The Data Analytics Blog

Our news and views relating to Data Analytics, Big Data, Machine Learning, and the world of Credit.

All Posts

6 Key Considerations When Managing Your IFRS 9 Forward-Looking Overrides

As part of the group that was the second company worldwide to become IFRS9 compliant, IFRS9 has been at the forefront of what we do.  We have assisted nearly 20 companies on their IFRS9 journey over the last two years.  This blog forms part of a more extensive series on IFRS9. In this blog, we explore the administering of management overrides.

Part of IFRS9 is creating data-driven models to most accurately predict credit losses.  On top of the technical models are the compulsory forward-looking economic models (empirical or expert based) and the optional management overrides.

IFRS

We are frequently asked about management adjustments, how many are acceptable, how do you manage them and when are they justified?  The technical models are derived on historical data and for various reasons, and because we need to consider forward-looking predictions, the future may not be exactly like the past.  Hence there is a requirement for occasional expert adjustments with management foresight.

IFRS 2

For an analytical purist, the movement from a scientific “deductive approach” utilising data to make your conclusion to a “motivated reasoning” approach where “convenient” data can be cherry-picked to draw a pre-determined conclusion ought to send shivers down a quant’s spine. Nevertheless, it may be necessary due to a variety of reasons below.  If one is going to apply expert adjustments, then there are some important principles to consider.

Examples of data that can be amended:

  1. Expected credit sales (for transactional accounts)
  2. Valuations (for assets)
  3. Macro-economic or regulatory events (consider both forecasted events and events that have already happened where the effects were not observed in the modelling data)
  4. Environmental factors (for example for agricultural loans)
  5. Industry related trends (downward/upward forecasts)
  6. Operational or technology changes (e.g. a change of collection system may have temporary adverse impacts and long-term favourable effects)
  7. Metrics for new sub-products/segments not catered for in the models
  8. Credit risk management policy or strategy changes that could change the risk mix or performance within a risk class
  9. Changes in accounting policies such as write-off policy
  10. Changes in debt-sale appetite and the related debt sale prices

Key considerations:

  1. Substantial change means strong motivation: Remember the adage that an IFRS 3“extraordinary adjustment requires an extraordinary motivation”.  Adjustments should generally be slight and conservative; if not there needs to be a strong motivation why. While historical data might not be able to showcase the event for which an override is being defined, data analytics should still be underpinning the expected size and timing of expected impacts.
  2. Stronger models mean fewer changes. If the original models were limited from a IFRS 4data-perspective or non-representative of the expected future trends, then the motivation is stronger, but this should be illustrated in the write-up. For Black-Swan (i.e. unprecedented) events, special adjustments may be necessary. Often a model focuses on macro-economic events, and as such, it may leave any expected strategic events unresolved.   Lagged impacts of an economic or strategic event could be modelled through early performance indicators.  These indicators predict typical model under or over-predictions.
  3. Expert touch. The individuals making the motivation for adjustment(s) should be IFRS 5those with good authority on the matters. The individual should be seen as acting objectivelyand not in favour of achieving a predefined goal/objective.
  4. Team involvement. IFRS 6The adjustments should be agreed by the larger team responsible for provisions within a formal governance committee with recorded conclusions.
  5. Show your evidence. The adjustments should be supported by suitable and additional analytics (if possible). If external IFRS 7factors are cited, then the use of external data is strongly advised.
  6. Objective approach. Under IAS39 conservatism was the preferred method. IFRS9 IFRS 8asks for absolute objectivity.   It continues to ask that forward-looking overrides are based on scenario-weighted outcomes.   In other words, it requires all forward-looking overrides to be applied to the extent that they are likely to realise (based on probability).

It is required by IFRS9 (within reasonable cost and effort) to consider scenario weighted forward-looking adjustments.  Per inference, this would require you to adopt multiple forward-looking scenarios. Our advice is to stick to a modelled approach as described under the “economic forecasting and credit-loss relationship modelling” section and avoid the use of management judgements (where possible) unless manual adjustments are shown to be entirely necessary. When regular management adjustments become the norm, it may be worth considering rebuilding your models.

Principa offers a wide-range of products and services relating to IFRS9.  These include analytical, advisory, model templates, reporting packs, deployment support, model software, etc.  We’ve also been able to help companies post-IFRS9 to make the most of their modelsContact us to find out how Principa might be able to help your business.

Get IFRS 9 Compliant

Thomas Maydon
Thomas Maydon
Thomas Maydon is the Head of Credit Solutions at Principa. With over 13 years of experience in the Southern African, West African and Middle Eastern retail credit markets, Tom has primarily been involved in consulting, analytics, credit bureau and predictive modelling services. He has experience in all aspects of the credit life cycle (in multiple industries) including intelligent prospecting, originations, strategy simulation, affordability analysis, behavioural modelling, pricing analysis, collections processes, and provisions (including Basel II) and profitability calculations.

Latest Posts

Will Your Machine Learning Models Pass The William Tell Test?

Machine learning models can be used very successfully in many different contexts to predict outcomes for different use cases accurately. These predictions can be used within the business to make better decisions or to operate more efficiently (or both) and can give you an edge over your competitors. Predictive models all follow the same recipe – i.e. train a model on historical data and then apply this model to unseen data to get predictions. If your model generalises well, you have a prediction that you can trust and use to decide "do this, not that" with some degree of accuracy.

Everything You Need To Know About MarketWise: Your FAQs Answered

What is MarketWise? MarketWise is an audience of online users within South Africa which can be segmented and arranged according to a brand or advertisers target market. Online ads can then be served to this audience as they browse the internet.