Response to consultation on draft ITS amending Commission Implementing Regulation (EU) 2024/3117 regarding Credit Risk and IFRS 9 Benchmarking reporting
General 1: Are the instructions and templates clear to the respondents? If not, what should be clarified?
We welcome the overall objective of simplifying and aligning the benchmarking framework with COREP definitions, which is a positive step towards enhancing consistency across reporting exercises. However, despite this improvement, several aspects of the proposed templates and instructions would benefit from further clarification to ensure a consistent and operationally feasible implementation across institutions.
First, important conceptual ambiguities remain regarding key notions embedded in the templates. This is notably the case for the identification and treatment of counterparty risk within the portfolio clustering approach, as well as for the distinction between mandatory and optional fields. In addition, certain metrics and definitions lack sufficient precision at this stage, such as the notion of “Hyp LGD senior unsecured without negative pledge”, the treatment of RWA add-ons, or the calculation of the long-run average realized loss rate. Without additional guidance, institutions may interpret these concepts differently, which could ultimately impair the comparability of benchmarking results.
Second, the operational implementation raises material concerns. The introduction of a new Portfolio ID structure significantly increases the complexity of the clustering framework, with a combination of multiple keys, including several based on newly required data. In practice, this will reduce the readability and usability of the portfolios, as the identifiers will consist of technical codes that require systematic reference to documentation. This represents a step back compared to the current framework, where portfolio identification remains more intuitive. Similarly, further clarification is needed regarding the definition of facility types and the precise calculation perimeter of certain metrics, such as “RWA before SME supporting factor”.
Finally, key components of the implementation package have not yet been specified, including the timing of the disclosure taxonomy, the associated reporting formats, and the validation and quality control rules. These elements are critical for institutions to assess the full implementation effort, particularly given the impact on disclosure tools and reporting chains.
In this context, we also seek confirmation that securitization exposures remain excluded from the benchmarking exercise, in line with the current scope.
Overall, while the structure of the templates is broadly understandable, further clarification is necessary to ensure consistent interpretation, comparability of results, and efficient implementation across institutions.
The instructions require several important clarifications.
On the construction of Portfolio IDs: The new dynamic construction logic ExposureClass_Geography_Rating_Sector_LTV_ExposureType_CollatType_FacilityType raises significant concerns regarding comparability. In the absence of an exhaustive standardized list provided by the EBA, each institution may understand the segmentation rules differently, generating divergences that could undermine cross-institutional comparability. We therefore request that the EBA provide an exhaustive and standardized list of valid Portfolio IDs, as was the case under the previous framework.
On the mapping between old and new versions: The instructions do not adequately document the field-by-field evolution between the former templates (C102.00, C103.00, C105.02) and the new C102.01. We request a detailed matrix specifying for each field:
- Fields retained without modification (unchanged definition, format and scope)
- Fields retained with modifications (precise description of changes)
- New fields (construction logic and practical examples)
- Deleted fields (clear identification to avoid mapping errors)
This systematic approach will prevent divergences in interpretation across institutions and will contribute to the simplification objective by establishing a shared implementation reference.
General 2: Do the respondents identify any discrepancies between these templates and instructions and the calculation of the requirements set out in the underlying regulation?
At this stage, we do not identify material discrepancies between the proposed templates and instructions and the calculation of the requirements set out in the underlying regulation.
General 3: Do the respondents agree that the amended ITS on Supervisory Reporting allows to fulfil the requirements and purpose of the Article 78 of the CRD?
We do not express specific concerns regarding the ability of the amended ITS to fulfil the objectives of Article 78 of the CRD.
General 4: what is the information (please refer to single columns) that you see as particularly costly to be provided? Please provide detailed information about the implementation and running costs of the proposed changes, also compared to the current total costs.
While the proposed framework is presented as a simplification, it introduces a number of structural changes that are expected to generate significant implementation and operational costs for institutions.
In particular, the introduction of the new loss rate template C104.00 represents a major effort. This template requires the reconstruction of long historical time series and their alignment with current portfolio definitions and modelling frameworks, which were not designed for such retrospective analysis. Similarly, the new segmentation framework, based on Portfolio ID and Model ID dimensions, significantly increases the level of granularity required, leading to higher data volumes and more complex data models.
The merger of the former templates C102 and C103, although conceptually coherent, will also result in a substantial increase in volumetry and processing time. In addition, the introduction of new metrics in template C102.01, such as RWA without margin of conservatism (MoC), RWA without supervisory measures, RWA add-ons, and long-run realized loss rates, is particularly challenging, as these metrics are not currently embedded in standard RWA production systems. Producing them would require either deep transformations of existing IT systems or the development of parallel calculation frameworks.
