BoA Decision 2020-03 on Scope Ratings v ESMA

D E C I S I O N given by the
BOARD OF APPEAL
OF THE EUROPEAN SUPERVISORY AUTHORITIES under Article 60.2 Regulation (EU) No 1095/2010 and the Board of Appeal’s Rules of Procedure (BoA 2019 04) In the appeal case brought by Scope Ratings GmbH against The European Securities and Markets Authority Decision Ref.: 2020-D-03 Board of Appeal Marco Lamandini (President and Co-Rapporteur) Pat McArdle Katalin Mero Niamh Moloney (Co-Rapporteur) Beata Maria Mrozowska Michele Siri Place of this decision: Paris Date of this Decision: 28 December 2020

APPEAL under Article 60 of Regulation (EU) No 1093/2010, Regulation (EU) No 1094/2010 and Regulation (EU) No 1095/2010 of the European Parliament and of the Council (the “ESAs Regulations”)

This is the decision of the Joint Board of Appeal of the European Supervisory
Authorities on the appeal filed with notice of appeal of 28 August 2020 (the “Notice of Appeal”) by Scope Ratings GmbH (“Scope Ratings” or “appellant” or “applicant”) under Article 60 of the ESAs Regulations. The appellant Scope Ratings is represented in the appeal by Nils Ipsen and Lars Röh of the law firm Lindenpartners. The respondent is the European Securities and Markets Authority (“ESMA”) established by Regulation (EU) No 1095/2010 and represented in the appeal by ESMA in-house Legal Counsels Gerasimina Filippa, Federica Cameli and Mar Huertas. Background of facts
Scope Ratings is a credit rating agency (“CRA”) established in Germany, with
branches in the United Kingdom, Italy, France and Norway. It is registered as a CRA with ESMA since 24 May 2011 (originally under the name “PSR Rating”, then renamed as Scope Ratings in January 2012). The aim of Scope Ratings, in the words of the appellant (Notice of Appeal, paragraph 10), is to offer a European alternative to the three dominant US-based CRAs. The appellant acknowledges in its Notice of Appeal that it is to be considered an important CRA, “taking into account its increase in physical geographical footprint, rise in revenues, number of staff, organisational complexity, and rating issuances” (Notice of Appeal, again paragraph 10).
Central to the facts of this appeal is the appellant’s first Covered Bond
Methodology (the “2015 CB Methodology”). The appellant started the development of the 2015 CB Methodology in 2014; the final draft of the 2015 CB Methodology was approved for publication by the appellant on 11 February 2015 and on the same day the appellant duly informed ESMA of its intention to publish this final draft for public consultation. On 12 February 2015, the appellant issued therefore a press release with a link to the final draft of the 2015 CB Methodology, inviting market participants to submit comments until 3 April 2015. On 26 February 2015 and 10 March 2015, the appellant held conference calls with stakeholders on the proposed 2015 CB Methodology. On 3 July 2015, the final version of the 2015 CB Methodology was finally approved for publication by the appellant, published on the appellant’s website, and notified to ESMA. The appellant also issued a press release including a link to the summary of comments received during the consultation and the resulting clarifications and/or amendments to the initial draft of the 2015 CB Methodology adopted with its final version. The (published, final version of the) 2015 CB Methodology is part of the file of this appeal, as Exhibit 6 of the appellant’s submissions.
The purpose of the 2015 CB Methodology was to set the framework for the
rating assessment and the regular monitoring of covered bonds by the appellant, “tak[ing] into account the significantly changed regulatory and

supervisory framework applicable to financial institutions, in particular the introduction [rectius: entry into effect] of the Bank Recovery and Resolution Directive (BRRD) in the European Union on 1 January 2015 as well as similar resolution regimes in other countries. It also reflects the importance of covered bonds in the regulatory and supervisory frameworks and for central banks’ monetary policy” (2015 CB Methodology, page 3). The notion of dual recourse is intrinsic to the structure of covered bonds. The 2015 CB Methodology accordingly applies “to all covered bonds that benefit from a dual recourse to both a financial institution and a ring-fenced cover pool. A financial institution is responsible for the timely and full payment of interest and principal (first recourse). In contrast to other parts of the bank’s liability structure, covered bonds are excluded, and hence protected against an issuer’s restructuring or resolution, which is the envisaged rescue [rectius: crisis management] mechanism for a bank in distress. Even in the unlikely event of an issuer default, covered bonds should in general not accelerate and a cover pool of eligible assets remains available to substitute for the issuer’s obligation to service the bonds (second recourse)” (2015 CB Methodology, page 3).
5. The 2015 CB Methodology describes the appellant’s rating approach to
covered bonds as follows (2015 CB Methodology, page 4). The Issuer Credit Strength Rating (“ICSR”) is “the fundamental anchor point for the covered bond analysis”. Scope Ratings ICSR “represents a credit opinion on a bank’s ability to meet its contractual commitment on a timely basis, and in full, as a going concern” and it signals “the relative risk of a default-like event” (2015 CB Methodology, page 4, footnote 4). However, there are two additional supporting elements for the covered bond analysis which reflect the dual recourse feature of covered bonds and which are taken into consideration by the appellant under the 2015 CB Methodology, specifically (a) “the combination of the legal and resolution frameworks” as “the most important supporting element for the covered bond rating” (2015 CB Methodology, page 4), because “covered bond ratings for highly rated banks are driven primarily by the fundamental benefits of the regulatory framework applicable to banks and their covered bonds” and (b) “the benefit of the cover pool [which] represents a second recourse coming after a chain of events affecting the issuer. The benefit of the cover pool is limited but it provides additional security and stability to the rating” (2015 CB Methodology, page 4). The final rating of a covered bond could, under the 2015 CB Methodology, accordingly, be supported by two building blocks: (a) analysis of the legal and resolution regime; and (b) analysis of the cover pool.
6. Based upon these two supporting elements, Scope Ratings’ 2015 CB
Methodology provided that, in the covered bonds rating, the bank’s ICSR can be uplifted up to 9 notches due to credit risk enhancements arising from the two building blocks as follows: (i) up to 6 notches, if the bank is resolvable, as a consequence of the legal framework concerning credit differentiation (i.e. asset segregation upon insolvency of the issuer, uninterrupted continuation of payments of interest and principal on the covered bonds after the issuer insolvency, market and liquidity risk management principles applied prior to and after insolvency, overcollateralization fully available after insolvency and independent monitoring by a trustee or supervisor: 2015 CB Methodology, p. 7) (up to 2 notches) and of the resolution regime (up to 4 notches) and (ii) up to 3

additional notches “above the uplift already provided by the fundamental status of covered bonds and the regulatory framework of the issuer”, based upon the cover pool analysis1. In this respect the 2015 CB Methodology acknowledges that “Scope […] performs a thorough analysis of the cover pool because it provides key information about the robustness of the covered bond second recourse, and ultimately, the magnitude of the expected loss for the instrument” (2015 CB Methodology, page 4). More specifically, although “the supportive benefit of the cover pool only becomes relevant when the credit quality, and thus the bank ratings, start to migrate downwards”, Scope Ratings clarifies in the 2015 CB Methodology (at page 5) that it “performs and publishes a detailed quantitative analysis of the cover pool for programmes from both highly and lowly rated issuers. This is because the cover pool analysis helps to understand the likely stability of the covered bond rating, the efforts issuers must make to manage risks prudently or the levels of overcollateralization they have to provide to mitigate these risks”.
7. To summarise, using the words of the appellant itself in its 2015 CB
Methodology, “Scope’s covered bond rating is closely linked to the ICSR of the bank issuer but is generally higher”. “Higher-than-ICSR ratings for covered bonds predominantly reflect their different treatment in bank resolution and recovery regimes, provided the covered bonds fall under the relevant regulatory definition in the resolution regime. They also rest on the assumption that a dedicated, ring-fenced cover pool of typically low credit-risk assets can replace the bank’s obligation to ensure that payments on the covered bonds can be made in full and in a timely manner”. Indeed, “covered bond ratings reflect the probability of insolvency of both the issuer and the cover pool, and the associated expected loss”. Therefore, the “methodology considers the dynamics and constraints that may be specific to each region, market or market segment”.
8. Legal frameworks analysis, in this context, “seeks to determine whether (i) the
framework provides sufficient protection to legally allow for uninterrupted payments on the covered bonds; and (ii) the structure could be affected by a moratorium or insolvency of the issuer. Furthermore, (iii) the analysis must identify whether the covered bonds can benefit from the preferential treatment of the resolution regime. Lastly, (iv) the analysis of the legal covered bond framework also supports the quantitative analysis, as [Scope Ratings] will be able to identify whether and how the framework further reduces the probability of default and mitigates the loss given default of the covered bonds” (2015 CB Methodology, p. 6). In this respect, the appellant acknowledges that “asset isolation is at the core of the covered bond definition, alongside the dualrecourse principle. It is one of the areas where most diligence is performed when a covered bond framework is set up by regulators or a specific covered bond structure is set up by issuers, often taking into account the preferences of investors as well”.
9. Cover pool analysis, in this context, “focuses primarily on identifying the
collateral characteristics driving the magnitude and patterns of assets defaults, However, the ratchet is one-way as a covered bond rating may be augmented by the cover pool analysis but not reduced by it.

the severity of losses upon an asset default and the dependency structure in cases of cover pools spanning multiple asset classes and geographies” (2015 CB Methodology, page 9). Although the 2015 CB Methodology considered that “covered bonds issued by high investment-grade-rated resolvable banks can exhibit a credit quality commensurate with AAA level, because of the covered bond status in the bail-in, regardless of the level of overcollateralization provided in their cover pool” and that “the use of the cover pool to fulfil the payment obligations under the covered bond only becomes necessary when a resolution has failed and the issuer has defaulted”, it also noted that “however, a thorough analysis of the cover pool needs to be performed for all rated covered bonds. The findings inform us on how specific features of the covered bond structure, as well as other country specific aspects, may affect the probability of default and the loss given default. It also provides information on the likely rating sensitivity resulting from it” (2015 CB Methodology, page 9). 10.The 2015 CB Methodology was applied until the adoption of a new version of the methodology on 22 July 2016 (the “2016 CB Methodology”). 11.More specifically, under the 2015 CB Methodology the appellant issued ratings concerning 17 covered bond programmes, for an overall total number of 622 ratings. 15 programmes were unsolicited and 2 were solicited. The 559 unsolicited ratings were issued on 22 September and 26 November 2015, whereas the 63 solicited ratings were issued on 4 May and 8 July 2016. 12.The cover pool analysis required under the 2015 CB Methodology was however performed only in the two solicited covered bond programmes. It was not performed for any of the 15 unsolicited covered bonds programmes and thus not performed for any of the 559 unsolicited ratings issued in September and November 2015 as part of these programmes. Most of these ratings were still outstanding when the 2016 CB Methodology was adopted. 13.Alongside the nature of analysis required and in practice applied under the 2015 CB Methodology, the process of adoption of the 2016 CB Methodology is another relevant aspect of this appeal. Prior to the adoption of its 2016 CB Methodology, the appellant did not publish the proposed changes to the 2015 CB Methodology on its website; neither did it invite stakeholders to submit comments. Further, the appellant did not notify ESMA about those changes and could not publish on its website the results of a consultation with stakeholders as such a consultation did not take place. 14.Based upon the facts summarised above, and following a complaint made to ESMA (as noted in Exhibit 19 of ESMA’s submissions), in a report dated 28 March 2018, ESMA concluded that there were serious indications of the possible existence of facts liable to constitute several infringements of the CRA Regulation (the potential infringements being listed in Annex III of the Regulation). That same
day, the Executive Director of ESMA appointed an Independent Investigating Officer (“IIO”), pursuant to Article 23e(1) CRA Regulation, to investigate the matter.

15.On 16 May 2019, the IIO sent her initial statement of findings to the appellant. The document included the findings that the appellant had committed the infringements set out at: (i) point 43 of Section I of Annex III of the CRA Regulation (infringement of Article 8(3) by not applying the 2015 CB Methodology in a systematic way); (ii) point 3a of Section II of Annex III of the CRA Regulation (infringement of Article 14(3) by not having notified ESMA of the intended material changes to the 2015 CB Methodology); (iii) point 3b of
Section II of Annex III of the CRA Regulation (infringement of Article 8(5a) by
not having published on its website the proposed material changes to the 2015 CB Methodology) and (iv) point 4a of Section III of Annex III of the CRA Regulation (infringement of Article 8(6)(aa) by not having informed ESMA and not having published immediately on its website the results of a consultation about the 2015 CB Methodology). 16.On 24 June 2019, the appellant provided written observations in response to the initial statement of findings of the IIO. 17.On 25 July 2019, the IIO issued an amended statement of findings confirming the findings that the appellant committed the infringement set out at points 43 of Section I, 3a and 3b of Section II and 4a of Section III of Annex III of the CRA Regulation. 18.On 1 April 2020, based on the file containing the IIO’s findings and taking account of the submissions made on behalf of the appellant, the ESMA Board of Supervisors sent a statement of findings to the appellant. 19.On 22 April 2020, the appellant provided written submissions in relation to the statement of findings of ESMA Board of Supervisors. 20.On 28 May 2020, the Board of Supervisors adopted the appealed decision (ESMA 41-356-77 of 28 May 2020: the “Contested Decision”) in which it fully agreed with the IIO and: (a) found that the appellant negligently committed the infringements at points 43 of Section I, 3a and 3b of Section II and 4a of Section III of Annex III of the CRA Regulation; (b) adopted a supervisory measure in the form of a public notice pursuant to Article 24 of the CRA Regulation; and (c) imposed on Scope Ratings a fine of EUR 640,000, pursuant to Article 36a of the CRA Regulation. The appeal against the Contested Decision and the proceedings before the Board of Appeal. 21.On 28 August 2020, the appellant filed the Notice of Appeal with the Joint Board of Appeal of the European Supervisory Authorities pursuant to Article 60 of Regulation (EU) No 1095/2010 (the “ESMA Regulation”) in which it requested the Board of Appeal not to confirm the Contested Decision and to remit the case to the ESMA Board of Supervisors to adopt an amended decision based on its findings. 22.On 1 September 2020, the Secretariat of the Board of Appeal notified to the parties the following communication from the President of the Board of Appeal:

Dear Parties,
The President thanks the Appellant for the appeal received by the Board of Appeal Secretariat on 28 August 2020. The President, having consulted with the Board of Appeal, makes the following observations:
Subject to the view of the parties, in light of Article 6 of the BoA Rules of Procedure, the President proposes giving (i) ESMA three weeks from the notice of these directions to respond to the appeal, (ii) the appellant two weeks to reply to ESMA response; (iii) ESMA two weeks to present, if any, a rejoinder to the appellant’s reply. According to Article 18 of the BoA Rules of Procedures, parties are entitled to make oral representations. In the absence of a request, the Board of Appeal may require oral representations if it considers it to be necessary for the just determination of the appeal. Both parties are invited therefore to communicate to the Secretariat of the Board of Appeal, one week from the expiry of the deadline for ESMA to present a rejoinder to the appellant’s reply if they intend to make oral representations. The Board of Appeal shall issue further case management directions after the expiry of the above-mentioned deadline concerning the hearing and, if deemed necessary by the Board, further submissions in the merits. The President shall act as Rapporteur of the case. The parties are asked to confirm this proposal with the Secretariat and raise any other points they wish to raise at this stage. The President wishes also to inform the parties that the filing and service of any further communication between the Parties and between the Parties and the Board of Appeal and its Secretariat (including the filing and service of the Respondent’s response pursuant to Article 6 of the Rules of Procedure and of any other submissions of the parties) may take place by email. The acting secretariat of the Board of Appeal (boardofappeal@eba.europa.eu) must always be copied. The parties are hereby informed about the sitting composition of the Board of Appeal according to Article 3(4) of the BoA Rules of Procedure:

Marco Lamandini (President and Rapporteur)
Pat McArdle
Katalin Mérő
Niamh Moloney
Beata Mrozowska
Michele Siri
23.On 7 September 2020, ESMA filed an application for directions with the Board of Appeal pursuant to Article 11 of the Rules of Procedure in which it requested an extension of the deadline to submit its response to the Appeal (the “ESMA Response”) in accordance with Article 6 of the Rules of Procedure of the Board of Appeal. 24.On 8 September 2020, the Board of Appeal issued directions pursuant to which ESMA was given five weeks to respond to the Notice of Appeal, starting from the notice to both parties of those directions, as follows:

Dear Parties,
The President hereby informs the parties that he has received, on 7 September 2020, an application from the Respondent, ESMA, to amend the directions furnished to the parties on 1 September 2020. ESMA believes that the three-week period in the 1 September directions is not sufficient to adequately address the Notice of Appeal and its annexes and to deliver its Response and has asked the President to amend the directions as follows:
“The Respondent to submit the Response within 2 months of receipt of the Notice of Appeal in full and in any case no later than 30 October 2020”.
Article 8(1) of the Board of Appeal (BoA) Rules of Procedure provides:
“Any time limit prescribed by or imposed under these Rules may be extended by the Board of Appeal or by the President”. The President, having consulted with the Board of Appeal, makes the following observations:
The Board of Appeal acknowledges that the present appeal relates to an issue of great importance and that the outcome of this Appeal could have a significant impact on ESMA’s supervisory powers and practices. It notes further that the Notice of Appeal is extensive and complex and needs careful consideration by ESMA’s Legal Team. The Board of Appeal also concedes that, in the current exceptional circumstances, the ESMA Legal Team – which is not a large team – may be under extreme pressure, since it must also address, alongside its regular work, a pressing work agenda relating to Covid19 and Brexit. The Board of Appeal further acknowledges that, unfortunately, Exhibit 7 was not attached to the Notice of Appeal notified to ESMA by the Secretariat on 1 September 2020 and that, accordingly, the supporting material was not complete. The Board of Appeal considers that these circumstances justify an extension of the original term of three weeks proposed by the President in the first case management directions. However, the Board of Appeal considers that, in order to ensure a timely adjudication of the matter, the extension should be shorter than the one requested by ESMA, also considering that the next meeting of the ESMA Supervisory Board, is scheduled for 23 September 2020. Subject to the views of both parties, which are kindly requested to confirm their agreement, and in light of Article 6(6) and Article 8(1) of the BoA Rules of Procedure, the President, having consulted with the Board of Appeal, proposes to amend the original directions to give ESMA five weeks to respond to the Notice of Appeal, starting from the notice to both parties of these new directions, to which Exhibit 7 is now also attached. Subject to the views of the parties, it remains confirmed that, after the filing of the ESMA’s Response, (i) the Appellant is granted two weeks to reply to the ESMA response and (ii) ESMA is granted two weeks to present, if any, a rejoinder to the Appellant’s reply. The Board of Appeal shall issue further case management directions after the expiry of the above-mentioned deadlines concerning the hearing and, if deemed necessary by the Board, further submissions on the merits.

