Response to addendum to Joint Consultation on draft Implementing Technical Standards on the mapping of ECAIs’ credit assessments
Go back
The definition of “sufficient” should also be adapted to the number of rated items provided by smaller ECAIs. The European Union wants to increase competition and therefore promotes the use of smaller ECAIs/CRAs, as set out (for example) in Article 8c(1) of the CRA Regulation (EU) No 462/2013.
The long-run and short-run default rate benchmarks applied in the mapping process correspond to those provided by S&P’s and Moody’s, as defined in Annex 2 (Standardised Approach - Implementing the Mapping Process) of Basel II (International Convergence of Capital Measurement and Capital Standards; Basel Committee on Banking Supervision, June 2006). It is questionable whether these benchmarks are still valid (for example) due to changes during and after the global financial crisis, and whether benchmarks based on default rates of multinational companies fully reflect the risk associated with (for example) Western European corporates or mid-cap markets.
However, the application of a “different measure of creditworthiness” may be less reliable. First, we partially agree with the application of scoring models to the mapping process. We would like to point out, though, that scoring models should be directly comparable to the genuine credit rating assigned by an ECAI in order to fully reflect the risk relationship between genuine ratings and scoring models.
We do not agree with the application of CQS and default rates of benchmark ECAIs to the mapping process. This approach presumably implies that rating methodologies, rating analysts and the rating universe of both, already rated items/asset classes and future rated items/asset classes are fully comparable and almost identical. We assume that this will not be the case. Any ECAI can purchase data on corporates and default information from data providers and is therefore able to develop a scoring approach adjusted with qualitative market data as an initial measure of credit risk. Applying such scoring models adjusted by qualitative information on industry trends would therefore better reflect the ECAI’s own rating performance.
Also, an uplift by 100% applied to a default rate of (for example) 1.5% in ECAI’s rating category BBB (equaling CQS 3 with an average default rate of 1.0%) would result in an adjusted default rate of 3%, which equals CQS 4 with an average default rate of 7.5%. Thus, the risk associated with the ECAI’s BBB rating category would be 400% more (5 times higher) than initially calculated instead of only 100% more (2 times higher). This could be mitigated if a more refined CQS scale is applied taking into account all rating and notch categories generally used by ECAIs (see also Q1).
In addition, it is crucial to consider “any other relevant information that can describe the degree of risk expressed by a rating category”, as set out in Article 8(g) of the draft ITS when assigning an uplift to the initial default rate; for example: ECAIs with broad information available are able to monitor any default events related to the rated entities, even if a rating is withdrawn or not monitored through information provided by the rated entity. It is also necessary to apply a judgement-driven approach to the assessment of the definition of default and the observed defaults, meaning that all qualitative information relevant to assess the risk associated with a specific rating category should be incorporated.
We, hence, suggest applying a more sophisticated benchmark taking into account an ECAI’s rating universe, incorporating any additional information on defaulted entities available to an ECAI and/or applying an uplift to an extent, which is based on observed market data and standards provided by industry experts.
It should be made more transparent what could be considered as “material” and in which cases a reassessment of an ECAI’s mapping will be proposed. In addition, a reassessment should not only be proposed based on quantitative information, especially when changes in the market environment, exogenous shocks or similar developments might affect default rates. This requires a more sophisticated judgement-based evaluation.
In addition, we are concerned that some issues related to the mapping approach might not fully reflect the risk associated with specific rating categories, particularly with respect to smaller ECAIs. We suggest applying a more judgement-based mapping approach, which is able to incorporate additional qualitative information. This requires a deep insight into the nature of different rating businesses and rating processes across all ECAIs.
Finally, we would highly recommend to take into account the amendments proposed by the European Association of Credit Rating Agencies.
Q2. Do you agree with the proposed definition of sufficient number of credit ratings and rest of requirements imposed to the calculation of the short run default rate when a sufficient number of credit ratings is available?
The definition of “sufficient” with respect to the number of ratings should be made more transparent. In addition, the mapping process should always include “necessary expert judgement”. However, there should be a more transparent description to what extent this expert judgement is applied to the definition of “sufficient” and to what extent the decision whether the number of ratings is considered sufficient or not is based on quantitative or judgement-driven decisions.The definition of “sufficient” should also be adapted to the number of rated items provided by smaller ECAIs. The European Union wants to increase competition and therefore promotes the use of smaller ECAIs/CRAs, as set out (for example) in Article 8c(1) of the CRA Regulation (EU) No 462/2013.
