2.1. FSB’s Key Attributes and TLAC

International community addressed the banking resolution issues (or rather non-existence of banking resolution scheme) already in 2011 and published Financial Stability Board’s (FSB’s) Key Attributes of Effective Resolution Regimes for Financial Institutions (KA, updated in 2014).^[16] This document not only acknowledged the need for special resolution regime for financial institutions, distinct from corporate insolvency law, but also set out a framework to be implemented by national regulators.

First, it stresses the need for designation of a resolution authority. Further, in addition to general resolution powers,^[17] the KA include detailed description of three main tools: transfer of assets and liabilities, bridge institution and the most innovative bail-in tool. The latter allows for debt to be written down or converted into equity. In other words, it forces shareholders and creditors to bear losses, it ‘bails them in’. As this procedure was expected to raise eyebrows from the outset, FSB also braced KA with safeguards, such as segregation of client assets or ‘no creditor worse off’ principle.

As the issue of contagion and systemic risk has become especially relevant since the crisis, the KA did not omit the aspect of global systemically important financial institutions (G-SIFIs). Cross-border cooperation between home and host resolution authorities, as well as specific requirement of recovery and resolution planning constitute two main improvements encompassed by the FSB’s document. However, the KA do not elaborate on the issue of bail-inable capital, that would ex-ante enhance credibility of resolution tools.

To address this shortcoming, the FSB issued the ‘Principles on Loss-absorbing and Recapitalisation Capacity of G-SIBs in Resolution’ along with its Total Loss Absorbing Capacity (TLAC) Term Sheet.^[18] Its goal is to establish a framework for systemically important banks, depriving them of their ‘too big to fail’ feature and ensuring public funds will not be used in resolution.^[19] Thus, TLAC should be set on firm-by-firm basis by national regulators, but taking into account the internationally recognised floor of 18% of the resolution group’s risk-weighted assets and 6.75% of the Basel leverage ratio denominator.^[20] Additionally, TLAC is to be held in addition to minimum capital requirements (and the buffers)^[21] and TLAC liabilities need to fulfil specific eligibility criteria in order to be counted towards this threshold.^[22] The FSB also compiled a closed list of exclusions from TLAC,^[23] including for instance insured deposits, short-term deposits and liabilities arising from derivatives. These TLAC features, the stacking order and quality of liabilities, and their implementation into EU law are analysed below and should be perceived as especially relevant in the context of ex-ante impact of resolution-related capital.

2.2. BRRD as Implementation

The original Bank Recovery and Resolution Directive of 2014 constituted EU’s implementation of FSB’s standards. It covered comprehensively all aspects addressed in KA (resolution and recovery planning, resolution tools, establishment of resolution authorities), as well as adopted the European’s take on bail-in capital – minimum requirement for own funds and eligible liabilities (MREL)^[24] – even before the FSB’s TLAC was officially agreed on.

Despite some differences, MREL and TLAC share the same main function: to absorb losses if the financial institution is to be resolved.^[25] However, they are not identical. MREL is binding for all institutions encompassed by the BRRD and TLAC only for G-SIBs. Furthermore, the latter is supposed to be set at an individual level above the prescribed minimum, in turn the former is also prescribed on a case-by-case basis by the resolution authorities, but without predefined minimum threshold over regulatory minima.^[26] There are also differences when it comes to eligibility criteria,^[27] or consequences of breach.^[28]

After its adoption, the BRRD was dubbed as giving ‘a clear set of rules,’^[29] and MREL was perceived as ‘part of the toolbox that allows us [SRB] to restore the principles of the market economy to the banking sector.’^[30] However, as time passed and this legal act’s factual efficiency has been challenged by troubled Spanish and Italian banks,^[31] more and more criticism fell on its solutions and EU legislature. Some raised doubts on harmonization struggles,^[32] but main concerns were more material and could be divided into three groups – the ones regarding bail-in, a subgroup on MREL and the general ‘too big to fail’ issues. The first set of problems included uncertainty regarding deposits’ hierarchy in insolvency and character of bail-inable depositors (their retail nature). The BRRD treats practically all savers, apart from small, insured depositors, as investors. Hence, they all incur losses in case of bank’s failure.^[33]

