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?
The instructions and templates are generally clear at a high level and provide a solid basis for understanding the revised benchmarking framework. However, several areas would benefit from further clarification to support consistent and accurate implementation across institutions.
In particular, we note the need for additional guidance on the treatment of portfolios where multiple PD, LGD and CCF models apply, as well as on certain collateral mappings (e.g. in template C 102.01). Clarification would also be welcome regarding the alignment between benchmarking templates and COREP references, including practical guidance on reconciliation with COREP data (e.g. C 08 and C 09), in order to ensure consistency across reporting frameworks.
Furthermore, greater transparency would be beneficial where existing fields have undergone material changes in their definitions. Clearly highlighting such changes would help avoid misinterpretation and facilitate implementation.
Additional clarification is also needed for certain data fields and metrics, including the definition of RWA excluding MoC and supervisory measures, as well as parameter- and portfolio-level adjustments. More detailed instructions would be particularly helpful for newly introduced templates, such as C 104, given their complexity and the challenges associated with sourcing and deriving the required data.
Finally, we note that the calculation of certain metrics, such as default and loss rates over a five-year horizon, may be difficult where portfolios have been newly defined. Reconstructing historical data under revised definitions could entail significant operational effort, and further guidance or proportionality considerations in this regard would be welcome.
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, ESBG members have not identified material discrepancies between the templates and instructions and the requirements set out in the underlying regulatory framework.
However, this assessment remains subject to further validation during the implementation phase. In particular, some elements of the benchmarking instructions, including specific segmentations and breakdowns, do not fully mirror the corresponding COREP structures. While this does not necessarily constitute a discrepancy with the underlying regulation, it may create challenges in ensuring full consistency and reconciliation between frameworks.
Further clarification and practical guidance in this area would therefore be beneficial to support consistent implementation and to confirm alignment with the regulatory requirements over time.
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 generally agree that the amended ITS on Supervisory Reporting allows supervisors to fulfil the requirements and objectives of Article 78 CRD, in particular by supporting the assessment of the consistency and comparability of internal models across institutions.
At the same time, several implementation aspects may affect the effectiveness and consistency of the framework in practice. In particular, the increased granularity and complexity of certain templates and instructions could lead to higher implementation effort and potential divergences in interpretation. This is especially relevant in areas such as collateral treatment, as well as in the absence of explicit reconciliation guidance between COREP and benchmarking templates.
In addition, data availability constraints may pose challenges for some of the newly introduced requirements. For instance, templates requiring long historical time series and detailed tracking of recovery processes (such as C104) may be difficult to populate consistently, particularly where portfolio definitions have evolved over time.
Finally, while the framework is broadly aligned with the objectives of Article 78 CRD, we note that the proposed changes appear to increase rather than reduce the overall complexity of the benchmarking exercise. This may reduce the extent to which the intended simplification objectives are achieved.
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.
The amended ITS is expected to increase both implementation and ongoing operational costs compared with the current benchmarking framework. The main cost drivers are the enhanced granularity of the templates, the introduction of new data requirements that are not currently available in existing reporting systems, and the need to establish new calculation, control and reconciliation processes.
A first major area of cost relates to template C102.01, in particular where the revised structure requires institutions to report information across a broader perimeter than under the current framework. The consolidation of high-default and low-default portfolios, as well as the former C102 and C103 templates, means that certain information that was previously reported only for a narrower scope must now be produced across the full benchmarking perimeter. This entails additional review work, increased validation effort and longer execution times. In particular, columns 0210, 0211 and 0212 are expected to require the creation of new processes from scratch. More generally, the multi-dimensional portfolio segmentation in C102.01 and the requirement to provide decomposed model outputs such as PD, LGD and RWA imply a substantial redesign of existing data structures and reporting logic.
Further significant costs arise from data fields whose definitions diverge from existing reporting frameworks. For example, the reporting of collateral value in C101.01 (column 0090) and C102.01 (column 0140) as the sum of uncapped market values differs from both previous benchmarking instructions and the values currently reported in COREP. As this information is not readily available in existing systems, institutions would need to develop dedicated IT processes solely for the purposes of the benchmarking exercise. This would require considerable time and resources and would create additional operational burden without generating broader simplification benefits in other regulatory reporting streams.
