How Jaywing's free syndicate and AI-powered modelling combats fraud
The PRA's IFRS 9 review - a complete guide
IFRS 9 – What you should you be thinking about now
13 June 2019
The 2018 IFRS 9 deadline has been and gone but the challenges continue for lenders, as they look to enhance models and infrastructure as part of BAU. Here, you can listen back to our recent webinar with UK Finance to find out more.
Over a year ago, most firms were solely focusing on developing IFRS 9, a time in which many other projects were put on hold as the regulation required transformational change to most lenders' existing infrastructures.
Since then, some firms have reaped multiple benefits from IFRS 9, beyond regulatory compliance by using the new models for forecasting, stress testing, IRB and profitability. But, for many, the main focus is simply on learning how to run IFRS 9 models effectively and build upon limitations in the first-generation models.
In a recent webinar with UK Finance, we covered the key areas currently surrounding impairment forecasting under IFRS 9:
- Monitoring performance of forward-looking models
- Best practice in Significant Increase in Credit Risk (SICR)
- Limitations of using historical data in macroeconomic modelling
These are especially pertinent given the recent PRA review, which provided an update and the main thematic findings from written auditor reporting.
Through surveying the c.300 attendees, we got some interesting insights. Here's a sample:
- Only 6% of firms surveyed feel their IFRS 9 solution is fully compliant and with no enhancements planned
- Almost half (45%) have automated IFRS 9 model monitoring processes either live or in development, a figure that we would expect to increase as industry best practice evolves
- The majority, 54.4%, said they use a simple approach to SICR with quantitative threshold(s) applied
- Only 16.9% said they had a sophisticated approach to economic modelling with no manual overlays required, in line with the PRA’s recent focus on the need for improvements to core models
For more stats and to see where you sit among your peers, you can listen back to the webinar recording here.