 2014 establishing a framework for the recovery and resolution of credit institutions and investment firms … (BRRD), O.J. 2014, L 13/190. 15. Although bail-in may be applied to entities other than credit institutions, this article will use exclusively the terms “credit institution” and “bank” when referring to entities to which a bail-in is or can be applied. 16. Sommer, op. cit. supra note 5, at 207, note 3 claims that D. Wilson Ervin, a banker at Credit Suisse, invented the concept. 17. See e.g. the so-called “CoCos” (contingent convertibles) or the legal requirements to qualify as Additional Tier 1 capital instruments. Capital instruments may qualify as Additional Tier 1 instruments according to Arts. 52 and 54 CRR if they include the definition of a trigger event at the occurrence of which write down or conversion may happen. CoCos allow in their contractual terms the conversion into equity with trigger events which do not refer to regulatory interventions; see Gleeson, “Legal aspects of bank bail-ins”, LSE Financial Markets Group Series, Special Paper 205, Jan. 2012, at 15; Kenadjian, “CoCos and bail-in” in Dombret and Kenadjian, op. cit. supra note 4, pp. 229–257, at 238; Rudolph and Zech, “Einsatzmöglichkeiten und Verlustabsorptionsfähigkeit von CoCos”, 68 Zeitschrift für das gesamte Kreditwesen (2015), 584–589.

96 2.

Wojcik

CML Rev. 2016

The newly created legal framework for bank recovery and resolution in the EU

The BRRD and the SRM Regulation, the delegated and implementing acts based on them, together with Directive 2001/2418 form the essential legal framework for the recovery and resolution of credit institutions and investment firms and certain other financial institutions.19 The adoption of this legal framework, and in particular the adoption of the BRRD, was heavily influenced by preparatory work at international level which, in the immediate aftermath of the Lehman failure, aimed at further convergence of resolution techniques. The Report and Recommendations of the Cross-border Bank Resolution Group20 of the Basel Committee of Banking Supervisors of 2010 and the Key Attributes of Effective Resolution Regimes for Financial Institutions,21 authored by the Financial Stability Board (FSB) in 2011, already set out the main principles and core elements which are considered to be essential for an effective resolution regime. Among these “core elements”, the FSB Report expressly referred to bail-in as a means to allocate losses to shareholders and creditors in a way that respects the hierarchy of claims and thus laid the ground for the future adoption of rules on bail-in in the EU.22 2.1.

The main features of the BRRD

BRRD’s main goal is to introduce and harmonize at EU level rules on recovery and resolution of credit institutions, investment firms and certain other financial institutions.23 Structurally, these rules deal with three main topics: the development of procedures for resolving credit institutions, the definition and conferral of specific resolution tools and powers which national resolution authorities will be able to apply, and the creation of national resolution financing arrangements. Member States to which the

18. Cited supra note 5. 19. For a description of the legislative antecedents of the BRRD and the SRM, see e.g. Gordon and Ringe, op. cit. supra note 10, at 1339 et seq. 20. Available at . 21. Available at , on which an update was published on 15 Oct. 2014, available at . 22. , at 11. 23. For ease of reference, the article will refer to banks only.

Bail-in

97

BRRD is addressed were obliged to transpose these rules into national law by 31 December 2014.24 2.1.1. Moving towards resolution The BRRD distinguishes three distinct stages: a preparatory phase, a phase of early intervention, and resolution. This distinction is made in line with the chronological order of events of a bank moving towards resolution (though not necessarily reaching resolution). The different stages are characterized by different powers conferred on public authorities vis-à-vis credit institutions, escalating in terms of extent and intensity the closer a bank moves towards resolution, and culminating in a situation of declared resolution. The preparatory phase (Arts. 4–18 BRRD) is essentially a “planning phase” during which recovery and resolution plans for each bank are set up in order to prevent resolution. While recovery plans are drawn up at least annually by each bank setting out the measures to be taken by the bank to restore its financial position in case of a significant deterioration of the latter and submitted to the bank’s relevant supervisor25 for a review, resolution plans are set up by resolution authorities for each bank and contain those actions which the resolution authority may take when a bank enters the resolution phase.26 The early intervention phase is characterized by a bank infringing, or being likely to infringe certain requirements of CRR, CRD IV or Regulation 600/2014. The BRRD assigns banking supervisors (and not resolution authorities!) a further set of intensified powers vis-à-vis such a bank in the

24. Not all EU Member States have fully transposed the BRRD at the time of writing. 25. This includes the ECB with regard to the specific tasks conferred on it by the SSM Regulation. 26. In setting resolution plans, resolution authorities enjoy far-reaching powers vis-à-vis the respective bank if there are substantive impediments to the resolvability of a bank, i.e. to the feasibility and credibility to either liquidate the bank or to resolve it with resolution tools. Such powers include the possibility to require the bank to divest specific assets and even to require changes to legal or operational structures of the bank so as to reduce its complexity. This may cause tensions between the supervisor and the resolution authority since the resolution authorities’ powers overlap to some extent with those of supervisors. Likewise, it will be interesting to observe to what extent resolution authorities will make use of these powers which possess the potential to tackle the “too complex to fail” problem in the current absence of a legislative measure on bank structural reform in the follow-up of the “Liikanen Report”, i.e. Commission, High-level Expert Group on reforming the structure of the EU banking sector: Final report, Brussels, Oct. 2012, available at . As a follow-up to the Liikanen Report, the Commission adopted on 29 Jan. 2014 a Proposal for a regulation of the European Parliament and the Council on structural measures improving the resilience of EU credit institutions, COM(2014)43 final.

98

Wojcik

CML Rev. 2016

early intervention phase, such as the power to require the management of the bank to implement arrangements foreseen in the recovery plan.27 Rules on the resolution phase form the heart of the BRRD. They set out under which conditions a bank may be considered under resolution, and they further detail the different resolution powers and tools that resolution authorities have at their disposal to deal with a bank under resolution. The exercise of such resolution powers needs to conform to resolution objectives and general principles set out in Articles 31 and 34 BRRD.28 Resolution powers can only be exercised where a bank finds itself in resolution. Resolution is not a state which any failing bank enters automatically. Instead, a resolution authority decides whether in respect of a bank the three resolution criteria set out in Article 32 BRRD are met. The first condition, in determining whether the bank is failing or likely to fail, relates to the financial situation of the bank. A bank is failing or likely to fail where, due to its financial situation, the supervisor would be entitled to withdraw its authorization. Moreover, a bank is failing or likely to fail where the bank is over-indebted, illiquid or where it needs (in principle) 27. See Art. 27 BRRD. 28. The resolution objectives stipulated in Art. 31(2) BRRD are: – to ensure the continuity of critical functions; – to avoid a significant adverse effect on the financial system, in particular by preventing contagion, including to market infrastructures, and by maintaining market discipline; – to protect public funds by minimizing reliance on extraordinary public financial support; – to protect depositors covered by Directive 2014/49/EU and investors covered by Directive 97/9/EC; – to protect client funds and client assets. When pursuing these objectives, the resolution authority must seek to minimize the cost of resolution and avoid destruction of value unless necessary to achieve the resolution objectives. Resolution principles, according to Art. 34 BRRD, are the following: – the shareholders of the bank under resolution bear first losses; – creditors of the bank under resolution bear losses after the shareholders in accordance with the order of priority of their claims under normal insolvency proceedings; – management body and senior management of the bank under resolution are replaced, except in those cases when the retention of the management body and senior management, in whole or in part, as appropriate to the circumstances, is considered to be necessary for the achievement of the resolution objectives; – management body and senior management of the bank under resolution shall provide all necessary assistance for the achievement of the resolution objectives; – natural and legal persons are made liable, subject to Member State law, under civil or criminal law for their responsibility for the failure of the bank; – creditors of the same class are treated in an equitable manner; – no creditor shall incur greater losses than would have been incurred if the bank had been wound up under normal insolvency proceedings; – covered deposits are fully protected; and – resolution action is taken in accordance with the safeguards set out in the BRRD.

Bail-in

99

extraordinary public financial support.29 The determination of whether a bank is failing or likely to fail is normally made by the banking supervisor having consulted the resolution authority. Member States have the option also to empower the resolution authority to make such a determination. The second condition asks whether the failure of the bank could be prevented within a reasonable timeframe by alternative private sector measures, including powers available in the early intervention phase, but also the write down or conversion of Additional Tier 1 and Tier 2 capital instruments by resolution authorities according to Article 59 BRRD et seq. The third condition requires the resolution authority to establish whether a resolution action is necessary in the public interest. To do so, the resolution authority needs to weigh whether, essentially, the application of resolution actions would better achieve the resolution objectives than the winding down of the bank under normal insolvency proceedings. It is obvious that such a comparison of, at the stage of decision, hypothetical future events involves a significant degree of discretion. This third criterion illustrates that “resolution” is an “alternative to normal insolvency proceedings”.30 Once resolution is determined, the resolution authority has at its disposal a range of resolution tools, including bail-in.31 In addition, resolution authorities may use write down and conversion powers with regard to certain regulatory capital instruments32 which are not “resolution tools” within the meaning of Article 37(3) BRRD,33 but which – as will be seen – share similar features. In order to implement the resolution tools, the BRRD grants additional resolution powers to national resolution authorities.34 These resolution powers are set out in a very detailed manner in Articles 63 BRRD et seq. They include for instance the power to take control of the failed bank and exercise all rights of shareholders or management and the power to transfer rights, assets or liabilities of that bank to another entity. 2.1.2. Financing Arrangements Next to the BRRD’s part on resolution tools and procedures, probably the most significant innovation in terms of policy is the obligation for Member States to establish national financing arrangements which together form a 29. See Gardella, “Bail-in and the financing of resolution within the SRM framework” in Busch and Ferrarini (Eds.), European Banking Union (OUP, 2015), pp. 373–407. 30. Binder, op. cit. supra note 5, at 93. 31. The sale of business tool, the bridge bank tool, the asset separation tool, the bail-in tool; Art. 37(3) BRRD. 32. Arts. 37(2) and 59–62 BRRD. 33. I.e. Additional Tier 1 and Tier 2 capital instruments; see Art. 51 et seq. and Art. 61 et seq. CRR. 34. See Arts. 63 et seq. BRRD.

100

Wojcik

CML Rev. 2016

European System of Financing Arrangements.35 The objective of these financing arrangements, which will typically be organized as funds,36 is to ensure the effective application of the resolution tools and powers by the resolution authorities. To avoid an abuse of the financing arrangement’s financial means, Article 101 BRRD contains an exhaustive list of authorized uses of the financing arrangement. Though financing arrangements are publicly managed, the financial means originate exclusively from contributions paid by banks.37 The creation of financing arrangements thus is another element of replacing taxpayer’s money with money from the banking sector when dealing with failing banks. To make this in a credible way, each national financing arrangement needs to attain a target level of at least 1 percent of covered deposits of all banks authorized in the relevant territory. Whether this is enough needs to be seen, in particular also because business models of banks vary from one Member State to another. In some States, banks rely more on deposits to finance themselves, with the consequence that the financing arrangement in such a Member State will be quite big compared to the overall size of the banking sector in that State. By contrast, in some Member States, banks rely more on other types of financing than deposits so that the question will arise whether the financing arrangement in such a State will be sufficient. By providing for mutual borrowing between financing arrangements38 and mutualization in case of cross-border resolution,39 the BRRD tries to tackle this problem within the European System of Financing Arrangements. 2.2.

The main features of the SRM Regulation

The SRM Regulation, which forms the second pillar of the Banking Union and applies since 1 January 2016, responds to the creation of the integrated supervision of credit institution within the SSM, aligning at European level the resolution framework for banks with that on the supervision of banks. It centralizes, for the euro area Member States, the decision-making powers in the field of resolution at the level of a newly created Single Resolution Board (SRB) and pools the national resolution funds in one newly created Single Resolution Fund (SRF). Since the BRRD only provides for minimum harmonization, leaving a considerable extent of discretion to national 35. Arts. 99–100 BRRD. 36. Exceptions are possible under Art. 100(6) BRRD. 37. See Commission Delegated Regulation (EU) 2015/63 of 21 Oct. 2014 supplementing Directive 2014/59/EU of the European Parliament and of the Council with regard to ex ante contributions to resolution financing arrangements, O.J. 2014, L 11/44. 38. Art. 106 BRRD. 39. Art. 107 BRRD.