As for MREL, many argued, that it is in fact doubling capital requirements^[34] and that no automatic consequences of the potential breach of this requirement are mentioned in the EU regime.^[35] Further, Tobias Tröger argued that MREL framework is too granular and complex to adequately price the risk underlying the MREL-eligible instruments and so, it should be separated from resolution strategies and drafted more like a capital layer.^[36]

Lastly, the BRRD also suffered from the lack of credibility regarding possible resolution of systemic institutions. No one really believes, that a ‘too big to fail’ bank could be resolved without any form of bail-out. Concetta Brescia Morra claims that not enough discretion is granted to the resolution authorities.^[37]John Armour and co-authors predict that resolution via sale of the business ‘plus an element of bail-out is likely to be the EU pattern, should a large and complex financial institution ever enter resolution.’^[38] Also lack of sufficient resources in the resolution fund undermines the case for BRRD-based resolution of systemic banks.

2.3. Banking Reform Package as a Way Forward

In light of these numerous shortcomings, EU regulators started working on the second version of the BRRD. However, it turned out that not only the BRRD requires some revisions and updates. Most of the legal acts introduced post-crisis also needed refreshing. The final Banking Reform Package adopted in 2019 encompassed Capital Requirements Directive V (CRD V),^[39] Capital Requirements Regulation II (CRR II),^[40] update to Single Resolution Mechanism Regulation (SRMRII)^[41] and finally, the new version of Bank Recovery and Resolution Directive (BRRD II).^[42] Most significant changes introduced by the CRD V and the CRR II included adoption of new standardised approach for counterparty risk, implementation of predefined level of leverage ratio, as well as liquidity enhancing net stable funding ratio.^[43] Additionally, third-country financial groups shall establish an intermediate parent undertaking within the European Union.^[44]

As for the resolution-related novelties, the revision was very wide-ranging – it covered most of the aspects addressed by the BRRD – from scope, definitions and competencies of resolution authorities, through preparatory activities and MREL to resolution of cross-border groups. The TLAC and MREL were streamlined by means of introduction of TLAC requirement via the CRR II and adjusting provisions on MREL in the BRRD. Also a new category of big banks was introduced – the so called ‘top-tier banks’, whose balance sheets exceeds 100 EUR billion.

Most importantly for this paper, the CRR II and the BRRD II shed new light on the character of MREL requirement. Actually, what Tröger advocates about MREL structured more like a capital layer in separation from resolution context, could be slowly materializing in the legal framework of MREL. Already in the previous version of the BRRD, some signs indicated that MREL’s role is actually reaching beyond the strictly resolution-related. Moreover, the new provisions on the quantity and quality of MREL instruments are included in the CRR II, the main legal act on prudential requirements.^[45] It looks like in case of this coin one side could be bigger than the other.

3.1. In Resolution

MREL has been originally linked to resolution and rightly so. The EU legislator embraced its resolution-related character by describing it in the BRRD, the legal act concentrated entirely on resolving financial institutions. Also, public Q&As published by the SRB explains that MREL is ‘to ensure an effective and credible application of the bail-in tool. Failure to meet MREL may negatively impact institutions’ loss absorption and recapitalisation capacity and, ultimately, the overall effectiveness of resolution. This requirement is part of the necessary steps needed to make institutions resolvable.’^[46] Such definition reflects the standard approach to MREL, only viewed as strictly functional to the implementation of resolution. Also, most resolution experts have shared this perspective. MREL – so the argument runs – is supposed to ‘limit the risk of contagion or bank runs, when it comes to conversions and write-downs,’^[47] and thus is clearly contingent on default. The essential character of MREL in the context of bail-in effectiveness is also stressed in the literature. As Wojcik stresses ‘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.’^[48]

The distinction between MREL and prudential requirements could also be derived from the conditions that ought to be fulfilled to trigger resolution. Namely, the first indicator of ‘failing or likely to fail’ status, the depletion of own funds, does not refer to MREL. Breach of MREL requirement itself, without infringement of the CRD requirements, could not lead to this assessment. However, it is worth noting that not all the infringement of prudential requirements leads to the designation of a bank as “failing or likely to fail”.^[49]

