Response to consultation on draft ITS on Pillar 3 disclosure

Go back

Question 1: Are the amended/new templates EU OV1, EU KM1, EU CMS1, EU CMS2 and the related instructions clear to the respondents? If no, please motivate your response.

The CRR3 text is very technical, so will be complex for all analysts and investors. Pillar 3 must therefore be as readable as possible for them. 

 

The internal models developed by banks and validated and regularly challenged by European supervisors remain the major standard of supervision, with the new add-on or output floor which will limit the reduction compared to the use of approaches standards. 

 

It is core to carefully segregate the two notions between RWA as published today and the add-on due to output floor which is ultimately calculated globally. This should clearly appear in the Template EU KM1 on institutions’ key metrics that provides the summary of the main prudential and regulatory information and ratios covered by the CRR. The instructions should be corrected.

 

Template OV1

 

We understand that this template requires banks to disclose TREAs by risk categories and by approaches for both opening and closing reference dates. However, the regulatory text requires banks to disclose only TREA at consolidated level. 

 

Shall we understand that there is a misleading denomination of what is required in the template? Indeed, banks have always claimed during the Basel 3 finalization phase that they would be ready to disclose U-TREA by risk and by approach but also that TREA in the sense of article 92 would be calculated only at consolidated level for output floor purposes.

We would then welcome a clarification detailing that only U-TREA are required to be breakdown in this template. Same logic applies for the column requesting own funds requirements. And in consequence of this modification, align the mapping tool we the correct formula.

 

Besides, for the sake of clarity in the reading of the template, we would recommend suppressing all rows that are labelled as “not applicable”.

 

Template EU KM1

 

The CRR3 text is very technical, so will be complex for all analysts and investors. Pillar 3 must therefore be as readable as possible for them. 

 

The internal models developed by banks and validated and regularly challenged by European supervisors remain the major standard of supervision, with the new add-on or output floor which will limit the reduction compared to the use of approaches standards. 

It is core to carefully segregate the two notions between RWA as published today and the add-on due to output floor which is ultimately calculated globally. This should clearly appear in the Template EU KM1 on institutions’ key metrics that provides the summary of the main prudential and regulatory information and ratios covered by the CRR. The instructions should be corrected. 

Please refer to question 4 for more detailed corrections/amendments that we would require.

 

Templates EU CMS1 & EU CMS 2

 

Accordingly, with BCBS recommendations and CRR article 438, the EBA proposes two new templates:

  • CMS 1 for which purpose is to compare full standardized risk-weighted assets (S-TREA) against normally calculated RWA ie U-TREA (splitting between standardized and IRB approaches). The template also provides the floored RWA taking into account the phase-in arrangements. The breakdown of these amounts is made according to risk categories, as defined by CRR.
  • CMS 2 for which purpose is to compare risk-weighted assets U-TREA and S-TREA for credit risk as above but with a breakdown at the asset class level as defined in Part III.

 

However, when checking the proposed template CMS 2, we observe that the breakdown is far more granular than the simple asset classes as defined in CRR articles 112 (standard approach) and also embeds categories under article 147 (IRB approaches), adding confusion and complexity for users.

 

Not only does it go way beyond CRR requirements as described in article 438, it also contravenes with Pillar 3 basic principles of clarity and transparency for the public. It is furthermore burdensome for banks, which have a short delay to adapt the breakdown of their reportings to the new granularities.

 

Consequently, we suggest coming back for CMS2 to the BCBS proposal with a simple breakdown by asset exposure classes according to the standardized approach which we deem is sufficient as information for the public. 

 

Furthermore, we question the relevancy of the row 7 (named “not applicable” with a typo) for the sake of clarity.

 

Besides, we note that for both CMS1 and CMS2, formulas used for the mapping tool are not updated for the new column EUd. They indeed refer to reporting C07 while the standardized approach does not apply to this template. A double consistency check shall be performed by the EBA teams to ensure that Pillar 3 templates are correctly filled with Pillar 1 reporting: EBA mapping tool shall indeed meet the self-imposed purposes of completeness and accuracy. 

Question 2: Do the respondents identify any discrepancies between these templates and related instructions and the calculation of the requirements set out in the underlying regulation?

The discrepancies we have observed are related to the mapping tool and its lack of updates and included in each question in regard of the template concerned.

Question 3: Do the respondents agree that the amended draft ITS fits the purpose of the underlying regulation?

As mentioned in our introduction, the granularity required for credit risk exposure classes beyond CRR requirements and the proposal to disclose a fully loaded solvency ratio are limits hindering us to claim the draft ITS fits the underlying regulation.