Bail-in

101

resolution authorities in the application of the rules and the national financing arrangements, the centralization and uniformization of the resolution procedures and rules by the SRM Regulation further strengthens the resolution regime in the Banking Union. 2.2.1. Scope of application and decision-making The scope of application of the SRM Regulation is closely aligned to that of the SSM Regulation and includes all banks40 established in participating Member States.41 That means that the uniform rules of the SRM Regulation do not apply to banks established outside the euro area, for instance to banks established in the UK. To those banks only the BRRD applies. The SRB, an independent EU agency based in Brussels, was created by the SRM Regulation as the central decision-making body within the SRM. The internal decision-making happens in either an executive or a plenary session, reflecting a compromise between a need for swift decision-making and stronger involvement of national representatives due to fiscal implications.42 2.2.2. Uniformization of resolution rules and of decision-making The SRM Regulation establishes uniform rules for the resolution of banks taking over the material rules of the BRRD on bank recovery and resolution and transforming them into directly applicable provisions. This was necessary not only to reduce Member States’ discretion in transposing the BRRD and the resolution authorities’ discretion in applying the national rules transposing the BRRD, but also to provide for rules which the SRB as an EU agency can apply directly, while respecting the Meroni case law of the ECJ.43 At the same time, the SRM Regulation creates a uniform procedure for determining the resolution of a bank by the SRB.44 In principle, the three material conditions for deciding on the resolution of a bank are identical to 40. As well as parent undertakings and holding companies, investment firms and financial institutions subject to consolidated supervisions carried out by the ECB according to the SSM Regulation. 41. Participating Member States are Member States whose currency is the euro or which has established a close cooperation in accordance with Art. 7 SSM Regulation. 42. Moloney, op. cit. supra note 7, at 1638. The SRB is composed of a Chair, four full-time members and a member appointed by each participating Member State, with permanent observers of the Commission and the ECB. For a more extensive discussion of the internal decision-making, see Wojcik and Ceyssens, op. cit. supra note 12, at 894. 43. Case C-10/56, Meroni v. High Authority, EU:C:1958:8, which was recently confirmed in Case C-270/12, United Kingdom v. European Parliament and Council, EU:C:2014:18, para 41. This case law essentially prohibits a delegation of a “discretionary power implying a wide margin of discretion which may, according to the use which is made of it, make possible the execution of actual economic policy”. 44. Art. 18 SRM Regulation.

102

Wojcik

CML Rev. 2016

those laid down in Article 32 BRRD.45 However, due to the important fiscal implications of resolution decisions, and legal constraints stemming from the Meroni doctrine, the SRM Regulation prescribes a sophisticated decision-making sequence, in which the SRB adopts a resolution scheme placing the bank under resolution, determining the resolution tools and, where necessary, the use of the SRF. This resolution scheme may however only enter into force where neither Commission nor Council have expressed objections to it within 24 hours, thus ensuring the involvement of EU institutions in the decision-making and still allowing for a speedy adoption process “over a weekend”.46 Though the SRB is responsible for the effective and consistent functioning of the SRM as a whole, the SRM as a mechanism requires a high degree of collaboration and sharing of tasks between the SRB and national resolution authorities. Firstly, the SRB is directly responsible for drawing up resolution plans and adopting resolution decisions only with regard to all cross-border groups and those (approximately 130) significant banks which are directly supervised by the ECB.47 In this respect, the SRM Regulation mirrors the scope of the SSM Regulation. Secondly, the SRB adopts resolution measures with regard to the banks falling within its direct responsibility. It is to be noted that the individual resolution tool, such as a bail-in, is implemented not by the SRB, but by the relevant national resolution authorities which act vis-à-vis a bank and its shareholders and creditors. 2.2.3. Single Resolution Fund Another central piece of the SRM Regulation is the creation of the SRF, owned and administered by the SRB. Its target level amounts to at least 1 percent of the covered deposits of all banks authorized in the participating Member States, i.e. approximately ¤ 55 billion, to be reached over 8 years and based on ex ante contributions by banks. If the ex ante contributions are not sufficient to cover the expenses incurred by the SRF, extraordinary ex post contributions may be raised from banks to cover these additional amounts. The use of the SRF is limited to the same purposes as the use of the financing arrangements under the BRRD. Due to its volume, the SRF can play a much more important role in effectively shielding taxpayers from financing bank

45. Note that in the SRM the condition whether a bank is failing or likely to fail is in principle to be determined by the ECB after consulting the SRB. 46. Moloney, op. cit. supra note 7, at 1639 et seq. 47. Art. 7(2) SRM Regulation.

Bail-in

103

crises and will thus contribute to the goal of the Banking Union to break the link between banks and sovereigns.48 Intergovernmental agreement on the transfer and mutualization of contributions to the single resolution fund The use of the SRF is however dependent on the entry into force of the “Intergovernmental Agreement on the transfer and mutualization of contributions to the Single Resolution Fund” (IGA) concluded on 21 May 2014 between 26 Member States (with the exception of the UK and Sweden) outside the EU law framework.49 Among others, this agreement, which was highly contentious between Council and European Parliament,50 deals with the transfer of contributions raised at national level in accordance with the BRRD and the SRM Regulation to national compartments within the SRF which will cease to exist after a period of 8 years, after which a full mutualization of the contributions will be achieved.51 As regards bail-in, the IGA contains two interesting rules: its Article 9 makes the existence of bail-in rules as laid down in the SRM Regulation and in the BRRD at the time of the conclusion of the IGA a prerequisite for the use of the SRF. This is completed by a declaration of intent by the contracting parties and observers of the intergovernmental conference. According to that declaration the principles and rules related to the bail-in tool are not repealed or amended in a way that is not equivalent and does not lead to, at least, the same and not less stringent result than that deriving from the SRM Regulation 2.2.4.

48. The size of the SRF is often criticized as having an insufficient target size, see e.g. Gordon and Ringe, op. cit. supra note 10, at 1354–1358; more optimistic, see Hadjiemmanuil, “Bank resolution financing in the Banking Union”, LSE Legal Studies Working Paper No. 6/2015 (March, 2015), available at , at 37; Huertas and Nieto, “How much is enough? The case of the Resolution Fund in Europe”, available at , 18 March 2014; Zavvos and Kaltsouni, op. cit. supra note 12, at 117 et seq. Member States therefore agreed to put in place a bridge financing arrangement and a common public backstop for the SRF, neither of which are yet in place. 49. Participating Member States representing not less than 90% of the aggregate of the weighted votes have ratified the IGA and deposited the ratification instrument by 30 Nov. 2015. 50. Some Member States strongly advocated the conclusion of an IGA agreement outside the scope of EU law. To shed some light on the underlying legal reasoning, see Calliess and Schoenfleisch, “Die Bankenunion, der ESM und die Rekapitalisierung von Banken”, 70 JZ (2015), 113–121, at 119 et seq.; see also Keppenne, “Institutional Report”, in Neergaard, Jacqueson and Danielsen (Eds.), The Economic and Monetary Union: Constitutional and Institutional Aspects of the Economic Governance within the EU, The XXVI FIDE Congress (Copenhagen, 2014), 179–257, at 230. 51. For a detailed analysis of the IGA, see Zavvos and Kaltsouni, op. cit. supra note 12, at 117 et seq.

104

Wojcik

CML Rev. 2016

as on the date of its initial adoption.52 All this is rather curious. It means that EU Member States pre-establish outside the framework of the EU Treaties and in a formal way their position and action within the Council of Ministers in case of a legislative proposal intending to amend the BRRD or SRM Regulation with regard to the bail-in rules. Furthermore, on the substance, the IGA introduces de facto the grandfathering of the bail-in provisions included in the BRRD and the SRM Regulation. 2.3.

The requirement of “burden-sharing” under State Aid rules: The Commission’s 2013 Banking Communication as a specific element of the EU legal framework on recovery and resolution of banks

According to Article 107 TFEU, any State aid is incompatible with the internal market, unless it qualifies as one of the narrow exceptions set out in Article 107(2) TFEU or unless it has been approved by the Commission for one of the reasons set out in Article 107(3) TFEU. The use of State money by EU governments to bail out numerous banks since the beginning of the financial crisis was considered State aid under Article 107 TFEU. The Commission was closely involved in the necessary bank rescue operations due to the obligation of the Member States to notify new aid before its implementation and the Commission’s exclusive competence to authorize the grant of State aid.53 Since bailouts of banks would not fall under any of the exceptions in Article 107(2) TFEU, every bailout had to be notified to and examined by the Commission to see whether it could be approved as being compatible with the Internal Market. When scrutinizing those bailouts, the main conceivable justification of this type of State aid could reside in Article 107(3)(b) TFEU which allows the use of “aid to . . . remedy a serious disturbance in the economy of a Member State”. The Commission’s approval practice thus principally focused on Article 107(3)(b) TFEU during the financial crisis. To give detailed guidance on the criteria for the compatibility of State aid with the Internal Market under Article 107(3)(b) TFEU during the financial crisis, the Commission initially adopted six communications (the so-called “Crisis Communications”), spelling out the conditions for access to State aid 52. Declaration No. 1 IGA. 53. Art. 108 TFEU. On the Commission’s State aid practice in the financial sector in the course of the crisis, see Angeloni and Lenihan, “Competition and State aid rules in the time of Banking Union” in Faia, Hackethal, Haliassos and Langenbucher (Eds.), Financial Regulation: A Transatlantic Perspective (CUP, 2015), pp. 89–124; Gerard, “Managing the financial crisis in Europe: The role of EU State aid law enforcement”, in Derenne, Merola and Rivas (Eds.), Competition Law at Times of Economic Crisis: In Need for Adjustment? (Bruylant, 2013), pp. 231–252, available at .

Bail-in

105

and the requirements for finding such aid compatible with the Internal Market in view of State aid principles set out in the TFEU.54 In the light of the regulatory developments regarding the creation of the Banking Union and the proposal for the BRRD, the 2013 Banking Communication55 added to this framework of Crisis Communications a further set of guiding principles, binding the Commission in the application of the State aid rules and the exercise of discretion when deciding on the approval of State aid measures in favour of banks. In particular, the 2013 Banking Communication introduces the concept of “burden-sharing”.56 It requires that State aid measures are granted on terms which make existing investors, both equity and junior debt holders, share the burden of rescuing failing banks, provided such an imposed contribution respects fundamental rights (including the “no creditor worse off ” principle), and financial stability is not put at risk.57 That means that State aid will not be approved unless, as a minimum, losses have first been absorbed by equity and that junior debt holders have participated in the sharing of losses through the conversion of their debt instruments into capital or by way of a write down of these debt instruments.58 “Burden-sharing” can be regarded as functionally equivalent to bail-in under the BRRD/SRM rules. However, while the bail-in-rules under the

54. Communication from the Commission of 25 Oct. 2008 on the application of State aid rules to measures taken in relation to financial institutions in the context of the current global financial crisis (“2008 Banking Communication”), O.J. 2008, C 270/8, Communication from the Commission of 15 Jan. 2009 on the recapitalization of financial institutions in the current financial crisis: limitation of aid to the minimum necessary and safeguards against undue distortions of competition (“Recapitalization Communication”), O.J. 2009, C 10/2; Communication from the Commission of 26 March 2009 on the treatment of impaired assets in the Community financial sector (“Impaired Assets Communication”), O.J. 2009, C 72/1; Communication of the Commission of 19 Aug. 2009 on the return to viability and the assessment of restructuring measures in the financial sector in the current crisis under the State aid rules (“Restructuring Communication”), O.J. 2009, C 195/9; Communication from the Commission of 7 Dec. 2010 on the application, from 1 Jan. 2011, of State aid rules to support measures in favour of financial institutions in the context of the financial crisis (“2010 Prolongation Communication”), O.J. 2010, C 329/7; Communication from the Commission of 6 Dec. 2011 on the application, from 1 Jan. 2012, of State aid rules to support measures in favour of financial institutions in the context of the financial crisis (“2011 Prolongation Communication”), O.J. 2011, C 356/7. 55. Communication from the Commission of 30 July 2013 on the application, from 1 Aug. 2013, of State aid rules to support measures in favour of banks in the context of the financial crisis (“Banking Communication”), O.J. 2013, C 216/1. 56. Para 15 of the Banking Communication. 57. Paras. 19, 40 and 42 of the Banking Communication. 58. Para 41 of the Banking Communication.

