Response to consultation on draft Guidelines on resubmission of historical data
Go back
However, we are of the opinion that the target set by the EBA itself in Recommendation No. 25 in the Study of the cost of compliance with supervisory reporting requirements, i.e. that the Guidelines on resubmissions that are the subject of this consultation should cut the ongoing reporting costs for SNCIs by 1.0 to 1.5%, cannot be achieved with the measures being proposed. To reduce reporting costs, we consider it necessary, among other things, to introduce materiality thresholds as well as further measures (see option 3a that was rejected according to the accompanying information on the draft Guidelines, our concrete proposals on resubmission periods (see the following and under question 5a) as well as on a general principle under question 3).
In particular, we welcome the definition of a period limited to one calendar year for resubmissions. This should be defined as the maximum period for resubmissions. A supervisory authority should only have the power to extend this period in very limited and justified exceptional cases.
For further discussions, we believe that close integration with the ECB’s activities on the topic of resubmissions is necessary. It is unclear from the Guidelines to what extent the ECB’s considerations on “significant resubmissions” are also being picked up by the EBA and are compatible with the Guidelines. The ECB’s approach purposefully provides for the introduction of thresholds based on a list of KRIs, which in our view represent a proportionate basis for requiring the resubmission of historical data (see ECB presentation entitled “Updates on the resubmission framework” of 26 April 2023). We ask that these considerations be taken into account when considering the inclusion of materiality thresholds. The Final Report of the EBA’s Cost of Compliance Study also states the following: “‘Immaterial’ or ‘unnecessary’ resubmissions should be avoided both from the reporting entities’ and data recipients' point of view”. In our opinion, the draft does not fully meet this objective. Limiting resubmission to one year does not prevent the submission of “immaterial” or “unnecessary” data. We can understand the need for reliable and accurate data, but we also believe that a large number of the required resubmissions, many of which result from breaches of validation rules, represent changes compared with the original submission that are so minor that they do not compromise the accomplishment of supervisory responsibilities. For example, there are no additional insights in our view if the resubmitted data only has a minor impact on relevant ratios (e.g. <0.1 percentage point change in the Common Equity Tier 1 ratio).
In light of this and the ECB’s ongoing 1-year trial phase for the KRIs, we consider a proposed initial application of the Guidelines as of 31 December 2023 to be premature. We suggest waiting for the ECB trial phase to be completed before finalising the EBA Guidelines and ideally following up with a second trial phase involving less significant institutions in consultation with the ECB, so as to build on the findings from the first phase of the ECB’s work and to assess whether the thresholds identified by the ECB can provide a basis for exempting LSIs from the resubmission of data points. The ECB’s approach assumes that changes below the relevant thresholds do not require explanation and are therefore immaterial. In our view, this approach can be applied to LSIs under a favourable proportionality approach, which could ultimately lead to a complete exemption from immaterial resubmissions under certain conditions.
In our opinion, the focus of resubmissions should generally be on correcting material errors. The present draft should therefore be supplemented by a definition of which data is erroneous. There is currently no classification as to whether core regulatory data or memo items are erroneous. When it comes to a materiality threshold, a distinction should be considered as to whether material changes arise, for example in own funds, or whether memo items that ultimately do not affect a minimum requirement that is relevant for management purposes are erroneous.
Furthermore, we also welcome the EBA’s clarification that newly published Q&As should not lead to resubmissions for previous reporting dates. Furthermore, we take the view that clarifications should be included in the Guidelines on how to deal with EBA validation rules that are cancelled, deactivated or suspended after a reporting date. In the past, institutions have had the experience that both EBA validation rules and ECB EGDQ checks could not be complied with in certain data constellations. Corresponding feedback to the supervisory authorities has led in part to the cancellation, deactivation, suspension or modification of the rules for future release cycles. For past reporting dates, however, the institutions are still required to resubmit in compliance with the applicable validation rules with the justification that these were still in effect on the corresponding reporting date. From the institution’s perspective, this sort of resubmission is de facto not viable and would lead to incorrect reports or validation errors elsewhere. Hence, a distinction would have to be made here as to whether an amendment to the supervisory requirements is the reason behind the cancellation of validation rules or the conclusion that the rules are not appropriate. Particularly in the latter case, the validation rule should not be applied retroactively for potential resubmissions, even though it was still effective at the reporting date.