25.On 9 September 2020, both parties informed the Secretariat in writing that they agreed with the directions issued by the Board of Appeal on 8 September 2020. 26.On 13 October 2020, ESMA submitted its Response, and related documents, in accordance with Articles 6, 7 and 8 of the Rules of Procedure and the directions issued by the Board of Appeal. 27.On 22 October 2020, the appellant filed an application for the amendment of the directions issued by the Board of Appeal on 8 September 2020, asking for an extension of one week of the two-week period originally granted to the appellant for filing a reply to ESMA Response. 28.On 23 October 2020, the Secretariat notified the parties the Board of Appeal directions following the appellant’s application as follows:
Dear Parties,
The President hereby informs the parties that he has received, on 22 October 2020, an application from the Appellant to amend the directions given to the parties on 1 and 8 September 2020. The Appellant asks for an extension of time for filing a reply to ESMA’s response of 13 October 2020, because the two-weeks period set in the directions of the Board of Appeal given to the parties on 1 and 8 September is not sufficient, in the Appellant’s view, to adequately address the Response and its annexes. The Appellant further notes that the appeal relates to issues of fundamental importance, the outcome of the appeal may strongly affect the applicant and ESMA response is extensive and complex and needs careful consideration by the Applicant in order to effectively exercise its right of defence. The Appellant asks therefore the extension of one week of the original deadline for the reply.
Article 8(1) of the Board of Appeal (BoA) Rules of Procedure provides:
“Any time limit prescribed by or imposed under these Rules may be extended by the Board of Appeal or by the President”. The President, having consulted with the Board of Appeal, hereby grants to the Appellant the required extension of one week, and, in order to ensure equal treatment to both parties, also grants to the ESMA three weeks, instead of the two weeks originally set with the directions given to the parties on 1 and 8 September 2020, for its rejoinder, if any, to the Appellant’s reply. The Board of Appeal shall issue further case management directions after the expiry of the above-mentioned deadlines concerning the hearing and, if deemed necessary by the Board, further submissions on the merits. 29.On 3 November 2020, the appellant submitted its reply to the ESMA Response, with related documents (“Appellant Reply”). 30.On 24 November 2020, ESMA submitted its rejoinder to the appellant’s reply, with related documents (“ESMA Rejoinder”). 31.On 25 November 2020, the Secretariat of the Board of Appeal invited the parties, in accordance with the case management directions already agreed, to communicate within one week if they intended to make oral representations,

anticipating that, due to the restrictions imposed to counteract the Coronavirus pandemic, the hearing, if requested, would have to be held via videoconference. On 26 November 2020, ESMA informed the Board of Appeal that it did not intend to make oral representations, in consideration of the extensive written submissions and documentation already exchanged between the parties and also because a hearing via videoconference could not fully preserve the principle of procedural immediacy. On 1 December 2020 the appellant, on the contrary, informed the Board of Appeal that it considered a hearing useful, also in light of the ESMA Rejoinder. The right for both parties to make oral representations at an online hearing was therefore granted by the Board of Appeal and the parties agreed to hold a hearing on 11 December 2020. 32.On 4 December 2020, the President issued case management directions as to the organisation and management of the hearing, which were agreed with the parties. The parties were also informed that, due to the importance of the matters in dispute, the President, having consulted with the Board of Appeal, had designated with her consent Professor Niamh Moloney as co-rapporteur to the purpose of Article 12 of the Rules of Procedure. 33.On 9 December 2020, the parties and the Board of Appeal made a pre-hearing test of the webex online platform to be used for the hearing and agreed on further details of the organisation and management of the forthcoming online hearing. 34.The hearing was held via webex videoconference on 11 December 2020. During the hearing the parties exercised their right under Article 60(4) of the ESMA Regulation to make oral representations, answered the questions raised by all sitting members of the Board of Appeal and were granted the opportunity to make short final replies. 35.At the end of the hearing, the Board of Appeal held that evidence was complete in relation to the appeal under Article 20 of Board of Appeal’s Rules of Procedure for the purposes of Article 60(2) of the ESMA Regulation and the President informed the parties that the appeal was considered lodged as of the date of 11 December 2020. The contentions of the parties. The appellant 36.In its Notice of Appeal and further in its Reply, the appellant complains that the Contested Decision (ESMA 41-356-77 of 28 May 2020, communicated to Scope on 29 May 2020 and published by ESMA on its website on 4 June 2020), is not well founded in law. The appellant argues that, contrary to the findings of the Contested Decision, it has not negligently committed: (a) the infringement set out at point 43 of Section I of Annex III of Regulation (EC) No 1060/2009 (as amended where relevant by Regulation (EU) No 513/2011 and Regulation (EU) No 462/2013, hereinafter the “CRA Regulation”) and (b) the

infringements set out at points 3a and 3b of Section II of Annex III and point 4a of Section III of Annex III of the CRA Regulation. 37.More specifically, regarding the infringement set out at point 43 of Section I of Annex III of the CRA Regulation, the Contested Decision finds that the appellant did not apply its 2015 CB Methodology in a systematic way and therefore breached Article 8(3) of the Regulation. On this, the appellant puts forward several grounds of appeal as follows. 38.In the appellant’s view, the wording of Article 8(3) of the Regulation, which is at the legal core of the alleged infringement, only concerns the design of a rating methodology. Article 8(3) does not relate to the application of a methodology. The grounding of the alleged infringement in point 43 of
Section I of Annex III of the Regulation is therefore wrong in law. The
appellant’s argument relies, inter alia, on the textual meaning of Article 8(3) and its limitation to the content of a methodology; the EU legislature’s deployment of a design/use distinction elsewhere in Article 8 (which relates to methodologies), and particularly Articles 8(2a) and 8(7); the objectives of the Regulation; and the dictates of the legal certainty principle. The appellant is also of the view that the interpretation adopted by ESMA in its Contested Decision of Article 5(1) of Commission Delegated Regulation (EU) No 447/2012 (hereinafter “Delegated Regulation No 447/2012”), which amplifies the meaning of “systematic” in Article 8(3), to complement its interpretation of Article 8(3), unlawfully extends the meaning of Article 8(3). 39.Alternatively, even if one assumes that Article 8(3) of the Regulation extends to the use of methodologies and so requires the systematic application of the rating methodology (and not only that the methodology be systematic), the appellant is of the view that this systematic application does not require that all of the steps of the analytical rating process laid out in the rating methodology should be applied, where some of those steps are not relevant for the determination of the final rating in an individual case. Specifically, if the highest rating in an individual case has already been achieved, the systematic application of a methodology does not require, in the appellant’s view, the fulfilment of additional, but at that point irrelevant, steps, even if they are envisaged in the methodology. Further, on the facts, the appellant argues that all of the ratings issued by the applicant at the time were correctly assigned and that the result of the analysis would not have changed, even if a cover pool analysis had been carried out by the applicant at the time. Accordingly, the appellant argues that the cover pool analysis was, in the specific context of each rating analysis where the 2015 CB Methodology was not (fully) applied, not relevant. Therefore, the applicant argues that it systematically applied the 2015 CB Methodology when it carried out only those steps that were relevant or necessary for the rating. In the appellant’s view, the application of a methodology is not an end in itself but a means to an end and, given also that judgment is to be exercised in the rating process, only relevant factors should be taken account of in the application of methodologies. The applicant also raises the principle of proportionality to argue that the enforcement action relating to Article 8(3) is disproportionate

as there was no decrease in the quality of the ratings and as full transparency was provided to the market. 40.Alternatively, even if one regards the approach of the applicant – of carrying out only those steps it regarded as relevant for the rating result – as a divergence from the 2015 CB Methodology, the decision to concentrate rating efforts only on relevant steps is objectively justified and amounts to an “objective reason” for the divergence within the meaning of Article 5 of the Delegated Regulation No 447/2012. The appellant argues inter alia that the meaning of objective reason implies, given also the proportionality principle, that any divergence from the methodology must not be arbitrary and must be suitable, necessary and appropriate in light of the objectives of the CRA Regulation. The requirements for a divergence are met as the divergence related to the omission of steps that were not relevant, it did not make any difference to the ratings, and the omission of the cover pool analysis was published to the market through press releases. 41.Alternatively, even if one assumes a violation of Article 8(3) of the Regulation, the appellant did not act negligently, as the appellant acted according to common market practices or standards and provided full disclosure to investors about the absence of a cover pool analysis. Moreover, the appellant could not foresee the interpretation of Article 8(3) of the CRA Regulation adopted by ESMA, as this interpretation goes beyond the wording of Article 8(3) of the CRA Regulation and no ESMA guidance existed at the time of the alleged infringement. The appellant also argues that any claim of negligence must be established in relation to the application of the methodology only and not in relation to the design phase. 42.Alternatively, even if one assumes that the appellant committed a negligent infringement, the amount of the fine imposed on the applicant was not calculated correctly. In particular, the reasoning of the Contested Decision concerning the duration of the infringement and the calculation of the basic amount of the fine, and the application of mitigation factors, is incoherent and to the detriment of the appellant. Additionally, ESMA did not correctly apply the cap provided for in Article 36a(4) of the CRA Regulation regarding the turnover of the applicant. If it had, the fine would have amounted to EUR 310,929. 43.Overall, the fine is, in the appellant’s view, clearly disproportionate to the alleged omission, because the omission did not have – and could not have – any negative effect on investor protection. Neither the quality of the ratings nor the level of transparency of the analytical approach that was provided to market participants were jeopardised. Such an alleged (minor) infringement cannot lead to one of the highest fines possible under the Regulation. This disproportionality is further evidenced by the fact that the fine significantly exceeds the revenues the appellant derived
from using the 2015 CB Methodology. In fact, the appellant argues that the fine is more than 3.5 times higher than the revenues that Scope Ratings derived from using the 2015 CB Methodology. In this way, in the appellant’s view, it clearly exceeds the

limits of what is necessary in order to attain the objectives of the CRA Regulation 44.The arguments raised by the appellant are supported by an expert opinion of Professor Dr Christoph Kumpan of Bucerius Law School of April 2020, presented by the appellant with the Notice of Appeal. 45.In particular, Professor Kumpan’s opinion reaches the following conclusions. (a) When assessing whether a rating agency has breached one of the provisions in Annex III of the Regulation, a distinction must be made between the design of a methodology and its concrete application. In the case of Scope Ratings GmbH, the design of the 2015 CB Methodology (as well as of the 2016 CB Methodology) was in accordance with Article 8(3) of the Regulation. The alleged infringement only concerns the correct application of the methodology. In such a case, however, a fine cannot be imposed pursuant to point 43 of Section I of Annex III of the Regulation. (b) For all credit ratings, the cover pool analysis can be regarded as a rating outlook within the methodology used by Scope Ratings GmbH in 2015, even if it is not described as a rating outlook in the 2015 CB Methodology. When issuing a rating outlook, point 43 of Section I of Annex III of the Regulation is not applicable (and thus cannot be violated), because Article 8(3) of the Regulation is not applicable to rating outlooks, as can be deducted from
Article 8(2) of the Regulation.
(c) The non-application of the cover pool analysis can be based on an objective reason within the meaning of Article 5 of the Delegated Regulation. In particular, not applying one of several investigation methods if the best possible rating has already been achieved on the basis of the other investigation methods does not appear arbitrary, but rather sensible from both an analytical and an economic point of view and also suitable, necessary and appropriate, because in this way a proper rating is established and an unnecessary waste of resources is avoided. (d) In its decision, ESMA has not taken sufficient account of the wording and the context of the provision, or of the objectives of the Regulation. ESMA’s extensive reading of Article 8(3) of the Regulation disregards fundamental principles of European Union law, in particular, the principle of legality (nulla poena sine lege) and the principle of proportionality that have to be observed when applying provisions that lead to the imposition of a fine. ESMA broadens the scope of Article 8(3) of the Regulation in a way that could not have been foreseen by the applicant and that is still not supported by the text of the Regulation today. (e) Moreover, ESMA interprets not only Article 8(3) of the Regulation but all other relevant provisions to the detriment of the applicant. As a result, the applicant is inflicted with a disproportionate fine considering that the (alleged) infringement in any case did not jeopardise the objectives of the

Regulation at any time. In particular, the applicant provided for high quality ratings (only abstaining from a cover pool analysis where the highest rating had already been achieved) and made its approach to the 2015 CB Methodology fully transparent in a press release. 46.Regarding the infringements of points 3a and 3b of Section II of Annex III and point 4a of Section III of Annex III of the Regulation, the Contested Decision finds that the applicant has made material changes without complying with the procedures set out in Article 8(5a) of the Regulation. 47.The appellant submits that there has not been a material change to the 2015 CB Methodology in the sense of Article 8(5a) of the Regulation. The changes were simply a clarification of the rating methodology that was already in use and/or concerned only the rating outlook. 48.With its Reply to the ESMA Response, the appellant reiterates its position as expressed in the Notice of Appeal and further expands on its arguments that (a) the scope of Article 8(3) of the CRA Regulation applies to the design of a methodology, but not to its application, noting in this respect that an incorrect application of a systematic methodology falls under Article 8(7) and not under Article 8(3) of the CRA Regulation; (b) pursuant to the Meronidoctrine, ESMA is strictly bound by the provisions of the Regulation, i.e. their wording, and (c) ESMA’s interpretation of Article 8(3) of the CRA Regulation violates the principles of legality and legal certainty as well as the principle of proportionality. If one assumed – quod non – that the wording of Article 8(3) of the CRA Regulation would allow for different interpretations, ESMA has to apply the principle of proportionality when deciding which interpretation to adopt. 49.Furthermore, the appellant reiterates that it acted with a high standard of care as would be expected from a regulated firm when it developed and approved the 2015 CB Methodology. In the appellant’s view, the design reflected the proprietary data and information the appellant had at that time with respect to covered bonds programmes. The design contained analytical building blocks which the appellant was able to perform as described by the 2015 CB Methodology. Further, the members of the credit ratings committees which met in September and November 2015 acted with high professional care and analytical diligence when applying the 2015 CB Methodology, with a deviation regarding the level of necessary cover pool analysis. 50.The appellant argues further, in this respect, that the appellant had sufficient public information to undertake a robust qualitative and quantitative analysis for the covered bonds programmes of the 12 European banks which were rated by it in two bulk rating actions on 22 September 2015 and 26 November
2015. The analytical approach which was taken by the members of the
appellant’s rating committees for the credit rating actions of 22 September 2015 and 26 November 2015 did include a deviation from the 2015 CB Methodology. This deviation was based, in the appellant’s view, on an objective reason. Whilst the appellant agrees with ESMA that any deviation

for objective reasons must preserve the quality and accuracy of the credit rating, the appellant is also of the opinion that (a) the rating committee’s analytical rationale which justified the deviation from the 2015 CB Methodology in September and November 2015 was not foreseeable at the time when the 2015 CB Methodology was developed and approved and (b) such rationale was derived from on an issuer specific interplay of analytical components (of the 2015 CB Methodology). By contrast, the rating committees for the credit rating actions of 3 May 2016 and 16 July 2016 needed the complete qualitative and quantitative cover pool analysis to reach the highest possible covered bond ratings and thus the analytical scenarios of both these credit rating actions were not comparable with those of the 2015 credit rating actions. 51.Finally, as to the amount of the fine, the appellant upholds the argument that ESMA did not factor in mitigating factors and incorrectly factored in aggravating factors when calculating the fine. There was no aggravating factor as regards the duration of the infringement extending for more than six months. The duration of the infringement, if any, in the appellant’s view, is limited to the dates of the credit rating actions on 22 September 2015 and 26 November 2015. In this regard, the application of a methodology when issuing a credit rating has to be differentiated from monitoring a credit rating. Finally, ESMA failed to apply a mitigating factor as regards the appellant’s full transparency during the investigation process and its enabling of ESMA to establish the facts of the alleged infringement. 52.For the reasons summarised above and more widely discussed in the Notice of Appeal and in the appellant’s Reply to ESMA’s Response, the appellant asks the Board of Appeal not to confirm the decision taken by the Board of Supervisors (ESMA41-356-77 of 28 May 2020) and remit the case to the Board of Supervisors to adopt an amended decision based on the findings of the Board of Appeal. ESMA 53.ESMA acknowledges that the appeal is admissible and argues that it is unfounded for several reasons, that can be summarized as follows. 54.With regard to the infringement of Article 8(3) of the CRA Regulation, ESMA makes the following submissions. 55.ESMA did not err in law in the interpretation of the scope of application of Article 8(3) of the CRA Regulation. In ESMA’s view, the legal basis that ESMA used to impose the fine, viz Article 8(3) of the CRA Regulation, is correct. The scope of application of Article 5(1) of the Delegated Regulation No 447/2012 is aligned with the scope of application of the above Article 8(3) of the CRA Regulation and therefore does not go beyond the scope of the powers conferred on ESMA and the Commission in Article 21(4)(d) of the CRA Regulation. 56.From the textual, contextual and teleological reading of Article 8(3) of the CRA Regulation, it is obvious that this provision does not only cover the design stage

of a methodology but also its application stage. Indeed, (i) from a textual reading of Article 8(3) of the CRA Regulation, also complemented by Article 5(1) of the Delegated Regulation and by the corresponding infringement provision2, ESMA argues that it is self-evident that this provision points to the application of the rating methodologies, as it refers to their use. (ii) Likewise, from a contextual and combined reading of the different provisions in Article 8 of the CRA Regulation, it is also clear that Article 8(3) of the CRA Regulation is dedicated to both the design and the application of the methodologies. Whereas Articles 8(2a) and 8(7) of the CRA Regulation, which are mentioned by the appellant as possible alternative legal bases for the infringement described in the Contested Decision, have different aims and, therefore, are not applicable in the present case. The contextual and combined reading of the CESR guidance published on ESMA’s website3 also supports ESMA’s conclusion. (iii) A teleological reading of Article 8(3) of the CRA Regulation leads to the same conclusion. Article 8(3) of the CRA Regulation should be interpreted and applied in accordance with the objectives of integrity, transparency, responsibility, and good governance pursued by the CRA Regulation. ESMA is tasked with the regular examination of the compliance by CRAs with the requirements under Article 8(3) of the CRA Regulation. If ESMA’s assessment was limited to the design of the methodology and not its application (as suggested by the appellant), ESMA’s task would be pointless. 57.For all the above reasons, ESMA argues that the appellant’s allegations, including the ones regarding the alleged violation of the principles of legal certainty and legality, are all unfounded and should be dismissed. 58.ESMA further notes that the appellant did not apply its 2015 CB Methodology systematically in the formulation of all the covered bond ratings that it issued under that methodology. ESMA argues that the 2015 CB Methodology, voluntarily designed and adopted by the appellant, included a clear and unequivocal commitment to perform the cover pool analysis for all credit ratings issued under that methodology. It was only in July 2016, when the 2015 CB Methodology was amended, that it was defined as an optional step. Therefore, irrespective of whether or not the analysis of the cover pool was needed to achieve a AAA rating, according to its own methodology the appellant had to perform it for all credit ratings. 59.However, for the 559 unsolicited ratings issued in September and November 2015 the appellant did not perform such an analysis even though it did not have an objective reason for this deviation. Most of these ratings remained outstanding at least until the adoption of the 2016 CB Methodology in July 2016. Contrary to the appellant’s argument that the only reason for the absence of the cover pool analysis was its lack of importance for the issued ratings and not their
unsolicited/solicited nature, the evidence demonstrates that there is a clear causal link between the unsolicited nature of the ratings and the lack of 2 Point 43 of Section I of Annex III of the CRA Regulation. 3 CESR’s Guidance on common standards for assessment of compliance of credit rating methodologies with the requirements set out in Article 8.3 (the ‘CESR Guidance’).