Q3. Do you agree with the proposed requirements imposed to the calculation of the long run default rate when a sufficient number of credit ratings is available?
In general, ECAIs apply a through-the-cycle rating approach. We believe that incorporating default rates, which only refer to a recessionary period might be too conservative.The long-run and short-run default rate benchmarks applied in the mapping process correspond to those provided by S&P’s and Moody’s, as defined in Annex 2 (Standardised Approach - Implementing the Mapping Process) of Basel II (International Convergence of Capital Measurement and Capital Standards; Basel Committee on Banking Supervision, June 2006). It is questionable whether these benchmarks are still valid (for example) due to changes during and after the global financial crisis, and whether benchmarks based on default rates of multinational companies fully reflect the risk associated with (for example) Western European corporates or mid-cap markets.
Q4. Do you agree with the proposed options to calculate the quantitative factors when a sufficient number of credit ratings is not available?
Basically, we agree with the “minimum required number of rated items”, as set out in table 1 in the Addendum to Section 4 of JC/CP/2014/01 entitled ‘Draft Implementing TS on the mapping if ECAIs’ credit assessments under Articles 136(1) and 136(3) of Regulation (EU) No 575/2013 (Capital Requirements Regulation-CRR)’ in cases where neither a sufficient number of credit ratings nor a sufficient number of items assigned a different measure of creditworthiness is available. We would welcome a more transparent description of the derivation of the required numbers of rated items for each rating category.However, the application of a “different measure of creditworthiness” may be less reliable. First, we partially agree with the application of scoring models to the mapping process. We would like to point out, though, that scoring models should be directly comparable to the genuine credit rating assigned by an ECAI in order to fully reflect the risk relationship between genuine ratings and scoring models.
We do not agree with the application of CQS and default rates of benchmark ECAIs to the mapping process. This approach presumably implies that rating methodologies, rating analysts and the rating universe of both, already rated items/asset classes and future rated items/asset classes are fully comparable and almost identical. We assume that this will not be the case. Any ECAI can purchase data on corporates and default information from data providers and is therefore able to develop a scoring approach adjusted with qualitative market data as an initial measure of credit risk. Applying such scoring models adjusted by qualitative information on industry trends would therefore better reflect the ECAI’s own rating performance.
Q5. Do you agree with the proposed use of the default definition use by the ECAI as a relevant factor for the mapping? Do you agree with the proposed assessment of the comparability of the default definition of an ECAI? If not, what alternatives would you propose? In particular, what are your views on the introduction of a specific benchmark for ECAIs relying on the regulatory definition of default and/or for ECAIs providing ratings for companies that do not rely on market financing? Could it be a robust alternative to the current benchmark? Do you think that the adjustment factor depends on certain characteristics of the rated firms such as size and credit quality and if so, how can this be reflected?
We basically agree that the definition of default used by ECAIs has a significant impact on the outcome of the analysis. Nevertheless, we assume that an uplift by up to 100% of the initial default rates is substantially too high, particularly in cases where only bankruptcy-related events are considered in the definition of default. In general, for most corporates, particularly SMEs, any missed or delayed (re-)payment is closely followed by bankruptcy. In addition, it is questionable whether a default with respect to missed or delayed payments or distressed exchanges on bank loans for which public information is not available may be consistently observable for publicly-listed companies. Large corporates may indeed try to make an agreement on bank loans as a first step avoiding any non-payment to bondholders. Therefore, we expect that on average, the same market mechanisms in terms of publicly available information on non-payments and distressed exchanges could be observed for both, SMEs and large corporates. Thus, any uplift should be avoided or made to a very low and judgment-based extent.Also, an uplift by 100% applied to a default rate of (for example) 1.5% in ECAI’s rating category BBB (equaling CQS 3 with an average default rate of 1.0%) would result in an adjusted default rate of 3%, which equals CQS 4 with an average default rate of 7.5%. Thus, the risk associated with the ECAI’s BBB rating category would be 400% more (5 times higher) than initially calculated instead of only 100% more (2 times higher). This could be mitigated if a more refined CQS scale is applied taking into account all rating and notch categories generally used by ECAIs (see also Q1).
In addition, it is crucial to consider “any other relevant information that can describe the degree of risk expressed by a rating category”, as set out in Article 8(g) of the draft ITS when assigning an uplift to the initial default rate; for example: ECAIs with broad information available are able to monitor any default events related to the rated entities, even if a rating is withdrawn or not monitored through information provided by the rated entity. It is also necessary to apply a judgement-driven approach to the assessment of the definition of default and the observed defaults, meaning that all qualitative information relevant to assess the risk associated with a specific rating category should be incorporated.
