Final Report on EMIR 3 RTS on model validations

9 October 2025 ESMA91-1505572268-4323 Final Report Draft RTS (under Article 49(5) of EMIR) on the conditions and the list of documents for an application for validation of changes to models and parameters under Articles 49 and 49a of EMIR

1 Table of Contents 1 Executive Summary ....................................................................................................3 2 Legislative references and abbreviations.....................................................................5 3 Introduction .................................................................................................................6 4 RTS on the determination of significance under Article 49(1i) of EMIR and on changes to models that can be considered as already covered by the approved model .................9 4.1 Introduction ..........................................................................................................9 4.2 Overall approach................................................................................................10 4.3 Condition (a).......................................................................................................13 4.4 Condition (c).......................................................................................................14 4.5 Condition (d).......................................................................................................14 4.6 Condition (e).......................................................................................................15 4.7 Condition (f)........................................................................................................17 4.8 Condition (g).......................................................................................................18 4.9 Condition (h).......................................................................................................19 4.10 Lookback period for Conditions (a), (c), (d), (e), (f), (g) and (h)...........................20 4.11 Condition (b).......................................................................................................23 4.12 Condition (i)........................................................................................................24 4.13 Changes to models that can be considered as already covered by the approved model 26 5 RTS on the list of required documents that are to accompany an application for validation of a change to models or parameters .............................................................30 5.1 Introduction/Background ....................................................................................30 5.2 Chapter 1 – General provisions..........................................................................30 5.3 Chapters 2 and 3 – Information to be provided for the validation of significant and non-significant changes ................................................................................................31 6 Annexes ....................................................................................................................37

2 6.1 Annex I - Legislative mandate to develop technical standards............................37 6.2 Annex II - Cost-benefit analysis..........................................................................38 6.2.1 Significant increase or decrease..................................................................38 6.2.2 Lookback period..........................................................................................39 6.2.3 Elements to be considered when assessing whether one of the conditions is met 39 6.3 Annex III - Advice of the Securities and Markets Stakeholder Group..................41 6.4 Annex IV - List of respondents ...........................................................................42 6.5 Annex V - Draft regulatory technical standards...................................................43

3 1 Executive Summary Reasons for publication Regulation (EU) No 2024/2987 of the European Parliament and of the Council of 27 November 2024 amending Regulations (EU) No 648/2012, (EU) No 575/2013 and (EU) 2017/1131 as regards measures to mitigate excessive exposures to third-country central counterparties and improve the efficiency of Union clearing markets (EMIR 3), which has amended Regulation (EU) 648/2012 of the European Parliament and of the Council of 4 July 2012 (EMIR), introduces several measures to make EU clearing services and EU CCPs more efficient and competitive, notably by streamlining and shortening supervisory procedures for validations of changes to models and parameters. Article 49(1) of EMIR now distinguishes between two approval procedures for a CCP intending to adopt a change to a model or parameters: for significant changes, a validation in accordance with Article 49 of EMIR is necessary, while for non-significant changes, the CCP can apply under a newly introduced accelerated procedure as set out under Article 49a of EMIR. ESMA is mandated under Article 49(5) of EMIR to develop draft Regulatory Technical Standards (RTS) further specifying what constitutes a significant increase or decrease for the purposes of Article 49(1i), points (a) and (c) to (h), of EMIR; the elements to be considered when assessing whether one of the conditions referred to in Article 49(1i) of EMIR is met, what constitute other changes that can be considered as already covered by the approved model and therefore not subject to the validation procedures under Articles 49 or 49a of EMIR, as well as the list of required documents and information that are to accompany an application for validation pursuant to Articles 49(1c) and 49a of EMIR. ESMA shall submit those draft RTS to the European Commission within 12 months from EMIR 3 entry into force, i.e. by 25 December 2025. ESMA conducted a public consultation on the draft RTS from 7 February 2025 until 7 April 2025, and received seven responses, one of which was confidential. The final draft RTS presented in this Final Report take into account the feedback received through the public consultation. In accordance with Article 49(5) of EMIR, the draft RTS have been prepared in close cooperation with the members of the ESCB. ESMA also sought the advice from the Securities and Markets Stakeholder Group (SMSG). Contents This Final Report presents the final draft RTS prepared by ESMA. Section 3 explains the background to ESMA’s proposals. Section 4 outlines ESMA’s proposal concerning the determination of significance under Article 49(1i) of EMIR and changes to models that can be considered as already covered by the approved model. Section 5 presents ESMA’s

4 proposal with regard to the list of required documents and information that are to accompany an application for validation of changes to models and parameters under Article 49 and Article 49a of EMIR. Finally, Section 6 contains all the relevant annexes (Annex I provides the legislative mandate for the development of the draft RTS; Annex II contains the cost-benefit analysis; Annex III includes the advice of the Securities and Markets Stakeholder Group; Annex IV provides the list of respondents to the public consultation; and Annex V contains the final draft RTS). Next Steps The Final Report, including the final draft RTS presented in Annex V, will be submitted to the European Commission. The European Commission has three months to decide whether to adopt the RTS in the form of a Commission Delegated Regulation. Following the adoption, they are then subject to non-objection by the European Parliament and the Council.

5 2 Legislative references and abbreviations CCP Central counterparty Delegated Regulation 153/2013 Commission Delegated Regulation (EU) No 153/2013 of 19 December 2012 supplementing Regulation (EU) No 648/2012 of the European Parliament and of the Council with regard to regulatory technical standards on requirements for central counterparties EMIR Regulation (EU) No 648/2012 of the European Parliament and of the Council of 4 July 2012 on OTC derivatives, central counterparties and trade repositories EMIR 3 Regulation (EU) No 2024/2987 of the European Parliament and of the Council of 27 November 2024 amending Regulations (EU) No 648/2012, (EU) No 575/2013 and (EU) 2017/1131 as regards measures to mitigate excessive exposures to third-country central counterparties and improve the efficiency of Union clearing markets ESMA Regulation Regulation (EU) No 1095/2010 of the European Parliament and of the Council of 24 November 2010 establishing a European Supervisory Authority (European Securities and Markets Authority), amending Decision No 716/2009/EC and repealing Commission Decision 2009/77/EC ITS Implementing Technical Standards NCA National competent authority RTS Regulatory Technical Standards

6 3 Introduction

1. Regulation (EU) 2024/2987 of the European Parliament and of the Council of 27 November 2024 1 (EMIR 3), which has amended Regulation (EU) 648/2012 of the European Parliament and of the Council of 4 July 20122 (EMIR), was published in the Official Journal of the EU on 4 December 2024 and entered into force on 24 December
2. It introduces several measures to make EU clearing services and EU CCPs more efficient and competitive, notably by streamlining and shortening supervisory procedures for validations of changes to models and parameters: An obligation for EU CCPs to submit all applications for validations of changes to models and parameters (Articles 49 and 49a of EMIR) via a CCP Central database set up and maintained by ESMA pursuant to Article 17c of EMIR.

• The introduction of revised and shortened timelines and procedures for these applications.
• The distinction between a normal procedure (for significant changes), an accelerated procedure (for non-significant changes), and changes that are considered part of a validated model and do not require a validation.

2. The revised Article 49(1) of EMIR now distinguishes between two approval procedures for a CCP intending to adopt a change to a model or a parameter: Where the CCP considers that the intended change is significant, pursuant to Article 49(1i) of EMIR, it is required to apply for validation in accordance with the Article 49 procedure. Where the CCP considers that the intended change is not significant, pursuant to Article 49(1i) of EMIR, then it is required to apply for validation in accordance with the newly introduced accelerated procedure, set out under Article 49a of EMIR.
3. The criteria for changes to models or parameters to be considered as significant are set out under paragraph 1i of Article 49 of EMIR.
4. In addition, Article 49(1h) of EMIR provides that changes to parameters, that are the result of applying a methodology that is part of a validated model, either due to external input or due to a regular review or calibration exercise, are not to be considered changes to models and parameters for the purpose of Article 49 and Article 49a of EMIR, and, therefore, such changes do not require a validation under Article 49 or Article 49a of EMIR. 1 Regulation (EU) 2024/2987 of the European Parliament and of the Council of 27 November 2024 amending Regulations (EU) No 648/2012, (EU) No 575/2013 and (EU) 2017/1131 as regards measures to mitigate excessive exposures to third-country central counterparties and improve the efficiency of Union clearing markets; OJ L, 2024/2987, 4.12.2024. 2 Regulation (EU) No 648/2012 of the European Parliament and of the Council of 4 July 2012 on OTC derivatives, central counterparties and trade repositories; OJ L 201, 27.7.2012, p. 1–59.

7 5. In this context, and in accordance with Article 49(5) of EMIR, ESMA has been empowered to develop, in close cooperation with the members of the ESCB, draft Regulatory Technical Standards (RTS) further specifying all of the following: what constitutes a significant increase or decrease for the purposes of Article 49(1i), points (a), (c), (d), (e), (f), (g) and (h), of EMIR;

• the elements to be considered when assessing whether one of the conditions referred to in Article 49(1i) of EMIR are met;
• other changes to models that can be considered as already covered by the approved model and are therefore not considered a model change and not subject to validation under Articles 49 or 49a of EMIR; the list of required documents and information that are to accompany an application for validation pursuant to Articles 49(1a) and 49a of EMIR. Article 49 Review of models, stress testing and back testing

5. ESMA, in close cooperation with the members of the ESCB, shall develop draft regulatory technical standards to further specify; (a) what constitutes a significant increase or decrease for the purposes of paragraph 1i, points (a) and (c) to (h); (b) the elements to be considered when assessing whether one of the conditions referred to in paragraph 1i is met; (c) other changes to models that can be considered as already covered by the approved model and are therefore not considered a model change and not subject to the procedures established in this Article or Article 49a; and (d) the lists of required documents that are to accompany an application for validation pursuant to paragraph 1c of this Article and Article 49a and the information that such documents are to contain to demonstrate that the CCP complies with all relevant requirements of this Regulation. The required documents and level of information shall be proportionate to the type of model validation but contain sufficient detail to ensure a proper analysis of the change. For the purposes of the first subparagraph, point (a), ESMA may set different values for the different points of paragraph 1i. ESMA shall submit the draft regulatory technical standards referred to in the first subparagraph to the Commission by 25 December 2025. Power is delegated to the Commission to supplement this Regulation by adopting the regulatory technical standards referred to in the first subparagraph of this paragraph in accordance with Articles 10 to 14 of Regulation (EU) No 1095/2010.

8 6. Having noted that all these mandates relate to content or conditions for the applications for validations of changes to models and parameters, ESMA has deemed relevant to group such mandates under a single consultation paper and final report (and under a single set of draft RTS). 7. For the avoidance of doubt, ESMA is also empowered under Article 49(6) of EMIR to develop draft Implementing Technical Standards (ITS) specifying the electronic format of the application to be submitted to the central database referred to in Article 17c of EMIR for the validation of changes under Articles 49 and 49a of EMIR. However, this mandate will be addressed in a separate Consultation Paper/Final Report.

