This document outlines the proposed amendments to the implementing technical standards regarding the benchmarking of internal models for the 2026 exercise. The changes mainly focus on market and credit risks, with adjustments to asset class definitions and data collection instructions. The aim is to enhance the assessment of the quality of internal approaches used by competent authorities for calculating capital…
The document is a final report from the European Banking Authority (EBA) dated August 2025, entitled "Amending draft ITS on benchmarking", concerning the amendment of Implementing Regulation (EU) 2016/2070 related to benchmarking internal models within the 2026 benchmarking exercise. The scope covers market and credit risks, focusing on banks using the Internal Model Approach (IMA) for market risk and the alignment of asset classes for credit risk according to revised reporting standards. The document comprises 29 pages and presents technical amendments, cost-benefit analyses, as well as feedback from the public consultation (p. 1-29).
The report addresses the update of the Implementing Technical Standards (ITS) for benchmarking internal models used by European banks in calculating capital requirements, pursuant to Article 78 of Directive 2013/36/EU (CRD). The 2026 exercise proposes targeted modifications: for market risk, data collection remains limited to IMA banks, focusing solely on the Alternative Standardised Approach (ASA) components, due to the postponement of the FRTB implementation. For credit risk, asset class definitions are aligned with the revised IRB templates of prudential reporting. These adjustments aim to ensure continuity, comparability, and relevance of collected data while reducing banks' operational burden. The benchmarking has three main objectives: assist competent authorities in assessing internal models, explain and monitor variability of risk-weighted assets (RWA), and provide institutions with comparative information on their risk assessments. The amended ITS will be submitted to the European Commission for approval and will enter into force 20 days after publication in the Official Journal. The report also includes a qualitative cost-benefit analysis concluding that the benefits of the changes outweigh the costs, notably for market risk. Finally, the public consultation revealed concerns about post-FRTB participation and suggestions to improve data quality, to which the EBA responded by adapting its proposals (p. 3-29).
This document was developed to update the Implementing Technical Standards governing the collection and processing of benchmarking data for banks' internal models, in response to regulatory developments and the postponement of the FRTB implementation. The objective is to maintain the quality, comparability, and relevance of data collected for annual exercises, while considering operational constraints of banks and competent authorities. The scope covers market and credit risks, with particular attention to consistency between definitions used in benchmarking and those in prudential reporting. The document also addresses limitations related to the FRTB postponement, which impacts participation of banks using the Alternative Standardised Approach (ASA), and measures to limit reporting burden. The challenges are to ensure a stable and effective framework for internal model assessment, pursuant to Article 78 CRD, while preparing for future regulatory developments (p. 4-7, 12-17).
Market risk:
- The FRTB postponement led to maintaining data collection limited to IMA banks, excluding for now the extension to ASA banks, to preserve data quality and comparability (p. 4-6).
- IMA banks must now report only ASA data, reducing burden while allowing monitoring of progress (p. 5-6).
- Technical amendments to annexes 5, 6, 7, and 10 notably introduce a new template 106.02 for SBM validation data collection, clarify instrument identifiers (LEI, ISIN), and adjust ASA validation portfolios (p. 5-7).
- The EBA considers making SBM validation data submission optional for banks demonstrating sufficient accuracy, to lighten the burden (p. 6).
Credit risk:
- Asset class definitions of benchmarking portfolios are aligned with those of the revised IRB templates of prudential reporting, in line with CRR3/CRD6 developments (p. 7).
- A precise mapping is proposed to avoid double counting, notably for Retail – SME and Retail – non-SME portfolios (p. 7).
- Four options were analyzed for this adaptation, with the EBA favoring a mapping solution between old and new definitions to minimize costs while ensuring alignment (p. 15-17).
Public consultation and feedback:
- Two responses were received, expressing concerns about post-FRTB participation, operational complexity, and data quality (p. 19-21).
- Suggestions to introduce operational metrics or clarifications on certain instruments were discussed, but the EBA maintained focus on comparability and standardization (p. 23-28).
- Clarification was provided on the use of SOFR rates (overnight daily compounded) for related instruments (p. 28).
Cost-benefit analysis:
- For market risk, template revision (option 2) is preferred despite initial cost, as it improves relevance and clarity of collected data (p. 13-15).
- For credit risk, the mapping option (option 3) is retained to ensure alignment with prudential reporting at lower cost (p. 15-17).
Findings:
- The FRTB postponement has a direct impact on the scope and nature of data collected in benchmarking, leading to limiting collection to IMA banks and focusing on ASA data (p. 4-6).
- Proposed changes in technical annexes aim to improve clarity, consistency, and accuracy of data, notably through new templates and standardized identifiers (p. 5-7).
- Aligning asset classes for credit risk with revised IRB templates is necessary to avoid inconsistencies in data (p. 7, 15-17).
Assumptions:
- ASA banks’ participation in benchmarking is postponed until the effective entry into force of FRTB requirements, an assumption confirmed by the European Commission (p. 4, 19).
- IMA banks will continue to provide ASA data, allowing progressive monitoring of developments (p. 6).
Interpretations:
- The decision to limit collection to IMA banks is justified by the need to preserve data quality and avoid operational overload, consistent with the spirit of legislative texts (p. 4-5).
- The EBA favors a pragmatic and proportionate approach, adapting requirements to market realities and institutions’ capacities (p. 5, 13-17).
Uncertainties:
- The precise impact of changes on capital requirements remains difficult to quantify, notably due to interactions with national decisions and future regulatory developments (p. 14).
- Future participation of ASA banks in benchmarking will depend on the effective implementation of FRTB and institutions’ ability to adapt (p. 19).
The EBA concludes that the proposed amendments to the ITS are necessary to ensure the relevance, consistency, and quality of benchmarking data for internal models in the context of the FRTB postponement. The temporary limitation of data collection to IMA banks, focusing on ASA data, is justified to avoid unnecessary burden and preserve comparability. Aligning asset class definitions for credit risk with revised IRB templates is essential to ensure consistency with prudential reporting. The EBA recommends adopting the proposed changes, which underwent public consultation and a qualitative cost-benefit analysis demonstrating that benefits outweigh costs. The amended ITS will be submitted to the European Commission for approval and will enter into force 20 days after publication in the Official Journal. Furthermore, the EBA considers eventually reducing the reporting burden related to SBM validation data for institutions demonstrating sufficient control. Finally, the EBA remains attentive to stakeholder feedback to adjust future developments, notably regarding FRTB implementation (p. 3-7, 12-29).
Synthesis note written from the full document by DataSAI Academy. This note comes from the scientific library of the DataSAI Academy, open to all.