performance of the cover pool analysis. This link is in a way recognised by the appellant in the press releases. 60.While ESMA acknowledges that a certain degree of judgement (notably, during the consideration of specific analytical factors) may be required when issuing ratings, such judgement is circumscribed by the methodology that the rating committees have to use to issue such ratings. Under the 2015 CB Methodology, the rating committees did not have any discretion as regards the building blocks. It is also clear from the text of the 2015 CB Methodology that the CB ratings in this methodology were defined as forward-looking opinions and therefore any reference to ‘stability’ in the methodology (which the appellant, ESMA argues, is seeking to link to the concept of ‘rating outlook’) must be read within this context. 61.Moreover, ESMA argues that the fact that the non-performance of the cover pool analysis did not have an impact on the result and quality of the ratings that the appellant issued under the 2015 CB Methodology is immaterial. The impact on the quality of the ratings is not a pre-condition for the establishment of the infringement. It only prevents the application of the corresponding aggravating coefficient, which in fact was not applied by the Board of Supervisors in the Contested Decision. 62.Lastly, ESMA notes that the appellant is trying to make a case of proportionality and of arbitrary and detrimental application of the CRA Regulation. However, ESMA applied the CRA Regulation in an objective way. Article 8 of the CRA Regulation applies to all registered CRAs, whether big or small, and the Board of Supervisors applied Article 8(3) of the CRA Regulation because, given the facts of the case, it was the right legal basis for the infringement committed by the appellant. ESMA precisely embraces the principle of undistorted competition between CRAs and therefore cannot offer preferential treatment when an infringement has taken place. Therefore, the arguments put forward by the appellant to try to demonstrate that it applied the 2015 CB Methodology systematically should also be dismissed. 63.ESMA further argues that the appellant did not have an objective reason for diverging from the 2015 CB Methodology. An objective reason justifying the divergence from the systematic application of a given rating methodology should be identified in an objective situation that arises during its application. Therefore, where, at the moment of the adoption of a methodology a potential situation is already known, as it is the case here, i.e. it is foreseeable that it would arise, this has to be integrated and explained in the methodology itself and, therefore, cannot be considered to be an objective reason to diverge from a methodology. 64.In the present case, the appellant considered – and consciously included in its 2015 CB Methodology – that the performance of the analysis of the cover pool for all highly and lowly rated issuers was not only
relevant but also necessary. Although there is no legal obligation to provide for a rating buffer, the appellant nonetheless committed to conduct the cover pool analysis in all cases to ensure the stability of the (forward-looking) CB ratings. Therefore, the fact that a AAA

rating could already be achieved before the performance of the cover pool analysis cannot be considered to constitute an objective reason to diverge from the methodology. 65.Lastly, ESMA reiterates that the fact that the quality of the ratings was not impacted by the non-performance of the cover pool analysis is immaterial because the impact on the quality of the ratings is not a pre-condition to establish the infringement and therefore cannot play a role in the determination of the existence of an objective reason to diverge either. 66.Similarly, the fact that the lack of performance of the cover pool analysis was disclosed to the public through the press releases cannot legitimise diverging from a methodology. Therefore, the arguments of the appellant trying to demonstrate the existence of an objective reason should also be dismissed. 67.As regards the appellant’s claim that it did not act negligently, ESMA argues that the appellant in fact acted negligently. The Board of Supervisors conducted a thorough assessment of the evidence gathered by ESMA’s Supervision Department and by the IIO throughout the entire investigation. 68.Part of the evidence even pointed to the existence of intention in committing the infringement. However, far from acting to the detriment of the appellant, the Board of Supervisors considered that the overall factual background was not sufficient to establish that the infringement was committed intentionally and, therefore, turned itself to assess whether the infringement had been committed with negligence. 69.With regards to the existence of negligence, the Board of Supervisors considered that — as confirmed by the Board of Appeal in the appeals of Svenska Handelsbanken AB, Skandinaviska Enskilda Banken AB, Swedbank AB and Nordea Bank Abp against ESMA’s decision (the Nordic Banks case)4 — taking into account the duty of special care required from professional firms who are used to having to proceed with a high degree of caution, and the nature and extent of the requirements imposed on CRAs under the CRA Regulation, a high standard of care was expected from the appellant. 70.Moreover, whether a natural or legal person is acting negligently has to be determined in light of all the circumstances of the specific case, taking into account, as part of the assessment, any specific acts or omissions identified, which, taken separately or as a whole, would amount to negligence. 71.The circumstances around the drafting and the adoption of the 2015 CB Methodology showed that, from the outset, the appellant demonstrated a clear lack of care. In particular, based on the evidence in the file, the Board of Supervisors found that the appellant already knew, during the drafting stage, that it would not be able to perform such assessments in all cases, notably due to the lack of sufficient public information on the cover pools (Contested 4 Ref. BoA D 2019 01, BoA D 2019 02, BoA D 2019 03 and BoA D 2019 04.

Decision, Annex I, para. 323). For instance, in response to the IIO, the appellant indicated that “(…) in light of the small amount of public information for all possible assets in a cover pool, (…) the analysts were aware that at the initial stage of establishing a pool of CB ratings it would not seem appropriate to start with the most complex covered bond structures” (paragraph 220 of the ESMA Response). Further examples of the evidence on negligence gathered are provided in paragraphs 218 to 221 of the ESMA Response. 72.As the design and the application of a methodology are strictly intertwined, the evidence demonstrating the existence of negligence during the drafting and adoption of the 2015 CB Methodology necessarily implies, due to a cascade effect, that the appellant did not care whether the methodology was going to be correctly applied thereafter. 73.Nevertheless, the Board of Supervisors did not focus its assessment exclusively on those aspects, but also took into account circumstances that transpired during the application, such as for instance the fact that the appellant did not carry out any assessment of what could constitute an objective reason to diverge from the systematic use of a methodology or that it did not contact ESMA to check in advance whether the reason to not perform the cover pool analysis could be considered as an objective reason. The lack of a careful assessment was confirmed by the appellant in its response to the second request for information of the IIO and is also demonstrated by the fact that the rating committee memoranda that were prepared for the unsolicited ratings that were issued in September and November 2015 are silent on this point (paragraph 223 of the ESMA Response). Moreover, taking into account that the wording of Article 8(3) of the CRA Regulation is clear and that there was also guidance publicly available on ESMA’s website, the appellant knew or ought to have known that the Board of Supervisors would interpret Article 8(3) of the CRA Regulation as also covering the application of the methodology. 74.ESMA further argues that, contrary to the appellant’s allegations that it acted according to an alleged common market standard/understanding, the ‘2015 ECBC European Covered Bond Fact book’ issued by the European Covered Bond Council - ECBC (i.e. a platform representing over 95% of the cover bond issuers in the EU) in 2015 shows that such an alleged market understanding actually did not exist. 75.Moreover, the fact that the press releases indicated that for the unsolicited ratings the appellant had not analysed the cover pools, cannot legitimate an incorrect action nor can it prove the good faith of the appellant, as the press releases do not serve the same purpose and cannot replace and substitute or supersede a rating methodology. Therefore, the appellant negligently infringed
Article 8(3) of the CRA Regulation.
76.ESMA also concludes that the calculation of the fine is correct. To fix the amount of the fine related to a specific infringement, ESMA must use the methodology set out in Article 36a of the CRA Regulation.

77.In the present case, the infringement started in September 2015, when the first unsolicited ratings without a cover pool analysis were issued by the appellant – in clear contravention of the 2015 CB Methodology – and lasted until 22 July 2016, when the 2015 CB Methodology was revised. In calculating the duration of the infringement, ESMA, in full consistency with the nature of the infringement that was linked to the application, took into account the overall period during which there was a clear mismatch between the 2015 CB Methodology (submitted to the supervisory authority and announced to the public), which foresaw the performance of a thorough cover pool analysis, and the content of the unsolicited ratings issued under that methodology (which did not include such an analysis). The unsolicited ratings were issued in September and November 2015 and many of them remained outstanding at least until the revision of the 2015 CB Methodology on 22 July 2016 (when de facto they became aligned with the new version of the CB Methodology). Therefore, since the infringement lasted for more than six months, the aggravating factor for duration applies. 78.The Board of Supervisors, also in line with its past practice, considered the ‘preceding business year’ that has to be taken into account to decide the basic amount of the fine under Article 36a(2) of the CRA Regulation and the legal cap in accordance with Article 36a(4) of the CRA Regulation to be the last business year preceding the end of the infringement (i.e. 2015). 79.Lastly, the appellant cannot benefit from the application of the mitigation factor for bringing quickly, effectively, and completely an infringement to ESMA’s attention because it has never recognised having committed an infringement. Therefore, it cannot be considered to have brought it to ESMA’s attention. In addition, in order to benefit from the application of this mitigating factor, a CRA must go beyond its legal obligations, such as the obligation to explain in a press release the key elements of a rating when announcing it to the public. Therefore, the Board of Supervisors correctly calculated the fine. 80.With regards to the infringement of Article 8(5a), 8(6)(aa) and 14(3) of the CRA Regulation, ESMA makes the following submissions. 81.ESMA did not err in law in the interpretation of the scope of the concept of ‘material changes’ and in finding that the changes introduced by the appellant to its 2015 CB Methodology were material for the purposes of the CRA Regulation. 82.The Board of Supervisors assessed the changes that the appellant introduced to the 2015 CB Methodology and concluded that the two intrinsically linked conditions for the application of Article 8(5a) of the CRA Regulation (i.e. materiality and potential impact on ratings) were met. Since it is sufficient that an impact could (at least in theory) exist as a result of a change, once the materiality of a change is established it is very likely that the
requirement of a potential impact is also met. 83.With regards specifically to the materiality of a change in a methodology, the Answer 7 of the Q&A on Implementation of the Regulation (EU) No 462/2013

on Credit Rating Agencies (the “Q&A”) – to which the appellant makes reference – only provides a non-exhaustive list of the cases where changes to specific elements in a methodology (such as changes to key criteria, key rating assumptions, key variables or driving factors) would, under any circumstances, be considered as material. However, whether a change is to be considered as material for the purposes of the CRA Regulation has to be assessed on a caseby-case basis and taking into account the importance or consequence of those changes for the methodology in question. The materiality of a change must thus be assessed considering the methodology that is subject to those changes. Whether the amendments were aimed at aligning the methodology with what the CRA was already doing in practice is irrelevant and should not be used as benchmark. 84.Contrary to the appellant’s allegations that the changes that it introduced to the 2015 CB Methodology were not material changes but mere clarifications, it is clear from a comparative analysis of the 2015 and 2016 CB Methodologies that the main changes were related to the performance of the cover pool analysis and were principally aimed at changing the way in which such analysis had to be performed. Moreover, the 2016 CB Methodology included an entirely new
appendix with completely new information describing how the covered bond
risk assessment was to be performed. The mere fact of having included such an appendix in the 2016 version of the CB Methodology should have been sufficient for the appellant to launch the procedure laid down in Article 8(5a), 8(6) and 14(3) of the CRA Regulation. Therefore, the arguments put forward by the appellant should be dismissed. 85.ESMA further claims that, also in this respect, the appellant acted negligently In ESMA’s view, some of the statements of the appellant raise doubts regarding the appellant’s willingness to comply with Articles 8(5a), 8(6)(aa) and 14(3) of the CRA Regulation and could be even considered to be constitutive of intent in committing the infringement. However, in the absence of further elements, in the Contested Decision, the Board of Supervisors did not conclude on the intent, but it did establish that the appellant had acted with negligence. In fact, ESMA argues that the lack of care demonstrated by the appellant is particularly striking in this case. 86.Despite the fact that the Independent Review Function (IRF) considered that the changes could be material, the appellant never performed and documented a thorough assessment of the materiality of the changes that were going to be introduced in the 2015 CB Methodology, before dismissing the possibility that they could be material just because no credit rating was actually impacted by those changes. The appellant’s conclusion was reached after an extremely limited and careless analysis of the non-exhaustive list of examples provided in Answer 7 to the Q&A. 87.In addition, despite the existence of internal conflicting views and the ongoing contacts with ESMA, the appellant never raised any question to ESMA in order to ascertain whether the changes should have been considered as material. Therefore, in ESMA’s view, the appellant negligently committed the infringements of Article 8(5a), 8(6)(aa) and 14(3) of the CRA Regulation.

88.ESMA concludes, also in this regard, that the calculation of the fine is correct. In respect to the appellant’s allegation, it is to be noted that the notification to ESMA of changes to a methodology following the annual review (in accordance with Article 8(5) of the CRA Regulation) and referring to the newly adopted methodology within the context of the periodic information to be submitted to ESMA, are not sufficient for the appellant to benefit from the application of the mitigating factor for bringing quickly, effectively and completely an infringement to ESMA’s attention. As explained in the Contested Decision, the appellant has never recognised that it committed an infringement. On the contrary, it has always challenged that the changes to the 2015 CB Methodology were material. Therefore, it cannot be considered to have brought the infringement to ESMA’s attention. 89.Lastly, the voluntary measures that the appellant took until the adoption of the Contested Decision by ESMA were not sufficient to ensure that a similar infringement could not be committed again. Therefore, the corresponding mitigation factor does not apply. Therefore, in ESMA’s view, the Board of Supervisors correctly calculated the fine to be imposed for the infringements of Articles 8(5a), 8(6)(aa) and 14(3) of the CRA Regulation. 90.In its Rejoinder ESMA stresses first, as regards the infringement of Article 8(3) of the CRA Regulation, that this is a case about a CRA which, without having an “objective reason”, deviated from the methodology that it had itself established, submitted to ESMA (its supervisor) and published on its website giving the impression to the public that an in-depth cover pool analysis would be performed in all covered bond ratings. In ESMA’s view, accepting such behaviour from a CRA would lead to the nullification of methodologies. ESMA further notes, as regards its interpretation of Article 8(3) as covering not only the design of methodologies but also their use, that it has followed the methods established by the Court of Justice of the EU (“CJEU“) for the interpretation of provisions of Union law; in so doing, it has not contravened the Meroni doctrine or infringed the proportionality principle. 91.ESMA further stresses in its Rejoinder that the appellant has not replied to some of the important points raised by ESMA in its Response, such as for instance the fact that in the 2015 CB Methodology the covered bonds ratings were described as ‘forward-looking ratings’ or the fact that there was no ‘common market understanding’ with regards to whether analysing the cover pool was needed. 92.As to the existence of an “objective reason to diverge”, ESMA reiterates that if a potential situation is already known at the moment of the adoption of a methodology, it has to be integrated and explained in the methodology itself. However, in ESMA’s view, it is clear from the evidence that at the moment of the adoption of the 2015 CB Methodology, the appellant
already knew that there would be situations where the Issuer Credit Strength Rating (ICSR) and the legal framework and resolution regime would be strong enough to support a AAA rating without undertaking a cover pool analysis. Therefore, this situation was not unpredictable or exceptional and, therefore, it could

have been reflected in the 2015 CB Methodology (as the appellant in fact did in 2016 when it revised the 2015 CB Methodology). 93.Furthermore, ESMA notes in the Rejoinder that it is also clear from the evidence that the appellant knew that the publicly available information on the cover pools (as a result of the entry into effect of the Capital Requirements Regulation No 573/2013 and the additional transparency requirements needed to obtain the European Covered Bond Council (ECBC)’s ‘Covered Bond Label’) was not sufficient to perform an in-depth qualitative and quantitative cover pool analysis. In fact, even after the unsolicited ratings of September and November 2015 were issued by the appellant, the situation as regards publicly available disclosures was not different. Despite these circumstances, during the design and adoption phase of the 2015 CB Methodology, the appellant decided that it would perform an in-depth qualitative and quantitative cover pool analysis in all cases and prominently indicated this in the 2015 CB Methodology. However, in the application phase, the appellant decided to diverge from its 2015 CB Methodology and purposely selected a series of jurisdictions and issuers that would allow it to reach AAA rating levels based on the credit differentiation provided by the ICSR together with the legal framework and the resolution regime, without performing the in-depth cover pool analysis prescribed in the 2015 CB Methodology. Therefore, ESMA concludes, also to support its finding of the appellant’s negligence, that the evidence in this case demonstrates that: (i) the appellant knew that the publicly available information on the cover pools was not sufficient to perform an in-depth qualitative and quantitative cover pool analysis and yet decided to include the performance of such an in-depth cover pool analysis for all credit ratings (thus also for unsolicited ratings issued without the interaction of the issuers) in the 2015 CB Methodology; and (ii) the appellant knew that in specific cases AAA ratings could be reached without the analysis of the cover pool and yet in the 2015 CB Methodology decided not to exclude its performance in those cases.
94. As to the calculation of the fine for the infringement of Article 8(3) of the
CRA Regulation, ESMA notes in the Rejoinder that where ratings are issued in breach of the obligation to apply a rating methodology in a systematic manner, the infringement of Article 8(3) of the CRA Regulation has to be considered as lasting until such ratings are either reviewed or withdrawn. Therefore, if overall the infringement lasts more than six months, the aggravating factor for duration needs to be applied. Furthermore, ESMA notes that the fine imposed by the Board of Supervisors for this infringement already reflects the fact that the appellant voluntarily took remedial actions (i.e. a EUR 200,000 reduction of the fine). 95.On the infringement of the third subparagraph of Article 14(3), Article 8(5a) and 8(6)(aa) of the CRA, ESMA upholds, in its Rejoinder, all the arguments put forward in the Response.

