Q1/Q2. Format of self-assessment template: Is the Excel format adequate for the bank to provide the information needed to assess the resolvability of the bank to provide justification? If not, please suggest alternatives.
We thank the SRB for this consultation, and we appreciate the extension of such practice.
Here are some general remarks on the proposed format. What are, according to us, the main potential hurdles we see. How we propose to remediate them.
1. Too much granularity with duplicates or near duplicates, often going beyond Expectations for Banks and guidance’s
We advocate to significantly reduce the number of capabilities, suppressing numerous duplicates or near duplicates, taking more into account proportionality, sticking to current expectations, and ensuring a consistent resolvability assessment over time.
The level 4 corresponding to ‘best practices’ of some banks should notably be removed and best practices be discussed with the banks through the IRT channel when relevant. This level 4 raises particular concerns in terms of proportionality, relevance and stabilisation of SRB expectations.
2. Bottom-up, bit by bit, approach bias
Very granular and mechanical approaches should be avoided as they could miss the objective of striking the right balance, by providing the illusion of precision while being hardly appropriate is many cases.
Resolvability should remain an idiosyncratic matter and integrate a global view of the articulation of capabilities for a given bank with its specific characteristics (business model, organisation, legal structure, and so on), rather than an addition of heterogeneous or potentially not relevant capabilities.
3. Validation of deliverables by the Board of Directors, notably testing plan, to be reconsidered
Validations requested by the SRB should be maintained at executive management level (i.e. senior executive responsible for resolution). A regular involvement of the Board of Directors on such technical and operational matters is not appropriate (not being part of its normal role) nor practically feasible, particularly for large groups.
4. Non EBA based scoring grades, heterogeneous across the grid
The semantic of the scoring proposed i.e. ‘compliant’, ‘largely compliant’, ‘materially non-compliant’, and ‘non-compliant’ raises concerns.
The grading “High”; “Medium”; “Low” proposed by the EBA looks like a fair base for an assessment at our level. If needed, the medium grade could be split in “Medium-High” and “Medium-Low”.
All capabilities should propose the same scoring grades.
5. Unrealistic expectation of operationalisation of all the liabilities included in the bail-in scope
The SRB should take into account the potential candidates for discretionary exclusions. We also advocate for the introduction of some materiality in the identification of the liabilities to be operationalised on Level 2 and in even more for all the liabilities targeted in the Level 3 (as mentioned Level 4 that should be suppressed).
Q3/4: Scope/frequency: Are the envisaged scope of application (at the resolution group level, covering also non-resolution entities) and frequency (yearly) for the submission of the self-assessment report well calibrated? If not, please explain.
The SRB should exclude non-resolution entities from the RSA scope, as in practice the resolution entity is the only entity where the resolution would occur.
We propose to stand by a yearly update at a maximum, articulated with the Budgeted Work Plan with no duplication of work i.e. testing action plan to be developed in the Budgeted Work Plan only and at a less granular level, not in the RSA grid per capability.
Q5/6: Granularity: Is the overall content and the number of capabilities described in the self-assessment template sufficiently detailed and comprehensive to cover the spectrum of progress made by banks? If not, please propose some concrete examples of new capabilities you would suggest introducing.
Please refer to our answer on question 2. We propose to significantly reduce the number of capabilities, eliminating duplicates or near duplicates, sticking to the ones in line with the current, published expectations and the proportionality principle, and ensuring stable resolvability assessment criteria over time.
Q7:Advanced capabilities: Level 4 represents advanced capabilities providing more granularity on the extent to which certain banks demonstrate their crisis preparedness. How can these capabilities be set out best to support banks’ work on ensuring resolvability? Would you recommend that additional capabilities be added to Level 4? If yes, please detail.
Level 4 capabilities go beyond the existing requirements and should be suppressed as they do not fit in a mandatory resolvability self-assessment with the potential implementation of remedial measures. Deemed best practices of certain banks, potentially linked to own specificities, should not be translated to an integral part of the assessment of all banks, which would be irrelevant and/or disproportionate in many cases. The assessment grid should be limited to what is strictly necessary for the purpose of resolving a bank.
Q8: Variant strategies (I): In order to assess progress on the operationalisation on the variant strategy, would you prefer to reflect such progress in one single column of the self-assessment template covering both Preferred Resolution Strategy (PRS) and Variant Resolution Strategy (VRS) (if applicable) or in two separate columns of the same template? Please explain.
A single column may be sufficient. In our view, variant strategies should be considered as complementary, adding optionality to the preferred strategy. Ideally, particularly for banks with Open Bank Bail-In as main strategy, variant strategies should be combined with it and become part of a revised, more optional, preferred resolution strategy.
Q9: Variant strategies (II): In your view, which resolvability capabilities included in the self-assessment template are the most relevant for assessing the operationalisation of the VRS as compared to the PRS?
Resolvability capabilities pertaining to dimension 7 of the self-assessment template (for banks with OBBI as current PRS).
Q10: Market transparency - aggregated level: What level of detail would you consider useful for benchmarking (e.g. by business model or bank size or by country)?
Distinction between G-SIB vs non-GSIB, country/group of countries, business model would be welcome.
Q11: Market transparency - individual level: Banks remain free, at their discretion, to disclose information on their resolvability progress and related activities and/or to make reference to potential authorities’ publications. Would you envisage such disclosures, if applicable, for your bank? If not, what considerations/concerns do you have in this regard?
A voluntary disclosure on resolvability self-assessment would entail a level playing field issue. We see no ground for such disclosure at bank level, nor potential interest for the market in general.
Q12: Scenarios-based assessment: Do you agree that the self-assessment could be completed under different scenarios or crisis events (e.g. defined by the resolution authority) to achieve better preparedness? Please comment.
We would strongly advise against scenario-based assessments, which would over-complexify the exercise and exceed any reasonable proportionality. We do not see the need of it in the resolvability self-assessment, as we are assessing many capabilities that should be available in potentially any crisis scenario.
Q13/14: Link to testing: Is the self-assessment template adequate to identify the areas that have been tested/ to be tested? If not, please explain.
The link to testing should be limited to tests performed, out of deep dives and on-site inspections that are of a different nature. Test to be performed should be tacked in the Budgeted Work Plan and excluded from the RSA grid.
Name of the organisation: BNP PARIBAS