Final Draft ITS 2023 on Benchmarking of Internal Models

FINAL REPORT ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS EBA/ITS/2022/04 05/05/2022 Final Report Draft Implementing Technical Standards on amending Commission Implementing Regulation (EU) 2016/2070 with regard to benchmarking of internal models

FINAL REPORT ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS Contents

1. Executive Summary 3
2. Background and rationale 4
3. Draft implementing standards 11
4. Accompanying documents 14
   4.1 Draft cost-benefit analysis for changes related to credit and market risk benchmarking 14
   4.2 Feedback on the public consultation 18

FINAL REPORT ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS

1. Executive Summary
   Article 78 of Directive 2013/36/EU (CRD) requires competent authorities to conduct an annual
   assessment of the quality of internal approaches used for the calculation of own funds requirements. To assist competent authorities in this assessment, the EBA calculates and distributes benchmark values that allow a comparison of individual institutions’ risk parameters. These benchmark values are based on data submitted by institutions as laid out in Commission Implementing Regulation (EU) 2016/2070 which specifies the benchmarking portfolios, templates and definitions to be used as part of the annual benchmarking exercises. For the 2023 benchmarking (BM) exercise the following changes are suggested:

• For credit risk, no changes are proposed with respect to the data collection templates, and
  two minor changes have been made to the portfolio’s definition. However, in order to improve further the data collection and benchmarking analysis, some further clarifications are included in the instructions.
• For market risk, in order to keep the exercise updated and informative for supervisors, the set
  of instruments is proposed to be extended. Therefore, to the previous set of instruments, which are mostly plain vanilla, the proposal is to add a more complex set of instruments that could provide additional information and analysis insights to supervisors and banks.
• For IFRS 9, no changes to existing templates are envisaged.
  Next steps
  The Annexes presented in these draft ITS replace or are added to the existing set of templates in order to create a consolidated version of the updated draft ITS package. These draft ITS will be submitted to the Commission for endorsement before being published in the Official Journal of the European Union. The technical standards will apply 20 days after publication in the Official Journal.

FINAL REPORT ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS
2. Background and rationale
2.1 Credit risk benchmarking

1. The templates for the data collection for credit risk (CR) benchmarking (BM) are specified in
   Annexes I-IV of the ITS. Annex I specifies the benchmarking portfolios via a set of characteristics and Annex II provides the relevant definitions for this. Annex III contains the actual parameters and metrics that institutions are to report for the portfolios defined in Annex I. Finally, Annex IV provides the definitions and descriptions relevant for Annex III. The proposed review of the ITS for the 2023 exercise has been developed with the objective to ensure that the CR BM data collection is:
   a. fit for purpose; b. adjusted to the nature of the information;
   c. specific to the analysis proposed; and
   d. expected to be stable for a foreseeable time horizon.
2. Starting from these objectives, the EBA proposes to keep the CR BM portfolios and metrics stable
   and shifting the focus to improving the clarity of the instructions. In this regard clarification is proposed for the reporting of default rates where the underlying definition of default has changed. In addition, a potential review of the data collection of loss rates is discussed – which will, however, rather be used for future reviews of the ITS.
   2.1.1 Clarifications on the data collection in the case of a change of the definition
   of default (DoD)
3. As part of the EBA’s roadmap on the review of the IRB approach 1
   , the relevant regulation for the identification of credit defaults has been reviewed. The deadline for the institutions for implementing both of the resulting policies is as follows: ‘the guidelines on the definition of default (DoD) and the RTS on the materiality threshold are aligned with all other regulatory products developed as part of the EBA’s regulatory review of the IRB approach, i.e. they apply at the latest from 1 January 2021 with regard to the application of the definition of default in default identification processes and have to be implemented in all rating systems by the end of 2021 or, in specific cases, the end of 2023. The final guidelines on the application of the definition of default across the EU and the final draft RTS on the materiality threshold for credit obligations past due were published in September 2016 2 .’ 1 EBA publishes report on progress made on its roadmap to repair IRB models | European Banking Authority (europa.eu) 2 In this regard, the EBA also published an opinion on the national discretion to use 180 days past due, which proposes that the continued application of the exemption should be disallowed. See https://eba.europa.eu/-/eba-advises-thecommission￾to-disallow-the-application-of-the-180-day-past-due-exemption-for-material-exposures