9 4 RTS on the determination of significance under Article 49(1i) of EMIR and on changes to models that can be considered as already covered by the approved model 4.1 Introduction 8. The revised Article 49(1) of EMIR now distinguishes between two approval procedures for a CCP intending to adopt a change to a model or a parameter: Where the CCP considers that the intended change is significant, pursuant to Article 49(1i) of EMIR, it is required to apply for validation in accordance with the Article 49 procedure. Where the CCP considers that the intended change is not significant, pursuant to Article 49(1i) of EMIR, then it is required to apply for validation in accordance with the newly introduced accelerated procedure, set out under Article 49a of EMIR. 9. In accordance with Article 49(5) of EMIR, ESMA is tasked to further specify in a draft RTS: what constitutes a significant increase or decrease for all relevant conditions set out in Article 49(1i) of EMIR (i.e. conditions (a), (c), (d), (e), (f), (g) and (h)); and

• which elements to consider when assessing whether one of the conditions referred to in Article 49(1i) of EMIR are met.

10. In other words, the RTS should specify numerical thresholds for each of the following conditions:

• Condition (a): a significant decrease or increase of the CCP’s total pre-funded financial resources, including margin requirements, default fund and (first) skin in the game.
• Condition (c): a significant decrease or increase of the output of the margin model at the CCP level following the introduction, removal or amendment of a component of the CCP’s margin model (including a margin parameter or an add-on).
• Condition (d): a significant decrease or increase of the total margin requirements for a set of financial instruments subject to portfolio margining following a change in the methodology used to compute portfolio offsets for these financial instruments.
• Condition (e): a significant decrease or increase in the size of any of the default funds or of any individual default fund contribution following a change in the methodology for defining and calibrating stress test scenarios for the purpose of determining the size of the CCP's default funds and the size of the individual clearing members' contributions to those default funds.

10

• Condition (f): a significant decrease or increase of the estimated liquidity needs in any currency or the total liquidity needs following a change to the methodology applied to assess liquidity risk.
• Condition (g): a significant decrease or increase in the CCP’s overall exposure to an individual counterparty following a change to the methodology applied to determine the concentration risk a CCP has towards this counterparty.
• Condition (h): a significant decrease or increase of the total value of collateral following a change to the methodology applied to value collateral or calibrate collateral haircuts.

11. As regards “the elements to be considered when assessing whether one of the conditions referred to in paragraph 1i are met”, ESMA notes that the conditions set out under points (a), (c), (d), (e), (f), (g) and (h) are clearly defined in Article 49(1i) of EMIR and linked to numerical thresholds further specified in the RTS. Therefore, in ESMA’s view, “the elements to consider” should focus in particular on further specifying conditions (b) and (i), namely:

• Condition (b): What would qualify as a change to the structure or structural elements of the margin model;
• Condition (i): Which type of change could have a material effect on the overall risk of the CCP.

12. Furthermore, in accordance with point (c) of Article 49(5) of EMIR, ESMA, in close cooperation with the members of the ESCB, is also empowered to develop draft RTS further specifying other changes to models that can be considered as already covered by the approved model and are therefore not considered a model change and not subject to the procedures established in Article 49 or Article 49a of EMIR.
13. ESMA notes that the empowerment under Article 49(5)(c) is in addition to Article 49(1h) of EMIR which already specifically provides that changes to parameters “that are the result of applying a methodology that is part of a validated model” are not to be considered changes to models and parameters and should therefore not be subject to a validation under Articles 49 or Article 49a of EMIR. 4.2 Overall approach ESMA initial proposal
14. As regards the numerical thresholds for conditions (a), (c), (d), (e), (f), (g) and (h), ESMA noted that the empowerment allows ESMA to set different values for the different points of Article 49(1i) of EMIR, therefore in principle the RTS should include at least one quantitative threshold per condition. However, for some conditions (e.g. impact on liquidity per currency vs. total liquidity (condition f); impact on individual default fund contributions

11 vs. total default fund (condition e)), ESMA proposed (in the Consultation Paper) different values for different metrics. 15. The RTS should also specify whether the quantitative thresholds should be defined in proportion (% of increase / decrease) or in absolute. ESMA noted that specifying percentage thresholds would ensure a proportional application of the conditions across all CCPs, and therefore decided to rely, in its proposal in the Consultation Paper, on percentage values. However, in order to ensure proportionality and avoid any ‘false positives’, where the percentages refer to shares of the CCP's resources (such as individual default fund contributions), ESMA also proposed (in the Consultation Paper) double thresholds, one on the individual level and one on the overall level. 16. ESMA, as well as NCAs, have been conducting validations of CCP’s models and parameters since the entry into effect of the initial EMIR, i.e. for over 10 years. Therefore, in setting the values for the thresholds (as proposed in the Consultation Paper), ESMA relied on that experience. 17. In order for a CCP to assess whether conditions (a), (c), (d), (e), (f), (g) and (h) are met on a transparent and comparable basis, and to ensure convergence across EU CCPs, in the Consultation Paper, ESMA also proposed a minimum lookback period over which the CCP should compute the impact of the change on actual cleared portfolios. 18. As regards “the elements to be assessed”, ESMA proposed (in the Consultation Paper) an exhaustive list of qualitative elements for conditions (b) and (i). Furthermore, ESMA noted that conditions (b) and (i) should act as “catch-all” conditions which could capture significant changes which have not triggered any of the numerical thresholds of conditions (a), (c), (d), (e), (f), (g) and (h). Therefore, in ESMA’s view, the risk of relying on a closed list was therefore limited. 19. In addition, ESMA (in the Consultation Paper) proposed that changes that result from the implementation of a specific prescriptive recommendation from ESMA, or the NCA in a previous validation decision under Article 49 or 49a of EMIR should not be subject to the procedures under Article 49 or 49a of EMIR. Feedback from respondents 20. ESMA received seven responses to the consultation (one of which was confidential), as follows: One response from an association representing CCPs;

• Five responses from CCPs (one of which was confidential);
• One response from an association representing clearing members and clients.

21. While the respondents that are CCPs or representing CCPs broadly agreed with the approach, they found some elements too prescriptive. These respondents were of the view that the quantitative thresholds were too low and definitions too narrow. According to

12 these respondents, most model changes would be considered significant thus requiring regular approval procedures, increasing costs and regulatory burden, and negatively impacting risk incentives. These respondents instead argued for: Allowing more non-significant changes to benefit from accelerated procedures (i.e. by increasing the significance thresholds); and

• Broadening the scope of “business as usual” (“BaU”) changes.

22. One respondent that is a CCP stated that in the past three years, there were three applications by the CCP that would have fallen under the standard procedure and three applications that would have fallen under the accelerated procedure under the EMIR 3 framework. Whereas under the EMIR 2.2 framework, only three applications fell under the standard procedure and the rest were not subject to any procedures.
23. As regards the other respondents that are CCPs or representing CCPs, while not providing any quantitative evidence/simulations, these respondents argued that virtually all changes to models and parameters, even if immaterial, would be subject to the Articles 49/49a procedures under the EMIR 3 framework. Specifically, they expect that a larger number of changes will be classified as significant, as it is likely that at least one of the significance criteria will be triggered. Moreover, a number of changes that were “exempted” under the EMIR 2.2 framework are likely to be part of the accelerated process.
24. On the other hand, the respondent representing clearing members/clients found some of the thresholds too high, and argued that CCPs should have safe, prudent, and conservative risk management frameworks and that it was also necessary to ensure sufficient supervision and scrutiny over changes to CCPs’ risk management frameworks.
25. Some of the respondents that are CCPs or representing CCPs also argued that where a CCP’s change triggers both conditions for a validation of model changes and conditions for an extension of authorisation, the CCP should not be required to submit two applications in respect of the same change. Those respondents would therefore welcome a clarification that the CCP can combine the application procedures and information to be provided such that it is not duplicative. ESMA assessment of feedback
26. ESMA has taken note of the different positions in relation to the overall approach.
27. ESMA notes that a number of respondents asked to revise the approach in order to allow more non-significant changes to benefit from the accelerated procedure, and to broaden the scope of what they consider “BaU” changes.
28. Regarding the first point, ESMA has taken note of the concerns expressed, and in order to reduce the burden on CCPs, ESMA suggests increasing most of the proposed quantitative thresholds under Articles 1 to 7 of the draft RTS to ease the conditions under which a change would be deemed as significant and to align the thresholds with the

13 proposals suggested by the respondents to the consultation (see relevant sections of this Report). 29. Regarding the second argument, however, ESMA notes that, unlike in respect of extensions of services or activities referred to under Article 15a of EMIR, the Level 1 (Article 49 of EMIR) does not explicitly refer to “BaU” changes or “exempted changes” for model changes, but to “changes not considered model changes” and thus “not subject to the procedures established in Article 49 or Article 49a of EMIR” (as per Articles 49(1h) and 49(5)(c) of EMIR). 30. Nevertheless, in order to limit the burden for CCPs and avoid that an unnecessarily high number of changes have to go through the accelerated procedure, ESMA suggests introducing a different set of quantitative thresholds, mirroring the conditions used for the purpose of determining significance, but with lower values. Where a change is below all the “low” thresholds, it would be considered as already covered by the approved model and exempted from the procedures under Articles 49 and 49a of EMIR (see more details under the relevant section). 31. ESMA also notes that some respondents have asked for more flexibility in applying the conditions and in relation to the procedures. However, ESMA would like to reiterate that the conditions as well as the procedures themselves are set out directly in Level 1 and ESMA is obliged to stay within the remit of the Level 1 mandate and ensure supervisory convergence, and therefore, it is of the view that the approach set out in this Final Report provides the right balance. 4.3 Condition (a) ESMA initial proposal 32. As regards condition (a), in the Consultation Paper, ESMA proposed that any decrease or increase of more than 15% of the CCP’s total prefunded financial resources should be qualified as significant. Feedback from respondents 33. The majority of respondents argued that the proposed threshold of 15% is too low and would capture non-material changes as significant, leading to increased regulatory burden, unnecessary Article 49 applications, and potential delays or avoidance in implementing risk improvements. Instead, these respondents proposed to increase this threshold to 20%. ESMA assessment of feedback 34. ESMA has taken note of the concerns expressed in relation to the proposed 15% threshold for condition (a) and suggests increasing this threshold to 20% to align it with the

14 suggestions expressed by respondents to the consultation and to avoid capturing non￾material changes. 4.4 Condition (c) ESMA initial proposal 35. As regards condition (c), in the Consultation Paper, ESMA proposed that any decrease or increase of more than 15% of the total margin requirements for a specific clearing service / default fund should be considered as significant. 36. ESMA also noted that a margin model should be understood as the complex of techniques used to compute the total margin requirements for a clearing service / default fund, and, therefore, explained that the CCP should measure the impact on the total margin requirement of the clearing service / default fund, as other margin models are not affected by the same change. Feedback from respondents 37. The majority of respondents argued that the proposed threshold of 15% is too low and would capture non-material changes as significant thus leading to extensive procedures for minor changes without meaningful risk benefit. Instead, these respondents proposed to increase this threshold to 20%. 38. In addition, two respondents argued that the threshold should be set at the output of the margin model at the CCP level instead of the clearing service/default fund level proposed by ESMA. In addition, one respondent argued that it should be set at total prefunded resources for a specific clearing service/default fund. ESMA assessment of feedback 39. ESMA has taken note of the different positions in relation to condition (c). ESMA suggests keeping the calculation at the clearing service / default fund level to ensure comparability across CCPs. Nevertheless, ESMA suggests increasing the threshold to 20% to avoid capturing non-material changes and to align the threshold with the views expressed by respondents to the consultation. 4.5 Condition (d) ESMA initial proposal 40. As regards condition (d), in the Consultation Paper, ESMA proposed that any decrease or increase of more than 20% of the total margin requirements for the financial instruments subject to portfolio offsets should be considered as significant. Feedback from respondents 41. The majority of respondents agreed with the proposed threshold of 20%.