106

Wojcik

CML Rev. 2016

BRRD/SRM are applicable as of 1 January 2016 at the latest,59 some Member States introduced or applied rules on bail-in equivalent “burden-sharing” already before the adoption of the BRRD60 in reaction to the 2013 Banking Communication if they wished to obtain a swift approval of a State aid measure from the Commission for a bail-out. This led in these Member States to some criticism and law suits by affected shareholders and debt holders regarding the validity of the 2013 Banking Communication.61 Since 1 January 2016 the relevance of the “burden-sharing” guidelines of the 2013 Banking Communication is vastly reduced. The main reason is that the full implementation of the Banking Union is supposed to substantially reduce the need for bail-outs and, hence, State aid approvals. In addition, the use of State aid will in most cases trigger resolution since the need for extraordinary public financial support means in principle that the bank is failing or likely to fail.62 3.

Bail-in according to BRRD and the SRM Regulation

Before examining the bail-in rules under the BRRD/SRM legal framework, some terminological explanations are necessary. This article employs the term “bail-in” as an umbrella term covering two distinct, but in many ways similar, powers under the BRRD/SRM legal framework: (i) bail-in as a “resolution tool” to be applied by the resolution authority where all the conditions for putting a bank into resolution are fulfilled63 and (ii) write down and conversion powers which legally are not a resolution tool and can be applied independently or in combination with a resolution tool.64 When referring to the former power, this article will use the terms “bail-in tool” or “resolution tool of bail-in”. When referring to the latter, this article will use the term “write down and conversion instrument”. This article will in the following base its analysis mainly on the provisions of the BRRD, making reference to the provisions of the SRM Regulation only where necessary. Within the Banking Union the BRRD remains relevant as the “single rulebook” in this area, and national resolution authorities will

59. Under the BRRD, Member States could decide to apply bail-in under the BRRD already as of 1 Jan. 2015. 60. The Banking Communication is applicable to aid measures notified after 1 Aug. 2013; see paras. 89 and 90 of the Banking Communication. 61. See e.g. the pending request for a preliminary ruling in Case C-526/14, Kotnik and Others. 62. See Art. 32(4)(d) BRRD. 63. Arts. 37(3), 43–55 BRRD and Art. 27 SRM Regulation. 64. Arts. 37(2), 59–62 BRRD and Art. 21 SRM Regulation.

Bail-in

107

implement resolution decisions by the SRB according to powers conferred on them under the national law transposing the BRRD. 3.1.

Functions of bail-in

Bail-in has in particular three interrelated functions within the resolution framework. (i) Bail-in restores the order of responsibility towards a bank, reducing the implicit State guarantee which banks enjoyed and reducing to a great extent the negative feedback loop between banks and sovereigns (“breaking-the-link-with-sovereigns function”). From this overarching function the next two functions flow as means to achieve the protection of public funds. (ii) Bail-in strengthens the responsibility of shareholders and creditors for their investments, being the corollary to the risks, which they have taken with an investment in the bank, and mirroring their treatment in a situation of a normal insolvency (“mirror function”). (iii) The resolution tool of bail-in restrains the use of the financial means pooled in the national financing arrangement/SRF (“gatekeeper function”), thus further contributing to the disentanglement of banks and sovereigns. 3.2.

The bail-in tool

3.2.1. Bail-in scenarios The bail-in tool may be applied in two distinct scenarios, where a bank has been declared to be “under resolution”: firstly, bail-in can be employed to recapitalize the bank.65 Secondly, the resolution tool of bail-in can be used in connection with and in support of the application of other resolution tools,66 namely to provide capital to a bridge bank67 and to complement the sale of business tool or the asset separation tool. In any of these scenarios the powers of the resolution authority are sweeping, “re-designing” the bank’s balance sheet. With regard to the bank’s equity holders, the resolution authority can decide that existing shares or other instruments of ownership are cancelled or are transferred to bailed-in creditors.68 With regard to debt holders, the 65. Art. 43(2)(a) BRRD and Art. 27(1)(a) SRM Regulation. 66. Mentioned in note 31 supra; Art. 43(2)(b)(i) BRRD and Art. 27(1)(b)(i) SRM Regulation. 67. Art. 43(2)(b)(ii) BRRD and Art. 27(1)(b)(ii) SRM Regulation. 68. Arts. 47(1)(a) and 63(1)(h) BRRD.

108

Wojcik

CML Rev. 2016

resolution authority may convert liabilities into shares or other instruments of ownership, thereby diluting existing shareholders.69 The principal amount of the bank’s liabilities may be reduced (written down), the liability’s maturity amended or its interest payments altered or suspended.70 Furthermore, debt instruments may be cancelled and derivatives closed out and terminated.71 3.2.2. Bail-inable financial instruments and liabilities Which financial instruments fall within the scope of the bail-in tool? That question is crucial. The scope of the bail-in tool determines its effectiveness to attain the resolution objectives and, at the same time, impacts very practically on existing and future investments in banks, both in economic and in legal terms. The wider the scope, the more equity and debt holders will contribute to rescuing a bank and the more effective the bail-in tool can probably be. On the other hand, a wide scope of bail-in may have important effects on the economy, depending on who the bailed-in creditors are (SMEs, pension funds, etc.).72 Unsurprisingly, it was one of the very contentious points during the legislative negotiations. While it is evident that equity is subject to bail-in, the BRRD stipulates that, in principle, all liabilities of a bank in resolution are subject to the bail-in tool, unless they are mandatorily or by decision excluded from it.73 Article 44(2) BRRD74 contains a list of liabilities which are mandatorily excluded from the bail-in tool. That list includes covered deposits75, secured liabilities, any liability arising from the holding by the bank of client assets or by virtue of a fiduciary relationship, liabilities to other banks, liabilities with a remaining maturity of less than 7 days towards securities settlement and clearing systems and liabilities towards employees (except variable remuneration components of material risk takers), towards commercial or trade creditors for critical operations of the bank, towards tax and security authorities and towards deposit guarantee schemes. The reason for their 69. Arts. 47(1)(b) and 63(1)(f) BRRD. 70. Art. 63(1)(e) and (j) BRRD. 71. Art. 63(1)(g) and (k) BRRD. 72. Awareness of possible contagion was raised during the bail-in exercised in 2013 in Cyprus, where the Troika required that losses be imposed on unsecured senior creditors and on uninsured depositors at Laiki Bank and Bank of Cyprus. 73. Art. 44(1) BRRD; eligible liabilities” are those not excluded from the scope of the bail-in tool. 74. Art. 27(3) SRM Regulation. 75. Pursuant to Art. 2(1), para 94 BRRD and Art. 2(1), para 5 of Directive 2014/49/EU of the European Parliament and of the Council of 16 April 2014 on Deposit Guarantee Schemes, “covered deposits” are such eligible deposits which do not exceed ¤100,000 per depositor per bank or, which exceed ¤100,000 if they stem from certain protected transactions and fulfil additional requirements set out in Art. 6 of Directive 2014/49/EU.

Bail-in

109

mandatory exclusion is that they are protected in normal insolvency proceedings or that practical reasons of continuity of business operations mandate their exclusion.76 According to Article 44(3) BRRD77 discretion is conferred on national resolution authorities to fully or partially exclude other liabilities.78 The exercise of this discretionary exclusion of liabilities from the scope of the bail-in tool is limited to exceptional circumstances and comes with many other strings attached.79 Where Article 44(3) BRRD is used, inevitably the amount of liabilities available for a bail-in will be reduced. Therefore, the resolution authority may decide to further increase the level of write down and conversion of eligible liabilities and to use contributions from the financing arrangements/SRF to cover resulting losses or purchase shares.80 It is against this background, that the resolution authority needs to duly consider that the exclusion may not disrupt the mirroring function of the bail-in, that a sufficient level of loss-absorbing capacity is maintained, and that sufficient resources from financing arrangements/SRF are available.81 Overall, these requirements reflect the need to address practical obstacles to the application of bail-in and to avoid undesired detrimental consequences of a strict application of the bail-in tool in certain cases. It thus embodies the result of a balancing exercise which intends to effectively safeguard the principle of proportionality.

76. Bliesener, “Legal problems of bail-ins under the EU’s proposed Recovery and Resolution Directive” in Dombret and Kenadjian, op. cit. supra note 4, pp. 189–228, at 199. 77. Art. 27(5) SRM Regulation. 78. That power was introduced only during the legislative negotiations. 79. These conditions are set out in Art. 44(3) BRRD, it is required e.g. that it is not possible to bail-in the respective liability within a reasonable time notwithstanding the good faith efforts of the resolution authority (a), that the exclusion is strictly necessary and is proportionate to achieve the continuity of critical functions and core business lines (b), that the exclusion is strictly necessary and proportionate to avoid giving rise to widespread contagion (c) and that the application of the bail-in tool to those liabilities would cause a destruction in value such that the losses borne by other creditors would be higher than if those liabilities were excluded from bail-in (d). They will be further specified in a Commission Delegated Act, based on Art. 44(11) BRRD. 80. Art. 44(4) BRRD and Art. 27(6) SRM Regulation. 81. Art. 44(9) BRRD and Art. 27(12) SRM Regulation. Another rationale for the possibility of discretionary exclusion is to alleviate, in a specific resolution situation, the fact that bail-in shifts the burden of loss from taxpayers to shareholders and creditors, among the latter being also individuals (directly or through financial institutions invested in the failing bank) and deposit holders, including SMEs, of deposits above the DGS (Deposit Guarantee Schemes) limit; see also Goodhart and Avgouleas, “A Critical Evaluation of Bail-in as a Bank Recapitalisation Mechanism”, in Allem, Carletti and Gray (Eds.), Bearing the Losses from Bank and Sovereign Default in the Eurozone (European University Institute & Wharton Financial Institutions Center, 2014), pp. 65–98, at 66.

110

Wojcik

CML Rev. 2016

Determination of the amount necessary for bail-in and sequence of bail-in Resolution authorities need to know to what extent the application of the bail-in tool is necessary. To this end, they have to determine the necessary amount for bail-in through a valuation of the bank’s assets and liabilities. The amount for bail-in consists first, where relevant, of an amount by which eligible liabilities must be written down or converted into shares to ensure that the net asset value of the bank is zero.82 By this operation, which is reducing the liabilities side of the bank’s balance sheet, a negative difference between the assets and the liabilities will be equalized. On that basis, secondly, an aggregate amount is calculated using the amount by which eligible liabilities must be converted into shares or other capital instruments in order to restore the Common Equity Tier 1 (CET1) ratio83 of the bank under resolution or the bridge bank to which they are conferred.84 This is necessary to ensure that the bank or the bridge bank will have, for at least one year, the necessary regulatory minimum capital in order to meet the conditions for (continuous or new) authorization and to sustain sufficient market confidence in the bank under resolution or the bridge bank.85 The assessment of these amounts is to be made on the basis of a fair, prudent and realistic valuation of the bank’s assets and liabilities to be carried out by a person independent from any public authority. The requirements which the valuation as a crucial and complex part of the application of the bail-in tool needs to meet are set out in Article 36 BRRD.86 Due to the urgency under which bank resolutions are in most cases carried out, it is often not possible to make a fully-fledged valuation. In such cases, Article 36 BRRD allows to make a provisional valuation to which specific, lighter requirements apply. Despite this urgency procedure, the BRRD requires to carry out as soon as practically possible an ex post definitive valuation which will finally comply with all valuation requirements and may result in later adjustments to the application of the resolution tools, including the bail-in tool. 3.2.3.

82. Art. 46(1)(a) BRRD and Art. 27(13)(a) SRM Regulation. 83. Each bank needs to have a CET 1 ratio of at least 4.5% against risk-weighted assets. 84. Art. 46(1)(b) BRRD and Art. 27(13)(b) SRM Regulation. 85. Art. 46(2) BRRD and Art. 27(13) subpara. 2 SRM Regulation. 86. Art. 20 SRM Regulation. These requirements are valid for resolution in general. In addition, Art. 49(4) BRRD contains special requirements regarding valuation of liabilities arising from derivatives. Liabilities arising from derivatives are fully bail-inable, unless excluded according to Art. 44(3) BRRD on a discretionary basis. On valuation, see Bliesener, op. cit. supra note 76, at 217 et seq.; Gardella, op. cit. supra note 29. Resolution authorities must first terminate and close out any derivative contract, before being able to write down or convert such liabilities (Art. 49(2) BRRD – see in more detail Binder, op. cit. supra note 5, at 112 et seq.