FINAL REPORT ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS
4. With the implementation of a common definition of default in institutions, the following
clarifications are set out:
a. Column c070 (default status) in template C 101 in Annex I is to be reported in line with the implemented DoD as of the reporting reference date. b. In templates C 102 and C 103 of Annex I, exposures shall be assigned to portfolio IDs based on their default status in line with the implemented DoD as of the reporting reference date.
c. For columns 0200 (default rate past 5 years) and 0220 (loss rate past 5 years) of
template C103 in Annex III, the institutions should reflect in the averages the defaults of a respective year with respect to the DoD that was in place in that year, i.e. defaults and default dates must be recorded under the DoD applicable at the time of the event. There is no expectation that banks report in these columns averages which relate to annual rates based on a DoD newly implemented (i.e. after the relevant reference date) for each year considered. Hence no backwards simulation of the new definition of default is expected. d. Similarly, for columns c0160 of C 102 and C 103, one-off credit risk adjustments in connection with the implementation of the new DoD should be reported as recorded in the bank’s database. Backward simulations should not be applied. e. Finally, columns c050 – c090 of template C 105.01 should be reported reflecting the implemented and approved rating systems at the reference date.
2.1.2 Clarification and completion of portfolio breakdowns
5. For the ITS 2023 one portfolio has been added in Annex 1 to complete the breakdown into FINREP
sectors with portfolios treated under the supervisory slotting approach where the sector of the counterparty is central governments under FINREP. Therefore, the following portfolio ID has been added:
COSP_GOV_0223_CT_SLSC_ x0_Rx0_ALL
6. For the ITS 2023 exposures with unfunded credit protection treated under the substitution
approach, which are already shifted to the corresponding guarantor exposures classes, shall now be reported under the portfolios specified along column 0120 in Annex II for templates C102 and C103. Furthermore, other funded credit protection with an effect on the exposure value shall now

FINAL REPORT ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS be considered in the collateral type breakdown. Further guidance on the reporting of these portfolios is provided in Q&A 2018_4093 3 and Q&A 2018_4091 4 .
2.2 Market risk
7. This year’s update to the market risk benchmarking exercise extends the existing set of instruments
and portfolios. This step is considered a natural upgrade of the exercise, as there has been stability in the composition of instruments and portfolios since the 2019 exercise. This extension of the exercise ensures that the exercise is kept informative and provides new insights for the supervisors and banks.
2.2.1 Change in reference numbering for instruments
8. The proposed changes in the CP 2023 introduce additional instruments to all asset classes in the
exercise (EQ, IR, Fx, CmD, CS and CTP). Since the proposal focuses on the new instruments (and new portfolios), there will be no fundamental change to the previous structure of the templates, i.e. only minor updates to Annex 6 (MR template instruction) and no update to Annex 7 (MR template). The main changes therefore solely relate to Annex 5, where new instruments are included.
9. In addition, there is a renumbering of the instruments. In the previous exercise, the instruments
were identified by sequential numbers (e.g. 1 to 81) irrespective of the asset class. This numbering implied that any addition or deletion of an instrument in the past led to a subsequent renumbering of the instruments. Hence, identification numbers and associated risk classes of previously existing instruments also changed. Such changes led to confusion and possible mistakes in submissions. 10.To rectify this issue, this year’s proposal introduces a new numbering, such that the instrument numbers reset for any asset class. Hopefully, this change will help in the future in case the EBA wants to amend/improve the composition of the instruments of a specific asset class, as it could do so without changing the reference of the others. The EQ asset class will be identified by the number 100, and the IR will be 200, Fx will be 300, Cmd will be 400, CS will be 500 and CTP will 600. The description of the instruments remains unchanged for the old instruments. This stability will help keep continuity in the exercise and avoid interpretation mistakes, providing a stable core of information to the EBA on the market model benchmarking. The table below will support the visualisation of the changes in the numbering of the portfolio. 3 2018_4093 Category on which the covered part of exposures should be reported | European Banking Authority (europa.eu) 4 2018_4091 Reporting of exposures whose collateral type is (g) credit derivatives, (h) guarantees or (i) unfunded credit protection | European Banking Authority (europa.eu)

FINAL REPORT ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS
2.2.2 New instruments
11.From the table above, it should be clear that there are a substantial number of new instruments in the exercise. To be specific: six new EQ instruments, four new IR instruments, four new Fx instruments, one new Cmd instrument, six new CS instruments and ten new CTP instruments. For a total of 31 new instruments are added to the exercise. 12.The expansion of the instruments in the exercise is needed to increase the scope of the observation provided to competent authorities on the performance of the market risk model. 13.Also, this increases the significance and representativeness of the instruments in the exercise, which have been criticised in the past to be too simple to represent an accurate composition of the assets in the banks’ trading books. 14.It should be noted that the EBA has also been in contact with the ISDA industry group, which is conducting a benchmarking exercise 5 alongside the EBA’s exercise, which was based initially on the EBA set of instruments. The ISDA benchmarking exercise evolved with time, from the market risk standardised method to SACCR and CVA, and so it evolved in the composition of instruments in the 55 https://www.isda.org/2020/06/17/isda-sa-benchmarking/ ITS 2022 1 - 18 19 - 38 ITS 2023 101 - 118 119 - 121* 201 - 220 221 - 224* ITS 2022 39 - 47 48 - 51 ITS 2023 301 - 309 310 - 311* 401 - 404 405* ITS 2022 52 - 79 80-81 ITS 2023 501 - 528 529 - 534* - 601 - 610* *added instruments (Industry input) CS CTP Instruments - Assets classes EQ IR Fx Cmd

