Position Papers
The voices of savings and retail banking
Learn more
You are here: > Newsroom > Position Papers > ESBG responds to EBA consultation on MREL framework
ESBG responds to EBA consultation on MREL framework

ESBG response to the EBA consultation on the report 

on implementation and design of the MREL framework

ESBG (European Savings and Retail Banking Group)

Rue Marie-Thérèse, 11 - B-1000 Brussels

ESBG Transparency Register ID: 8765978796-80

August 2016​


Doc 0720/2016                                                                esbgkleinzonderletters200dpi                                                              


Vers. 1.4

>> See .pdf version

Dear Sir/Madam,

Thank you for the opportunity to comment on the EBA report on implementation and design of the MREL framework. We would like to share with you the following reflections that we hope will be taken into account by the EBA.

The comments below address both the questions specifically addressed by the EBA to interested parties (A) as well as additional areas of the interim report we believe to be in need of comment (B).


A.   Answers to EBA specific questions

1.       Reference base for MREL

ESBG is generally in favour of changing the reference base to risk-weighted assets (RWAs). To us, it seems a logical step given the methodology for the determination of Minimum Requirement for Own Funds and Eligible Liabilities (MREL) in the EBA Regulatory Technical Standards (RTS) starting from the expression of the requirement in terms of RWAs, which is then translated into a requirement in terms of own funds and total liabilities. In addition, the change in the reference base would align the formulation of the requirement with Total Loss-Absorbing Capacity (TLAC) requirements.

At the same time, however, we would like to take this opportunity to stress that it is important to keep total liabilities as a reference base in another context, namely regarding the minimum bail-in rule to access resolution funds. According to Art 44(5) Bank Recovery and Resolution Directive (BRRD), resolution funds can make a contribution to an institution under resolution only where, among other conditions, shareholders and creditors have made a contribution of at least 8 percent of total liabilities and own funds (TLOF)[1]. ESBG finds it important to preserve the reference base as TLOF in the context of the minimum bail-in rule to access resolution funds.

Furthermore, we believe that the alternative solution (using leverage exposure as the denominator) has drawbacks where the calculation is undertaken at solo level for institutions that are part of a group. Leverage exposure can exclude intra-group liabilities, which – depending on the resolution strategy – should be considered as separate to the MREL setting. In addition, there is significant divergence in supervisory practices (at least at the moment) with respect to authorising the exclusion of intra-group exposures from leverage exposure at solo level, making this figure less comparable across jurisdictions and institutions than it might appear. In order to establish consistency with the TLAC requirements, the compliance of the MREL requirements with leverage ratio exposure as a 'backstop' should only apply to global systemically important banks (G-SIBs). Establishing it as a 'backstop' also for non-G-SIBs would be contradictory to the superior requirements of the BRRD.

ESBG notes that the EBA's interim report does not mention the Basel I floor as an alternative backstop measure for the calculation of MREL. We would request clarification whether this omission is done on purpose or whether the current treatment in the EBA RTS on MREL, which looks at both the leverage ratio and the Basel I floor as backstops, is expected to be maintained. We would support dropping the Basel I floor as it not only creates additional complexity but the measure itself is becoming more and more 'anachronistic' as other measures are being phased in. Furthermore, the future of the Basel I floor beyond its transitional end date of 31 December 2017 is uncertain and it would not, in our view, be appropriate to retain it as a backstop for MREL if it were no longer part of the capital framework. In any event, the interaction of the Basel I floor with minimum capital requirements, additional capital requirements and buffers as set out in the RTS on MREL should be aligned between the loss absorption and recapitalisation amounts. While the loss absorption amount clearly specifies the Basel I floor as a backstop to total capital requirements, including buffers, the wording for the recapitalisation amount is not clear on how the Basel I floor should be used as a backstop. Based on this, ESBG asks the EBA to specify that the Basel I floor should also be compared to total capital requirements, including buffers, with respect to the recapitalisation amount by clarifying the current wording in the RTS on MREL, which in fact lead to interpretation differences by resolution authorities.


