Response to consultation on amending ITS on the benchmarking of approaches for calculating own funds requirements – 2027 market risk benchmarking exercise

Q1: Do you see any issues or any missing information that should be required in the new templates suggested for the FRTB AIMA benchmarking exercise (i.e. Annex V & VI)?

The industry identifies a need for clearer guidance within the instructions for the new Template C 130.02 (MKR IMA SP) under Annex V.

Specifically, the industry interprets the requirements for populating Template C 130.02 (MKR IMA SP) as referring to the stress periods calibrated on the institution's actual, live trading portfolio, rather than a separate calibration performed solely for the prescribed hypothetical benchmarking portfolios.

To ensure consistent reporting across all participating firms, we would welcome explicit confirmation of this approach in the final instructions. This view is based on three key considerations:

• Preserving model integrity: The benchmarking exercise is designed to evaluate the performance of the actual, approved internal models used for regulatory capital. Calibrating dedicated stress windows for a small set of synthetic portfolios would run counter to this objective, potentially yielding results that do not represent the true behavior of the bank's production model.
• Methodological consistency: A 12-month stress period is inherently an enterprise-wide model parameter. Calibrating it against a limited set of hypothetical trades is methodologically inconsistent with how internal models are designed and validated.
• Operational feasibility: Requiring institutions to execute parallel, dedicated calibration cycles for temporary, benchmarking-specific portfolios would introduce a significant and disproportionate operational and computational burden.

Consequently, the industry would welcome the instructions to specify that the reported stress periods should reflect the live production parameters.

Q2: Do you see any issues with specifying the timeline in Annex II (i.e. 4-month postponement compared to 2026 timeline)?

Any issues with specifying the timeline in Annex II (i.e. 4-month postponement compared to 2026 timeline)

The industry welcomes the EBA's proposal to postpone the 2027 market risk benchmarking exercise compared to the 2026 timeline. This additional preparation time is particularly beneficial for institutions participating in the exercise for the first time and will support more effective planning and overall readiness.

However, the industry strongly recommends extending the submission window for the IMV module to three weeks (extension by one week) and preferably to four weeks (extension by two weeks) due to the following considerations:

1. To effectively manage competing priorities such as quarter-end and half-year-end reporting activities. The current proposed IMV (IMV, initial SbM sensi and SA validation portfolio) module remittance period (and RM remittance period) in the ITS coincides with quarter-end and half-year-end reporting requirements, which are already among the most resource-intensive periods for regulatory reporting teams.
2. Detailed review of IMV is critical to identify and rectify any trade booking errors. Accurate trade booking is essential for producing reliable benchmarking results.
3. Benchmarking submissions are subject to internal governance and approval processes, including management review and sign-offs. In addition, firms must complete validation checks against the latest Data Point Model (DPM) requirements before submitting reports through the NCA’s reporting portal. A longer submission period would allow firms to complete the governance and validation activities while reducing the risk of operational errors and re-submissions.
4. An extended timeline would be particularly beneficial for FRTB ASA-only institutions, many of which will participate in the benchmarking exercise for the first time. Additional time would support familiarization with the submission process, and the completion of necessary validation checks, thereby contributing to higher-quality submissions across the industry.
5. The IMV submission phase involves the submission of multiple components, including:
   1. The IMV, The initial SbM sensitivities; and
   2. The SbM validation portfolios.

With regard to the timeline for the whole exercise, the industry also considered the proposal of both advancing or further postponing the benchmarking exercise into the latter part of 2027. However, neither option was considered preferable. 

• Advancing the exercise would create significant overlap with the EBA stress-testing programme. For many institutions, particularly smaller firms, the same teams and subject matter experts are responsible for both benchmarking and stress-testing activities, resulting in material resource constraints. An earlier timeline would also reduce the readiness time for new participants.
• Further postponing the exercise would create challenges associated with the summer holiday period, with July representing a peak holiday season particularly across the Nordic region, and August being a key holiday month in many other European countries. A later start would also compress the overall benchmarking timetable and reduce the time available for subsequent phases of the exercise, including the RM module submission and ISDA dry run.

For these reasons, the industry considers that the proposed extension of the IMV submission window would significantly improve operational feasibility, manage competing regulatory priorities more effectively while maintaining the quality of benchmarking submissions, within the proposed overall timeline for the exercise.

In addition, the industry notes that the proposed IMV module calculation date, 25 June 2027 (the fourth Friday of June), coincides with Midsummer Day, a public holiday in Sweden. As a result, Swedish institutions would be unable to obtain market data and valuations as of the prescribed calculation date and would instead need to rely on market data from the preceding business day. This may reduce consistency across participating institutions and could affect the comparability of benchmarking results. 

