Don’t Let “Averages” Mess Up Your IFRS 9 Impact Forecasts
Blog: Enterprise Decision Management Blog
The clock is ticking on IFRS 9 compliance. Preparing for Day 1 IFRS 9 impacts before models are fully built, tested and validated is a bit like preparing for a scheduled car crash: You know the event is going to happen and you know it will be expensive and painful, but you don’t yet know the exact amount of damage.
It is tempting to seek comfort in averages from industry benchmarks or surveys. But just as you would never apply the brakes of your car during heavy rain based on the average stopping distance of an average car in average weather conditions, neither should you plan for your business’s IFRS 9 impacts based on average industry impacts alone.
Don’t Go by Published Numbers
A few widely published, well-intended surveys have attempted to prepare affected organisations for this event by providing estimated ranges and averages in respect of key impact measures. For example, the EBA’s initial IFRS 9 impact assessment report from November 2016 revealed that the “estimated increase of provisions compared to the current levels of provisions under IAS 39 is 18% on average and up to 30% for 86% (75th percentile) of respondents.” One in seven Deloitte survey respondents were preparing for provision increases of 50% or more.
Once shared widely, such numbers set an expectation that may leave certain entities under-prepared. To better plan for the challenges ahead, spend some time thinking about factors that may specifically affect your portfolio.
Key drivers of impact include:
- Current approach to IAS 39. The interpretation of IAS 39 may be the single largest determinant of relative provision impacts on IFRS 9 Day 1. Interpretations of IAS 39 from individual organisations, auditors and regulators vary widely. Most large UK banks begin impairment at the earliest stages of delinquency and also maintain a general provision to cover Incurred but Not Reported (IBNR) losses for accounts without individually identifiable signs of impairment. In contrast, the regulator in the Philippines (the BSP) requires zero impairment for accounts less than 90 days past due, 25% balance coverage at 90-119 days past due, 50% at 120-179 days past due and 100% coverage at 180 days past due.
- Portfolio composition and risk profile. Certain products are inherently more affected by the change to IFRS 9. The requirement to hold provisions against not just drawn balances but also undrawn commitments significantly impacts credit card portfolios due to their generally higher risk profile. Riskier asset classes with more sensitive stage allocation criteria will also be more heavily impacted.
- Write-off and debt sale policies. In general, provision levels for defaulted (Stage 3) accounts are the least impacted by IFRS 9. To the extent that banks maintain a “recovery book” of charged-off accounts, or otherwise prefer to manage rather than write-off or sell highly impaired assets, their provisions will tend to increase by a lower percentage overall.
- Expected macro¬-economic environment on Day 1. As the January 2018 deadline applicable to most organisations approaches, the uncertainty of Day 1 macroeconomic impacts may diminish. Yet recent world events have shown that the unexpected can and will happen, introducing uncertainty into both current and expected portfolio performance. This uncertainty will influence forward-looking projections even if their immediate effects are not yet widely observed in current portfolio performance.
- Product structure. Secured products are less impacted than unsecured (LGD), short-term loans are impacted less than long-term loans (lifetime), and term loans are impacted less than revolving loans (EAD).
- Interpretation of IFRS 9. As IFRS 9 is a principles-based standard, it is open to interpretation. The strictness of key decision points such as the definition of “significant increase in credit risk since initial recognition” and the choice of methodology to estimate lifetime losses will materially influence post-compliance impairment levels.
All of these considerations emphasise the need for swift, accurate model development, testing and validation, as well as ongoing stakeholder management. High-level impact assessments can be created relatively early in the process based on readily available data and management reporting, combined with high-level assumptions about model methodology.
Any impacts should be presented as a range and be heavily caveated and clearly explained. As key components become clearer, the range can be narrowed and the caveats removed one by one. For some portfolios, IFRS 9 provisions will be more appropriately expressed as a multiple of IAS 39 provisions rather than as a percentage increase.
Setting reasonable and specific expectations of IFRS 9 impairment impacts by considering all of the points above will help you prepare your organisation’s key decision makers for the impacts that lie ahead. FICO has a structured approach for helping you work through the IFRS planning and compliance stages, including post-compliance. Instead of driving blind, you’ll have a clear view of the road ahead — however bumpy it may be.
The post Don’t Let “Averages” Mess Up Your IFRS 9 Impact Forecasts appeared first on FICO.