Response to discussion on the simplification and assessment of the credit risk framework

Go back

Q4. Which other clarifications do you consider necessary to apply the new ECAI framework?

Credit Rating Agencies (“CRAs”) general approach to bank credit ratings

From a general perspective, CRAs use established and transparent rating methodologies, developed and applied independently by each agency, to derive bank ratings. These methodologies usually consider a wide range of quantitative and qualitative factors relevant to a bank's creditworthiness, typically including an assessment of its stand-alone credit profile. In the banking sector, the legal, regulatory and institutional environment in which banks operate is often credit‑relevant and reflected in the analytical assessment itself, rather than treated as a separate or discretionary support assumption. Depending on the circumstances, CRAs may also consider external support factors, such as the potential for extraordinary support (e.g. provided by shareholders or governments) to derive the final credit rating of a bank. 

Historic review

The Global Financial Crisis of 2008 and 2009 led to numerous bank failures in the European Union and several EU Member States had to bail-out ailing banks. Against this background, methodologies used by most CRAs that incorporated assumptions about government intervention in extreme stress scenarios were broadly consistent with observed outcomes.

In order to address the bank-sovereign link, the European Union introduced a wide range of regulatory reforms, notably the Bank Recovery and Resolution Directive, the Single Resolution Mechanism, the Single Supervisory Mechanism, as well as the Single Resolution Fund. These reforms were designed to reduce the likelihood that EU Member States would need to bail-out failing banks using taxpayers' money. To our knowledge, these instruments have not yet been fully tested in a severe systemic crisis. Nevertheless, in view of the concentration in the EU banking market, it cannot be excluded that some form of government intervention could be required in the event of the failure of a systemically important EU-wide bank, in order to avoid systemic stress to financial stability.

Furthermore, from a global perspective, it is unclear whether all jurisdictions have equally implemented such stringent reforms aimed at avoiding future government bail-outs of banks. It is therefore possible that assumptions regarding government intervention continue to play an important role in assessing the overall creditworthiness of a bank in certain non-EU jurisdictions. 

The Basel III reform includes in item 20.18 the requirement that credit ratings on banks “must not incorporate assumptions of implicit government support, unless the ratings refer to a public bank owned by its government”. This requirement was transposed into EU law in Article 138(g) of the Capital Requirements Regulation. 

Response to Q4

In order to facilitate the consistent and proportionate application of the new ECAI framework, EACRA considers that clarification would be helpful in particular on the following points:

  • That credit assessments excluding implicit government support are intended as prudential tools for the purposes of CRR Article 138(g) and do not redefine the nature or role of credit ratings used more broadly by market participants;
  • The intended boundary between assumptions of extraordinary government support and structural legal or regulatory features that may be credit‑relevant;
  • Labelling and disclosure expectations for such credit assessments, in order to mitigate the risk of market confusion or unintended use outside the prudential framework; and
  • That the proposed shortcut mapping approach is limited in scope and transitional and does not set a precedent for prescribing rating methodologies in other regulatory contexts.

Non-interference in credit ratings and rating methodologies

The CRA Regulation (as amended) includes in its Article 23 the core principle stating: “In carrying out their duties under this Regulation, ESMA, the Commission or any public authorities of a Member State shall not interfere with the content of credit ratings or methodologies.“ This principle is meant to ensure the independence of CRAs in assigning their credit ratings. 

It is therefore somewhat surprising that a change in EU legislation give rise to expectations of methodological changes by CRAs, notably expecting the exclusion of implicit government support from bank ratings. We therefore strongly hope that this change will not be used as a benchmark in other areas that would seek to prescribe how CRAs assess the creditworthiness of specific entities or to determine which credit-relevant factors may be included or excluded.

EBA’s pragmatic approach

In the Discussion Paper, EBA puts forward a pragmatic approach to the requirements embedded in the Level 1 legislation:

  • CRAs are requested to provide credit ratings that exclude government support on a stand-alone basis, but EBA does not prohibit credit ratings that include government support,
  • EBA requires that such credit ratings excluding government support should be issued based on existing rating scales (for example, by adding a suffix to highlight the difference from “full credit ratings”),
  • EBA proposes to use a pragmatic “shortcut approach” to map these credit ratings excluding government support,
  • EBA may review this pragmatic approach once sufficient data become available to carry out a full quantitative mapping process. 

We are supportive of this high degree of pragmatism by EBA. We would welcome a similar degree of pragmatism being applied in other areas of the mapping process of CRA rating scales to the regulatory credit quality steps.

Upload files

Name of the organization

European Association of Credit Rating Agencies ("EACRA")