15 42. However, one respondent stated that it is not clear what is meant under the term “portfolio” upon which the threshold is supposed to be set, and one respondent argued that the threshold should be set at total prefunded resources for a specific clearing service / default fund. ESMA assessment of feedback 43. ESMA suggests leaving the proposed 20% threshold for condition (d) unchanged, with a small editorial drafting to further specify the scope of the portfolio offsets, which, in any case, is aligned with the wording of the Level 1 text. 4.6 Condition (e) ESMA initial proposal 44. As regards condition (e), in the Consultation Paper, ESMA explained that it considered that three quantitative thresholds for defining the significance of a change should be specified, namely: i) a threshold on the decrease or increase in the size of any of the default funds in respect of CCPs that have multiple default funds; ii) a threshold on the decrease or increase in the size of the relevant liquidation group or segment within the default fund in respect of CCPs that have a single default fund; and iii) a threshold on the impact on individual contributions. 45. ESMA therefore proposed that the following would constitute a significant increase or decrease for the purpose of condition (e): a. As regards a CCP that has multiple default funds, where the change in the methodology for defining and calibrating stress test scenarios results in a decrease or increase of more than 20% in the size of any of the default funds; or b. As regards a CCP that has a single default fund, where the change in the methodology for defining and calibrating stress test scenarios results in a decrease or increase of more than 20% in the size of the relevant liquidation group or segment of the default fund; or c. Where the change in the methodology for defining and calibrating stress test scenarios results in one clearing member’s contribution to a default fund decreasing or increasing by an amount corresponding to both: i. more than 30% of the size of the clearing member’s contribution, and ii. more than 5% of the size of the default fund. Feedback from respondents 46. The majority of respondents expressed concern that the proposed thresholds are too low and may trigger unnecessary regulatory procedures under Article 49 of EMIR, without improving risk management or resilience.

16 47. As regards point (a) of condition (e), some respondents proposed to increase the threshold to 20%, and one respondent argued that the threshold should be set at total prefunded resources for a specific clearing service / default fund, with the threshold set at 15%, as these are the total resources that would be available to cover the risk of the stress testing scenarios. 48. As regards point (b) of condition (e), some respondents disagreed with differentiating between CCPs with a single default fund vs. CCPs with multiple default funds, arguing that using liquidation group thresholds creates an unlevel playing field. These respondents, therefore, proposed to delete point (b) of condition (e). 49. As regards point (c) of condition (e), some respondents argued against setting thresholds at any individual clearing member’s default fund contribution level. Some respondents instead proposed: • Increasing the threshold in (c)(i) to 50%, and in (c)(ii) to 15%;

• That the condition should be considered as met only if triggered by several clearing members or apply only to the top tier/largest clearing members;
• Adding a third “absolute” threshold related to the CCP’s own resources in the default waterfall.

50. The majority of respondents opposed differentiating between CCPs with a single default fund vs. CCPs with multiple default funds for the purpose of point (c) of condition (e). However, some respondents argued that the condition should apply to the resources available to cover the specific stress scenario subject to change by the CCP, and that if losses are only absorbed by specific funds, only the affected fund should be counted, to prevent CCPs from structuring default funds purely to avoid triggering the thresholds. ESMA assessment of feedback
51. ESMA has taken note of the different positions in relation to condition (e).
52. In relation to the distinction between CCPs with a single default fund and CCPs with multiple default funds (points (a) and (b)) for the purpose of this condition, ESMA is of the view that this approach is the best-suited to avoid any unlevel playing field between CCPs with different waterfall structures and to ensure equal treatment across EU CCPs, and therefore proposes to keep these points unchanged.
53. Regarding the proposal to set thresholds at any individual clearing member (point (c)), despite some concerns expressed by respondents, ESMA notes that the requirement under Article 49(1i)(e) is clearly referring to “the size of the individual clearing member’s contributions”, hence the introduction of these individual thresholds is not optional, but required by law.
54. However, in having noted the concerns expressed in relation to point (c), including the risk of triggering an unnecessary high number of significant changes, in particular as regards

17 small default funds with very few clearing members, ESMA has decided to revise this threshold as follows (it should also be noted that the three conditions under point (c) are cumulative):

• Maintaining the threshold on the size of the clearing member’s contribution at 30%;
• Increasing the threshold on the size of the default fund from 5% to 10%; Introducing an additional threshold based on the size of the CCP’s skin in the game. 4.7 Condition (f) ESMA initial proposal

55. As regards condition (f), in the Consultation Paper, ESMA proposed introducing two quantitative thresholds for defining the significance of a change, namely a threshold on the total liquidity needs, and a threshold on liquidity per currency. The following was therefore proposed as constituting a significant increase or decrease for the purpose of condition (f): a. Where the change leads to a decrease or increase of more than 15% of the CCP’s total liquidity needs across all currencies; or b. Where the change results in the liquidity needs in a single currency decreasing or increasing by an amount corresponding to both: i. more than 50% of the liquidity needs in the single currency, and ii. more than 5% of the CCP’s total liquidity needs across all currencies. Feedback from respondents
56. As regards point (a) of condition (f), the majority of respondents did not provide any feedback; while one respondent supported the proposed threshold of 15%.
57. As regards point (b) of condition (f), all respondents argued that the proposed thresholds are too low, and that minor changes (in particular for minor currencies) would trigger the full procedure under Article 49 of EMIR. These respondents proposed either applying point (b) of condition (f) only to major currencies or those deemed material by the CCP's risk framework; or

• removing the threshold in (b)(i) altogether; or applying the threshold in (b)(i) to several (two or three) currencies; or
• increasing the threshold in (b)(ii) to 10% - 20%. ESMA assessment of feedback

18 58. ESMA has taken note of the different positions in relation to condition (f). In relation to point (a), ESMA proposes to increase the threshold to 20%, in line with the other quantitative conditions. 59. In relation to point (b), ESMA finds that it would go against the mandate set out in Level 1 to limit the increase/decrease only to several currencies and/or to major currencies, as Article 49(1i)(f) of EMIR refers to “liquidity needs in any currency”. 60. Nevertheless, having noted the concerns regarding the proposed threshold, and in particular the risk of capturing a high number of changes or changes in respect of minor currencies, and in order to align it with the views expressed in the consultation, ESMA suggests increasing the threshold in point (b)(ii) from 5% to 10%. 4.8 Condition (g) ESMA initial proposal 61. As regards condition (g), in the Consultation Paper, ESMA noted that it found that the scope of condition (g) was not entirely clear, as it refers first to the methodology applied to determine the concentration risk of a CCP towards an individual counterparty and then to its effect on overall exposure to that counterparty. 62. However, ESMA explained that in its view this condition should capture changes to concentration limits for any entity towards which the CCP has a liquidity exposure including the entities listed in Article 32(4) of Delegated Regulation 153/2013. Such changes may generate changes in exposures to specific counterparties, which would also be reflected in liquidity resources. 63. Considering the above, in the Consultation Paper, ESMA proposed that the following would constitute a significant increase or decrease for the purpose of condition (g): Where the change results in the CCP’s liquidity exposure towards an individual counterparty, including the entities listed in Article 32(4) of Delegated Regulation 153/2013, decreasing or increasing by an amount corresponding to both: a. more than 20% of the CCP’s liquidity exposure to the individual counterparty, and b. more than 5% of the aggregate liquid resources of the CCP. Feedback from respondents 64. The majority of respondents argued that the proposed thresholds are too low, which would result in unnecessary regulatory burdens. 65. The respondents proposed:

19

• Removing the threshold under point (a) as there is already a threshold referring to the CCP liquidity under condition (f), and condition (g) refers to the overall exposure and not any exposure as it is the case in other conditions; or
• Increasing the threshold under point (b) to 15%; or
• Removing the threshold under point (b) to reduce disproportionate impact from outlier counterparties.

66. In addition, some respondents argued that there is some ambiguity with regards to the scope and category of changes which may be relevant for monitoring this condition, e.g. margin model change would not be relevant for this condition.
67. One respondent argued that condition (g) instead refers to the concentration add-ons used by many CCPs to account for concentrated positions, where margin increases if a clearing member presents portfolios with a high concentration in certain products. This respondent thus suggested focusing the condition on changes in concentration add-ons, with a proposed threshold of 15%. ESMA assessment of feedback
68. ESMA has taken note of the different positions in relation to condition (g). While noting the relative ambiguity of the drafting under Article 49(1i)(g) of EMIR, ESMA remains of the view that such condition captures changes to concentration limits for any entity towards which the CCP has a liquidity exposure.
69. Nevertheless, in order to reduce the impact of this threshold and the burden on CCPs, and align it with the views expressed in the consultation, ESMA suggests increasing the threshold on aggregate liquid resources from 5% to 10%. 4.9 Condition (h) ESMA initial proposal
70. As regards condition (h), in the Consultation Paper, ESMA proposed that any decrease or increase of more than 5% of the CCP’s total collateral value should be considered as significant. Feedback from respondents
71. All respondents argued that the proposed threshold is too low. Instead, they proposed to increase the threshold to 10% - 20%, depending on interpretation/scope of the condition.
72. As regards the interpretation/scope of the condition, some respondents asked ESMA to clarify that: The condition only applies to a change to the methodology applied to value collateral or calibrate collateral haircuts;

20 The change to the total value of collateral refers to the “total value of collateral after applying haircuts”;

• The condition does not apply to a change in the value of collateral that results from a change to total margin requirements and/or default fund contributions. ESMA assessment of feedback

73. ESMA has taken note of the different concerns expressed in relation to condition (h). However, ESMA finds it important to clarify that, in accordance with Level 1, this criterion would only be triggered in the context of a change to the methodology applied to value collateral or calibrate collateral haircuts, that would result in an increase/ decrease of more than 5% of the total value of collateral. Therefore, ESMA maintains that 5% seems to be adequate considering the very targeted changes, and the typical scale of the impact of valuation and haircut models on collateral value.
74. However, ESMA suggests slightly revising the wording to clarify (in accordance with the wording in Level 1) how the criterion applies. 4.10Lookback period for Conditions (a), (c), (d), (e), (f), (g) and (h) ESMA initial proposal
75. In the Consultation Paper, ESMA explained that for a CCP to assess whether conditions (a), (c), (d), (e), (f), (g) and (h) are met on a transparent and comparable basis, and to ensure convergence across EU CCPs, the RTS should also specify a minimum lookback period over which the CCP should compute the impact of the change on actual cleared portfolios. ESMA further noted that the lookback period should be sufficiently short and close to the planned implementation date of the change, but long enough to avoid any window-dressing from the CCP.
76. Therefore, in the Consultation Paper, ESMA proposed that a CCP should assess the impact of each change, comparing the outcome of current methodology with the outcome of the proposed methodology, using a continuous period of at least six months within the 12 months terminating at the application date. A condition would be considered to be met if the corresponding threshold is triggered at any time in the period. Feedback from respondents
77. All respondents raised some concerns regarding the proposed lookback period.
78. The majority of respondents argued that the proposed continuous production parallel simulation is resource-intensive, costly, and difficult to implement, especially for immaterial changes, and that simulations over entire risk frameworks would require major IT infrastructure upgrades and extend regulatory timelines. Additionally, some respondents argued that six months of continuous data for assessing the significance criteria may not

21 be available specially for model changes related to a new product with limited data. These respondents proposed instead the following options:

• A six-month lookback period within the last 24 months;
• Performing an assessment over a period of six months with at least six/10 observations that the CCP considers relevant within the last 24 months, including a requirement for the CCP to justify that the chosen dates are representative;
• Performing an assessment during dates at regular intervals (i.e. always the same amount of time between two observations) and for the CCP to explain in case it deviates from these frequencies;
• Performing an assessment on several representative recent dates as well as on relevant stressed dates; these dates could be adapted to the relevance of specific events or days for specific asset classes; including a requirement for the CCP to justify that the chosen dates are representative;
• Allowing qualitative assessments for immaterial changes.

