Guidelines on the recognition of External Credit Assessment Institutions

20 January 2006
Guidelines on the recognition of External Credit Assessment Institutions Executive Summary

The Capital Requirements Directive1 (CRD) allows institutions to use
external credit assessments to determine the risk weight of their exposures, provided the External Credit Assessment Institutions (ECAIs) that produce those assessments have been recognised as eligible for that purpose by the competent supervisory authorities. This recognition is granted only if the competent authorities judge an ECAI to meet the recognition criteria laid down in the CRD.
ECAI recognition for capital purposes does not in any way constitute a
form of regulation of ECAIs or a form of licensing of rating agencies to do business in Europe. Its sole purpose is to provide a basis for capital requirement calculations in the Standardised Approach and the Securitisation Ratings Based Approaches. 2 These approaches are intended to increase the risksensitivity of capital requirements relative to the current framework, and to ensure that institutions using these approaches have appropriate levels of regulatory capital to support their aggregate credit risk.
This paper sets out CEBS’ proposed common approach to the recognition
of eligible ECAIs. This covers:

The recognition process,
The implementation of the CRD recognition criteria, and
The criteria for 'mapping' external credit assessments to the CRD
risk weights.

The guidelines detail the high degree of convergence that has been
achieved among supervisors in recent months on both the procedural and 1 Recast 2000/12/EC and 93/6/EEC DirectivesVersion ECOFIN 29912890/05 2 Recital 34(a) of the CRD states that "with regard to the use of both external and an institution's own estimates or internal ratings account must be taken of the fact that, at present, only the latter are drawn up by an entity the financial institution itself which is subject to a European authorisation process. In the case of external ratings use is made of the products of what are known as recognised ratings agencies, which in Europe are not currently subject to an authorisation process. In view of the importance of external ratings in connection with the calculation of capital requirements under this Directive, appropriate future authorisation and supervisory process for rating agencies need to be kept under review."

substantive aspects of ECAI recognition. This includes a significantly enhanced common understanding of the recognition criteria set out in the CRD and of their implementation across the EU.
5. The intent of the guidelines is to provide the basis for consistent decision
making across jurisdictions, enhance the singlemarket level playing field, and reduce administrative burdens for all participants, including potentially eligible ECAIs, institutions, and supervisory authorities.
6. The guidelines set out agreed procedures for the application and
assessment process. Two modes of supervisory recognition are set out in the CRD: direct and indirect recognition. In direct recognition, supervisors make their own evaluation of an ECAI's compliance with the recognition criteria. In indirect recognition, supervisors recognise an ECAI based on recognition in another Member State, without carrying out their own evaluation process.
7. CEBS considers both of these approaches to be important and notes that
indirect recognition can be a valuable instrument for enhancing efficiency and reducing administrative burdens. CEBS believes that the common understanding set out in the guidelines will provide a sound foundation for supervisors' use of the indirect recognition approach in relevant circumstances.
8. Where recognition is sought in more than one Member State, competent
authorities will cooperate in a joint assessment process. The aim of the joint assessment process is to reach a shared view on compliance with the recognition criteria, while respecting the requirements of the CRD for individual decisions by supervisors.
9. Supervisors propose to carry out an overall assessment of ECAIs’ eligibility
according to the CRD recognition criteria, based on the common understanding set out in this paper and using information identified in a 'common basis application pack.'
10. Concerning the 'mapping' of external credit assessments to the CRD's
credit quality steps, CEBS considers the Basel Committee's guidance for supervisors, set out in Annex 2 of the Basel II framework published in June 2004, to be valuable and appropriate, and recommends that supervisors follow it. A common approach to mapping is important for ensuring consistency across the EU and reducing the risk of regulatory arbitrage. 11. In developing its guidelines, CEBS has benefited from ongoing dialogue and meetings with a large number of market participants. The guidelines went through a threemonth public consultation in 2005. Special attention was given to the technical issue of the mapping of securitisation positions and collective investment undertakings (CIUs), which went through a dedicated onemonth public consultation.
12. The comments received have been published on the CEBS website unless
the respondents requested otherwise. Feedback on the responses received have been published in a separate document (see www.cebs.org/pdfs/CP07_feedback.pdf ) .
13. CEBS participated as an observer to the Committee of European Securities
regulators (CESR) task force on Credit Rating Agencies, which drafted

technical advice to the European Commission on possible measures concerning credit rating agencies in March 2005. The dialogue between the two Committees has continued since then, in order to ensure complementary approaches.
14. In view of the approaching deadline for implementation, it is desirable that
the recognition of ECAIs should advance as quickly as possible. While formal processes cannot begin before finalisation and transposition of the CRD, many supervisors intend to be in a position to accept informal applications in the near future. Those supervisors are expecting to start this process on the basis of these guidelines from the beginning of February 2006.

Table of contents 4
Introduction 5
Part 1: The recognition process 7
General Principles 7
Application 8
The level of recognition 9
The form of recognition 9
Application in more than one Member State 10 Provision of information to support the application 12 Review of application by the competent authority 13 Disclosure by the competent authority 13 Ongoing review of eligibility 13
Part 2: Common understanding of the ECAI recognition 14
criteria laid down in the CRD
General Principles 14
The technical criteria 16
Part 3: Mapping 24
General Principles 25
Credit assessments of exposures others than securitisation positions and CIUs 26 Credit assessments of securitisation positions 29 Shortterm credit assessments Collective Investment Undertakings 31
Part 4: Export Credit Agencies 31
Annex I: Common Basis Application Pack 33
General Information 33
Methodology 34
Individual Credit Assessments 36
Mapping 37

Introduction

An eligible ECAI is an entity, other than an Export Credit Agency 3 , that
issues external credit assessments, and that has been determined by the competent authorities to meet the eligibility requirements set out in the Capital Requirements Directive4 (CRD). Only the credit assessments of an eligible ECAI, and for some exposures, the credit assessments of Export Credit Agencies, may be used by credit institutions and investment firms 5 (institutions) for the purposes of determining risk weights under the Standardised Approach and the Securitisation Ratings Based Approaches.
The CRD allows Member States to recognise an ECAI as eligible in two
ways: direct recognition, in which the competent authority carries out its own assessment of the ECAI’s compliance with the CRD’s eligibility criteria; and indirect recognition, in which the competent authority recognises the ECAI without carrying out its own evaluation, relying instead on the recognition of the ECAI by the competent authority of another Member State.
A common understanding of the recognition criteria and processes has
been developed to support consistency in direct recognition decision making across the EU and to increase the scope for indirect recognition. In order to avoid the inefficiencies of sequential direct recognition processes in cases where applications for the same ECAI are received by a number of competent authorities, those competent authorities will participate in a ‘joint assessment process’ to assess together the ECAIs’ eligibility. Recognising that the CRD requires a decision by each competent authority, where a shared view is achieved, this should form the basis for national decisionmaking.
By adopting common procedures and reaching a common understanding of
the CRD recognition criteria, competent authorities seek to ensure the consistency of the recognition process. In particular:

All ECAI applications will have to be supported by evidence that the
credit assessments will be used for regulatory capital purposes under the Standardised and/or Securitisation Ratings Based Approaches.
The applicant will have to fill out an application pack. CEBS has
developed a 'common basis application pack' which will provide supervisors with an adequate level and amount of information for their assessment. Competent authorities will be able to collect additional information needed to address countryspecific issues. 3 Annex VI, Part 1, paragraph 7 of the CRD states the criteria that a credit assessment of an Export Credit Agency must meet in order to be recognised for determining the risk weighting of exposures to central governments and central banks. See Part 4 of this paper. 4 The CEBS Guidelines are based on the Capital Requirements Directive as adopted by the European Institutions (ECOFIN 29912890/05) which will replace the existing Consolidated Banking Directive (2000/12/EC) and Capital Adequacy Directive (93/6/EEC). 5 See Article 2(2)(a) of the recast Directive 93/6/EEC.
Supervisors will assess the information provided in the application
pack in accordance with the common understanding of the CRD recognition criteria laid out in this paper. The purpose of the recognition criteria is to identify ECAIs that produce external credit assessments of sufficiently high quality, consistency and robustness to be used by institutions for regulatory capital purposes under the Standardised Approach and the Securitisation Ratings Based Approaches.

