A primeira Revista de Direito Público Online em Portugal

O bail-in enquanto novo paradigma para a resolução bancária: discricionariedade e dever de cuidado

Bail-in as new paradigm for bank resolution: discretion and the duty of care


1. The shift from bail-out to bail-in

The public outcry against bail-outs of banks, as a direct consequence of the financial crisis, brought to the international regulatory agenda the need to pursue a different avenue without taxpayers’ money being involved as a rule. Under this approach, State funds should only be injected in ailing banks in very exceptional circumstances, notably related to systemic crisis scenarios. The new paradigm for bank resolution rests on private creditors being bailed-in, and banks required to prepare ‘living wills’ so as to streamline their resolvability and facilitate the orderly winding down of the impaired parts of resolved banks. The ‘new normal’ for dealing with banking crisis is thus rooted on a ‘no fiscal cost’ commitment anchored on an ex-ante burden imposed on shareholders and the various classes of creditors to finance resolution of banks.

The shift to a bail-in model, in contrast to the initial response shaped by bail-outs, is foremost a political choice – and arguably an understandable one. Empirical data on the banking public recapitalisations between 2007 and 2009 helps put in context the substantial financial resources deployed to cater for the capital shortfalls of banks. For the sake of conciseness, the data provided below will be restricted to Europe.

One commonality among the 2007-09 banking crises is that they mostly affected advanced economies with large internationally integrated financial institutions that were deemed too large or interconnected to fail. The large international networks and cross-border exposures of these financial institutions fuelled contagion to other countries. Failure of any of these large financial institutions could have resulted in the collapse of other systemically important institutions, either directly by imposing large losses through counterparty exposures or indirectly by triggering a panic that could generate bank runs. These exposures prompted large-scale government interventions in the financial sector,1 and in particular massive public sector recapitalizations.

Tackling solvency problems with public money tends to elicit intense political debates, thus making policymakers to often delay direct injection of public funds and prolong the use of liquidity support and guarantees in the hope that problems in the banking sector subside. In contrast, bank recapitalizations in the timeframe 2007-09,2 although not more common than in past crisis episodes, were more speedy by comparison with previous trends.3 The direct fiscal costs associated with financial sector recapitalization during that period surpassed the 5% GDP threshold in Iceland (13%), the Netherlands (12,5%), United Kingdom (9%), Ireland (7,5%) and Luxembourg (7,5%),4 which clearly shows that Europe – and in particular its most advanced economies 5 – have footed a severe budgetary bill.6 It is worth noticing that the increase in public debt was quite spectacular, rising more than 50% in three European countries: Iceland (80%), Ireland (50%) and Latvia (50%).7

Albeit encompassing the direct fiscal outlays that supported aid to the financial sector, the surge in public debt is broadly associated with fiscal stimulus packages.8 This shows that high-income countries complemented their public recapitalization measures with expansionary budgetary policies to deal with their banking crises, thus leading to the conclusion that taxpayer money also indirectly supported the financial sector by stimulating aggregate demand, in turn propping up loan demand and lowering the risk of loan defaults. 9 In any case, and in spite of State ample backing, output losses were considerable, and hit mostly severely those countries with larger financial systems, especially those that experienced rapid expansion before the crisis (such as Iceland, Ireland and Latvia).

The reliance of the banking sector on taxpayers’ money at the initial stage of the financial crisis heated the discussion on moral hazard and unscrupulous behaviours, and shifted the focus to bail-in instead of bail-outs. This was a much controverted issue at the time of the negotiation of the Irish financial and economic assistance program.10 It is well known that the Irish economy and financial system were sternly affected by the financial crisis due to an overleveraged and oversized banking system, which collapsed after the bursting of the property bubble. The dramatic end of the property boom, fuelled by domestic and cross-border banking credit, resulted in losses on large commercial real estate loans of over 50%.11 The banking crisis in the second half of 2008 focused on two institutions, the Anglo-Irish Bank and the Irish Nationwide Building Society. From the start, there was uncertainty and lack of agreement whether the problems of other Irish banks were remotely as severe.12 In an attempt to quell a nascent panic, the Irish government in late 2008 pledged to guarantee all the banks’ liabilities, including senior bonds, for a period of 2 years.13

Whilst the Irish blanket guarantee was in force, the Irish government provided up to €64 bn. to restore the capital base of the Irish banking system (amounting to 40% of GDP). Besides moral hazard problems, the sovereign unlimited backing nurtured what became the nexus of the crisis, the fatal feedback loop between bank and sovereign creditworthiness. Ireland, whose government debt-to-GDP ratio in 2007 was one of the lowest in Europe, became by 2010 the third-most-indebted country in the euro area after Greece and Italy.14

When the guarantee expired at the end of 2010, the European Central Bank (ECB) exerted pressure for Ireland to negotiate a Troika program. Letters released by the ECB in November 2014 confirm that the ECB forced Ireland to apply for aid.15 In the course of the negotiations of the Troika program, the Irish government attempted to force the unguaranteed holders of debt securities issued by banks to accept haircuts. Put differently, the Irish authorities proposed a bail-in of bank bondholders so as to reduce the fiscal cost of rescuing the banking system. The ECB seems to have opposed fiercely any burden-sharing involving bondholders on grounds of the negative impact this would have had over financial stability in Ireland and potential spill-over effects on the banking systems of other European countries – Eichengreen16 and Pisani-Ferry17 refer to recounts testifying Trichet’s fears linked respectively to the French and German banks’ significant exposure to Irish paper, and to contagion coupled with deteriorating risk perception of the euro area.

In 2012, a member at the time of the ECB’s Executive Board referred to this issue in a public speech in Dublin. More concretely, Jörg Asmussen voiced the ECB’s concerns back in late 2010 under the following terms (emphasis ours):18

I know that the decisions concerning the repayment of bondholders in the former Anglo Irish Bank have been a source of controversy. Decisions taken by the Irish authorities such as these are not taken lightly. And the consequences of subsequent actions are weighed carefully. It is true that the ECB viewed it as the least damaging course to fully honour the outstanding senior debts of Anglo. However unpopular that may now seem, this assessment was made at a time of extraordinary stresses in financial markets and great uncertainty. Protecting the hard-won gains and credibility from the early successes in 2011 was also a key consideration, to ensure no negative effects spilled-over to other Irish banks or to banks in other European Countries. Determined action and a willingness to take tough, even controversial decisions, has placed Ireland’s financial system on a steadier footing.