79. One respondent argued that while using six out of 12 months may be acceptable and coherent for most segments, certain instruments would require a different lookback period to enable adequate assessment of the stated conditions to capture the seasonal effect for any asset class with such features (e.g. soft commodities and Milling Wheat, which have significant deliveries in both September and December each year), and a six-month period might lead a CCP to selectively avoid these two critical months. The respondent proposed instead a “consecutive nine months out of 18 months period” or “consecutive 12 months out of 18 months period”.
80. One respondent also argued that the proposal gives CCPs too much flexibility, and instead proposed that ESMA should clearly define the exact lookback period, rather than allowing CCPs to pick a benign one.
81. The majority of respondents also expressed concerns regarding the requirement to consider a change significant if any condition is met at any point in time in the period, as it may capture outliers or ‘false positives’. Instead, these respondents proposed:

• threshold-based or “majority of times” criteria; Replacing “any point in time” with “majority of times”; Allowing a certain a number of breaches;
• Allowing 99% thresholds or risk appetite-based exceptions.

82. The majority of respondents also argued that the assessment should focus only on the affected model components, as follows:

22

• For changes to margin models: conditions laid down in Article 2 and Article 3 of the draft RTS (as proposed in the Consultation Paper) would need to be analysed;
• For changes to stress testing models: conditions laid down in Article 1 and Article 4 of the draft RTS (as proposed in the Consultation Paper) would need to be analysed;
• For changes to liquidity stress testing models: conditions laid down in Article 5 and Article 6 of the draft RTS (as proposed in the Consultation Paper) would need to be analysed.
• For changes to collateral haircut models: conditions laid down in Article 7 of the draft RTS (as proposed in the Consultation Paper) would need to be analysed.

83. Some respondents also suggested aligning the lookback period with the period for credit and liquidity stress testing results documentation as required by Article 17 of the draft RTS (as proposed in the Consultation Paper). ESMA assessment of feedback
84. ESMA has taken note of the different positions expressed in relation to the proposed lookback period.
85. In order to reduce the operational burden for CCPs in preparing their applications, ESMA suggests reducing the requirements on the lookback period to a set of representative dates during the lookback period.
86. As a result, ESMA suggests:

• Extending the lookback period requirement to 12 months over the last 24 months; and Requiring the significance assessment to be performed on 12 dates, each representing one month, spaced at regular intervals, spanning this period.

87. In addition, ESMA suggests clarifying that for a given change, only the relevant conditions should be tested, as set out in the following table: Art. 1 Art. 2 Art. 3 Art. 4 Art. 5 Art. 6 Art. 7 Art. 8 Art. 9 CCP’s total prefunded financial resources Output of margin model Total margin req Size of DF or indiv contribution Liquidity needs Exposure to cpty Total value of collateral Structural elements Overall risk of CCP Margin models x ST models x LST models x Collat haircut/valuat ion models x

23 4.11Condition (b) ESMA initial proposal 88. As regards condition (b), in the Consultation Paper, ESMA proposed relying on an exhaustive list of changes/elements that should be considered as a change to the structure or structural elements of a CCP’s margin model: a. Implementation of a new type of model including moving from/to a parametric model, historical simulation and Monte Carlo simulation; b. Changes to the model’s lookback window, including the introduction of a filtering scheme where one was not previously used; c. Changes to the model’s confidence level or chosen tail statistic, such as percentile or expected shortfall; d. Changes to number of days used for the margin period of risk; and e. Changes to the choice of APC option (e.g. changing from option (a) to option (b) under Article 28(1) of Delegated Regulation 153/2013). 89. Furthermore, ESMA proposed that when the CCP’s proposed change fulfils at least one of the elements, condition (b) would be considered as met. Feedback from respondents 90. The majority of respondents agreed that point (a) (i.e. implementation of a new type of model) is a structural element of a CCP’s margin model and should therefore be included under condition (b). 91. Some respondents also proposed to include “(volatility) filtering scheme” under point (a) (or to include it as a separate point) as supposed to under point (b) as initially proposed by ESMA. 92. Most respondents argued that point (b) (lookback window), point (c) (confidence level) and point (d) (margin period of risk, MPOR) are parameters and not structural elements of a CCP’s model, and should therefore not be included under condition (b). 93. As regards point (e) (APC option), while most respondents agreed that changes to anti￾procyclicality options represent a structural change to the margin model, they did not believe it should be included under condition (b), i.e. that it should not trigger an Article 49 procedure. 94. Furthermore, most respondents argued that, in general, changes within scope of EMIR (e.g. point (e)) should not require a validation under Article 49 of EMIR. 95. One respondent agreed that all points (i.e. points (a) to (e)), as proposed by ESMA, should be included under condition (b).

24 ESMA assessment of feedback 96. ESMA has taken note of the contrasted feedback in relation to condition (b). 97. First, ESMA has noted some broad support for points (a) and (e) (now points (a) and (d)), and suggests leaving them unchanged. 98. With regard to point (b), having taken note of the feedback received, ESMA proposes to delete this point, and instead add the introduction of ‘filtered historical simulation’ under point (a) as suggested by several respondents. 99. In relation to points (c) and (d) (now points (b) and (c)), i.e. the changes to the “core parameters” of the CCP’s margin model, ESMA disagrees with the argument that any change to these parameters, if compliant with the requirements set out under EMIR and the relevant RTS, should be deemed as not material, as compliance with EMIR is always the starting point. 100. However, in order to limit the burden on CCPs, ESMA suggests revising the draft RTS to consider changes in points (c) and (d) (now points (b) and (c)) that by construction make the model more conservative (e.g. increase of the confidence level, moving from percentile to expected shortfall and increase of the MPOR) as already covered by the approved model. 101. Additionally, changes in relation to points (c) and (d) (now points (b) and (c)), which consist of adjustments to the confidence interval and/or the length of the liquidation period where the methodology in place at the CCP (which has already been validated by the CCP’s NCA and ESMA) requires such adjustments with defined rules and based on predetermined, objective criteria, should be considered as already covered by the approved model. 4.12Condition (i) ESMA initial proposal 102. As regards condition (i), in the Consultation Paper, ESMA proposed that the elements to be considered when assessing whether condition (i) is met should be any of the following: a. The CCP intends to change its risk models in order to accommodate a new type of participant or a new clearing access model, including a sponsored access model; and b. The CCP intends to change its risk models in order to extend the list of eligible collateral to accept collateral: i. in a new asset class; or ii. in the form of gold; or

25 iii. in the form of public guarantees, or public bank guarantees, or commercial bank guarantees; or iv. issued by corporate issuers where the CCP currently only accepts collateral issued by sovereign issuers; or v. issued by sovereign issuers where the CCPs currently only accepts collateral issued by corporate issuers. 103. ESMA further proposed that when the CCP’s proposed change fulfils at least one of the elements, condition (i) would be considered as met. Feedback from respondents 104. The majority of respondents did not agree with point (a) as they argued that the criterion on “a new type of participant or a new clearing access model” is better suited for the extension of authorisation procedure rather than a validation under Article 49 of EMIR. 105. Most respondents also disagreed with point (b), in particular with sub-point (v). 106. In addition, some participants argued that it is not clear what is meant by “a new type of participant” or “new asset class”. 107. One respondent agreed that both points (a) and (b), as proposed by ESMA, should be included under condition (i). 108. The respondents did not provide any alternatives as regards the further specification of condition (i), with the exception of a proposed rewording of point (b), if that point was to be kept, as follows: “The CCP intends to change its risk models related to collateral, i.e. collateral haircut and collateral valuation models, in order to accommodate an extension of the list of eligible collateral.” ESMA assessment of feedback 109. ESMA has taken note of the different concerns in relation to condition (i). 110. Regarding point (a), ESMA suggests revising the RTS to limit the criterion to a change in the CCP’s risk model in order to accommodate a new clearing access model (which modifies the CCP’s exposure towards participants), but not a new type of participant (which may be better suited for extensions of authorisation). 111. Regarding point (b), ESMA suggests further specifying the criterion to clarify that it would only apply in case of changes to the methodology to set haircuts or value collateral. ESMA also suggests removing sub-point (v) on collateral issued by sovereign issuers.

26 4.13Changes to models that can be considered as already covered by the approved model ESMA initial proposal 112. In addition to the provision set out in Article 49(1h) of EMIR, which is not subject to a further specification via the draft RTS, ESMA (in the Consultation Paper) proposed that changes that result from the implementation of a specific prescriptive recommendation from ESMA, or the NCA in a previous validation decision under Article 49 or 49a of EMIR should not be subject to the procedures under Article 49 or 49a of EMIR. For the avoidance of doubt, the recommendation would need to: a. explicitly mention that the future change would be considered as already covered by the approved model; and b. prescribe the precise scope and approach that the change should follow in addressing the recommendation. Feedback from respondents 113. All respondents supported ESMA’s proposal. One respondent, however, found the conditions in points (a) and (b) too prescriptive. 114. In addition to ESMA’s proposal, respondents proposed that the following changes should also be considered as changes to models that can be considered as already covered by the approved model: Two respondents proposed to leverage the format of quantitative conditions (conditions (a), (c), (d), (e), (f), (g) and (h)) but modifying the thresholds to reflect that these changes are sufficiently minor to be considered as already covered by the approved model, as follows: • Condition (a): use a threshold of 10% of total pre-funded resources; • Condition (c): use a threshold of 10% of total margin as a CCP level; • Condition (d): use a threshold of 10% of the total margin requirements for the financial instruments subject to portfolio offsets; • Condition (e): use a threshold of 10% of the size of any of the default funds; • Condition (f): use a threshold of 10% total liquidity needs; • Condition (g): use a joint threshold of 10% of the CCP’s liquidity exposure to an individual counterparty AND 5% of the aggregate liquid resources; • Condition (h): use a threshold of 5% of the CCP’s total value of collateral.