To ensure the transparency of the recognition process, supervisors are
required to disclose an explanation of their recognition process and a list of eligible ECAIs. These disclosures will be part of the CEBS supervisory disclosure framework, which has been designed to allow for a meaningful comparison 6 of the disclosed information.
To calculate the risk weight of exposures under the Standardised Approach
and the Securitisation Ratings Based Approach, competent authorities have to determine which credit quality steps the external credit assessments are to be associated with. Those determinations shall be objective and consistent, and based on both qualitative and quantitative factors.
In general terms, and consistent with the CRD (which requires the use of a
benchmark), the approach to mapping set out in the guidance provided by the Basel Committee (Basel II framework, June 2004, Annex 2) is considered to represent the appropriate basis for mapping under these guidelines. It incorporates the use of threeyear cumulative default rates together with qualitative analysis and appropriate flexibility of supervisory response.
The CRD requires supervisors to carry out a separate mapping of
securitisation positions, using both quantitative and qualitative factors. Supervisors will consider quantitative factors such as default and loss rates, and qualitative factors such as the methodologies adopted by ECAIs, the range of transactions assessed, and whether market participants view ECAIs' ratings of securitisation products as being equivalent.
The guidelines adopt a tailored approach to the spirit and use of the credit
assessments of Collective Investment Undertakings (CIUs) in the Standardised and Securitisation Ratings Based Approaches, in order to capture the variety of funds in the market. The guidelines state that, to be eligible, a CIU credit assessment must depend primarily on the credit quality of the underlying assets CIU credit assessments will not be subject to a separate mapping approach, but will be mapped using the approach referred to in paragraph 7 above.
This paper has four parts and an annex:

Part 1 describes the process for applying for ECAI recognition and
outlines how applications will be handled by competent authorities. 6 See Article 144 of the CRD
Part 2 sets out CEBS' common understanding on interpretation and
application of the CRD criteria for assessing whether an ECAI is eligible for the purposes of the CRD.
Part 3 proposes guidance to competent authorities for mapping the
credit assessments of eligible ECAIs to the credit quality steps in the CRD. This section addresses specific issues relating to the mapping of particular portfolios, such as shortterm credit assessments and the credit assessments of securitisation positions and CIUs.
Part 4 provides clarification on the use of Export Credit Agencies’
assessments for regulatory capital purposes in accordance with the provisions of the CRD.
Annex I lists the information requirements that form the 'common
basis application pack' to be used by supervisors.

Part 1: The recognition process
General principles
11. The responsibility for assessing whether the CRD recognition criteria are
fulfilled and an ECAI should be recognised as eligible lies with the competent authority of the country where the institution that intends to use that ECAI’s credit assessment is authorised and supervised.
12. Recognition of an ECAI by a competent authority should not be taken as
indicating suitability for any purpose other than the calculation of regulatory capital requirements using the Standardised Approach or the Securitisation Ratings Based Approach. Institutions retain full responsibility for their internal risk management.
13. The recognition process is initiated when the competent authority receives
an application for recognition. The competent authority in each Member State will indicate and disclose from which type of entity it will accept applications: ECAIs and/or institutions that intend to use the ECAI's credit assessments for risk weighting purposes. Further details are provided below.
14. In any event, before undertaking an assessment of an ECAI's eligibility, the competent authority will have to determine that at least one institution
within its jurisdiction intends to use the ECAI's credit assessments for risk weighting purposes. This will ensure that competent authorities need only consider applications of ECAIs whose credit assessments would actually be used under the Standardised or the Securitisation Ratings Based Approaches.
15. An institution will not be allowed to apply on behalf of, or to nominate for
its own capital purposes, an ECAI which is its subsidiary. This prohibition is judged necessary for consistency with the policy of avoiding any institution using ‘external’ credit assessments which are issued within its group.
16. Competent authorities must be provided with all material information they
need in order to assess whether an ECAI meets the CRD eligibility criteria. Consequently, regardless of how an application is initiated, the recognition process will require the full cooperation of the ECAI, in terms of its willingness and promptness in providing necessary information.
17. Some ECAIs may apply for recognition in more than one Member State, or
equivalently, institutions in more than one Member State may submit an application on behalf of the same ECAI. Competent authorities will adopt a single joint approach to the assessment of such applications. This approach has been designed to avoid duplication of work, promote supervisory efficiency, and reduce the overall burden of the recognition process. The aim is to reach a shared view on the ECAI’s eligibility and on the mapping of its credit assessments to the CRD’s credit quality steps. Each national competent authority will then take its own decision on the basis of the joint assessment and any other information it deems relevant.

If additional competent authorities receive applications concerning the
ECAI at a later date, they can choose whether to recognise indirectly the decisions already made by other competent authorities or to undertake their own direct assessment of the ECAI.
Both of these approaches are considered to be important. Indirect
recognition can be a valuable instrument for enhancing efficiency and reducing administrative burdens. The common understanding below should provide a robust foundation for the use of the indirect recognition approach by competent authorities in appropriate circumstances.
Each competent authority shall disclose an explanation of its recognition
process and a list of eligible ECAIs. This information should be included in the information disclosed in the common supervisory disclosure framework set out by CEBS 7 : http://www.cebs.org/SD/Rules_ECAI.htm . Application
The recognition process is initiated when the competent authority receives
an application for recognition. In order to provide the flexibility required by different national circumstances, each competent authority can decide whether to adopt an application process in which the applicants are ECAIs, a process in which the applicants are institutions that intend to use the ECAI's credit assessments for riskweighting purposes, or a process in which the applicant can be either one. 8 Each competent authority shall disclose which application process applies within its jurisdiction. This information should also be included in the information disclosed in the common supervisory disclosure framework set out by CEBS.
When the applicant is an ECAI, it should demonstrate that at least one
institution in the competent authority’s jurisdiction intends to use its credit assessments for prudential riskweighting purposes. This could be achieved by including in the application the name of at least one institution that proposes or intends to use the ECAI’s credit assessments for capital purposes.
As explained in paragraphs 85 to 87 below, an ECAI’s application will be
assessed separately for recognition in each of three main market segments: public finance, commercial entities (including corporates and financial companies), and structured finance (including securitisation). Applications shall indicate in which market segment recognition is being sought, and whether recognition is sought for use of credit assessment for the risk weighting of securitisation positions.
In cases where recognition is being sought in more than one Member
State, it will be important for competent authorities to be made aware of all jurisdictions in which recognition is being sought, so that appropriate 7 CEBS has published guidelines on a common European supervisory disclosure framework. More information can be found on the CEBS website: www.cebs.org 8 Depending on the national legal setting, some competent authorities may wish to ask institutions to channel their applications through their national banking association.

cooperation arrangements between competent authorities can be organised (see paragraphs 33 to 49 below).
25. All applications should be supported by comprehensive, transparent, and
appropriately concise documentation, as indicated below. When applications are initiated by institutions that intend to use an ECAI’s credit assessments, it will be highly desirable for the ECAIs to ensure that all the relevant and material information deemed necessary for the sole purpose of ECAI recognition has been delivered to the competent authorities. The level of recognition
26. Some ECAIs have subsidiaries in different Member States. The CRD does
not specify the level at which the ECAI recognition process should apply:
at the group level or at the subsidiary level.
27. A central question in deciding whether to give recognition at the level of
the group or at the subsidiary level will be whether a given credit assessment grade is judged to represent the same opinion as to the creditworthiness of a given rated entity, regardless of the geographical location where the credit assessment has been issued.
28. If an ECAI group can demonstrate that each of the subsidiaries for which it
seeks recognition adheres to practices and procedures that are set at a groupwide level, then it will not be required to make separate applications for each subsidiary. However, separate applications will be required if subsidiaries use credit assessment practices, methodologies, and procedures that are materially different from those of the group.
29. This approach recognises the organisation of certain crossborder ECAIs,
which apply the same 'core' credit assessment methodology consistently throughout the group. This organisation is considered to ensure the comparability of credit assessments undertaken in different countries, regardless of the legal structure of the ECAI group and notwithstanding the fact that the process of assigning credit assessments may involve credit analysts and rating committees located in a wide range of geographical locations.
30. However, 'grouplevel' applications shall not include 'affiliates' or joint
ventures. Even if they use the same methodologies or comply with the same Code of Conduct as the group they are associated with, some of their characteristics may differ from those of the group under consideration. For instance, their ownership structure may differ from that of the ECAI group. Competent authorities will therefore need to assess those ‘affiliates’ or joint ventures separately. The form of recognition
31. The CRD allows Member States to recognise ECAIs as eligible in two ways:
direct recognition, in which the competent authority carries out its own assessment of the ECAI’s compliance with the CRD eligibility criteria; and indirect recognition, in which the competent authority recognises the ECAI

without carrying out its own direct recognition process 9 , relying instead on the recognition of the ECAI by the competent authority of another Member State. Member States may use a combination of the two approaches.
32. The CRD does not provide for Member States to recognise an ECAI on the
basis of recognition by the competent authorities in a nonEU country. Therefore, in order to be eligible within Europe, ECAIs will have to be recognised by the competent authority of at least one Member State. Application in more than one Member State
33. As mentioned in paragraph 17 above, some ECAIs may apply for
recognition in more than one Member State, or, equivalently, institutions in more than one Member State may submit an application on behalf of the same ECAI. This is likely to occur when the ECAI provides credit assessment services in several Member States, but it may also occur when an ECAI assesses the credit quality of issuers located in only one Member State and those assessments are used by institutions in other Member States.
34. General legal considerations dictate that a separate application for ECAI
recognition must be made in each Member State in which a credit institution intends to use the ECAI’s credit assessments for prudential risk weighting purposes, and that each national competent authority must make its own decision on the ECAI's eligibility.
35. In order to make the recognition process as efficient as possible in such
cases, CEBS recommends that competent authorities seek the greatest possible convergence in approach, both by seeking a consensus on minimum eligibility criteria (this being the goal of the present document), and by cooperating closely in the joint assessment process. 36. Consistency in national supervisory approaches to ECAI recognition has the advantages of:
a. Avoiding unnecessary duplication of effort on the part of competent authorities, and b. Reducing the compliance burden for applicants.
37. In order to facilitate a fully inclusive and coordinated approach, each
competent authority must be aware of all other competent authorities that will be assessing the eligibility of the same ECAI. a. When applications are initiated by an ECAI, the ECAI's application should include a list of any other Member States in which it is seeking or plans to seek recognition in the future. b. When applications are initiated by institutions, competent authorities will take the necessary steps to ascertain which other competent authorities have received applications for the same ECAI, and to i.e. without conducting its own assessment of the ECAI’s compliance with the eligibility criteria of the CRD.