Bail-in

111

A simplified example may illustrate the quantitative calibrations and the subsequent sequence of events in order to achieve a zero net asset value and a sufficient regulatory capital basis for continuous authorization: A bank before resolution had a balance sheet with assets valued at 150. Its shareholders had 20 in equity. Its liabilities amounted in total to 130, of which 5 were Additional Tier 1 (AT1) capital instruments, 10 Tier 2 instruments, 10 in junior (i.e. subordinated) debt and 105 in other liabilities, 60 of which are not bail-inable. The bank enters resolution. A valuation according to Article 36 BRRD is carried out. It detects a loss of 50 due to non-performing loans. The necessary amount for bail-in is at least 54.5, since 50 are required to bring the net asset value of the bank to zero and further 4.5 are required to restore the CET1 ratio. The sequencing of the bail-in powers follows, in principle, a “reverse order of priority of claims” as laid down in the applicable national insolvency law.87 Article 48 BRRD establishes this “waterfall”: first, CET1 items will be reduced,88 i.e. all the shareholders will lose their 20. If this is not enough to cover the losses, subsequently, the principal amount of both AT1 and Tier 2 capital instruments will be written down to zero.89 Where this is still not sufficient to bring the bank to a net asset value of zero, remaining subordinated debt is written down.90 Then, all the remaining eligible liabilities will be written down or converted to attain the necessary amount for bail-in.91 When doing so, debt which included contractual terms explicitly allowing write down or conversion will be written down or converted first.92 As a consequence, in this example 5 of eligible liabilities are written down fully. In order to reach the CET1 ratio of 4.5 percent, 4.5 of the remaining eligible liabilities are converted according to a conversion rate representing appropriate compensation to the affected creditors. In any case, the pari passu principle of treating creditors of the same rank in the same way is to be respected during the sequencing of write down and conversion.93 3.3.

The write down and conversion instrument

As explained above, bail-in – as an umbrella term – includes (in addition to the bail-in tool) the write down and conversion instrument, laid down in Articles

87. 88. 89. 90. 91. 92. 93.

Art. 48 BRRD. See clearly Art. 27(15) and Art. 17 SRM Regulation. Art. 48(1)(a) and 60(1) BRRD. Art. 48(1)(b) and (c) BRRD. Art. 48(1)(d) BRRD. Art. 48(1)(e) BRRD. Art. 48(3) BRRD. Art. 48(2) BRRD.

112

Wojcik

CML Rev. 2016

37(2), 59–62 BRRD.94 The write down and conversion instrument is not a resolution tool.95 Nevertheless, many of its features and the basic principles are very similar to the bail-in tool. The two main differences are the following: the write down and conversion instrument only covers relevant capital instruments, namely AT1 and Tier 2 capital instruments.96 It does not include other eligible liabilities and is thus more limited in scope. The write down and conversion instrument can be applied in combination with a resolution tool where a bank is formally under resolution. But it can be also applied independently from resolution actions and even before the conditions of resolution are met. Actually, a bank should not be put “under resolution” where the application of the write down and conversion instrument would effectively restore the bank’s financial position so that it would not be considered as failing: the use of the write down and conversion instrument can therefore prevent the application of the bail-in tool. This aspect of priority of the write down and conversion instrument vis-à-vis the application of the bail-in tool is further emphasized by Articles 60(5), 59(3) BRRD. They require the resolution authority to apply the write down and conversion instrument to AT1 and Tier 2 capital instruments before the application of resolution tools, if, for instance, the resolution conditions are met or where the bank would no longer be viable.97 A bank is no longer viable when the relevant authority determines that, if AT1 and Tier 2 capital instruments were not written down or converted, the bank would be failing or likely to fail and no other action would be able to prevent the failure of the bank.98 Since the application of the write down and conversion instrument may actually prevent the application of the bail-in tool, AT1 and Tier 2 capital instruments play an important role of additional “buffers”,99 emphasizing the sequence of involvement of equity and debt holders in a crisis situation of a bank. Where the write down and conversion instrument is applied, the necessary amount to be written down or to be converted will need to be calculated on the basis of a valuation which complies with Article 36 BRRD.100 The sequencing follows a similar “cascade” to that applicable under the bail-in tool.

94. Art. 21 SRM Regulation. 95. See Art. 37 BRRD which does not mention the write down and conversion instrument in the exhaustive list of resolution tools. 96. Arts. 59(1), 2(1) para 74 and Art. 21(1) SRM Regulation. 97. See Art. 59(3)(a) and (b) BRRD and Art. 21(3)(a) and (b) SRM Regulation. 98. Recital 81 BRRD. 99. Binder, op. cit. supra note 5, at 108. 100. See Art. 59(10) BRRD.

Bail-in 4.

113

Minimum requirement for own funds and eligible liabilities

Bail-in can only attain its objectives where a bank has sufficient own funds and eligible liabilities which can be effectively bailed-in in a resolution situation. The requirements for banks to meet, at all times, a minimum of bail-inable own funds (i.e. regulatory capital consisting of CET1, AT1 and Tier 2)101 and eligible liabilities (MREL) are laid down in Article 45 BRRD.102 MREL serves at least two purposes: first, to ensure that bail-in can be effectively applied (even where liabilities are excluded from bail-in by law103 or by a decision104 of the resolution authority)105 and, second, to increase the predictability of the bail-in operation for the bank’s investors, thus enhancing the stability of the operation, incentivizing the right pricing of the investment and reducing the risk of contagion due to uncertainties in the market.106 MREL is, in short, a necessary corollary to make bail-in work. To this end, resolution authorities will determine for each bank an amount of own funds and eligible liabilities which will be expressed by a percentage figure. That determination rests on criteria107 which prescribe that the bank has sufficient eligible liabilities “to ensure that . . . losses could be absorbed and the Common Equity Tier 1 ratio of the bank could be restored at a level necessary to enable it to continue to comply with the conditions for authorization …”.108 They force the resolution authority to base itself on an individual assessment of the specific situation of each bank.109 MREL includes a number of conditions with regard to the quality of eligible liabilities

101. Art. 2(1), para 38 BRRD referring to the definition of “own funds” in Art. 4(1), para 118 CRR. 102. Art. 12 SRM Regulation. 103. Art. 44(2) BRRD and Art. 27(4) SRM Regulation. 104. Art. 44(3) BRRD and Art. 27(5) SRM Regulation. 105. See also Recital 79 of BRRD, stressing the need to avoid that banks structure their liabilities in a manner which impedes the effectiveness of the bail-in tool. 106. For the sake of completeness it is to be noted that, with similar goals, the BRRD requires banks to maintain at all times a sufficient amount of authorized CET1 capital instruments and/or to have been granted the necessary prior authorization to issue a relevant number of CET1 instruments; see Arts. 54 and 60(4) BRRD. 107. These criteria will be further specified in Draft Regulatory Technical Standard on criteria for determining the minimum requirement for own funds and eligible liabilities under Directive 2014/59/EU, EBA/RTS/2015/05, 5 July 2015, to be adopted by the European Commission. 108. Art. 45(6) BRRD and Art. 12(6) SRM Regulation. 109. E.g. the “size, business model, the funding model and the risk profile”; see Art. 45(6)(d) BRRD and Art. 12(7) SRM Regulation.

Wojcik

114

CML Rev. 2016

in order to create a pool of liabilities which will allow the effective application of the bail-in tool.110 In that context, two particularities should be highlighted: Article 45(13) BRRD allows banks to partially meet MREL through contractual bail-in instruments, i.e. instruments which contain a contractual term providing that, where a resolution authority decides to apply the bail-in tool to that bank, the instrument shall be written down or converted and is subject to a binding provision pursuant to which in the event of normal insolvency proceedings it ranks below other eligible liabilities and cannot be repaid until other eligible liabilities outstanding at the time have been settled. That is an important element, since it can reduce the risk of litigation based on an alleged breach of the right to property. Moreover, liabilities governed by the law of a third country shall not be counted towards meeting MREL, if it cannot be established that the bail-in decision of a European resolution authority would have automatic effect in the third country’s legal order.111 This latter requirement reflects the high demands aiming at ensuring the effective bail-in. MREL will need to be complied with by banks on an individual basis.112 In a group situation, it will need to be complied with on a consolidated basis as well as by each subsidiary.113 These requirements can be waived on an exceptional basis.114 Although MREL pursues valid and necessary goals, the determination of MREL raises several legal and practical issues, namely the relationship between resolution and supervisory authorities and the interaction with regulatory capital requirements under CRD IV/CRR,115 the coordination of 110. See Art. 45(4) BRRD: “Eligible liabilities shall be included in the amount of own funds and eligible liabilities referred to in paragraph 1 only if they satisfy the following conditions: the instrument is issued and fully paid up; the liability is not owed to, secured by or guaranteed by the institution itself; the purchase of the instrument was not funded directly or indirectly by the institution; (d) the liability has a remaining maturity of at least one year; (e) the liability does not arise from a derivative; (f) the liability does not arise from a deposit which benefits from preference in the national insolvency hierarchy in accordance with Art. 108. (a) (b) (c)

For the purpose of point (d) where a liability confers upon its owner a right to early reimbursement, the maturity of that liability shall be the first date where such a right arises”. 111. Art. 45(5) BRRD and Art. 12(17) SRM Regulation. 112. Art. 45(7) BRRD and Art. 12(9) SRM Regulation. 113. Art. 45(8)(10) BRRD and Art. 12(9) SRM Regulation. 114. Art. 45(11)(12) BRRD and Art. 12(10) SRM Regulation. 115. For an extensive discussion of the interaction between bail-in rules and regulatory capital requirements, see Joosen, “Bail in mechanisms in the Bank Recovery and Resolution

Bail-in

115

MREL in groups, and the interaction with the FSB’s recently adopted requirement for G-SIBs (i.e. Global Systemically Important Banks) on a “total loss absorbing capacity” (TLAC). While in the group context the BRRD provides for sophisticated mechanisms to reach coordinated and consensual joint decisions, involving, where necessary, EBA’s binding mediation powers,116 the setting of MREL by resolution authorities will, to the extent that own funds are concerned, impact unavoidably on the banking supervisor’s obligations in enforcing regulatory capital requirements under CRD IV/CRR.117 Being aware of this issue, BRRD and SRM Regulation try to solve it by obliging resolution authorities to consult the relevant banking supervisor, including the ECB within the remit of the SRB’s MREL setting powers.118 Whether mere consultation will be sufficient to bring about efficient coordination and will avoid potential conflicts seems however doubtful. The current solution rather creates a stronger role for the resolution authority which may ultimately set the MREL in contrast to the opinion of the banking supervisor. Similar problems will need to be solved if and when the EU incorporates the non-binding TLAC standards into EU law.119 In this context, it should be recalled that MREL actually precedes work at international level on adequate loss-absorbing capacities for banks. Loss-absorbing capacities for G-SIBs were already considered in the FSB’s “Too big to fail” Report to the G-20 of September 2013120 as well as in the FSB Guidance on Developing Effective Resolution Strategies of November 2013.121 In November 2015, the FSB finally published its “Principles on loss-absorbing and recapitalization capacity for G-SIBs in Resolution” together with a “Total Loss-absorbing

Directive”, Oct. 2014, available at , at 1 et seq.; Joosen, “Regulatory capital requirements and bail in mechanisms”, March 2015, available at . The EBA has published draft guidelines on the treatment of liabilities in bail-in for public consultation, available at . 116. See Art. 45(9) and (10) BRRD in connection with Art. 19 SRM Regulation. The problem of coordinated decision-making with regard to MREL (but also in general) will become less acute in the Banking Union context, at least with regard to purely Banking Union based banking groups, since to this extent the SRB will be the sole decision-making resolution authority. 117. Joosen, “Bail in mechanisms …”, op. cit. supra note 115, at 12. 118. Art. 45(6) BRRD and Art. 12(1) SRM Regulation. 119. In its Communication “Towards the completion of the Banking Union”, COM(2015)587 final, the Commission announced that it would bring forward a legislative proposal in 2016 so that TLAC can be implemented by 2019. 120. See , at 14. 121. See , at 7–8.

116

Wojcik

CML Rev. 2016

Capacity (TLAC) Term Sheet”.122 The FSB requires G-SIBs to hold a “Minimum TLAC” that should amount to at least 16 percent of the G-SIB resolution group’s risk-weighted assets (RWA) as of 1 January 2019 and at least 18 percent as of 1 January 2022. This amounts to roughly two times the Basel III regulatory capital requirements of 8 percent of RWA. The FSB adds that Minimum TLAC must be at least 6 percent of the Basel III leverage ratio denominator as of 1 January 2019 and 6.75 percent as of 1 January 2022. In addition, authorities in charge may impose additional TLAC requirements for individual banks in respect of their individual position. The TLAC term sheet document includes requirements as to the quality of TLAC liabilities and deals with the location of TLAC within a banking group. TLAC is conceived as an additional requirement to minimum regulatory capital requirements set out in the Basel III framework. Items that count towards satisfying these minimum regulatory capital requirements may also count towards the minimum TLAC. Moreover, it is expected that the TLAC’s part of eligible liabilities will be at least 33 percent of the amount of the Minimum TLAC requirement. The FSB’s TLAC standard does not answer the question of who – banking supervisor or resolution authority – will set TLAC. Therefore, any legislative follow-up at EU level to implement the TLAC standard will need to assess and give an answer as to how the EU will comply with TLAC standard considering the existence of MREL and the respective enforcement mandate for banking supervisors and resolution authorities.123 5.