2.      Relationship with regulatory requirements

From ESBG's point of view, the EBA rightly mentions that extending TLAC requirements to all institutions, and consequently increasing MREL requirements, could cause problems, especially for smaller banks with limited access to capital markets.

Apart from that, the subsequent paragraphs present further aspects that should be taken into consideration by the EBA when aligning the MREL requirements with the TLAC framework.

As the EBA indicates, there is potential interplay between the position of the buffers (and the requirement that CET1 capital used for buffers cannot be used for MREL) and the determination of the maximum distributable amount (MDA) and potential distribution restrictions when combined buffer requirements are breached. The EBA implicitly (this is not stated explicitly in the report) seems to assume that distribution restrictions would be triggered if the capital and other eligible liabilities other than own funds of a bank are insufficient to meet Pillar 1 and Pillar 2 capital requirements, MREL requirements and buffer requirements. In order for this to occur, the relevant articles in the Capital Requirements Directive IV (CRD IV) would need to be amended; however, in this respect, there is no reference made in this report. Furthermore, ESBG is sceptical about an automatic MDA restriction as soon as a bank has breached the combined buffer requirement (CBR), which sits on top of MREL, when the bank still fulfils all capital requirements (including CBR) – at least for the coupon distribution – as otherwise it would have a detrimental effect on the ability to issue own funds instruments.

ESBG is concerned about the consequences of this possible, new EBA approach and we find it essential to consider the interaction with the MDA before potentially moving any further with this approach. It must be kept in mind that buffers should not operate in a way that using part of a buffer to satisfy MREL would result in a buffer breach with MDA-restrictions as long as the institution fulfils its capital requirements. Moreover, if double counting is no longer possible, the calibration of the MREL requirement should reflect the restricted base of MREL-eligible instruments.

What is more, it would be important to clarify how a bank would determine the amount of capital required to fulfil MREL requirements sitting 'underneath' the CBR, particularly in cases where banks have an excess of MREL-eligible liabilities and own funds compared to their requirements. In this context, could banks assume to first 'fill in' the MREL requirement with eligible liabilities other than own funds and apply own funds only for any residual amount of the requirement (thus leaving the maximum possible amount of CET1 to cover the buffers)? In contrast, ESBG would not appreciate a 'composition requirement'-approach for MREL similar to those for Pillar 2, imposing a mandatory proportion of CET1 to be applied to MREL. This would be an undue requirement especially for deposit-funded banks with 'excess deposits'.

When aligning the MREL requirements with the TLAC framework, the current RTS on MREL would have to be changed to ensure that the buffers are not considered to be part of neither the loss absorption nor recapitalisation amounts, i.e. not part of MREL.

3.      Breach of MREL

Generally speaking, ESBG is of the opinion that a breach of MREL should be met with a patient and soft approach by the relevant authority since MREL in nature is aimed at solving a gone concern situation rather than a going concern situation (for instance, any measures and solutions should always reflect the actual situation, in particular as some 'breaches' of MREL can be purely technical or just a question of timing, e.g. when an anticipated bond issue is delayed by a few weeks, leading to available MREL dropping below the requirement). In line with this view, a breach of MREL should not be an automatic indication of failing or likely to fail, unless other requirements (e.g. minimum capital requirements, the leverage ratio or the liquidity coverage ratio [LCR]) are breached at the same time.

In particular, in cases where the preferred resolution strategy is bail-in and MREL is set at a level of twice the minimum capital requirements (potentially including twice the Pillar 2 add-on plus – at least at the moment – elements of capital buffers), a breach of the MREL requirement does under no circumstances mean that the bank is not sound on a going-concern basis. The breach of MREL may present an impediment to resolvability. However, this in itself cannot be interpreted as an indicator of failing or likely to fail. 

Additionally, it should be kept in mind that the final decision on sanctions because of a breach of MREL depends on the final suggestions of the European Commission regarding the harmonisation of MREL and TLAC. According to our information, both of MREL's Pillar 1 and Pillar 2 requirements are being discussed. At the end of the day, a breach of MREL's Pillar 1 requirements could be dealt with by applying the same sanctions as for the own funds requirements, which would be in line with the TLAC requirements. Any decision on sanctions due to a breach of MREL's Pillar 2 requirements could be left to the competent authorities (supervisory and resolution), as it is a requirement to ensure the resolvability of a bank.