ensure that all of them are aware that a coordinated approach is being initiated.
c. In either case, the CEBS Secretariat could provide logistical support
for the joint process if necessary.
38. Within a month of receiving such an application, all the competent
authorities concerned will meet to determine a single joint process for reviewing the applications.
39. The actual arrangements decided upon will vary on a casebycase basis.
In some cases – for example, when an ECAI operates predominantly in a single Member State it may be appropriate for the assessment to be undertaken by a single competent authority. In other cases, it may be appropriate for a subset of interested competent authorities to undertake a joint assessment, with other competent authorities choosing – on a voluntary basis – not to participate directly. In yet other cases, all the competent authorities concerned may wish to be involved in a single joint process.
40. In any case, for the sake of efficiency, a ‘process facilitator’ will be
appointed who will carry out the tasks of coordinating and ultimately producing the joint assessment. The ECAI concerned will be informed of the type of recognition process to be undertaken and of the identity of the process facilitator.
41. The authorities involved will agree among themselves on the process
facilitator. For this purpose, the following nonexhaustive set of criteria may be used:

The extent to which an ECAI is focused on or will be used in a
particular territory or territories.
The relationship between the supervisor and the ECAI. Some ECAIs
may maintain regular contacts via businessrelated issues and events.
Administrative convenience, burden optimisation, and an appropriate
distribution of tasks.

The role of the process facilitator consists of coordinating and ultimately
producing the joint assessment report. It includes:

Ensuring that all the information necessary to carry out the joint
process is shared among the members,
Ensuring that a precise timetable and workplan are established for
carrying out the joint assessment within a reasonable timeframe. This should include facilitating agreement on the division/allocation of tasks amongst participants,
Acting as the point of contact for the interested parties – including
ECAIs, institutions and other competent authorities – collecting and disseminating information as required,
Facilitating agreement on requests for further information,
Ensuring that supervisors adhere to the agreed working plan and
ensure the timely delivery of the report, and
Ensuring that the report is produced and made available to all
interested competent authorities. (see para 47 to 49 below).

The aim of the joint process is to reach a shared view on the ECAI’s
eligibility and on the mapping of its credit assessments to the CRD’s credit quality steps. This shared view will be the outcome of a joint assessment of the ECAI’s methodology and its credit assessments.
Each competent authority will then take its own decision as to whether to
recognise the ECAI, on the basis of the joint assessment and any other information it deems relevant.
Participation in the joint assessment does not preclude a national
competent authority from undertaking any additional assessment it deems appropriate. In particular, competent authorities may judge that an ECAI's organisation or credit assessment methodology raises countryspecific issues that are not fully captured by the broader consideration of the ECAI's eligibility at a group level during the joint assessment. Competent authorities may wish to take such national specificities into account in coming to their own eligibility decisions.
The joint assessment will be made on the basis of the ECAI's answers to
the ‘common basis application pack’ outlined in Annex I of this paper, any additional supporting information provided by the applicant, and any other information requested by the process facilitator. The joint assessment will also be based on the guidelines on recognition criteria issued by CEBS.
Once the joint assessment of an ECAI has been completed, it will be made
available to all competent authorities that have received applications for recognition of that ECAI, or where the ECAI has indicated that it intends to make an application.
The joint assessment report will be provided upon request to the
competent authorities that did not receive an application. 10 In view of the confidentiality of the supporting documentation of the application pack provided by the ECAIs and upon which the report has been drafted, this will be subject to the prior consent of the ECAI.
The joint assessment report itself will be communicated to the ECAI with
the agreement of all the supervisors involved in the joint process. The way in which the ultimate decision of a competent authority is communicated to the applicant will be determined in accordance with the national legal setting. Provision of information to support the application
It is important that the minimum information to be provided by applicants
is commonly agreed, in order to reduce inconsistent information requirements by individual competent authorities. 10 The CRD does not require competent authorities using indirect recognition to conduct their own assessment.
The adoption of similar information requirements will also facilitate
cooperation between competent authorities and encourage greater use of indirect recognition, since competent authorities will be more inclined to rely on each other if they have some comfort that decisions are based on similar procedures and types of input.
Without prejudice to domestic language requirements, in order to facilitate
the joint recognition process, documentation should be provided in a language of mutual understanding. The language will be determined on a casebycase basis.
Competent authorities should be guided by two principles in determining
how much and what level of information to request from ECAIs. On the one hand, they should obtain sufficient information to be able to make a wellinformed decision as to whether an ECAI satisfies the CRD’s recognition criteria. On the other hand, they should request only that information they need to exercise their responsibilities under the CRD, i.e. only that information that is necessary to assess an ECAI’s eligibility for prudential capital purposes.
In order to balance these twin objectives, CEBS has developed the
‘common basis application pack’ set forth in Annex 1 to this paper. CEBS proposes that competent authorities use this pack as the basis for their requests for information from ECAIs. If necessary, competent authorities can ask for complementary information and documentation on a caseby case basis, to be discussed, if needed, in additional contacts with the ECAI.
The joint assessment will be based on the documentation provided by the
applicant in the application pack. All communication with the ECAI concerning the joint assessment, including any request for additional information deemed necessary for it, will be made by the process facilitator to the ECAI on behalf of the competent authorities involved.
The use of the common basis application pack during the joint assessment
process does not preclude the competent authorities from asking for the information necessary to address any countryspecific issues they have identified, before they take their decision.
The application pack permits competent authorities to consider the
granularity 11 of the ECAI's methodology, on both an asset class and market segment and geographical basis. For crossborder ECAIs, such information will be used to assess the extent to which separate recognition processes are needed for specific asset classes or geographical locations.
CEBS believes that much of the information called for in the application
pack will already be available in ECAIs’ existing documentation, and that they should therefore be able to provide concise answers. However, some additional information may be needed, in particular concerning default 11 ECAIs generally use different rating methodologies for exposures in different asset classes and/or different geographical regions. ‘Granularity’ refers to the fineness of these divisions. For example, using a distinct methodology for French residential mortgagebacked securities is more granular than using one methodology for all European structured finance.

data and other quantitative tools on which ECAIs base their opinion of the creditworthiness of an entity. Review of application by the competent authority
59. Competent authorities shall base their recognition decisions on an
assessment of the objectivity, independence, ongoing review, and transparency of the ECAIs’ methodologies and the credibility and transparency of their credit assessments.
60. CEBS has developed a set of guidelines for meeting these criteria, to be
followed by all competent authorities. These guidelines are set out in Part 2 of this paper. Disclosure by the competent authority
61. Article 81(4) of the CRD requires competent authorities to make publicly
available an explanation of the recognition process and a list of eligible ECAIs.
62. In line with the CEBS supervisory disclosure framework mentioned in
paragraph 20 above, this disclosure should include the name of each eligible ECAI along with the mapping that the competent authority has established between the ECAI’s credit assessments and the ‘credit quality steps’ set out in the CRD. The disclosure should also include information on how the recognition process has been initiated, whether a joint process has been carried out, the market segments for which recognition has been sought, and the mapping used. The disclosures should be kept updated. Ongoing review of eligibility
63. To ensure that the credit assessments used by institutions in calculating
their capital requirements remain of sufficiently high quality, competent authorities will need to assess whether ECAIs to which they have granted recognition continue to meet the eligibility criteria of the CRD on an ongoing basis.
64. As with initial assessments, ongoing assessments should be tailored to the
purposes of the CRD and should be limited to ensuring that eligible ECAIs continue to meet the criteria that led to their initial recognition.
65. Specifically, competent authorities shall monitor the effects of any material
changes that ECAIs have reported to them in accordance with Annex VI, Part 2, paragraph 6 of the CRD and as spelled out in paragraph 107 below.
66. In addition, competent authorities will undertake an analysis in an
appropriate depth of each ECAI’s eligibility every five years.
67. Competent authorities which have been directly involved in an initial joint
recognition process will cooperate and jointly carry out the ongoing review. A process facilitator shall act as a contact point with the ECAI and facilitate coordination among the relevant competent authorities.

68. In cases where the ECAI has been directly recognised by additional
competent authorities at a later date (see paragraph 18 above), these competent authorities will be invited to participate in the joint review.
69. Competent authorities shall withdraw the recognition of any eligible ECAI
that ceases to comply with the CRD recognition criteria, after first discussing the matter with the ECAI. Any withdrawal will be communicated to the other competent authorities.
Part 2: Common understanding of the ECAI recognition criteria laid down
in the CRD
General principles
70. The key purpose of the recognition criteria is to identify ECAIs that
produce external credit assessments of sufficiently high quality, consistency and robustness to be used by institutions for regulatory capital purposes under the Standardised Approach and the Securitisation Ratings Based Approaches.
71. For this purpose, competent authorities will investigate whether the ECAI
has processes and procedures in place which ensure that credit assessments meet the standards stated above.
72. In determining which ECAIs are eligible for the Standardised Approach,
competent authorities shall take into account the technical criteria set out in Annex VI, Part 2 of the CRD. 12
73. Moreover, according to Article 97(2) of the CRD, competent authorities
shall recognise an ECAI as eligible for the purposes of Article 96 – calculation of the riskweighted amount of a securitisation position only if they are satisfied that it has complied with the requirements laid down in
Article 81, taking into account the technical criteria in Annex VI, Part 2,
and that it has demonstrated ability in the area of securitisation which may be evidenced by a strong market acceptance. In addition, the credit assessments shall comply with the principles of credibility and transparency as elaborated in Annex IX, Part 3.
74. Competent authorities must treat all ECAIs equally. However, given the
different business models adopted by individual ECAIs, competent authorities may need to take a differentiated approach to assessing how they satisfy the CRD recognition criteria. Accordingly, they may place different weights on the various criteria for different ECAIs – subject of course to their all satisfying these CRD criteria – if this serves the ultimate objective of establishing whether an ECAI's methodology and credit assessments are suitable for calculating regulatory capital requirements. 12 Article 81(2): "competent authorities shall recognise an ECAI as eligible for the purposes of Article 82 only if they are satisfied that its assessment methodology complies with the requirements of objectivity, independence, ongoing review and transparency, and that the resulting credit assessments meet the requirements of credibility and transparency. For those purposes, the competent authorities shall take into account the technical criteria set out in Annex VI, Part 2."