96.For all the above reasons, ESMA requests the Board of Appeal to dismiss the Appeal as entirely unfounded and to confirm the Contested Decision in its entirety. Discussion by the Board of Appeal of the parties’ contentions 97.The Board of Appeal is part of the governance structure of ESMA (and the other European Supervisory Authorities of which it is a joint body under their founding regulations). The members are required to be independent in making their decisions and undertake to act independently and in the public interest (Article 59 of the ESMA Regulation). 98.As an appeal body of ESMA, the Board of Appeal must decide whether the decisions of the Board of Supervisors were correct or not and may confirm the decisions or remit the cases to the Board of Supervisors (Article 60(5) of the ESMA Regulation). 99.This is a complex appeal, and the parties have filed a considerable volume of written submissions, with supporting documents, and have exercised their rights under Article 60(4) of the ESMA Regulation to make oral representations. The initial submissions of both parties, the appellant’s Reply and the ESMA’s Rejoinder, as well as the oral representations at the hearing (including related powerpoint presentations) were all particularly helpful in illuminating the issues and have been carefully considered in detail by the Board of Appeal.
100. The Board of Appeal acknowledges the high quality of the submissions
presented by both parties. All the parties’ contentions have been taken into account, whether expressly referred to herein or not.
101. The findings of the Board of Appeal are as follows.
I – Grounds of appeal regarding the part of the Contested Decision concerning the infringement of Article 8(3) of the CRA Regulation (a) Whether or not ESMA erred in law in the interpretation of the scope of application of Article 8(3) of the CRA Regulation.
102. Article 8 of the CRA Regulation is as follows:
Article 8
Methodologies, models and key rating assumptions

A credit rating agency shall disclose to the public the methodologies,
models and key rating assumptions it uses in its credit rating activities as defined in point 5 of Part I of Section E of Annex I.
A credit rating agency shall adopt, implement and enforce adequate
measures to ensure that the credit ratings and the rating outlooks it issues are based on a thorough analysis of all the information that is available to it and that is relevant to its analysis according to the applicable rating methodologies. It shall adopt all necessary measures so that the information it uses in assigning credit ratings and rating outlooks is of sufficient quality and

from reliable sources. The credit rating agency shall issue credit ratings and rating outlooks stipulating that the rating is the agency’s opinion and should be relied upon to a limited degree. 2a. Changes in credit ratings shall be issued in accordance with the credit rating agency’s published rating methodologies.
3. A credit rating agency shall use rating methodologies that are rigorous,
systematic, continuous and subject to validation based on historical experience, including back-testing.
4. Where a credit rating agency is using an existing credit rating prepared by
another credit rating agency with respect to underlying assets or securitisation instruments, it shall not refuse to issue a credit rating of an entity or a financial instrument because a portion of the entity or the financial instrument had been previously rated by another credit rating agency. A credit rating agency shall record all instances where in its credit rating process it departs from existing credit ratings prepared by another credit rating agency with respect to underlying assets or securitisation instruments providing a justification for the differing assessment.
5. A credit rating agency shall monitor credit ratings and review its credit
ratings and methodologies on an ongoing basis and at least annually, in particular where material changes occur that could have an impact on a credit rating. A credit rating agency shall establish internal arrangements to monitor the impact of changes in macroeconomic or financial market conditions on credit ratings. Sovereign ratings shall be reviewed at least every six months. 5a. A credit rating agency that intends to make a material change to, or use, new rating methodologies, models or key rating assumptions which could have an impact on a credit rating shall publish the proposed material changes or proposed new rating methodologies on its website inviting stakeholders to submit comments for a period of one month together with a detailed explanation of the reasons for and the implications of the proposed material changes or proposed new rating methodologies.
6. Where rating methodologies, models or key rating assumptions used in
credit rating activities are changed in accordance with Article 14(3), a credit rating agency shall:
(a) immediately, using the same means of communication as used for the distribution of the affected credit ratings, disclose the likely scope of credit ratings to be affected; (aa) immediately inform ESMA and publish on its website the results of the consultation and the new rating methodologies together with a detailed explanation thereof and their date of application; (ab) immediately publish on its website the responses to the consultation referred to in paragraph 5a except in cases where confidentiality is requested by the respondent to the consultation; (b) review the affected credit ratings as soon as possible and no later than six months after the change, in the meantime placing those ratings under observation; and (c) re-rate all credit ratings that have been based on those methodologies, models or key rating assumptions if, following the review, the overall combined effect of the changes affects those credit ratings.
7. Where a credit rating agency becomes aware of errors in its rating
methodologies or in their application it shall immediately:

(a) notify those errors to ESMA and all affected rated entities explaining the impact on its ratings including the need to review issued ratings; (b) where errors have an impact on its credit ratings, publish those errors on its website; (c) correct those errors in the rating methodologies; and (d) apply the measures referred to in points (a), (b) and (c) of paragraph 6
103. At the heart of this appeal is whether the requirement under Article 8(3)
that credit rating agencies shall use rating methodologies that are systematic implies also that the use or application of the methodologies must be systematic.
104. The Board of Appeal considered all the arguments raised by both parties
on this issue. It finds that Article 8(3) requires that credit rating agencies use rating methodologies that are systematic, and that credit rating agencies use (or apply) methodologies systematically. It is of the view that the notion of “systematic” in the context of Article 8(3), and also of the CRA Regulation generally, cannot be dis-associated from the notion of application or use; to do so would render Article 8(3) devoid of meaning. This would, in the Board of Appeal’s view, defeat the intention of the EU legislature as it would render ESMA powerless to supervise the ongoing application by rating agencies of methodologies in a systematic and consistent way. Specifically, it finds that a rating methodology is systematic (and continuous, as also required by Article 8(3) of the CRA Regulation) when it is designed in such a way that it can be applied systematically and continuously, unless there are exceptional reasons to deviate from it, and that these requirements for the rating methodology refer both to the methodology’s design (in the ex ante supervisory examination) and its use (in the ex post supervisory examination).
105. This conclusion is supported, in the first place, by the literal meaning of
Article 8(3), also considering its versions in the other official languages (see
e.g. the German version: “ Eine Ratingagentur wendet Ratingmethoden an, die streng, systematisch und beständig sind”, the French version: “Les agences de notation de crédit utilisent des méthodes de notation rigoureuses, systématiques, sans discontinuités”; The Italian version : “Un’agenzia di rating del credito utilizza metodologie di rating rigorose, sistematiche, continuative”; the Spanish version : “Las agencias de calificación crediticia emplearán métodos de calificación que sean rigurosos, sistemáticos, continuados”; the Portuguese version: ”As agências de notação de risco devem utilizar metodologias de notação rigorosas, sistemáticas e contínuas”). A similar text is also used in recital (23) of the CRA Regulation. In the Board of Appeal’s view, it is clear from the literal meaning of the provision of Article 8(3) that the rating methodology must be considered not only in its design but also in its use. This is confirmed by the initial words of Article 8(3), which sets out that “A credit rating agency shall use rating methodologies that are rigorous, systematic, continuous” (emphasis added), and by point 43 of Section I of Annex III of the CRA Regulation (“a credit rating agency infringes Article 8(3) by not using rating methodologies that are (…) systematic”). Similarly, recital 23 provides that “credit rating agencies should use rating methodologies that are rigorous, systematic, continuous (…). Further, according to the Cambridge Dictionary,

systematic means “using an organized method that is often detailed” and continuous means “without a pause or interruption”. The notion of use is, accordingly, on a clear reading of the text, intrinsic to the Article 8(3) requirement that credit rating agencies shall use methodologies that are systematic; this translates into a requirement that not only the design of the methodology, but also its use, must be systematic and continuous. Furthermore, recital 34, by requiring that credit rating agencies “should ensure that methodologies (…) used (emphasis added) for determining credit ratings are properly maintained, up-to-date and subject to a comprehensive review”, clearly implies that rating methodologies must be used as described and must be maintained and updated in consideration of their use. The design and the use of the methodologies are, therefore, equally relevant to the purpose of assessing the methodology’s compliance with the necessary qualifications set out in Article 8(3). Finally, the Board of Appeal notes that, under Article 8(3), a rating agency must use methodologies that are rigorous, systematic, continuous and subject to validation based on historical experience, including back-testing (emphasis added). The Board of Appeal notes that if a methodology is not applied or used systematically, there is no way to back-test the results that follow from the methodology. Accordingly, there is no way, if design is disassociated from use in the interpretation of Article 8(3), of using methodologies that are “systematic…….and subject to validation based on historical experience, including back-testing”. The provision would be inoperable as regards the back-testing element and this would defeat the clearly-expressed intention of the co-legislators. The Board of Appeal is therefore of the view that the insertion of back-testing into Article 8(3) further underlines the inseparability of the design of the methodology and its application in the structure of Article 8(3).
106. This textual interpretation is supported by the 2010 CESR guidance on
common standards for assessment of compliance of credit rating methodologies. The Board of Appeal finds these standards to be of strong persuasive force, as authoritative guidance in force since 2010, and applied by ESMA in its implementation of its supervisory mandate over credit rating agencies. The guidance provides that credit rating agencies, to demonstrate that methodologies are systematic, must have procedures so that “pre-defined methodologies are applied consistently (emphasis added) in the formulation of ratings in a given asset class, or appropriate records of the reasons why a rating has diverged from the pre-defined methodology are kept” (paragraph 42). The guidance also notes that the requirement for methodologies to be “continuous” means that “rating methodologies should remain globally consistent and appropriate over time unless there is a robust reason for not treating the methodology consistently” (paragraph 50). In other words, as also rightly acknowledged by the expert opinion submitted by the appellant (at paragraph 52), the rating methodology must, in principle, “be followed consistently and implemented as planned”. To do so, the methodology must, in the first place, be designed in such a way that it can be conducive to such consistent application. But once it is designed as systematic, the methodology also has to be applied consistently with its systematic character, because a systematic design not followed by a consistent application would simply prove illusory and would circumvent the requirement that the credit rating agency must use a

methodology which is systematic (and continuous). The Board finally notes that the CESR Guidance highlights that a rating agency is responsible for demonstrating compliance with the requirements on an ongoing basis (paragraph 7). This requirement, which reflects the foundational obligation on rating agencies to comply at all times with the conditions for initial registration (CRA Regulation, Article 14(3)), together with the CESR Guidance read as a whole, reinforces the clear textual implication from Article 8(3) that ongoing application/use cannot be dis-associated from design.
107. In a related point, the Board of Appeal does not consider that this textual
interpretation may be leading to a risk of circularity, as the expert opinion submitted by the appellant suggests (at p.63: “If a rating methodology would only be classified as systematic within the meaning of Article 8(3) of the CRA Regulation and Article 5 of the Delegated Regulation No 447/2012 if it were also systematically applied, this would mean that a rating methodology could never be classified as systematic when it was developed”). Indeed, this argument from the appellant seems to neglect, in the Board of Appeal’s view, that, at the stage of development of the methodology (and before its initial application), the adjectives “systematic” and “continuous” mean that the methodology must be designed in such a way that, once it is applied, the methodology’s application will be systematic and continuous. In other words, the methodology must be developed so as to ensure the possibility of its systematic and continuous application. The factual compliance with these requirements may be assessed by the supervisor (which is given the task of regularly examining the methodology’s compliance with Article 8(3) under
Article 22a(1) of the CRA Regulation) ex ante, before the application of the
methodology, by considering whether the design of the methodology is convincingly appropriate and apt to ensure the required outcome, and ex post, judging from its factual application.
108. This is also in line with the finality of the provision and its contextual,
teleological interpretation. It is indeed settled case-law that in interpreting a provision of EU law it is necessary to consider non only its wording but also, where appropriate, the context, in which it occurs and the objects of the rules of which it is part (judgment of 30 January 1997, C-340/94, de Jaeck and Staatsecretaris van Financien, ECLI:EU:C:1997:43, paragraph 17 and caselaw cited herein). The history, purpose and legislative intent of the CRA Regulation confirm the centrality of rating methodologies, and their consistent application, to ensuring a high quality of ratings, which was the aim of the inclusion of credit rating agencies into the European regulatory framework (see e.g., to this effect, recitals (1), (10), (11), (23), (25), (26), (34), (37) and (38) of the CRA Regulation).
109. As regards the context, the Board of Appeal notes that the CRA
Regulation is one of the pillars of the EU’s financial-crisis-era reform programme, and also that, in an indication of its importance, Article 8(3) and its strict requirements on rating methodologies was adopted as part of the first rating agency regulation in 2009. Weaknesses in, and failures of, methodologies were strongly associated with the financial crisis, and were among the main drivers of the CRA Regulation’s adoption. The 2008 Proposal,

from which Article 8(3) derives, states that the new regime has the objective of improving the quality of the methodologies used by credit rating agencies and the quality of ratings.5 In this context, and given in particular the dependence of rating quality on the application of methodologies, it is a core objective of the Regulation, as expressed in Article 8(3), to address the quality of ratings through review of methodologies and also their application in practice. Further evidencing the centrality of the quality of methodologies to the design of the CRA Regulation, Delegated Regulation No 447/2012, which implements Article 8(3), introduced regulatory technical standards (RTS) on rating methodologies. Furthermore, ESMA has adopted Guidelines on the validation and review of Credit Rating Agencies’ methodologies (Guidelines ESMA/2016/1575). The Guidelines provide for measures that ESMA typically expects a CRA to use in implementing the RTS, as well as examples of complementary measures which a CRA should consider appropriate.
110. The Board of Appeal also notes, by way of background, that ESMA has
highlighted that methodologies are critical to ensuring high quality ratings, while Article 8(3) has been recognised to be one of the most important requirements of the CRA Regulation, given the centrality of methodologies to the rating decision7. Limiting Article 8(3) to the design phase, and thereby disabling ESMA from supervising whether the application of a methodology was systematic, would, in the view of the Board of Appeal, defeat the clear objective of the CRA Regulation to ensure that rating activities are “conducted in accordance with the principles of integrity, transparency, responsibility, and good governance in order to ensure that resulting credit ratings used in the Community are independent, objective and of adequate quality” (recital 1). It would also defeat the Regulation’s specific incorporation of rules expressly designed to address methodologies.
111. Further, and considering Article 8(3) within the context of Article 8, which
covers different aspects of the regulation of methodologies, the Board of Appeal notes that Article 8 deploys a series of regulatory techniques, including disclosure and operational rules, alongside supervisory rules, to address the dependence of high-quality ratings on methodologies. It is constructed with a wide application and with regards to design and application. The Board of Appeal is of the view that it would accordingly defeat the design and purpose of Article 8 to limit Article 8(3), which is the central element of Article 8, to the design phase, and thereby cut out the critical application period during which methodologies have a live impact on ratings decisions. Were it otherwise, a credit rating agency could choose which of the elements of a methodology to apply, thereby defeating the objectives of the CRA Regulation as regards contributing to the quality of ratings issued in the Union, while achieving a high level of consumer and investor protection (CRA Regulation, Article 1). Allowing such a choice to a rating agency would also defeat the purpose of Article 8 which, inter alia, requires transparency to the market on methodologies and 5 COM(2008)704, 4. 6 ESMA Supervision, Annual Report 2019 and Work Programme 2020, 37. 7 See, eg, the view of some respondents to ESMA’s consultation on the development of Delegated Regulation No 447/2012, summarised in ESMA, Final Report, Draft RTS on the assessment of compliance of credit rating methodologies with the CRA Regulation (2011), 6.

their use (Article 8(1)) and which requires any changes to ratings to be issued in accordance with the credit rating agency’s published methodologies (Article 8(2a)). Consistency and predictability are accordingly core concerns of Article
8. Finally as regards the Article 8 context, the Board of Appeal is of the view
that Article 8(7) and Article 8(2a), which are relied on in argument by the appellant as evidencing the EU legislator’s intention to make a distinction between design and use/application, speak to different aspects of the Article 8 methodology regime and so do not qualify or restrict the scope of Article 8(3).
112. The Board of Appeal further notes, with regard to the structure of the
CRA Regulation, that the CRA Regulation, in Article 22a, makes express and distinct reference to ESMA’s supervisory obligations as regards examination of compliance with methodologies. Specifically, in its exercise of its ongoing supervision of credit rating agencies, ESMA shall examine regularly compliance with Article 8(3). The conferral of this targeted supervisory obligation on ESMA as regards methodologies, and the related reference to “examine regularly”, reinforces the Board of Appeal’s view that the systematic application and use of methodologies forms part of Article 8(3)’s requirement for methodologies to be systematic.
113. This interpretation is also in line with legislative history prior to the
adoption of the CRA Regulation. Article 8(3) of the CRA Regulation reflects, in the European context, a principle on the quality of rating which was already part of the 2004 IOSCO Code of Conduct Fundamentals for Credit Rating Agencies.
Article 1.2. of the Code set out that “the CRA should use rating methodologies
that are rigorous, systematic and, where possible, result in ratings that can be subjected to some form of objective validation based on historical experience”.
Article 1.3. of the Code, in turn, set out that “in assessing an issuer’s
creditworthiness, analysts involved in the preparation or review of any rating action should use methodologies established by the CRA. Analysts should apply a given methodology in a consistent manner, as determined by the CRA”.
114. The Board of Appeal finds therefore that, contrary to the appellant’s
claim, Article 8(3), and its qualifications “systematic” and “continuous”, apply both to the design and application of methodologies. The Board of Appeal also shares the argument put forward by ESMA in this respect, that a methodology is nothing but the description of what a credit rating agency is expected to do in practice, and that assessing compliance of the concept or design of a methodology alone, without assessing whether it is used effectively or not, is vain. The Board of Appeal further shares the view put forward by ESMA that the systematic application of a methodology does not imply the mechanistic application of the methodology and allows for an appropriate margin of judgment.
115. The Board of Appeal concludes, therefore, that ESMA’s interpretation of
Article 8(3) in the Contested Decision is correct.
116. The Board of Appeal is also of the view that this interpretation, being
grounded on solid arguments derived from the literal, contextual and teleological interpretations of Article 8(3), does not violate the fundamental

principles of legal certainty and legality, contrary to the argument raised by the appellant.
117. In this regard, the Board of Appeal notes that EU legislation benefits from
the presumption of legality, but acknowledges (and agrees) that, as is clear from the case-law of the CJEU, the principle that penalties must have a proper legal basis is a corollary of the principle of legal certainty, which constitutes a general principle of European law and requires in particular that any European legislation, when it imposes or permits the imposition of sanctions, must be clear and precise so that the persons concerned may know without ambiguity what rights and obligations flow from it and may take steps accordingly (see to this effect judgment 5 April 2006, T-279/02, Degussa v European Commission, ECLI:EU:T:2006:103, at paragraph 66). That principle, which forms part of the constitutional traditions common to the Member States, must be observed in regard both to provisions of a criminal nature and to specific administrative instruments imposing or permitting the imposition of administrative sanctions (judgment 5 April 2006, T-279/02, Degussa v European Commission, ECLI:EU:T:2006:103, at paragraph 67, and case-law cited herein). It is relevant in this appeal accordingly.
118. However, in the instant case, the Board of Appeal finds that credit rating
agencies, as addressees of the relevant provision, are in a position, on the basis of the clear wording of the relevant provision, to know or at least to reasonably foresee which acts or omissions would trigger the violation of Article 8(3) and make the credit rating agency liable for the administrative sanction provided for in Annex III of the CRA Regulation.
119. The Board of Appeal is further of the view that a rating agency, as a
professional market participant, could foresee that it could make a request to ESMA if it was unclear as to the scope of a provision of the CRA Regulation before taking any action which could amount to a breach of the relevant provision.
120. The Board of Appeal is, further, not persuaded that there was a material
ambiguity as regards the interpretation of Article 8(3) that would offend against the principle of legal certainty. The Board of Appeal shares with ESMA the view that it is not sufficient for a party to raise an alternative interpretation for legal certainty to be offended, as previously also determined by this Board of Appeal in the Nordic Banks case.
121. Further, it is well settled that the principles of legality and legal certainty
cannot be interpreted as precluding the gradual, case-by-case clarification of the rules by way of interpretation, provided that the result is reasonably foreseeable at the time the violation was committed. As the CJEU already clarified (see judgment of 22 October 2015, C-194/14, AC Treuhand v European Commission, ECLI:EU:C:2015:717, at paragraphs 42 and 43), the scope of the notion of foreseeability depends to a considerable degree on the content of the text in issue, the field it covers and the number and status of those to whom it is addressed; and a law may still satisfy the requirement of foreseeability even if the person concerned has to take appropriate legal advice

to assess, to a degree that is reasonable in the circumstances, the consequences which a given action may entail. This is particularly true in relation to persons carrying on a professional activity, and even more so for credit rating agencies whose professional activity requires registration or certification and, in the words of the CRA Regulation “play[s] an important role in global securities and banking markets, as credit ratings are used by investors, borrowers, issuers, and government as part of making informed investment and financing decisions” (recital 1 CRA Regulation). Credit rating agencies, therefore, are used to having to proceed with a high degree of caution when pursuing their activity and can therefore be expected to take special care in evaluating the risk that such an activity entails, including as regards compliance with relevant rules.
122. The CJEU also clarified that these principles apply also in the context
where courts had not yet had the opportunity to rule specifically on the relevant conduct, especially where – like in the instant case – the addressee of the sanction could have expected, if necessary after taking appropriate legal advice, its conduct to be declared in violation of the relevant provision (judgment of 29 March 2012, T-336/07, Telefonica v European Commission, ECLI:EU:T:2012:172, paragraph 323; judgment of 8 September 2016, T472/13, Lundbeck v. European Commission, ECLI:EU:T:2016:449, paragraph 763 and 767). The Board of Appeal is therefore of the view that the appellant could have been expected to take special care in evaluating the risks of its approach to Article 8(3) and of its approach to the application of the 2015 CB Methodology, particularly as regards the likelihood of non-compliance.
123. The Board of Appeal is also of the view that ESMA’s approach to the
supervision of Article 8, and of Article 8(3) specifically, was reasonably foreseeable and that interpretive material was available, including the 2010 CESR Guidelines, but also through Articles 5 and 6 of Delegated Regulation No 447/2012, as discussed in the following paragraphs. The Board of Appeal notes further, and by way of relevant background, that in its first major report on the supervision of credit rating agencies in 2012, ESMA noted that rating agencies should make sure that relevant control functions effectively contribute to ensuring the consistent application of credit rating methodologies.8
124. Our interpretation of Article 8(3) is further confirmed (as a
complementary argument but in no way as the only reason to justify such interpretation) by Article 5 (and, to some extent, also Article 6) of Delegated Regulation No 447/2012, which are as follows:
Article 5
Assessing that a credit rating methodology is systematic
1. A credit rating agency shall use a credit rating methodology and its
associated analytical models, key credit rating assumptions and criteria that are applied systematically in the formulation of all credit ratings in a given asset class or market segment unless there is an objective reason for diverging from it. 8 ESMA, Report on the Supervision of Credit Rating Agencies (2012), 10.

