In the summer, we looked at IFRS 9 from the point of view of the investors. Here, we bring you the insightful and valuable findings shared by our local regulator CSSF in their IFRS 9 follow-up study covering 126 banks. In this blog we focus on their findings related to the qualitative aspects of IFRS 9. If you are interested in the quantitative impact, read more here.
Calculating ECL (Expected Credit Losses)
The CSSF found that most banks covered in general all required points regarding the methodology used to calculate ECL, but one or more items (e.g. the calibration of input parameters, the discounting factor or the scenarios) have been disregarded by some banks. It identified that 23% of banks had inaccuracies in their ECL calculation including minor rounding errors to issues around the formula itself (no discounting or omission of interest accruals, for instance).
The regulator identified that forward-looking/macroeconomic information, which should be incorporated during the assessment of significant increase in credit risk (SICR) and/or ECL calculation, had been missed by over a quarter of banks.
The challenge here is that the information needs to be relevant for the particular financial instruments (or group of financial instruments), and it is not easy to determine relevant factors to be considered in all cases. What's more, the information should be reasonable and supportable, and sometimes it is judgmental to decide what information satisfies the mentioned criteria.
Imagine a bank wants to factor in an uncertain future event (e.g. a no-deal Brexit scenario). It would need to assess the impact on the credit losses associated with the event. On the other hand, a bank shouldn't skip forward-looking/macroeconomic information just because the modeling of the impact is difficult. Another important finding is that some banks missed important items including selection and correlation of relevant factors as well as the risk-weights attributed to scenarios.
SICR & transfer between stages
The CSSF found that a large majority of banks appropriately defined the SICR and the criteria for transferring between different stages. All reasonable and supportable information should be considered during the assessment of SICR. This matter has not been properly addressed by some banks as they were not detailed enough in their procedures. For example, a bank may not have considered qualitative factors or over-relied on the 30 days past due backstop, which is not an absolute indicator that lifetime expected credit losses should be recognized. It is, however, assumed to be the latest point to account for them.
Definition of default
The CSSF found the definition of default determined by the banks was adequate and they used the same definition for IFRS 9 and prudential reporting.
Classification of financial assets
The CSSF identified Business Model Assessment as an area for improvement. Why? 17% of banks omitted some important aspects during the analysis of relevant factors including sales level, remuneration targets and risk management. Some also failed to analyze solely payments of principal and interest (SPPI) requirements properly (either comprehensively or on a risk-based sampling basis) to get to the right classification of the financial assets.
Governance and modeling
Application of IFRS 9 is not a pure accounting exercise; strong governance and cooperation between finance and risk teams are crucial. It also means appropriate policies and procedures, which reflect segregation of duties (SoD) principle. The CSSF found that banks set up governance frameworks and internal procedures which have not been materialized yet beyond draft versions or group level. Take the test!
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.