MREL’s mere ex post nature, meaning its usefulness only after resolution is triggered, has been also underlined by the prescribed consequences that given institution must face in case of violation of this requirement. According to the original version of the BRRD, MREL breaches were not sanctioned by any legal prescription, institutions was supposed to be urged to restore it. Precisely, ‘resolution authorities, in coordination with competent authorities, [were supposed to] require and verify that institutions meet the minimum requirement for own funds and eligible liabilities.’^[50] Some scholars derived indirect consequences from the stacking order, namely the relation between MREL and combined buffer requirements. For instance, Tröger argues that unsuccessful issuance of MREL could lead to depletion of CET1, and thus trigger restrictions on payouts from article 141 of CRD IV.^[51] However, no provision directly prescribed such restrictions for the breach of MREL standard.

3.2. Preventive

On the other hand, it can be argued that MREL’s preventive character has been noticeable from the very beginning. Indeed, already accusations of ‘doubling capital requirements’ could be perceived as granting MREL its preventive title. Even the SRB in its Q&A referenced above, does not deny that it could be seen as yet another capital layer, it even calls it a separate minimum requirement.^[52]

Regardless of the signs that holding MREL might be bearing some ex-ante effect, its preventive potential has so far failed to be recognised. It has been perceived mostly as putting additional burden on the credit institutions, or through the lens of market discipline potential. In contrast, we contend that MREL has in its nature a preventive effect. Its impact goes much beyond compliance challenges that banks have to face, or the influence of bail-inable creditors on risk-taking. The preventive potential of MREL should be perceived at both micro- and macroprudential level. As for the former, MREL determination framework and consequences of its breach would definitely influence individual bank’s operations or even business models. In the context of the latter, MREL seems to complement prudential capital requirements and so to raise the resilience of the financial system in good times, achieving countercyclical objectives.

In line with our claim, the BRRD II (along with the CRR II) sets up a viable legal framework to operationalise such preventive nature. It seems to acknowledge MREL’s ex-ante potential and not only at one level.

Firstly, the preventive influence of MREL is noticeable from the general and functional perspective. MREL for G-SIIs, along with MREL quality requirements, has been included in the CRR II, among the capital requirements of strictly preventive nature. Furthermore, also aspect of MREL’s composition and counting of the CET 1 capital could indicate its similarity to the combined buffer requirements (CBR).^[53]

Secondly, MREL’s preventiveness has been vastly developed by the particular material provisions introduced in the BRRD II and the CRR II. First, the abovementioned determination of MREL both at qualitative and quantitative levels suggests its transformation into preventive requirement. Second and maybe most convincing in that regard is the BRRD II’s formulation of consequences, should MREL be breached. It undoubtedly presents the broadest range of tools in the whole of EU banking law regime, mixing resolution-related powers with supervisory competences. It seems logical, that the EU regulators want to incentivise banks to maintain MREL at all times, as much as they should do with other capital minima and CBR. If they wanted MREL to be only resolution-connected tool, they would maintain the solution present in the US legal regime in the context of TLAC – its breach triggers distribution limitations only.

These two main areas where MREL slowly but steadily becomes more a preventive than a resolution-dependent requirement require in-depth analysis and assessment, starting from the relation between MREL and capital requirements, through the changes in its determination, to the broad scope of discretion granted to authorities in case MREL is breached.

Before delving deeper on the design of the new MREL framework, a methodological caveat is warranted. Our analysis demonstrates the existence of a preventive component of MREL across multiple dimensions: in MREL determination and enforcement, with effects on both micro- and macro-prudential policies. We also argue that such a preventive component positively complements the existing prudential framework.

However, we do not posit that the SRB and other resolution authorities shall always put their ‘preventive hat’ on and approach MREL from a preventive perspective. On the contrary, this is a problematic issue since the SRB probably even lacks a preventive mandate. Some of the issues related to the implementation of the preventive potential of MREL are discussed in Section 5.