Compliance of bail-in with the fundamental right to property?

History shows that bail-in-equivalent measures are often contested in court by affected shareholders or creditors. Applicants invoke – apart from alleging the illegality of the bail-in decision – their constitutionally protected property rights.124 It is against this background that this article examines the compatibility of bail-in, as laid down in the BRRD/SRM Regulation, with the requirements resulting from Article 17 of the EU Charter of Fundamental Rights on the right to property. While the bail-in measures – such as cancellation or transfer of shares, the reduction of the principal amount of liabilities, their suspension or conversion 122. See . 123. EBA will submit a report on the functioning of MREL which could lead to legislative proposals by the EU Commission; see Art. 45(18)-(20) BRRD. 124. For a recent example, see the request for a preliminary ruling, Case C-526/14, Kotnik and Others, pending.

Bail-in

117

into shares – are measures interfering with the right to property, as protected by Article 17 Charter, the main question is whether such interference respects the imposed legal limits and can be validly justified. 5.1.

Interference with property rights

The Charter and thus its Article 17 are applicable according to Article 51(1) Charter, where institutions, bodies, offices and agencies of the Union or where Member States are implementing Union law. That means that it is not only where Council and European Parliament as the co-legislators adopted the relevant provisions on bail-in in both the BRRD and the SRM Regulation or where the SRB adopts the resolution framework applying bail-in, that the Charter is applicable. The Charter is also applicable and thus needs to be complied with where Member States transpose the BRRD in their national law and where national resolution authorities decide and implement bail-in measures. This is because these actions are measures of implementation of Union law.125 Article 17 Charter concerning the right to property guarantees rights corresponding to those guaranteed by Article 1 of Protocol No. 1 to the ECHR. According to Article 52(3) Charter the meaning and scope of Article 17 Charter are to be the same as those laid down by Article 1 of Protocol No. 1 to the ECHR, as interpreted by the case law of the European Court of Human Rights.126 It is however possible to provide more extensive protection under EU law. Article 17(1) Charter provides that: “[e]veryone has the right to own, use, dispose of and bequeath his or her lawfully acquired possessions. No one may be deprived of his or her possessions, except in the public interest and in the cases and under the conditions provided for by law, subject to fair compensation being paid in good time for their loss. The use of property may be regulated by law in so far as is necessary for the general interest.” Article 1 of Protocol No 1 to the ECHR stipulates: “1. Every natural or legal person is entitled to the peaceful enjoyment of his possessions. No one shall be deprived of his possessions except in the public interest and subject to the conditions provided for by law and by the general principles of international law. 2. The preceding provisions shall not, however, in any way impair the right of a State to enforce such laws as it deems necessary to control the use of 125. According to the explanations relating to Art. 51 Charter, which, in accordance with Art. 6(1)(3) TEU and Art. 52(7) Charter, have to be taken into consideration for the purpose of interpreting it, “the provisions of the Charter are addressed to the Member States only when they are implementing EU law”; as to the interpretation of Art. 51(1) Charter, see Case C-206/13, Siragusa, EU:C:2014:126, paras. 20 et seq. 126. See Art. 52(3) Charter.

118

Wojcik

CML Rev. 2016

property in accordance with the general interest or to secure the payment of taxes or other contributions or penalties”. The European Court of Human rights interpreted Article 1 of Protocol No. 1 ECHR in Sporrong and Lönnroth v. Sweden as comprising three distinct rules: “The first rule, which is of a general nature, enounces the principle of peaceful enjoyment of property; it is set out in the first sentence of the first paragraph. The second rule covers deprivation of possessions and subjects it to certain conditions; it appears in the second sentence of the same paragraph. The third rule recognizes that the States are entitled, amongst other things, to control the use of property in accordance with the general interest, by enforcing such laws as they deem necessary for the purpose; it is contained in the second paragraph”.127 It is established case law that shares in a company128 and claims resulting from debt instruments129 constitute property rights. Due to the cancellation of shares or other titles of ownership in a bail-in, the owner will definitely cease to enjoy the rights corresponding with the share, in particular voting rights, rights to influence the commercial activity of the bank and the right to participate in the (potential) profits. Creditors whose liabilities have been reduced or otherwise altered in a bail-in cease to receive (partially or in full) the payment of interest and of the principal at maturity. Creditors whose liabilities have been converted into shares will lose their agreed payment of principal and interest, receiving instead a contingent right to participate in profits (if any). In addition, after a conversion into equity their position in insolvency proceedings is downgraded since equity ranks lower than debt in such proceedings. 5.2.

Proportionality?

For an interference with property to be permissible, such limitations must be provided for by law and respect the essence of the right. In accordance with the principle of proportionality it must not only serve a legitimate aim in the public interest, but the reasonable aim must be sought with by proportionate means.130 A fair balance must be struck between the demands of the general interest of the community and the requirements of the protection of the 127. ECtHR, Sporrong and Lönnroth v. Sweden, Appl. No. 7171/75, judgment of 23 Sept. 1982, para 61. 128. ECtHR, Sovtransavto Holding v. Ukraine, Appl. No. 48553/99, judgment of 25 July 2002, para 92. 129. ECtHR, Fomin and Others v. Russian Federation, Appl. No. 34703/04, judgment of 26 Feb. 2013, para 25. 130. ECtHR, Grainger and others v. United Kingdom, Appl. No. 34940/10, decision of 10 July 2012, para 35.

Bail-in

119

individual’s fundamental rights.131 The ECHR grants Member States to the ECHR a wide margin of appreciation when it comes to general measures of economic or social strategy.132 Due to their direct knowledge of their society and its needs, the Member States to the ECHR are in principle better placed than the international court to assess what is in the public interest on social or economic grounds, and the ECtHR will generally respect the legislature’s policy choice unless it is “manifestly without reasonable foundation”.133 The ECtHR is most stringent in its proportionality test with regard to the interference with the right to property which takes the form of expropriations. According to its case law, the taking of property, i.e. the deprivation of possession, without payment of an amount reasonably related to its value will normally constitute a disproportionate interference and a total lack of compensation can be considered justifiable under Article 1 of Protocol No. 1 only in exceptional circumstances.134 Whether the write down or conversion through bail-in actually qualifies as a “deprivation of possession” can be debated. One could indeed argue that a bail-in is not an expropriation, but a “mere” limitation of “peaceful enjoyment of property”, because the write down or cancellation of a share or of a liability does not involve a transfer of property from the shareholder or creditor to another public or private person. Moreover, it could be reasoned that, for instance, a share which is worth nothing under normal insolvency rules does not have any economic value due to insolvency rules, and that therefore the share either does not need any protection under the right to property or that the cancellation of such a worthless share can be justified more easily than an expropriation. On the other hand, and in favour of a qualification of bail-in measures as “deprivations of possession”, one could argue – and this is the view which this article favours as being legally more convincing – that, formally, the status of a share or of a liability disappears or changes through bail-in. Regardless of economic considerations, the legal position of the shareholder or creditor after bail-in has decreased compared to its legal position ex ante, which is tantamount to the taking of property. Bail-in under BRRD/SRM pursues the resolution objectives set out in Article 31(2) BRRD.135 The most relevant objectives for bail-in are to “avoid 131. ECtHR, Sporrong and Lönnroth v. Sweden, Appl. No. 7171/75, para 73. 132. ECtHR, James and Others v. United Kingdom, Appl. No. 8793/79, judgment of 21 Feb. 1986, para 46. 133. ECtHR, Grainger and Others v. United Kingdom, Appl. No. 34940/10, para 35. 134. ECtHR, The Holy Monasteries v. Greece, Appl. No. 13092/87 and 13984/88, judgment of 9 Dec 1994, para 71; ECtHR, The former King of Greece and Others v. Greece, Appl. No. 25701/94, judgment of 23 Nov 2000, para 89; ECtHR, Jahn and Others v. Germany, Appl. No. 46720/99, 72203/01 and 72550/01, judgment of 30 June 2005, para 117. 135. Art. 14 SRM Regulation.

120

Wojcik

CML Rev. 2016

a significant adverse effect on the financial system, in particular by preventing contagion . . . and by maintaining market discipline” and “to protect public funds by minimizing reliance on extraordinary public financial support”. These two goals – financial stability and fiscal protection – can be regarded as legitimate aims within the public interest.136 The analysis whether these legitimate goals are sought with by proportionate means, i.e. by means which are effective, the least intrusive and appropriate, has two underlying dimensions, an abstract and a specific one: on the one hand, the legal provisions concerning bail-in as such need to be proportionate; on the other hand, the resolution authority will need to decide in a proportionate way on the application of the bail-in provisions in each resolution case. While, in principle, the legal provisions on bail-in in abstracto can be considered as proportionate (for the reasons to follow), any assessment of a specific application of the bail-in provisions by resolution authorities can only be made ex post with regard to the specific resolution case. However, since in both situations, which are closely intertwined, similar aspects will need to be considered, major pitfalls to be avoided by the resolution authority when deciding on a bail-in can already be highlighted. 5.2.1. The “no creditor worse off” principle The foundation on which bail-in is based – and which provides for the elementary justification why bail-in can be considered as a proportionate mean to achieve the aims pursued by it – is the principle that, when bailed-in, shareholders and creditors are treated in an identical way compared to the treatment they would have faced under normal insolvency proceedings. Article 34(1) lit. g BRRD137 explicitly lays down this principle, commonly described as the “no creditor worse off ” principle, prescribing that “no creditor shall incur greater losses than would have been incurred if the bank had been wound up under normal insolvency proceedings.” If this goal is effectively achieved, the cancellation of a share or the write down or conversion of a liability does not alter the economic situation of the respective creditor or shareholder with regard to an insolvency procedure. This is well illustrated with the example of shareholders who would not receive anything from the liquidation of the bank’s assets under normal insolvency proceedings and whose shares were thus worth nothing: the cancellation of their shares through bail-in will not put them in a worse position. Similarly, if the bank’s creditors were only partially redeemed under normal insolvency proceedings, then a write down to the extent of what they would economically lose in 136. ECtHR, Bugajny and others v. Poland, Appl. No. 22531/05, judgment 6 Nov 2007, para 63; ECtHR Grainger and others v. United Kingdom, Appl. No. 34940/10, para 42. 137. Art. 15 SRM Regulation; see also Art. 73(b) BRRD.

Bail-in

121

insolvency proceedings is in any case justified. In the same logic, the value of the equity stake which creditors would receive through the conversion of their claims needs to reflect the value of their claim under normal insolvency proceedings. This idea is reflected in Article 50 BRRD, which prescribes that the conversion rate for a liability into equity needs to appropriately compensate the affected creditor for any loss incurred. At this point, creditors may try to argue that the introduction of bail-in as such altered their legal and economic situation, because, due to implicit State guarantees materializing in outright bail-outs of a failing bank, it was – before the adoption of the legal framework on recovery and resolution – very improbable that a systemically important bank entered into insolvency. This allegation does not change the assessment above. In the EU, State aid is, as the general rule set out in Articles 107 and 108 TFEU, prohibited. Therefore, investors in a bank cannot rely on legitimate expectations for the granting of public financial support or claim the right to benefit from a State aid measure. Obviously, the “no creditor worse off ” principle needs to be specified in such a way that the intended result is effectively achieved. It is to comply with this requirement that the sequencing of the bail-in measures follows the reverse order of the priority of claims under the applicable national insolvency law, affecting shareholders first and then moving on to other capital instruments, then junior and, only finally, implying senior debt holders, 138 including pari passu-treatment of creditors.139 Curiously enough, Articles 34(1) lit. g and 73 BRRD which lay down the “no creditor worse off ” principle by their wording only apply to resolution tools and resolution powers. Since the write down and conversion instrument is not a resolution tool or resolution power, one could argue that the “no creditor worse off ” principle did not apply to the write down or conversion instrument. Such a result would however make the application of the write down and conversion instrument incompatible with the right to property. In the light of this fundamental right, the only appropriate interpretation is to extend the application of the “no creditor worse off ” principle to the write down and conversion instrument. 5.2.2. Additional safeguards, including compensation Furthermore, the BRRD in its Articles 73–75 provides for important additional “safeguards” to give real effect to the “no creditor worse off ” principle. While Article 73 BRRD mainly repeats the “no creditor worse off ” principle, Article 75 BRRD introduces a payment claim for bailed-in 138. Arts. 34(1)(a)(b), 48, 60 BRRD and Arts. 15, 21, 27 SRM Regulation. 139. Arts. 34(1)(f) BRRD and Arts. 15, 21, 27 SRM Regulation.