75. In coming to an assessment, competent authorities will take into
consideration the ability of the ECAI to produce robust credit assessments, based on quantitative methods and a proven data track record.
76. Competent authorities will also take account of market indications of the
ECAI's standing. For example, strong market acceptance and the existence of a long track record may be viewed as indications that the market has a favourable opinion of the ECAI's methodology and credit assessments. As indicated in paragraphs 112 to 116, the relevant threshold for eligibility is that the assessments are recognised as credible and reliable by market users. Stronger market acceptance is not a prerequisite, but rather a factor which is relevant in determining the intensity of assessment of other factors.
77. In addition, the extent to which an ECAI adheres to a code of conduct
which is in line with market standards and internationally recognised principles such as the IOSCO Code of Conduct Fundamentals for Credit Rating Agencies and the public disclosure of such a code may contribute to satisfying competent authorities that the ECAI conforms to certain CRD criteria, such as the independence criterion. This could reduce the amount of analysis that the competent authorities themselves need to undertake in order to verify the ECAI's eligibility in these areas.
78. The implementation of a recognition regime may be perceived as creating
barriers to entry in the market for external credit assessments. While seeking to avoid any unnecessary interference with the market, competent authorities recognise that their primary objective is the prudential need to identify which ECAIs are eligible for risk weighting purposes. The technical criteria13
1. Methodology
1.1. Objectivity
Competent authorities shall verify that the methodology for assigning credit assessments is rigorous, systematic, continuous, and subject to validation based on historical experience.
79. The purpose of this criterion is to ensure that an ECAI’s credit assessment
methodology produces an informed and wellfounded opinion on the creditworthiness of the rated entities, and that its credit assessments are based on all information deemed relevant and available at the time they are issued.
80. In meeting this criterion, an ECAI will need to demonstrate that its
methodology incorporates factors known to be relevant in determining an entity's creditworthiness. This demonstration should, to the fullest extent possible, be supported by statistical evidence that the methodology has produced accurate credit assessments in the past. 13 The boxed text represents the technical criteria set out in Annexes VI and IX of the CRD. When there are differences between the two Annexes, they are explicitly mentioned.

81. The ECAI must also implement and follow procedures which ensure that its
predefined credit assessment methodology is applied consistently in the formulation of all credit assessments in a given asset class, such that two identical companies would receive equivalent credit assessments, and different analysts or rating committees within the ECAI would assign equivalent credit assessment to any given entity.
82. Preliminary discussions with some ECAIs indicate that they do not have a
single credit assessment methodology. This is to some extent evidenced by their publication of numerous papers describing their credit assessment methods for different asset classes (e.g. corporates versus institutions), for different subclasses (e.g. automobile manufacturers versus oil companies) and for different geographical regions (e.g. European auto manufacturers versus US auto manufacturers).
83. However, ECAIs appear to apply a similar ‘core’ credit assessment
methodology within broad asset classes or market segments. The same factors are identified as significant in determining a credit assessment for all entities within the broad asset class, although different emphasis may be put on the importance of individual factors when assessing different companies within that broad asset class.
84. In defining broad asset classes and/or market segments, it is not
necessary for the core assessment factors to be evaluated in an identical and mechanical way for all entities within a group. Indeed, it is expected that an ECAI will place different emphasis on the importance of individual factors when assessing different companies and/or markets, and will take such differences into account. What is important is that the same core factors are always considered, to some extent, when assessing an entity within the given asset class and/or market segments.
85. With regard to the definition of these broad groups, it appears that ECAIs
generally use similar credit assessment methodologies within each of the following three broad asset classes or market segments: structured finance, public finance, and commercial entities (including corporates and financial companies).
86. CEBS proposes that such broad asset classes or market segments form the
basis of the ECAI recognition process with separate assessment of the ECAI's methodology by competent authorities in each of the broad asset classes. This is not intended to prevent ECAIs from seeking recognition for a methodology which is more specifically focused, e.g. on SMEs.
87. When assessing an ECAI’s methodology in any of these broad asset
classes or market segments, competent authorities will avoid making a direct judgment as to whether an ECAI's methodology is objectively correct. They should not be seen as endorsing any particular type of methodology.
88. Instead, competent authorities should concentrate on assessing whether
the credit assessment processes adopted by an ECAI produce credit assessments that embody a sufficient level of consistency and discrimination to provide the basis for capital requirements under the Standardised and Securitisation Ratings Based Approaches.

89. This assessment should focus on three factors:
a. Quantitative evidence of the discriminatory power of the ECAI's credit assessment methodology, using statistical techniques such as default studies and transition matrices to demonstrate the robustness and predictive power of credit assessments over time and across different asset classes; b. The ECAI’s demonstration that it has processes in place to assess factors driving creditworthiness and to ensure that these factors are incorporated into the credit assessment methodology; and
c. The ECAI’s demonstration that it has procedures which ensure that its
predefined methodology is applied consistently in the formulation of all credit assessments.
90. Where appropriate, competent authorities should use quantitative
evidence of the consistency and predictive power of an ECAI's credit assessments (the outputs of their methodological process) as an indicator of the objectivity of its methodological processes (the inputs of the methodological process). When ECAIs have a demonstrable track record of producing robust credit assessments (outputs) using quantitative methods such as default or transition studies, competent authorities should view this as a good indication that its methodological processes (inputs) are sufficiently objective for the purposes of the CRD. This may reduce the level of assessment that competent authorities themselves have to undertake.
91. In cases where there is less quantitative evidence to support the
robustness of an ECAI's credit assessments (outputs), competent authorities will need to undertake a greater assessment of the ECAI’s methodological process (inputs) in order to be satisfied that the ECAI’s methodology meets the objectivity criterion.
92. As indicated above, the assessment of the ECAI’s methodology should
remain highlevel. Competent authorities should not undertake a detailed assessment of the exact methodology used by the ECAI, but should instead satisfy themselves that the credit assessment drivers used in the ECAI's methodology are sensible predictors of creditworthiness, and that the ECAI's internal procedures ensure that its predefined credit assessment methodology is applied consistently in the formulation of all credit assessments within each broad asset class or market segment.
93. The CRD requires competent authorities to verify that ECAIs validate their
methodologies based on historical experience. As indicated above, quantitative validation will need to be based on the ECAI's credit assessments (the outputs of the methodology) rather than on the methodology itself. ECAIs should demonstrate that the methods they use in their quantitative assessment confirm the robustness, discriminatory power, and consistency of their credit assessments over time and across different market segments. In addition, ECAIs should demonstrate that procedures are in place to ensure that systematic rating errors highlighted by backtesting will be incorporated into credit assessment methodologies and corrected.

1.2. Independence
Competent authorities shall verify that the methodology is free from external political influences or constraints, and from economic pressures that may influence the credit assessment. Independence of the ECAI’s methodology shall be assessed by competent authorities according to factors such as the following:
(a)ownership and organisation structure of the ECAI (b)financial resources of the ECAI (c) staffing and expertise of the ECAI (d)corporate governance of the ECAI
94. This criterion is intended to ensure that all credit assessments issued by
ECAIs are independent and objective in all circumstances, including when conflicts of interest may arise.
95. Conflicts of interest may arise as a result of external political or economic
pressures. Examples include the following situations:

The ECAI is owned by a government, trade association, or political
body that has an interest in securing favourable credit assessments for its constituent entities.
The ECAI is owned by a private company which could use its position
to secure favourable credit assessments.
The ECAI's financial position depends on revenue from key customers
who could seek to leverage their position to secure favourable credit assessments.
The ECAI provides ancillary services to rated entities or has other
business relationships with them that could undermine the objectivity of its credit assessments.
An ECAI employee is in a managing position in a rated entity.
The ECAI’s staff is compensated in a way, or they have business
relationships with the rated entities, that could lead to nonobjective credit assessments.