Similarly, the reporting of maturity in C102.01 (column 0170) creates additional burden where the benchmarking instructions diverge from existing COREP logic. Excluding maturity for exposures where it is not a component of the RWA calculation, while COREP continues to report maturity for such exposures, would require specific adjustments to current processes and controls in order to maintain consistency between reporting streams.
Additional burden also arises from the requirement to report RWA excluding MoC and add-ons, especially in columns 0203 and 0204, as these values are not typically available in current systems and would therefore require additional calculations and dedicated implementation work.
A second major cost driver is template C104, which is expected to generate significant implementation effort as well as recurring operational burden. The information required in this template is generally not part of existing regulatory reporting and would therefore need to be generated on an ad hoc basis or through newly established processes. This requires the involvement of additional teams beyond those usually responsible for benchmarking submissions, including data, modelling, workout or recovery-related functions. The template also relies on long historical time series and detailed monitoring of recovery processes, which may be difficult to source and reconstruct consistently. Within this template, column 0044 appears particularly burdensome, as it is not commonly used for reporting purposes and would require dedicated process development.
Overall, we expect the proposed changes to lead to higher one-off implementation costs, including IT development, redesign of data architecture, adjustments to controls and reconciliation frameworks, and mobilisation of additional teams, as well as higher recurring running costs, due to more complex data collection, more extensive validation and longer reporting cycles. Compared with the current framework, the amended ITS is therefore expected to increase the overall cost of the benchmarking exercise. At the same time, the precise magnitude of these costs will depend on each institution’s existing IT landscape, degree of automation and data availability.
Finally, the implementation timeline may further increase costs, as the expected publication of the final ITS late in the year would leave institutions with limited time to develop, test and operationalise the required changes.
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. In particular, it avoids unnecessary duplication of reporting and reduces the need for complex reconciliation processes between entities within the same group.
By contrast, requiring reporting at individual entity level would increase complexity and reporting costs without providing commensurate benefits in terms of comparability or supervisory insight. It could also introduce inconsistencies across entities and reduce the overall coherence of the data.
Finally, we note that the existing structure of the templates, including portfolio segmentation and geographic breakdowns where relevant, already provides an appropriate level of granularity for supervisory purposes. Requiring additional reporting at entity level would therefore not be necessary.
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?
ESBG considers that supplementing the sample of counterparties in template C 100.00 with Large Exposures counterparties reported in template C 27.00 is feasible in principle and could contribute to expanding the sample of low-default counterparties.
However, this approach would entail additional operational effort and costs, as it would require the integration of data from different reporting streams and the establishment of new processes to ensure consistency across templates.
In particular, ESBG notes that differences in scope and data definitions may create challenges in ensuring comparability across institutions. Leaving the definition of the supplemented sample at the discretion of each institution could result in divergent perimeters, thereby reducing the overall consistency of the benchmarking exercise.
Further challenges arise from data integration aspects. For example, counterparty identifiers may differ across templates, with internal codes used in benchmarking templates and external identifiers such as Legal Entity Identifier (LEIs) used in template C 27.00. This could complicate reconciliation and require additional mapping and validation processes.
While the proposed approach is feasible, ESBG considers that clear and detailed guidance would be necessary to ensure consistent implementation across institutions and to preserve the comparability of results. Without such guidance, the additional complexity and costs could outweigh the benefits of expanding the sample.
CR 2: Do you identify any errors in the list of counterparties in C 100.00? If so, please report it to us.
ESBG has identified a specific inconsistency in the list of counterparties included in template C 100.00. In particular, Royal Caribbean Cruises Ltd. (counterparty code LC_0001575) is reported under the geographical area US, whereas its legal jurisdiction of incorporation is Liberia. As regulatory reporting is typically based on the country of incorporation, this classification appears inconsistent and should be clarified or corrected in the ITS amendment.