In a nutshell, the ECB’s stance at the time of the Troika program was opposed to bail-in of creditors of banks, and notably senior bondholders, on grounds of concerns with financial stability of the euro area.

Curiously enough, another member of the ECB’s Executive Board highlighted in 2013 the virtues of bail-in as a means to mitigate systemic risk. Benoît Cœuré flagged the merits as follows (emphasis ours):19

By redistributing the risk between tax-payers, depositors and debt holders, bail-in rules can reduce the overall risk in the system. The corporate finance literature has pointed out the role of secured bank debt as an equilibrium response to the agency problems of equity financing. This highlights the risk of a “race to the bottom” towards a liability structure that would leave bank managers unchecked. It is therefore crucial that all sources of bank financing are subject to adequate monitoring.

It is curious to note how the early opposition of the ECB to bail-in, based on concerns with financial stability, was converted into the defence of that tool as a means to foster effective resolution systems and to properly realign incentives towards risk. In short, from a risk to financial stability bail-in became a powerful instrument to deter excessive-risk taking and thus contain systemic risk.

Be that as it may, the changing mood of the ECB towards bail-in became aligned with the new paradigm that started gaining momentum in 2011 with the adoption by the Financial Stability Board (FSB) of the ‘Key Attributes of Effective Resolution Regimes for Financial Institutions’. 20 The approach of the G20 and the FSB to resolution was shaped by the concern to avoid the use of fiscal resources when dealing with financial crises. The Key Attributes require inter alia jurisdictions to ensure they have designated resolution authorities with a broad range of powers to intervene and resolve a financial institution that is no longer viable, including through transfers of business and creditor-financed recapitalisation (‘bail-in’ within resolution) that allocate losses to shareholders and unsecured and uninsured creditors in their order of seniority.

At European level, the new regulatory paradigm on resolution, ending the culture of bail-out and ushering in a culture of bail-in, was fully embraced with the adoption of the Banking Recovery and Resolution Directive (BRRD).21 The ECB’s Vice-President qualified this change in emphatic terms right after its adoption (emphasis ours):22

It represents a true paradigm change. As of 2016, in all resolution cases, the BRRD will require a bail-in of shareholders and creditors equal to at least 8% of total liabilities of a given bank, including own funds. Only after the 8% threshold can money from the resolution fund be used and for a maximum amount of 5% of total liabilities (including own funds) of the bank under resolution. Public money, either from national governments or from direct European recapitalisation of banks, can only be used at the very end of the process which, in practice, should happen exceedingly rarely . Bail-in of shareholders and creditors plus the use of the Resolution Fund should in most conceivable cases, be enough to ultimately cover for the losses incurred by the bank .

The “Government financial stabilisation tools" that the Directive introduces open the possibility of broader public interventions in the case of serious systemic risk situations. It remains however an instrument of last resort. We are still far from the initial plan of direct European recapitalisations regarding this important question but the goal of avoiding to overburden the sovereign with banks’ rescues has in practice been achieved.

This sets the tone for the remainder of this paper.23 Bail-in represents indeed a new regulatory paradigm for bank resolution.24 The strive for having banking crises without public money involved is matched by a wide margin of discretion being granted to resolution authorities to decide on the extent of stakeholders that are to be included in financing resolutions and thus subject to bail-in. The ample discretionary powers enjoyed by resolution authorities encompass also the ability to decide on exclusions. Although key to the effectiveness of of bail-in, it will be argued that the exercise of such discretionary powers must be carefully counterbalanced by a duty of care incumbent upon resolution authorities subject to bail-in galling short of its allure. Before entering into such analysis, it is worth recalling the main tenets of bail-in.


2. Bail-in tool: the cornerstone of the new resolution regime

The bail-in powers differ starkly from the waiver of property rights that underpinned the resolution model enshrined in the UK Banking Act of 2009, which was adopted in the aftermath of the failure of Northern Rock in 2007. At the core of this latter model is a series of mechanisms for waiving ordinary property rights to effect a transfer of the troubled firm (or its assets and liabilities) in return for a payment of compensation. 25 The UK Banking Act of 2009 is based on sweeping disapplications of ordinary property law so as to bring about transfers by operation of law. The powers extend to waiving contractual termination provisions, and to imposing obligations on the transferor entity in relation to the transferee on a post-transfer basis. In contrast, the bail-in tool established under the BRRD 26 empowers the resolution authorities to force a failing or failed bank’s shareholders and creditors – although in the latter case not all – to contribute to the financial cost of resolution through a write-down or conversion of their claims against the bank. The bail-in tool imposes the internalization of bank losses by allocating them primarily to shareholders and creditors. 27 As explained by Hüpkes, “bail-in may take the form of either (i) a ‘direct bail-in’, i.e. the write-down and/ or exchange of a legal entity’s unsecured debt for equity in the same legal entity, without closing the legal entity; or (ii) an ‘indirect bail-in’ or ‘bridge bail-in’, i.e. the exchange of a legal entity’s unsecured debt for equity in a newly established bridge institution to which all or a portion of the first legal entity’s assets have been transferred”.28

Conceptually, when applied to bank failures, the threat of ex post compensation through the write-down and conversion of own funds and eligible liabilities is intended to work as an ex-ante governance tool, deterring excessively risky conducts and generating incentives for sound and prudent management. 29 This leads Gardella to view the bail-in tool as not just remedial in nature but meant to achieve ex ante beneficial effects.30 More concretely, by extending write down effects not only to shareholders but also to creditors, it shifts to these latter a significant burden of monitoring the debtor’s creditworthiness. On the other hand, when compared with the voluntary approach to haircuts or write-downs underpinning reorganization measures typical of ordinary insolvency or restructuring procedures, the losses borne by creditors via the bail-in tool are imposed by the resolution authority on the assumption that is best placed to pursue the objectives to be achieved by a resolution decision. In contrast to ordinary insolvency or restructuring proceedings, the safeguard of private interests such as the recovery of creditors’ claims ranks only ancillary to the pursuance of public goals related to the preservation of financial stability and the continuity of critical functions coupled with the protection of public finances and covered depositors.