A key structural difficulty lies in the requirement to isolate the impact of MoC, supervisory measures and add-ons. In most institutions, these elements are integrated within model parameters or applied through overlays that are not explicitly tracked at the level of granularity required by the templates. Extracting this information would therefore imply significant changes to data architecture, parameter storage and calculation engines.
In addition, the introduction of Expected Loss add-ons in template C105.1 will require further modelling developments and data sourcing efforts.
In light of these elements, the proposed timeline for a first implementation by Q4 2027 does not appear realistic. The scale of the required IT developments, methodological adjustments and operational changes suggests that a longer implementation period would be necessary to ensure robust and consistent reporting.
The following items are expected to generate the highest implementation and operational costs:
- Template C104 (15-year realized losses): Complex historical reconstruction using the new Basel 4 exposure classes for data predating 2025. The choice of reference year is critical to estimating the full deployment cost of this requirement
- Extension of 5Y default and loss rates to all portfolios: Similarly requires complex historical reconstruction under the new CRR3 exposure class definitions for data predating 2025. The reference year will be a determining factor in assessing the implementation effort
- Template merger and implementation of new controls with regulatory reportings
- Columns 0203, 0204, 0205 (RWA excluding MoC, supervisory measures, add-ons): these calculations require significant system development and implementation of new controls
Given the scale and cost of the technical developments required, we reiterate that producing data under the new format for the Q4 2026 submission does not appear feasible, particularly as FINREP/COREP changes will not become mandatory until September 2027.
General 5: what is the reporting level for banking groups that you deem more appropriate?
We consider that reporting at the highest level of consolidation is the most appropriate approach for the benchmarking exercise. This is consistent with the regulatory capital framework and provides a coherent and comprehensive view of risk and model usage across banking groups.
Limiting reporting to the consolidated level would significantly reduce operational burden and complexity, while remaining fully aligned with the objectives of the benchmarking exercise.
CR 1: Do you foresee any issues with supplementing the sample of counterparties listed in C 100.00 by the Large Exposures counterparties reported in the Template C 27.00 at the same reference date in order to expand the sample of low default counterparties for the reporting for template C 101.01/.02?
NA
CR 2: Do you identify any errors in the list of counterparties in C 100.00? If so, please report it to us.
NA
CR 3: Do you foresee any issues with the transfer of the portfolio/model split from the former template C 105.02 to the template C 101.01 and the new template C 102.01? Do you agree with the EBA’s assessment that the portfolio/model split in the new template C 102.01 is the preferred option compared to designing a separate template increasing the reporting details of the former templates C 105.01 and C 105.02?
We understand and broadly support the rationale for integrating the portfolio/model split directly into templates C101.01 and C102.01, as this contributes to a more streamlined and coherent reporting framework. However, this structural change also introduces significant operational challenges.
The merger of former templates and the integration of model-level information lead to a substantial increase in data volumes and processing complexity. In particular, the requirement to report data at the combined level of PD, LGD and CCF models represents a higher level of granularity than in the current framework, which did not systematically require such combinations.
Moreover, the introduction of the Model ID dimension raises questions regarding the comparability of reported data across institutions. Given the diversity of modelling practices, segmentation choices and model architectures, it is not clear how consistent benchmarking outcomes can be ensured when data is reported at such a granular level. Without additional guidance on how comparability will be achieved and interpreted, there is a risk that increased granularity may paradoxically reduce the usefulness of the benchmarking exercise.
CR 4: Do you think it is possible to provide the details of the impacts of the RWEA add-ons and MoCs (columns 0203, 0204, 0205 of Template C 102.01) at the required segmentation level?
Providing the requested breakdown of RWA into components excluding MoC, supervisory measures and add-ons at the required level of granularity is expected to be particularly challenging.
This information is generally not available in current RWA production systems, where MoC and other adjustments are often embedded directly within model parameters or applied through aggregated overlays. Producing such a breakdown would therefore require substantial changes to systems, including the redesign of data models, the storage of additional parameter versions, the adaptation of calculation engines, the reporting processes, the controls and the reconciliation framework.
These challenges are further amplified in complex environments, such as retail portfolios or in cases where multiple RWA production systems coexist, notably between central and local models. In such cases, ensuring consistency across different calculation chains would add an additional layer of complexity.
While technically feasible, the implementation of this requirement would involve a disproportionate level of effort and cost, particularly within the proposed timeline.