Finally, we would welcome a tight cooperation and coordination between resolution and competent authorities in order to avoid diverging actions and therefore limiting the impact.


4.      Adequacy and calibration

The EBA provisionally recommends that the calibration of MREL should in all cases be closely linked to and justified by the institution's resolution strategy. ESBG fully agrees with this approach and would, in particular, like to highlight the need for justification by authorities setting MREL as – at least in our experience – this is not yet sufficiently the case. This would also include the procedure set out in the BRRD stating that the determination of any consolidated requirement would need to be made by a resolution college, taking into account the resolution strategy and the MREL levels set (or contemplated) at individual entity level. Notably, the reduction of own funds deductions and RWAs due to the resolution strategy envisaged for each institution should be taken into account.

We believe that this creates particular challenges for institutions headquartered in the Banking Union, but with significant operations in non-Eurozone Member States at the same time, as the Single Resolution Board's (SRB's) approach to setting MREL does not appear to align with this envisaged procedure. In this respect, we would also like to highlight that we do not consider that the SRB has really 'publicly communicated its policy intentions for setting MREL'. Statements so far have been very unspecific, in some cases at odds with the EBA RTS (e.g. when stating that G-SIIs would have to hold at least 8% of own funds and eligible liabilities as MREL). Additionally – and most importantly – the outcome of the dialogue with individual institutions does not appear to reflect the public statements made in all cases. Further convergence of the approach therefore seems to be necessary, in ESBG's opinion.

The EBA moreover states that business models should be considered in the determination of MREL when they lead to differences in resolution strategies. We agree with this statement, but would also like to point out that the chosen resolution strategy should respect the business model, e.g. deposit-based banks should not be forced to change their business model, strategy and funding profile as a result of a national resolution authority (or the SRB) determining a 'preferred  resolution strategy'.

Furthermore, the EBA writes that MREL should be set at the higher of the requirement resulting from the assessment at individual level as outlined in the BRRD (firm-specific Pillar 2 requirement) and any Pillar 1 requirement, should one be introduced. The EBA is, however, not clear on the character of such a Pillar 1 requirement, talking about 'floors' in the report. In our opinion, the only Pillar 1 requirement to be considered should be the TLAC requirement for G-SIIs and the 'higher of' test should thus only apply to G-SIIs subject to TLAC. No Pillar 1 minimum requirement should be introduced for any other type of institution, where the firm-specific approach currently contained in the BRRD should be maintained.

As the ECB introduced a new mechanism for the supervisory review and evaluation process (SREP) to be set for 2017, ESBG assumes that only the Pillar 2 requirement (and not the Pillar 2 guidance) will count towards the capital requirement in the calibration of the loss absorption and recapitalisation amount required, in particular with regard to non-G-SIBs.

​So far, institutions and authorities have been unfamiliar with the preparation of resolution plans and the calibration of MREL requirements. No minimum MREL requirements should be determined hastily, in our opinion.

Furthermore, according to the BRRD, resolution authorities are responsible for calibrating any MREL requirements taking into account the specific conditions of an institution. The needed recapitalisation amount in case of resolution depends on the particular resolution strategy.

Moreover, from ESBG's point of view, membership in an institutional protection scheme (IPS) should be taken into account in an appropriate manner. Up to now, being member of an IPS does not seem to be considered at all. The recent ECB guide regarding the recognition of IPSs for prudential purposes underlines their importance. The ECB stresses that an IPS should ensure that its member institutions abide by the regulatory own funds and liquidity requirements. An intervention by an IPS is deemed to be triggered, at the latest, where there is no reasonable prospect that any alternative measures, including the recovery measures provided for in the recovery plan, would prevent the failure of that institution. If an IPS is recognised according to Art. 113(7) CRR and is hence supposed to prevent a resolution, this fact must be taken into account when calibrating MREL.