To achieve these objectives, the bail-in tool is construed to bypass or disregard basic assumptions and limitations of general insolvency regimes, marking a rather decisive departure from private law’s common assumptions regarding the roles of shareholders and debt holders in a distressed business enterprise. Moulded to address the pivotal problem of resolution financing, the bail-in tool combines a distinctive mix of three different tenets: the internalization of costs by stakeholders; the mandatory nature of liquidation; and the radical financial restructuring of distressed companies. 31 As pertinently noted by Hadjiemmanuil, whilst all three elements may be found in general corporate insolvency law and/or pre-insolvency reorganization proceedings, the second and the third do not generally appear in combination.32 In this respect, the same author stresses that whilst the loss-absorption function through mandatory write-down or conversion of claims is reminiscent of liquidation procedures, when viewed from the perspective of reorganization proceedings bears another function: it is designed to neutralize the veto power arising from the strategic behaviour of stakeholders (shareholders, debt holders and, to a certain extent, board members, especially those entrusted with executive powers) of failed banks in bail-out scenarios. In such situations, those actors have an incentive to hold out by vetoing, or at least obstructing the proposed reorganization plans so as to receive some benefit from the State support that ultimately will improve their position substantially or leave unaffected their patrimonial interests. 33 The compulsory nature of bail-in coupled with it being administration-based – as opposed to a judiciary-driven process – removes the veto power of stakeholders.

Notwithstanding being a loss-absorption mechanism that serves to impose losses on certain classes of stakeholders, the purview of the bail-in tool encapsulates a vaster purpose than remedying the shortfall directly associated with the failed bank. It is a tool designed to restore viability to a going-concern entity that will emerge from the ruins of the failed entity. To ensure business continuity and restore the confidence of the market, the nascent institution or the institution acquiring partly its business, according to the resolution modality chosen by the resolution authority, have to be sustained by confortable capital buffers. This means that bail-in not only serves the purpose of loss-absorption, but is also a means of recapitalization either via write-down or conversion.34

The formidable array of powers vested with resolution authorities, when resorting to bail-in, is a manifestation of an almost unlimited ability of depriving private interests and claims of patrimonial meaning for financing resolutions and attaining public goals related to systemic stability and State budgetary protection. One can hardly find a parallel in Western public law on administrative authorities being lawfully entitled to exercise powers of an expropriatory nature without an upfront right to due compensation being recognized to the holders of such claims. It is true that bail-in is not assimilated to an expropriation; it works rather as a burden-sharing mechanism operating on the basis of an obligatory financing cascade serving not a merely a loss-absorption purpose but also the recapitalization of the entity that emerges on a going-concern scenario. Seen from this angle, bail-in may impose a burden on the affected shareholders, but in particular on creditors, that they would not otherwise incur in a liquidation scenario. In this sense, it is argued that bail-in might entail an expropriation effect.

The BRRD mitigates the ablative reach of patrimonial loss to shareholders and creditors by establishing the so-called ‘no creditor worse off’ principle, which requires that no creditor shall incur greater losses than would have been incurred if the ailing bank would have been wound up under normal insolvency proceedings.35 It remains to be seen whether this compensation is to be assimilated to the right of compensation that a private party is entitled to in case its patrimonial rights are affected by means of expropriation or mandatory writing-down by whatever form. Be that as it may, it is the expression that resolution bears a genetic link to insolvency proceedings, and thus creditors should not be treated in a more detrimental way than under these latter proceedings. For the BRRD requires that the winding-up of a failing institution through normal insolvency proceedings should always be considered before resolution tools are applied.36

The ‘no creditor worse off’ principle appears thus justified under the BRRD as a means of granting protection to property rights as guaranteed by Article 52 of the EU Charter of Fundamental Rights (Charter). Problems might arise however by the leeway that the BRRD bestows on resolution authorities to differentiate between creditors of the same class by allowing those authorities to treat them in an equitable manner 37 – and not strictly on equal terms.


3. Discretion and exclusions under bail-in powers

The view sustained by the European Court of Justice (ECJ) regarding judicial scrutiny of exercise of discretionary powers by administrative authorities tends not be of high intensity. In contrast with a comprehensive court control, such as that prevailing in Germany, where such control is preferred to a freedom of decision and arrangement on the part of the administration as deduced from the separation of powers, the ECJ tends to adopt a more restrained approach in revising discretionary decisions. This ‘low’ approach, which seems to be based on the recognition of ‘administrative assessment leeway’ when revising complex economic and technical evaluations, only limited to manifest errors and cases of arbitrariness, tends to be compensated in the ECJ’s jurisdiction by the intensive court control attached to procedural guarantees.38 The ruling issued by the ECJ in Technische Universität München39 set a landmark on a range of protective procedural standards in the face of administrative discretion. It is interesting to note that, under said preliminary ruling, the Bundesfinanzhof questioned the ECJ whether the up till then fairly restrictive approach involving the assessment of complex technical issues was compatible with the constitutional principle guaranteeing effective legal protection as recognized by Community law – implying a fortiori that the more difficult the technical questions to be decided the more immune from challenge administrative decisions would be. 40 In the face of this legal test, the Court explicitly extrapolated the doctrinal concept of ‘good administration’, which justifies the imposition of a duty of care as counterbalance to discretion, to similar procedural principles such as the giving reasons requirement and the right to be heard.41 Based on this ECJ approach, Nehl emphatically suggests that, albeit not clearly expressed in the judgement, ‘(…) the ECJ felt the relevance of the concept of care as a ground for review which would allow a more intensive degree of judicial scrutiny of the exercise of administrative discretion’ .42 These implications are, according to the same author, encapsulated in the following paragraph of the ruling (emphasis ours):43

(…), where the Community institutions have such a power of appraisal, respect for the rights guaranteed by the Community legal order in administrative procedures is of even more fundamental importance. Those guarantees include, in particular, the duty of the competent institution to examine carefully and impartially all the relevant aspects of the individual case, the right of the person concerned to make his views known and to have an adequately reasoned decision. Only in this way can the Court verify whether the factual and legal elements upon which the exercise of the power of appraisal depends were present.

The findings by the Court in TU München – which by the way lead to the annulment of the Commission decision on grounds of an array of procedural flaws related to the duty of care, the right to be heard and the giving reasons requirement – deliver a very fierce benchmark as regards the exercise of discretionary powers that are vested in administrative authorities by EU legislation: the wider the margin of appraisal, the more stringent must be the observance of the duty of care (diligence) translated in the careful and impartial examination of all relevant aspects of the individual case. In this respect, it is worth stressing that the duty of care, together with the right to be heard, the obligation of adequate reasoning of a decision and the public right of access to documents, integrates the principle of good administration,44 which is currently acknowledged as a right under Article 41 of the Charter in relation to the action of institutions, bodies, offices and agencies of the Union.