More generally, ESBG notes that the removal of certain identifiers, such as the Credit Register Code, Commercial Register Code and ISIN, may create challenges in mapping counterparties across internal systems. This could increase the risk of incomplete coverage and make it more difficult for institutions to verify the accuracy and completeness of the counterparty list.
ESBG therefore considers that maintaining a sufficient set of identifiers would be important to support robust mapping, validation and consistency of the reported data.
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?
ESBG understands and supports the integration of the portfolio/model split into templates C101.01 and C102.01, as well as the removal of template C105.02. This approach has the potential to streamline the reporting architecture, reduce duplication across templates and enhance the overall consistency and analytical value of the benchmarking exercise.
At the same time, ESBG considers that the proposed implementation introduces additional complexity compared to existing processes. In particular, the integration of the portfolio/model split into C102.01 would require a significant redesign of current reporting processes, including the development of new controls, validation layers and reconciliation mechanisms. This is expected to entail considerable implementation effort in terms of time and resources.
ESBG also notes that important aspects of the implementation remain unclear, particularly for portfolios where multiple PD, LGD and CCF models apply. Further clarification would be needed on how such situations should be reported, including whether all combinations of model identifiers are expected to be reported at portfolio level. Without clear guidance, this could lead to divergent approaches across institutions and increase the reporting burden.
Although there are potential benefits to the revised structure from an architectural standpoint, further clarification would be necessary to ensure consistent and proportionate implementation across institutions.
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?
In principle, it is possible to provide the details of the impacts of RWEA add-ons and margins of conservatism (MoC) at the required level of segmentation.
However, the definition and calculation of these elements at the requested level of granularity are not sufficiently clear. In particular, isolating MoC at the parameter level may pose practical challenges and would require further clarification to ensure a consistent interpretation across institutions.
In addition, meeting these requirements may require additional calculations or simulations, especially where the relevant figures are not readily available in existing reporting frameworks. This would increase implementation effort and may introduce additional complexity in the reporting process.
Further clarification on definitions and calculation methodologies would be needed to ensure consistent and adequate implementation.
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?
ESBG acknowledges the objective of the proposed simplification for column 0202 (RWEA Standardised) and considers that the level of aggregation appears sufficient to highlight the key differences between the IRB and Standardised approaches.
Nevertheless, ESBG observes that the implementation of this change may introduce additional operational complexity. In particular, the need to apply filters to ensure that the information is reported only for relevant portfolios would require adjustments to existing data processing logic. This would also necessitate the introduction of additional validation checks and controls to avoid misreporting.
As a result, the proposed change may require modifications to existing processes and the integration of new controls into current workflows, leading to increased time and resource requirements.
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?
The revised split by collateral type introduces a more structured and standardised classification compared to the current Benchmarking framework.
However, we observe that certain elements of the new definition and instructions are not sufficiently clear and may create interpretation challenges when mapping existing reporting to the new structure.
In particular, a potential inconsistency in the instructions has been identified, whereby the typology “Credit derivatives” does not appear to be fully aligned with the relevant columns. In addition, further clarification would be warranted on several aspects of the proposed classification. These include the treatment of portfolios currently reported under “Other funded credit protection”, which no longer appears explicitly in the new framework, the reporting of ADC exposures within mortgage-backed portfolios, and the interpretation of collateral types 50 and 60. Indeed, it is not clear whether certain categories represent aggregate values (i.e., total sum of collateral types) or residual components.
These elements may lead to different interpretations across institutions and complicate the consistent mapping of existing data to the revised classification.
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)?
The information required for column 0302 can, in principle, be derived from existing LGD data, as long-run average realised loss rates are generally available within institutions.
It should be noted, however, that these data are not typically embedded in automated reporting processes and would therefore need to be extracted and processed on an ad hoc basis. This would require manual intervention, the involvement of specialised teams and additional validation steps, leading to increased operational effort and longer processing times.
In terms of its implementation, we observe that the requested level of granularity, in particular at Portfolio ID level, may present significant challenges. In many cases, long-run average realised loss rates are currently calculated at a more aggregated level, such as country, rating grade, LGD model ID or LGD pool. Moving to the requested level of detail would require a substantial degree of data disaggregation, which may not be supported by existing data structures.