4.1 Capital Requirements and Eligible Liabilities

The main goal of capital requirements is to safeguard the solvency of banks and foster their prudent and sound management.^[54] In so doing, the EU regulatory framework determines the level and quality of the capital banks must issue and hold.^[55]

On the contrary, as discussed in Section 3, the main and initial policy goal of setting MREL is to have enough eligible liabilities to resolve a bank should it get failing or likely to fail^[56]. In this view, meeting MREL in good times serves for maintaining the bank resolvable at all times^[57]. To this end, the resolution authority (RA) determines the required level of eligible liabilities and, as a consequence, restricts the entrepreneurial freedom of banks to determine their funding strategy, requiring to issue a certain amount of long-term, non-runnable debt.^[58]

Despite the difference in the primary policy goals pursued by the two sets of requirements; functionally, setting MREL does not differ much from regulating capital. Also in the case of capital requirements, banks must abide by a certain funding structure and follow the quantitative and qualitative requirements set by the the CRR and the CRD.

This functional resemblance suggests that MREL’s effects go beyond bank resolvability and intersects prudential policies. The Banking Reform Package of July 2019 goes in the direction of recognizing MREL as a two-face Janus. On the one hand, it acknowledges its preventive nature, including the requirements for own funds and eligible liabilities in the CRRII^[59] as a complement of capital requirements, at least for G-SIIs. On the other hand, BRRRII confirms and details the nature of MREL as a requirement to guarantee the resolvability of the bank.^[60]

Thus, the first step to evaluate the preventive potential of MREL is to assess its relationships with capital regulation in the current EU legal framework. In this perspective, it is important to differentiate between minima and buffers. Minimum capital requirements are the cornerstone of bank prudential regulation as of Basel I^[61] and are now set in the CRRII.^[62] These requirements shall be met at all times. Failing to meet those requirements can have severe consequences, up to the withdrawal of the banking licence.^[63] On the other hand, buffers are a new form of requirements introduced in the aftermath of the financial crisis with the aim of countering the procyclicality of capital regulation and the macro-economic externalities of banks’ activities.^[64]

Article 128(1)(6) CRDV introduces the combined buffer requirements (CBR),^[65] an add-on to minimum requirements that must be met with common equity.^[66] Differently from minimum requirements, buffers are designed to build excessive capital capacity in good time that can be depleted in bad times without endangering the solvency of the troubled institution.^[67] Accordingly, the consequences of breaching the CBR are milder and aim at restoring the buffer once the turmoil is overcome. Specifically, banks breaching the combined buffer requirement are restricted in their voluntary distributions^[68] and must present a “capital conservation plan”.^[69]

This legal design is consistent with the macro-economic aim to consider the credit and financial cycle in regulating bank capital,^[70] imposing higher requirements in good time to reduce risk-shifting and excessive risk-taking incentive and relaxing those in bad times to alleviate the possible credit crunch.^[71] Building excessive capacity in good times and restricting distributions when the bank is still relatively far from insolvency represent a coherent policy response to the Minsky’s theory of financial instability and endogenous creation of financial crises.^[72] In this view, financial crises result from imbalances that are driven by, but also feed by, an unsustainable economic expansion, which manifests itself in unusually rapid growth of credit and asset prices.^[73]

A third layer of capital requirement is the so called “Pillar 2” capital. The competent authority can require individual banks, on a case-by-case basis, to hold capital to cover additional risks^[74], as a result of the annual evaluation process.^[75]

The problematic aspect in coordinating a multilayer capital regulation rests in the stacking order of the various capital components. This aspect is almost irrelevant in good times for a well-capitalised bank but becomes extremely relevant when the situation starts to deteriorate. The ways the various components are stacked changes the threshold for breaching the CBR and, hence, trigger the limitation on distributions. Clearly, minimum requirements are the first to be complied with. Then, the CRDV specifies that Pillar 2 requirements sit on top regulatory minima^[76]. Finally, the CBR sits on top of the Pillar 2 capital. Therefore, the limitation on distributions can be triggered early and this has itself a positive macro-prudential effect as it restrains excessive distributions in a phase where capital needs to be strengthened without issuing fresh equity, as it might send negative signals to the market^[77].

Figure 1

Different option in stacking Capital Requirements, Capital Buffers and MREL.(Scale not meaningful, just illustrative).