It is argued that the standard of care inaugurated by the ECJ in TU München may not be disregarded by resolution authorities in the exercise of bail-in powers, in spite of the situations of extreme urgency dealt with. Recital 89 of the BRRD mirrors this concern in the following terms (emphasis ours):

Crisis management measures taken by national resolution authorities may require complex economic assessments and a large margin of discretion. The national resolution authorities are specifically equipped with the expertise needed for making those assessments and for determining the appropriate use of the margin of discretion . Therefore, it is important to ensure that the complex economic assessments made by national resolution authorities in that context are used as a basis by national courts when reviewing the crisis management measures concerned. However, the complex nature of those assessments should not prevent national courts from examining whether the evidence relied on by the resolution authority is factually accurate, reliable and consistent, whether that evidence contains all relevant information which should be taken into account in order to assess a complex situation and whether it is capable of substantiating the conclusions drawn therefrom .

Interestingly, the EU lawmakers have not felt the need at this instance to quote the right to good administration as laid down in Article 41 of the Charter in contrast to other instances of the preamble where explicit references to Article 16 (freedom to conduct a business),45 Article 47 (right to an effective remedy and to a fair trial),46 and Article 52 (scope and interpretation of rights and principles) 47 of the Charter are made, or otherwise to the right to property, the right to an effective remedy and to a fair trial and the right of defence as enshrined in the Charter, without citing the specific legal provisions thereof. 48 It is true that the BRRD is premised on requiring that any interference with rights of shareholders and creditors which results from resolution action should be compatible with the Charter, in particular where creditors within the same class are treated differently in the context of resolution action – which distinctions should be justified in the public interest and proportionate to the risks being addressed and should be neither directly nor indirectly discriminatory on the grounds of nationality. 49 Under this reading, the lack of any reference to the right to good administration as laid down in Article 41 of the Charter, in contrast to the explicit references to other provisions of the Charter throughout the preamble of the BRRD, would be compensated by the general statement contained in its recital 13.

The reason for not referring to Article 41 of the Charter may be rooted in the ratione personae ambit of that provision. For its paragraph 1 restricts its scope to ‘institutions, bodies, offices and agencies of the Union’, whilst paragraphs 3 and 4 only speak of ‘institutions’ and ‘servants in the performance of their duties’ respectively. Considering that the BRRD is addressed to Member States and in particular vests resolution powers on national resolution authorities, Article 41 would not apply. This however is a reading that entails a fragmented approach to resolution, which ignores its intrinsic link to prudential supervision and the broader placement of resolution within the Banking Union. To be sure, resolution is initiated when a supervisory authority, after consulting a resolution authority, determines that an institution is failing or likely to fail. 50 Exceptionally, the BRRD offers Member States the possibility that, in addition to the supervisory authority, the determination that the institution is failing or likely to fail can be made also by the resolution authority, after consulting the former. 51 In any case, resolution is triggered by a decision of the supervisory authorities, or after consultation of these latter, on the failure or likely failure of a bank. It must be noted that prudential supervision today is exercised, in the case of euro area Member States, at EU level through the Single Supervisory Mechanism (SSM). Indeed, Article 6(1) of Council Regulation (EU) No 1024/2013 (SSM Regulation)52 stipulates that the ECB shall carry out its tasks within a single supervisory mechanism composed of the ECB and national competent authorities (NCAs), and that it shall be responsible for the effective and consistent functioning of the SSM. Irrespective of the division between significant credit institutions, which are directly supervised by the ECB, 53 and the less significant, which direct supervision is retained by NCAs subject to the oversight of the ECB,54 prudential supervision is exercised within the SSM by means of a composite procedure in which components of one single administrative procedure are conducted by administrative authorities from different jurisdictions – both from national and European levels each using different procedural rules.55 This means that both the ECB and the NCAs exercise their supervisory powers squarely within the scope of EU law, thus requiring these latter also to abide inter alia by Article 41 of the Charter – even if a prima facie reading of this provision would seem only to oblige the ECB.

Although the resolution framework is autonomously laid down in the BRRD, it is argued that the umbilical connection to the supervisory assessment on failing or likely to fail situations makes resolution powers fall within the scope of EU law, thus imposing on resolution authorities the duty to act pursuant to good administration as established in Article 41 of the Charter. The ECJ confirmed a similar broader reading in Ākerberg by referring that ‘since the fundamental rights guaranteed by the Charter must therefore be complied with where national legislation falls within the scope of European Union law, situations cannot exist which are covered in that way by European Union law without those fundamental rights being applicable. The applicability of European Union law entails applicability of the fundamental rights guaranteed by the Charter.’56

Against this backdrop, and notwithstanding the BRRD’s silence on the right to good administration, the ample discretionary powers enjoyed by the resolution authorities cannot depart from the duty of care as a fundamental tenet of that right. As Nehl points out, an examination of the national public laws shows that an essential justification for the imposition of this latter duty on the administration is to act as counterbalance to administrative discretion. 57 The concern with the settlement of a procedural framework regulating the administrative decision-making is usually associated, in historical terms, with a precise constitutional matrix: in a democratic constitutional framework every public decision-making must be submitted to a certain process.58 Moreover, the administrative procedure served to reconcile broad grants of administrative discretion typical of the post-war administration with the growing distrust on bureaucratic/expertise power and the preservation of pluralistic values enshrining post-war societies.59 In the design of such a procedural framework a reasonable degree of care by the administration, in particular in relation to individuals who are likely to be adversely affected by the outcome of administrative decision-making, constitutes the basic expression of any modern legal order subjected to the rule of law. The right of good administration, either construed as a general principle or as subjective right, is anchored in the constitutional traditions common to Member States. The resolution authorities have thus to abide, when basing its resolution actions on the wide margin of discretion granted by the BRRD, on the standard of care flowing from the right to good administration laid down in Article 41 of the Charter.60