2. A credit rating agency shall use a credit rating methodology which is
capable of promptly incorporating the findings from any review of its appropriateness.
Article 6
Assessing that a credit rating methodology is continuous A credit rating agency shall use credit rating methodologies shall that are designed and implemented in a way that enables them to:
(a) continue to be used unless there is an objective reason for the credit rating methodology to change or be discontinued; (b) be capable of promptly incorporating any finding from ongoing monitoring or a review, in particular where changes in structural macroeconomic or financial market conditions would be capable of affecting credit ratings produced by that methodology; (c) compare credit ratings across different asset classes.
125. The Board of Appeal agrees with both parties that the interpretation of
the Delegated Regulation has to be in accordance (and in full conformity) with the CRA Regulation (which granted the European Commission in Article 21(4)(d) the delegated power to this effect) and that the Delegated Regulation may be used as a tool of interpretation. It is settled case-law that an implementing regulation must be given, if possible, an interpretation consistent with the basic regulation (however that case-law does not apply in the case of a provision of an implementing regulation whose meaning is clear and unambiguous and therefore requires no interpretation: judgment of 25 November 2009, T-376/07, Germany v. Commission, ECLI:EU:T:2009:467, paragraph 22 and the case-law cited herein; judgment of 28 February 2017, T157/14, JingAo Solar v Council of the European Union, ECLI:EU:T:2017:127, paragraph 151). The Board of Appeal further notes, however, that in so far as the provisions of the Delegated Regulation No 447/2012 are compliant with the principles laid down in the CRA Regulation, the Delegated Regulation may also supplement the CRA Regulation within the limits of the delegation conferred by
Article 21(4)(d) and Article 290 TFEU. In the instant case, the Board of Appeal
finds that, even without the need to consider the presumption of legality of the acts of EU institutions and the fact that only the CJEU can declare illegal such acts, Articles 5 and 6 of the Delegated Regulation No 447/2012 are fully aligned with Article 8(3) and within the limits of the delegation conferred to the European Commission.
126. Specifically, Article 5(1), which interprets the meaning of a “systematic”
methodology, links systematic to the use of the methodology in the formulation of all ratings in a given asset class or market segment (unless there is an objective reason not to), in compliance with the meaning and finality of Article 8(3) of the CRA Regulation. It makes clear, accordingly, that application is inextricably linked with design. The Board of Appeal notes, incidentally, that there is no material evidence of industry confusion on the meaning of “systematic” from the consultation ESMA carried out over the initial proposal

and development of Article 5(1). Indeed, specifically on Article 5(1), ESMA noted that most respondents to its proposed language for Article 5(1) (which was adopted in its entirety in the Delegated Regulation) did not foresee major issues with the text as regards systematic application.
127. The Board of Appeal therefore concludes on this point that the Contested
Decision is based on the correct interpretation of Article 8(3) and that ESMA correctly relied primarily, therefore, on such legal basis to properly substantiate its conclusions. Article 5(1) of the Delegated Regulation No 447/2012 was rightly used as a complementary source, without introducing any expansion of the scope of application of Article 8(3) of the CRA Regulation by way of application of the Delegated Regulation. (b) Whether or not the appellant applied its 2015 CB Methodology systematically in the formulation of all the covered bonds ratings that is issued under it.
128. The Contested Decision found that, although the 2015 CB Methodology
indicated that the cover pool analysis had to be performed for all credit ratings issued under that methodology, it was clear from the evidence gathered that the appellant did not conduct the cover pool analysis foreseen by the 2015 CB Methodology for the unsolicited CB ratings issued in September and November 2015, whereas it conducted the required cover pool analysis for the solicited ratings of May and July 2016.
129. The appellant argues that the application of methodologies is not a
mechanistic operation but requires the exercise of judgment, and that the systematic application of a methodology does not require a credit rating agency to carry out all the steps of the analytical rating process laid down in the methodology, where some of these steps are considered to be not relevant for the determination of the final rating in the individual case. In particular, in the instant case, the appellant claims that the lack of the cover pool analysis where the covered bonds being rated had already achieved AAA ratings did not constitute an infringement, because such an analysis would be irrelevant for the final result of the ratings. All the (unsolicited) ratings issued by the appellant at the time were correctly assigned and the result of the analysis would not have changed, even if a cover pool analysis had been carried out. The appellant in addition claims that the Contested Decision disregards the principle of proportionality. Furthermore, the appellant claims that the cover pool analysis required by the 2015 CB Methodology could only serve as a rating outlook and that the provisions of Article 8(3) of the CRA Regulation are not applicable to rating outlooks.
130. The Board of Appeal refers to paragraphs 3 to 9 of this decision for the
short description of the 2015 CB Methodology and notes that, in the express words of the 2015 CB Methodology:
(a) the Issuer Credit Strength Rating or ICSR is “the fundamental anchor point for the covered bond analysis”. Scope Ratings ICSR “represents a credit 9 ESMA, Final Report, Draft RTS on the assessment of compliance of credit rating methodologies with the CRA Regulation (2011), 7.

opinion on a bank’s ability to meet its contractual commitment on a timely basis, and in full, as a going concern” and it signals “the relative risk of a default-like event” (2015 CB Methodology, page 4, footnote 4). However, there are two additional supporting elements for the covered bond analysis which are taken in consideration by the 2015 CB Methodology and in particular (i) “the combination of the legal and resolution frameworks” as “the most important supporting element for the covered bond rating” (2015 CB Methodology, page 4), because “covered bond ratings for highly rated banks are driven primarily by the fundamental benefits of the regulatory framework applicable to banks and their covered bonds” and (ii) “the benefit of the cover pool [which] represents a second recourse coming after a chain of events affecting the issuer. On the latter, the 2015 CB Methodology is explicit in stating that: “The benefit of the cover pool is limited but it provides additional security and stability to the rating” (2015 CB Methodology, page 4). (emphasis added); (b) “covered bonds issued by high investment-grade-rated resolvable banks can exhibit a credit quality commensurate with AAA level, because of the covered bond status in the bail-in, regardless of the level of overcollateralization provided in their cover pool”. In turn, “the use of the cover pool to fulfil the payment obligations under the covered bond only becomes necessary when a resolution has failed and the issuer has defaulted”. However, the 2015 CB Methodology expressly states that: “However, a thorough analysis of the cover pool needs to be performed for all rated covered bonds. The findings inform us on how specific features of the covered bond structure, as well as other country specific aspects, may affect the probability of default and the loss given default. It also provides information on the likely rating sensitivity resulting from it” (2015 CB Methodology, page 9). (emphasis added).
131. In light of these provisions of the 2015 CB Methodology, the Board of
Appeal finds that the 2015 CB Methodology clearly required the cover pool analysis in all cases, including where the covered bond already merited a AAA rating, without introducing any distinction between solicited and unsolicited ratings. The Board of Appeal further finds that, under the 2015 CB Methodology, the cover pool analysis for AAA ratings should have remained
part of the overall rating assessment because it was described by the 2015 CB
Methodology as a necessary step for all ratings and, as already noted, in the words of the 2015 CB Methodology “it provides additional security and stability to the rating” and “informs on how specific features of the covered bond structure, as well as other country specific aspects, may affect the probability of default and the loss given default. It also provides information on the likely rating sensitivity resulting from it.” The 2015 CB Methodology is, in the Board of Appeal’s view, unequivocal as regards the need for the cover pool analysis in all cases, including “covered bonds issued by high investment-grade-rated resolvable banks [that] can exhibit a credit quality commensurate with AAA level, because of the covered bond status in the bail-in”, although the methodology itself concedes that in such a case the cover pool analysis may have less impact and significance for the rating; the methodology shows, however, that also in these cases the information deriving from the cover pool analysis may prove helpful to inform on the probability of default and loss given default and on the rating sensitivity.

Relatedly, the appellant’s argument as to the cover pool analysis
becoming a non-mandatory rating outlook in circumstances where the analysis is not necessary because the highest rating has already been achieved is, in the Board of Appeal’s view, not made out. As already noted, the 2015 CB Methodology expressly required that the cover pool analysis be carried out in all cases and regardless of the rating already achieved. ESMA, in the Board of Appeal’s view, rightly emphasises that the appellant’s ratings according to the 2015 CB Methodology were forward looking and “the cover pool, as conceived by the appellant in the 2015 CB Methodology, added stability to the rating itself….and was not just a driver of the determination of the rating outlook” (ESMA Response, paragraph 171).
The Board of Appeal notes that methodology design is a proprietary
matter for rating agencies (Article 23 of the CRA Regulation clarifies that ESMA, the Commission or any public authority “shall not interfere with the content of credit ratings or methodologies”), albeit that it must be carried out within regulatory parameters. It was certainly up to the appellant to devise and adopt a different rating methodology for covered bonds which could better ensure, in the appellant’s expectations, its compliance with Article 8(3) and that could also be considered rigorous, systematic and continuous. That the appellant had reservations regarding the cover pool building block is clear from the evidence presented to the Board of Appeal, including as regards the necessity for a cover pool analysis where the issuer had already achieved the highest possible rating, and also as regards the availability to it of sufficient information to carry out the analysis in all cases. That the appellant could and should foresee and understand that it could design the methodology differently is also clear from its decision to amend the 2015 CB Methodology in 2016, with the aim of expressly excluding the cover pool analysis in those cases where the issuer had already achieved the highest possible rating or where insufficient information was available (as may have happened with the unsolicited ratings). This does not change the fact that the 2015 CB Methodology, as notified to ESMA and published to the market, expressly included the embedding of the cover pool analysis as a necessary and mandatory step for incorporating into the rating the second recourse expectations associated with covered bonds generally, and the rated bonds specifically. The cover pool building block of the 2015 CB Methodology could not, accordingly, be considered irrelevant and disregarded as such in the practical implementation of the 2015 CB Methodology, simply because the rating of the issuer in question was already of the highest quality, irrespective of the cover pool analysis. In other words, the cover pool analysis is presented by the 2015 CB Methodology as constituting a necessary part of the rating, and not of the rating outlook, and the 2015 CB Methodology also clarifies that the cover pool analysis may be of use to investors even in the case of covered bonds issued by high investment-grade-rated resolvable banks that can exhibit a credit quality commensurate with AAA level, because of the covered bond status in the bail-in. In the words of the 2015 CB Methodology itself, “a thorough analysis of the cover pool needs to be performed for all rated covered bonds. The findings inform us on how specific features of the covered bond structure, as well as other country specific aspects, may affect the probability of default and the loss given default. It also provides information on

the likely rating sensitivity resulting from it” (2015 CB Methodology, page 9). The cover pool analysis was, therefore, required to be carried out in each case. We are of the view that appellant’s different justificatory arguments fail accordingly.
134. The Board of Appeal finds therefore that the lack of the cover pool
analysis for the unsolicited ratings issued in September and November 2015 ran contrary to the requirements of Article 8(3) of the CRA Regulation and of the 2015 CB Methodology.
135. The Board of Appeal further holds that the Contested Decision, in finding
(and sanctioning) such an infringement, simply applied the relevant legal provisions in an objective manner, by examining the facts and ascertaining that these facts engaged Article 8(3) and the related infringement provision in point 43 of Section I of Annex III of the CRA Regulation. In so doing, ESMA could not have, in law, adjusted the application of the relevant provisions to the size or other special features of the appellant. ESMA had to apply the relevant legal provisions objectively and in a non-discriminatory way. Under Article 21, ESMA must ensure the CRA Regulation is applied. This requires ESMA to apply the relevant rules in a consistent manner and as they present themselves on the facts to specific instances. ESMA does not have a discretion to calibrate how rules apply to specific fact patterns depending on the status of the relevant party. The appellant’s claim that, in applying the rules, ESMA violated the proportionality principle is therefore unwarranted. Application of the proportionality principle does not imply that an otherwise applicable rule is disapplied because of the size or other characteristic of a party. The principle of legal certainty implies that rules must be applied in a consistent and evenhanded manner.
136. The Board of Appeal also notes that the proportionality principle is
already embedded within the CRA Regulation, which contains several provisions which, in an expression of the proportionality principle, are designed to exempt certain categories of rating agencies (see Article 6(3) to the effect of points 2, 5, 6,and 9 of Section A of Annex I and Article 7(4); Article 36a and
Annex IV). None of these provisions refers to a proportionate application of the
requirements of Article 8(3) of the CRA Regulation.
(c) Whether or not the appellant had an objective reason for diverging form the 2015 CB Methodology.
137. The appellant claims that the Contested Decision is mistaken because,
even if the lack of the cover pool analysis for unsolicited ratings of issuers with the highest possible rating in September and November 2015 was a divergence from the 2015 CB Methodology, this would be justified under Article 5 of the Delegated Regulation, which allows a divergence for an “objective reason”. The appellant argues that an objective reason is to be interpreted as a means for ensuring that any divergence is not arbitrary and is suitable, necessary, and appropriate in light of the objectives of the CRA Regulation. In this case, the concentration by the appellant on only relevant steps from the 2015 CB Methodology amounts to an objective reason: the approach taken by the

appellant provided for high quality ratings and economic efficiencies, as it only refrained from an analysis that was not necessary for the issued ratings. Adequate quality was ensured since the ratings in question were fully supported by the fundamental benefits of the regulatory framework; the assessment of the cover pool would not have made any difference to the issued ratings. The appellant further notes that the reason for the deviation can be found in two concomitant factors. First, in September and November 2015 Scope Ratings assigned credit ratings only to covered bonds of a subset of issuers, in particular the strongest banks in those countries with a strong covered bond framework. Second, the stable or positive outlooks assigned to the respective banks’ ratings provided insight that over the forward-looking period of 12-18 months covered by the outlook, none of the respective issuers were expected to become downgraded. The appellant further underlines that the press releases about the ratings issued under the 2015 CB Methodology for which no cover pool analysis was conducted were entirely transparent in informing the market as to the lack of cover pool analysis.