FINAL REPORT ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS exercise. Therefore, in order to increase the variety of the instruments, the EBA decided to adapt the new instruments to some of the instruments already applied by the industry. 15.By adapting existing portfolios, this should facilitate the introduction of new instruments, as institutions will already be familiar with these instruments. Moreover, these instruments have already shown at least some degree of consensus on their significance among banks.
16. It is, however, clear that not all of the instruments can be directly replicated, as there is an EU focus
in the EBA exercise, which should prevail when choosing specific instruments. Therefore, where needed, these industry instruments were updated while keeping the substantial structure of the instruments, but adapting it to the EU market (e.g. the underlying index or issuer were replaced with a European equivalent where needed). 17.In this regard, as can be seen in Annex 5, sections 2 and 5 for the details, the following instruments are new with respect to the previous exercise. a. Equity instruments 18.Instrument 119 – long call on Euro Stoxx 50; instrument 120 – long call on Euro Stoxx 600; instrument 121 – long call on VIX. b. Interest rate instruments 19.Instrument 221 – 10-year IR swap on ESTER; instrument 222 – inflation-linked bond; instrument 223 – 5-year zero coupon inflation swap; instrument 224 – 2-year EUR swaption (OTM) on 5-year interest rate swap.
c. FX instruments
20. Instrument 310 – 6-month EUR/DKK forward contract; instrument 311 – 6-month EUR/BRL non￾deliverable forward contract.
d. Commodities instruments
21.Instrument 405 – long call on gold. e. CS instruments 22.Instrument 529 – long CDS on iTraxx Europe; instrument 530 – short put on iTraxx Europe; instruments 531 – 534 – long positions on callable bonds with different maturities. CTP instruments 23.Instruments 601 – 610 – a series of positions on iTraxx Europe and CDS on iTraxx Europe.

FINAL REPORT ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS
2.2.3 New portfolios with updated numbering and Annex 6 amendment
24.As noted in the previous sections, a series of instruments was added to all assets classes. Moreover, the numbering of the old instruments (i.e. same as the previous exercise) was updated. Consequently, the references of instruments in the portfolios had to be updated. 25.The reference number of the portfolios is also updated, following the same logic asthe instruments. The numbering of the individual and aggregated portfolios is now defined in ‘thousands’ and restarts from 1 for each asset class (i.e. the first portfolio of equity instruments is 1001, the first portfolio of IR instruments is 2001, etc.). 26.Also, new portfolios had to be added to reflect the new instruments, as described in the previous
section. Moreover, a few portfolios were added to reflect some hedging strategies as suggested in
the EBA 2014 CP 6
.
27.The new individual portfolios’ reference numbers compared to the old ones (ITS 2022) are reported in the table below. 28.Finally, In Annex 6, in the section “C 108.00 - Profit & Loss Time Series” of Annex VI, in the second
table, in the first row of the fourth column (Instructions), the date ‘28 January 2022’ is replaced by
the more general provision ‘the RM (and final SBM) final reference date, Annex 5, Section 1 letter (b)(v) of this Implementing Regulation’. In a nutshell, every year banks with historical simulation models are required to provide a series of one-year P&L, to allow the EBA to generate a VaR and ES comparable with the data submitted by the banks. Since this date merely rolls over from one year to another with respect to the submission dates of the risk measure in the exercise, it simplifies the ITS amendment process to just put the reference to the date specified in Annex 5. 6
Annex VII.a – EBA market risk benchmark portfolios.pdf (europa.eu)

FINAL REPORT ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS
2.2.4 Changes in the final ITS with respect to the consultation paper
29.Instruments 122, 123 and 124 were removed as suggested in the feedback received in the ITS consultation. 30.The remittance dates were reset to match the deadlines provided in the 2022 exercise. 31.Equity instruments notional was better specified in order to avoid misalignment in the book phase. 32.Annex 6 was amended to address a consistency issue in the reporting of curvature risk factor in templates 106.01 and 120.01. 33.A series of minor amendments was provided for instruments: 117, 202, 204, 205, 220 – 224, 311, 532 – 4, 602, 604 – 8, 610. Instrument 611 was deleted. Portfolios 2018,2019 and 5019 were also slightly amended to keep them consistent, and portfolios 6006 and 6007 were removed. Full details of the changes are reported in the feedback table. ITS 2022 1 - 10 11 - 27 ITS 2023 1001 - 1010 1011 - 1013* 1014 - 1016 2001 - 2017 2018 - 2019* 2020 - 2023 ITS 2022 28 - 32 33 - 35 ITS 2023 3001 - 3005 3006 - 3007 4001 - 4003 4004 ITS 2022 36 - 53 37 - 59 54-55-56 ITS 2023 5001 - 5018 5019* 5020 - 5022 5023 - 5027 - 6001 - 6005 *added portfolios (EBA 2014 CP input) added portfolios (Industry input) CS CTP Portfolios - Assets classes EQ IR Fx Cmd

FINAL DRAFT PAPER ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS 11
3. Draft implementing technical
standards
EBA/ITS/2022/04
05/05/2022
Draft implementing technical standards amending Commission Implementing Regulation (EU) 2016/2070 on benchmarking of internal models