In this respect, the implications of some prominent case law by the European courts dealing with the extent to which the principle of good administration confers rights upon individuals must also be taken into account. In Nölle, the Court of First Instance overtly stated that the protective nature of the principle of care cannot be questioned,61 thus acknowledging the aptitude of such principle to confer subjective rights on which the individuals could rely against the administration. In Tillack, the General Court held that subjective rights may arise from the principle of sound administration only insofar as it ‘constitutes the expression of specific rights such as the right to have affairs handled impartially, fairly and within a reasonable time, the right to be heard, the right to have access to files, or the obligation to give reasons for decisions, for the purposes of Article 41 of the Charter of fundamental rights of the European Union’. 62 In the SPM case, the Court stated that the right to good administration confers subjective rights on individuals when it constitutes the expression of specific rights such as, inter alia, the right of an individual to have his or her affairs handled impartially, fairly and within a reasonable time or his or her right to be heard.63

This string of case law is important to understand the place of the duty of care (as an emanation of the right of good administration) within the dichotomy of rights and freedoms enshrined in Article 52 of the Charter.64 In this respect, it is worth quoting the stance and associated reasoning defended by Hofman and Mihaescu (emphasis ours):65

The reason for the distinction between rights/freedoms in Article 52(1) and principles in 52(5) CFR is that the latter merely constitute programmatic objectives which have to or might be implemented, being therefore incapable to confer subjective rights on individuals. While the umbrella notion of good administration is a principle under the concept of 52(5) CFR, the sub-concepts may be rights or freedoms under Article 52(1) CFR. Its umbrella notion is made concrete by the constitutional legislature in Article 41 CFR, by the ordinary legislature in specific legislation and by courts applying general principles of EU law.

The identification of subjective individual rights, in this view, is not the result of any abstract dogmatic classification. Instead subjective rights, as has always been acknowledged since the early case-law of the Court of Justice, could be identified by the simple question as to whether the definition of the principle or its sub-component is sufficiently precise or with other words, ‘clear and unconditional.’ Where that is not the case, the principle requires specification. This has been the requirement of the courts regarding general principles of law and is now also codified in Article 52(5) CFR. This approach should, in our view, guide the definition of whether good administration is a subjective right of individuals or whether it merely is an objective principle – irrespective of whether a specific sub-notion of good administration is explicitly referred to in Article 41 CFR or not. Using the question of whether a right or principle is sufficiently ‘clear and unconditional’ as to be capable of directly conferring rights on individuals or not saves searching for difficult-to-define notions of legislative intention in creating a right.

On this basis, the same authors attach a subjective nature to the duty of care in such convincing terms that are again worth citing (emphasis ours): 66

On the basis of the approach to distinguish subjective rights which are characterized by their ‘clear and unconditional’ content and objective principles which arise as obligations of the administration only due to their lack of clarity or conditionality (i.e., relying on further acts or sub-principles to make them appear sufficiently clear and unconditional) we might differ from these opinions. The obligation of diligent and impartial investigation in the form of the Court’s interpretation of the duty of care is interpreted as the obligation to undertake a full and impartial assessment of all relevant facts of the case prior to taking a decision . This is an obligation which in our view is sufficiently clear and precise to be interpreted also to contain a right of an individual to claim such investigation. This approach is in line with the Court, which found in the leading case on the duty of care as a sub-component of the general principle of good administration that there are elements of the principle of good administration which, although having an ‘objective’ nature, also confer ‘subjective’ rights on individuals.

The duty of care incumbent upon resolution authorities when exercising bail-in powers is particularly relevant in situations of discretionary exclusions. Further to mandatory exclusions contained in Article 44(2) of the BRRD, paragraph 3 of the very same provision enables resolution authorities, in exceptional circumstances, to exclude fully or partially certain liabilities from the application of the write-down or conversion powers. Such discretionary exclusion may be exercised on grounds of the following situations:

(a) it is not possible to bail-in a liability within a reasonable time notwithstanding the good faith efforts of the resolution authority;

(b) the exclusion is strictly necessary and is proportionate to achieve the continuity of critical functions and core business lines in a manner that maintains the ability of the institution under resolution to continue key operations, services and transactions;

(c) the exclusion is strictly necessary and proportionate to avoid giving rise to widespread contagion, in particular as regards eligible deposits held by natural persons and micro, small and medium sized enterprises, which would severely disrupt the functioning of financial markets, including of financial market infrastructures, in a manner that could cause a serious disturbance to the economy of a Member State or of the Union; or

(d) 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.

In the event of a discretionary exclusion on such grounds, the level of write-down or conversion applied to other eligible liabilities may be increased, subject to the ‘no creditor worse off’ principle.67 As noted by Hadjiemmanuil, ‘it is precisely because of exclusions (whether mandatory or discretionary) that a need for external resolution financing may arise. The exclusions deplete the pool of resources available for loss absorption and recapitalization, while leaving an increased volume of liabilities behind. If all capital instruments and liabilities were fully subject to bail-in in accordance to their ranking, there would never be a need for external financing, because the pool of liabilities would be allowed to disappear to the degree necessary for ensuring a full absorption and internalization of losses. Liability holders would be treated in roughly the same way as in a normal liquidation’.68

The resort to discretionary exclusions shows clearly how bail-in might fall short of what it promised to deliver, since the exercise of such discretion might run counter to the problem of resolution financing. If a significant funding gap is to persist as the outcome of the exercise of such discretionary exclusions, recourse to the resolution fund and ultima ratio to the State might be needed. Equally grave, if not more, is the impact the exercise of discretionary exclusions might have on the different class of creditors. Firstly, creditors ranking in a different class might be required to suffer a higher loss to compensate for those that were excluded. This overcompensation is most likely prone to distort the level of burden-sharing. Secondly, discretionary exclusions might affect in differentiated terms creditors placed in the same class, thus hampering the principle of pari passu satisfaction within the same class. As referred above, the BRRD requires the equitable, and not equal, treatment of creditors of the same class. This said, pari passu is built-in under the BRRD as a fundamental constraint on the discretionary powers by resolution authorities when delimiting the bail-in circle.69

By nature, the exclusion of certain claims worsens the position of those left behind in a reduced pool of bail-inable liabilities, which hampers both the pari passu principle and the recovery maximization objective of insolvency law.70 This is particularly harmful when resulting from the applicability of discretionary exclusions, since those affected were most likely not provided with ex ante information that could have been relevant for their investment decision-making. Moreover, discretionary exclusion might go as far as to have senior claims written down or converted before junior ones have been extinguished completely. This opens a terrain of uncertainty that clearly clouds investment decision-making.

