In life, it does not matter how much work you put into something if you are heading in the wrong direction, as one Olympic athlete once discovered to his dismay:
In the 1912 Stockholm Olympics, Japanese marathon runner Shizo Kanaguri suffered from hyperthermia and stumbled off course and into a garden where a picnic was in full swing. The family gave him a drink and put him to bed, where he fell asleep. When he woke up hours later, he was too ashamed to tell anyone what had happened so caught a train back to Stockholm and then the boat to Japan. Swedish authorities considered him missing for 50 years before discovering that he was living in Japan and had competed in several Olympics marathons.
So how does this relate to IFRS 9?
Well, just like poor Kanaguri failed to study the local route, if you fail to pay attention at a local level, you are setting yourself up for a costly detour. IFRS 9 requires you to look at how a range of economic scenarios could affect your portfolio. The direct route to accurate loss forecasting is to consider this at a granular level, understanding how different segments will be impacted in different ways.
For example, as well as the differences in people’s household finances, the economy itself also varies according to geography or industry sector. You will need to model economics into your loss forecast at this granular level to take into account all of these variations.
One challenge you might face is the availability of appropriate data. This could be because of a lack of depth or breadth in the information you hold on the historic performance of your accounts.
The recession is a further complication: if you have enough data on the performance of your accounts, going back the several years needed for lifetime loss modelling, it may span the period of recession and so may not be truly indicative of the behaviour you can expect going forward.
That is why you will need a robust loss-forecasting model with sufficient granularity to recognise the impact of these economic stresses on the various subsets of your portfolio.
Modelling at this granular level improves the accuracy of your loss forecast and therefore your provision, allowing your business to free money for investment which would otherwise be sitting in provisions unnecessarily.
Experian can help, by using our wealth of economic data and expertise in modeling at the granular level to create highly accurate loss forecasts.
For more guidance on how Experian can help you improve the accuracy of your loss predictions and provisions , read our more on our ‘Loss forecasting under IFRS 9’ webpage.