To ensure an integrated approach to reporting, we would also like to suggest that the EBA coordinates with the ECB and the SRB before finalising these Guidelines, so that rules on resubmissions apply consistently to supervisory, statistical and resolution-related reporting in future.
An institution is free to act more conservatively than the minimum requirements stipulated by the supervisory regime. Such a procedure is not harmful from a supervisory perspective. There is therefore no error in the derived reported data. The institution fulfils the legal requirements and meets data quality standards. Consequently, historical reported data is only erroneous and must be resubmitted if – from a supervisory perspective – it has presented an overly optimistic picture of the risk situation, especially with regard to equity and liquidity.
The EBA should therefore clarify what it means by an “error” or an “inconsistency”. A data inconsistency identified after a report has been submitted whose resubmission would, for example, lead to an improvement in the supervisory ratios, should not be designated as an error, but rather as a positive change, and should be excluded from any resubmission requirement.
Furthermore, we believe that the existing tolerance thresholds based on the filing rules are insufficient. The draft should also take the balance between costs and benefits into account to an even greater extent (see our proposal to incorporate the thresholds per KRI in accordance with the ECB approach under 1).
In principle, we consider it necessary to define significant reporting points (KRIs) that are relevant for the resubmission. For example, data/amounts that are not relevant for management purposes should be excluded from retrospective submissions (e.g. memo items that have no impact on own funds requirements). This would also be consistent with the objective of preventing immaterial and unnecessary resubmissions.
Under the draft Guidelines, monthly resubmissions should be limited to seven resubmissions, unless the resubmissions as at the previous year-end are not included. As a result, up to 11 resubmissions of monthly reports could be necessary in these cases.
In our opinion, the cap referred to above (max. seven resubmissions) should also apply in these cases. These resubmissions provide a sufficient historical basis for analysing the data to be reported monthly.
As described under a), EBA risk indicators (see ECB presentation “Updates on the resubmission framework” of 26 April 2023) are a suitable option for enabling materiality without increasing complexity. Specific materiality thresholds could be based on the potential change in a metric. Under the ECB’s approach, institutions supervised by the ECB are required to provide an explanation of the extent of the resubmission if the defined thresholds are exceeded. It can be concluded that variations below or within the thresholds are not material and, can provide a proportional approach to avoid unnecessary and insignificant resubmissions entirely.
1. What are your general views on the proposed approach to the resubmission of historical data?
In general, we welcome the EBA’s clarifications, as they are likely to lead to greater legal certainty in dealing with resubmissions. As we understand it, the scope of application of the Guidelines is limited to the EBA’s reporting framework and does not extend to the resubmission of data in external reporting and in Pillar 3 reports under Part 8 of the CRR. We would ask you to clarify this accordingly in the final Guidelines.However, we are of the opinion that the target set by the EBA itself in Recommendation No. 25 in the Study of the cost of compliance with supervisory reporting requirements, i.e. that the Guidelines on resubmissions that are the subject of this consultation should cut the ongoing reporting costs for SNCIs by 1.0 to 1.5%, cannot be achieved with the measures being proposed. To reduce reporting costs, we consider it necessary, among other things, to introduce materiality thresholds as well as further measures (see option 3a that was rejected according to the accompanying information on the draft Guidelines, our concrete proposals on resubmission periods (see the following and under question 5a) as well as on a general principle under question 3).
In particular, we welcome the definition of a period limited to one calendar year for resubmissions. This should be defined as the maximum period for resubmissions. A supervisory authority should only have the power to extend this period in very limited and justified exceptional cases.
