The European Central Bank recently published the findings from its targeted review of internal models (TRIM), and its outcome serves as a wake-up call. In total, 5800 weaknesses were identified, of which 30% had a high severity and in total translated into the issuance of 253 supervisory decisions.
The Trim project, launched in 2016 and completed in April, had two main objectives:
- Ensure bank’s internal models met regulatory requirements
- Reduce non-risk based risk-weighted assets (RWA) variability across financial institutions.
In total, 200 on-site model investigations across 65 top European institutions were conducted covering credit, market and CCR models.
The final outcome of the exercise will lead to a whopping 275 billion euro (330 billion USD) increase in aggregated RWA’s, translating into an average 71 basis point erosion of CET1 ratios for the institutions in the scope of the exercise.
When diving deeper into deviations and subsequent supervisory decisions being issued within the general topics, the areas of concern quickly become apparent. The graph below shows that validation and internal governance were the areas with the largest number of deviations for Counterparty Credit Risk. For the non-model specific topics, where credit risk was the focus most deviations where found in: organization and activities of the internal validation function, roll-out of PPU and de overall management of model changes.
And although the study acknowledges that effort has been made at the institutions to ensure sound data management and quality frameworks are in place, it also clearly indicates that amendments are still needed to ensure compliance with: data quality management and internal control and the allocation of roles and responsibilities relating to data management.
In light of the current COVID crisis, it clearly shows that this could have proven disastrous as buffers at many of these top lenders might have proved insufficient to absorb the losses incurred on their lending books in the first months of the COVID pandemic. If not for the quick action of governments, which guaranteed loans and an avalanche of regulatory changes by European authorities, it could have spelt doom for the European banking sector merely a decade after the previous crisis.
The question which presents itself now is: have banks learned enough from TRIM to ensure better compliance and reduced underestimation of their risks? The key to accomplishing this is improvements in the data management process capabilities. Shortcomings around internal model data where found in all onsite investigations, with problem areas situating around internal control frameworks and weaknesses in the allocation of roles and responsibilities within data management. Further reduction in model deficiencies could be accomplished by strengthened independent internal validation functions with a goal to prevent capital requirement “gaming” by model users.
The Trim clearly highlights area’s where bank executives should focus manpower and investments. Failure to do so could saddle banks with huge costs in the form of higher RWAs. These investments could be part of the overall Basel IV efforts as the remediation of TRIM findings is complementary to the layers of safeguarding against inadequate internal models as set out by the new Basel IV regulation.
As we gradually emerge from the COVID crisis, and with EU banks profitability under stress, it’s easy to prioritize initiatives that translate into higher earnings. However, failure to act might could come back to haunt you in the long run.