5.      Eligibility

In this context, we would like to point out again that the European Commission is working on a possible harmonisation of the bail-in hierarchy of claims. Based on that, ESBG suggests waiting for the official proposal (expected in autumn 2016) before deciding on any subordination requirements for MREL.

​What is more, ESBG agrees that information on creditor hierarchies should be available to investors, regardless of the form of subordination employed. However, we are not sure that the proposed disclosure items necessarily lead to the desired outcome. Below we would like to comment on each of the proposed types of disclosure:


  • Bank balance-sheet structure: ESBG does not think that the disclosure of bank balance-sheet structure would improve transparency with respect to the insolvency ranking, particularly given potential differences between resolution and insolvency law. In addition, the resulting disclosures would be significantly more complex than current disclosures in order to provide sufficient detail on, e.g. the difference between MREL-eligibility and bail-in-ability, potential exclusions from bail-in and the insolvency hierarchy. This would not be in the interest of investors or debtors.
  • Banks' MREL: disclosure of MREL itself is insufficient for investors to determine their position in the insolvency hierarchy, especially as bail-in does not stop at MREL-eligible instruments but encompasses all instruments that could be subject to bail-in (in particular debt with a remaining maturity below one year).
  • Availability of standardised information on statutory creditor hierarchies: this information will, to some extent, help to provide transparency on creditor hierarchies and is regarded as more positive by ESBG than the other two items above. 

In summary, ESBG would be in favour of standardised information on statutory creditor hierarchies.

In case it is considered that banks need to take care of providing standardised information on statutory creditor hierarchies, ESBG would like to highlight that an additional public consultation process with the stakeholders from the banking sector would be imperative as they would be faced with very burdensome consequences.

With regard to reporting obligations, we believe that they should apply only to institutions, which would be resolved if they are failing; particularly if the resolution plan clarifies that bail-in would be the preferred resolution strategy. This, in contrast, should not be the case where a resolution authority comes to the conclusion that an institution would not be resolved, but normal insolvency proceedings would apply. If common reporting standards will be developed one day, they should be proportionate and not overly burdensome. They should only be set to a minimal possible extent, ensure clear and unambiguous definitions, and provide an adequate timeframe for implementation.

As a final point, the EBA mentions that there is a need to allow for the write-down or conversion of internal MREL instruments in cases where the subsidiary holding the internal MREL is not itself in resolution (particularly important in SPE resolution strategies). ESBG believes that in most cases a contractual solution should suffice, as internal MREL will be subject to internal arrangements within banking groups and does not involve any particular issues with respect to contractual clauses. Using Art. 59 BRRD to create a statutory framework is, in our view, not necessary at this stage. We would also like to mention that, while we understand the rationale set out in the report to extend Art. 59 BRRD to internal MREL, the most recent non-paper by the European Commission proposed the extension to all MREL-eligible liabilities, i.e. including external MREL. ESBG opposes to such a blanket extension as bail-in of external MREL should only occur in resolution and a write-down power outside of resolution is thus not necessary for external instruments (as it is issued out of resolution entities).


6.      Third-country recognition

One aspect that should be considered is the issue that potentially arises through the UK's decision to leave the EU. If the UK does leave the EU, it would become a third country. In our opinion, there should be considerations about what requirements would apply in case a country leaves the Union, e.g. would all contracts automatically become subject to Art. 55 BRRD requirements or would this apply only to contracts entered into after the actual 'leaving date'?

A more general issue pertains to the question of 'equivalence'. The regime would be significantly easier to apply if there was a centralised process for assessing the equivalence of third-country resolution regimes. For example, if the Financial Stability Board (FSB) concluded that jurisdictions implemented the key attributes with respect to bail-in powers in line with the FSB proposals, the EBA could consider them equivalent for the purposes of Art. 55 BRRD and publish this acknowledgement so that banks would not need to include contractual language in contracts under the law of these jurisdictions. Similarly, this regime could be applied in the UK's situation, i.e. the EBA could still consider the UK an equivalent jurisdiction based on an assessment of its implementation of the BRRD.

