On 1 January 2018, banks applying IFRSs move into a new era with the implementation of IFRS 9 on financial instruments, which will supersede the existing IAS 39.
Although adopting this new standard will lead to many changes, it was phase 2 of the standard, on provisions, that interested the Basel Committee on 11 October, since the new impairment model will undeniably have an impact on regulatory capital, in both the short and long term.
Readers will recall that under IFRS 9 credit losses will be recognised earlier, with an “expected credit loss” approach (ECL) in place of the incurred loss approach that was applied under IAS 39. This will affect (a) all balance sheet financial assets measured at amortised cost or at fair value through other comprehensive income and (b) given commitments and financial guarantees. To anticipate these impacts, the Basel Committee is putting forward a number of approaches to the management of regulatory capital which we will discuss in this new article.
Before setting out some of the potential consequences of IRFS 9 phase 2 on regulatory capital, here are some key dates:
– 24 July 2014: IASB (International Accounting Standards Board) published the final version of IFRS 9 (not yet endorsed by the European Union)
– 18 December 2015: BCBS (Basel Committee on Banking Supervision) guidance on the recognition of Expected Credit Losses (ECL)
– 26 July 2016: draft EBA (European Banking Authority) guidance on the practices of institutions as regards credit risk and the recognition of expected credit losses
– 1 January 2018: implementation date of IFRS 9 for banks
Note that insurance companies will apply IFRS 9 from 1 January 2021 in conjunction with the introduction of IFRS 4 phase 2 (insurance contracts), so the fate of bankinsurers is still unclear, the IASB having rejected the postponement of IRFS 9 while the European Commission should give a ruling by the end of 2016.
Existing treatment of provisions in regulatory capital
Since the transition to Basel II, there have been two different approaches to the management of provisions in the prudential sphere:
Entities using the standardised approach:
– No calculation of the regulatory expected loss
– Specific provisions are deducted from the gross carrying amount, i.e. Risk-Weighted Assets (RWA) are calculated on the basis of an exposure at default (EAD) net of specific provisions.
– General provisions are compared with 1.25% of RWA in the standardised credit risk method (excluding securitisation and other assets that do not correspond to credit obligations) and the lesser of the two amounts is eligible for inclusion in Tier 2 capital – article 62 (c) of the CRR
Entities using the IRB approach:
– There are several calculations for the regulatory expected loss (the formula PD x LGD x EAD / flat rate/ zero), depending on the exposure category.
– There is no difference between the treatment of general and specific provisions.
If total provisions exceed the EL, this excess is compared to 0.6% of RWA in the IRB approach to credit risk (excluding securitisation and equities) and the lesser of the two amounts is eligible for inclusion in Tier 2 capital – article 62 (d) of the CRR.
If total provisions are lower than the EL, the shortfall is deducted directly from CET1 capital –article 40 of CRR.
This mechanism, while now well-established, has some limitations, particularly in relation to the distinction between general and specific provisions, which does not exist under accounting frameworks and which varies from one jurisdiction to another, or even from one institution to the next. That defeats the European regulator’s aim of reducing the variability of the models used and promoting a level playing field for European banks.
How will the arrival of IFRS 9 affect future regulatory treatment?
Although we do not really know the impacts yet, it is expected that, overall, IRFS 9 will have a materially negative effect on the level of CET1 capital because of a much higher level of provisioning, due to:
– Calculation of expected loss at 12 months on performing loans (bucket 1) whereas no provision was previously made;
– Calculation of expected loss at maturity for significantly deteriorated loans (bucket 2)
– Calculation of expected loss at maturity for doubtful loans (bucket 3) (the impact of the transition to IFRS 9 will be more modest than for buckets 1 and 2)
Now, CET1 capital is very sensitive, for reasons of both regulatory requirements (floor rate fixed at 4.5% + capital buffers) and communication with the market.
The Basel Committee has put forward a number of short-term solutions for transition to IFRS 9:
Approach 1 – Comparison between CET1 regulatory capital at 31 December 2017 (under IAS 39) and at 1 January 2018 (under IFRS 9). Where this shows a reduction in CET1, this decline could be temporarily adjusted and spread over 3 to 5 years (net of tax effect). This would make it possible to level out the negative effect but would freeze the impact over time and make necessary a new transitional adjustment coupled with the current filters and deductions applied to CET1 capital (phase-in).
Approach 2 – Determination of a percentage equal to the decline in CET1 capital due to the transition to IFRS 9 calculated on the total amount of provisions at moment of transition. The transitional adjustment mechanism used in approach 1 would be applied, but calculated on an amount equal to the stock of provisions at a given date multiplied by that percentage. This would permit more dynamic management, despite a percentage fixed at the outset.
Approach 3 – Taking account of the total actual amount of accounting provisions for buckets 1 and 2 at the moment of transition to IFRS 9 by applying a percentage determined by the Basel Committee at each reporting date, depending on the duration of the transitional period.
Whichever approach is taken, entities applying the IRB method would have to deduct any EL shortfall from the amount of provisions before transition to IFRS 9.
In any case, the Basel Committee’s proposals embody the current dilemma between seeking more convergence between accounting standards and prudential standards, despite their different bases (principles-based versus rules-based, respectively), and avoiding too severe an impact on regulatory capital, at a time when the finalisation of the Basel III corpus has exposed the Basel Committee to lively criticisms, nourished by fears of a significant increase in capital requirements.
Looking beyond the short-term, the Basel Committee has also launched a more general review of the regulatory treatment of provisions and expected losses by proposing more long-term changes:
– Aligning the treatment of provisions by entities using the standardised credit risk approach with the treatment applied in the IRB approach, meaning that specific provisions would no longer be deducted from the gross carrying amount in the solvency ratio. This would require entities to take account of the impact on other ratios (leverage and major risks, principally).
– Revising the classification of general and specific provisions, as mentioned above, to the point where it is no longer possible to distinguish them, particularly if the standard methodology is aligned with the IRB approach.
– Introducing an EL calculation in the standardised approach, determining an EL rate per asset class using the same granularity as for RWA rates, except for loans in default where the EL would be also based on the ECL calculation. The rate would then be a function of the PD and the LGD in the IRBF methodology.
– Removal of the option to add back excess provisions to Tier 2 capital
– A review of the treatment of provisions under Pillar 2, particularly excess provisions, the recognition of which could reduce the additional need for capital.
With this consultation, the Basel Committee is launching a review that many banks were waiting for, given the tsunami of IFRS 9 and the sensitivity of regulatory capital, under economic and regulatory circumstances that call for increasingly close capital monitoring.
It remains to be seen how professionals in the sector will react to these proposals and how the timetables for implementation will combine with the other reviews in progress, led by the reviews of the standardised approach and internal credit risk models.