An institution will not be allowed to nominate for its own capital purposes
an ECAI which is its subsidiary.
Competent authorities must be comfortable that ECAIs have procedures in
place to ensure that their methodologies are free from political influences or constraints and from economic pressures that may influence the credit assessments. In order to satisfy competent authorities on this point, ECAIs will need to demonstrate:
a. That they have adopted, monitored, and successfully applied internal procedures to ensure that all credit assessments are formulated in a

consistent and objective manner, particularly in situations where conflicts of interest may arise and could threaten objectivity; b. That they have mechanisms in place to identify actual and potential conflicts of interest and take reasonable measures to prevent, manage and eliminate them, so that they do not impair the production of independent, objective, and highquality credit assessments.
98. CEBS considers that ECAIs themselves are in the best position to design
internal procedures, fee policies, staff management practices, corporate governance rules, and internal codes of conduct that manage potential conflicts of interests and ensure that their credit assessment methodologies are free from political and economic influences. It should be demonstrated that ECAIs have adopted appropriate internal practices and procedures in this respect, and in particular in the following areas:
a. It should be demonstrated that ECAIs have put in place, and apply, adequate safeguards to ensure their independence from ownership, and to prevent external pressure or constraints – either political and economic – from jeopardising the objectivity of the credit assessment process. b. It should be demonstrated that ECAI’s organisational structure separates its credit assessment business – operationally, personally and potentially legally – from any other business, such as consulting services, that undermine the objectivity of the credit assessments.
c. With regard to financial resources, it should be demonstrated that
ECAIs are sufficiently robust and have adequate safeguards in place to ensure independence from key customers and issuers, and prevent nonobjective credit assessments. d. In terms of staffing and expertise, it should be demonstrated and selfcertified that ECAI’s staff has the levels of skills and experience necessary to perform the tasks required of them for example, that at least one person involved in the rating decisionmaking process has at least three years experience as a rating analyst or in a comparable function (e.g. as an analyst in a credit institution). The ECAI should also have enough resources to carry out consistent assessments and to have frequent contacts with the rated companies when this represents a necessary part of their methodology. e. ECAIs should have an independent internal audit function. Those that do not (e.g. nonpublicly listed entities) should have a function that plays the same role and carries out the same tasks. f. The integrity of the credit assessment process should be ensured by adequate written internal procedures, corporate governance rules, fee policies, and, where relevant, an internal code of conduct. g. In order to promote independence through transparency and market scrutiny, ECAIs should consider disclosing situations where conflicts of

interest have arisen or may potentially arise, and the mechanisms in place to identify, prevent, manage, and eliminate conflicts of interest.
1.3. Ongoing Review
Competent authorities shall verify that ECAI's credit assessments are subject to ongoing review and shall be responsive to changes in the financial conditions. Such review shall take place after all significant events and at least annually. Before any recognition, competent authorities shall verify that the assessment methodology for each market segment is established according to standards such as the following:
(a) The backtesting must be established for at least one year. (b) The regularity of the review process by the ECAI must be monitored by the competent authorities. (c) The competent authorities must be able to receive from the ECAI the extent of its contact with the senior management of the entities which it rates. Competent authorities shall take the necessary measures to be promptly informed by ECAIs of any material changes in the methodology they use for assigning credit assessments.
99. The purpose of the ongoing review criterion is to ensure that the ECAI's
external credit assessments remain appropriate over different periods of time and through changes in market conditions.
100. The CRD does not define the terms ‘changes in financial conditions’ or
‘significant events.’ However, these terms are linked; both refer to any event (financial or otherwise) that is large enough to potentially or actually change the credit assessment assigned by an ECAI to an entity.
101. Competent authorities will not themselves undertake any ongoing review
of the credit assessments of the ECAI. They will instead verify that ECAIs have procedures in place to ensure that their credit assessments remain appropriate over different time periods and market conditions. In particular, competent authorities will require ECAIs to demonstrate that they have processes in place that:
a. Reliably detect changes in conditions facing a rated entity that are large enough to potentially change its assignment to a credit assessment category, and b. Ensure that a credit assessment is indeed revised when the change in operating conditions is large enough to warrant a revision.
102. It will also have to be demonstrated that the ECAI reviews each credit
assessment at least annually (regardless of whether a reassessment has already been undertaken in response to a significant change in financial conditions). The ECAI should provide a detailed summary on how these

reviews are conducted, including the extent of contacts with the senior management of the rated entity.
103. The backtesting requirement in paragraph 5(a) of Annex VI of the CRD is
viewed as an additional criterion that ECAIs must satisfy before they are granted recognition. One year of backtesting has been deemed necessary to fulfil the criterion "subject to validation based on historical experience." ECAIs will therefore be required to demonstrate and certify that their backtesting has been in place for at least one year.
104. The term 'backtesting' means an analysis of 'outcomes' visàvis rated
entities/issues designed to assess the 'performance' (e.g. the discriminatory power) of the credit assessments. Backtesting is thus synonymous with the ‘validation based on historical experience’ that is mentioned in the CRD’s ‘objectivity’ criterion. For the sake of consistency, backtesting should be undertaken for each of the ‘market segments’ for which an ECAI is seeking recognition (as explained in paragraphs 85 to 87 above).
105. The requirement for ECAIs to inform competent authorities of material
changes in their credit assessment methodology is intended to enable competent authorities to assess whether the methodologies continue to meet the CRD criteria on an ongoing basis after initial recognition has been granted.
106. Preliminary discussions with market participants indicate that it is
important to clarify the term ‘material change in methodology used for assigning credit assessments’ in order to avoid overwhelming competent authorities with information.
107. Competent authorities should require ECAIs to inform them immediately of
any significant event that would change their performance on any criteria upon which initial recognition was granted. This should be construed to mean any change in methodology that could change a significant proportion of credit assessments in a given market segment.
1.4. Transparency and disclosure
Competent authorities shall take the necessary measures to assure that the principles of the methodology employed by the ECAI for the formulation of its credit assessments are publicly available as to allow all potential users to decide whether they are derived in a reasonable way.
108. The CRD does not specify either the level of detail that is required or the
manner in which the information should be disclosed. Nor does the CRD require competent authorities to elaborate disclosure principles for ECAIs, or to provide a comprehensive disclosure framework where all the information that ECAIs are required to submit could be published. ECAIs are not required to disclose the same information to the public as to competent authorities.
109. ECAIs should disclose the principles of their methodology to the public.
This shall be an overall yet thorough description of their credit assessment

methodologies, presented in a way that is easily understandable to potential users.
110. ECAIs should also disclose as promptly as possible material changes in
methodology referred to in paragraph 6 of Annex VI of the CRD and in paragraph 107 above.
111. ECAIs should use appropriate methods of disclosure to ensure public
access to the abovementioned information. These methods could include display in the public area of the ECAIs’ Internet website or free of charge distribution of written publications on request.
2. Individual credit assessments
2.1. Credibility and Market Acceptance
Competent authorities shall verify that ECAI's individual credit assessments are recognised in the market as credible and reliable by the user of such credit assessments. Credibility shall be assessed by competent authorities according to factors such as the following:
(a) market share of the ECAI;
(b) revenues generated by the ECAI, and more in general financial resources of the ECAI; (c) whether there is any pricing on the basis of the rating. (ca) in case at least two banks use the ECAI’s individual credit assessment for bond issuing and/or assessing credit risks. An ECAI credit assessment may be used to determine the risk weight of a securitisation position (Article 97(1) of the CRD) only if the competent authorities are satisfied that the ECAI has a demonstrated ability in the area of securitisation, which may be evidenced by a strong market acceptance.
112. Evidence of widespread use in the market – by investors, for example –
indicates that market participants have a favourable opinion of the credibility and reliability of the ECAI’s credit assessments.
113. The fact that the market regards an ECAI’s credit assessments as credible
and reliable may provide competent authorities with a significant degree of confidence as to the appropriateness of the credit assessments as the basis for capital requirement calculations under the Standardised and Securitisation Ratings Based Approaches. The greater the credibility and reliability, the higher this level of confidence is likely to be.
114. Conversely, a lower degree of market standing means that competent
authorities may themselves need to undertake a greater level of assessment before they can be satisfied that an ECAI fulfils the recognition requirements.

The factors outlined in the boxed text above will be considered in the
context of the market in which the ECAI is operating and in the context of which recognition is sought.
Competent authorities may also wish to consider other indicators of
market credibility not mentioned in the CRD. For example, evidence that a large number of institutions plan to use an ECAI's credit assessments for regulatory capital or other purposes may be viewed as an indication of market credibility for the purpose of ECAI recognition.
2.2. Transparency and disclosure of individual credit assessments
Competent authorities shall verify that individual credit assessments are accessible at equivalent terms at least to all credit institutions having a legitimate interest in these individual credit assessments. In particular, competent authorities shall verify that individual credit assessments are available to nondomestic parties on equivalent terms as to domestic credit institutions having a legitimate interest in these individual credit assessments.
Annex IX, Part 3, Paragraph 1 states that credit assessments of securitisation
positions shall be available publicly to the market. Credit assessments are considered to be publicly available only if they have been published in a publicly available forum and they are included in the ECAI’s transition matrix. Credit assessments that are made available only to a limited number of entities shall not be considered to be publicly available.
The transparency criterion is intended to create a level playing field by
ensuring that all institutions "having a legitimate interest" in credit assessments published by the ECAI (public credit assessments), in whatever jurisdiction, have equal and timely access to them.
Institutions "having a legitimate interest" are those institutions that use
the Standardised Approach and the Securitisation Ratings Based Approaches to calculate their regulatory capital requirements, and that intend to use the credit assessments of the respective ECAI for risk weighting purposes. ECAIs that wish to be recognised as eligible must make their public credit assessments accessible at least to all institutions fulfilling these criteria.
"At equivalent terms" does not mean that every institution must be
provided access to the credit assessments on identical terms and in particular, that there be no discrimination in terms of the pricing for access – but rather that under the same (economic) circumstances, there should be no undue (price) discrimination. Competent authorities should pay particular attention to differences in the terms and prices offered to domestic versus nondomestic institutions.
CEBS considers it necessary to differentiate between situations where
ECAIs do not charge subscribers for access to their public credit assessments and situations where they do.