ESMA, in its Contested Decision and its submissions for this appeal,
argues that, in light of settled EU case law on the interpretation of EU law and, relatedly, the context and objectives of the CRA Regulation, the possibility of diverging from a methodology on the basis of objective reasons should be interpreted in a narrow way. Specifically, it argues that an objective reason justifying the divergence from the systematic application of a rating methodology could not consist of, as in the instant case, a potential situation already known at the moment of adoption of a methodology (because in this case, it should be integrated and explained in the methodology itself) but should be limited to an objective situation that occurs in the moment of the application of the methodology. In the instant case, the potential situation of highly-rated issuers was known to the appellant when developing the 2015 CB Methodology. Furthermore, a divergence can only be something which occurs in an exceptional circumstance, and not something that happens with regard to all, or a vast majority of the ratings in question and continuously for months, otherwise the rule becomes the divergence and not the compliance with the methodology that the appellant itself established and published (and in the case at hand, the ratings for which the cover pool analysis, foreseen in the 2015 CB Methodology, was not performed amounted to 559 ratings out of the 622 ratings assigned on the basis of the 2015 CB Methodology, i.e. almost 90 per cent of the ratings). ESMA is also of the view that press release disclosures as to the absence of an element do not allow a rating agency to disregard an element provided for in a methodology for any reason – otherwise the requirement for a systematic application of methodologies would have no substance.
The Board of Appeal notes that, according to Article 5(1) of Delegated
Regulation No 447/2012 “a credit rating agency shall use a credit rating methodology and its associated analytical models, key credit rating assumptions and criteria that are applied systematically in the formulation of all credit ratings in a given asset class or market segment unless there is an objective reason for diverging from it” (emphasis added). It notes further that the term “objective reason” is not defined. In its view, the correct interpretation can be supported by the text and context of the Delegated Regulation and of

the CRA Regulation, in accordance with settled principles of EU law statutory interpretation (judgment 4 July 2006, Konstantinos Adeneler e.a.v ELOG, C212/04, ECLI:EU:C:2006:443, paragraph 60). From the clear language of
Article 5(1) and its context, and in light also of its view of the need for the use
of a methodology to be systematic, the Board of Appeal supports the view taken by the Contested Decision that the CRA Regulation has as objectives ensuring the transparency of methodologies and rating agency processes, and that ratings of rating agencies subject to the CRA Regulation are of adequate quality and issued by rating agencies subject to stringent requirements (CRA Regulation, recitals 37 and 2); and that divergences from methodologies should be exempted only on narrow grounds. The Board is also of the view that, in light of the purpose of the CRA Regulation, and also in light of settled EU case law on the operation of exemptions (judgment 19 July 2012 A Oy, C-33/11, ECLI:EU:C:2012:482, paragraph 49), divergences from notified and published methodologies are to be regarded as exceptional and to be justified by situation-specific and reasoned explanations.
140. The Board of Appeal is further of the view that the final determination of
an objective reason is a supervisory one by ESMA as to whether or not the nature of the divergence is acceptable, given the purpose and objectives of the CRA Regulation. The objective reason qualification to Article 5(1) is accordingly, in the Board of Appeal’s view, a supervisory tool, designed to accommodate appropriate exceptions and to be applied to specific fact patterns. It therefore entails a necessary degree of supervisory appreciation. The Board of Appeal further notes that an appropriate margin of operational latitude must be afforded to ESMA’s technical appreciation, as the expert and independent supervisor of rating agencies. In line with past decisions, the Board of Appeal is indeed of the view that, in technical matters, the decision of ESMA, as a specialist regulator, is entitled to some margin of appreciation (FinancialCraft Analytics Sp. zo.o. v ESMA, BoA 2017 01, paragraph 45).
141. The Board of Appeal is therefore of the view that the determination by
ESMA as to whether or not a reason to diverge from a methodology, put forward by a credit rating agency, may be considered “objective” in the context of the application of the CRA Regulation entails in principle such margin of appreciation of ESMA’s Board of Supervisors. This need for a margin of appreciation is all the more compelling given the open-textured nature of the term “objective reason”. Since the Board of Appeal is not in functional continuity with ESMA’s Board of Supervisors (in the sense that, unlike other boards of appeal of European agencies, e.g. EUIPO, the Board of Appeal does not enjoy the same powers as the ESMA Board of Supervisors and there is not, thus, in the merit, full “continuity of its functions” with the agency decision-maker: see to this effect, a contrario, judgement of 12 December 2002, Procter & Gamble
v. OHIM, T-63/01, EU:T:2002:317, paragraphs 21-22), nor is it empowered to
second guess decisions of the Board of Supervisors which entail a margin of appreciation, the Board of Appeal’s review is then limited to verifying whether ESMA, in adopting its determination on this, (i) complied with all applicable procedural rules, (ii) duly stated its reasons, (iii) accurately stated the facts or (iv) committed a manifest error of assessment or a misuse of powers (see, among others, judgment 4 June 2015, Versorgungswerk der Zahnärztekammer

Schleswig-Holstein v. European Central Bank, T-376/13, EU:T:2015:361, paragraph 53; judgment 29 November 2012, Thesing and Bloomberg Finance v ECB, T-590/10, EU:T:2012:635, paragraph 43).
142. The Board of Appeal finds, on its review of the facts presented to it, that
there is no evidence that ESMA has adopted a decision that is not sound as regards procedure, reasons, and the application of relevant law and facts. The Board of Appeal notes in particular its view that the overall design of Article 8 of the CRA Regulation underlines the importance of consistency and predictability in the design and application of methodologies. We are, accordingly, of the view that it is reasonable for ESMA to apply qualifications to what can be regarded as an objective reason that are, inter alia, related to the inapplicability of facts of a nature that were obvious at the time of the methodology’s development and so could have been integrated into the model (notably, it was known in advance that the assessment of the rating quality of certain issuers could not have been strengthened by the cover pool building block).
143. Specifically, the Board of Appeal is of the view that ESMA rightly
emphasises in its Contested Decision that if the appellant did not find it relevant to conduct the cover pool analysis for highly-rated issuers, it should have made this clear in the methodology, instead of expressly committing to performing the assessment also for highly-rated issuers (Contested Decision, paragraph 268). We also note as regards the relevance of the cover pool analysis that ESMA further rightly emphasises that, especially in the case of a number of ratings where there was no rating buffer, that the cover pool analysis could have provided useful information but was not carried out (Contested Decision, paragraphs 270 to 274).
144. The Board of Appeal is further of the view that the appellant’s contention
that an objective reason could be applied to some 90% of the ratings would drain the objective reason of real content and undermine the foundational 2015 CB Methodology, contrary to the purpose and meaning of Article 8(3).
145. The Board of Appeal accordingly regards ESMA’s qualifications as to the
type of facts that, in the instant case, can ground an objective reason as representing a reasonable exercise by ESMA of its margin of supervisory appreciation, particularly in light of the need for exemptions to Article 8(3) to be narrow. This is particularly so in light of the centrality of consistent methodologies to the ratings process, and also given the expectations created on the initial disclosure of methodologies to the market as to how the methodologies were to be applied in practice. The Board finds accordingly, and with due regard to the margin of supervisory appreciation afforded to ESMA, that ESMA did not err in law or on the facts in deciding that the appellant did not have an “objective reason” for deviating from the 2015 CB Methodology.
146. The Board of Appeal is also persuaded by ESMA’s argument that public
disclosure as to the non-application of the cover pool building block does not serve as an objective reason for the omission of this step.

(d) Whether or not the appellant acted negligently
147. The appellant claims that, in any event, it did not act negligently and that
the Contested Decision is therefore mistaken where it finds that the infringement set out in point 43 of Section I of Annex III of the CRA Regulation was committed with negligence (Contested Decision, paragraphs 303-344).
148. ESMA, in the Contested Decision and in this appeal, relies on CJEU
case-law on the concept of negligence and on the duty of special care required from professionals. It considers negligence to be established in circumstances where the CRA, as a professional firm in the financial services sector subject to stringent regulatory requirements, is required to take special care in assessing the risks that its acts or omissions entail, and has failed to take that care; and as a result of that failure, the CRA has not foreseen the consequences of its acts or omissions, including particularly the infringement of the CRA Regulation, in circumstances when a person in such a position who is normally informed and sufficiently attentive could not have failed to foresee those consequences (Contested Decision, paragraph 313). It concludes, inter alia, that (i) the circumstances around the drafting and the adoption of the 2015 CB Methodology are relevant for the establishment of negligence, because they show at least a lack of care from the outset by the appellant, and (ii) the negligence in the moment of the application of the 2015 CB Methodology constituted the necessary consequence, due to a “cascade effect”, of the lack of care of the appellant in the process of drafting and adopting the 2015 CB Methodology (Contested Decision, paragraph 331). Moreover, ESMA notes that the appellant did not contact ESMA in advance to check whether there were reasons which could be considered objective to diverge from the systematic application of the 2015 CB Methodology and also did not carry out any assessment of what could be considered an objective reason for the purposes of the CRA Regulation and the Delegated Regulation No 447/2012, as is confirmed by the appellant’s response to the second request for information of the IIO and also by the rating memoranda prepared for the unsolicited ratings issued in September and November 2015, which were silent on this point (Contested Decision, paragraphs 332 to 335). ESMA also argues that the common market standards/understanding relied on by the appellant are not proven and would not, in any case, constitute a justification for the infringement or exclude the existence of negligence; and that the disclosure of the absence of a cover pool analysis in the relevant press releases issued by the appellant do not cure the earlier infringement and lack of care.
149. The appellant claims, on the contrary, that (i) any lack of care in the
drafting of the 2015 CB Methodology cannot be linked to the application of the same, which occurred months later, (ii) it acted according to purported generally accepted common market standards, as described in paragraphs 24 to 26 of the Notice of Appeal, (iii) it provided full transparency about the lack of the cover pool analysis with the press releases, and (iv) it could not have foreseen ESMA’s interpretation of Article 8(3) and, accordingly, that its approach would lead to an infringement, as in the appellant’s view, ESMA’s interpretation goes beyond the wording and no ESMA guidance existed at the time of the alleged infringement.

The Board of Appeal notes, in the first place, that under the CRA
Regulation ESMA can impose fines only for infringements which have been committed intentionally or negligently (Article 36a(1) of the CRA Regulation). It is not in dispute that the action was not committed intentionally. The question then arises as to whether the appropriate negligence standard was applied by ESMA to the facts. The concept of negligence is not defined in the CRA Regulation.
As to the standard that governs whether the appellant’s failure to apply
the 2015 CB Methodology in the relevant cases amounts to a negligent breach of the CRA Regulation, the Board of Appeal is of the view that the case law of the Court of Justice is of compelling clarity. The CJEU, in its judgment of 3 June 2008, C-308/06, The Queen on the application of Intertanko v Secretary of State for Transport, ECLI:EU:C:2008:312, paragraph 75 has held that all EU legal systems have recourse to the concept of negligence, which refers to an unintentional act or omission by which the person responsible breaches his duty of care. The CJEU has further held in paragraph 77 that “serious negligence” (the legislative concept at issue in that case) must be understood as entailing an unintentional act or omission by which the person responsible commits a patent breach of the duty of care which he should have and could have complied with in view of his attributes, knowledge, abilities and individual situation.
The EU courts have in addition made clear that a high standard is to be
expected of certain professional actors, including as regards the taking of advice and the assessment of risks. Specifically, in its judgment of 29 March 2012, T-336/07 Telefónica, SA and Telefónicade Espaňa, SA v Commission, ECLI:EU:T:2012:172, paragraph 323, the CJEU found that it can be expected that, in certain circumstances, appropriate legal advice is taken to assess, to a degree that is reasonable in the circumstances, the consequences that a given act may entail. The Court found that this is particularly true in relation to persons carrying on a professional activity, who are used to having to proceed with a high degree of caution when pursuing their occupation. They can, on that account be expected to take special care in assessing the risks that such an activity entails (ex multis: judgment of 28 June 2005, Dansk Rørindustri A/S, C189/02 P, ECLI:EU:C:2005:408).
Finally, the Board notes that the CJEU has held that an undertaking may
not escape imposition of a fine even where the infringement has resulted from that undertaking erring as to the lawfulness of its conduct on account of the terms of legal advice given by a lawyer (judgment of 18 June 2013, C-681/11, Schenker and Others, ECLI:EU:C:2013:404, paragraph 43).
The CJEU has further found that the negligence condition is satisfied
where the undertaking concerned cannot be unaware that it is infringing the relevant rule, with the further precision that the fact that the undertaking concerned has characterised wrongly in law its conduct upon which the finding of the infringement is based cannot have the effect of exempting it from the imposition of the related fine, in so far as it could not be unaware of the

infringement of the relevant rule (judgment of 26 October 2017, T-704/14, Marine Harvest v. European Commission, ECLI:EU:T:753, paragraphs 237- 238).
155. In addition, the absence of a precedent on the application of a specific
rule does not prevent an authority, such as ESMA, from finding an infringement (judgment of 8 September, Lundbeck, Case T-472/13, ECLI:EU:T:2016:449, paragraph 782).
156. The Board of Appeal accordingly finds that ESMA, in light of relevant
CJEU case law and the purpose and objectives of the CRA Regulation, correctly assessed the relevant standard of care imposed on the appellant as of being required to take special care in assessing the risks that its acts or omissions entailed and failing to take that care; and accordingly not foreseeing the consequences of its acts/omissions, particularly as regards infringements of the CRA Regulation, where a normally informed and sufficiently attentive person in the position of the appellant could not have failed to foresee those consequences.
157. The Board of Appeal in particular finds persuasive ESMA’s
determination that, in imposing a high level of care, regard should be had to the objectives and purpose of the CRA Regulation. The Board of Appeal further notes that EU-regulated rating agencies play an important role in financial markets (note recital 1 of the CRA Regulation) and, correspondingly, have special responsibilities in the issuance of ratings for regulatory use and that, as holders of a regulatory license via registration or certification to operate under the CRA Regulation, rating agencies are subject to a duty to undertake their activities with a high level of care, and appropriate and robust organisational safeguards including as regards their compliance with relevant rules. Undertaking a highly important (in global securities and banking markets) and regulated business as a rating agency is a voluntary determination, but once such business is undertaken, it brings with it specific regulatory duties, including with respect to assessing the risks of acts/omissions and foreseeing the consequences of acts/omissions, in particular as regards infringements of the Regulations.
158. The Board of Appeal further notes that it has previously accepted that a
high standard of care is expected of financial services providers and rating agencies. In the Nordic Banks case, and in relation to the facts of that appeal, the Board of Appeal stated that “ESMA rightly emphasises that financial services providers and CRAs play an important role in the economy of the EU, as well as in the financial stability and integrity of the financial markets and that “[a] high standard of care is to be expected of such persons”. (paragraph 285).
159. As to the application of the standard, the Board is of the view that the
determination of whether an act is committed negligently as against the relevant standard of care is a question of fact.
160. On its review of the facts presented to it, the Board of Appeal is of the
view that the appellant could reasonably have been expected to take special

care in assessing the risks of its acts and omissions in designing and applying the 2015 CB Methodology, specifically as regards Article 8(3) (and Article 5 (1) of the Delegated Regulation), but failed, on the evidence, to do so. Accordingly, it did not foresee the consequence of these actions being in breach of its regulatory obligations, in circumstances where a normally informed and sufficiently attentive person in the position of the appellant could not have failed to so foresee. The appellant could also reasonably have been expected to seek advice from ESMA, with which it is in a direct supervisory relationship, including as to whether objective reasons were in place to ground an exception to the requirement under Article 8(3) that the methodology was to be applied systematically.
161. Specifically, the Board is persuaded that ESMA’s finding of a negligent
breach of duty in the appellant’s failure to apply the 2015 CB Methodology systematically is based on adequate factual evidence and on a correct interpretation and application of the relevant legal provisions applicable to such facts, recalling in this regard what the Board of Appeal regards as a clear and unequivocal commitment by the appellant, in the 2015 CB Methodology, to carry out the cover pool analysis in all cases, as well as the Board of Appeal’s view that the meaning of Article 8(3) as regards the systematic use of methodologies was clear.
162. The Board is further persuaded, as a complementary but not primary
consideration, and given in particular its view that negligence is to be assessed with regard to all the circumstances relating to the breach of duty, that this negligent application of the 2015 CB Methodology arose as a “cascade effect” from an initial lack of care in the design of the 2015 CB Methodology (Contested Decision, paragraph 331), and that, given that the design and use of the Methodology are “strictly intertwined” (ESMA Response, paragraph 237), it was reasonable, accordingly, of ESMA to take into consideration circumstances relating to the adoption of the Methodology.
163. Finally, the Board finds it persuasive to the finding of a breach of duty
that the appellant failed to take advice from ESMA before issuing unsolicited ratings without the cover pool analysis, particularly as regards the potential existence and application of exempting objective reasons. The appellant was in a direct supervisory relationship with ESMA, and it was foreseeable to the appellant that the consideration of whether an objective reason was in place was highly relevant in the circumstances, given the omission of the cover bond analysis — the performance of which, in all rating cases, was clearly indicated in the 2015 CB Methodology. In these circumstances, it was foreseeable that advice in this regard could reasonably have been expected to be sought. This is even more so, if one considers that the ESMA clarifications, if timely sought, would not have prevented the appellant from issuing in the future unsolicited ratings without cover pool analysis for issuers with the highest rating; the request for clarification may have, perhaps, delayed the intended issuance of the unsolicited ratings without cover pool analysis but would have timely alerted the rating agency to the prior necessity to amend the 2015 CB Methodology, before issuing such ratings (something that the appellant did in 2016, but only after having disregarded the 2015 CB Methodology in September and

November 2015). There was therefore no significant risk, for the appellant, in seeking timely appropriate guidance from its supervisor; nor that the appellant, in doing so, would have been deprived of its right not to give evidence against oneself (assuming that, according to the Opinion of Advocate General Pikamae of 27 October 2020, C-481/19, DB v. Consob, ECLI:EU:C:2020:861, the right “nemo tenetur se detegere” could be invoked not only in the context of criminal sanctions but also of administrative sanctions, departing from settled case-law of the CJEU in its judgment of 18 October 1989, Orkem v Commission, C374/87, ECLI:EU:C:1989:387 and judgment of 18 October 1989, Solvay v Commission, C-27/88, ECLI:EU:C:1989:388). Indeed, a diligent behaviour would require that the necessary clarification from the supervisor is sought before issuing the ratings, when there is no risk of sanction, and not afterwards.
164. The Board also notes that the disclosure in the relevant press releases
of the absence of a cover pool analysis does not legitimise the breach, or serve to evidence that the appellant did, on the facts, take due care. Disclosure of a breach does not eliminate the breach or cure the prior lack of care in its commission. The Board of Appeal is of the view that the appellant infringed
Article 8(3) in failing to follow the 2015 CB Methodology and that, on the facts,
ESMA made an appropriate determination as to the existence of negligence in the failure to follow the 2015 CB Methodology. The disclosure to the market of the absence of the cover pool analysis is not relevant to the determination of either the breach, or a relevant fact as regards the determination of whether the breach was committed negligently. (e) Whether or not ESMA’s calculation of the fine is correct
165. The appellant claims that ESMA did not calculate the fine related to the
infringement of Article 8(3) of the CRA Regulation correctly. In particular, the appellant claims that the fine is not proportionate to the seriousness of the infringement committed by the appellant, because (i) ESMA erred in the calculation of the duration of the infringement (which ESMA considered existent) during the entire period that the 2015 CB Methodology was in place and governed the ratings issued without a cover pool analysis, i.e. from 22 September 2015 to 22 July 2016, whereas the appellant, arguing that ESMA’s interpretation represents a broad reading and calling for a narrower interpretation, contends that the alleged violations of the 2015 CB Methodology took place only over September to November 2015 (when the Methodology was applied to specific rating issuances), and also thereby offended the principle of proportionality, (ii) ESMA should have applied the mitigating factor set out at point III.3 of Annex IV of the CRA Regulation, notably as regards the appellant’s transparency in disclosing the absence of a cover pool analysis in the relevant press releases, (iii) the fine exceeds the 20% of annual turnover cap provided for in Article 36a(4) of the CRA Regulation, when that cap is applied, as the appellant argues it should be, against turnover figures for 2014 as the “preceding business year”, and not against 2015 (as applied by ESMA), and, as regards the calculation of annual turnover, ESMA erred in that it included all of the appellant’s activities in the calculation, and did not limit the turnover calculation to rating activities only, (iv) ESMA did not duly consider whether what the appellant regards to be a minor omission, which did not have an impact