FINAL DRAFT PAPER ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS 12 COMMISSION IMPLEMENTING REGULATION (EU) No …/... of [date] amending Implementing Regulation (EU) 2016/2070 as regards benchmark portfolios, reporting templates and reporting instructions to be applied in the Union for the reporting referred to in Article 78(2) of Directive 2013/36/EU of the European Parliament and of the Council THE EUROPEAN COMMISSION, Having regard to the Treaty on the Functioning of the European Union, Having regard to Directive 2013/36/EU of the European Parliament and of the Council of 26 June 2013 on access to the activity of credit institutions and the prudential supervision of credit institutions, amending Directive 2002/87/EC and repealing Directives 2006/48/EC and 2006/49/EC 7 , and in particular Article 78(8), the third subparagraph thereof, Whereas:
(1) Commission Implementing Regulation (EU) 2016/2070 8 specifies the reporting requirements for institutions to enable the European Banking Authority (‘EBA’) and competent authorities to monitor the range of risk-weighted exposure amounts or own funds requirements for the exposures or transactions in the benchmark portfolio resulting from the internal approaches of those institutions, and to assess those approaches as required by
Article 78(3) of Directive 2013/36/EU.
(2) Considering that, pursuant to Article 78(1) of Directive 2013/36/EU, the benchmarking exercise is of at least annual duration and that the focus of the competent authorities’ assessments and of EBA’s reports has changed over time, in order to identify areas where further regulatory guidance is needed exposures or positions that are included in the benchmark portfolios, and therefore also reporting requirements, need to be adapted accordingly. It is therefore appropriate to amend Annexes IV, V and VI to Implementing Regulation (EU) 2016/2070. (3) Implementing Regulation (EU) 2016/2070 should be amended accordingly. (4) This Regulation is based on the draft implementing technical standards submitted to the Commission by the EBA. (5) EBA has conducted open public consultations on the draft implementing technical standards on which this Regulation is based, analysed the potential related costs and benefits, and 7 OJ L 176, 27.6.2013, p. 338–436. 8 Commission Implementing Regulation (EU) 2016/2070 of 14 September 2016 laying down implementing technical standards for templates, definitions and IT solutions to be used by institutions when reporting to the European Banking Authority and to competent authorities in accordance with Article 78(2) of Directive 2013/36/EU of the European Parliament and of the Council (OJ L 328, 2.12.2016, p.1).

FINAL DRAFT PAPER ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS 13 requested the advice of the Banking Stakeholder Group established in accordance with
Article 37 of Regulation (EU) No 1093/2010 of the European Parliament and of the Council 9
.
HAS ADOPTED THIS REGULATION:
Article 1
Implementing Regulation (EU) 2016/2070 is amended as follows:
(1) Annex I is replaced by the text in Annex I to this Regulation; (2) Annex II is replaced by the text in Annex II to this Regulation; (3) Annex IV is replaced by the text in Annex IV to this Regulation; (4) Annex V is replaced by the text in Annex V to this Regulation; (5) Annex VI is replaced by the text in Annex VI to this Regulation.
Article 2
This Regulation shall enter into force on the twentieth day following that of its publication in the Official Journal of the European Union. This Regulation shall be binding in its entirety and directly applicable in all Member States. Done at Brussels, For the Commission The President On behalf of the President [Position]
ANNEX
Annex I (Credit Risk Benchmarking)
Annex II (Credit Risk Benchmarking)
Annex IV (Credit Risk Benchmarking)
Annex V (Market Risk Benchmarking)
Annex VI (Market Risk Benchmarking)
9
Regulation (EU) No 1093/2010 of the European Parliament and of the Council of 24 November 2010 establishing a European Supervisory Authority (European Banking Authority), amending Decision No 716/2009/EC and repealing Commission Decision 2009/78/EC (OJ L 331, 15.12.2010, p. 12).

FINAL DRAFT PAPER ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS 14
4. Accompanying documents
4.1 Draft cost-benefit analysis for changes related to credit and
market risk benchmarking
Article 78 of Directive 2013/36/EU (CRD IV) requires competent authorities to conduct an annual
assessment of the quality of internal model approaches, used for the calculation of own funds requirements, and requires the EBA to produce a report to assist them in this assessment. The report of the EBA relies on data submitted by institutions in accordance with EU Regulation 2016/2070, which specifies the benchmarking portfolios, templates, definitions and IT solutions to be used by the institutions as part of the annual benchmarking exercise, when using internal model approaches for market and credit risk. The current draft ITS aim to update the previous ITS for the benchmarking data collection with the purpose of improving the exercises and adapting to the relevant policy changes which will be applicable by end-2022 and thus relevant for the 2023 exercise. Regarding the EBA’s market risk benchmarking data collection, the purpose is to extend the set of instruments to keep the exercise relevant and informative for banks and supervisors joining the exercise. For the credit risk and IFRS 9 benchmarking data collection no material changes have been made with respect to the portfolio definitions or metrics. Therefore, an in-depth impact assessment is not considered relevant. As per Article 15(1) of the EBA Regulation (Regulation (EU) No 1093/2010 of the European Parliament and of the Council), any ITS developed by the EBA shall be accompanied by an impact assessment (IA) annex which analyses ‘the potential related costs and benefits’ before submitting to the European Commission. Such an annex shall provide the reader with an overview of the findings as regards the problem identification, the options identified to remove the problem and their potential impacts. For the purposes of the IA section of the Consultation Paper, the EBA prepared the IA with cost￾benefit analysis of the policy options included in the regulatory technical standards described in this Consultation Paper. Given the nature of the study, the IA is mainly high-level and qualitative in nature, including quantitative analysis when possible.