On the other hand, creditors within the same class have an expectation to be treated in the same manner, and not be differentiated in spite of the equitable clause enshrined in the BRRD coupled with the ‘no creditor worse off’ principle. This the more so in the case of debt holders that have placed their investments before the entry into force of the BRRD. A recent speech by the Governor of Banca d’Italia addressed in vocal terms the concern of insufficient attention having been given to the need of having an adequate transition period before the full-fledged implementation of the BRRD: 71

During this delicate changeover at European level, insufficient attention was given to the transition period. At the technical meetings that laid the ground for the BRRD, the Ministry of Economy and Finance and the Bank of Italy argued, without mustering the necessary support, that an immediate and, more importantly, retroactive enforcement of the burden-sharing mechanisms to 2015, and subsequently of the bail-in, could – in addition to a costlier and less plentiful supply of credit to the economy – have posed risks to financial stability also in relation to the treatment of creditors who had subscribed bank liabilities many years ago, at a time when the possibility of losing the original investment was very remote. Our views were expressed in the Bank of Italy’s official publications. A gradual, less abrupt transition would have been preferable. This would have enabled investors to be fully acquainted with the new rules and to adapt their choices to the new regulatory environment. A targeted approach, with the application of the bail-in to only those financial instruments with a specific contractual clause to that effect, and a sufficient transition phase, would have allowed banks to issue new liabilities subject to express bail-in conditions.

Still on the differentiation between creditors of the same class, the recent decision taken by Banco de Portugal on imposing losses on certain senior bondholders of Novo Banco, the bridge bank resulting from the resolution of Banco Esp ito Santones hav ebeen ively exd from Novo Banco to BES akenng been given to the transtion fore junior ones hav ebeen ively exírito Santo,72 has sparked a chorus of disapproval amongst institutional investors.73 This decision, although not qualified by the Portuguese resolution authority as a bail-in but rather a retransmission of liabilities to the bad bank,74 amounts in substance to a loss-absorption measure imposed on certain qualified investors by means of ‘cherry-picking’ within the same class with the aim of solving the capital shortfall of Novo Banco.75 The Banco de Portugal’s decisions prompted the IMF to consider, in its last Global Financial Stability Report, that the treatment of select senior debt holders of Novo Banco has led to a perception of uneven handedness and increased uncertainty that has dented confidence.76 This authoritative view certainly triggers the question whether said decision lives up to what should be one of the core goals of a resolution measure, notably to safeguard financial stability. A consideration that is reminiscent of the concerns that led the ECB, as seen above, to resist fiercely the bail-in of senior creditors in the Irish case. This shows that bail-in, although raised to new paradigm of the resolution edifice, might face restrictions of its own as regards the effective fulfilment of the objectives it is intended to pursue. This limitation thus manifestly reinforces the need for discretion powers exercised by resolution authorities, when resorting to the bail-in tool, to be scrutinised against the duty of care.


4. Conclusion

The set of concerns on discretionary differentiation and exclusions related to bail-in, which were only flagged above in quite perfunctory terms, makes the case for the duty of care to be effectively observed by the resolution authorities. As said, this duty amounts to a true counterbalance to the wide margin of discretion that resolution authorities are granted under the BRRD, and in particular for implementing the bail-in tool. The ‘no creditor worse off’ may not act as a palliative to enable the resolution authorities to make recourse in particular to discretionary exclusions on bail-ins e.g. if the ‘exceptional circumstances’ referred to in Article 44(3) are not verifiable. Resolution authorities are indeed best placed to make complex assessments on financial stability and continuity of critical functions coupled with the likely impacts in terms of contagion when implementing resolution decisions. In a resolution scenario, the exercise of discretionary exclusions might indeed be the proper means to address the risks and threats that resolution authorities have identified in their ratio decidendi, and which circumstances of exceptional natural might fully justify such exclusions. But the wide discretion on the assessment of the facts related to resolution scenarios must be duly compensated by the duty to investigate properly the case in the interest of those that are likely to bear a higher burden-sharing than would result from the application of insolvency law priority rules, or that will be singled-out in deviation of pari passu rules. In short, discretion enjoyed by resolution authorities is to be exercised in deference to the protective nature of Article 41 of the Charter. For the duty of care constitutes the justiciable requirement to collect impartially and completely the relevant facts forming the basis of the discretionary decision. Ultimately it is the expression that resolution is subject to the rule of law.


*Instituto dos Valores Mobiliários, Faculdade de Direito de Lisboa ; machado71@icloud.com.

1 Laeven, Luc/Valencia, Fábian (2014) ‘Resolution of Banking Crises: The Good, the Bad, and the Ugly’ in Claessens, Stijn et al., Financial crises: causes, consequences, and policy responses. IMF, Washington DC, pp. 404-405.

2 As per the approach proposed by Laeven/Valencia, for bank recapitalizations only ‘comprehensive’ recapitalization packages in which public funds were used are considered, thereby excluding ad hoc interventions. From 2007 to 2009, three recapitalization programs targeted specific banks: Iceland (three largest banks), Luxembourg (Fortis and Dexia) and Latvia (Parex). Laeven, Luc/Valencia, Fábian (2014) ‘Resolution of Banking Crises: The Good, the Bad, and the Ugly’, p. 410, ft. 17.

3 Laeven, Luc / Valencia, Fábian (2014) ‘Resolution of Banking Crises: The Good, the Bad, and the Ugly’, pp. 410-411.

4 Laeven, Luc/Valencia, Fabián (2008), Systemic Banking Crises: A New Database, IMF Working Paper 08/224, IMF, Washington DC, available at https://www.imf.org/external/pubs/ft/wp/2008/wp08224.pdf.

5 During this period, the less advanced peripheral economies were more modestly impacted: Greece (4%), Spain (2%), Portugal (0,1%).

6 For the purposes of this article, the costs with contingent liabilities are excluded from the analysis. For further information on the sizeable volume issued during the period in reference, see Appendix Table 13A.3 in Laeven, Luc / Valencia, Fábian (2014) ‘Resolution of Banking Crises: The Good, the Bad, and the Ugly’, pp. 424-426.

7 Laeven, Luc/Valencia, Fabián (2008), Systemic Banking Crises: A New Database.

8 This explains why the impact on less advanced peripheral economies was stronger when compared with direct fiscal costs associated with public banking recapitalizations: Greece (30%), Spain (40%) and Portugal (25%).