While the text is rather straightforward in regard of what is required, the amended draft ITS exceeds the limits of simplicity goals set by the regulators, particularly about:

  • Pillar 3 aims to provide clear and relevant guidance to the public
  • Paragraph 1 of the introduction « Pillar 3 of the Basel framework aims to promote market discipline through regulatory disclosure requirements. These requirements enable market participants to access key information relating to a bank’s regulatory capital and risk exposures in order to increase transparency and confidence about a bank’s exposure to risk and the overall adequacy of its regulatory capital”
  • Pillar 3 is based on Pillar 1 and not the contrary.
  • Paragraph 2 of the introduction “The revised Pillar 3 disclosures in this document focus on regulatory measures defined in Pillar 1 of the Basel framework, which requires banks to adopt specified approaches for measuring credit, market and operational risks and their associated resulting risk-weighted assets (RWA) and capital requirements.”
  • The output floor, which is required to be calculated at consolidated level only and taken into account all relevant provisional transitory arrangements. However, several templates (such as OV1) seem to require either fully loaded solvency ratios/own funds requirements (i.e. with a 72,5% percentage on S-TREA) while others seem to require a very granular breakdown (such as CMS2)
  • The granularity of credit risk exposures, e.g. CMS2 compiling both exposures classes according to the SA with the ones according to the IRB approach(es). It adds confusion and contravenes with the purpose of clarity set by regulators.

 

For the sake of clarity, we would recommend aligning notions and concepts in this draft ITS with the ones detailed in the underlying regulation (CRR3).

 

Question 4: In particular, regarding the disclosure of the output floor, do respondents agree with the inclusion of rows EU 5c, EU 6c, EU 7c in template EU KM 1 and the column EU d in templates EU CMS1 and EU CMS2? Please provide the rationale behind your answer.

Europe has recognized the need for long transitional measures, some of which could be further extended. This is to allow banks to adapt to the international context despite the lack of relevance of the far too high loss measurements which constituted their basis (notably on residential real estate). 

 

The multiplication of ratios ((i) “normal” and (ii) “fully loaded” without the phasing of the output floor and (iii) “full fully loaded” without the application of the transitional measures, far from bringing transparency, generates confusion and non-experts systematically align themselves with the least favorable ratios for banks.

 

As detailed in general remarks, we do not share the choice of option 3b which would go completely against the European text. Given the complexity of the CRR3 text, in the event of publication of two ratios, one being current and the other said to be fully loaded, analysts and investors would align themselves - as was the case in other circumstances - on the weaker ratio and the contribution of transitional measures would be purely erased. 

 

It’s a major attention point for French banks.

 

The Output floor is a global calculation and should not be broken down by assets or type of assets. When there are IRB measures in a bank, this involves showing the total standard RWA at 100% (which should be detailed separately) and displaying the formula with 72.5% to produce the impact of the output floor in a unique single figure which depends on all the components of risks used.

 

Besides, the Output floor is an overall calculation and depends on all risk components used in the relevant banking entity. The output floor therefore has the original characteristic of being non-additive so as to consolidate. 

 

In the case of a group where the Member State authorizes the exemption from the application of the output floor at the level of the sub-entities making up the group in its country, it will be very important to specify the absence of declaration of output floor elements at the level of these exempt entities. Indeed, they are not relevant at the highest level of consolidation as there will be this specific calculation at this group level.

 

On this question, the EBA indicated during the virtual public hearing on January 23 that the reporting would only apply at consolidated level and that they will consider making this clear in the instructions.

Accordingly, we would welcome that this clarification is added to the ITS instructions. 

 

As for added rows EU5c, EU6c and EU7c in the template KM1, we find the mention “fully loaded output floor” very misleading and confusing. The underlying regulation requires only an output floor taking into account all transition arrangements (including phase-in). We consequently do not agree with the introduction of these rows for the sake of clarity and efficiency of the Pillar 3 disclosures. We ask the EBA to amend the template KM1 accordingly.

 

As for column EUd added to the templates CMS1 and CMS2, we welcome the addition as far as it responds to disclosure requirements described in article 438.e. However, for the sake of clarity, we recommend suppressing the column d requiring to disclose the amounts related to S-TREA in fully loaded version (full standardized approach) as far as:

  • it is not a metric required by the regulation for disclosures,
  • the metric required by the regulation is the one disclosed in the column EUd
  • “no useless duplicate for the Pillar 3” principle 

 

Question 5: Are the amended templates EU CR 4, EU CR 5 and the related instructions clear to the respondents? If no, please motivate your response.