FINAL DRAFT PAPER ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS 15 A. Problem identification With regard to the market risk benchmarking data collection, the previous ITS for benchmarking data collection have remained substantially stable, in terms of the set of instruments in scope of the exercise, since the change in the instruments/portfolios definition for the 2019 exercise. B. Policy objectives The general objective of the current ITS is to update the previous ITS for benchmarking data collection. The main objective of the implementation of the current draft benchmarking ITS is the The specific objective for market risk is to extend the set of instruments and portfolios to be benchmarked, to keep them relevant and informative for banks and supervisors participating to the exercise, without changing the framework of the data collection. harmonisation of the current reporting framework rules amongst EU institutions. This would foster the strategic objective is to create a supervisory and reporting environment to ensure that institutions apply consistent modelling and valuation techniques. The following sections examine the options that could create such an environment, as well as the net impact that the implementation of such solutions implies.
C. Baseline scenario
For the market risk part of the exercise, for most EU institutions, the current status of reporting the results of modelling and valuations implies the usual potential operational costs and miscalculations, which lead to overvaluation or undervaluation of the reported values for the purposes of the benchmarking exercises. Since the extent and magnitude of overvaluations or undervaluations cannot be identified, the impact assessment focuses on the assessment of the net impact on the institutions’ operations. D. Options considered When developing the draft ITS, the EBA considered the following options:
Option 1: do nothing
This option implies that credit institutions continue reporting data for the benchmarking exercise:

• using just the same hypothetical portfolios as defined for the exercises to date;
• using the current guidance, templates and portfolios for the credit risk exercise.

FINAL DRAFT PAPER ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS 16 For the market risk part of the exercise, the continuation of the application of just the previous set of instruments assumes that credit institutions and the EBA have the usual operational cost assigned to providing clarifications and ensuring the consistent submission of data. The ‘do nothing’ option would imply leaving the Implementing Regulation on market risk benchmarking unchanged, Annex V, would result in obtaining almost the same results as the previous exercise, with a loss of relevance and significance for banks and competent authorities in the data collection. Option 2: revision of the guidance relating to the benchmarking exercises The main arguments that support the revision of the composition of the instruments in the benchmarking exercises are:
(i) to enhance the significance of the benchmarking exercises across all EU credit institutions; (ii) potentially providing new insights into the different functioning of the market risk model. For the market risk part of the exercise, the current ITS could achieve the objective by expanding the set of instruments and portfolios collected. With some new additional instruments, slightly more complex than the usual set of plain vanilla instruments, the data collection could be more relevant, in terms of being closer to the instruments the banks actually trade in their trading book, and more informative, providing new elements of analysis, for banks and competent authorities. E. Cost-benefit analysis The principle of proportionality applies to all aspects of the impact assessment, including methodology, depth of analysis, level of detail and necessity of quantitative analysis. Being consistent with this principle, the EBA staff follow the principle of proportionality when conducting the cost-benefit analyses. Given that the implementation of the current ITS would not have a detrimental impact, the following analysis focuses on the qualitative characteristics. In doing so, it provides rough estimations on the net monetary impact that relates to the conduct of benchmarking exercises. The net impact on capital requirements, implied by the implementation of the current guidelines, cannot be precisely assessed because, substantially, it would depend on further actions agreed by institutions with national competent authorities in response to the benchmarking exercise results; however, it is expected to be on average close to zero due to the hypothetical market portfolio exercise framework. Market risk:

FINAL DRAFT PAPER ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS 17 Option 1 Costs: a possible loss of informativeness in the data collection, which would be substantially identical to the previous one. Benefits: one-off benefits (reduction of the existing operational costs) of not dedicating human resources to the drafting the present ITS. Option 2 Costs: the one-off cost of dedicating EBA staff resources to the drafting of the ITS. There is also a source of negligible cost that relates to the need for the EBA to explain the new set of instruments to the national competent authorities and, through them, the participating credit institutions. However, it is to be noted that the data requested with the new instruments could not be too burdensome, since the instruments are not too exotic and some banks already know them well because some banks apply them (or very similar instruments) in the industry benchmarking exercise. Benefits: the benefits of this option arise from providing new information and data on new instruments, which would trigger the provision of additional insights to competent authorities and would keep relevant the exercise for the banks involved. F. Preferred option The EBA considers that, although these benefits are not directly observable and are spread over time, they are not negligible, and they are considered more important than the costs enumerated above. For this reason, the preferred option is Option 2.