For further discussions, we believe that close integration with the ECB’s activities on the topic of resubmissions is necessary. It is unclear from the Guidelines to what extent the ECB’s considerations on “significant resubmissions” are also being picked up by the EBA and are compatible with the Guidelines. The ECB’s approach purposefully provides for the introduction of thresholds based on a list of KRIs, which in our view represent a proportionate basis for requiring the resubmission of historical data (see ECB presentation entitled “Updates on the resubmission framework” of 26 April 2023). We ask that these considerations be taken into account when considering the inclusion of materiality thresholds. The Final Report of the EBA’s Cost of Compliance Study also states the following: “‘Immaterial’ or ‘unnecessary’ resubmissions should be avoided both from the reporting entities’ and data recipients' point of view”. In our opinion, the draft does not fully meet this objective. Limiting resubmission to one year does not prevent the submission of “immaterial” or “unnecessary” data. We can understand the need for reliable and accurate data, but we also believe that a large number of the required resubmissions, many of which result from breaches of validation rules, represent changes compared with the original submission that are so minor that they do not compromise the accomplishment of supervisory responsibilities. For example, there are no additional insights in our view if the resubmitted data only has a minor impact on relevant ratios (e.g. <0.1 percentage point change in the Common Equity Tier 1 ratio).
In light of this and the ECB’s ongoing 1-year trial phase for the KRIs, we consider a proposed initial application of the Guidelines as of 31 December 2023 to be premature. We suggest waiting for the ECB trial phase to be completed before finalising the EBA Guidelines and ideally following up with a second trial phase involving less significant institutions in consultation with the ECB, so as to build on the findings from the first phase of the ECB’s work and to assess whether the thresholds identified by the ECB can provide a basis for exempting LSIs from the resubmission of data points. The ECB’s approach assumes that changes below the relevant thresholds do not require explanation and are therefore immaterial. In our view, this approach can be applied to LSIs under a favourable proportionality approach, which could ultimately lead to a complete exemption from immaterial resubmissions under certain conditions.
In our opinion, the focus of resubmissions should generally be on correcting material errors. The present draft should therefore be supplemented by a definition of which data is erroneous. There is currently no classification as to whether core regulatory data or memo items are erroneous. When it comes to a materiality threshold, a distinction should be considered as to whether material changes arise, for example in own funds, or whether memo items that ultimately do not affect a minimum requirement that is relevant for management purposes are erroneous.
Furthermore, we also welcome the EBA’s clarification that newly published Q&As should not lead to resubmissions for previous reporting dates. Furthermore, we take the view that clarifications should be included in the Guidelines on how to deal with EBA validation rules that are cancelled, deactivated or suspended after a reporting date. In the past, institutions have had the experience that both EBA validation rules and ECB EGDQ checks could not be complied with in certain data constellations. Corresponding feedback to the supervisory authorities has led in part to the cancellation, deactivation, suspension or modification of the rules for future release cycles. For past reporting dates, however, the institutions are still required to resubmit in compliance with the applicable validation rules with the justification that these were still in effect on the corresponding reporting date. From the institution’s perspective, this sort of resubmission is de facto not viable and would lead to incorrect reports or validation errors elsewhere. Hence, a distinction would have to be made here as to whether an amendment to the supervisory requirements is the reason behind the cancellation of validation rules or the conclusion that the rules are not appropriate. Particularly in the latter case, the validation rule should not be applied retroactively for potential resubmissions, even though it was still effective at the reporting date.
To ensure an integrated approach to reporting, we would also like to suggest that the EBA coordinates with the ECB and the SRB before finalising these Guidelines, so that rules on resubmissions apply consistently to supervisory, statistical and resolution-related reporting in future.
2. How do you see the proposed approach in relation to your existing resubmission policies set out in your institutions, agreed with internal audit and control functions?
Far-reaching control mechanisms have already been established at the institutions to ensure the highest possible data quality. In this context, processes have been introduced that mean overall that the data being delivered to the supervisory authority on an aggregated basis already has a very high data quality and, at the same time, reflects the institution’s risk profile. We welcome in principle the clarification the EBA has now made that retroactive resubmissions do not have to be made indefinitely in the case of identified errors. However, we are of the opinion that additional measures are necessary to further optimise the balance between costs and benefits (see our comments on Question 1).3. How do you see the proposed approach in relation to actual practices for the resubmission of data also considering the legal requirements set out in existing legislation (e.g. Article 3(5) of Commission Implementing Regulation (EU) 2021/451)?