CR 5: Do you agree with the proposed simplification in the reporting for column 0202 (RWEA Standardised) of the template C 102.01? Do you deem the segmentation level for which the information is requested sufficient to highlight the key differences between the IRB approach and the Standardised Approach?
We do not observe a material simplification in the proposed approach for reporting RWA under the Standardized Approach.
Although the reporting is limited to a subset of rows, institutions would still need to perform the calculations across the full relevant exposure perimeter. Furthermore, the proposed segmentation at model level appears conceptually misaligned with the nature of the Standardized Approach, which is not based on PD, LGD and CCF combinations in the same way as IRB models.
From an analytical perspective, the comparison between IRB and Standardized approaches is more meaningful at portfolio level rather than at the level of model combinations. In this context, the added granularity may not improve the understanding of differences between approaches, while increasing implementation complexity.
Additional clarification would also be needed on whether and how allocation methodologies, such as pro-rata approaches, are expected to be used.
CR 6: The split by collateral type (column 0061 of new template C 102.01) differs from the split used in the current Benchmarking portfolios. Do you foresee any issues or disadvantages with the new definition and instructions as compared to the ones currently in use?
In principle, we consider the proposed approach to be appropriate. However, we would appreciate further guidance for the type of exposures that should be reported under category 60 “Exposures with partial collateralisation” to ensure a consistent interpretation and application across institutions.
CR 7: Do you see material obstacles in providing the information required by the column 0302 of the template C 102.01 (Long-run average realized loss rate of assigned facility grades)?
At this stage, additional guidance is required regarding the definition and calculation methodology of the long-run average realized loss rate. Without further clarification, it is difficult for institutions to assess the feasibility of this requirement and to determine the necessary implementation steps.
CR 8: Do you see material obstacles in providing the information required by the new loss rate template (C 104.00)? Would alternative measures of loss rates be more appropriate?
The introduction of template C104.00 raises significant methodological and operational questions. While the underlying data may exist within institutions, it is not necessarily structured in a way that allows for an easy extraction and aggregation in line with the proposed requirements.
In particular, the need to combine information from different sources, including local and central systems, may add complexity to the reconciliation and consolidation process. Furthermore, aligning historical data with current portfolio definitions and modelling frameworks may require extensive reconstruction efforts.
At this stage, institutions are not in a position to confirm that all required data points can be produced with sufficient reliability and consistency.
We identify several material obstacles to the implementation of template C104.00:
- Conceptual obstacle: the inclusion of "cured" facilities in a template described as capturing realized losses creates a conceptual inconsistency. We recommend separating the four recovery outcomes into two distinct groups: successful recoveries (cured) and effective losses (repossession, write-off, disposal).
- Technical obstacle: Reconstructing 15 years of historical data under the new CRR3 exposure class definitions is particularly complex for data predating 2025.
We recommend that the historical series begins from 2025, with progressive extension over subsequent years as data under the new framework accumulates.
CR 9.1: The split by facility type (column 0062 of template C 102.01) is designed to collect information on the functioning of the CCF models. The classification into buckets (facility types) of off-balance-sheet items is based on Annex I of Regulation (EU) No 575/2013. Do you think there exist other classifications (for example product types) of off-balance-sheet items that would be more useful for representing the CCF models?
NA
CR 9.2: Template C 101.01 and C 102.01 propose two alternatives to collect the information on the CCF, either by one column (column 0110 in C 101.01 and column 0120 in C 102.01) or by two separate columns (columns 0111/0112 in C 101.01 and columns 0121/0122 in C 102.01). Which of these reporting alternatives do you prefer?
NA
CR 10: Do you support the use and the proposed granularity of the common master scale to limit the number of reported rating grades? What alternative would you propose.
We support the principle of the common master scale R01-R98, which improves comparability across institutions.
However, we request clarification on its consistency with the master scale used in other COREP reporting frameworks, to ensure coherent application across the full regulatory reporting chain.
IFRS9 1: Do you agree with the proposed definition of SA institutions as relevant for the benchmarking reporting?
We agree with the proposed definition, but clarification would be welcome on the treatment of subsidiaries within banking groups.
Where a parent institution is included in the scope, it would be useful to confirm whether subsidiaries applying the Standardised Approach, which are not explicitly listed, should be considered within scope at the consolidated level of the parent, or instead treated as out of scope.
IFRS9 2: Do you agree with the proposal to refer to SA exposure classes in defining the portfolios for SA institutions?
From a simplification and implementation perspective, we support the option based on a single CCF column, as it would reduce the number of data points to be reported while remaining aligned with the overall objective of streamlining the framework.