9 Laeven, Luc/Valencia, Fabián (2008), Systemic Banking Crises: A New Database, p. 30.

10 Hereinafter ‘Troika program’.

11 Schoenmaker, Dirk (2015) Stabilising and Healing the Irish Banking System: Policy Lessons, Paper prepared for the CBI-CEPR-IMF Conference, available at https://www.imf.org/external/np/seminars/eng/
, p. 2.

12 Eichengreen, Barry (2015), The Irish Crisis and the EU from a Distance. Paper prepared for the CBI-CEPR-IMF Conference, available at https://www.imf.org/external/np/seminars/eng/2014/ireland/pdf/
, p. 4.

13 Pisani-Ferry, Jean (2013), The Euro Crisis and its Aftermath, Oxford, p. 90.

14 Pisani-Ferry, Jean (2013), The Euro Crisis and its Aftermath, p. 90.

15 The correspondence is available at http://www.ecb.europa.eu/press/html/irish-letters.en.html

16 Eichengreen, Barry (2015), The Irish Crisis and the EU from a Distance, p. 6.

17 Pisani-Ferry, Jean (2013), The Euro Crisis and its Aftermath, pp. 90-92.

18 Asmussen, Jörg (2012), The Irish case from an ECB perspective, available athttps://www.ecb.europa.eu

19 Cœuré, Benoît (2013) The implications of bail-in rules for bank activity and stability, available at https://www.ecb.europa.eu/press/key/date/2013/html/sp130930.en.html

20 Hereinafter ‘Key Attributes’. The FSB adopted the Key Attributes at its Plenary meeting in October 2011. The G20 Heads of States and Government subsequently endorsed the Key Attributes at the Cannes Summit in November 2011 as “a new international standard for resolution regimes”. The Key Attributes were updated in October 2014. The newly adopted guidance documents have been incorporated as annexes into the 2014 version of the Key Attributes document. No changes were made to the text of the twelve Key Attributes of October 2011. The twelve Key Attributes remain the umbrella standard for resolution regimes covering financial institutions of all types that could be systemic in failure. The updated version of the Key Attributes is available at http://www.fsb.org/wp-content/uploads/r_141015.pdf

21 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 and amending Council Directive 82/891/EEC, and Directives 2001/24/EC, 2002/47/EC, 2004/25/EC, 2005/56/EC, 2007/36/EC, 2011/35/EU, 2012/30/EU and 2013/36/EU, and Regulations (EU) No 1093/2010 and (EU) No 648/2012, of the European Parliament and of the Council, OJ L 173, 12.6.2014, p. 190–348.

22 Constâncio, Vitor (2014), Banking Union: meaning and implications for the future of banking, available at https://www.ecb.europa.eu/press/key/date/2014/html/sp140424_1.en.html

23 The Single Resolution Mechanism (SRM) falls outside the scope of this article. For information, the SRM legal framework was adopted by Regulation (EU) No 806/2014 of the European Parliament and of the Council 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) No 1093/2010, OJ L 225, 30.7.2014, p. 1–90.

24 The BRRD lists four resolution tools, which the resolution authorities may apply individually or in combination: the sale of business tool, the bridge institution tool, the asset separation tool and the bail-in tool. The asset separation tool may however only be used together with another resolution tool. Articles 37(3)–(5) BRRD.

25 Armour, John (2015) ‘Making Bank Resolution Credible’ in Moloney, Niamh et al., The Oxford Handbook of Financial Regulation. Oxford,p. 462.

26 Articles 43-55 BRRD.

27 For a thorough description of the main features of the bail-in tool, see Gardella, Anna (2015), ‘Bail-in and the Financing of Resolution within the SRM Framework’ in Busch, Danny/Ferrarini, Guido (eds.), The European Banking Union, Oxford, pp. 392 ss.

28 Hüpkes, Eva (2015) ‘Adequate loss-absorbing and Recapitalisation capacity of G-SIBs in resolution’ in ECB Legal Conference 2015 - From Monetary Union to Banking Union, on the way to Capital Markets Union: New opportunities for European integration , p. 200, available athttp://www.ecb.europa.eu/pub/pdf/other/frommonetaryuniontobankingunion201512.en.pdf

29 Gardella, Anna (2015) ‘Bail-in and the two dimensions oF burden-sharing’ in ECB Legal Conference 2015 - From Monetary Union to Banking Union, on the way to Capital Markets Union: New opportunities for European integration , p. 207, available at http://www.ecb.europa.eu/pub/pdf/other/frommonetaryuniontobankingunion201512.en.pdf

30 Idem.

31 Hadjiemmanuil, Christos (2015) ‘Bank Stakeholders’ Mandatory Contribution to Resolution Financing: Principle and Ambiguities of Bail-in’, pp. 232-233 in ECB Legal Conference 2015 - From Monetary Union to Banking Union, on the way to Capital Markets Union: New opportunities for European integration, p. 207, available at http://www.ecb.europa.eu/pub/pdf/other/frommonetaryuniontobankingunion201512.en.pdf

32 Idem, p. 233.

33 For further developments on this topic, see de Weijs, Rolef (2013) ‘Too Big to Fail as a Game of Chicken with the State: What Insolvency Law Theory Has to Say About TBTF and Vice Versa’, 14 European Business Organization Law Review, pp. 201–224.

34 We follow the approach proposed by Hadjiemmanuil of treating bail-in as a short form for the exercise of either the power of writing-down the claims or converting the claims or debt instruments into equity, albeit mindful of the fact that such power is treated separately from bail-in under Articles 59-62 BRRD. Such power is however regulated separately under the BRRD because of the possibility of write-down or conversion being exercised without placing the ailing bank in resolution. See Hadjiemmanuil (2015), p. 234.

35 Articles 34(1)(g) and 73-75.

36 Recital 46 BRRD.

37 Article 34(1)(f) BRRD.

38 Brenner, Michael (1997) ‘Administrative Judicial Protection in Europe: General Principles’,European Review of Public Law/Revue Europeène de Droit Public 9(3), pp. 614-615.

39 Case C-269/90, Technische Universität München v Hauptzollamt München-Mitte 1991 ECR I-5469.

40 The relevant transcription of the order for reference by the Bundesfinanzhof may be found in the Opinion of AG Jacobs, Case C-269/90, Technische Universität München v Hauptzollamt München-Mitte 1991 ECR I- 5469, para. 11. For a more detailed overview, see Nolte, G. (1994) ‘General Principles of German and European Administrative Law - A Comparison in Historical Perspective’, Modern Law Review 57, pp. 207 ss.