a. Competent authorities should require ECAIs that do not charge subscribers for access to their public credit assessments to ensure that a full list of their public credit assessments is available and updated whenever a new credit assessment is issued or an old assessment is revised. One possibility would be for the ECAI to publish a full list of its public credit assessments in the public section of its Internet website and to update the list whenever such a credit assessment is newly issued or revised. b. Competent authorities should require ECAIs that permit only paying subscribers to access their credit assessments to ensure that the complete range of its public credit assessments is potentially available to all subscribing institutions and that the list is updated as soon as such a credit assessment is newly issued or revised.
Part 3: Mapping
121. When determining which of the credit quality steps the relevant credit
assessments of an eligible ECAI are to be associated with, competent authorities shall apply the technical criteria laid down in the CRD14 . Article 82(1) and Article 98(1) of the CRD state that those determinations shall be objective and consistent.
122. Objectivity and consistency in mapping are necessary in order to ensure
appropriate levels of capital under the Standardised Approach and the Securitisation Ratings Based Approaches, a level playing field for institutions, and fairness of treatment for ECAIs.
123. At the same time it is recognised that absolute accuracy in the mapping
process is likely to be neither possible given data constraints and differences in methodology nor necessary to achieve the objectives of the Standardised Approach and the Securitisation Ratings Based Approaches. General Principles
124. The mapping process should not imply the imposition of additional
eligibility requirements on ECAIs.
125. For the purposes of benchmarking and monitoring ECAIs' credit
assessments (other than for securitisation and structured transactions), it is recommended that competent authorities use the guidance provided by the Basel Committee in the Basel II framework published in June 2004, Annex 2 (‘the Basel II text’). To access the document, please click on either http://www.bis.org/publ/bcbs107.pdf or http://www.c ebs.org/documents/Basel.pdf 14 Annex VI, Part 2 for the Standardised Approach; Annex IX for securitisation positions

126. The use of threeyear Cumulative Default Rates (CDRs), evaluated over
the longer term (see paragraph 140 below) and on an ongoing basis, is considered to provide an appropriate measure of the predictive power of credit assessments in relation to creditworthiness. This choice is consistent with the requirements of the CRD.
127. Where significant amounts of quantitative data are available, they will
form the central basis of the mapping process. However, the mapping process will also take into account qualitative factors which influence the comparability of the credit assessments’ CDRs with the benchmark CDRs (e.g. differences in the definition of default, the methodology for calculating CDRs, etc.).
128. Where significant amounts of quantitative data are not available, competent authorities will form their judgement based on both whatever
quantitative information is available and an assessment of the meaning of the ECAI's rating scale in comparison with the benchmark. In this situation, competent authorities may take into account the ECAI’s own comparison. This judgement will use whatever quantitative information is available, but will be based mainly on a qualitative comparison, incorporating appropriate conservatism where uncertainty remains.
129. Supervisory authorities will base their assessments on the credit
assessment models, processes, and methodologies presented by the ECAIs, and will in no way seek to influence or change these models, processes, or methodologies.
130. The granularity of the mapping process is not linked to the granularity of
the methodology used by an ECAI. As long as an ECAI uses the same rating scale (i.e. the same interpretation of the different rating categories) for their broad asset classes, the mapping need not be conducted separately.
131. ECAIs will be required to communicate their default rates and the data
related to the mapping of securitisation positions annually to the competent authorities, in order to allow them to assess whether the mapping of credit assessments to credit quality steps needs to be updated or changed.
132. The cooperative arrangements set up for the recognition process will also
be followed by the competent authorities when carrying out the mapping jointly. For the ongoing monitoring of the mapping referred to in paragraph 131 above, this implies that a process facilitator will receive the data and dispatch them to all competent authorities concerned.
133. This will contribute to a consistent assignment of credit assessments to
credit quality steps across Member States.
Credit assessments of exposures other than securitisation positions
134. The CRD sets out quantitative and qualitative factors to be taken into
consideration.

1. Quantitative factors
135. Quantitative data are the key to ensuring consistency between the credit
assessments of different ECAIs and to differentiating between the relative degrees of risk expressed by each credit assessment.
136. The CRD suggests that competent authorities use, as quantitative data,
“the longterm default rate associated with all items assigned the same credit assessment.”
137. Moreover, the CRD requires the competent authorities to “compare default
rates for each credit assessment of a particular ECAI and compare them with a benchmark built on the basis on default rates experienced by other ECAIs on a population of issuers that the competent authorities believes to present an equivalent level of credit risk.”
138. CEBS considers the work conducted to date by the Basel Committee on
quantitative factors to be both relevant and appropriate, and proposes to adopt the benchmark and monitoring guidance set out in Annex 2 of the Basel II text.
139. In this context, the key variable will be the “cumulative default rates”
(CDRs) over a threeyear period: that is, the sum of all defaults that have occurred in a given threeyear period for all rated items belonging to the same bucket.
140. Supervisors will be provided with two separate measures of CDRs: the ten
year average of the threeyear CDR as an indicator of the longterm default experience of individual ECAI’s credit assessments, and the two most recent threeyear CDRs, where available.
141. Using the data provided by the ECAI, competent authorities will compare
the most recent tenyear average of the threeyear CDR with the proposed longrun ‘reference’ threeyear CDRs in Table 2 of Annex 2 of the Basel II text15
142. In additional, competent authorities will monitor the two most recent
threeyear CDRs and compare them with the two different CDR levels established in the Basel II text: the ‘monitoring level’ and the ‘trigger level’. 16 The methodology laid out in Annex 2 of the Basel II text will help supervisors assess whether the ECAI’s default rates are materially and systematically higher than these benchmarks.
143. Based on this assessment, competent authorities will decide whether to
assign a less favourable credit quality step. If the ECAI can demonstrate that higher observed or estimated CDRs are not due to weaker assessment standards or miscalculations, competent authorities may decide to leave the initial mapping unchanged. 15 It should be noted that the numbers provided for the longterm benchmarks are midpoint numbers. Consequently, supervisory authorities will not expect the data provided by ECAIs to coincide exactly with these numbers. 16 In this case, both benchmarks are considered as upper limits.

144. For recently established ECAIs and those that have compiled only a short
record of default data, competent authorities will ask the ECAI for its two most recent CDRs and a projection of the tenyear average of the three year CDR, i.e. the value that the ECAI believes to be the longterm default rate associated with all items assigned the same credit assessment.
145. Competent authorities will review this assessment on the basis of the
availability of data and the methodology used by the ECAI in question, comparing it with those used to calculate the benchmark. Based on the consideration of such qualitative factors, competent authorities may then adjust the mapping of the ECAI accordingly. Where uncertainty remains, competent authorities should incorporate appropriate conservatism into the final mapping.
146. For ECAIs that adopt significantly different approaches, supervisors should
consider adjusting their assessment on the basis of qualitative factors as set out below.
2. Qualitative factors
147. The CRD requires competent authorities to consider qualitative factors
such as the pool of issuers covered by the ECAI, the range of credit assessments that it assigns, the meaning of each credit assessment, and the ECAI’s definition of default.
148. Qualitative factors will play a crucial role in the mapping process in the
following situations:

When the ECAI uses methodologies (e.g. definition of default, etc.)
similar to those used by the international entities upon which the Basel Committee constructed its benchmarks, competent authorities will use qualitative factors to adjust their quantitative assessment before finalising the assignment of each of the ECAI’s credit assessments to the credit quality steps established in the CRD.
When the ECAI uses different methodologies, it will be required to
provide its own assessment of whether and to what extent its methodology differs from that used to calculate the benchmarks in the Basel II text. In this way, supervisors will get a better understanding of what a credit assessment represents and the risk level associated with it. This may lead to the assignment of similar CDR data to different credit quality steps from those set out in the Basel text.
Recently established ECAIs and ECAIs that have compiled only a short
record of default data would need to demonstrate to what extent they believe that the default data they use are a longterm default rate. Qualitative factors would be particularly importance in making that demonstration.
149. In their assessment, competent authorities will take into account the
qualitative variables indicated in the CRD:

a. The definition of default. An ECAI using a more stringent definition of default than that used in the international benchmark will report more default events, meaning that CDRs could be overstated. The opposite situation could also occur: i.e an ECAI using a less stringent definition of default. b. The pool of issuers covered by the ECAI. ECAIs may use a static pool of issuers or adjust the pool periodically, for example for withdrawn credit assessments.
c. The statistical significance of ECAIs’ default rates. In particular, the
number of rated issues shall be sufficiently large to ensure the statistical significance of CDRs. Particular attention will be paid to situations where the ECAI is sectorallyfocused or geographically specialised, or where the ECAI rates portfolios for which default data are very scarce. d. The meaning of the credit assessment, i.e. the substance of the opinion represented by a particular rating grade.
150. In addition to the qualitative factors set out in the CRD, competent
authorities should consider other relevant factors such as:
e. The variable used to weight default events. Different variables, such as the number of issues, the currency value of exposures rated, or other characteristics, can be used to weight default events. The choice of variable may have an impact on the results. f. Geographic coverage: the use of regional or global data. g. Dynamic properties and characteristics of the rating system or methodology (a ‘pointintime’ rating system or a ‘through the cycle’ system). This can be assessed in general terms without entering into the details of the ECAI’s default model.
151. Supervisors may consider the mapping on the basis of additional
information and analysis provided by the ECAI.
Credit assessments of securitisation positions
152. The CRD17 requires a separate mapping of credit assessments
securitisation positions. For ease of reference, the paragraphs below refer to this simply as ‘securitisation mapping.’
153. Securitisation mapping and the mapping of credit assessments discussed
above follow the same principles of objectivity and consistency. However, there are likely to be important differences. First, securitisation transactions have unique characteristics, and the market is highly innovative and constantly evolving. Second, securitisation mapping under 17 Article 98(3) requires competent authorities to determine with which of the credit quality steps set out in Annex IX the relevant credit assessments of an eligible ECAI are to be associated. Those determinations shall be objective and consistent.