We, hence, suggest applying a more sophisticated benchmark taking into account an ECAI’s rating universe, incorporating any additional information on defaulted entities available to an ECAI and/or applying an uplift to an extent, which is based on observed market data and standards provided by industry experts.
Q6. Do you agree with the proposed use of the time horizon of the rating category as a relevant factor for the mapping? Do you agree with the proposed use of transition probabilities to identify the expected level of risk during the 3-year time horizon?
As both measures are common for the evaluation of different levels of risk, we agree with the proposed use of the time horizon and transition probabilities. In order to assess the different risk levels with respect to the through-the-cycle approach applied by most ECAIs, we also suggest taking into account a 5-year horizon.Q7. Do you agree with the proposed use of the range and meaning of rating categories as a relevant factor for the mapping? Do you agree with the proposed restriction of this factor to adjacent rating categories?
Basically, we agree with the proposed approach. However, we are concerned that the range of CQS and the range of rating categories assigned by most CRAs do not fully match in terms of the number of rating categories (see also Q1). Therefore, the mapping process should include some information associated with specific rating (notch) categories. This is evidenced by typically disproportionately increasing empirical default rates across (for example) BBB+, BBB and BBB- notch categories.Q8. Do you agree with the proposed use of the risk profile of a rating category as a relevant factor for the mapping?
In our opinion, the mapping approach is designed to enable an unbiased and consistent mapping of the rating categories of different ECAIs. Therefore, the risk associated with a rating category should not depend on the risk profile of the rated entities when no default data is available. ECAIs, which predominantly provide ratings in mid-cap markets are expected not to assign the highest rating categories to SMEs but are still able to assign these highest rating categories to (for example) very large and highly diversified corporates or covered bonds.Q9. Do you agree with the proposed use of the estimate provided by the ECAI of the long run default rate associated with all items assigned the same rating category as a relevant factor for the mapping? Do you agree with the proposed role played by this factor depending on the availability of default data for the rating category?
We generally agree with the application of estimates provided by ECAIs, given the shortcomings pointed out under Q4. However, a “different measure of creditworthiness” should not be applied to the calculation of the long-run default rate, as set out in Article 7 of the draft ITS because different measures will not fully reflect any analyst-driven rating approach but bias the risk associated with a rating category.Q10. Do you agree with the proposed use of the internal mapping of a rating category established by the ECAI?
We believe that this issue needs further discussion with respect to differences between rating scales applied by ECAIs.Q11. Do you agree with the proposed specification of the long run and short run benchmarks? In particular, do you think that different benchmarks should be introduced to fully take into account the specific situation of ECAIs relying on the regulatory definition of default and/or of ECAIs providing ratings for companies that do not rely on market financing, including in particular SMEs? Do you agree with the proposed mechanism to identify a weakening of assessment standards?
As set out in Q3, the long-run and short-run default rate benchmarks applied in the mapping process correspond to those provided by S&P’s and Moody’s, as defined in Annex 2 of Basel II. It is questionable whether these benchmarks are still valid (for example) due to changes during and after the global financial crisis, and whether benchmarks based on default rates of multinational companies fully reflect the risk associated with (for example) Western European corporates or mid-cap markets.It should be made more transparent what could be considered as “material” and in which cases a reassessment of an ECAI’s mapping will be proposed. In addition, a reassessment should not only be proposed based on quantitative information, especially when changes in the market environment, exogenous shocks or similar developments might affect default rates. This requires a more sophisticated judgement-based evaluation.
Q12. Do you agree with our analysis of the impact of the proposals in this CP? If not, can you provide any evidence or data that would explain why you disagree or which might further inform our analysis of the likely impacts of the proposals?
In our opinion, the mapping approach should be made very transparent and an identical mapping process should be applied consistently across all ECAIs. This is crucial as investors, banks and other third parties should be able to have a deep insight into possible differences between ECAIs and a full understanding of the methodology applied to the mapping process.In addition, we are concerned that some issues related to the mapping approach might not fully reflect the risk associated with specific rating categories, particularly with respect to smaller ECAIs. We suggest applying a more judgement-based mapping approach, which is able to incorporate additional qualitative information. This requires a deep insight into the nature of different rating businesses and rating processes across all ECAIs.
Finally, we would highly recommend to take into account the amendments proposed by the European Association of Credit Rating Agencies.