27

• One respondent proposed the following changes: • Recalibration of collateral haircuts; • Certain model recalibrations; • Corporate actions-driven; • Minor extensions within a group; • Change in parameters within collateral risk framework; • Managing exposures and limits; • Amendment of scenarios suite; • Any extension of a model to new underlying. One respondent proposed to reuse the qualitative conditions (conditions (b) and (i)) to ensure the change is not material.
• One respondent asked for clarification that changes ‘already covered’ under a procedure under Articles 15, 17, 17a of EMIR or ‘exempted’ under Article 15a of EMIR can be considered as ‘already covered’ or ‘exempted’ for purposes of Articles 49 and 49a of EMIR.

115. The majority of respondents also pointed out that Article 49(1h) of EMIR is clear regarding changes to parameters that are part of a validated model and that no further clarification is needed in the RTS in this regard.
116. One respondent argued that the regular ex-post notification of exempted changes which should be considered as already covered by the approved model, combined with annual model validation and review by the NCA under Article 21 of EMIR provide sufficient safeguards to ensure the CCP remains compliant with EMIR without creating unwarranted burden for minor change activities relating to models. ESMA assessment of feedback
117. ESMA has taken note of the feedback received in relation to other changes to models that can be considered as already covered by the approved model and thus not subject to the procedures under Articles 49 and 49a of EMIR.
118. First, ESMA would like to point out that, unlike in respect of extensions of services or activities referred to under Article 15a of EMIR, the Level 1 (Article 49 of EMIR) does not explicitly refer to “BaU” changes or “exempted changes” for model changes, but to “changes not considered model changes” and thus “not subject to the procedures established in Article 49 or Article 49a of EMIR” (as per Articles 49(1h) and 49(5)(c) of EMIR).

28 119. In addition, unlike in respect of the “exempted extensions” (i.e. extensions falling in scope of Article 15a of EMIR), Article 49 of EMIR does not require changes already covered by the approved model to be reported / notified to the NCA and ESMA, and Article 21 of EMIR does not require such changes to be subject to an ex-post review within the annual review and evaluation. 120. However, having taken into account the comments received, ESMA suggests revising the approach to include low quantitative thresholds for each relevant criterion, set out in Article 49(1i) of EMIR, to identify changes that can be deemed as ‘already covered’ when all are met. 121. Under this approach, a change would be deemed as already covered by the approved model if it does not meet any of the conditions for significance and it meets all of the following elements: − decrease or increase of less than 2% of the CCP’s total pre-funded financial resources, including margin requirements, default fund and (first) skin in the game; − decrease or increase of less than 2% of the output of the margin model at the CCP level following the introduction, removal or amendment of a component of the CCP’s margin model (including a margin parameter or an add-on); − decrease or increase of less than 2% of the total margin requirements for a set of financial instruments subject to portfolio margining following a change in the methodology used to compute portfolio offsets for these financial instruments; − decrease or increase of less than 2% in the size of any of the default funds; − decrease or increase of less than 2% of the estimated total liquidity needs across all currencies following a change to the methodology applied to assess liquidity risk; − decrease or increase of less than 2% in the CCP’s liquidity exposure to an individual counterparty; − decrease or increase of less than 1% of the total value of collateral following a change to the methodology applied to value collateral or calibrate collateral haircuts. 122. In addition, ESMA proposes that a change that meets one of the elements set out in Article 8(1), points (c) and (d) (now points (b) and (c)), of the draft RTS, should be considered as already covered by the approved model where the methodology in place at the CCP, which has already been validated by the CCP’s NCA and ESMA, requires adjustments to the confidence interval or the length of the liquidation period with defined rules and based on predetermined, objective criteria. Similarly, ESMA also proposes that a change that meets one of the elements set out in Article 8(1), points (c) and (d) (now points (b) and (c)), of the draft RTS, should also be considered as already covered by the approved model where the change makes by construction the model more conservative (e.g. increase of

29 the confidence level, moving from percentile to expected shortfall and increase of the MPOR). 123. In addition to the above, ESMA also suggests clarifying that changes with no direct impact on the CCP’s total or liquid financial resources (i.e. the resources collected from clearing members, the liquid resources, or collateral values) should not be subject to the procedures set out under Articles 49 and 49a of EMIR. More specifically, investment risk models generally do not affect margin or default fund requirements, and do not directly alter liquid resources. Similarly, a change in a credit scoring model will not directly impact the CCP’s resources. Any impact may occur only in an indirect way, through other choices made by the CCP in its risk management framework.

30 5 RTS on the list of required documents that are to accompany an application for validation of a change to models or parameters 5.1 Introduction/Background 5.2 Chapter 1 – General provisions ESMA initial proposal 124. ESMA’s initial proposal on the general provisions applying to the submission of both types of applications for validation of model changes, regardless of whether they concern significant or non-significant model changes, specifies the language requirements and the need to provide each document submitted for any application with a unique reference number. Furthermore, ESMA proposed that any application is accompanied by a statement from the CCP’s board certifying the accuracy and completeness of all submitted documents. Finally, considering that in some Member States fees may apply when an entity initiates the process to validate a change to a model, ESMA also requested a proof of payment of national fees related to the application when they are due before the model change is validated. Finally, ESMA also requested that applications include an index and a correspondence table to facilitate the identification of the relevant information for the assessment of the applications. Feedback from respondents 125. While language and fees provisions were broadly accepted, all respondents raised concerns primarily around the board certification requirement. Several respondents proposed alternative governance confirmations that, according to these respondents, would be less burdensome. 126. The key concerns raised in relation to the board certification requirement include the disproportionate burden, in particular for routine or technical changes, the misalignment with the role of boards, especially in two-tier board structures, and the risk of slowing down decision making processes for model changes. 127. Some respondents suggested that ESMA explores the possibility of requiring confirmation that internal governance procedures have been completed, including the date of approval. Others proposed that certification is signed by the executive management, the relevant internal committees, or senior officers, rather than the board. 128. In relation to the proposal to submit an index and a correspondence table the feedback reflects broad consensus that this requirement is clear and appropriate. However, one

31 respondent raised concerns about regulatory burden, suggesting the need for proportionality, particularly when combined with other procedural requirements. ESMA assessment of feedback 129. ESMA understand the comments raised regarding the requirement to have a document approved by the board of the applicant CCP certifying the accuracy and veracity of all the documents submitted. 130. However, ESMA consider it essential to ensure that the responsibility of the applicant CCP is assumed in relation to the complete application. In order to provide for more flexibility while keeping the spirit of the requirement, ESMA suggests replacing the board by “the chief executive officer or other person within the governance of the applicant CCP occupying a senior managerial position and engaging the responsibility of the CCP regarding the accuracy and veracity of all the documents submitted”. 5.3 Chapters 2 and 3 – Information to be provided for the validation of significant and non-significant changes Required documentation in relation to the general information ESMA initial proposal 131. For both significant and non-significant model changes, ESMA initial proposal requested that the applicant CCPs provide identifying information, including details of the responsible contact and the purpose of the model change. Furthermore, applications must describe the model’s function within the CCP and provide a forecasted timeline for the change implementation, including key milestones, risks, and mitigations. Feedback from respondents 132. All respondents agreed with the overall objective of providing sufficient detail on the general information related to the model, however none of them fully supported ESMA’s proposal in Article 14. The main areas of concern were the granularity of the implementation timeline (Article 14(d)) and the requirement to assess compliance against all of EMIR for each model change (Article 14(f)). Respondents suggested a more proportionate and targeted approach especially for non-significant changes. 133. Concerning the timeline (Article 14(d)), all respondents opposed the level of detail required, particularly the inclusion of milestones, project risks and mitigating measures. Some of these respondents indicated that internal project management process is dynamic and evolves over time, therefore making detailed snapshots could be misleading or incomplete.

32 134. On the scope of Compliance Assessment (Article 14(f)), there was unanimous concern that requiring CCPs to assess the model change against the entirety of EMIR is duplicative, particularly in view of the annual review foreseen in Article 21 of EMIR, and inefficient, especially if changes only impact a limited part of the regulatory framework. Respondents recommended limiting the assessment to only those EMIR provisions directly affected by the proposed model change. 135. One respondent recommended removing the request for telephone number (Article 14(a)) to align with data minimisation principles under EMIR’s central database provisions. 136. Another respondent suggested that only relevant significance criteria should be assessed for each change (Article 14(e)). ESMA assessment of feedback 137. ESMA agrees with the suggestions to simplify Article 14(d) to require only a high-level timeline without project risk analysis or mitigation detail. The high-level timeline should allow supervisors to have an adequate view of the time-horizon of the model change, without being unduly burdensome for the applicant CCP. 138. ESMA has noted the comments regarding the need to provide further simplification by requiring only an assessment of EMIR provisions directly impacted by the model change. ESMA has amended its initial proposal to simplify and provide further proportionality. 139. ESMA agrees that only the relevant significance criteria should be assessed, which might not imply all significance criteria. ESMA has amended its initial proposal to make it clearer. Required documentation in relation to the description of the model change ESMA initial proposal 140. ESMA’s initial proposal in relation to the information needed to obtain a detailed description of a model change (independently of whether the application is for a significant or a non-significant change), included a request for an overview of the model change, including the rationale, potential benefits, and how it relates to existing models. This section also addresses whether the change replaces or refines previous models. This description must include a list of the contracts within the model’s scope, as well as impacted products, default funds, and asset classes as applicable. 141. Furthermore, ESMA also requested applicant CCPs to include a detailed technical breakdown covering calculations, logical steps, and mathematical and statistical elements of the model. Practical, worked-out examples are also requested in order to illustrate how the model behaves after a significant change. 142. For significant model changes, other elements are required as part of the description. Indeed, the applicant CCP must provide the standards chosen to calibrate the model and

33 justify the choice, including where relevant the confidence level, the lookback period, the time horizon for the liquidation period, the definition of extreme but plausible scenarios. 143. Finally, the applicant CCP must also provide a comprehensive list of parameters and assumptions in relation to a significant model change, along with any limitations identified in relation to those parameters and assumptions. A sensitivity analysis must equally be provided, detailing the qualitative and quantitative impact of these parameters on the model’s performance Feedback from respondents 144. All respondents supported the general intent of Article 15, to provide sufficient information for regulators to assess the impact of model changes on CCP risk management. However, none fully agreed with the level of detail requested, and suggested some refinements. 145. Respondents noted confusion between the model change description and the model documentation already submitted or published under EMIR. They suggested limiting Article 15(c) to only describe the change itself, and not the entire model. 146. In relation to the parameters and assumptions, points (f) and (g) of Article 15, all respondents suggested limiting the information to be provided to parameters and assumptions impacted by the model change. Respondents considered that it would be otherwise burdensome for non-significant changes and adds little incremental value for the assessment. 147. Some respondents indicated that the technical documentation referred to in Article 15(i) is ambiguous. They suggest clarifying it by limiting it to a model description that is sufficient to replicate the model and avoiding internal system specifications and IT implementation documentation which would be too burdensome. 148. One respondent mentioned that, for models with numerous parameters, providing sensitivity analysis for all parameters would become a very extensive exercise. This same respondent suggested that requiring examples to illustrate the model change be optional, as they may not always be relevant or feasible. 149. Another respondent clarified that existing model specifications should be accepted to satisfy unchanged elements (e.g., assumptions, parameters), reducing duplication and documentation overhead. 150. Furthermore, views were split in relation to EMSA’s proposal to require all policies and procedures with relevance to the model even if not amended. Some respondents disagreed, while others agreed with the rationale to understand how the model change fits into the overall risk framework. 151. Some respondents have suggested that CCPs only submit and highlight any amended sections of existing policies and procedures that are relevant for the model change, provided amendments have been conducted, to ensure a more proportional approach.