122

Wojcik

CML Rev. 2016

shareholders and creditors140 against the relevant resolution financing arrangement, to compensate them in case bail-in imposed a greater loss on them than they would have had to bear under normal insolvency proceedings.141 According to Article 74 BRRD the difference to be compensated is to be established by a valuation carried out by an independent person as soon as possible after the bail-in was implemented. This valuation has to be distinct from the valuation under Article 36 BRRD and focuses exclusively on the question whether the treatment of shareholders and creditors would have been better under normal insolvency proceedings than under resolution. This compensation possibility takes into account the case law of the ECtHR on the requirement of compensation in case of expropriations. The compensation solution under Article 74 BRRD is necessary, but also sufficient to comply with the requirement of proportionality of the bail-in provisions. By limiting the compensation claim in Article 75 BRRD to situations in which bail-in would result in imposing higher losses on shareholders or creditors than they would have to bear under normal insolvency proceedings, the bail-in rules do not give rise to expropriation without any compensation. If the shares of a bank are under normal insolvency proceedings worth nothing, no compensation is warranted where the bank’s losses are absorbed by equity. In a similar vein, if all the losses cannot be absorbed by equity and a bank continues to fall short of the minimum regulatory capital required to continue to be allowed to operate, no compensation is warranted for the creditors of that bank as long as the economic value of those creditors’ claims is not reduced to below the level at which those claims would have stood if no State aid was given to the bank. As was stated above, the ECtHR has held in relation to Article 1 of the First Protocol: “That Article does not, however, guarantee a right to full compensation in all circumstances, since legitimate objectives of ‘public interest’ may allow to reimburse less than the full market value”.142 As such, the full market value of the shares or claims of creditors, as the case may be, constitutes an upper cap on the level of compensation which is required by that provision in cases of deprivation of possessions. However, since the BRRD requires that the “no creditor worse off ” principle is respected, shareholders and creditors will receive the full market value of their claims against the bank, including through payments of compensations according to Article 75 BRRD.

140. Including Deposit Guarantee Schemes. 141. Such a compensation payment against the relevant resolution financing arrangement should not be an incentive to exclude on a discretionary basis liabilities from bail-in. 142. See case law cited supra note 134.

Bail-in 5.3.

123

Specific legal issues

5.3.1. Exceptions to the “no creditor worse off” principle Specific attention needs to be paid to the exceptions to the “no creditor worse off ” principle, i.e. the exclusions of certain liabilities from bail-in and the possibility to deviate from pari passu-treatment of creditors. This exception will effectively result in a “juniorization” of certain liabilities. Liabilities excluded from bail-in will not be written down or converted and will thus be treated in a different way from eligible liabilities. Since the “pool of liabilities” which will bear the losses thus gets smaller, the likelihood increases that eligible liabilities that otherwise would not or not to the extent have been touched upon, will actually be written down or converted. Such difference of treatment in the light of the creditors’ right to property can be justified with respect to covered deposits. If covered depositors were to be bailed-in, the risks to financial stability would be enormous and the possible damages would outweigh the benefits resulting from the bail-in of such liabilities. The legislature explicitly highlights the importance of protecting depositors by naming this one of the resolution objectives in Article 31 BRRD. Other mandatory exclusions of liabilities in Article 44(2) BRRD can be justified by the need to prevent contagion for the financial system, the continuity of the core operation of the failing bank with a view to avoid the destruction of value or the need to maintain the financial stability of tax or social security systems. 5.3.2. Appropriate quantitative calibration of bail-in Similar reasons to justify discretionary exclusions of liabilities under Article 44(3) BRRD are mentioned explicitly in that provision, so that one could argue that Article 44(3) BRRD would comply in general with Article 17 Charter. However, it will be the application of Article 44(3) BRRD in a specific resolution situation which will need a very careful and thorough assessment of the facts and circumstances by the resolution authorities. This will definitely be a challenge for any resolution authority. Another difficulty is the appropriate quantitative calibration of bail-in. Article 46(2) BRRD requires to establish an amount of bail-in which is adequate to sustain sufficient market confidence in the bank and enable it to continue to meet, at least for a year, the conditions for authorization. In case of recapitalizations the bail-in should also restore the bank to “financial soundness and long-term viability”. These terms are pretty generic and rather vague, involving a high degree of prognosis. The resolution authority will need to make a very complex decision based on a (sometimes preliminary) valuation and on other assumptions. More importantly, the forward looking

124

Wojcik

CML Rev. 2016

perspective of the criteria to be used for this bail-in which may advocate for a higher extent of write down and conversion of eligible liabilities “to be on the safe side for the future” is not easy to reconcile with the “gone concern” perspective which underlies any comparison with the treatment of creditors in a normal insolvency proceeding according to the “no creditor worse off ” principle.143 This problem will already arise when the independent valuer needs to do its valuation pursuant to Article 74 BRRD. Furthermore, this will make it even more difficult for resolution authorities to defend a specific bail-in decision in case of litigation brought by affected creditors who will certainly and among others claim the correctness of a different valuation. It remains to be seen whether these difficulties will provide incentives to be very cautious in order to avoid litigation risks or whether resolution authorities will simply pass on this problem to the relevant financing arrangement which will need to pay compensation according to Article 75 BRRD if the “no creditor worse off ” principle is not respected. Legal and practical issues in the application of the “no creditor worse off” principle Finally, while the principle of “no creditor worse off ” is fundamental for establishing the proportionality of the interference of bail-in measures with the right to property, the application of this principle is not as straightforward as might seem on the paper. Instead, it raises several legal and practical issues which may impact on the proportionality of the bail-in operation. The first issue relates to the fact that in order to establish the treatment which shareholders and creditors would have received had the bank been wound up under normal insolvency proceedings, a comparison will be made with a hypothetical situation. Unavoidably, such a comparison will need to simulate the outcome of a hypothetical insolvency proceeding and will be based on a number of assumptions,144 since insolvency proceedings usually take time, depend on the application of specific rules by courts or administrative authorities which decide depending on specific motions or situations of creditors. That may result in rather unsatisfactory outcomes due to the uncertainties necessarily involved. Obviously, this will contribute to a higher risk of legal challenge. The second issue relates to the fact that substantive insolvency law, including the order of priority of claims, is not comprehensively harmonized 5.3.3.

143. Adolff and Eschweg, “Lastenverteilung bei der Finanzmarktstabilisierung”, 177 ZHR (2013), 902–977, at 969–970. 144. On more technical valuation matters, such as the appropriate “cut off ” date for valuation, see Athanassiou, “Valuation in resolution and the ‘no-creditor-worse-off principle’”, 29 Butterworths Journal of International Banking and Financial Law (2014), 16–20.

Bail-in

125

in the EU and that Member States are thus, in principle, free to regulate in this area. It is true and already an important achievement that the BRRD in its Article 108 harmonizes partially the order of priority of claims in insolvency proceedings, namely requiring that the part of eligible deposits from natural persons and micro, small and medium-sized enterprises which exceeds ¤100,000 and that deposits which would be eligible deposits from natural persons, micro, small and medium-sized enterprises were they not made through branches located outside the EU of banks established within the EU have to have the same priority ranking, which is higher than the ranking provided for the claims of ordinary unsecured, non-preferred creditors. At the same time, covered deposits and Deposit Guarantee Schemes subrogating to the rights and obligations of covered depositors in insolvency need to have the same priority ranking, but this ranking needs to be higher than the ranking for the claims mentioned before. Apart from this, EU law as it stands now only harmonizes provisions on jurisdiction, recognition and applicable law.145 This legal situation means that every resolution authority, when complying with the “no creditor worse off ” principle, will need to determine the applicable national insolvency law for the bank under resolution. In principle, this will be, according to Article 10 of Directive 2001/24, the laws, regulations and procedures applicable in its home Member State. However, in exceptional circumstances other legal orders may be applicable to specific legal aspects of the insolvency proceedings, adding another layer of complexity given the various international links of today’s banking business.146 While such an assessment may still be feasible in practical terms for a national resolution authority carrying out a bail-in measure, the situation is much more demanding for the SRB when it is deciding in the framework of the Banking Union and moreover on significant banks, whose structure will be even more international and more complex. In practical terms, the SRB will need to know and be able to take into account the insolvency regimes of the 19 participating Member States. While Article 17(2) SRM Regulation requests participating Member States to notify to the SRB (and the Commission) the latest ranking of claims against banks in their national insolvency proceedings, thus facilitating the monitoring of these

145. See in general Council Regulation (EC) 1346/2000 of 29 May 2000 on insolvency proceedings, O.J. 2000, L 160/1; the Recast Regulation on Insolvency 2015/848 will apply, in principle, as of 26 June 2017. Specifically for credit institutions, see Directive 2001/24/EC of 4 April 2001 on the reorganization and winding up of credit institutions, supra note 6. 146. See e.g. the list of exceptions in Arts. 20–33 of Directive 2001/24; for the interpretation of Art. 30 of Directive 2001/24, see the judgment of the EFTA Court of 17 Oct. 2014 in Case E-28/13, LBI v. Merrill Lynch International Ltd.

126

Wojcik

CML Rev. 2016

complicated rules, the task of the SRB will be extremely difficult and the risk of committing errors considerable. Lastly, although the lack of EU law harmonization of the order of priority of claims in national insolvency proceedings will not make the bail-in rules incompatible with Article 17 Charter, it increases the risk of a disproportionate application of bail-in rules in a specific resolution situation, making it more difficult for creditors to predict the likelihood of their liabilities being bailed-in. Moreover, it might also contribute to a certain fragmentation of the internal market with respect to investments in banks. Member States might try to regulate the ranking of claims against banks in insolvency proceedings in such a way that incentives might be created for creditors to provide debt capital to banks established in that Member State due to a more senior statutory ranking of that claim in insolvency proceedings, since this would reduce the probability of their claims being bailed-in. To sum up: the bail-in rules under BRRD/SRM Regulation do respect the right to property as laid down in Article 17 Charter. However, the correct application by resolution authorities of the safeguards necessary to come to that assessment will be extremely challenging. 6.

Will and can bail-in achieve its objectives in the light of its legal and economic implications?

The main objectives of bail-in as the crucial part of the newly established bank recovery and resolution framework in the EU are to protect taxpayers from rescue operations to the benefit of failing banks, to avoid significant adverse effects on the financial system, in particular by preventing contagion, and to restore the orderly functioning of markets without competition distortions and moral hazard, by taking away the implicit State guarantee for banks. To break the link between banks and sovereigns, bail-in is conceived as an instrument mirroring for shareholders and creditors alike the (hypothetical) outcome of normal insolvency proceedings in a dedicated administrative procedure which is triggered by decisions of resolution authorities and as a prerequisite for the use of resolution financing arrangements. While economic analysis has shown – retrospectively – that the existence of bail-in during the financial crisis would have alleviated the recourse to taxpayers’ money to a great extent,147 bail-in can in the real world only 147. Conlon and Cotter, “Anatomy of a bail-in”, 14 Journal of Financial Stability (2014), available at , at 14; for an economic analysis, see Galliani and Zedda, “Will the bail-in break the vicious circle between banks and their sovereign?”, 45 Computational Economics (2015), 597–614.

Bail-in

127

achieve its goals where it functions properly in light of the legal and economic consequences it will trigger and in light of the challenges and risks which its application will bring. Since ultimately bail-in is an accounting operation changing the balance sheet structure of a bank,148 this article’s view is that bail-in can achieve its goals to a large extent depending on the legal and economic circumstances and if certain conditions are respected. 6.1.