on the relevant ratings and was disclosed to the market, and which did not jeopardise the objectives of the CRA Regulation, should lead to such a material fine - and thereby also breached the proportionality principle, and (v) the fine is excessive if compared to fines imposed for other infringements and to the revenues that the appellant derived from the 2015 CB Methodology.
166. ESMA argues that it calculated the fine correctly and in accordance with
Article 36a of the CRA Regulation. As regards the relevant time period and the
related application of the aggravating factor set out in point I.2 of Annex IV of the CRA Regulation, it argues that the Contested Decision adopted, as the starting date for the calculation of the duration of the infringement, the date on which the appellant issued the first CB ratings without an assessment of the cover pool (22 September 2015), but considered that the infringement lasted until the appellant adopted the 2016 CB Methodology, which was however more than six months after the starting date (and triggered therefore the aggravating factor set out in point I.2. of Annex IV of the CRA Regulation). It therefore, and in full consistency with the nature of the infringement, took into the account in the calculation of the time period of the infringement the overall period during which there was a mismatch between the 2015 CB Methodology and the content of the unsolicited ratings that were issued under that methodology but without the required analysis of the cover pool being carried out. As to the mitigating factor set put in point II.3 of Annex IV of CRA Regulation, ESMA argues that the publication of the press releases informing the market that the cover pool analysis was not performed does not amount to quickly, efficiently, and completely bringing the infringement to ESMA’s attention, as required by the relevant provision of Annex IV, and, further, was done in compliance with
Article 10(2) of the CRA Regulation and point I.5. of Section D of Annex I, and
thus for a different purpose. ESMA further notes that the 20% turnover cap was respected, and this was so taking into consideration either the annual turnover of the business year preceding the end of the infringement (2015) (or, in the alternative but not noted in the Contested Decision) the one preceding the date of the decision (2019); and that Article 36a(4) refers to annual turnover without differentiating between revenues originating from rating services and revenues originating from ancillary services Finally, ESMA argues that the fine was calculated in accordance with predefined criteria set out by the CRA Regulation and also that the agency, due to the constraints to its discretion flowing from settled case-law of the CJEU (notably in the cases C 9/56, Meroni v High Authority and C-270/12, United Kingdom v Council, ECLI:EU:C:2014:18, paragraphs 41, 51-53) can only apply the aggravating and mitigating factors set out in Annex IV to the CRA Regulation, without any further possibility to adjust or calibrate in a discretionary manner the amount of the fine.
167. The Board of Appeal finds, on review of the CRA Regulation and the
facts as presented, that the fine was correctly applied.
168. First, the Board of Appeal is of the view that the adjustment coefficient
for infringements that have been committed for more than six months (CRA Regulation, Annex IV, point I.2) was applied correctly. At issue, in the view of the Board of Appeal, is whether the infringement was continuous in nature (relating to the period of time over which ratings remained outstanding that were

issued under the 2015 CB Methodology, but without the required cover pool analysis: 22 September 2015 to 22 July 2026) or, alternatively occurred at the points in time when ratings were formulated under the 2015 CB Methodology in a non-systematic manner (the 22 September and 26 November 2015 issuances of ratings).
169. The duration of the infringement is a key condition precedent for the
application of the adjustment coefficient relating to infringements committed for more than six months: specifically, was the infringement continuous or related to a specific, point-in-time application. It is ESMA’s contention that the infringement was continuous in nature; it is the appellant’s contention that the infringement relates to two instances of the application of the 2015 CB Methodology in September and November 2015. In order to assess the nature of the infringement by the appellant, the Board of Appeal has regard to the purpose of Article 8(3). The purpose of Article 8(3) is to ensure that methodologies, once adopted by a rating agency (and thereby also disclosed to the market and notified to ESMA), are applied in a systematic (and continuous) manner. It is the view of the Board of Appeal that it is therefore a logical corollary that any failure to comply with this obligation — failure to ensure that all ratings issued by a rating agency reflect a systematic application of relevant methodologies — continues for as long as the defect in the systematic application of the methodology continues. Any breach continues, therefore, until it is cured by action which ensures that the methodology is, in fact, being applied systematically, and that outstanding ratings reflect the analysis required by the methodology. The Board of Appeal notes, as a complementary but not primary point, the obligation under the CRA Regulation on rating agencies to review ratings and methodologies on an ongoing basis (Article 8(5)). While this separate obligation is distinct from the Article 8(3) obligation to apply methodologies systematically, it underlines the overall purpose of the CRA Regulation as regards the ongoing compliance of methodologies and ratings issued under such methodologies with the CRA Regulation. Further, Article 1 of the CRA Regulation provides that the Regulation “lays down conditions for the issuance of ratings”, one of which is the Article 8(3) requirement that methodologies be applied systematically. The legislative scheme of the CRA Regulation, therefore, implies that it is correct to interpret the nature of an Article 8(3) infringement relating to a methodology not being applied systematically, as being ongoing until: the methodology in question is revised such that ratings issued under the methodology are aligned correctly with the methodology such that the methodology applies systematically; or until the relevant ratings are either withdrawn or revised to align with the methodology such that, again, the methodology applies systematically. The Board of Appeal in this regard also places emphasis on the argument made by ESMA that the purpose of the aggravating coefficient is “to penalise slow reaction time” (ESMA Rejoinder, paragraph 182). The enforcement regime established under the CRA Regulation is designed to ensure that the regime is applied and that its pivotal requirements regarding the quality of ratings and the reliability of
methodologies are observed; it accordingly penalises ongoing breaches by means of an aggravating co-efficient.

As to the facts, the Board of Appeal finds that, as regards the contested
application of the 1.5 adjustment coefficient for infringements that have been committed for more than six months (CRA Regulation, Annex IV, point I.2), it is factually accurate to find that the breach obtained from 22 September 2015, when the appellant issued the first ratings under the 2015 CB Methodology without a cover pool analysis, to 22 July 2016, when the 2016 CB Methodology was adopted. The Board of Appeal notes in this regard ESMA’s correction of the argument made by the appellant in its Notice of Appeal that the time period adopted in the Contested Decision related from the point at which the 2015 CB Methodology was adopted in July 2015; ESMA notes that the time period was in fact applied by ESMA from the first issuance without the cover pool analysis

that is, from 22 September 2015 (Response by ESMA, paragraph 286). Over
this time period, the facts indicate that ratings were still in issue that had been adopted without the covered bond analysis (Exhibits 2 and 3 of ESMA’s submissions) and that were, accordingly, not aligned with the 2015 CB Methodology correctly; and that the 2015 CB Methodology had not been revised, so as to bring the outstanding ratings into alignment.

The Board of Appeal is accordingly of the view that ESMA applied the
aggravating coefficient correctly in law and on the facts by relating the time period for the infringement from the first unsystematic application of the 2015 CB Methodology (22 September 2015) to the adoption of the 2016 CB Methodology (22 July 2016), when the outstanding ratings became aligned with their governing methodology (the 2016 CB methodology) and so the application of the methodology could be regarded as systematic in accordance with Article 8(3). The linking by ESMA of the fine to the time period of 22 September 2015 to 22 July 2016 was therefore factually correct and in accordance with the CRA Regulation: over this time, the ratings which had not been subject to the cover pool analysis and which remained in issue were not in conformity with the analytical method set out in the governing 2015 CB Methodology nor were the errors in the application of the 2015 CB Methodology corrected so that the outstanding ratings represented a systematic application of the Methodology.
Relatedly, the Board of Appeal finds that ESMA applied the facts and the
law correctly in deciding that the infringement ended on 22 July 2016 when the 2016 CB Methodology was adopted. At that point, the ratings that had previously not been aligned with the 2015 CB Methodology came into conformity with the 2016 CB Methodology in that the 2016 Methodology did not require a cover pool analysis to be carried out for highly rated issuers which had already achieved a AAA rating (the ratings in September and November 2015 all being AAA rated). The 2016 CB Methodology was therefore aligned with the outstanding ratings at this point. The Board of Appeal is therefore of the view that ESMA applied the aggravating co-efficient of 1.5 correctly: the infringement was committed for a period of ten months, clearly leading to the mandatory application by ESMA of the 1.5 aggravating co-efficient.
The Board of Appeal further notes that had the appellant at any point
either undertaken the cover pool analysis, and thereby aligned the ratings in practice with the 2015 CB Methodology, or revised the 2015 CB Methodology as regards the cover pool analysis such that the ratings then became aligned

with the amended methodology, the time period of the infringement would have been shortened.
174. The Board of Appeal notes further that had the appellant taken advice
from ESMA at the point at which the first failure to apply the cover pool analysis arose (September 2015), and taken remedial action to cure the breach (e.g. by amending earlier the 2015 CB Methodology), the time period could have been substantially shortened to the point that the aggravating coefficient would not potentially have applied. The dis-application of the aggravating co-efficient was therefore within the appellant’s power to achieve, particularly as, as previously noted above in this Decision, the appellant could reasonably have been expected to check with ESMA as to whether “objective reasons” were in place that would have supported an exemption from Article 5(1) and Article 8(3).
175. The appellant argues that the infringement of Article 8(3) was not of a
continuous nature, but instead related to the specific applications in September and November 2015 of the 2015 CB Methodology (the points in time at which the ratings were formulated under the 2015 CB Methodology). The appellant argues that, at a maximum, the infringement applied from the start of the analytical process until the public assignment of the ratings (Appellant Reply, paragraph 92). The Board of Appeal in this regard highlights that, while ESMA does not admit that the appellant’s interpretation of the nature of the infringement is correct, ESMA notes in its Response that the appellant’s argument would not be to the benefit of the appellant (ESMA Response, paragraph 290). ESMA notes that if the specific applications of the 2015 CB Methodology to the ratings at issue were to be regarded as constituting the infringement, the fine would be significantly larger, to the appellant’s detriment. ESMA argues that under the appellant’s interpretation, either each individual covered bond rating (599 ratings) or, at the least, each separate CB programme (15 programmes), issued without the cover pool analysis, would be regarded as a separate instance of infringement. This would amount to a repetition of the infringement and, consequent on the required application of an aggravating coefficient of 1.1 (point I.1 of Annex IV, CRA Regulation) for each repetition, would lead to a fine of either EUR 1,000,000 (the application of the 1.1 coefficient x 14) or EUR 28,200,000 (the application of the 1.1 coefficient X 558) (ESMA Response, paragraph 290). The Board of Appeal gives weight to this argument. It is of the view that the infringement was, in law and in fact, continuous in nature and related to the ongoing failure to apply the 2015 CB Methodology systematically (by either withdrawing the ratings or revising the Methodology). But, were there to be doubt as to this point, and had ESMA a margin of supervisory appreciation in this regard then, and in accordance with settled CJEU case law (for example, judgment of 24 June 2015, C-293/13 P Fresh Del Monte Produce v Commission ECLI: EU: C:2015:416, paragraph 149), the principle of in dubio pro reo would require that the interpretation most favourable to the appellant be adopted. That is, the interpretation that the nature of infringement was of a continuous nature, relating to the continued misalignment of the outstanding ratings with the 2015 CB Methodology, and to the related failure to cure the defect by either revising the Methodology or withdrawing/revising the ratings. This interpretation is to the benefit of the appellant, leading to a materially lower fine than would an interpretation based

on the infringement being grounded in specific and repeated applications of the 2015 CB Methodology to the formulation of ratings where the cover pool analysis was not applied.
176. As regards the turnover cap, the Board finds that, on the facts, the
related requirements were correctly applied by ESMA. Article 36a(4) applies a cap of 20% of annual turnover of the rating agency concerned “in the preceding business year” to any fine imposed under the CRA Regulation. In the Contested Decision, ESMA considers this requirement as against the turnover of the appellant in 2015 (EUR 4,351,165) and finds that the fine does not exceed the cap (being less than EUR 870,233, 20% of 2015 turnover) (paragraph 392). The appellant, to the contrary, argues that the reference year for the calculation of turnover is 2014, the year preceding the start of the infringement (2015) and that ESMA did not accordingly use the correct turnover figure. The appellant argues, also basing the turnover figure on rating business only, that the correct
figure (from 2014) was EUR 1,554,646 and that the related cap amounted to
EUR 310,929, a cap exceeded by the fine imposed by ESMA (EUR 550,000). .
177. Critical to the question of the correct application of the turnover cap is
the reference year adopted to determine the turnover. ESMA in its Contested Decision uses the “full business year preceding the end of the infringement (i.e. 2015” (ESMA Response, paragraph 298). This determination is in accordance with its consistent practice as regards the calculation and imposition of monetary penalties (ESMA Response, paragraph 301). The Board of Appeal takes due note of ESMA’s practice in this regard, but is also mindful that the design and operation of ESMA’s practices regarding monetary penalties should reflect relevant EU law. The Board of Appeal notes in this regard that ESMA’s approach to “preceding business year” is not in accordance with cognate caselaw of the CJEU as regards the interpretation of “preceding business year” in the Regulation currently governing the calculation of monetary penalties in the competition field (Regulation (EU) No 1/2003, specifically Article 23(2)). The CJEU has held, in interpreting the precursor provision to Article 23(2) in Regulation No 17 (which also uses “preceding business year”), that the “preceding business year” is the year which precedes the date of the relevant enforcement decision, and has noted that it is logical to refer to that business year (judgment of 16 November 2000, C-291/98 P Sarrió SA v Commission ECLI: EU: C:2000:631, paragraph 85). The Board of Appeal notes that this ruling and the provision in question relates to Commission enforcement of competition law. Given, however, the identity of language and context (the application of a cap to a monetary penalty based on turnover) between Article 36a(4) and the provision interpreted by the CJEU in Sarrió, the Board of Appeal finds this caselaw to be of strongly persuasive authority in the interpretation of “full business year” in Article 36a(4). It finds, accordingly, that ESMA erred in interpreting Article 36a(4), and with regard to “preceding business year”, as relating to the year preceding the end of the infringement. In accordance with persuasive EU case law, the Board of Appeal finds that the correct reference year was 2019, being the year preceding the date of the Contested Decision. The Board of Appeal takes note of the argument made by ESMA that different language governs the calculation by ESMA of the ‘basic fine’ (Article 36a(2)) to that which governs the Commission’s equivalent powers as regard the

calculation of the basic fine (the calculation of fines by the Commission not being related to turnover in the equivalent Article 23(3) of Regulation No 1/2003). The Board is nonetheless of the view that the CJEU caselaw is unequivocal as regards the interpretation of “preceding business year” to the application of a cap on fines that is related to turnover, and in relating it to the year preceding the relevant enforcement decision. The Board finds, therefore that the correct reference year for the application of the cap was 2019.
178. The application of 2019 as the reference year does not, however, have
an impact on how the turnover cap applies to the fine imposed on the appellant by the Contested Decision. According to Exhibit 21 of ESMA’s submissions, the appellant’s turnover for 2019 was EUR 9,786,285. Using this figure to calculate the cap of 20% of turnover on the fine does not alter the calculation, as the fine remains within the cap. With the 2019 turnover figure, the cap amounts to EUR 1,957,257; the fine therefore remains within the cap at EUR 550,000 (ESMA Response, paragraph 298). No order is accordingly made as regards the calculation of the fine or its remittance to ESMA in this regard. The Board of Appeal recommends, however, that ESMA reviews its fining policies and practices to assure itself they align with CJEU jurisprudence.
179. Further, the Board of Appeal agrees with ESMA that, contrary to the
argument of the appellant (Notice of Appeal, paragraph 156), there are no grounds for calculating annual turnover, for the purposes of Article 36a(4), with reference only to revenues arising from rating services and excluding ancillary services. The Board of Appeal notes that “annual turnover” is not defined by the CRA Regulation. The Board of Appeal has, therefore, in accordance with settled EU law principles, adopted a textual, teleological and contextual approach to the term’s interpretation. From the textual perspective, annual turnover/turnover, while not defined, is referred to in a number of provisions across the Regulation. These include as regards the application of the rules governing the maximum duration of a contractual relationship with a rating agency (Article 6b(5)); ESMA’s obligation to monitor the total market share of registered rating agencies (Article 8d(3)); registration and supervisory fees (Article 19); the quantum of fines and periodic penalty payments (Articles 36a and 36b); and the annual disclosures required of rating agencies as regards revenue (Annex I, Section E, III.7). In all but three cases (Article 8d(3), Article 6b(5) and Annex I, Section E, III.7), the term turnover does not distinguish between turnover relating to rating services and turnover relating to other ancillary business. Article 8d(3), by contrast, specifies that, as regards ESMA’s annual publication of the total market share of registered rating agencies, the calculation of total market share is to be undertaken “with reference to annual turnover generated from credit rating activities and ancillary services”. Annex I,
Section E, III.7 similarly requires a rating agency to disclose annually its total
turnover, divided into fees from rating and ancillary services. Ancillary services are defined as not forming part of rating activities and as comprising services relating to market forecasts, estimates of economic trends, pricing analysis and other general data analysis as well as related distribution services (CRA Regulation, Annex I, Section B, 4). The Board of Appeal notes the specification in Article 8d(3) and in Annex I, Section E, III.7 of turnover as including rating services and ancillary services. It acknowledges that it might be textually

inferred from this that, where the CRA Regulation does not expressly specify that turnover includes rating services and ancillary services, annual turnover should be read as referring only to rating services. In Article 6b(5), however, turnover is expressed directly in terms of “turnover generated from credit rating activities”. There are, therefore, different approaches taken in the CRA Regulation. On balance, the Board is of the view that the reference to annual turnover in Article 36a(4) should, from a textual perspective, not be limited to rating services only. First, it might, given that Article 8d(3) and Annex I, Section E, III.7 relate turnover to all rating agency business, reasonably be assumed that the intention of the co-legislators was for annual turnover to be interpreted as relating to all rating agency business, including ancillary services, and that a distinction not be made, whenever annual turnover is referred to, between rating services and ancillary services. Further, the clear qualification with regard to turnover relating to rating activities only in Article 6b(5) suggests that were the co-legislators minded to expressly reference turnover as relating to rating services only, they would have done so. Second, to limit the term annual turnover in Article 36a(4) to rating services only, and thereby reduce the scope of the term turnover, would amount to a material change to the language of
Article 36a(4). In this regard, the Board of Appeal places weight on the need to
give effect to the intention of the co-legislators. Had the co-legislators intended to make a distinction between rating services and ancillary services in Article 36a(4), the Board of Appeal is of the view that it is reasonable to assume that they would have done so (as they did in Article 6b(5)). Further, the reference to annual turnover in Article 36a(4) relates to the quantification of a sanction. Given the materiality of this obligation and its third party effects, the Board of Appeal is of the view that it is reasonable to assume that the clear language of the text must be followed; this language does not include a qualification limiting the definition of annual turnover to rating services only. To insert such a qualification would, in the Board’s view, represent a material change to the clear text of the provision.
180. It would also run counter to a teleological interpretation of the CRA
Regulation. The CRA Regulation expressly permits rating agencies to engage in non-rating ancillary services (CRA Regulation, Annex I, Section B, 4; recital 6)). This implies that the CRA Regulation is designed to apply widely to rating agency activities. Accordingly, it is reasonable to infer that annual turnover, unless otherwise qualified, relates to all rating agency turnover.
181. Further, and from a contextual as well as teleological perspective, the
Board notes the CRA Regulation was designed in part to address the failures which followed from conflicts of interest within rating agencies, prior to the financial crisis, as between rating and non-rating advisory business (particularly as regards the rating of securitised debt). The CRA Regulation accordingly expressly addresses the entirety of rating agency business. CRA Regulation,
Annex I, Section B, 4, for example, provides that a rating agency shall ensure
that the provision of ancillary services does not present conflicts of interest with rating activities; Annex I, Section B, 3c provides that fees for rating business and ancillary services should be based on actual cost; and Annex I, Section E, I.2 and II.2 require a rating agency to disclose a list of its ancillary services as well as (annually) fees from ancillary services Similarly, recital 6 notes that the

performance of ancillary activities should not compromise the independence or integrity of rating agencies’ rating activities; and recital 22 notes that rating agencies should not be allowed to carry out consultancy or advisory services but should be able to provide ancillary services where this does not create potential conflicts of interest with the issuance of credit ratings. The enforcement regime also applies to the entirety of rating agency business. Under point 23 of Section I of Annex III, a rating agency infringes the CRA Regulation where it does not ensure that the provision of an ancillary service does not present a conflict of interest with its rating activities. The Board of Appeal is therefore of the view that the CRA Regulation applies to the entirety of a rating agency’s business and that, accordingly, the reference to annual turnover in Article 36a(4) should be interpreted as relating to the entire turnover of a rating agency and not be limited to turnover relating to ratings business.
182. While recognizing the textual ambiguity relating to the definition of
annual turnover, the Board is of the view that ESMA did not err in law in interpreting annual turnover under Article 36a(4) as relating to rating services and ancillary services. In reaching this view, the Board is mindful of not exceeding its mandate, and so of not inserting a qualifying condition into the meaning of annual turnover for the purposes of Article 36a(4) where such has not been expressly inserted by the co-legislators.
183. As regards the 0.4 mitigating coefficient relating to a rating agency
bringing an infringement quickly, effectively, and completely to ESMA’s attention (CRA Regulation, Annex IV, point II.3), the Board finds that ESMA was correct in not applying the mitigating coefficient as it finds it clear on the facts that the appellant did not acknowledge that it had committed (or believe it could have committed) an infringement, and done so quickly, effectively, and completely. There is no evidence to this effect from the facts reviewed by the Board of Appeal. The Board of Appeal agrees with ESMA that the relevant press releases do not meet the requirement as regards bringing the infringement to ESMA’s attention, and within the terms of point II.3. The press releases served a different purpose (disclosure under CRA Regulation Article 10(2)), did not expressly note the existence of an infringement, and were not addressed directly to ESMA. Moreover, in the Board of Appeal’s view, the appellant should also have brought to the attention of ESMA this situation, but it did not .
184. Finally, and as regards the appellant’s argument that the fine was
disproportionate, the Board does not find this argument to be made out. ESMA does not have discretion under the CRA Regulation to alter or calibrate fines depending on its subjective view of the seriousness or otherwise of an infringement or based on factors beyond those identified in the Regulation. The list of infringements set out in Annex III governs the infringements which must be subject to enforcement action by ESMA once a breach is identified. Further, the CRA Regulation contains a number of design features precisely adopted in order to support appropriate finessing of sanctions in light of individual circumstances. These include the set of aggravating and mitigating co-efficients set out in Annex IV to the Regulation and applied by ESMA in this case. In this regard, the Board of Appeal notes that recital 19 of Regulation (EU) No