In this already complex conundrum,^[78] the position of MREL represents an additional problem to consider. Both own funds (i.e.: capital instruments)^[79] and other eligible liabilities^[80] constitute MREL. Therefore, it may arise the question of how to count CET1 instruments in excess of minimum requirements and Pillar 2 requirements. There are three possibilities, as graphically depicted in Figure 1. First, common equity in excess should first cover the MREL shortfall and, in turn, increase the amount of capital needed to comply with the CBR. Second, it should cover the CBR without triggering the limitations on distribution. In this case, a breach of MREL requirement is likely to materialise. Third, common equity in excess counts for both MREL and CBR (double counting).^[81]

The European legislator solved this conundrum adopting a compromise in between the first two regulatory options. The solution adopted by the BRRDII in this regard will be widely discussed in Section 4.3, as it directly pertains to MREL breaches.

Interestingly, among these three possibilities, only the third one would be consistent with a mere ex-post approach to MREL, where the amount of eligible liabilities held by the institution is only relevant for the purpose of performing a resolution should the bank be declared “Failing or Likely to Fail”. Interestingly, the double counting was the only option ruled out since the very beginning of the negotiation towards the BRRDII.^[82] This demonstrates, once again, that the double nature of MREL is inherent to its nature and, willingly or unwillingly, is also embraced by the European Legislator. Therefore, the prudential potential constitutes an important component of the MREL framework. The rest of this article discusses whether the legal framework resulting from the 2019 Banking Reform Package already fully exploit such a (macro)preventive potential.

Before moving forward and specifically discuss the MREL framework, it is important to stress once again that MREL regulation is not functionally different from capital regulation. It restrains the funding options of banks, regulating (part of) the composition of the liability side of their balance sheet. The regulatory framework on capital and the one on eligible liabilities are inherently integrated and their preventive effectiveness significantly depends on the quality of such integration.

Bearing in mind this take away point, the proceeding of the section discussed in detail the two crucial aspects of MREL determination (Section 4.2) and MREL breaches (Section 4.3) and highlights their prudential component.

4.2 MREL Determination

The determination of MREL represents the first feature dense of preventive potential. Two aspects are crucial: the qualitative requirements to be counted as eligible liability and the rules to determine the quantitative level of MREL.

The Banking Reform Package innovates significantly on the qualitative requirements in two main directions. First, it introduces the structural subordination requirement: eligible liabilities must be subordinated to any other non-eligible liability.^[83] This clearly goes in the direction of making banks more resolvable, smoothening the relationship between the resolution framework and national insolvency laws.^[84]

Furthermore, the qualitative requirements for financial instruments to count toward MREL were also updated. Such an update goes in the direction of making MREL requirements more aligned with the qualitative requirements for capital instruments.^[85] In particular, the new regulation forecloses the possibility to include many of the standard debt covenants on eligible liabilities. For instance, the contractual arrangements governing the instruments cannot include any provision incentivising an early call or redemption of the instruments;^[86] or provisions to speed up the repayment for conditions other than insolvency or liquidation.^[87]

These qualitative requirements parallel the requirements that were previously applied only to capital instruments^[88]. In contrast, the first version of the BRRD established very light qualitative requirements mainly focusing on the requirements of having at least one year of residual maturity.^[89] These closely resembled the criteria to consider a liability as bail-inable in resolution^[90] and departed significantly from the qualitative requirements for capital instruments.

This transition has a profound functional meaning. It implicitly says that MREL instruments are not only functional to perform a bail-in should the bank enter in resolution, but these also serves the maintenance of the safety and soundness of the institution in good time, such as the (preventive) capital requirements.^[91]

These provisions, at a first sight, might appear as a very technical and legalistic set of specifications. However, these make a remarkable functional difference also for the ability of creditors to discipline the bank. The corporate finance literature has recognised the nature of debt contracts as a way to contingently allocate control over corporate activities, shifting it from shareholders to creditors.^[92] The main statutory contingency in non-financial institution is insolvency. But contractual agreements, covenants, can determine partial control shifts also when the borrower is solvent in order to minimise the agency costs of debt.^[93] On the other hand, allowing for financial covenants implies a likely fund outflow when the situation of the bank starts deteriorating.^[94] This would, consequently, pose issues of financial stability.

