Response to consultation on draft ITS amending Regulation (EU) 2021/453 with regard to the specific reporting requirements for market risk
Go back
- The FBF welcomes the opportunity to contribute to the European Banking Authority (EBA) Consultation by providing its response to the key aspects of this FRTB reporting ITS. The FBF response represents the consolidated view from all FBF Members. Our comments mainly focus on identifying issues that are considered crucial for the industry.
Firstly, we note that the Draft ITS suggests 30th of September 2024 as the first reference date for reporting using the final ITS. The industry is in favor of the alignment of the reporting requirements with the calculation requirements, and therefore, believes it is more appropriate to align the date of application of the ITS amending the reporting requirements for market risk with the implementation of CRR3 FRTB requirements. Contrarily to the reason invoked by the EBA for early reporting requested to institution to disentangle the reporting requirements from the CRR3 one, we believe that such early landing of reporting requirement is creating more complexity for institution, while having a common reporting timeline would simplify the implementation of the new approaches.
-In general, for the banking industry, there is a multiplication of reporting and an increase in the amount of information to be provided. Indeed, the number of reports to be completed is very significant, and the information requested is highly detailed for a large number of entities. Indeed, banks are currently required to submit high-level information on the basis of the Alternative Standardised Approach for market risk (A-SA) since 2021 by filling two reporting templates, while the Draft ITS suggests that twenty-nine templates are required under the CRR3 market risk standards. The number of templates banks are asked to provide, combined with the very high granularity requested (displaying all the risks, all the buckets, for each type of underlying), as well as the very detailed level of information on every intermediate calculation step, are particularly burdensome for institutions and would trigger high costs of compliance.
- The stated objectives of the EBA are twofold [ITS article 20]: to have templates that reflects each and every step of the market risk own funds requirements calculation process and to facilitate the monitoring of risks. We understand those objectives, but the banking industry feels that some elements of proportionality should be introduced to avoid undue complexity (hence meeting the intentions of the EBA when meeting the objectives as expressed at article 12 of the ITS. Proportionality should not be understood only as adapting the reporting requirements to the size of the institution, but as well as setting the granularity of the reported calculation steps or risk metrics to levels of significance towards the bank overall risks and own funds requirements.
- In addition, the industry questions whether these reporting requirements will be maintained in the final version of CRR3, and how they will be implemented, given that in some versions of CRR 3, article 430b mandating the EBA has been deleted. The industry is also wondering about the possibility of transposing these requirements into a supervisory ITS.
Implementation of the offsetting group-concept:
The reporting breakdown at offsetting group level would exponentially multiply the reporting requirements and is seen as particularly burdensome.
The industry is worried about the implementation of this offsetting group-based reporting and does not consider it acceptable to have to fulfill all templates unless they are significant.
The industry suggests various alternative options:
- To report all templates for the most material offsetting group and to have lighter reporting requirements for non-material offsetting groups such as simplified set of templates.
- To report all templates for the most material offsetting group and to report all templates for non-material offsetting groups on an aggregated basis (i.e., through a second offsetting group which would be the sum of all non-material offsetting groups).
- To set a materiality threshold to assess accordingly non-material offsetting groups.
- Additional capital and detail per offsetting group could be provided on an ad-hoc basis, if necessary, thus avoiding an unnecessary increase and complexity in the reporting requirements. This should come together with some consideration of proportionality, whereby: calculation levels below a materiality threshold may require only a simplified reporting; low materiality calculation levels may be reported in aggregation.
In many areas, the set of templates requires data that are more granular than those provided for the semi-annually Basel Estimates. For instance, the worksheet “TB risk class” panel A on General Interest Rates Risk does not require a breakdown by tenor. The Basel Estimates template requires only the equivalent of the columns 1000 to 1080 (delta, vega and curvature per bucket and per scenario). In general, we see the Basel Estimates as sufficient to meet the objective of the EBA, some insights into the process for calculating own funds requirements and sufficient information to monitor risks embedded in the trading portfolios. Keeping the reporting requirements in term of amount of data and template layout close to those of the Basel estimates will have the additional benefit of a simpler and faster implementation.