FINAL DRAFT PAPER ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS 18
4.2 Feedback on the public consultation
The EBA publicly consulted on the draft proposal contained in this paper.
The consultation period lasted for nine weeks and ended on 18 February 2022. Three responses were received, of which two were published on the EBA website. This paper presents a summary of the key points and other comments arising from the consultation, the analysis and discussion triggered by these comments, and the actions taken to address them if deemed necessary. Changes to the draft ITS have been incorporated as a result of the responses received during the public consultation. Summary of key issues and the EBA’s response The respondents welcome the EBA efforts to improve the effectiveness of the exercise such as the renumbering of instruments / portfolios and note that the EBA has incorporated some of the instrument portfolios also included in the industry exercise in the scope of the ISDA’s FRTB-SA benchmarking exercise. Notwithstanding the improvements made compared with the instruments now in scope, the growth of the hypothetical portfolios year on year requires considerably more maintenance and computation capacities for the IMA exercise, therefore the respondents kindly request that the number of instruments be kept stable going forward. The respondents have identified two of the additional instruments as problematic and request that these be removed from the EBA exercise. In addition, the IMV (and initial SBM) as well as the risk measure (and final SBM) submission dates have been brought forward and that will result in a shorter period for banks’ submission as well as for performing the necessary data validation; the respondents recommend that the timelines from the 2022 exercise be retained. The EBA carefully reviewed the suggestions and amended the Consultation Paper text where it was deemed needed. The complete list of changes is reported below in the table.

FINAL DRAFT PAPER ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS 19 Summary of responses to the consultation and the EBA’s analysis Comments Summary of responses received EBA analysis Amendments to the proposals General comments PD/LGD excl. supervisory measures, PD/LGD excl. supervisory measures & MoC, LGD excl. supervisory measures & MoC & downturn RWA add-ons in C105.01 One respondent claimed that there might have been a mistake in the Consultation Paper and final draft ITS as they contain the data fields relating to conservatism in the estimation, contrary to the requirements that were ultimately published in the Official Journal of the EU (Commission Implementing Regulation (EU) 2021/1971) where these extra metrics are not contained. The EBA would like to point out that the requirements that were published in September 2021 in the Official Journal of the EU (Commission Implementing Regulation (EU) 2021/1971) reflect the legal implementation of the EBA’s draft final ITS for the 2020 benchmarking exercise. The draft final ITS for the 2022 exercise are published on the EBA website and are expected to be published in the Official Journal of the EU in Q2 2022. None Two instruments problematic and submission dates The respondent welcomes the EBA efforts to improve the effectiveness of the exercise such as the renumbering of instruments / portfolios and notes that the EBA has incorporated some of the instruments and portfolios also included by the participants in the ISDA’s FRTB-SA benchmarking exercise. Notwithstanding the improvements made compared with the instruments now in scope, the growth of the hypothetical portfolios year on year requires considerably more The EBA agrees with the removal of the two problematic instruments and reset the submission date as they were set for the 2022 exercise. Text amended as reported in the specific comment below, in line with the suggestions

FINAL DRAFT PAPER ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS 20 Comments Summary of responses received EBA analysis Amendments to the proposals maintenance and computation capacities for the IMA exercise, therefore the respondent kindly requeststhat the number of instruments be kept stable going forward. The respondent has identified two of the additional instruments as problematic and requests that these be removed from the EBA exercise. In addition, the IMV (and initial SBM) as well as the risk measure (and final SBM) submission dates have been brought forward and that will result in a shorter period for banks’ submission as well as for performing the necessary data validation; the respondent recommends the timelines from the 2022 exercise be retained. Question 1: For the purpose of reporting the above-mentioned fields, would you make use of the possibility to report the default and loss rate in template C 103 of Annex III with respect to a consistent but back-simulated definition of default, or would you report these fields with respect to the definition of default that was in place at the time of the event? Please briefly explain the underlying reasons and your motivation. Use of backwards simulated default and loss rates The respondents were critical regarding the possibility to report the default and loss rate in template C 103 of Annex III with respect to a consistent but back-simulated definition of default. The added value of such a voluntary reporting would be very limited, according to one respondent, as the extent to which the default definition can be simulated backwards can differ from bank to bank or even from model to model. The EBA is mindful of the balance between the consistency of the definition of default underlying the average default and loss rates in the reporting of a single institution and the comparability of the reported metrics referring to historical data across institutions. In the light of the feedback the EBA has chosen to adjust the instructions in favour of requesting the information related to historical defaults with respect to the definition of default that was in place at the time of the event. Clarification