As rightly mentioned in the EBA interim report, some reduction in the burden of compliance with third-country recognition requirements is necessary, and could be achieved by narrowing the scope of the requirement, while maintaining the effectiveness of contractual recognition for MREL liabilities.

In this regard, Art. 55 BRRD states that Member States shall require institutions and entities to include a contractual term by which the creditor or party to the agreement creating the liability recognises that liability may be subject to the write-down and conversion powers.

ESBG is concerned about the potential broad interpretation of the term 'liabilities' because it could cover certain types of contract such as: (a) bank guarantees or documentary credits; (b) financing agreements in which the financing entity is committed to deliver funds to its customers or to issue contingent payment commitments on its behalf; or (c) contracts that may only involve a payment obligation by the bank in case of a sentence obliging the bank to pay because of a breach of the contractual obligation (whether monetary or not). From an accounting perspective, these contracts are not considered liabilities.

ESBG therefore suggests that only liabilities that can absorb losses and are therefore included as liabilities in the balance sheet from an accounting perspective should be under the scope of Art. 55 BRRD. We would appreciate if the EBA could clarify this aspect.

What is more, for the purpose of avoiding legal uncertainties regarding a bank's suitability of being resolved and giving resolution authorities all the flexibility that they may need to exclude liabilities from bail-in (considering a bank's business model and resolution strategy), ESBG believes that a combination of options iii (limit the scope of Art. 55) and ii (introduce a power for resolution authorities to grant waivers from Art. 55), as proposed by the EBA (page 62), would be best.

The assessment of various existing contracts have shown that it will be nearly impossible to adapt many of the potentially-affected contracts. The abovementioned proposal, i.e. a combination of ii and iii, would help ensure that liabilities and claims do not need to be adapted in the future, nor would complex expert opinions be imperative.


B.   Further Comments on the EBA Interim Report not specifically raised

1.          Approval for redemption of MREL instruments

The EBA's interim report considers a number of options for approvals concerning the redemption of MREL-eligible liabilities. Before discussing the issues with each of the options considered below, ESBG would like to recall that there is currently no approval requirement for redemptions of MREL-eligible instruments in the BRRD and that the TLAC term sheet, which includes an approval requirement, only requires this in cases where the redemption would lead to a breach of the TLAC requirement.

Consequently, this raises the question whether there should be an approval requirement for MREL-eligible liabilities introduced in the BRRD at all, obviating the need for any of the proposed options. If the objective was to introduce a requirement in order to facilitate the implementation of TLAC, any requirement should not exceed the provisions of the TLAC term sheet.

Some more detailed comments on the proposed options:


  • Extending approval for redemption to all external MREL-eligible instruments: we consider this approach too strict – especially for banks with a very high amount of MREL-eligible liabilities. In addition, we do not consider it workable in cases where deposits constitute eligible liabilities. Requiring approvals in all these cases would significantly impede banks' liquidity and funding risk management in addition to creating significant administrative burden for both institutions and competent authorities. As the current process of obtaining approval for redemptions of capital instruments has already proven to be extremely cumbersome and time-consuming (with approvals taking significant amounts of time), ESBG has serious doubts whether an extension to all MREL-eligible liabilities would be workable.
  • Further extending approval for redemption to internal TLAC/MREL: in general, this option considers two separate issues, i.e. an approval might be required in all instances where redemptions are contemplated (similar to the previous point) or only in cases where the redemption would lead to a breach of the internal TLAC/MREL requirement. With respect to the first issue, the points made right above apply equally, with the volume of potential approvals further multiplied. In respect of the second issue, it would, first and foremost, be necessary to establish a binding internal MREL requirement (which is currently not included in the BRRD and the report itself does not provide any information on how internal MREL would be established). Given that the TLAC term sheet also does not foresee any approval requirement for redemptions of internal TLAC, ESBG recommends not introducing this requirement for MREL.
  • Powers for resolution authorities to monitor, and potentially enforce, the MREL maturity structure: we agree that the resolution authority should monitor the MREL maturity structure and potentially make recommendations. However, there is a need to consider the work undertaken by competent authorities with respect to a review of the funding and liquidity management of the institution, which will include a review of the maturity profile of funding and also include recommendations as appropriate. Any duplication of analysis and – in a worst case – contradicting recommendations should be avoided. As such, ESBG is of the opinion that it should be the primary responsibility of the competent authorities to monitor the funding maturity profile, while the resolution authorities provide a 'second pair of eyes', only making recommendations deviating from those of the competent authorities if these can be justified from a resolvability perspective. Resolution authorities should not have the power to enforce an MREL maturity structure since the funding maturity structure (and strategy) of a bank is an important part of its liquidity and funding risk management framework. Maturity structures will also be influenced by market conditions (which may make longer-term funding more or less attractive/available) and pricing considerations in addition to risk considerations. In addition, the only requirement for MREL in terms of a maturity is a remaining maturity in excess of one year, so imposing any maturity structure that goes beyond this requirement is in ESBG's point of view not consistent with the legal requirements.
  • Requiring approval from competent authority only if a breach of the MREL requirement would occur as a result of the redemption of the instruments (as in TLAC term sheet): as already outlined in the introduction, we question the need to impose an approval requirement for MREL. Besides, this is currently not foreseen in the BRRD. The option to align the requirement with the TLAC requirement represents the maximum that could be required with respect to MREL. Nevertheless, even under this option, the question arises of how the approval process should be designed. In our view, this should not follow the same rules, documentation requirements or approval processes as those required for capital instruments, as this process has proven too time-consuming for both banks and competent authorities with decisions taking several months to be made. The EBA should consider which parts of the process could be simplified since eligible liabilities do not have the same characteristics as own funds. In addition, the requirements might be differentiated depending on the size and/or anticipated duration of the breach.

​An important point to consider with respect to the approval for redemptions is the interaction with the consequences of an MREL breach. Clearly, if option 4, as set out above, is considered and the resolution authorities approve the (temporary) MREL breach resulting from redemption, there should be no consequences for banks resulting from this breach.


2.           Comments on impact analysis

The EBA's interim report includes a section on quantitative findings which is based on a sample of banks from across the EU. The analysis highlights issues due to differences in business models, banking systems, etc. and analyses the impact of MREL under a number of scenarios. While the analysis appears relatively comprehensive, it still suffers from a number of weaknesses, according to our observations:


  • Completeness: the analysis does not cover the entire EU, i.e. a number of countries are missing entirely. While they, in aggregate, do not account for more than 30% of banking assets in the EU, issues that may affect individual countries are not considered. For instance, to one ESBG member, the exclusion of Croatia, Romania, Slovakia and Slovenia is of particular importance, as these markets represent significant operations for their group.
  • Consolidated vs. solo level: since MREL is a requirement at individual institution level, any quantitative impact analysis needs to be undertaken at individual institution level as well in order to determine whether there are potential issues related to specific markets, business models or types of institutions. The majority of the analysis is undertaken (as in previous exercises) at consolidated level and the EBA acknowledges that the analysis only includes 'a few' subsidiaries of EU parent institutions. As a result, the reported shortfall may significantly underestimate the actual shortfalls at legal entity level.
  • Pillar 2 requirements: while we appreciate the data constraints that led the EBA to consider a 'standard' Pillar 2 add-on of 2% for all banks, this vastly underestimates the true Pillar 2 add-on for many institutions and – as a result – leads to a significant underestimation of MREL requirements in all scenarios. The issue is particularly acute in some jurisdictions that set comparatively high Pillar 2 requirements and – coupled with the lack of analysis at solo-institution level – mask potential problems for individual institutions or individual markets.
  • Buffer requirements: the analysis only includes the capital conservation buffer and G-SII buffers where applicable, but excludes countercyclical buffers, O-SII buffers and systemic risk buffers. As, in particular, systemic risk buffers are significantly higher than G-SII buffers in most cases, and apply to a larger number of banks, the analysis vastly underestimates the actual MREL requirements of institutions across the EU and again masks effects in specific jurisdictions. This further exacerbates the issues already highlighted with respect to Pillar 2 requirements.
  • Deposit-funded banks[2]: the EBA's analysis considers the issue of deposit-funded banks, however, does so in an insufficient manner. The EBA chose two thresholds for the identification of a deposit-funded bank (30% and 40%). In ESBG's view, this approach is insufficient as there are banks for which retail deposits represent more than 50% of their funding base. The difference of over 10% of the funding base compared to the most conservative scenario employed by the EBA, and the resulting lack of MREL-eligible liabilities, is another factor contributing to the underestimation of the overall funding needed in the analysis.
  • Calibration scenarios: the report includes three calibration scenarios with the first and third scenarios described as unrealistic (and therefore for illustrative purposes only). While we agree that scenario 1 could be considered a 'best-case', scenario 3 is still far from a 'worst-case' scenario that could occur on the basis of the current RTS on MREL. Both scenario 2 and scenario 3 include only a standardised Pillar 2 add-on, no countercyclical, O-SII or systemic risk buffers and any buffer requirement (capital conservation and G-SII buffer where applicable) only once in the loss absorption amount. As already outlined above, actual buffer requirements (Pillar 2 and others) are higher than those assumed in the analysis and the EBA RTS also allows for the inclusion of all buffers in the recapitalisation amount. Including significantly higher buffer requirements and including them twice in the total MREL requirement would show the 'real' worst case scenario for European banks (and actually a scenario that is contemplated in interactions with the SRB), and should as such be quantified, since the analysis otherwise underestimates the true effects on the European banking system. In addition, the EBA mentions at several points in the report that the scenarios 'are likely to be too conservative' for some banks and then makes 'more realistic' assumptions. While ESBG agrees with this assessment, the scenarios are also not conservative enough for some banks and a more realistic assumption in their case would result in much higher MREL requirements (this is especially relevant with respect to Pillar 2 add-ons and systemic risk buffer requirements).

As a result of the shortcomings listed above, ESBG believes that the current quantitative analysis is not fully appropriate to convey a true picture of the impact of MREL requirements in the EU. Not only does it lack detail in terms of requirements at individual institution and/or jurisdiction level, 'glossing over' real national differences and issues for banks with particular business models, but, in our view, it also vastly underestimates true funding needs by ignoring significant amounts of buffer requirements and the possibility of the EBA RTS (and emerging practice of some national resolution authorities and the SRB) to require buffers also as part of the recapitalisation amount.



[1] Or the equivalent of 20% of RWA if the conditions of Art. 44(8) BRRD are met.

[2] Related to this point, we have concerns about the construction of Figure 6 in the report, which depicts the composition of MREL for different categories of banks. We assume the 'stacks' represent aggregate data, although this is not clear. More importantly, the stacks (at least those for O-SIIs and Other) would – in ESBG's view – look significantly different if disaggregated by jurisdiction, as the preponderance of deposit funded banks in some markets would be evident. As presented at the moment, this graph does not accurately represent the funding structure in many jurisdictions and as such creates an inaccurate picture of the situation in the EU banking sector, according to our impression.​

About ESBG (European Savings and Retail Banking Group)

ESBG brings together nearly 1000 savings and retail banks in 20 European countries that believe in a common identity for European policies. ESBG members represent one of the largest European retail banking networks, comprising one-third of the retail banking market in Europe, with 190 million customers, more than 60,000 outlets, total assets of €7.1 trillion, non-bank deposits of €3.5 trillion, and non-bank loans of €3.7 trillion. ESBG members come together to agree on and promote common positions on relevant regulatory or supervisory matters.

European Savings and Retail Banking Group – aisbl

Rue Marie-Thérèse, 11 ■ B-1000 Brussels ■ Tel: +32 2 211 11 11 ■ Fax: +32 2 211 11 99

Info@wsbi-esbg.org ■ www.wsbi-esbg.org


Published by ESBG. August 2016.

>> See .pdf version

European Institutions; European Supervisory Authorities (EBA-ESMA)