Reconstructing historical loss data at this level of granularity may also be difficult where the necessary information has not been consistently captured over time, which could limit the ability of institutions to provide robust and comparable figures.
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?
ESBG considers that the information required for template C104.00 can, in principle, be derived from existing LGD data. However, these data are not typically embedded in automated reporting processes and would need to be produced through ad hoc extraction and processing. This entails manual intervention, additional validation steps and the involvement of specialised teams, resulting in higher operational effort, longer processing times and increased operational risk.
Regarding implementation, ESBG identifies important challenges related to the scope and structure of the template. The requirement to provide long historical time series, combined with harmonised definitions of default and recovery, implies significant data integration and tracking efforts. In many cases, the necessary data may not be consistently available over the full-time horizon.
ESBG further highlights that the requested breakdown of realised losses, including cured exposures, write-offs, disposals and repossessions, may be difficult to deliver at the level of detail required. Current calculation frameworks do not generally produce these components at exposure class level, and aligning existing data structures with the template requirements would require substantial adjustments.
Long time horizons between default and realised losses may also affect the comparability of results across institutions. A shorter observation period could provide a more proportionate and consistent basis for analysis.
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?
We agree with the proposed split by facility type in template C 102.01 and considers it suitable for capturing the functioning of CCF models.
The alignment with Annex I of Regulation (EU) No 575/2013 ensures consistency with the COREP framework and supports a coherent treatment of off-balance-sheet exposures across reporting templates. This contributes to greater consistency in interpretation and strengthens comparability across institutions.
No material implementation challenges have been identified in relation to this classification, and limited added value would be expected from introducing alternative approaches, such as product-based classifications.
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?
ESBG expresses a preference for the single-column approach for reporting CCF information in templates C 101.01 and C 102.01. This option aligns with existing practices and avoids the need for additional derivations, in particular with regard to Standardised Approach CCF values. It also simplifies data processing and reduces unnecessary complexity in the reporting framework.
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.
ESBG supports the use of a common master scale, as it contributes to greater comparability and consistency across institutions. That said, the proposed level of granularity raises several concerns. In particular, the breakdown of default status into 90 days past due and unlikely-to-pay introduces additional effort and may create a misalignment with COREP, where such distinctions are not consistently reflected. This complicates the reconciliation between benchmarking and regulatory reporting frameworks.
It is also noted that, depending on the structure of existing internal master scales, the proposed approach may increase rather than reduce the number of portfolios for some institutions.
IFRS9 1: Do you agree with the proposed definition of SA institutions as relevant for the benchmarking reporting?
ESBG agrees with the proposed definition of SA institutions, based on the classification of large institutions pursuant to Article 4(1)(146) of Regulation (EU) No 575/2013 (CRR), provided that these institutions are subject to IFRS 9 requirements.
It is expected that the EBA will continue its established practice of publishing, on an annual basis, the list of institutions falling within the scope of the benchmarking exercise, including for the 2027 exercise with reference date 31 December 2026.
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?
ESBG would welcome clarification on the use of SA exposure classes across templates C 101.01 and C 102.02.
In template C 102.02, only exposure classes starting with “Z**” appear to be in scope for both relevant SA institutions and IRB institutions for exposures under the Standardised Approach, namely Z21, Z30, Z31, Z40 and Z60. However, according to the definitions provided for template C 101.01, these codes are associated with SA institutions only. It would therefore be helpful to confirm whether exposures under the Standardised Approach in IRB institutions are intended to be reported in template C 101.01, or only in template C 102.02.
Clarification would also be useful regarding the definition of the Portfolio ID in template C 102.02. The instructions indicate that, for both SA institutions and IRB institutions under the Standardised Approach, the segmentation is based on exposure classes listed in template C 101.01, in combination with geographical area and rating information. Particularly, the reference to geographical area in template C 102.02 appears unclear and may require confirmation.
A similar question arises in relation to the definitions applicable to IRB institutions under the IRB approach, where the same reference is made.