FINAL DRAFT PAPER ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS 21 Comments Summary of responses received EBA analysis Amendments to the proposals Reporting of five years’ average default and loss rates where the underlying definition of default has changed One respondent makes the point that the time window underlying the metrics relating to averages of default rates and loss rates should be reduced until both values have five years of historical information, considering the revised definitions of default. The EBA will be mindful of the difficulty of comparing institutions’ metrics, which may relate to non-consistent underlying processes for default identification over time. However, the EBA considers that the monitoring of the implementation of the revised definition of default will serve the purpose of restoring the trust in the IRB approaches. Question 2: To evaluate the complexity as well as the costs and benefits of a change in the definition of loss rate in the context of the CR BM data collection, the EBA seeks views on enhancing the CR BM exercise with respect to its ability to reveal significant underestimation of LGD on portfolios with comparable characteristics. Industry views are welcome as regards the following questions:
a) Please comment on the expected operational burden if a reporting of realised losses / realised LGDs with respect to closed cases would be required (e.g. either by benchmarking portfolio as specified in c103 of Annex 1 or by LGD model as specified in C105 of Annex III). b) Which alternative metrics could be used for the benchmarking of LGD estimates? Responses to this question will not have a direct impact on the 2023 ITS. The input will therefore only be used as input for future reviews of the ITS. No responses were received with respect to question 3. Question 4: Do you agree with the new instruments added, according to section 3.3.2? Please indicate any issues, specified by instruments, and provide an actual suggestion of potentially missing information in the instruments’ descriptions (as provided in sections 2 and 5 of Annex 5 to these ITS). Instruments 122 and 123 The respondent has identified issues with certain instrument specifications and is proposing the amendments below. The EBA agrees with the removal of the two problematic instruments. Two instrument were removed in the final text.

FINAL DRAFT PAPER ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS 22 Comments Summary of responses received EBA analysis Amendments to the proposals The respondent requests that instruments 122 and 123 be dropped from the exercise as both instruments are non-standard products which could increase the dispersion of the benchmarking results. These were instruments 1004 and 1005 in the Phase 3 benchmarking exercise conducted by the ISDA and both instruments have subsequently been dropped from the ISDA’s Phase 4 benchmarking exercise. As a minimum, instrument 123 should have additional specifications in section 5. Instrument 117 Instrument 117 – Nikkei futures, expiry date should be 8 June, year T. The EBA agrees with the suggestion. Text amended Instrument 123 Instrument 123 – Total return swap, should be Pay 1-month EURIBOR. The instrument was removed based on the previous suggestion. Text amended Instrument 202 Instrument 202 – Swaption, should reference instrument 201. It is not clear if the swaption should hedge the swap, therefore the respondent recommends restating the swap details. The EBA agrees with the suggestion. Text amended Instrument 205 Instrument 205 – Collared note, should explicitly state ‘Base Currency USD’. The EBA agrees with the suggestion. Text amended Instrument 220 Instrument 220 – MtM reset xccy swap, the respondent suggests making this forward starting (6 months) so the exchange at the effective date has not happened and there can be no confusion over The EBA agrees with the suggestion. Text amended in
section 2 and section
5

FINAL DRAFT PAPER ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS 23 Comments Summary of responses received EBA analysis Amendments to the proposals the cash balance to be included or not. N.B. in
section 5 the coupon rate should be 3.
Instrument 221 Instrument 221 – Receive ESTER pay floating rate, the description should be amended as ESTER is also The EBA agrees with the suggestion. Text amended Instrument 223 Instrument 223 – ZC inflation swap, should explicitly state ‘base currency EUR’. The EBA agrees with the suggestion. Text amended Instrument 224 Instrument 224 – Swaption (+100bps), should reference instrument 219. It is not clear if the swaption should hedge the swap, therefore the respondent recommends restating the swap details. The EBA agrees with the suggestion. Text amended Instrument 311 Instrument 311 – EUR / BRL NDF, the instruction ‘cash settled’ should be deleted. The EBA agrees with the suggestion. Text amended Instruments 532, 533, 534 Instrument 532 – AT&T callable bond, USD 1 000 000 Instrument 533 – Bayer callable bond, EUR 1 000 000 Instrument 534 – AT&T callable bond, USD 1 000 000 The EBA agrees with the suggestion:
instrument 534 is in EUR.
Text amended
Instrument 602
Instrument 602 – iTraxx CDS, hedges instrument 601, not 6002. In addition, the respondent proposes:
CDS on iTraxx Europe index most recent on-the-run series.
The EBA agrees with the suggestion.
Text amended

FINAL DRAFT PAPER ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS 24 Comments Summary of responses received EBA analysis Amendments to the proposals Also, the Notional Instruments 604, 606, 608, 610 As suggested for 602. The EBA agrees with the suggestion. Text amended Instruments 605, 607 Instrument 605 – Short position in iTraxx Europe, contrary to how this was specified in the ISDA Phase 3 exercise, the respondents would now recommend an attachment point of 12%. Instrument 607 – Long position in iTraxx Europe with an attachment point of 12%. The EBA agrees with the suggestion. Text amended Question 5: Do you agree with the new portfolios added, according to section 3.3.3? Please indicate any issues, specified by portfolio, and provide an actual suggestion to clarify the potential misspecification in the portfolios’ composition (as provided in sections 3 and 4 of Annex 5 to these ITS). Portfolio 1013 Portfolio 1013 – Long 10 000 Bayer shares, short 1 000 Stellantis futures (1 000 shares). The issue should be fixed by the better specification of the equity instruments. No change in the portfolio definition Portfolio 2015 Portfolio 2015 – Swap USD 10m, US Govt bond USD 1m Is the mismatch in notional amounts designed to be as stated? The swap was reset to 1m notional to match the bond position. Instrument 203 was amended to 1m notional. Portfolio 2018 Portfolio 2018 – 209 - 10 instruments. The change made for the 2022 exercise to normalise all notional amounts such that all portfolios had just one of each instrument was a significant simplification and greatly reduced the chance of booking errors. The number of instruments (instrument 209) in portfolio 2018 was reset to one, to be consistent with the rest of the portfolios in section 3. The number of instruments (instrument 209) in portfolio 2019 was reset to one, to be consistent with the rest of the portfolios in section 3. The notional of Portfolios 2018 and 2019 and instrument 219 were amended, as provided in the final version of the ITS.