Complementing these provisions on the stipulation of the contract, the CRR disciplined a crucial aspect of its executions. The new provisions resemble qualitative requirements for capital instruments also in relation to calling, redeeming or repurchasing instruments. Such activities are now subject to stringent conditions and to the prior approval of the competent authority.^[95]

Therefore, the functional relevance of reforming qualitative requirements for eligible liabilities lies in striking a balance between financial stability considerations and market discipline.^[96] After the Banking Reform Package, the financial stability considerations prevailed not only when it comes to capital instruments but also for other eligible liabilities. As a by-product of this new approach, the regulatory and supervisory agencies gained the upper end over the market participants in making the MREL framework effective ex-ante. This is consistent with the approach to banking regulation according to which market discipline cannot work because of the special features of banks; hence, the supervisor should substitute the market participants, mimicking the behaviours these would have had in normal market conditions.^[97]

Building on this last observation, the rules for the quantitative determination of MREL represent a further step toward its preventive understanding. In the first version of the BRRD, the RA determined the MREL on a case-by-case basis.^[98] In contrast, the BRRDII states that MREL ought to be met at all times, such as minimum capital requirements, and that the determination of MREL must be expressed as a percentage of both the risk-weighted and non-risk-weighted exposures of each institution.^[99]

This second aspects represents an important innovation compared to the previous regime, insofar as imposing a plain leverage ratio as a complement of the traditional risk-weighted ratio better capture the leverage dynamics over the financial cycle. This approach constitutes a cornerstone of the post-crisis countercyclical (preventive) financial regulation.^[100] A leverage ratio requirement was already in place for own funds in the first version of the CRR^[101]. Again, the alignment with capital requirements testifies a new understanding of the preventive potential of the MREL framework.

The specific determination depends upon the systemic relevance of each bank. Nonetheless, article 45c(1) BRRDII sets down some general principles that should guide the resolution authority in the quantitative determination of MREL. First of all, each institution should have sufficient eligible liabilities to implement the preferred resolution strategy envisaged in the resolution plan.^[102] This principle clearly looks at assuring the resolvability of the institution and is in line with the main-stream characterisation of MREL. Besides assuring the resolvability of the institutions, the RA must also consider the “size, the business model, the funding model and the risk profile” of the institutions.^[103] The RA shall also mind the negative spillovers on financial stability that the failure of an institution may have, with specific regard to counterparty risk.^[104]

Reading together these last two provisions clearly shows that – in determining MREL quantity – the resolution authority can, and should, go beyond resolvability consideration. The RA should consider the specific risks posed by each institution, trying to prevent these from failing. Furthermore, the reference of the interconnectedness of counterparties as a criterion for MREL determination trespasses the border of micro-prudential policy and enters directly into the macro-prudential area. Such a reference allows to set MREL considering the cross-sectional contribution to systemic risk posed by each institution.^[105]

The case-by-case determination that characterised the first version of the BRRD is now only applied to less significant institutions.^[106] Those should be required to hold enough eligible instruments to cover the expected losses and to appropriately recapitalise the bank after resolution. In this instance, the resolution authority has little room for considering preventive aspects in determining the level of MREL.^[107] For “top-tier” European banks that do not qualify as G-SIIs, the case-by-case determination only partially applies. The BRRDII set a regulatory floor to the level of own funds and eligible liability, that must be at least 13,5% of risk weighted exposures and 5% of non-risk weighted exposure.^[108] Such a floor can, in itself, be considered countercyclical. Even if in good times the level of expected losses is particularly low because of the economic condition, the MREL cannot fall below the predetermined amount, limiting leverage and risk-shifting. Finally, the requirement for globally systemic banks is fully harmonised to the TLAC standard. Therefore, G-SII banks shall hold 18,0% of risk weighted exposures and 6,75% of non-risk weighted exposure^[109]. Notably, the introduction of different categories of banks proves the awareness of the EU legislator about the macro-prudential effects of MREL determination and reinforces the argument proposed in this article.