Based on Article 3 of the CRR, we consider it necessary to add the following principle to the Guidelines:An institution is free to act more conservatively than the minimum requirements stipulated by the supervisory regime. Such a procedure is not harmful from a supervisory perspective. There is therefore no error in the derived reported data. The institution fulfils the legal requirements and meets data quality standards. Consequently, historical reported data is only erroneous and must be resubmitted if – from a supervisory perspective – it has presented an overly optimistic picture of the risk situation, especially with regard to equity and liquidity.
The EBA should therefore clarify what it means by an “error” or an “inconsistency”. A data inconsistency identified after a report has been submitted whose resubmission would, for example, lead to an improvement in the supervisory ratios, should not be designated as an error, but rather as a positive change, and should be excluded from any resubmission requirement.
Furthermore, we believe that the existing tolerance thresholds based on the filing rules are insufficient. The draft should also take the balance between costs and benefits into account to an even greater extent (see our proposal to incorporate the thresholds per KRI in accordance with the ECB approach under 1).
4. Would the proposed approach be feasible from the technology perspective considering the current reporting solutions?
We see challenges relating to a need for resubmissions to be made over a reporting date when there has been a change to the DPM. The IT service providers cannot maintain all the processing logic required for all DPMs and deploy it again when needed, so any necessary resubmission would have to be extremely manual. This is already the case under the current situation.a. If not, please provide concrete and realistic proposals for improving the proportionality element that can be efficiently implemented in the reporting systems without unreasonable costs or increasing the overall complexity.
Metrics inherently reflect the size of the institutions and form the basis for decisions or assessments by the data user. The materiality assessment with regard to resubmissions should be based on whether a resubmission can influence a decision or assessment by the data user. This is the concept the IASB uses, for example, in order to take into account the conflicting factors of costs and benefits. In this respect, an option that is easy to implement would be to define proportionality and materiality on the basis of metrics and their potential change through resubmissions. An example of a suitable set of indicators are the EBA Risk Indicators, which are used similarly by the ECB “significant resubmissions (pilot phase)” initiative.In principle, we consider it necessary to define significant reporting points (KRIs) that are relevant for the resubmission. For example, data/amounts that are not relevant for management purposes should be excluded from retrospective submissions (e.g. memo items that have no impact on own funds requirements). This would also be consistent with the objective of preventing immaterial and unnecessary resubmissions.
Under the draft Guidelines, monthly resubmissions should be limited to seven resubmissions, unless the resubmissions as at the previous year-end are not included. As a result, up to 11 resubmissions of monthly reports could be necessary in these cases.
In our opinion, the cap referred to above (max. seven resubmissions) should also apply in these cases. These resubmissions provide a sufficient historical basis for analysing the data to be reported monthly.
b. If such additional proportionality proposals are to be based on any threshold(s), please provide examples of such thresholds (relative and absolute) in relation to the size and complexity of your institution, and the reasoning behind that threshold.
We would welcome the materiality thresholds being expressed as a combination of percentage change and absolute values. Using relative/percentage thresholds has the advantage over absolute amount thresholds because it ensures that these thresholds simultaneously ensure proportionality with regard to the size of the institution as well as the reported information in question. The effort associated with a resubmission, which comprises internal and — in the case of the usual involvement of reporting software providers – external working hours, should also be taken into account.As described under a), EBA risk indicators (see ECB presentation “Updates on the resubmission framework” of 26 April 2023) are a suitable option for enabling materiality without increasing complexity. Specific materiality thresholds could be based on the potential change in a metric. Under the ECB’s approach, institutions supervised by the ECB are required to provide an explanation of the extent of the resubmission if the defined thresholds are exceeded. It can be concluded that variations below or within the thresholds are not material and, can provide a proportional approach to avoid unnecessary and insignificant resubmissions entirely.