34 152. Several respondents suggested it would be helpful to clarify that the validation of the model change should not be confused with a revalidation of the existing model or a re￾assessment of the existing risk framework to ensure that the regulatory authorities stay within the confines of the scope of the application according to Article 49 and Article 49(1) respectively. ESMA assessment of feedback 153. ESMA has noted the request for further proportionality and avoidance of duplication. For this reason, ESMA is suggesting a new drafting of Article 15(a) and (c) to focus exclusively on the model change. 154. Furthermore, ESMA has also amended the drafting of points (f) and (g) to include only parameters and assumptions which are affected by the model change. 155. ESMA has also noted that the drafting of Article 15(i) might cover an extensive number of documents which might not always be necessary in the context of the assessment of the model change. ESMA has clarified the requirements for background documentation needed in assessing the model change, specifying this should not include IT and operational documents. However, the CCP should provide methodology documents describing the wider model beyond the component where the change is applied, and the associated policies and procedures. 156. Finally, ESMA has also noted the comments from respondents in relation to point (d) (i.e. worked-out examples). However, ESMA would like to highlight the importance of the provision of these worked-out examples as part of the elements which should allow for the assessment of the model change. Required documents in relation to the governance of the model change ESMA initial proposal 157. Only in relation to significant model changes, ESMA’s initial proposal requested applicant CCPS to include the governance framework that oversaw the model adjustment, including risk committee advice and independent validation outcomes. Furthermore, ESMA suggested that the applicant CCP describe the process that will be followed to analyse and monitor the model performance following the model change. Feedback from respondents 158. Respondents raised concerned about the scope and granularity of the documentation required under Article 16(c). Most of the respondents agreed that operational-level procedures should not be required at the application stage, and that a proportionate, high￾level approach would be more appropriate.

35 159. Several respondents indicated that the term “processes” could be problematic, as it could be interpreted to include daily operational procedures, which are typically not finalised at the time of application. Indeed, these respondents noted that regulatory governance and IT implementation processes run in parallel, and full operational documentation is often not yet available when the application is submitted. 160. Some respondents recommended limiting the required documentation to a high-level description of how performance monitoring will be conducted, specifically for the parts of the model affected by the change, while detailed operational procedures can be reviewed after implementation as part of ongoing supervision. 161. Most of respondents saw no added value in requiring more detailed information on the independent validation process, which is already governed under EMIR. According to them, additional granularity would increase regulatory burden without improving risk oversight. ESMA assessment of feedback 162. ESMA has taken note of the comments raised in particular in relation to the information from the applicant CCP’s risk committee and the processes, including methodologies, frequency and governance arrangements, that the applicant CCP will use to analyse and monitor the model performance following the model change. In relation to information from the risk committee, ESMA sees value in understanding which concerns might have been highlighted by the applicant CCP’s risk committee. However, in order to provide further proportionality and burden reduction, ESMA has amended the initial proposal to request only a high-level description of how the model change will be monitored after validation. Testing methodologies ESMA initial proposal 163. ESMA’s initial proposal requested the comprehensive testing results, including back￾testing, stress testing, and any analyses related to procyclicality of the model change. Furthermore, the proposal also included a request for the applicant CCP to provide results from credit and liquidity stress tests over a 12-month period to ensure the change supports the CCP’s resource adequacy. Feedback from respondents 164. All respondents agreed on the importance of robust testing methodologies to assess model changes, but none fully supported the requirements as currently drafted. Respondents found some of the requirements operationally burdensome, overly prescriptive, and lacking clarity, especially regarding procyclicality, the treatment of hypothetical portfolios, and the scope of stress testing.

36 165. Most respondents raised concerns about the complexity and feasibility of using hypothetical portfolios in testing. According to them, such portfolios are difficult to simulate, especially for metrics like Cover-2 and default fund sizing that rely on legal segregation and portfolio hierarchies. Furthermore, they suggested that implementing them may require parallel production environments, which is resource-intensive and difficult to maintain. 166. Respondents generally agreed with the need for back-testing and stress testing, but they suggested increased proportionality, they recommended avoiding a 12-month lookback period and instead using representative and stressed dates. Respondents also suggested that stress testing should only be triggered when the model in question directly affects credit or liquidity risks (e.g., changes to default fund sizing or liquidity arrangements). 167. Respondents asked for more explicit guidance on how to assess procyclicality, suggesting the use of international standards (e.g. Proposal 6 from the BCBS-CPMI-IOSCO final report on transparency and responsiveness of initial margin in centrally cleared markets, which provides standardized measures of margin responsiveness). 168. On the period for credit and liquidity stress tests, most of respondents suggested aligning with the six-month lookback period proposed in Article 10 of the draft RTS. ESMA assessment of feedback 169. ESMA has taken note of the concerns raised in relation to some of the requirements included in Article 17. ESMA has provided further clarity and efficiency by specifying the type of testing results required, depending on the change for which the applicant is seeking validation. ESMA has also aligned the lookback period for liquidity stress test with the one proposed in Article 10 of the draft RTS.

37 6 Annexes 6.1 Annex I - Legislative mandate to develop technical standards Article 49(5) of EMIR 5. ESMA, in close cooperation with the members of the ESCB, shall develop draft regulatory technical standards to further specify; (a) what constitutes a significant increase or decrease for the purposes of paragraph 1i, points (a) and (c) to (h); (b) the elements to be considered when assessing whether one of the conditions referred to in paragraph 1i is met; (c) other changes to models that can be considered as already covered by the approved model and are therefore not considered a model change and not subject to the procedures established in this Article or Article 49a; and (d) the lists of required documents that are to accompany an application for validation pursuant to paragraph 1c of this Article and Article 49a and the information that such documents are to contain to demonstrate that the CCP complies with all relevant requirements of this Regulation. The required documents and level of information shall be proportionate to the type of model validation but contain sufficient detail to ensure a proper analysis of the change. For the purposes of the first subparagraph, point (a), ESMA may set different values for the different points of paragraph 1i. ESMA shall submit the draft regulatory technical standards referred to in the first subparagraph to the Commission by 25 December 2025. Power is delegated to the Commission to supplement this Regulation by adopting the regulatory technical standards referred to in the first subparagraph of this paragraph in accordance with Articles 10 to 14 of Regulation (EU) No 1095/2010.

38 6.2 Annex II - Cost-benefit analysis 6.2.1 Significant increase or decrease Specific objective The objective is to further specify what constitutes a significant increase or decrease for the purposes of Article 49(1i), points (a) and (c) to (h) of EMIR. Policy option 1 A first policy option would be to specify relative figures (i.e. percentages) for the increase / decrease of each condition. Different values may be used for each condition. Policy option 2 A second policy option would be to specify absolute figures (i.e. amounts in EUR equivalent) for the increase / decrease of each condition. Different values may be used for each condition. A sub-option may include a combination of relative and absolute figures for each threshold. Preferred option Policy option 1. Impact of the proposed policies Option 1 Benefits / drawbacks Specifying percentage thresholds would ensure a consistent and proportional application of the quantitative conditions and a level playing field across all CCPs. In addition, in order to ensure that only meaningful changes are captured under the significance criteria, the proposal includes, where relevant, a double (or triple) threshold assessing the impact of a change against two (or three) different metrics (usually individual and aggregate). Compliance costs Such option should not generate additional costs for the CCPs. Supervision costs The same would apply for the supervisory costs. Option 2 Benefits / drawbacks Specifying absolute figures for the thresholds and/or absolute minimums would decrease costs for small CCPs as the number of standard validations under Article 49 of EMIR would decrease. However, this would also increase the risks posed by smaller CCPs. Compliance costs Lower than option 1 for smaller CCPs, as described above. But costs could be higher for bigger CCPs if the fixed thresholds are set too low. Supervision costs Unchanged.

39 6.2.2 Lookback period Specific objective The objective is to specify the period on the basis of which the significant increase or decrease should be calculated. In doing so, ESMA considered three different dimensions: the length of the lookback period, the degree of flexibility left to CCPs, and the metrics for the calculation of the impact (maximum vs. average). 1st policy dimension: length The lookback period needs to be long enough to provide sufficient coverage and an accurate reflection of the model’s behaviour over diverse market conditions. It should however be short enough to reduce the burden on the CCPs and facilitate the computation of the impact. 2 nd policy dimension: flexibility In order to provide some flexibility for the CCPs and cater, among other, for seasonality effects, the prescribed lookback period could be selected within a wider time range, using a set of X dates, spaced at regular intervals. It should however remain as close as possible to the date of submission of the validation application. 3 rd policy dimension: metrics In assessing the impact of a change, one option could be to compute the average increase or decrease over the lookback period. However, in order to better reflect the risks posed by the proposed models, another option would be to make the assessment on the maximum impact over the lookback period, even if this may trigger a higher number of significant changes. Preferred option Considering the above, ESMA proposes to rely on a lookback period using a set of 12 dates (chosen by the CCP), each representing one month, spaced at regular intervals over a continuous period of 12 months within the 24 months terminating at the date on which the CCP submits an application for validation pursuant to Article 49 or Article 49a of EMIR (to provide a degree of flexibility), and by considering the maximum impact over the period. 6.2.3 Elements to be considered when assessing whether one of the conditions is met Specific objective Having noted that conditions (a) and (c) to (h) are specifically linked to quantitative thresholds which are to be specified under Article 49(5)(a) of EMIR, ESMA considered it necessary to specify these elements only for the two ‘qualitative’ conditions set out under points (b) and (i) of Article 49(1i) of EMIR. The objective is to determine the elements to be assessed in order to consider whether either of these conditions is met.

40 Policy option 1 A first policy option is to rely on a closed list of elements. It would be sufficient for the CCP to meet one element for the respective condition to be fulfilled. Policy option 2 A second policy option is to rely on an indicative open list of elements, that would be complemented by the CCP’s and NCA’s expert judgment to determine whether the respective condition is met. Preferred option Policy option 1. Impact of the proposed policies Option 1 Benefits / drawbacks Option 1 would ensure clarity and legal certainty to CCPs when making the significance determination. It would also ensure that applications under the accelerated procedure are not ultimately rejected by the NCA or ESMA. It would however provide little flexibility for CCPs and NCAs and ESMA and may lead to a higher number of significant changes. Compliance costs Compliance cost is expected to be low. Supervision costs The same would apply for the supervisory costs, as the check from the NCA and ESMA will be straightforward based on the CCP’s assessment. Option 2 Benefits / drawbacks Option 2 would provide some flexibility to the CCP, NCA and ESMA when determining whether a change is significant. However, it could lead to applications being ultimately rejected by the NCA or ESMA despite the CCP having determined the change as non-significant. In addition, it could lead to divergent interpretations and implementation across EU CCPs and NCAs. Compliance costs Could be higher than option 1 as a larger number of applications may be ultimately rejected. Supervision costs Higher than option 1 as the NCA and ESMA would be required to perform an extra assessment on top of the CCP’s determination.