Besides, for the SBM templates (from 92.01 to 92.07.01), the industry would like it clarified whether it’s the “own fund requirement” (Art. 325f(8)) or the “bucket-specific sensitivity” (Art. 325f(7)) which is intended in the columns 1000 to 1080. Indeed, the “own fund requirement” is calculated at the global level (delta, vega, curvature) when the templates are at the buckets level. In addition, the “own fund requirement” by correlation scenarios are already reported in the template 91. The “bucket-specific sensitivity” seems more appropriate for the SBM templates.
With regard to FX risk reporting (Templates C 92.07.1 (FX1) and C 92.07.2 (FX2)):
The industry considers that it would be operationally cumbersome and irrelevant to have to report on all existing currency pairs. The breakdown requested by all currencies is too granular and adds little supervisory value, and it is therefore suggested to report at risk currency level with the major cross currencies such as the top 10 currencies and to create a category ‘Other’ for insignificant currencies.
Reporting on RRAO:
The industry wants to highlight the very high granularity requested, which displays very detailed information on underlyings and instruments bearing other residual risks, and also needs clarifications about this granularity. Indeed, rules should be defined to enable the input of options with products that may fall into several categories. Similarly, the industry points to the problem of back-to-back operations, which would have to be reported twice, which could be avoided.
- Lastly, at places the ITS refers to the CRR2 framework and reporting requirement. We understand that it will have to be updated to meet the final provisions of CRR3. It is important that this is taking place fully and in due time before the ITS becomes applicable. For instance, the expression of the k-coefficient in point 52 of page 20 will have to be changed likely to k=0.5∙(∑_(i∈Y)▒〖SA〗_i )/(∑_(i∈G,Y)▒〖SA〗_i ).
- See also the answer to question 1 on the granularity of offsetting group-based reporting.
If the aim is to capture data relevant to P&L Attribution (PLA), then the reporting requirement should only be relevant to A-IMA banks and should only be used as a supervisory tool to ensure bank specific PLA test works appropriately.
Furthermore, the additional proposed requirement to provide daily P&L data for each offsetting group means a significant increase in effort, especially as the breakdown by risk classes exceeds the current requirements for internal models. The industry suggests that the P&L should only be provided as of the end of the quarter according to OFR. If the EBA was to proceed with additional P&L reporting for FRTB specifically, allocation of the P&L to risk classes should be based on internally used P&L explanation functionalities.
In addition, the requirement described in the ITS for this template is also for horizontal review purpose. Such horizontal review can be performed outside of the COREP reporting. Those templates have been filled in the past on ad hoc requests from the ECB, without having the huge burden of a quarterly production within the COREP templates.
As this report would add a huge burden for the Supervisory reporting, the industry is wondering what would be measured through such template and why it should be incorporated into the Supervisory reporting framework.
Additional information on the boundary between TB/BB should be provided on an ad-hoc basis, if necessary, rather than on a regular basis, thus avoiding an unnecessary increase in the reporting requirements.
a) Did you identify any issues regarding the implementation and use of the offsetting group-concept of Article 325b CRR in the context of these ITS?
General remarks:- The FBF welcomes the opportunity to contribute to the European Banking Authority (EBA) Consultation by providing its response to the key aspects of this FRTB reporting ITS. The FBF response represents the consolidated view from all FBF Members. Our comments mainly focus on identifying issues that are considered crucial for the industry.
Firstly, we note that the Draft ITS suggests 30th of September 2024 as the first reference date for reporting using the final ITS. The industry is in favor of the alignment of the reporting requirements with the calculation requirements, and therefore, believes it is more appropriate to align the date of application of the ITS amending the reporting requirements for market risk with the implementation of CRR3 FRTB requirements. Contrarily to the reason invoked by the EBA for early reporting requested to institution to disentangle the reporting requirements from the CRR3 one, we believe that such early landing of reporting requirement is creating more complexity for institution, while having a common reporting timeline would simplify the implementation of the new approaches.
-In general, for the banking industry, there is a multiplication of reporting and an increase in the amount of information to be provided. Indeed, the number of reports to be completed is very significant, and the information requested is highly detailed for a large number of entities. Indeed, banks are currently required to submit high-level information on the basis of the Alternative Standardised Approach for market risk (A-SA) since 2021 by filling two reporting templates, while the Draft ITS suggests that twenty-nine templates are required under the CRR3 market risk standards. The number of templates banks are asked to provide, combined with the very high granularity requested (displaying all the risks, all the buckets, for each type of underlying), as well as the very detailed level of information on every intermediate calculation step, are particularly burdensome for institutions and would trigger high costs of compliance.