513/2011 (which empowers ESMA to act as supervisor of rating agencies and take enforcement action) provides that co-efficients linked to aggravating and mitigating circumstances are established in order to give the necessary tools to ESMA to decide on a fine which is proportionate to the seriousness of an infringement committed by a credit rating agency, taking into account the circumstances under which the infringement was committed. Proportionality is accordingly embedded within the design of the fines regime. Finally, the Board of Appeal notes that, as an EU agency, and so an entity which must operate under the legislative framework established by the co-legislators, ESMA must apply its enforcement powers within the terms of the CRA Regulation. The Regulation does not contain an over-arching discretion for ESMA to calibrate fines in the manner argued for by the appellant.
185. The Board of Appeal finds accordingly that the amount of the fine for the
relevant infringement (point 43 of Section I of Annex III of the CRA Regulation) was correctly calculated. II – Grounds of appeal regarding the part of the Contested Decision concerning the infringement of the third subparagraph of Article 14(3), the first subparagraph of Article 8(5a) and point (aa) of Article 8(6) of the CRA Regulation (a) Whether or not ESMA erred in law in the interpretation of the scope pf the concept of ‘material changes’ and in finding that the changes introduced by the appellant to the 2015 CB Methodology were material.
186. In the Contested Decision ESMA also found that the appellant had
breached its obligations under Articles 8(5a), 8(6)(aa) and 14(3) of the CRA Regulation by: (i) not having notified ESMA about the intended material changes to the 2015 CB Methodology; (ii) not having published those changes on its website and (iii) not having informed ESMA and not having published immediately on its website the results of the consultation about the intended material changes to the 2015 CB Methodology. The Contested Decision also held that ESMA answer 7 to the Q&A provided a non-exhaustive list of examples of changes which are considered to be material changes and that the mere fact that no existing rating is de facto impacted by the changes introduced to a methodology does not per se exclude that these changes are material. This needs to be assessed on a case-by-case basis.
187. The appellant argues that there has not been a material change within
the terms of Article 8(5a); the changes were simply a clarification of the rating methodology that was already in use and/or concerned only the rating outlook. It claims that Article 8(5a) of the CRA Regulation only applies where two cumulative conditions are met independently: (i) the intended changes to the methodology are material and (ii) they could have an impact on a credit rating, and denies that this was the case. The appellant further notes that answer 7 to the Q&A shows that there is a high threshold for a change to be considered material (noting that answer 7 includes as an example a change that has a direct or indirect impact on a significant number of credit ratings), and that a change cannot be considered to be material if it does not (and could not) have any impact on credit ratings. The appellant argues that the changes introduced

in the 2016 CB Methodology did not and could not have an impact on the ratings carried out, at the most having an impact on the rating outlook, and not the rating itself. The appellant further notes that the changes did not alter the way in which an assessment of the cover pool was to be performed, as they only clarified and laid down what was common practice in the rating market. The changes had the intention of describing more precisely the methodology that has been used by the appellant since July 2015.
188. The Board of Appeal finds that ESMA did not err in law on the
interpretation of the scope of “material changes”, included under Article 8(5a). This term is not defined in the CRA Regulation, although “material changes” in
Article 8(5a) (which governs Articles 8(6) and 14(3)) are related to changes that
“could have an impact on a credit rating”. As argued in earlier paragraphs of this Decision, the term is accordingly to be interpreted in accordance with settled principles governing the interpretation of EU legislation, including as to text and context. Textually, the co-legislators did not qualify the term material change by reference to change that would, in fact, have an impact on a credit rating. As regards context, the purpose and objectives of the CRA Regulation imply, in the Board of Appeal’s view, that material change should be interpreted broadly, and on a case-by-case basis, so that the objectives of the CRA Regulation, including as regards notification of ESMA and consultation with stakeholders are met. The Board of Appeal notes in particular the reference in the Contested Decision to recital 27 of the 2013 CRA Regulation No 462/2013, from which Article 8(5a) of the CRA Regulation derives (Contested Decision, paragraph 420). The Board of Appeal is of the view that the Contested Decision correctly interpreted Article 8(5a) in adopting a case-by-case approach to the notion of material change, and in linking the notion of material change to whether or not there could have been a potential impact on a rating given the materiality of the change, whether or not the impact actually exists (Contested Decision, paragraph 423).
189. The Board of Appeal is also of the view that the inclusion in the ESMA
Q&A, answer 7 of “a change that has a direct or indirect impact on a significant number of credit ratings” as an example of a material change, is illustrative only, particularly in light of the non-exhaustive nature of the Q&A.
190. The Board further notes that the Q&A is a soft instrument and, while of
authoritative status as to ESMA’s thinking, it cannot, in accordance with settled law, limit the clear and express meaning of legislative provisions.
191. On the facts, and while being mindful of the operational latitude in
supervisory decisions that should be afforded to ESMA’s margin of appreciation, and including with regard to the application of facts to opentextured provisions, the Board of Appeal is of the view that ESMA was correct, in assessing materiality, in conducting a comparative analysis of the 2015 and 2016 CB Methodologies. Specifically, the Board of the Appeal is of the view that ESMA was correct on the facts in finding that removing the requirement for the cover pool analysis, and changing how it was to be applied where it was deployed (Contested Decision, paragraphs 431-442), amounted in practice to a material change to the design of the 2015 CB Methodology as regards how

the cover pool assessment was to be performed. While the Board of Appeal acknowledges that in practice these changes would not have had an impact on the rating of issuers already with the highest rating, and would also reflect the previous practice of the appellant, it is of the view that these factors are not relevant. The changes represented a significant change to the design of the 2015 CB Methodology and therefore to how the Methodology was to be applied. In effect, the changes removed, for a class of issuers, a building block previously seen as necessary for all ratings, and thereby amounted to a material change to the manner in which the assessment of the cover pool was to be performed for cover bonds. The Board of Appeal also notes that the 2016 CB Methodology could have had an impact for those issuers whose rating level could not have reached the highest level in the absence of the cover pool analysis (ESMA Response, paragraph 344). The Board of Appeal concludes, accordingly, that ESMA’s decision was based on a reasonable interpretation of the facts and in accordance with the law. ESMA accordingly acted correctly in its application of Article 8(5a). (b) Whether or not the appellant acted negligently
192. The Contested Decision concluded that the appellant acted negligently
at the time of the revision of the 2015 CB Methodology, including by not undertaking with sufficient care an assessment of whether or not the changes were material and by failing to raise any questions with ESMA as to whether the changes could be considered material. The Contested Decision notes, inter alia, the appellant’s failure to have in place, at the time of the changes to the 2015 CB Methodology, a procedure explaining how it would assess whether a change was material and how it would ensure compliance with its obligations under the CRA Regulation, including as regards notification of ESMA and publication, and finds this to denote a clear lack of care. The Contested Decision also reveals a lack of care in the assessment of the materiality of the specific changes made to the 2015 CB Methodology, particularly as regards supporting documents, contrary to what a professional firm in the financial services sector, subject to stringent regulatory requirements, would have done.
193. The appellant claims that the lack of a written description of the
procedure for material changes cannot denote a lack of care at the time of revision of the 2015 CB Methodology and that such a written procedure was developed in 2016, as soon as the change of methodologies became relevant, and in parallel with the development of the 2016 CB Methodology. The appellant also claims that the lack of a formal assessment of the materiality of the changes introduced to the 2015 CB Methodology was due to the fact that these were considered by the appellant as pure clarifications.
194. The Board of Appeal refers to its earlier findings as to the high level of
care required of the appellant. As to the standard of care applicable and whether or not it has been met, for all the reasons already stated above it agrees with ESMA’s Contested Decision that the relevant standard of care and its application to the appellant can be characterised in the following manner:
“As a professional firm in the financial services sector subject to stringent regulatory requirements, [the appellant] is required to take special care in

assessing the risks that its acts or omissions entail, and has failed to take that care; and as a result of that failure, it has not foreseen the consequences of its acts or omissions, including particularly its infringement of the Regulation, in circumstances where a person in such a position who is normally informed and sufficiently attentive could not have failed to foresee those consequences” (Contested Decision, paragraph 480).
195. As to the application of the standard to the facts, the Board of Appeal
finds that there is factual evidence of a lack of care sufficient to amount to negligent infringement of points 3a and 3b of Section II of Annex IIII and Points 4a of Section III of Annex III of the CRA Regulation as regards the material changes made to the 2015 CB Methodology. The Board of Appeal notes in particular that there was no internal policy in place prior to 2016 governing the assessment of “material changes” and related compliance with the consequential requirements of the CRA Regulation; and that the policy changes developed in parallel with the 2016 changes to the 2015 CB Methodology did not address the consequential regulatory obligations relating to ESMA notification and public consultation. It is of the view that ESMA was correct in relying on these internal policy gaps as evidence at the relevant time of a lack of care. Specifically, the Board of Appeal finds that the absence of appropriately comprehensive policies suggests that the appellant had not shown the care required of it, as a regulated actor in a market (the ratings market) subject to a high level of regulation, as regards ensuring it was in compliance with the CRA Regulation with respect to changes to methodologies.
196. The Board of Appeal also finds on the facts a lack of a sufficiency of
evidence as to a diligent and careful assessment, as would be expected of a regulated and professional actor operating in a highly regulated environment, being carried out by the appellant of whether or not the changes were material, including as regards the critical Methodology Committee Memorandum of July 2016.
197. The Board of Appeal accordingly finds that ESMA was correct in drawing
a finding of negligence from this insufficiency of evidence as to appropriate levels of care and diligence.
198. Finally, the Board of Appeal notes the failure of the appellant to consult
with ESMA in circumstances where a reasonable rating agency, operating under the CRA Regulation, could have been expected to foresee the need for consultation, particularly given the consequential nature of a finding (or not) of materiality given the related regulatory obligations which follow and the related potential for an infringement to arise.
199. The Board of Appeal is of the view, accordingly, that on the facts ESMA
was correct in finding that the infringement of Points 3a and 3b of Section II of
Annex IIII and Points 4a of Section III of Annex III were committed negligently.
(c) Whether or not ESMA’s calculation of the fine is correct

Finally, the appellant claims that ESMA erred in the calculation of the
fine related to this infringement, because ESMA did not apply both (i) the mitigating factor set out at point II.3 of Annex IV of the CRA Regulation and (ii) the mitigating factor set out at point II.4 of Annex IV. Under Point II.3 of Annex IV a mitigating adjustment coefficient of 0.4 applies where a rating agency has brought quickly, efficiently, and completely the infringement to ESMA’s attention. The appellant argues that it is sufficient for the conditions of this mitigation to be met if a rating agency brings all relevant factors to ESMA’s attention. The appellant claims this requirement is met as it notified what it regarded as clarifications of the 2015 CB Methodology to ESMA and receipt of this notification was confirmed by ESMA; ESMA accordingly received all necessary information. Under point II.4 of Annex IV, a mitigating adjustment coefficient of 0.6 applies where the rating agency has voluntarily taken measures to ensure that similar infringements cannot be committed in the future. The appellant relates compliance with these mitigation conditions to its Validation Policy distinguishing, since 2016, between material and non-material changes. The appellant disagrees with the view of ESMA in the Contested Decision that the definition of “material changes” in the Policy is too narrow. It argues accordingly that the point II.4 mitigating coefficient should have been applied.
In the Contested Decision, ESMA does not apply the mitigation
adjustment coefficient under Point II.3 of Annex IV as it is of the view that the infringements in question (points 3a and 3b of Section II of Annex III and point 4a of Section III of Annex III) came to ESMA’s attention only through the investigation conducted by ESMA’s Supervision Department and the subsequent IIO investigation; they were not brought to ESMA’s attention directly by the appellant. Specifically, the notification of what the appellant termed clarifications to the 2015 CB Methodology to ESMA did not indicate an infringement (or an issue of concern) and so could not be considered as sufficient to ground mitigation. As regards the mitigation adjustment coefficient under point II.4, ESMA does not apply this mitigation in the Contested Decision as it is of the view that the governing definition of “material change” in the appellant’s 2016 Validation Policy does not cover all cases of material change for the purposes of the CRA Regulation. Specifically, the notion of materiality is linked in the Validation Policy to changes that would impact existing ratings, or substantial changes of a key rating factor; relatedly, the notion of materiality in the appellant’s Rating Methodologies Process Manuals (2017, 2018, and 2019) relates material changes to methodologies to substantial changes to one or more key rating factor(s) or their weight, or to changes that impact already assigned ratings. Accordingly, the relevant policies do not cover the range of changes which can be regarded as material under the CRA Regulation, specifically by their linking materiality to actual impact on existing/already assigned ratings. The Contested Decision notes that the materiality assessment should take into consideration whether there could be a potential impact on ratings, and not whether the impact would actually materialise (Contested Decision, paragraph 534). Accordingly, the Contested Decision finds that despite the appellant voluntarily adopting changes to its processes regarding methodology changes, it was not established that these measures would prevent a similar infringement being committed in the future.
The Board of Appeal finds, on review of the facts and the CRA
Regulation, that ESMA correctly applied the relevant mitigation rules under the CRA Regulation. Specifically, the Board of Appeal finds as regards the mitigation coefficient adjustment set out in point II.3 of Annex IV that it is clear that the appellant did not quickly, efficiently, and completely bring the infringement to ESMA’s attention. The relevant notification of clarifications to ESMA did not in any way indicate expressly to ESMA that an infringement had been committed. Further, on the facts presented to the Board of Appeal, the notification in question was provided in the course of the appellant’s ongoing supervisory relationship with ESMA and as part of its periodic disclosures; it was not presented in the form of an express acknowledgement of an infringement that is clearly required by point II.3 of Annex IV. The Board of Appeal notes and gives weight in this regard that ESMA only came to have notice of the infringements following supervisory and subsequently IIO action (following, in turn, a complaint). On the facts, therefore, ESMA was correct in finding that this coefficient could not be applied.
As regards the application of the mitigation coefficient set out in point II.4
of Annex IV, the Board of Appeal finds that it is clear and unequivocal that this mitigation requires that measures have been taken by the rating agency in question to ensure that similar infringements cannot be committed in the future. Central to securing this outcome in the instant case, in the Board of Appeal’s view, is that the appellant’s policies and procedures are clear as to the potential range of “material changes” under the CRA Regulation. The Board of Appeal finds, however, that there is sufficient doubt as to the capacity of the changes made by the appellant to relevant internal policies and procedures to prevent a future infringement such as to render the mitigation inapplicable. While the Board of Appeal acknowledges the voluntary efforts the appellant has made to revise relevant policies and procedures as regards the management of material changes, and compliance with relevant rules under the CRA Regulation, it is not convinced that the approach adopted to the critical determinative notion of material change means that it is ensured that similar infringements will not be committed in the future. The Board of Appeal is of the view that ESMA correctly interpreted the CRA Regulation as relating the notion of a “material change” to a case-by-case assessment and, in particular, as not being dependent on the actual existence of an impact on a rating. The Board of Appeal is of the view that, on the facts, the measures taken by the appellant do not sufficiently ensure that it will in the future be in compliance with its obligations under the CRA Regulation as regards material changes to methodologies; these measures, on the facts, do not sufficiently reflect the required linkage between materiality and a potential impact on ratings. The Board of Appeal therefore finds that ESMA was correct in law, and on the application of the facts, in finding that there was not sufficient evidence to establish mitigation as regards the adoption of measures to ensure the infringements under points 3a and 3b of Section II of
Annex IIII and points 4a of Section III of Annex III cannot be committed in the
future.
The Board of Appeal draws weight in its conclusion from the evidence
that the appellant has not accepted that its interpretation of a material change

as requiring actual impact on a rating is not in accordance with the correct interpretation (in the view of the Board of Appeal) of this term under the CRA Regulation, specifically Article 8(5a).
205. The Board of Appeal finds accordingly that the amount of the fine for the
relevant infringements under points 3a and 3b of Section II of Annex IIII and points 4a of Section III of Annex III was correctly calculated. The decision On these grounds the Board of Appeal unanimously decides to dismiss the appeal. The original of this Decision is signed by the Members of the Board in electronic format, as authorised by Article 22.2 of the Rules of Procedure and countersigned by hand by the Secretariat. Marco Lamandini (President and Co-Rapporteur) (SIGNED) Pat McArdle (SIGNED) Niamh Moloney (Co-Rapporteur) (SIGNED) Katalin Mero (SIGNED) Beata Mrozowska (SIGNED) Michele Siri (SIGNED) On behalf of the Secretariat Tijmen Swank (SIGNED) A signed copy of the decision is held by the Secretariat.

Lineage

Similar documents from other regulators