As templates referring to exposures for which the standardized approach is applied, CR4 and CR5 shall refer to article 112 of CRR3 in regard of breakdown exposure classes.

As a recall of our remarks brought to the consultation in regard of supervisory reportings, the granularity of the exposure class “secured by mortgage immovable property” is quite problematic as it oversteps the article 112 and adds burdens in the production process for banks, all the more as the new framework is imposed in a very short delay (less than one year between the final draft ITS and the first production).

To comply with the underlying regulation, we advise to stick with the proposition of Basel in its 2018 Updated Disclosure Framework.

This answer is to be recalled in the one to Question 8.

 

Besides, for the denomination of rows, we advise to uniformize commas and dots for numbers (namely 9.3 instead of 9,3 for instance).

 

Question 6: Do the respondents identify any discrepancies between these templates and instructions and the calculation of the requirements set out in the underlying regulation?

No comments at this stage.

Question 7: Do the respondents agree that the amended draft ITS fits the purpose of the underlying regulation?

Same answer as for Question 3.

 

Question 8: In particular, for templates EU CR 4 and EU CR 5, do the respondents agree with the reconciliation of the row numbering with the Basel one in the corresponding templates? Please provide the rationale behind your answer.

As mentioned above, we do not agree with the fact that the templates proposed by the EBA go beyond requirements from Basel Disclosure Framework. We deem that the EBA oversteps its mandate by detailing a granularity not required by the underlying regulation.

 

Furthermore, we question the relevancy of rows named “not applicable”. The set of templates must be easily readable and usable for the public, the inclusion of useless data and/or format contravenes with this purpose.

Question 9: Are the amended templates EU CR 6, EU CR 6-A, EU CR 7, EU CR 7-A and the related instructions clear to the respondents? If no, please motivate your response.

No comments at this stage.

Question 10: Do the respondents identify any discrepancies between these templates and instructions and the calculation of the requirements set out in the underlying regulation?

No comments at this stage.

Question 11: Do the respondents agree that the amended draft ITS fits the purpose of the underlying regulation?

Same answer as for Question 3.

Question 12: Regarding the template EU CR 7, do the respondents agree with reconciliation of the row numbering with the Basel one in the corresponding templates? Please provide the rationale behind your answer.

As a template referring to exposures for which the IRB approach is applied, CR7 shall refer to article 147 in regard of breakdown exposure classes.

 

We do not have specific comment to bring in regard of this aspect even though we question the relevancy of rows named “not applicable”. The set of templates must be easily readable and usable for the public, the inclusion of useless data and/or format contravenes with this purpose.

Question 13: Do the respondents agree with the deletion of the rows on SMEs in templates EU CR 6-A, EU CR 7 and EU CR 7-A?

We welcome the decision to simplify and alleviate the burden in disclosure requirements.

Question 14: Are the amended/new templates EU MRA, EU MRB, EU MR1, EU MR2, EU MR3 and the related instructions clear to the respondents? If no, please motivate your response.

Template EU MR2 - Market risk under the alternative internal model approach (AIMA)

 

In Template EU MR2 related to Market risk under AIMA, for rows 11 to 16 the expected value for columns a & e is the own funds requirement not the most recent measures (applicable for rows 1 to 10 only), hence, this template, as is, might be quite misleading as this precision is only specified in Annex X and it mixes contradictory and uncorrelated information.

 

To ease the reading of information and to avoid the conjunction of uncorrelated information, would it be possible to reshape this template to better differentiate information related to rows 1 to 10 (daily values) vs those related to rows 11 to 16 (own fund requirements)?

 

Template EU-MRB – Qualitative disclosures requirements for institutions using the A-IMA

 

In theory, qualitative disclosures are provided by institutions on the last quarter of each year. 

Respondents feel that, to ease the understanding of the new figures for market participants, it would be more relevant to start to report the template EU-MRB once the first remittance of quantitative disclosures related to the A-IMA starts (narrative support accompanying the figures).

 

In this sense, the EBA should let the opportunity to disclose this template to those institutions which might want to send a narrative with the 1st figures.

Question 15: Do the respondents identify any discrepancies between these templates and instructions and the calculation of the requirements set out in the underlying regulation?

Reporting requirements if FRTB is postponed. 

 

If FRTB is postponed, the respondents understood from the EBA during the Public Hearing that they shall use the current version of market risk disclosures. Instructions should be amended to precise this information and reflect the reporting requirements for Pillar 3 under this scenario.

 

Question 16: Do the respondents agree that the amended draft ITS fits the purpose of the underlying regulation?

No comments at this stage.