- The stated objectives of the EBA are twofold [ITS article 20]: to have templates that reflects each and every step of the market risk own funds requirements calculation process and to facilitate the monitoring of risks. We understand those objectives, but the banking industry feels that some elements of proportionality should be introduced to avoid undue complexity (hence meeting the intentions of the EBA when meeting the objectives as expressed at article 12 of the ITS. Proportionality should not be understood only as adapting the reporting requirements to the size of the institution, but as well as setting the granularity of the reported calculation steps or risk metrics to levels of significance towards the bank overall risks and own funds requirements.
- In addition, the industry questions whether these reporting requirements will be maintained in the final version of CRR3, and how they will be implemented, given that in some versions of CRR 3, article 430b mandating the EBA has been deleted. The industry is also wondering about the possibility of transposing these requirements into a supervisory ITS.
Implementation of the offsetting group-concept:
The reporting breakdown at offsetting group level would exponentially multiply the reporting requirements and is seen as particularly burdensome.
The industry is worried about the implementation of this offsetting group-based reporting and does not consider it acceptable to have to fulfill all templates unless they are significant.
The industry suggests various alternative options:
- To report all templates for the most material offsetting group and to have lighter reporting requirements for non-material offsetting groups such as simplified set of templates.
- To report all templates for the most material offsetting group and to report all templates for non-material offsetting groups on an aggregated basis (i.e., through a second offsetting group which would be the sum of all non-material offsetting groups).
- To set a materiality threshold to assess accordingly non-material offsetting groups.
- Additional capital and detail per offsetting group could be provided on an ad-hoc basis, if necessary, thus avoiding an unnecessary increase and complexity in the reporting requirements. This should come together with some consideration of proportionality, whereby: calculation levels below a materiality threshold may require only a simplified reporting; low materiality calculation levels may be reported in aggregation.
Is it clear how positions in CIUs are to be reflected in the three template groups (SBM, RRAO, DRC) of the A-SA templates? If you identify any issues, please suggest how to clarify their treatment in the templates and/or instructions.
Instructions on how positions in CIUs shall be reflected are clear.a) Did you identify any issues regarding the representation of A-SA (policy) framework in the reporting templates?
For banks using internal models, precisions are expected on reporting requirements for A-IMA desks under A-SA. Indeed, instructions (particularly points 47 and 48 of the EBA consultation paper) are open to interpretation and hence further clarification is needed on which information should be populated on A-SA templates for A-IMA desks (all SA templates with all steps of calculations and metrics or only the results? if so, which specific cells and/or templates are expected?).clear? If you identify any issues, please clearly specify the affected templates and instructions and include suggestions how to rectify the issues.
Reporting on Sensitivities Based Method:In many areas, the set of templates requires data that are more granular than those provided for the semi-annually Basel Estimates. For instance, the worksheet “TB risk class” panel A on General Interest Rates Risk does not require a breakdown by tenor. The Basel Estimates template requires only the equivalent of the columns 1000 to 1080 (delta, vega and curvature per bucket and per scenario). In general, we see the Basel Estimates as sufficient to meet the objective of the EBA, some insights into the process for calculating own funds requirements and sufficient information to monitor risks embedded in the trading portfolios. Keeping the reporting requirements in term of amount of data and template layout close to those of the Basel estimates will have the additional benefit of a simpler and faster implementation.
Besides, for the SBM templates (from 92.01 to 92.07.01), the industry would like it clarified whether it’s the “own fund requirement” (Art. 325f(8)) or the “bucket-specific sensitivity” (Art. 325f(7)) which is intended in the columns 1000 to 1080. Indeed, the “own fund requirement” is calculated at the global level (delta, vega, curvature) when the templates are at the buckets level. In addition, the “own fund requirement” by correlation scenarios are already reported in the template 91. The “bucket-specific sensitivity” seems more appropriate for the SBM templates.
With regard to FX risk reporting (Templates C 92.07.1 (FX1) and C 92.07.2 (FX2)):
The industry considers that it would be operationally cumbersome and irrelevant to have to report on all existing currency pairs. The breakdown requested by all currencies is too granular and adds little supervisory value, and it is therefore suggested to report at risk currency level with the major cross currencies such as the top 10 currencies and to create a category ‘Other’ for insignificant currencies.