41 Nehl, Hanns-Peter (1999) Principles of Administrative Procedure in EC Law. Oxford, Hart, pp. 132-133.

42 Idem, p. 133.

43 Case C-269/90, para. 14.

44 Nehl (1999), pp. 27-37; Azoulaï, Loïc (2007) ‘Le principe de bonne administration’, in J.-B. Auby and J.Duteil de la Rochère (eds.) (2007) Droit administratif européen, Brussels, Bruylant, p. 489; Hofmann, Herwig C.H. (2007) ‘Good Administration in EU Law – A Fundamental Right?’, 13 Bulletin des Droits de l’Homme, pp. 44-52.

45 Recitals 24 and 29 BRRD.

46 Recital 88 BRRD.

47 Recitals 24 and 49 BRRD.

48 Recital 130 BRRD.

49 Recital 13 BRRD.

50 Article 32(1)(a) BRRD.

51 Article 32(2) BRRD.

52 Council Regulation (EU) No 1024/2013 of 15 October 2013 conferring specific tasks on the European Central Bank concerning policies relating to theprudential supervision of credit institutions (OJ L 287, 29.10.2013, p. 63).

53 Article 6(4) and (5) SSM Regulation.

54 Article 6(5) and (6) SSM Regulation.

55 For further references, Hofmann, Herwig H.C. (2009) ‘Decision-Making in EU Administrative Law – The Problem of Composite Procedures’, 61 Administrative Law Review, Special Edition, pp. 199-221.

56 Case C-617/10 Åklagaren v. Hans Åkerberg Fransson 2013 ECR I-nyr of 26 Feb. 2013, paras. 19-21.

57 Nehl (1999), p. 115.

58 Mannori, Luca / Sordi, Bernardo (2001) Storia del diritto amministrativo, Rome, Laterza, pp. 464-465.

59 Reich, Robert B. (1985) ‘Public Administration and Public Deliberation: An Interpretative Essay’, Yale Law Journal 94, pp. 1618-1619.

60 Defending that the absence of pre-intervention rights in resolution situations is to be ‘measured against the right to good administration guaranteed under Article 41 of the Charter’, Bornemann, Alexander (2014) ‘Resolution Regimes for Financial Institutions and the Rule of Law’ in Haentjens, Matthias/Wessels, Bob, Bank Recovery and Resolution – A Conference Book, The Hague, Eleven, p. 95.

61 Case T-167/94 Nölle v. Council and Commission 1995 ECR II-2589, paras. 69 and 76

62 Case T-193/04 Tillack v. Commission 2006 ECR II-3995, para. 127.

63 Case C-39/09 P SPM v. Council and Commission 2010 ECR I-38, paras. 65-66. See also Case T-128/05, SPM v. Council and Commission2008 ECR II-260, para. 127.

64 Article 52 of the Charter is worded as follows:
1. Any limitation on the exercise of the rights and freedoms recognised by this Charter must be provided for by law and respect the essence of those rights and freedoms. Subject to the principle of proportionality, limitations may be made only if they are necessary and genuinely meet objectives of general interest recognised by the Union or the need to protect the rights and freedoms of others.
2. Rights recognised by this Charter for which provision is made in the Treaties shall be exercised under the conditions and within the limits defined by those Treaties.
3. Insofar as this Charter contains rights which correspond to rights guaranteed by the Convention for the Protection of Human Rights and Fundamental Freedoms, the meaning and scope of those rights shall be the same as those laid down by the said Convention. This provision shall not prevent Union law providing more extensive protection.
4. In so far as this Charter recognises fundamental rights as they result from the constitutional traditions common to the Member States, those rights shall be interpreted in harmony with those traditions.
5. The provisions of this Charter which contain principles may be implemented by legislative and executive acts taken by institutions, bodies, offices and agencies of the Union, and by acts of Member States when they are implementing Union law, in the exercise of their respective powers. They shall be judicially cognisable only in the interpretation of such acts and in the ruling on their legality.
6. Full account shall be taken of national laws and practices as specified in this Charter.
7. The explanations drawn up as a way of providing guidance in the interpretation of this Charter shall be given due regard by the courts of the Union and of the Member States.

65 Hofmann, Herwig C.H./Mihaescu, Bucura C. (2013) ‘The Relation between the Charter's Fundamental Rights and the Unwritten General Principles of EU Law: Good Administration as the Test Case’, European Constitutional Law Review, 9, pp. 73-101, at 93-94 (footnotes omitted).

66 Idem, at 95.

67 Article 44(3), second subpara. BRRD.

68 Hadjiemmanuil (2015), p. 239.

69 Recital 77 BRRD.

70 Idem, p. 242.

71 Speech by Governor Ignazio Visco at the 22nd ASSIOM FOREX Congress, p. 8, available at https://www.bancaditalia.it/pubblicazioni/interventi-governatore/integov2016/en-Visco-300116.pdf?language_id=1 .

72 The press release related to the decision by Banco de Portugal, adopted on 29 December 2015, on certain bond issaunces that were retransmitted from Novo Banco to BES is available at https://www.bportugal.pt/en-US/OBancoeoEurosistema/ComunicadoseNotasdeInformacao/Pages/combp20151229-2.aspx

73 See e.g. ‘ECB under fire as Portugal hits Novo Banco bondholders’, FT, 7 January 2016.

74 Interestingly, the Governor of Banco de Portugal told in a recent interview, in replying to a query on the retransfer of senior debt from Novo Banco to BES, that ‘it was the first time the BRRD was applied’. See E-Revista do Expresso, 5 March, p. 58.

75 Although not mentioned by Banco de Portugal that the aim of the retransfer decision was to cater for the capital needs of Novo Banco, the Commission in its Country Report Portugal 2016 refers that ‘(…) the Portuguese banking system was impacted by the announcement of the Banco de Portugal's decision to transfer five selected senior bonds out of Novo Banco to remedy its capital shortfall identified under the European Central Bank's stress test of November 2015. The actions in relation to Novo Banco were taken to safeguard the stability of the financial system and to take away the uncertainty surrounding the bank's capital position in the previous sales process’. See 2016 Country Report for Portugal, p. 20, available at http://ec.europa.eu/europe2020/pdf/csr2016/cr2016_portugal_en.pdf

76 Global Financial Stability Report: Potent Policies for a Successful Normalization, April 2016, p. 36, available at https://www.imf.org/External/Pubs/FT/GFSR/2016/01/pdf/text_v2.pdf