Yes, the scope of application of the guidelines is considered clear and appropriate.
Yes, assuming that IFRS 9 application date in the EU will be the same.
The consultation paper establishes that credit entities should ensure that the application of these Guidelines considers their “nature, scope and complexity of their activities and, more generally, all other relevant facts and circumstances of the credit institution and the group (if any) to which it belongs”.
We understand that this principle is to be applied individually to each entity within a consolidated group. Each of these entities may operate in different geographies, with different products and have a different mix of portfolios than other subsidiaries of the same financial group.
Taking other approaches to proportionality, especially those based on the size of the consolidated group, could lead to an unlevel playing field and an unfair competition in the local market in which each subsidiary operates.
This interpretation is consistent with other paragraphs of this guidelines, which already contemplate the existence of differences among the legal entities that compose a consolidated group, even at portfolio level within those units. For instance, regarding the implementation of Significant Increase in Credit Risk assessment, paragraph 98 of these guidelines states that “The need for consistency should not be interpreted as a requirement that the practice be identical across a group”. Supporting this point, EBA’s Guidelines also mention that “within a consistent framework there may be differences across jurisdictions and products”. Hence, it seems reasonable that the application of proportionality principle takes into account the diversity of portfolios and credit entities, even when those entities operate under the same consolidated group.
To avoid misunderstandings, we would appreciate an explicit clarification of the interpretation of the application of the proportionality principle in the final Guidelines and its application at the segment level.
Furthermore, the proportionality principle should be applied consistently across the document, for instance:
(p.129) “…credit institutions which are both smaller and less complex may reasonably rely more on the use of practical expedients…”
We would find more appropriate to refer to the complexity of the portfolio, and not the credit institution.
Accompanying documents D.2) Option 2.3 should be discouraged, as being a systemically important institution is measured at a consolidated level, with the cons explained before.
EBA’s Guidelines completed the principle of materiality as defined in the final version of BCBS guidelines by adding this reference in paragraph 18: “under the applicable accounting framework” which we support.
We are however of the view that the resulting EBA Guidelines’ wording would not be in alignment with accounting framework as IAS 1 states that: “Material Omissions or misstatements of items are material if they could, individually or collectively, influence the economic decisions that users make on the basis of the financial statements”. In this sense, IAS 1 definition neither restricts the use of methodology to estimate materiality nor states which classes of exposures should be considered as material. IAS 1 definition seems more reasonable as the concept of materiality in the implementation of an expected credit losses model should not be necessarily linked with those large and highly collateralized exposures.
This would be consistent with other previous documents. For instance, EBA’s “Guidelines on Materiality, proprietary and confidentiality and on disclosure frequency under Articles 432(1), 432 (2) and 433 of Regulation 575/2013” define materiality as information whose “omission or misstatement could change or influence the assessment or decision of a user relying on that information for the purpose of making economic decisions”.
For all the reasons stated above, we recommend that the materiality principle is reworded in the EBA’s guidelines in order to properly address the definition provided in the accounting framework. The reference to the impact on the P&L as well as the examples of large portfolios of highly collateralized lending exposures like real estate mortgages should be removed.
Consideration of large portfolios per se as “material” in our view is in conflict with the application of the materiality principle stipulated in the IFRS.
In addition, we would also like to mention that there should be an alignment between the proportionality and materiality principles with the intention that both principles enable a proper implementation of these Guidelines. This means that materiality concept should also be applied according to the “nature, scope and complexity of their activities” of every credit institution and portfolios.
• Given that the EBA Guidelines will set out the regulatory expectations in the EU, we would suggest to omit the references to the Basel Guidance.
• We have noticed a difference in wording between BCBS guidance and the EBA consultation paper that makes us believe of a restrictive use of the “more-than-30-days-past-due” rebuttable presumption. To ensure a level playing field, we believe the paragraph 135 of the EBA Guidance should be aligned with the wording of paragraph A 52 of the Basel Guidance.
• Concerning paragraph 19, an additional text has been added to the EBA Guidance, “for example, systemic delay in the recognition of credit losses”. IFRS 9 permits a variety of approaches and it is unlikely that there is a common, generally accepted view on what is “ideal”. In situation where no better information is available, credit institution will not be in the position to know whether recognition of losses is being delayed, neither whether it is done systematically.
• Paragraph 23 requires “robust justification where, in exceptional circumstances, information relevant to the assessment of credit risk is not reasonable and supportable”. While the Basel Committee also “expects” banks to provide a clearly documented, robust justification, the EBA requirement for documentation together with the very broad scope of possible relevant information could be read as requiring justification for not including every possible future scenario, which is not tenable and not in keeping with IFRS 9 which does not require all possible circumstances to be considered.
• Similarly, it is unclear how and to what extent banks can demonstrate that additional cost /effort to consider all reasonable and supportable information would not contribute to a high-quality implementation of IFRS 9 as required in paragraph 70. This creates a high potential for time-consuming discussions with reviewers as well as findings from regulatory reviews.
• Paragraph 59 of the Basel Guidance says “and not result in materially different allowance measurements” while the EBA Guidance says “and not result in materially different allowance amounts”. We suggest reverting back to the Basel Guidance text.
• The EBA wording of paragraph 109 could be read as requiring both ratio and absolute change to always be taken into account. An obligation to consider an absolute width when determining significant increases in credit risk is not in line with IFRS 9 that defines significant increase in credit risk as a relative change. Consequently, in considering the regulatory guidelines banks would not determine significant increases in compliance with IFRS 9. The text should be reworded for the significance to be defined appropriately.
• The requirements under “Principle 2 – sound ECL Methodologies” contain several requirements, which can be read as checklists rather than parts of a principle-based guidance. We would recommend to retain the wording of the Basel Guidelines in paragraphs 33, 36, 37 of the Consultation Paper where “in particular” should be replaced with “generally” and “should” be replaced with “could” as in the text of the Basel Guidance.
• Paragraph 41 requires that the probation period as defined in Commission Implementing Regulation (EU) 2015/227 shall be considered for impairment purposes as well while IFRS 9 requires symmetry in the credit risk assessment (increase/decrease). Regulation 2015/227 is only meant for reporting (to authorities) purposes and not for accounting. The use of probation periods in accounting will inevitably result in biased accounting conflicting with the neutrality principle. It should be still subject to the analysis of the credit institution what probation period is appropriate to conclude that the credit risk on the exposure has decreased instead of using standard probation periods that are not commensurate to the risks.
• The detailed requirements of paragraph 38 are akin to a mandatory checklist despite the stated EBA intention to avoid any “checklist approach”. The additional sensitivity analysis as required by 38 a) would not add any substantial benefit that would outweight the cost of its application.
• We are also concerned with the regulatory expectation to demonstrate risk-adjusted pricing for all loans and throughout all business segments. It should not be used as a factor for determining significant increases in credit risk and consequently as an indicator for the assessment of the quality of the implementation of the ECL model (paragraph 107).
Furthermore paragraph 107 has deviated from the Basel Guidance by deleting of “while it is netither possible nor desirable for universally applicable criteria to be developed” and together with the wording “credit institutions should” could be read as mandatory checklist which we understood is not the intention of the EBA.