41 6.3 Annex III - Advice of the Securities and Markets Stakeholder Group In accordance with Article 16 of the ESMA Regulation, ESMA has requested the advice of the Securities and Markets Stakeholder Group (SMSG). The SMSG has not provided any comment.

42 6.4 Annex IV - List of respondents Central Counterparty (CCP) 1 Cboe Clear Europe N.V. 2 Deutsche Boerse Group 3 European Association of CCP Clearing Houses (EACH) 4 LSEG 5 SIX Group Banking sector 6 International Swaps and Derivatives Association (ISDA)

43 6.5 Annex V - Draft regulatory technical standards COMMISSION DELEGATED REGULATION (EU) No …/.. supplementing Regulation (EU) No 648/2012 of the European Parliament and of the Council with regard to regulatory technical standards specifying what constitutes a significant increase or decrease for the purpose of Article 49(1i), the elements to be considered when assessing whether one of the conditions referred to in Article 49(1i) is met, other changes to models that can be considered as already covered by the approved model, and the list of documents that shall accompany an application for validation pursuant to Article 49 and Article 49a. of [ ] (text with EEA relevance) THE EUROPEAN COMMISSION, Having regard to the Treaty on the Functioning of the European Union, Having regard to Regulation (EU) No 648/2012 of the European Parliament and of the Council of 4 July 20123 in particular Article 49(5), fifth subparagraph, thereof, Whereas: (1) In order to provide legal certainty to Union CCPs as regards the appropriate procedure for the validation of model changes, improve the efficiency of the validation process, as well as ensure a level playing field across CCPs and a consistent application of Regulation (EU) No 648/2012, this Regulation should provide clear quantitative thresholds for the criteria set out under points (a) and (c) to (h) of Article 49(1i) of Regulation (EU) No 648/2012 in order to determine what would constitute a significant increase or decrease. (2) Different values should be set for each of the quantitative threshold, in order to cater for the specific characteristics of each condition under points (a) and (c) to (h) of Article 49(1i) of Regulation (EU) No 648/2012. In order to ensure a proportionate approach, the quantitative thresholds should be set as a percentage of the increase or decrease of each metric. To reduce the number of false positives, some conditions should be subject to double thresholds, both on 3 Regulation (EU) No 648/2012 of the European Parliament and of the Council of 4 July 2012 on OTC derivatives, central counterparties and trade repositories; OJ L 201, 27.7.2012, p. 1–59.

44 the individual level and the overall level of the CCP’s resources, to make sure that only material changes are captured. (3) With regards to the condition set out under point (e) of Article 49(1i) of Regulation (EU) 648/2012, the thresholds should take into account whether the CCP has a single default fund, which is segmented or organised in liquidation groups, or whether the CCP has multiple default funds. A segmented default fund or a default fund organised in liquidation groups should be understood as a default fund consisting of multiple segments, corresponding to groups of products with similar risk characteristics. The funds of each segment will be used as a priority to cover losses incurred in liquidating products in that group, with surpluses and deficits from other groups potentially included in the calculation with a lower priority. (4) In order to ensure convergence across the Union when assessing the quantitative conditions, the Regulation should specify a minimum lookback period over which the CCP should compute the impact of the change on actual cleared portfolios. The lookback period should be sufficiently short and close to the planned implementation date of the change, but long enough to avoid any window-dressing from the CCP. (5) Given the quantitative nature of conditions (a) and (c) to (h) of Article 49(1i) of Regulation (EU) No 648/2012, other elements to be assessed should only be specified for conditions (b) and (i) of Article 49(1i) of Regulation (EU) No 648/2012, the description of which is more qualitative. In doing so, and to provide legal certainty to CCPs when performing their assessment, this Regulation should establish a closed list of elements to be assessed by the CCP. Where the CCP assesses that at least one of the elements is fulfilled, the respective condition should be considered as met. (6) In order to ensure proportionality in the supervisory assessment of changes to models used by CCPs, it is appropriate to further specify which types of changes can be considered as already covered by the approved model, and are therefore not subject to the procedures set out in Article 49 or Article 49a of Regulation (EU) No 648/2012. In particular, changes resulting from the implementation of specific, prescriptive recommendations issued by the CCP’s competent authority or ESMA in the context of a prior model validation decision should not be treated as model changes, provided that such recommendations clearly state that the change is covered by the existing model approval and define the scope and methodology for the implementation of the change. Similarly, minor technical changes with limited quantitative impact on the CCP’s key financial risk metrics should not trigger the validation procedures. This ensures that minor, low-impact changes are not subject to unnecessary supervisory procedures, while preserving the integrity of the CCP’s risk management framework. Furthermore, changes to the CCP’s confidence interval or liquidation period resulting from the application of a methodology in place at the CCP, already validated by the CCP’s competent authority and ESMA, which requires such adjustments with defined rules and based on predetermined and objective criteria, should be considered as already covered by the approved model. Similarly, changes in relation to the confidence interval or liquidation period that by design increase the CCP’s model conservativeness, including those involving an increase in the CCP’s confidence level, a move from percentile to expected shortfall, or an extension of the CCP’s margin period of risk, should be considered as covered by the approved model. Finally, to ensure clarity and reduce unnecessary regulatory burden, it is appropriate to confirm that changes to models with no direct impact on a CCP’s financial or liquidity

45 resources, such as credit rating or investment risk models, should not be subject to the procedures laid down in Articles 49 and 49a of Regulation (EU) No 648/2012. (7) For significant model changes, the procedure for validation of which is set out under Article 49 of Regulation (EU) No 648/2012, the documentation requirements should be more extensive, covering technical details, governance, and testing results. Regarding non￾significant model changes, the procedure for validation of which is set out under Article 49a of Regulation (EU) No 648/2012, this Regulation should set forth a streamlined set of requirements tailored to the limited impact of such changes. This proportional approach avoids unnecessary administrative burdens for CCPs while ensuring that non-significant changes still adhere to the necessary standards of compliance and transparency. (8) When providing testing results in the context of an application for the validation of a significant model change to a margin model, CCPs should provide back testing results using an observation window which is in line with their own back testing policy, unless they can justify the choice of a different window. Any different observation window should comply with paragraph 2 of Article 60 of Delegated Regulation (EU) No 153/2013 with regard to regulatory technical standards on requirements for central counterparties. (9) This Regulation is based on the draft regulatory technical standards submitted by the European Securities and Markets Authority (ESMA) to the Commission. (10) ESMA has developed the draft regulatory technical standards in close cooperation with the European System of Central Banks (ESCB). In accordance with Article 10 of Regulation (EU) 1095/2010 of the European Parliament and the Council of 24 November 2010 establishing a European Supervisory Authority (European Securities and Markets Authority)4 , ESMA has conducted open public consultations on such draft regulatory technical standards, analysed the potential related costs and benefits and requested the advice of the Securities and Markets Stakeholder Group established in accordance with Article 37 of Regulation (EU) No 1095/2010, HAS ADOPTED THIS REGULATION: TITLE I CONDITIONS FOR THE VALIDATION OF CHANGES TO CCP’S MODELS AND PARAMETERS CHAPTER I SIGNIFICANT CHANGES TO CCP’S MODELS AND PARAMETERS 4 OJ L 331, 15.12.2010, p. 84.

46 Article 1 Significant decrease or increase of the CCP’s total pre-funded financial resources For the purpose of Article 49(1i), point (a), of Regulation (EU) 648/2012, any decrease or increase of more than 20% of the CCP’s total pre-funded financial resources shall be considered as significant. Article 2 Significant decrease or increase of the output of the margin model For the purpose of Article 49(1i), point (c), of Regulation (EU) 648/2012, any decrease or increase of more than 20% of the total margin requirements for a specific clearing service or default fund shall be considered as significant. Article 3 Significant decrease or increase of the total margin requirements for financial instruments For the purpose of Article 49(1i), point (d), of Regulation (EU) 648/2012, any decrease or increase of more than 20% of the total margin requirements for the financial instruments within the portfolio across which the offsets apply shall be considered as significant. Article 4 Significant decrease or increase in the size of any of the default funds or of any individual default fund contribution For the purpose of Article 49(1i), point (e), of Regulation (EU) 648/2012, any of the following shall be considered as a significant decrease or increase: a. as regards a CCP that has multiple default funds, the change in the methodology for defining and calibrating stress test scenarios results in a decrease or increase of more than 20% in the size of any of the default funds; b. as regards a CCP that has a single default fund, the change in the methodology for defining and calibrating stress test scenarios results in a decrease or increase of more than 20% in the size of the relevant liquidation group or segment of the default fund; or c. the change in the methodology for defining and calibrating stress test scenarios results in one clearing member’s contribution to a default fund decreasing or increasing by an amount corresponding to all of the following: i. more than 30% of the size of the clearing member’s contribution, ii. more than 10% of the size of the default fund, and iii. more than the CCP’s own resources calculated in accordance with Article 35 of Commission Delegated Regulation (EU) 153/2013.

47 Article 5 Significant decrease or increase of the estimated liquidity needs in any currency or the total liquidity needs For the purpose of Article 49(1i), point (f), of Regulation (EU) 648/2012, any of the following shall be considered as a significant decrease or increase: a. the change in the methodology applied to assess liquidity risks results in a decrease or increase of more than 20% of the CCP’s total liquidity needs across all currencies; or b. the change in the methodology applied to assess liquidity risks results in the liquidity needs in a single currency decreasing or increasing by an amount corresponding to both: i. more than 50% of the CCP’s liquidity needs in the single currency, and ii. more than 10% of the CCP’s total liquidity needs across all currencies. Article 6 Significant decrease or increase of the CCP's overall exposure to a counterparty For the purpose of Article 49(1i), point (g), of Regulation (EU) 648/2012, where the change results in the CCP’s liquidity exposure towards an individual counterparty, including the entities listed in Article 32(4) of Delegated Regulation 153/2013, decreasing or increasing by an amount corresponding to both: a. more than 20% of the CCP’s liquidity exposure to the individual counterparty, and b. more than 10% of the aggregate liquid resources of the CCP, that decrease or increase shall be considered as significant. Article 7 Significant decrease or increase of the total value of collateral For the purpose of Article 49(1i), point (h), of Regulation (EU) 648/2012, any decrease or increase of more than 5% of the CCP’s total value of collateral (including cash and non-cash collateral) resulting from a change to the methodology applied to value collateral or calibrate collateral haircut shall be considered as significant. Article 8 Changes to the structure or the structural elements of the margin model

1. For the purpose of Article 49(1i), point (b), of Regulation (EU) 648/2012, the elements to be considered when assessing whether that condition is met shall be all of the following: a. Implementation of a new type of model including moving from or to a parametric model, historical simulation, filtered historical simulation or Monte Carlo simulation; b. Changes to the model’s confidence interval in accordance with Article 24 of Delegated Regulation (EU) No 153/2013 or switch between percentile or expected shortfall; c. Changes to the number of days used for the liquidation period in accordance with Article 26 of Delegated Regulation (EU) No 153/2013; and