In addition, the level of required MREL can be further increased if the RA deems it necessary to fulfil the general conditions of article 45c,^[110] i. e.: assuring the resolvability of the institution and, crucially for this paper, considering the specificities of each institution and the specific systemic risks its failure would pose.^[111]

In the jargon resulting from the Banking Reform Package, this was referred as Pillar 2 MREL, considering it in parallel with Pillar 2 capital requirements.^[112] However, such reference is incorrect. This additional layer of MREL lack some of the prerequisites of Pillar 2 capital. According to the CRD, the competent authority can require to hold own fund in excess^[113] mainly to “ensure a sound management and coverage of [banks’] risks”^[114] if the outcome of the annual review process reveals that those are not sufficiently accounted for by regulatory requirement.

Article 45d (3) refers to the general principles of MREL determination, i. e.: ensuring resolvability, considering the business model and risk-profile of the bank and the systemic concerns it poses. Ensuring resolvability is a resolution specific principle. However, the other two are conceptually closer to capital buffers than to Pillar 2 capital. Contrarily to capital buffers, the additional amount of eligible instruments required pursuing article 45d(3) is structurally part of MREL and, accordingly, shall be met at all time. Hence, its legal design differs sharply from buffers.

This ambiguity in the legal nature of MREL dates back to its first formulation and closely relates to the regime on the enforcement of MREL breaches. In the first version of the BRRD, the RA had no enforcement mechanisms to handle a potential breach. This is clearly at odds with the legal determination of MREL as a requirement to be met at all times. The European Legislator tackled this gap in the BRRDII, providing the RA with intrusive powers to enforce MREL. Nonetheless, some ambiguity in the design of the rules and the remedies still persists, signalling that the understanding of the legal and functional nature of MREL is not yet perfect.

4.3 MREL Breaches

The new enforcement powers in case of an MREL breach constitute the most innovative aspect of the MREL framework in the BRRDII. Regulatory enforcement represents a quintessential element to make preventive regulation effective and efficient: “The enforcement of regulation and government controls is complex and to a large extent determines the effects of the law. Yet it is surprising how many discussions of government regulation implicitly assume that laws are somehow self-enforcing and that there is full compliance”.^[115] This was largely the case in the first version of the BRRD, where the resolution authority lacked clearly defined and proportionate legal tools to preventively address a breach of MREL.^[116]

The Banking Reform Package tried to solve this shortcoming by granting the RA with extensive powers to address any breach of MREL requirements.^[117] In so doing, the resolution authority can implement the powers to remove impediments to resolvability;^[118] restrict certain distributions;^[119] impose one of the Pillar 2 measures attributed to the competent authority;^[120] use the early intervention powers provided by the BRRD;^[121] impose monetary sanctions and other administrative measures.^[122]

From a legal perspective, it seems that these powers should be interpreted as strictly tied to the goal of addressing MREL breaches. So, for instance, breaching MREL does not automatically trigger the early intervention phase but the resolution authority can use those powers to address a breach. Furthermore, breaching MREL can be qualified as a material impediment to resolvability^[123] for which the law provides wider enforcement powers as compared to other impediments to resolvability.

The imposition of penalties, though legally relevant, does not seem to have a substantive ex-ante potential. The vast literature on law enforcement highlights how the two crucial components to determine the effectiveness of a sanction are its magnitude and the probability of detection.^[124] In the context of banking, optimizing the deterrence level of penalties would likely mean to increase the magnitude of the sanction to a point that would, itself, endanger the stability of the institution. Therefore, behavioural remedies seem to better fit the case of MREL breaches and can be complemented, if necessary, by the sanctions provided by Articles 110 and 111 of the BRRD. The use of early intervention powers does not seem to have great ex-ante potential. It is only worth mentioning that it allows for an early trigger of the recovery plan.^[125] The recovery play may encompass planned measures to reduce risk and leverage or restructure liabilities, so that the preventive effect can be indirect.^[126]

The other three sets of powers are crucial to deeply understand the ex-ante nature and the preventive potential of MREL. Hence, these deserve a detailed discussion. First, the RA can use the power to remove impediments to resolvability. The resolvability assessment consists in a state of the art, periodical exercise during which the resolution authority assesses whether the preferred resolution strategy, as envisaged in the resolution plan, is credible and feasible.^[127] Breaching MREL constitutes a material impediment to resolvability and, consequently, the resolution authority can certainly use the powers for removing such impediments.^[128]

Before discussing the specific powers to remove impediments to resolvability, it is worth noting how the BRRD itself recognises the ex-ante and potentially macro-prudential impact of those powers. In identifying the suitable power, the resolution authority must “duly consider the potential effect of those measures on the particular entity, on the internal market for financial services, and on the financial stability in other Member States and in the Union as a whole”.^[129] This formulation seems to hint at the possible adverse consequences of removing the impediments to resolvability. However, as this paper have argued, this may also be interpreted with a positive connotation: between two equally effective measure to remove impediments to resolvability, chose the one that has also a preventive, micro- or macro-prudential, effect.