RM initial calculation date and final calculation date

The proposed dates for the Risk Measure (RM) calculation window will create an inconsistency in the period available for calculating current IMA risk measures (VaR, Stressed VaR and IRC).

Specifically, setting the RM initial calculation date as the third Monday of September and the RM final calculation date as the fourth Friday of September will result in only a 5-day calculation window in 2027 and 2028, whereas the period extends to 10 days in 2029:

• 2027: 20–24 September (5 days)
• 2028: 18–22 September (5 days)
• 2029: 17–28 September (10 days)

Templates C107–C109 appear to consider a 10-day calculation period. The industry would welcome clarification on the intended calculation period ensuring a consistent 10-day period.

Q3: Do you see any issues or unclarity with the reference date for new ASA institution in the exercise as defined in the suggested draft of Article 4.1.(b)?

In general, the industry welcomes the clarity provided regarding the reference date applicable to newly authorised ASA institutions participating in the 2027 exercise. Newly in-scope firms would benefit from further clarity regarding the participation process.

Q4: Do you see any issues with the new instruments / portfolios introduced in the Annex III?

The industry notes below three points:

• For the aggregated FX Cumulative Portfolio 13000, combining portfolios 3301 and 3303 would result in instrument 305 being included twice.
• The proposed instrument deletion includes all instruments with instrument currency as GBP or with EUR/GBP spot underlying. The industry considers GBP to be a material currency and therefore proposes retaining a selection of GBP-denominated instruments to ensure adequate coverage and representativeness, even if the number of GBP instruments included is relatively limited.
• Several instruments report the maturity date using the format “Booking Date + [X] Months”; however, this convention does not appear to be applied consistently across all instruments. Examples include Instruments 106, 115, 406, 505, 507, 514, 515, 602, 604, 606, and 608. Unless this difference is intentional, the industry would welcome a consistent approach to the presentation of maturity dates across all instruments and to ensure that the maturity date is updated accordingly following any changes to the booking date.

Q5: Despite some improvements, we note that the instructions set out in Annex I, point JJ, are still not fully followed by part of the submitters. Could you please explain the reasons for this and suggest any changes that could help ensure compliance?

• A number of reasons have been cited by the industry on the differing approaches taken across firms. Some firms noted that their systems automatically generate FX risk components as part of the standard risk calculation process, making it operationally challenging to exclude FX risks solely for benchmarking reporting purposes.
• Some firms noted that instruction was not clear.
• In addition, some firms indicated that their reporting approach was influenced by guidance received from their Joint Supervisory Teams (JSTs). Example when it was observed that systematic lower risk measures were submitted and contribution of FX risk was not considered.

In general, to promote consistency and comparability of benchmarking submissions across institutions, the industry would welcome clarification in the instructions via simpler language such as that FX risk needs to be reported only for the FX asset class portfolios if that is the intention of the EBA. Such clarification needs to apply consistently to both RM portfolio-level submission (as part of C120 templates) as well as the initial SBM sensitivities submission (as part of C106.01 template) as part of IMV module at instrument level.  To further support consistent implementation and compliance, the industry recommends that the instructions be supplemented with practical examples. 

Q6: Are the instructions and templates clear to the respondents? If not, what should be clarified? Do the respondents identify any discrepancies between these templates and instructions and the calculation of the requirements set out in the underlying regulation?

European Commission’s  3^rd Delegated Act (DA) transitional provisions

The industry would welcome clarification that, for benchmarking submissions, firms should reflect all elements of the European Commission’s  3^rd Delegated Act (DA) transitional provisions,  except for the bank-specific capital-neutral multiplier applied to the overall OFR.

Certain transitional provisions may already be embedded within banks’ FRTB SA calculation engines (for example, tenor correlations for carbon trading risk commodity risk factors), making it operationally challenging and potentially burdensome to perform separate calculations solely for benchmarking purposes.

Whilst the industry is comfortable with not applying the 0.9 SbM multiplier if required, it would appreciate explicit instructions to ensure a consistent interpretation and application of the transitional provisions for benchmarking submissions across all participating firms. Such clarification would support a harmonised approach for the IMV module, RM module, and ASA validation portfolio submissions, thereby enhancing the comparability and consistency of benchmarking results.

Benchmarking of current IMA (Basel2.5 – VaR / sVaR measures)

• The industry would welcome clarification on whether institutions that have implemented the FRTB Standardised Approach without bank-specific capital-neutral multiplier are expected to continue participating in the CRR2 IMA benchmarking data collection exercise. Our understanding is that such institutions no longer fall within the intended scope of this CRR2 IMA benchmarking exercise, particularly where they are not implementing the bank-specific multiplier. Given the supervisory purpose of the exercise, we believe CRR2 IMA benchmarking should be limited to institutions for which the resulting benchmark remains directly relevant and would welcome confirmation of this.
• To avoid potential ambiguity, the industry recommends the amendment of Annex 1 Section 1 instruction (aa) to explicitly include the requirement to submit current IMA (Basel 2.5 – VaR/sVaR) measures where applicable.