Question 17: Regarding the template EU MRB, do the respondents agree with the reconciliation of the row numbering with the Basel one in the corresponding template? Please provide the rationale behind your answer.

No comments at this stage.

Question 18: Are the new templates EU CVAA, EU CVA 1, EU CVAB, EU CVA2, EU CVA 3, EU CVA 4 and the related instructions clear to the respondents? If no, please motivate your response.

As for the template EU CVA2, we note that the row 3 is named “total”, implying it would be the sum of rows 1 and 2. 

However, it is not the case: the total of RWAs under full basic approach contains amounts that are reported under neither the rows 1 nor 2. 

Consequently, we would welcome a more explicit name for this row so that neither producers nor users are misled by this confusing denomination.

Question 19: Do the respondents identify any discrepancies between these templates and instructions and the calculation of the requirements set out in the underlying regulation?

We have not identified major discrepancies between the template’s requirements and the underlying regulation but 2 points:

  • CCR2 template is not included into the Pillar 3 consultation but as this template is fully mapped with report C25 and as C25 is totally modified, we considered that this template should be cancelled.
  • Template EU CVA4’s title indicates that it is related to the standardized approach, but the mapping indicates that all data from the total of all approaches is reported.

Question 20: Do the respondents agree that the amended draft ITS fits the purpose of the underlying regulation?

In order to provide the required disclosure on CVA risk, Annex XLI has been developed, in accordance with the Pillar 3 requirements of CRR3 and with its changes in principles on CVA Risk.

As far as CVA risk and the related new templates, we have not identified breaches with the underlying regulation.

 

On the other hand, the template EU CCR2 “Transactions subject to capital requirements for CVA risk” relating, under CRR2, on CVA risk is not included in the statements delivered in the Consultation Paper Step 1 – Part 1 of the EBA. If this template were to be maintained for the first CRR3 publications in 2025, this would raise obvious problems since it’s not in its current format compliant with CRR3 (e.g. row 1 “Advanced method” refers to the Article 383 of the CRR which will now relate to the Standardized Approach and row 4 “Standardized Method” refers to Article 384 “of the CRR which will now relate to the Basic Approach”).

 

We believe that template CCR2 should be deleted, or at least amended, in the final version of the CRR3 Pillar 3 templates. Otherwise, we need new instructions from EBA to explain us how to complete this template which will no longer comply with the CCR3 principles.

 

Question 21: Do the respondents consider that the “mapping tool” appropriately reflects the mapping of the quantitative disclosure templates with supervisory reporting templates?

On Pillar 3, we have identified issues on the mapping tool proposed and more globally consistency of column / row asked mentioned in the previous questions.

 

We would add that, as a recall of our remarks brought to the consultation about reporting sheets are still not all updated (for instance, the one referring to “Purchased receivables” asset class, still denominated as s00XX) so that it is difficult to assess the consistency to date of the mapping tool.

 

REMARKS ON TEMPLATE CCR2

 

This template aims at disclosing EADs and RWAs for transactions subject to own funds requirements for CVA risk. It is not in the scope of modifications/additions/suppressions brought by the upcoming regulation.

However, this regulation also requires the implementation of brand-new templates dedicated to CVA risk. Hence a risk of useless duplicate in disclosures. We the request, for the sake of Pillar 3 clarity, the suppression of the template EU CCR 2. 

 

REMARKS ON TEMPLATE CCYB2

 

The template is not in the scope of the ones modified, added, or suppressed by the CRR3 regulation.

However, we note that the first row is still labelled as “total risk exposure amount” (TREA) while this term is now associated with a specific requirement from CRR3.

This name could be confusing for both the users and the producers of Pillar 3.

Which is why we kindly ask the EBA to clarify this mention by denominating the row as “unfloored total risk exposure amount, U-TREA” as far as it is the metric required by regulators for this section.

 

REMARKS ON TEMPLATE CR6A

 

Regarding the template used for mapping tool of column “a”, there are gaps with regulatory template:

  • Rows 5.1 and 5.3 are grayed in mapping tool template and ungrayed out in the regulatory template.
  • Rows 6.3 and 6.4 are ungrayed and not empty in mapping tool template (“{C 08.07, r0125, c0010}” for row 6.3 and “no mapping to C 08.07” for row 6.4) and grayed in the regulatory template.

There is no mapping tool integrated for column “d” (existing under CRR2). They may probably be mapped to C08.07?

 

REMARKS ON TEMPLATE CR6

 

The mapping tool for column j has not been modified to take into account the changes of column references. In our opinion, formula (Column j / Column e) should be replaced by (Column i / Column d).

 

Upload files

Name of the organization

Fédération Bancaire Française