Impact on banks’ costs of funding and the pricing of debt

The removal of the implicit State guarantee and its replacement with a credible bail-in mechanism will inevitably impact on banks’ funding costs. Creditors will reassess their investment in a bank’s debt instruments according to the probability of a write down or conversion of their debt instruments. Such assessment will focus, inter alia, on the bank’s general creditworthiness, its business model, its structure, its systemic importance, whether it is part of an institutional protection scheme and also on the expected behaviour of a resolution authority to “pull the trigger” and apply bail-in. Moreover, prospective and existing creditors will examine the liability structure of the bank, including the amount and type of liabilities excluded from bail-in, in order to understand which order of priority their debt instrument would take in the applicable normal insolvency proceeding, thus trying to predict the position of their claim in a bail-in. However, it would not be sound to state that the funding costs of all banks in all cases would increase in a resolution framework which includes bail-in. If creditors come to the conclusion that the probability of a bail-in is higher than a loss given default through normal insolvency proceedings, then they would be expected to ask for a higher remuneration for accepting a bank’s debts instruments. This effect could in some cases however be reversed since the costs to creditors for the destruction of value through normal insolvency proceeding might fall due to an expected higher recovery rate in resolution compared to bankruptcy.149 In this context, the introduction of MREL150 may have specific implications for a bank’s balance sheet and funding costs due to MREL’s 148. Bates and Gleeson, “Legal aspects of bank bail-ins”, May 2011, available at , at 19. 149. Eliasson, Jansson and Jansson, “The bail-in tool from a Swedish perspective”, 2 Sveriges Riksbank Economic Review (2014), 23–51, at 34; see also Commission Staff Working Document of 6 June 2012, “Impact Assessment on the BRRD proposal”, SWD(2012)166 final, available at , at 137 et seq. 150. The possible implications of TLAC will not be discussed, since the FSB’s TLAC standard has to be yet transposed into binding laws.

128

Wojcik

CML Rev. 2016

effect on the pricing of debt. MREL will be an indicator for creditors as regards the probability whether their liability will be affected by a bail-in operation if a bank gets into trouble. Already the determination of which liabilities are bail-inable or not will probably lead to a “shift of interest expenses between liability classes”,151 since excluded liabilities will be perceived as being less risky, while bail-inable liabilities will, in principle, be perceived as more risky and thus pricier. MREL will further enhance this process, given that according to Article 45(4) BRRD152 only specific bail-inable liabilities can be counted towards MREL, for instance eligible liabilities with maturity of at least 1 year. Thus, Article 45 BRRD creates another subset of bail-inable liabilities, which are even more likely to be bailed-in than other eligible liabilities and, hence, should be even more costly for banks to issue. The introduction of MREL might create specific difficulties for banks to find enough investors to take on subordinated senior debt which would meet the MREL quality requirements, thus driving up funding costs. This is because potential investors might be deterred from making MREL-compatible investments at all, fearing they might not have the necessary capacities to manage “potential equity exposures” after a conversion of debt.153 Moreover, the investor base might be reduced due to restrictions in the investment mandate as to the purchase of MREL-compatible debt.154 The amount of liabilities available is further reduced by the fact that liabilities governed by the law of a third country will not count towards meeting MREL, if it cannot be established that the bail-in decision of a European resolution authority would have automatic effect in the third country’s legal order.155 Finally, some banks which currently have financing models which did not rely on MREL-compatible debt will need to change their liability structure. Since they will need to issue a considerable amount of MREL-compatible debt for the first time and at a moment where other banks will do the same, the potential number of investors will be even further diminished, thus increasing funding costs. To sum up: if as a consequence of the introduction of bail-in in European legislation “creditor inertia”156 ends, and debt holders start actively assessing their investment and continue monitoring it, this would clearly be a positive development, since a stronger involvement of debt-holders with the management of a bank will contribute to an enhancement of prudent 151. Eliasson, Jansson and Jansson, op. cit. supra note 149, at 36. 152. Art. 12 SRM Regulation. 153. Eliasson, Jansson and Jansson, op. cit. supra note 149, at 39. 154. Gleeson, op. cit. supra note 17, at 9. 155. Art. 45(5) BRRD and Art. 12(17) SRM Regulation. 156. Goodhart and Avgouleas, “Critical reflections on bank bail-ins”, Jan. 2015, available at , at 4.

Bail-in

129

behaviour of the management of a bank.157 From a macroeconomic perspective, the disciplinary effects of stronger creditor involvement in the affairs of a bank which will add to shareholder induced discipline may overall reduce excessive risk taking by banks and thus be conducive to a reduced likelihood of systemic bank crises.158 Obviously, increased credit assessment and monitoring by debt-holders will, despite the general difficulties attached to it,159 be easier to carry out for institutional investors than for non-professional individuals. In this sense, an investment in a bank, through the acquisition of bonds for instance, will become more risky for the latter. However, bail-in is expected to incentivize behaviour which should be appropriate even if bail-in did not exist. Prices of debt would more correctly reflect the real risk of any investment in a bank, strengthening the responsibility of shareholders and creditors for their investments. 6.2.

Risks of contagion

According to Article 31(2) lit. b BRRD, one of the explicit resolution objectives applicable to bail-ins is “to avoid a significant adverse effect on the financial system, in particular by preventing contagion”. In general, bail-in will contribute to avoiding contagion, because losses are precisely imposed on those who can expect to take them, thus preventing asset sales by banks to finance losses.160 However, it has been argued that, in certain situations, the application of bail-in in a systemic crisis might actually create contagion, 161 since creditors might sell debt and refrain from taking on new debt from banks. As a consequence, banks might be forced to (fire-)sell assets in order to adjust their balance sheets. Because the current resolution framework is likely to incentivize institutional investors (i.e. often other banks or insurance undertakings) to take on bail-inable debt and may thus create interconnectedness, the failure of one bank might, through the write down and

157. Goodhart and Avgouleas, op. cit. supra note 81, at 36. 158. See Basel Committee on Banking Supervision, TLAC Quantitative Impact Study Report, pp. 22 and 23, financialstabilityboard.org/2015/11/bcbs-tlac-quantitative-impactstudy-report/. 159. Goodhart and Avgouleas, op. cit. supra note 81, at 30. 160. Commission Staff Working Document, “Impact Assessment on the BRRD proposal”, see supra note 149, at 131. 161. Goodhart and Avgouleas, op. cit. supra note 81, at 37.

130

Wojcik

CML Rev. 2016

conversion of liabilities which are other banks’ assets, be transmitted to these other banks and provoke a further need for bail-in.162 Against this background, it is true that bail-ins will be most effective when applied to non-systemic, smaller domestic banks.163 However, it should be kept in mind that in a systemic crisis, the lack of bail-in would probably lead to even more contagion, since in such a situation liquidation – unless avoided by bail-outs – would destroy even more value and have even stronger knock-on effects.164 6.3.

Effectiveness of bail-in

The effectiveness of bail-in will very much depend on whether its application is predictable for the markets and whether its application will produce stable results. Predictability and stability of result are key prerequisites for a successful bail-in, since uncertainty about the course of bail-in and its final outcome may endanger financial stability and produce sub-optimal results. 6.3.1. Managing expectations Predictability starts for shareholders and creditors with a reasonably correct comprehension about the moment when the resolution authority will trigger the write down and conversion instrument or the bail-in tool. To this end, resolution authorities, together with the banking supervisors, which both enjoy some discretion with regard to the determination of the triggering events for resolution, should communicate transparently and in a reliable manner on the circumstances under which they would exercise the resolution powers.165 Furthermore, resolution authorities should try to make it clear who will be affected, and to what extent, by a specific bail-in measure. While in theory the application of the bail-in instrument seems to be rather straightforward, in fact, resolution authorities still enjoy discretion when it comes, for instance, to the exercise of discretionary exclusion of liabilities from a bail-in.166 Moreover, as was shown above, due to the lack of harmonized insolvency laws in the EU, the predictability of and trust in the correct application of the “no creditor worse off ” principle, may be questioned, if investors do not precisely know which national law will be applicable and what the treatment

162. Acknowledging this risk of contagion, the FSB TLAC standard (p. 7) emphasizes the importance to strongly disincentivize internationally active banks from holding TLAC issued by G-SIBs. 163. Goodhart and Avgouleas, op. cit. supra note 81, at 37. 164. Gleeson, op. cit. supra note 17, at 24. 165. Huertas, op. cit. supra note 4, at 175 et seq.; Gleeson, op. cit. supra note 17, at 9. 166. See Art. 44(3) BRRD and Art. 27(5) SRM Regulation.

Bail-in

131

of their claims under that national law would be. This is exacerbated by the fact that over time the applicable national insolvency law may change. Ideally, this problem could be addressed by harmonizing at EU level the ranking of claims under normal insolvency proceedings. While some harmonization in this respect was done in the “depositor preference provision” of Article 108 BRRD, it is only partial. In addition, such a harmonization would be a long-term legislative project.167 Therefore, it is paramount that resolution authorities make thorough use of resolution plans to prepare and be prepared for the event. This alone will however not create the necessary degree of predictability for the public, since resolution plans contain a lot of confidential information and are not to be disclosed. However, these plans could be the basis for resolution authorities or indeed the respective bank to communicate with the interested public. 6.3.2. Legal stability of results Another element contributing to the effectiveness of bail-in is that creditors, counterparties and customers need to be convinced that once bail-in is applied, the bank is robust again,168 and that the result of the bail-in operation will not be put in question or altered at a later moment. Litigation challenging the resolution authorities’ decision on a bail-in can be expected, given the important legal and economic consequences for the affected persons.169 Such litigation will most probably relate to the compatibility of both bail-in decision and of the BRRD/SRM provisions on bail-in with fundamental freedoms, namely the right to property under Article 17 Charter, discussed above. In particular ex ante judicial review, given the urgency and time constraints of any resolution procedure, poses a considerable challenge to the need for effectiveness of any bail-in.170 The EU legislature was aware of these litigation risks and sought solutions to contain them. As a consequence, it was intended to limit ex ante judicial review as much as legally feasible and to shift the centre of the legal battles to an ex post review of the adequacy of compensation. First of all, Article 85 BRRD reaffirms the right to an effective remedy as guaranteed under Article 47 Charter, including the possibility of ex ante judicial review. However, in light of the necessities to guarantee an effective bail-in, which is paramount to achieving financial stability, Article 85 BRRD requires an expeditious ex ante judicial review and emphasizes the fact that 167. However, it should be noted that the Commission announced in its Communication of 30 Sept. 2015, “Action plan on building a Capital Markets Union”, COM(2015)468 final, at 25 and 30, that it would make a legislative proposal on business insolvency by the end of 2016. 168. Gleeson, op. cit. supra note 17, at 24. 169. Goodhart and Avgouleas, op. cit. supra note 81, at 36. 170. Bliesener, op. cit. supra note 76, at 225.

132

Wojcik

CML Rev. 2016

bail-in decisions are based on complex economic assessments of facts which are usually not reassessed in every detail by the courts. Secondly, appeals brought by affected persons against a resolution decision do not entail any automatic suspension of the effects of the challenged decision. The decision of the resolution authority is still immediately enforceable. Moreover, Article 85(3) BRRD prescribes that the annulment of a decision of a resolution authority does not affect any subsequent administrative acts or transactions concluded by the resolution authority concerned which were based on the annulled decision, if it is necessary to protect the interests of third parties acting in good faith who have acquired shares, other instruments of ownership, assets, rights or liabilities of an bank under resolution by virtue of the use of resolution tools or exercise of resolution powers by a resolution authority. Remedies for a wrongful decision or action by the resolution authorities are limited to compensation for the loss suffered by the applicant as a result of the decision or act. This illustrates the intention to focus on a remedy for compensation as a means to leave the legal results of the bail-in operation unaltered. Thirdly, Article 75 BRRD grants a payment claim for bailed-in shareholders and creditors171 against the relevant resolution financing arrangement to compensate them, in case bail-in imposed a greater loss on them than they would have had to bear under normal insolvency proceedings. The difference to be compensated is to be established by a valuation to be carried out, according to Article 74 BRRD, by an independent person as soon as possible after the bail-in was implemented. Again, this demonstrates that the legislature’s intention was to maintain the effectiveness of the legal result of the bail-in operation, by adjusting possible issues with the quantitative calibration of bail-in through an ex post pecuniary compensation claim to alter the economic, but not the legal result of the bail-in operation. Finally and beyond the measures taken by the EU legislature, one could consider introducing special contractual clauses in capital and debt instruments in order to attain further stability of the legal result of the statutory bail-in. Such contractual clauses should define the exercise of the resolution authority’s bail-in powers as a trigger event according to which the principal amount of the liability will be written down or the liability converted into equity in line with a conversion rate determined by the resolution authority.172 This contractual clause could be interpreted as a waiver by the creditor to 171. Including Deposit Guarantee Schemes. 172. Such debt instruments are to be distinguished from so called “contingent convertible bonds”, which provide for contractual terms allowing the conversion into equity with trigger events which do not refer to regulatory interventions; see Gleeson, op. cit. supra note 17, at 15; Kenadjian, op. cit. supra note 17, at 238.