FINAL DRAFT PAPER ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS 25 Comments Summary of responses received EBA analysis Amendments to the proposals It would be good to keep this approach for the 2023 exercise. The respondent notes that instrument 209 is defined as EUR 1m so it is sized appropriately for portfolio 2016. However, to avoid the need to re-introduce instrument quantities all notional amounts could be set to 10 000 000 where appropriate. instrument 2019 was reset to 1m notional to match the bond position. Portfolio 2022 Portfolio 2022 – No instruments defined. Should this be the below:
Instr. 209 – Long Italy Govt
Instr. 222 – Long Italy Govt (HICP)
(tests FRTB-SA aggregation of GIRR rates & inflation) The empty row was a typo. Portfolio 2022 was removed, and portfolios 2023 and 2024 were renamed 2022 and 2023. Portfolio 3002 Portfolio 3002 – Long USD 1m cash, Long USD/EUR option EUR 10m Is the mismatch in notional amounts designed to be as stated? The mismatch was addressed, amending the instrument in the portfolio, matching to 10 m notional. Instrument 303 was reset to 10 million notional. Portfolio 6002, 6003 and 6004 Portfolio 6002 – Need to specify ‘Base currency is EUR’ Portfolio 6003 – Need to specify ‘Base currency is EUR’ Portfolio 6004 – Need to specify ‘Base currency is EUR’ Portfolio 6005 – Need to specify ‘Base currency is EUR’ The EBA agrees with the suggestion. Text amended

FINAL DRAFT PAPER ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS 26 Comments Summary of responses received EBA analysis Amendments to the proposals Portfolio 6007 Portfolio 6007 – Long FTD swap, should be 611, not
612. Furthermore, the respondent recommends the
CTP portfolios be hedged with CDS and not bonds.
Portfolio 6007 and instrument 611 were dropped from the exercise, in the final version of the ITS, since they are no longer needed for the purpose of the exercise. Text amended; portfolio removed Question 6: Do you agree with the current wording provided in the instruction in Annex 5, section 1, letter (aa)? Please provide any suggestion that would be required, in your opinion, to update the instruction. Instruction ‘aa’ and Instruments 204 and 220 Paragraph (aa) leaves it up to institutions to substitute the prescribed reference rate with the alternative rate. Instead, for 2023 the instrument descriptions in section 2 should be amended to explicitly specify the alternate rate where necessary:
Instrument 204 – GBP swap, replace LIBOR with
SONIA rate compounded and paid annually.
Instrument 220 – MtM reset xccy swap, the respondent suggests making this SOFR/ESTER as this is the new market standard, with adjusted details as below:
EUR: ESTER, pay quarterly compounded with a payment lag 2D USD: SOFR, receive quarterly compounded with a payment lag 2D Leg 1: USD Notional EUR 10 000 000 equivalent adjusted on a quarterly basis Leg 2: EUR Notional EUR 10 000 000 The EBA agree with the suggested changes in instruments 204 and 220. Instruction ‘aa’ was also amended accordingly. Instruction ‘aa’ in
section 1 and
instruments 204 and
220 were amended in the final ITS.

FINAL DRAFT PAPER ON AMENDING ITS ON BENCHMARKING OF INTERNAL MODELS 27 Comments Summary of responses received EBA analysis Amendments to the proposals Other observations Submission dates The IMV (and initial SBM) submission date has been brought forward from 15 October in the 2022 exercise to 4 October in the 2023 exercise. Likewise, the risk measure (and final SBM) submission date is brought forward from 4 March to 24 February. That means the time during which the banks may produce the submission and do the necessary validation to ensure correct data would become shorter. Considering the significant increase in the number of instruments and portfolios, the period between reference and submission dates should not be reduced. Also given the impending Basel IV and FRTB implementation, the EBA benchmarking exercise will be key, and care should be taken to make sure the outcome is not undermined by a shorter-than-necessary process. The EBA agrees to reset the submission dates as they were set for the 2022 exercise. Submission dates are postponed by two weeks with respect to the CP text, and in line with the 2022 final implementation exercise.
Annex 6 Templates 106.01
and 120.01
The respondent would like clarification on the guidance for the risk factor identifier in the reporting of templates 106.01 and 120.01. Specifically for CSR curvature, which requires specification of curvature by period applied (0.5 years – 30 years), which is a different approach to what is used for GIRR curvature. The EBA agrees with the suggestion. Text amended