Reporting on RRAO:
The industry wants to highlight the very high granularity requested, which displays very detailed information on underlyings and instruments bearing other residual risks, and also needs clarifications about this granularity. Indeed, rules should be defined to enable the input of options with products that may fall into several categories. Similarly, the industry points to the problem of back-to-back operations, which would have to be reported twice, which could be avoided.
a) Did you identify any issues regarding the representation of A-IMA (policy) framework in the reporting templates?
The aim of the A-IMA templates is to assess and monitor the quality of the models. In this sense, the industry is questioning the relevance of the level of granularity of the information requested by EBA.clear? If you identify any issues, please clearly specify the affected templates and instructions and include suggestions how to rectify the issues.
- On templates C 98.01.1 (DRC1), C 98.01.2 (DRC2): Information are requested on intermediary steps of calculation, and it seems to be very far from capital information. We suggest reporting only figures needed for capital purposes and relevant from a capital standpoint.- Lastly, at places the ITS refers to the CRR2 framework and reporting requirement. We understand that it will have to be updated to meet the final provisions of CRR3. It is important that this is taking place fully and in due time before the ITS becomes applicable. For instance, the expression of the k-coefficient in point 52 of page 20 will have to be changed likely to k=0.5∙(∑_(i∈Y)▒〖SA〗_i )/(∑_(i∈G,Y)▒〖SA〗_i ).
- See also the answer to question 1 on the granularity of offsetting group-based reporting.
Does this approach work for you? Or do you need any further, or different, guidance regarding the elements of the P&L and the scope of the positions to be covered by that P&L? Which additional specifications could facilitate your compliance with this reporting requirement? Which general methodology would you envisage to allocate P&L to the risk classes of the sensitivities-based method?
The industry suggests that this template should be deleted from the scope of the ITS. We believe that this information is being delivered in other COREP templates, which while unrelated to FRTB, still contain the official risk information and the same measure sought in this proposed report. Therefore, this potential template would only add more reporting burden and will duplicate the information delivered to supervisory authorities.If the aim is to capture data relevant to P&L Attribution (PLA), then the reporting requirement should only be relevant to A-IMA banks and should only be used as a supervisory tool to ensure bank specific PLA test works appropriately.
Furthermore, the additional proposed requirement to provide daily P&L data for each offsetting group means a significant increase in effort, especially as the breakdown by risk classes exceeds the current requirements for internal models. The industry suggests that the P&L should only be provided as of the end of the quarter according to OFR. If the EBA was to proceed with additional P&L reporting for FRTB specifically, allocation of the P&L to risk classes should be based on internally used P&L explanation functionalities.
In addition, the requirement described in the ITS for this template is also for horizontal review purpose. Such horizontal review can be performed outside of the COREP reporting. Those templates have been filled in the past on ad hoc requests from the ECB, without having the huge burden of a quarterly production within the COREP templates.
a) Did you identify any issues regarding the representation of the prudential framework for reclassifications and the associated own funds requirement in the reporting template?
There might be exceptional and very rare cases where a prudential reclassification is triggered by an accounting reclassification under IFRS 9 in case of a modification of the business model under which the financial asset is held (for instance in or outside of the Held for Trading business model). The determination of a business model is not done when a particular instrument is classified for accounting purposes but shall be carried out at a higher level, namely at the level that reflects how groups of financial assets are managed together to achieve a particular business objective. In such an exceptional and rare case, regarding the scope of transactions, the template should reflect the granularity of this higher accounting level (Business Model Organization of financial assets) and not a reporting at the level of each individual instrument.b) Which benefits and challenges do you foresee as regards this reporting? Which issues should be taken into account or addressed, to maximise the benefit and reduce the cost of compliance with this particular reporting requirement?
The industry is opposed to this template. Indeed, it is of no interest and does not meet the CRR3 criteria. According to French banks, it is not part of the EBA's mandate and is not relevant in the short term.As this report would add a huge burden for the Supervisory reporting, the industry is wondering what would be measured through such template and why it should be incorporated into the Supervisory reporting framework.
Additional information on the boundary between TB/BB should be provided on an ad-hoc basis, if necessary, rather than on a regular basis, thus avoiding an unnecessary increase in the reporting requirements.