48 d. Changes to the choice of anti-procyclicality option in accordance with Article 28 of Delegated Regulation (EU) No 153/2013. 2. Where the CCP’s change fulfils at least one of the elements listed in paragraph 1, the condition set out in Article 49(1i), point (b), of Regulation (EU) No 648/2012 shall be considered as met. 3. By way of derogation from paragraph 2, the elements listed under points (b) and (c) of paragraph 1 shall not be considered as fulfilled where: a. the change results from the application of a methodology in place at the CCP, already validated by the CCP’s competent authority and ESMA, which requires adjustments to the confidence interval or the length of the liquidation period, as applicable, with defined rules and based on predetermined, objective criteria; or b. the change makes, by construction, the model more conservative. Article 9 Changes that could have a material effect on the overall risk of the CCP

1. For the purpose of Article 49(1i), point (i), of Regulation (EU) 648/2012, the elements to be considered when assessing whether that condition is met shall be all the following: a. The CCP intends to change its risk models in order to accommodate a new clearing access model which modifies the CCP’s exposure towards participants; b. The CCP intends to change its methodology to set haircuts or value collateral in order to extend the list of eligible collateral to accept collateral: i. in a new asset class; or ii. in the form of gold; or iii. in the form of public bank guarantees, or commercial bank guarantees or non-EU public guarantees; or iv. issued by corporate issuers where the CCP currently only accepts collateral issued by sovereign issuers.
2. Where the CCP’s change fulfils at least one of the elements listed in paragraph 1, the condition set out in Article 49(1i), point (i), of Regulation (EU) No 648/2012 shall be considered as met. Article 10 Lookback period
3. For the purpose of Article 49(1i), points (a), (c), (d), (e), (f), (g) and (h), of Regulation (EU) 648/2012, a CCP shall assess the impact of each change, comparing the outcome of current methodology with the outcome of the proposed methodology, using a set of 12 dates, each representing one month, spaced at regular intervals over a continuous period of 12 months within the 24 months terminating at the date on which the CCP submits an application for validation pursuant to Article 49 or Article 49a of Regulation (EU) 648/2012.
4. A condition shall be considered to be met where any of the corresponding thresholds, as set out in Articles 1 to 7 of this Regulation, is triggered on any of the 12 chosen dates.

49 3. The CCP shall only assess the impact of a model change against the thresholds which are applicable to the type of model change in accordance with Article 49(1i) of Regulation (EU) 648/2012. CHAPTER II OTHER CHANGES TO MODELS THAT CAN BE CONSIDERED AS ALREADY COVERED BY THE APPROVED MODEL Article 11 Other changes to models that can be considered as already covered by the approved model

1. For the purpose of Article 49(5), point (c) of Regulation (EU) 648/2012, the following changes shall be considered as other changes to models that can be considered as already covered by the approved model and are therefore not considered a model change and not subject to the procedures established in Article 49 or Article 49a of Regulation (EU) 648/2012: a. A change that results from the implementation of a specific prescriptive recommendation from the CCP’s competent authority or ESMA contained in a previous validation decision adopted by the CCP’s competent authority or ESMA pursuant to Article 49 or Article 49a of Regulation (EU) 648/2012, provided that the following conditions are met: i. the recommendation explicitly states that the future change is to be considered as already covered by the approved model; and ii. ii. the recommendation prescribes the precise scope and specifications for the future change to address the recommendation; b. A change that does not meet any of the conditions for significance set out in Article 49(1i) of Regulation EU) No 648/2012 and that fulfils all of the following elements: i. decrease or increase of less than 2% of the CCP’s total pre-funded financial resources, including margin requirements, default fund and (first) skin in the game; ii. decrease or increase of less than 2% of the output of the margin model at the CCP level following the introduction, removal or amendment of a component of the CCP’s margin model (including a margin parameter or an add-on); iii. decrease or increase of less than 2% of the total margin requirements for a set of financial instruments subject to portfolio margining following a change in the methodology used to compute portfolio offsets for these financial instruments; iv. decrease or increase of less than 2% in the size of any of the default funds; v. decrease or increase of less than 2% of the estimated total liquidity needs across all currencies following a change to the methodology applied to assess liquidity risk; vi. decrease or increase of less than 2% in the CCP’s liquidity exposure to an individual counterparty; and vii. decrease or increase of less than 1% in the CCP’s total value of collateral (including cash and non-cash collateral) resulting from a change to the methodology applied to value collateral or calibrate collateral haircuts.

50 c. A change referred to in Article 8(3) of this Regulation; and d. A change to a model with no direct impact on the CCP’s total or liquid financial resources, e.g. credit rating models, investment risk models. 2. For the purpose of the elements referred to in paragraph 1, point (b), sub-points i to vii, the CCP shall use the lookback period and methodology set out in Article 10(1) of this Regulation. The elements shall be considered as fulfilled where the applicable decrease or increase is less than the specified thresholds on all of the 12 chosen dates. TITLE II LIST OF DOCUMENTS CHAPTER I GENERAL PROVISIONS Article 12 Language, reference of documents submitted, document index and certification of the application for validation of a model change

1. Documents submitted by an applicant CCP for an application for the validation of a model change pursuant to Article 49 or Article 49a of Regulation (EU) No 648/2012, shall: a. be submitted in a language customary in the sphere of international finance; b. be provided with a unique reference number for each document.
2. Any application submitted for the validation of a model change shall be accompanied by an index, including all the documents and their unique reference number.
3. The applicant CCPs shall provide a correspondence table allowing to identify the section of the document or the document submitted where the information required in the different articles of this Regulation can be found.
4. Any application submitted for the validation of a model change shall be accompanied by a written statement signed by the chief executive officer, or other person within the governance of the applicant CCP occupying a senior managerial position and engaging the responsibility of the CCP, certifying the accuracy and veracity of all the documents submitted in accordance with this Regulation. Article 13 Proof of payment of fees Where national laws of Member States provide for the imposition, by competent authorities, of administrative or any other fees in relation to an application for validation of a model change pursuant to Article 49 or Article 49a of Regulation (EU) No 648/2012, before the application is

51 reviewed or decided upon, the applicant CCP shall also include the proof of payment of any such fees. CHAPTER II DOCUMENTS FOR THE VALIDATION OF A SIGNIFICANT MODEL CHANGE Article 14 General information of the application for the proposed model change An application for validation of a model change pursuant to Article 49 of Regulation (EU) No 648/2012 shall include the following information: a. The identification of the person responsible for the model change application, including full name, function, email address and telephone number; b. The identification of the applicant CCP, including the corporate name of the applicant CCP, its Legal Entity Identifier (LEI) and registered address in the Union; c. The model purpose, including the function of the applicant CCP where the model is used; d. The expected timeline and milestones of the change implementation; e. An assessment of the model change against the relevant significance criteria set forth in Article 49(1i) of Regulation (EU) No 648/2012 and further specified in Articles 1 to 9 of this Regulation. Where the assessment incorporates any assumptions, these assumptions shall also be listed and described. For the purposes of this assessment, the relevant significant criteria shall be the following: i. Where the application concerns a change to a margin model, the criteria specified in Articles 1, 2, 8 and 9 of this Regulation, and the criteria specified in Article 3 if the change concerns the methodology used to compute portfolio offsets; ii. Where the application concerns a change to a stress test model, the criteria specified in Articles 1, 4 and 9 of this Regulation; iii. Where the application concerns a change to a liquidity stress test model, the criteria specified in Articles 5, 6 and 9 of this Regulation; iv. Where the application concerns a change to a collateral haircut or valuation model, the criteria specified in Articles 1, 7 and 9 of this Regulation; f. An assessment of the model change against the requirements of Regulation (EU) No 648/2012 and of Delegated Regulation (EU) No 153/2013 which are impacted by the proposed change. For each requirement, the applicant CCP shall explain how it complies with the requirement, referencing as appropriate the relevant section of the document or documents submitted in the application. Article 15 Description of the model change An application for validation of a model change pursuant to Article 49 of Regulation (EU) No 648/2012 shall include:

52 a. A description of the model change, of a sufficient standard to enable the reader to replicate the model, that shall include the motivation of the model change, the relationship of the proposed model change with the broader model of which it is a part, or other models if applicable, including whether it refines or replaces an existing model, the expected benefits of the proposed model change, and an overview of the technical aspects of the model change; b. The list of contracts in the scope of the model change, including where applicable the products, default funds and asset classes affected by the model change; c. Worked-out examples illustrating the behaviour of the model change; d. The model standards chosen to calibrate the model and their justification, including where relevant the confidence level, the lookback period, the time horizon for the liquidation period, the definition of extreme but plausible scenarios; e. A comprehensive list of all parameters with relevance to the model change with a description of their function in the model; f. A list of assumptions used with relevance to the model change, and the consequent limitations; g. A sensitivity analysis presenting qualitative and quantitative estimates of the materiality of key parameters, assumptions and limitations on relevant model outcomes; h. The policies, procedures and broader model documentation necessary to assess the model that is subject to change with tracking of the changes made for the purpose of the model change. Article 16 Governance An application for validation of a model change pursuant to Article 49 of Regulation (EU) No 648/2012 shall include: a. A description of the methodologies, frequency and governance arrangements, that the applicant CCP will use to analyse and monitor the model performance following the model change; b. The independent validation of the model change conducted in application of paragraph 1b of Article 49 of Regulation (EU) No 648/2012, including any points raised by the independent validation and associated actions planned. Article 17 Testing results An application for validation of a model change pursuant to Article 49 of Regulation (EU) No 648/2012 shall include: a. Where the application concerns a change to a margin model, test results evidencing the impact of the model change on the CCP’s total prefunded resources and on margin requirements for actual and selected hypothetical portfolios. The results shall be supplemented by back testing results for actual and selected hypothetical portfolios over an

53 appropriate observation window, including an evaluation of the coverage after the model change and of the testing exceptions observed, and of the behaviour of the model change in relation to procyclicality and periods of stress. b. Where the application concerns a change to a stress test model, test results evidencing the impact of the model change on the CCP’s total prefunded resources and on stress test results for actual portfolios evidencing the sufficiency of pre-funded available financial resources; c. Where the application concerns a change to a liquidity stress test model, test results evidencing the impact of the model change on liquidity stress test for actual portfolios evidencing the sufficiency and timely availability of liquid resources; d. Where the application concerns a change to a collateral haircut or valuation model, test results evidencing the impact of the model change on the CCP’s total prefunded resources and on collateral values for actual and selected hypothetical portfolios; The test results regarding the impact of the model change in points (a) to (d) above shall be provided at least with reference to the lookback period specified in Article 10. Where the results of the tests incorporate any assumptions, these assumptions shall also be listed and described. CHAPTER III DOCUMENTS FOR THE VALIDATION OF A NON-SIGNIFICANT MODEL CHANGE Article 18 Information to be provided by an applicant CCP for the validation of a non￾significant model change in accordance with Article 49a of Regulation (EU) No 648/2012 An application requesting the validation of a model change under the procedure set out in Article 49a of Regulation (EU) No 648/2012 shall include: a. The document and information requested under Article 14, points (a) to (e); b. The documents and information requested under Article 15, points (a) and (b); c. For the requirements under Regulation (EU) No 648/2012 and under Delegated Regulation (EU) No 153/2013 impacted by the model change, a description of how the applicant CCP achieves compliance with those requirements including the legal references of those requirements. TITLE III FINAL PROVISIONS Article 19 Entry into force This Regulation shall enter into force on the twentieth day following that of its publication in the Official Journal of the European Union.

54 This Regulation shall be binding in its entirety and directly applicable in all Member States. Done at Brussels For the Commission The President