Since 2018 firms have been required to run stress tests on an IFRS 9 basis, but are lenders really doing this?

In a recent webinar, Jaywing used learnings gained from the first round of stress tests run under the new accounting standards, both Concurrent Stress Tests (CST) and Internal Capital Adequacy Assessment Processes (ICAAP), to offer best practice guidance on IFRS 9 stress testing modelling, setting capital buffers and more.

IFRS 9 stress test key technical aspects

The estimation of stage transition  is an aspect of IFRS 9 that has required significant adaptation to existing stress testing frameworks. Defining a suitable approach to stage migration, compliant while proportional to business size, is a fine balancing act and most firms are likely to still be refining their approach.

Unsurprisingly, the majority questioned have implemented at least a partially strategic solution to comply with the immediate requirements of running stress testing under IFRS 9 in 2018. Most firms now plan to improve on the exisitng methodologies, either because they have been told by auditors/regulators to do so, or because they recognise the limitations of their first generation of models - only 3 per cent of attendess stated that they expected no change to their current approach.

Capital impact of IFRS 9

From a computational point of view, the different timing and quantum of impairments under IFRS 9 is likely to have an impact on a firms? capital requirements under stress.

On the whole, some form of increased capital impact has been observed, or is expected, by the audience.

The Prudential Regulation Authority (PRA) set the amount of the stress capital requirement for individual firms based on the largest capital drop experienced during the stress. As IFRS 9 implies, a larger drop in the early years of the stress could potentially lead to higher stress capital requirements.  This seems counterintuitive, as the higher (and earlier) accumulation of losses in the stress should guarantee more resilience in a downturn.

Showing a good understanding of the IFRS 9 process and results is key as a lack of confidence in the IFRS 9 stress testing results, or perceievd gaps in the process, can trigger the imposition of financial penalties through additional capital requirements.

IFRS 9 models integration

Model integration seems the right path to follow given the clear benefits in terms of consistency and efficiency of a framework based on common model inputs, economic impacts, governance and data infrastructure, but it is easier said than done. The initial investment of time and resources required, coupled with the possible separation between different functions, can make it an attractive but unattainable proposition.

We were pleasantly surprised that nearly half (47 per cent) surveyed said their stress testing models were either mostly of completely integrated with their IFRS 9 and/or IRB models. While some still recognise the need to improve, there was a clear trend that most of the audience regard their models to be at an advanced stage of integration with other business areas.

These are encouraging results, but we strongly believe this to be an area which will continue evolving as regulatory expectations evolve.

For additional guidance on running IFRS 9-based stress tests, download Jaywing's guide.