the Internal RiskBased (IRB) Approach would be more finely graduated than the mapping of general credit assessments under the Standardised Approach.
154. Unlike the mapping of other ECAI credit assessments, the CRD does not
require competent authorities to create a 'benchmark' for default rate comparison. This reflects the difficulties that would surround the creation of such a benchmark at this stage in the development of the securitisation market. The CRD18 requires competent authorities to consider “quantitative factors, such as default rates and loss rates and qualitative factors such as the range of transactions assessed by the ECAI and the meaning of the credit assessment.”
155. Preliminary discussions suggest that individual ECAIs construct their
methodologies specifically to give credit assessments a consistent meaning across different structured product classes. This implies that competent authorities could treat structured products as a single market segment for mapping purposes, and that the mappings used for securitisation transactions will be those derived for all structured products.
156. Securitisation mapping, like the mapping of other credit assessments,
should be carried out within the framework of the joint assessment process outlined above.
157. In mapping securitisation position credit assessments into the credit
quality steps of the CRD, supervisors will take into consideration quantitative factors and qualitative information, including those set out in the following paragraphs. 1.Quantitative factors
158. Quantitative factors will be a key consideration in mapping securitisation,
as they are in mapping other credit assessments. CEBS recognises that many potential ECAIs do not target quantitative outcomes for their assessments, seeking instead to achieve consistent rank ordinal assessments. Nonetheless, consideration of quantitative 'performance' studies of those assessments over time, in line with the CRD’s requirement, is a key element in providing a mapping in a consistent and objective manner.
159. Competent authorities will consider data relating to the default/impairment
rates associated with different credit assessments. Competent authorities remain open as to the extent to which impairment rates can provide an appropriate proxy for the measurement of the 'performance' of securitisation credit assessments over time in the absence of more complete recovery rate data. Competent authorities will also consider transition matrices when this provides additional useful information.
160. In comparing default/impairment rates, competent authorities will work
with the ECAIs in question to seek to understand fully the definition of 18 Annex IX, Part 3, paragraph 7

default/impairment on the basis of which they carry out their data analysis. It will be important for competent authorities to understand the ECAI's approach to this issue. In view of the long maturity of many securitisation transactions and the fact that contractual default/impairment may be tied to this long maturity, ECAIs may use varying definitions of default/impairment as alternatives to or proxies for contractual default.
161. It is expected that most ECAIs will produce ratings performance data using
a 'cohort' approach – i.e. an approach that incorporates the effect of ratings migration in its analysis of the performance of the rating. While ECAIs may also produce data based on an 'original rating' analysis, CEBS considers that – as for other credit assessments a 'cohort' approach is likely to be the most meaningful for the purposes of mapping securitisation assessments.
162. In considering quantitative factors, competent authorities will also consider
the approach of the ECAI to aspects such as 'curing' (the subsequent repayment of missed payments) and withdrawn credit assessments, and how these affect the ECAI's ratings 'performance' studies.
163. 'Seasoning' is another factor that competent authorities will consider. In
particular, given the possible difference between the lossdistribution curve for assetbacked securities as compared with corporate and other debt, the period over which rating performance is considered – e.g. three years versus five years – may be significant.
164. Different ECAIs have different approaches to the meaning of their
securitisation credit assessments. For example, some seek to produce a rank ordering with respect to the loss that may be suffered by the tranche in question, while others base their rank ordering to a greater extent on the likelihood of the tranche suffering 'first euro' impairment. Nonetheless, there seems to be a broad consensus that the question of loss is an important factor to be taken into consideration. At this stage in the development of the market, it seems likely that the amount of loss data available will continue to grow. Competent authorities will seek to take into account the loss/recovery rate data that are available in relation to the different ECAIs ratings. It is expected that these data will improve in significance over time and that recovery rate studies will become an increasingly rich source of information. 2.Qualitative factors
165. The CRD requires competent authorities to take qualitative as well as
quantitative factors into consideration in mapping securitisation credit assessments into credit quality steps. Competent authorities believe that this is likely to be an important aspect, particularly when quantitative data are less than conclusive, as noted above in paragraph 148.
166. In assigning securitisation credit assessments, ECAIs often adopt an
'indicative' approach. That is, they indicate what is required in order for a particular tranche of a transaction to achieve a particular credit assessment level. This means that an ECAI's assignment methodology for

ABS credit assessments can provide important insights in the mapping process. It is also likely to be useful, where relevant, to consider the relationship between an ECAI's securitisation credit assessments and its other credit assessments – in particular in relation to the 'meaning' of the different assessments.
167. A notable feature in the development of the securitisation market over
recent years has been the degree to which the market has been 'ratingsdriven.' That is, the credit assessments assigned by ECAIs have played an important role in the structuring and marketing of transactions and in the provision of investor information.
168. In this context, competent authorities think that it will be highly relevant
to consider the way in which market participants view the published credit assessments of different ECAIs. Accordingly, in mapping an ECAI's securitisation credit assessments to the CRD’s credit quality steps, competent authorities will take into account market information concerning the degree to which the published credit assessments of the ECAI in question are regarded as being similar in meaning, as an indicator of creditworthiness, to those of its peers. There is some evidence to indicate that market participants regard the published securitisation credit assessments of a number of relevant ECAIs as being in many respects equivalent. It is expected that studies on market information e.g. credit spreads on securitisations rated by an ECAI as compared to its peers will also become an increasingly rich source of information. Shortterm credit assessments
169. Competent authorities propose to base the mapping of shortterm credit
assessments on the mapping of longterm credit assessments explained above, and on the internal mapping of shortterm to longterm credit assessments undertaken by the ECAI. Should any inconsistencies arise, competent authorities will seek to adjust the mapping accordingly (e.g. by adopting a conservative approach). Credit Assessments of Collective Investment Undertakings (CIUs)
170. According to Annex VI, Part 1, paragraph 72 of the CRD, exposures in the
form of CIUs for which a credit assessment by a nominated ECAI is available shall be assigned a risk weight in accordance with the assignment by the competent authorities of the credit assessments of eligible ECAIs to six steps in a credit quality assessment scale
171. For CIUs, however, ECAIs usually issue several assessments with distinct
meanings (e.g. assessments of the asset quality of a fund, of the quality of the management of the fund, or of the volatility of the fund). It has therefore been deemed necessary to define which of the assessments of a CIU should be eligible for risk weighting purposes in the context of the CRD and how they should be mapped to the individual credit quality steps. Eligible assessments

In order to be eligible for the purposes of the CRD, credit assessments for
CIUs must fulfil the following criteria:

The assessment of the credit quality of the CIU must depend primarily
on the credit quality of the underlying assets.
Where other factors have a significant influence on the assessment, supervisors shall consider the extent and nature of that influence in
determining whether the assessment remains a credit assessment for these purposes and whether any adjustment to the mapping may be required.
Only assessments for fixedincome CIUs should be eligible, since the
CRD does not allow the use of credit assessments for other asset classes (e.g. equity) within the Standardised Approach. Mapping

An assessment of the credit quality of a CIU which meets the criteria set
out above can be mapped similarly to the other fundamental credit assessments of the respective ECAI. It has therefore not been considered necessary to develop an alternative mapping approach for CIU assessments.
Part 4: Export Credit Agencies
Article 80(1) of the CRD states that credit quality may be determined by
reference to the credit assessments of ECAIs in accordance with the provisions of Article 81 to 83 or the credit assessments of Export Credit Agencies as described in Annex VI, Part 1. The CRD limits the use of ECA’s credit assessments to exposures to central governments and central banks. Therefore, institutions are allowed to use Export Credit Agency credit assessments to calculate the risk weight of their exposures to central governments and central banks, in addition to ECAIs’ credit assessments for the other types of exposures.
Paragraph 7, Part 1, Annex VI of the CRD provides that the credit
assessments of an Export Credit Agency can be used for calculating capital requirements if either of two conditions are met:
a. The credit assessment is a consensus risk score from an Export Credit Agency participating in the OECD Arrangement on Guidelines for Officially Supported Export Credits, or b. The Export Credit Agency publishes its credit assessments, the Export Credit Agency subscribes to the OECD agreed methodology, and the credit assessment is associated with one of the eight minimum export insurance premiums (MEIP) that the OECD agreed methodology establishes.
It has not been deemed necessary to set up a recognition process for
Export Credit Agencies equivalent to the one required for ECAIs. Competent authorities shall simply ask the institutions that wish to use an Export Credit Agency’s credit assessments to demonstrate that one of the above conditions is met. Thus, eligible credit assessments are either:
a. Consensus risk scores from the OECD Arrangements, or b. Any credit assessments of participants in the OECD Arrangements following the agreed methodology that are not consensus risk scores, regardless of whether the country in question has been assigned a consensus risk score.
The rules set out in Annex VI, Part 3, Paragraphs 2 and 3 of the CRD
should apply to Export Credit Agencies’ credit assessments. This means that an institution which decides to use the eligible credit assessments of an Export Credit Agency or OECD participants' consensus risk scores must use those credit assessments consistently for all exposures belonging to the exposure class. Moreover, an institution which decides to use the eligible credit assessments of an Export Credit Agency must use them in a continuous and consistent way over time.
To avoid arbitrage and ensure consistency between the two types of
external credit assessments, the rules set out in Annex VI, Part 3, Paragraphs 6 and 7 of the CRD will also apply. The CRD states that if two credit assessments are available and the two correspond to different risk weights for a rated item, the higher risk weight shall be applied. If more than two credit assessments are available for a rated item, the two assessments generating the two lowest weights shall be referred to. If the lowest risk weights are different, the higher risk weight shall be applied; if the two lowest risk weights are the same, that risk weight shall be applied.
Finally, Annex VI, Part 1, Paragraph 8 of the CRD links the eight minimum
export insurance premiums (MEIP) established under the OECD methodology and the risk weights available under the Standardised Approach.