The powers aiming at removing impediments to resolvability can be split into MREL-specific and other general powers. Among the MREL-specific powers, the resolution authority can ask the bank to issue new eligible liabilities^[130] and to renegotiate the contractual terms and the maturity profile of existing ones.^[131] Moreover, the RA can require to submit a plan to restore compliance with MREL,^[132] as a complement to other measures imposed to the bank or agreed upon during the informal phase.^[133] Among the non-MREL-specific powers, the resolution authority can require drawing or updating the intragroup financing arrangements; require the entity to limit its exposures; divest specific assets; limit or cease complex and risky activities or products; require to change the legal or operational structure.^[134]

Looking at the formulation of the law and the current EBA Guidelines on the matter, it seems that the resolution authority has full discretion in choosing the measure to address the impediment, so long as the measure is proportionate and the decision is justified.^[135] Therefore, the resolution authority should not be restricted to the MREL-specific measures but could use whole menu of powers.^[136]

The resolvability assessment clearly showcases the double nature of MREL. On the one hand, its first goal is to make resolution credible and feasible; hence, looking at the ex-post nature of MREL. However, this has two preventive spill-overs. First, making resolution as credible as possible increases the level of market discipline ex-ante. Second, the powers to address the impediments to resolvability shape banking activities and funding model ex-ante in the direction of increasing bank resilience. This latter observation is particularly important when positioned over the financial cycle. As the EBA Guidelines point out: “Resolution authorities may apply the measures solely to address impediments to resolvability without any existing breach or likely breach of prudential regulation requirements by the institution being required”.^[137]

This means that the powers for removing impediments to resolvability can be used timely, when the bank is still solvent and profitable, limiting the build-up of systemic risk in good time well before such risk materialises. At that point, when the situation starts deteriorating also other, more intrusive, measures can be used. However, from a preventive perspective, it is already too late.

The new power to limit certain distributions in relation to MREL or CBR breach fits in this framework of early policing of non-sustainable practices. The resolution authority has the power to prohibit the voluntary distributions exceeding the Maximum Distributable Amount related to the minimum requirement for own funds and eligible liabilities (‘M-MDA’).^[138] The BRRDII provides two legal bases for the resolution authority to restrict distributions. The first, as discussed in Section 4.1, is the case when the bank is breaching the CBR when MREL is counted on top of Pillar 2 capital. The second case is to address MREL breaches. Article 16a describes the first case, linking the restriction on distributions to CBR breaches. In contrast, article 45k empowers the resolution authority to use this power also in case of MREL breaches, regardless of the bank’s compliance with the CBR. This interpretation is important from a functional standpoint – it enlarges the scope of the power of the resolution authority and enhances its ability to tackle cases of early risk building.

The last relevant set of powers to address MREL breaches consists of the possibility to resort to the supervisory measure referred to in article 104 CRD. Those are the solutions allowing the supervisor to take early actions as a result of the evaluation process^[139] or if the bank is likely to breach regulatory requirement in the following 12 months.^[140] The BRRDII allows to use these powers earlier, when capital requirements are complied with but MREL is not. The formulation of article 45k BRRDII suggests that the competent authority is the only one entitled to use those powers. The interpretation of the rule seems to suggest that supervisory measures can be triggered not only by the likelihood to breach capital requirements, but also by MREL breaches. In this case, the competent and resolution authorities must coordinate so that the competent authority uses its powers upon requests of the resolution authority. The coordination and overlaps in the powers and the mandates of the relevant authorities involved represents one of the most problematic aspects of the entire framework, as will be detailed in Section 5.