Bail-in

133

attack the resolution authority’s decision through an ex ante judicial remedy. But it is rather doubtful whether the creditor may renounce from exercising any remedy against the resolution authority through a contractual clause agreed with the bank. This would not seem to be in the objective and equitable interest of the creditor. When agreeing to the contractual clause as set out above, the creditor will, despite his willingness to keep the result of the bail-in operation stable, implicitly only agree to a lawful exercise of the resolution powers by the resolution authority.173 It follows from this that the contractual clause will not exclude the possibility to request compensation from the resolution authority for a wrongful exercise of the bail-in powers. This is enough, though, to achieve a sufficient degree of stability of the bail-in operation and in line with the EU legislature’s desire to have the battles fought only on ex post pecuniary compensation. As to the practicalities of this approach, one needs to highlight that the insertion of such contractual clauses would need to be done on a consensual basis by the bank and its creditors, both with regard to existing and new liabilities. De lege lata, there are no legal requirements which could force them to do so: for the time being, the BRRD only requires similar contractual clauses to be included in debt instruments which are governed by the law of a third country.174 Likewise, Articles 52(1) lit. n, 54 CRR require a contractual clause providing for write down or conversion to be inserted in a financial instrument to be qualified as an AT1 item. However, such a contractual clause may, but will not necessarily, be synchronized with the resolution authorities’ decision to bail-in: the trigger event mandatorily requested by Article 54 CRR may not capture the situation where the resolution authority may apply bail-in. CRR does not require any contractual clause for Tier 2 items.175 Liabilities governed by the law of a jurisdiction different from the law of the failing bank The effectiveness of statutory bail-in measures taken by resolution authorities with respect to debt instruments governed by the law of a jurisdiction different from the law of the (failing) bank’s home Member State will depend on the (timely) recognition of the exercise of the bail-in, i.e. the write down or conversion of the liabilities by the home Member State’s resolution authority.176 The treatment of these situations depends on whether, assuming 6.3.3.

173. This is even more true if a contractual clause is mandatory under a legal requirement (e.g. Art. 55 BRRD), since the law may not require anybody to waive a remedy in advance. 174. See Art. 55 BRRD. 175. See also Joosen, “Bail in Mechanisms …”, op. cit. supra note 115, at 9 et seq. 176. See FSB, “Cross-border recognition of resolution action”, Sept. 2014, available at , at 13. This article does not

134

Wojcik

CML Rev. 2016

that the bank is established in the EU, the liability to be bailed-in is governed by the law of another EU Member State (often UK law) or by the law of a third country (often New York State law). In the former case, Article 3 of Directive 2001/24 stipulates that a reorganization measure, which after the amendment of Article 2 of Directive 2001/24 by the BRRD now specifically mentions bail-in measures, is to be automatically recognized and be effective across the EU. In the latter case, Article 55 BRRD provides for a two-step-approach: either the third country jurisdiction recognizes and enforces the decision of the resolution authority of the bank’s EU home Member State, including the SRB, or Member States will require banks to include recognition clauses for statutory bail-in measures in the contractual terms of the debt instrument. Although there cannot be complete certainty, courts will generally enforce contractual provisions properly entered into unless contrary to public policy.177 6.3.4. Banking groups Finally, the effectiveness of bail-in, in the same manner as the effectiveness of any resolution procedure, will very much depend on whether the resolution is applied to a single bank entity or in the context of a banking group. This is particularly important since most global systemically important banks are organized in groups of different structure and complexity operating in numerous jurisdictions. For such groups the necessary coordination and cooperation between different regulatory authorities across countries can be very challenging and complex to organize.178 While for purely Banking Union-based banking groups, the SRB will be the only resolution authority, so that – depending on the efficiency of the SRB’s internal governance – the decision-making in resolution situations is expected to be less complex, banking groups which are solely based outside the Banking Union or in and outside the Banking Union will face more challenging coordination and collaboration demands. In such situations, it is the home Member State’s resolution authority of the individual banking entity under resolution which is empowered to take the resolution actions, including bail-in, and is obliged to have due regard to the impact of its decision in one or

deal with the topic of recognition of resolution procedures in general, which include situations where subsidiaries or branches are located in a different jurisdiction. 177. Ibid., 14. 178. On policy challenges for the resolution of cross-border banking groups in general, see e.g. Gleeson, “The importance of group resolution” in Dombret and Kenadjian, op. cit. supra note 4, pp. 25–38, at 25, as well as for an analysis of the BRRD’s rules on group resolution, see Randell, “Group resolution under the EU Resolution Directive”, in Dombret and Kenadjian, ibid., pp. 39–68, at 53 et seq.

Bail-in

135

more other Member States.179 Resolution colleges are set up, presided by the group level resolution authority, i.e. the resolution authority in the Member State where the consolidating supervisor is situated, to facilitate the decision-making of the Member States’ resolution authorities,180 without being itself a decision-making body. Matters become even more complicated where third country jurisdictions are involved, either when subsidiaries or branches of EU based banks are situated in third country jurisdictions or vice versa. The EU’s approach is to enter into international agreements on the cooperation, recognition and enforcement with third countries.181 Depending on the conclusion and scope of such international agreements, non-binding cooperation agreements in lieu or in addition may be concluded between the EBA and relevant third-country authorities or between national resolution authorities and relevant third-country authorities.182 In the absence of such agreements and even despite a joint decision of a European resolution college, the home Member State’s resolution authority of the individual banking entity under resolution remains in any case competent to refuse to recognize or enforce third country resolution proceedings.183 It is obvious that such cooperation and collaboration procedures can be cumbersome and lengthy, their outcome difficult to predict and thus the effectiveness of bail-in in a specific situation can be impeded. Recourse to such complex cooperation procedures will in particular happen the more decentralized the legal structure of the banking group is.184 One way of reducing the complexity of group resolution procedures is to structure a banking group in such a way that a “single point of entry” resolution (SPOE)185 can be applied. SPOE actually requires a group structure with a non-operative top level company and subsidiaries operating in the banking business. It was developed and is applied predominantly in the USA where for historical reasons banking groups are organized in such a way.186 In SPOE, resolution powers, such as bail-in, will only be applied to the top-level company which absorbs the losses incurred within the group, i.e. with certain 179. Art. 87 BRRD. This power remains with the home Member State’s resolution authority even where a group resolution scheme was adopted; see Arts. 91(8) and 92(8) BRRD. 180. Recital 98 BRRD. 181. Art. 93 BRRD. 182. Art. 97 BRRD. 183. Art. 95 BRRD. 184. Binder, op. cit. supra note 5, at 124. 185. See FSB, “Recovery and resolution planning for systemically important financial institutions: Guidance on developing effective resolution strategies”, July 2013, available at , at 13. 186. Gordon and Ringe, op. cit. supra note 10, at 1330 et seq.

136

Wojcik

CML Rev. 2016

operating subsidiaries.187 As a consequence and in parallel, operating subsidiaries should be able to continue as a going concern without entering resolution.188 By contrast, “multiple point of entry” resolution (MPOE) involves the application of resolution powers by two or more resolution authorities to different entities of the group, depending on which of them meets the resolution conditions. MPOE requires actions to be coordinated across jurisdictions so as to avoid conflicts or inconsistencies that undermine the effectiveness of the separate resolution actions.189 These strategies are not mutually exclusive, but can be combined. The EU resolution framework allows explicitly for SPOE and MPOE.190 As a matter of fact, however, the type of legal structure conducive to the SPOE approach is not prevalent in the EU, at least not in the Banking Union. Many, also global systemically important banks based in the EU actually have operating top-level banking entities. To address the complexities and possible impediments to bail-in within a banking group, at least two approaches are conceivable. The first answer would be “preparation”. That means to focus and devote much diligence to drawing up resolution plans, in particular group resolution plans.191 The second answer could be to find ways in order to change the legal structure of the banking group to make it suitable for the SPOE approach.192 Obviously, banks could decide themselves to change their corporate structures. In addition, regulatory measures could be contemplated as well. Apart from imposing structural changes through new legislation,193 one way of imposing legal structures more conducive to resolution, including bail-in, could be the use of the powers to address impediments to resolvability conferred on resolution authorities in the resolution planning phase.194 To sum up: the analysis shows that a bail-in will have many legal and economic consequences. Bail-in has a considerable guiding impact before the 187. Ibid., at 1323 et seq., for a description of such a process. 188. For additional requirements for a good functioning of SPOE, see FSB, op. cit. supra note 185, at 16. 189. Ibid., at 14. 190. Recital 80 BRRD. 191. On the need for recovery and resolutions plans in general, see Hüpkes, “‘Living Wills’: An international perspective”, in Dombret and Kenadjian, op. cit. supra note 4, pp. 71–86, at 80 et seq. 192. This approach is vigorously advocated by Gordon and Ringe, op. cit. supra note 10, at 1364 et seq. 193. See Commission Proposal for a Regulation to improve resilience, cited supra note 26. 194. Art. 17 BRRD and Art. SRM Regulation; see also Binder, “Resolution planning and structural bank reform within the Banking Union” in Castaneda, Karamichailidou, Mayes and Wood (Eds.), European Banking Union: Prospects and Challenges (Routledge, 2015), available at , at 26 et seq.

Bail-in

137

actual resolution has even begun, for instance by emphasizing the importance of the recovery and resolution planning phase and by influencing decisions on how to design the corporate structure of a banking group. Its application can be challenging and will require the existence of some pre-conditions. Despite such challenges it is this article’s view that bail-in is a huge achievement and a game-changing step forwards which can achieve its goals to a large extent. With regard to the objective of breaking the link between banks and sovereigns, it is clear that bail-in will diminish these ties to a great extent and will thus contribute to reducing the costs of bank failures. However: the existence of bail-in rules alone will not make recourse to government financed bail-outs totally obsolete. There are several reasons for this. Since national resolution authorities lack practical experience in applying bail-ins, they may try to apply it to the minimum extent necessary to be able to use the resolution financing arrangement. However, the financing arrangement’s contribution may, in principle, not exceed a ceiling of 5 percent of the total liabilities including own funds of the bank under resolution, so that losses will be absorbed through the financing arrangement only to a certain extent, leaving resolution authorities with the choice to further extend bail-in and/or ask for government money to finance a bail-out.195 There might also be situations of systemic failures where the application of bail-in might exacerbate and further spread the problem through the financial system. In view of these circumstances, the BRRD itself allows in its Articles 56–58 BRRD to provide extraordinary public financial support to a bank under resolution if it is the last resort and in line with the EU State aid framework. 7.

Conclusions

The mandatory introduction of bail-in in the EU as of 1 January 2016 marks a fundamental change in the way a crisis of a bank is dealt with. This article has shown the complexity of the EU legal framework into which bail-in rules are embedded in the Banking Union, which is in particular composed of the closely intertwined BRRD and the SRM Regulation. While the prospect of bail-ins will certainly impact decisively on the financing of banks and influence decisions on how to design the corporate structure of a banking group, the answer to the question whether or not bail-in can achieve its goals and will effectively deliver on them, will depend on several factors. Two of the most important parameters to make bail-in effective 195. Art. 44(7) BRRD and Art. 27(9) SRM Regulation. This ceiling may, however, exceptionally be exceeded, where all unsecured, non-preferred liabilities, other than eligible deposits, were written down or converted.

138

Wojcik

CML Rev. 2016

and thus contribute to the success of the Banking Union as a whole are its predictability and the stability of the legal result produced by a bail-in. This article suggests that, while the legal result of a bail-in can probably be maintained, the application of bail-in in a way which is legally sound and compatible with fundamental rights, such as the right to property, is challenging, both in legal and practical terms. This is due to complexities, which global systemically important banks, with business in jurisdictions all over the world, pose and due to the ensuing risks of contagion. Furthermore, challenges result from the difficulties of determining the adequate amount of bail-in through a valuation to be carried out under immense urgency and in particular from the fact that harmonization of substantive insolvency law, including the hierarchy of claims, in the EU is almost non-existent, thus making the application of the crucial “no creditor worse off ” principle extremely difficult. Therefore, this article suggests that thorough preparation through recovery and resolution planning and appropriate transparency vis-à-vis the markets in order to create predictability will become essential in ensuring the success of bail-in. Finally, this article comes to the conclusion that, although bail-in will alleviate the burden on taxpayers substantially, it will not make public interventions obsolete. Bail-in is a game changer and a cornerstone of the Banking Union. Like any new regulatory tool, it will prove its utility only in real life and in the course of time.