Annex 1
Common Basis Application Pack
GENERAL INFORMATION
The type of application: to use ECAI credit assessments for riskweighting in the Standardised Approach, or for riskweighting of securitisations. The market segments for which the applicant is seeking recognition. The type of credit assessments provided: solicited or/and unsolicited, with a brief explanation of the rationale behind the policy. The competent authorities where the applicant is seeking recognition or intends to seek recognition. The countries where the applicant is active. Presentation of the ECAI An overview of the legal structure of the ECAI and the group to which it belongs: ownership, major subsidiaries, ancillary or other services provided, etc. The information on ownership should include a list of shareholders that hold more than, for example, 10 percent of the ECAI’s equity. This threshold may vary depending on the ownership structure of the ECAIs. The total number of fulltime employees. The total number and percentage of revenues from major customers and/or subscribers (e.g. customers or subscribers accounting for 5% or more of total revenues. The threshold may vary depending on the ECAIs). Financial information demonstrating the financial soundness of the ECAI: the ECAI’s financial statements from the past three years and forecasts for the next three years where applicable; alternatively, letter of support from the parent entity. Do you adhere to a code of conduct similar to market accepted standards or which is in line with internationally recognised principles? Technical criteria laid down in the CRD The applicant shall include in its application a description of the core rating process for each market segment or securitisation position and each geographical area in which it is seeking recognition. The applicant is not required to provide duplicate answers and information for this application pack, but will clearly indicate for each recognition criteria what differs from one area of recognition to another. Competent authorities are interested only in information that is relevant to the market segments and/or securitisation positions for which the application is made. METHODOLOGY

Objectivity
Question:
How do you ensure that the methodology used for assigning credit assessments is rigorous, systematic and subject to validation based on historical experience? Minimum information to be provided to the competent authorities to enable them to verify that the criterion is met:
1) A highlevel description of the credit assessment methodology and processes
and how the methodology is determined, implemented, and changed. This description shall include a description of processes in place to ensure the consistent application of the assessment methodologies across all credit assessments, in particular the role of rating committees and guidelines governing them, the extent of input from rated entities, the access to nonpublic information, etc.
2) For each of the asset groupings within which a core methodology is applied
consistently (for example, structured finance, public finance, or commercial entities, as mentioned above), a highlevel description of quantitative inputs: key variables, data sources, assumptions and quantitative techniques used, extent of input from rated entities, etc.
3) For each of the asset groupings within which a core methodology is applied
consistently (for instance structured finance, public finance, commercial entities, as mentioned above), a highlevel description of qualitative inputs in particular the scope of qualitative judgement e.g. regarding the strategy, business plans of the rated entities, etc.
4) A summary by geographical area of the major differences in the core
methodologies.
5) A description of the methodology used to verify the accuracy, consistency, and
discriminatory power of the rating systems, with details on the results and conclusions generated by such analysis.
Independence
Question:
How do you ensure that the methodology used is free from external political influences or constraints and from economic pressures that could influence the credit assessment? Minimum information to be provided to the competent authorities to enable them to verify that the criterion is met:
1) A description of the procedures aimed at ensuring fair and objective credit
assessments: mechanisms to identify, prevent, manage and eliminate actual or potential conflicts of interest.
2) A detailed description of the safeguards in place when shareholders, subsidiaries, or other entities belonging to the group are rated.

3) Demonstration and selfcertification of the existence of an internal audit
function and/or that there are means to ensure that internal procedures are implemented effectively.
4) Demonstration and selfcertification that members of the rating teams and
committees have appropriate and requisite skills – including quantitative expertise – and experience in credit assessment, and that these skills are maintained or improved over time through adequate training programmes. 5) A description of the main features of the ECAI’s internal code of conduct.
6) Demonstration and selfcertification that the remuneration policy of the staff
involved in credit assessment does not affect the production of independent and objective credit assessments: e.g. certification that analysts’ remuneration is not tied to credit assessment decisions, fees from issuers, or revenues from investors or subscribers.
7) Details of the ECAI’s fee policy.
8) Selfcertification that the staff involved in the credit assessment process are
not engaged in any business relationships with rated entities which could hinder the issuance of independent and highquality credit assessments.
3. Ongoing review
Questions:
1) Are your credit assessments subject to ongoing review which is carried out
at least annually and after all significant events?
2) To what extent are your credit assessments responsive to changes in the
financial conditions?
3) Do you have procedures in place that ensure that competent authorities are
promptly informed of material changes, and if so, what are they? Minimum information to be provided to the competent authorities to enable them to verify that the criterion is met:
1) General information on rating reviews: e.g. the process in place, main
characteristics, scope, frequency, people/teams involved, means used, treatment, main phases of the monitoring process, data updates, information from rated entities taken into account, automatic warning systems, mechanisms that allow systematic errors in credit assessments to feedback into potential changes in ratings method, etc.
2) A summary of the outcome of the reviews carried out
3) Demonstration that a backtesting system is in place and has been up and
running for at least one year. 4) The extent of contacts with the senior management of the rated entities (this information is to be provided upon request of the competent authority).
4. Transparency and disclosure
Question:

How (by what means and in what language) and to whom do you disclose the principles of the methodology you use? Minimum information to be provided to the competent authorities to enable them to verify that the criterion is met:
1) A demonstration that the principles of the methodology employed by the ECAI
for the formulation of its credit assessments are disclosed. 2) Descriptions of the ways used to make methodologies publicly available, and of the terms of access to the credit assessments by all potential users
3) A description of transparency policy with regard to the types of credit
assessment: solicited or unsolicited.
INDIVIDUAL CREDIT ASSESSMENTS
5. Credibility and market acceptance
Question:
How could you prove your credibility and market acceptance?
Minimum information to be provided to the competent authorities to enable them to verify that the criterion is met:
1) Any evidence demonstrating market reliance on the credit assessments, such
as market share, number of issuers, how long the ECAI has been active in the market, the revenues generated by the rating activities, or any other proof.
6. Transparency and disclosure
Questions:
1) How do you ensure that credit assessments are accessible at equivalent terms
at least to all institutions having a legitimate interest in them?
2) In particular, how do you ensure that credit assessments are accessible at
equivalent terms to both domestic and nondomestic parties having a legitimate interest? Minimum information to be provided to the competent authorities to enable them to verify that the criterion is met:
1) A highlevel description of the disclosure procedures in place
MAPPING
Question: None
Minimum information to be provided to the competent authorities to enable them to perform the mapping the credit assessments of exposures other than securitisation positions:

1) The definition of default
2) The CDR over a threeyear period for each credit assessment category (to be
provided annually if the ECAI is recognised as eligible), at least the two most recent CDRs, if available
3) The tenyear average of the threeyear CDR. If not available, an indication of
the ECAI's expectation concerning the long term default rate. 4) If a target probability of default is used, the target probability of default for each credit assessment category,
5) Description of the methodology to calculate the CDRs: selection of pool (static
versus dynamic/adjusted), definition of default, aggregation of defaults (weighting mechanism), 6) The statistical significance of the default rates,
7) Dynamic characteristics of the rating methodology (pointintime or through
the cycle),
8) The meaning of the credit assessment categories,
9) The range of credit assessments that the ECAI assigns,
10) The time horizon of the credit assessment,
11) Transition matrices,
12) Geographic coverage. Minimum information to be provided to the competent authorities to
enable them to perform the mapping of securitisation positions
1) the definition of default/impairment on the basis of which the
default/impairment rates are computed
2) Ratings’ performance data, accompanied by an explanation of its main
features (e.g. the reasons underlying the determination of the time horizon over which the study has been carried out and how curing and withdrawn credit assessments impact the rating performance studies; how seasoning is taken into account)
3) loss/recovery data
4) Information referred to point 8 to 12 above
Additional information for CIUs
1) Presentation of the CIU ratings considered as assessing primarily the credit
quality of the underlying assets
2) Description of the factors and the extent to which they have been taken into
account
3) Information referred to point 8 to 12 above

Lineage: No known changes

Similar documents from other regulators