Terminology inconsistency between Annex I and Annex II, and its impact on template C 130.02 (MKR IMA SP)

The industry highlights a drafting discrepancy between the definitions in Annex I (Booking Instructions) and the relevant dates specified in Annex II (Relevant dates of the exercise) of the draft ITS 2027. This inconsistency creates ambiguity in how institutions must populate Template C 130.02 (MKR IMA SP).

1. The terminology discrepancy (Annex I vs. Annex II)

While Annex II updates its terminology replacing "reference dates" with "calculation dates" (e.g., “RM initial calculation date” and “RM final and ASA calculation date”) – these updates are not reflected in Annex I points (a), (b), and (aa).

Annex I continues to define and refer to the ITS 2026 terms "RM initial reference date" and "RM final reference date", which no longer exist in the Annex II timeline.

2. Impact on template C 130.02 (MKR IMA SP)

This terminology mismatch introduces an ambiguity in the reporting of stress periods in Template C 130.02 (MKR IMA SP), specifically for Column 0040 (First date as of which the stress period was used in the calculation of the measure) and column 0050 (Number of reviews of the stress period during the reference period).

The instructions for column 0040 require institutions to:

“...indicate the date corresponding to the first day of the three months period ending with the reference date (reference period)...”

Because of the terminology update, it is unclear which "reference date" is the anchor for this 3-month lookback window, which in turn distorts both columns:

• Interpretation A: The RM module reporting reference date (last calendar day of September, per Annex II, point vii).
  • Impact on column 0040: The 3-month lookback window is bounded by the calendar quarter (e.g., July 1st to September 30th) and aligned with COREP.
  • Impact on column 0050: The reference period spans exactly three full calendar months. Under the quarterly stress-period selection mandate, and in case of no significant change in the markets, institutions would consistently report 1 review.
• Interpretation B: The RM final and ASA calculation date (fourth Friday of September, per Annex II, point vi – which represents the actual final calculation date, formerly known as the "final reference date").
  • Impact on column 0040: The 3-month lookback window is bounded by the actual execution period (e.g., June 24th to September 24th, 2027).
  • Impact on column 0050: Because this period cuts across calendar quarters, the number of captured quarterly review cycles is highly sensitive to the exact date an institution runs its selection process. Depending on the schedule, identical quarterly processes could capture 0, 1, or 2 reviews within this window.

A difference of even a few days between these two interpretations (e.g., June 24th vs. July 1st) determines whether a model recalibration or stress-period update is classified as occurring before or during the active benchmarking window. Furthermore, the ambiguity in column 0050 will result in institutions reporting different number of reviews for identical quarterly governance processes. 

In conclusion the industry would welcome:

• Alignment of the terminology in Annex I with the updated "calculation date" definition established in Annex II to ensure consistency across the draft ITS.
• Further clarification in the instructions for template C 130.02 to clearly link the "reference date" used for establishing the 3-month lookback window with the Annex II terminology.

Q7: 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?

The industry strongly reiterates the point to allow for bilateral consultation with the relevant supervisory team to assess if a firm is required to submit ASA validation portfolio results given they have submitted in previous benchmarking rounds and considering the diminishing returns observed from the execution of SBM Validation portfolios recurring on a year-on-year basis. This will be aligned to the EBA’s objective of promoting ‘simpler, smarter and proportionate’ reporting.

The industry welcomes the postponement of FRTB AIMA benchmarking due to the limited initial uptake of AIMA and expected transitional period. As only an extremely small number of banks in the EU are currently approaching CRR3 IMA readiness, near-term CRR3 IMA benchmarking would generate no meaningful peer statistics or comparative data. Consequently, the industry would welcome a deferral of making these specific templates mandatory in the 2027 exercise to allow the underlying calculation methodology to stabilise first.

Q8: 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 industry is supportive of removing repetitive instruments. However, changes to instrument specifications, portfolio structures, and instrument-to-portfolio mappings can result in significant implementation effort for participating institutions. In many firms, benchmarking portfolio hierarchies and mappings are managed separately from production risk management structures and desk hierarchies. As a result, even targeted changes may require portfolio hierarchies, mappings, controls, and validation processes to be updated and retested.

Industry participants estimate that such changes can typically result in an additional one to two weeks of effort ahead of the benchmarking exercise. The industry would therefore welcome the avoidance of significant year-on-year reshuffling of instruments and portfolio structures, where possible, to reduce recurring implementation effort and operational burden for participating institutions.

Upload files

Name of the organization