Banking Institutions IFRS 9 ECL Provisions

Blog Article By Claxon Actuaries Team

Banking Institutions IFRS 9 ECL Provisions COVID 19 impact and considerations


The current global crisis as result of COVID 19 has affected businesses and lives in unimaginable ways. Globally, countries are affected differently with many being at different stages of the ‘curve’. Some experts anticipate a first and second wave over a prolonged cycle. In this scourge; banks, which are known to be a pivotal stimulant for development and growth for economies are at the heart of the crisis.

IFRS 9 ECLs stress testing and scenario analysis is expected to provide a range of outcomes to allow for the current crisis. However, even with macro-economic and forward-looking scenarios in place, it would be hard to imagine if any model forecasted compellingly a ‘pandemic’ scenario we woke up to. ECLs are expected not to be overly too optimistic or too cautious; being unbiased probability weighted best estimate outcomes. This could imply drawing from a set of plausible scenarios. COVID 19 could arguably qualify as an outlier event.

This article looks at IFRS 9; focusing on expected credit loss provisions for loan exposures as one key area of financial disclosures under the spotlight. It looks at possible impacts and key considerations under this crisis. Similar principles could apply to other bank exposures or financial sectors.

Governments and Regulators guiding the way

For most jurisdictions, including ours; public policy makers and regulators are leading the way through providing guidelines. Irrefutably, they should. Reasons are multiple fold but a few include;

COVID 19 could turn out to be a short-term phenomenon. Hence, adjusting IFRS 9 ECL models could be premature. The models have lifetime default likelihoods which could be difficult to properly adjust for with supportable information, past and future. Consistent and stable approaches are preferred.

Similarly, there is need to stonewall any adrenaline rush which could lead to overprovisioning on ECLs due to some of the assumptions built in the models. Thus, any effects to businesses which are COVID 19 induced are broadly being exempt from being mechanically allowed for in models. Rightly so, most borrowers are ready to play the ‘force majeure’ card when coerced. Where considerations for adjustments are being made these are being cautiously done. For now!

Banks operating in Zimbabwe have been asked to come up with business impact assessments by the 30th of June 2020. IFRS 9 ECLs could be potentially one of the tough exercises to conduct confidently.

Impact on ECL models; an in-depth look at some of the aspects

Assuming there are no external measures as outline previously then undoubtedly COVID 19 may warrant a worsening off of PD, LGD and EAD estimation parameters. The key driver being any outcomes of credit assessments that would lead to Significant Changes in Credit Risk (SIRC) for different borrowers.

Credit Risk Assessments

This would, however, impact different sectors or portfolio segments to varying extents. Even in the aftermath of the crisis some sectors may recover easily. For example, with stay home and social distancing orders the Transport sector is severely impacted on. On the flip side, the need to utilize different IT platforms could draw a different picture for the IT related sectors. Similar arguments could be made for individual loans.

One big challenge is the bank’s ability to separate genuinely those borrowers for certain sectors who are severely impacted by COVID 19 from those who had already been due for a downgrade to stage 2 or 3. This may imply thorough credit assessment with perhaps a revision to current credit policies. In Romania, some sectors are issued with Certificates of the Emergency Situation by a government ministry which could be relied on by banks to predict expected loss impacts. Such a certificate could also assist banks in crafting their internal policies on who to grant concessions to, deem performing and non-performing when carrying out classifications.

Model Adjustments for COVID 19

Some considerations if adjusting the model is conceived;

Macroeconomic scenarios and associated weights that are applied by banks in their models need to be developed. However, these would need to be done relying on supportable information. IASB has advised that, at the moment, it is difficult to allow for specific effects of COVID 19 and government measures being put in place. They have advised that if the scenarios could not be developed easily then other approaches such as post-model overlays or adjustments would have to be considered.

An approach to use overlays would, however, need to be backed by sound credit risk policies and governance issues which may then form part of the key disclosures made.

European Securities and Market Authority (ESMA) notes that in cases where issuers encounter challenges in isolating the COVID 19 effects at instrument level, it could be sufficient to perform the assessment on a collective basis [IFRS9 paragraphs B5.5.1-B5.5.5]. Furthermore, ESMA also note that if temporary relief measures to borrowers are not impacting the net economic value of the loan significantly then accounting modifications required could be considered unsubstantial.

European Central Bank (ECB) recommended that banks give greater weight to long-term stable outlook as evidenced by past information and consider the various relief measures being granted by the public institutions.

Temporary nature of measures being suggested to counter COVID 19 may have a minimal impact on other aspects such as lifetime ECLs. Most temporary measures are aimed at minimizing liquidity strain on borrowers due to health related measures being taken to curb and contain the virus.

For these temporary reliefs, care is needed to avoid procyclical assumptions to avoid undue short-term volatility in the reported ECLs. This could also make parameter back testing in models difficult to understand and justify.

General guidelines being given by different regulators suggest against moving borrowers into stage 2 ECL due to measures such as repayment holidays etc.

The different cocktail of measures could also mean banks may revisit and could rebut the 30 days past due presumption in their staging as one of their judgmental calls.

Different products will behave differently in terms of magnitude of increases in risk e.g. secured versus unsecured loans, and revolving facilities versus term loans. Furthermore, the perceived effort required to carry out individual assessments could be untenable influencing many banks to carry out adjustments which are of a probabilistic nature.

There could be need for re-modelling of parameters after an observation period. How long the waiting period would be is contingent on the pandemic being declared over and normalcy restored sustainably. This is a judgmental area for decision makers. When it comes to modelling the past information which is readily available, easily supportable and obtained mostly without undue costs still remain the main source of information to build and test robustness of models.

In most jurisdictions, there could be a lot of sovereign backed guarantees as stability levers by the state. These could result in LGD adjustments. The same could be said of Bank guarantees.

Collateral valuations could also result in significant haircuts impacting ECL adjustments. These volatilities could enhance uncertainty over ECL outcomes.

Going forward;

The pandemic still unravels. Some global projections are looking at V, U and L shaped curves with high uncertainty. Governments and regulators are likely to continue taking steps to both contain the spread and support their economies. Unfortunately, this is being done in steps as the experts cannot attest to having contained the virus. Under such circumstances IFRS9 ECL models face many practical adjustment challenges.

Most banks globally have built in buffers already in their capital reserves; would regulators consider softening and allowing banks to use these reserves to provide further lending during this crisis period? BCBS advises that a Transitional relief on IFRS 9 could be applied. They advise for this to be jurisdictional and would also require enhanced reporting which also illustrates full capital positions.

Where loan modification has taken place, any change in the present value of the modified contractual cashflows would be reflected as modification gain or loss, potentially changing the exposure amount on which risk weighted assets would be calculated for regulatory purposes.

What will the post pandemic period be like?

How suitable would the current models be for the potential new world likely to be born out of this crisis. New loan originations may become constrained as banks may reduce appetite for certain exposures unless Government provides a guarantee scheme to support borrowers. Collections could also become a challenge and loss given defaults may increase due to fall in asset values.

Isolating the impact of Covid19 from genuine business failures could be a challenge. Even facilities which remained in grade 1; in a good state, may start experiencing their own business challenges if they are in sectors that may not be adaptable easily. What would sectoral recovery periods be. E.g Tourism and Agricultural sectors may have prolonged recovery cycles.

IFRS 9 standard does not provide bright line rules or a mechanistical way to model for these stressed times. There are a lot of judgmental areas within the standard itself. If a bank develops a bespoke approach for their exposures and can support and apply them, the ECL provisions still stand as a powerful tool to communicate with consumers of financial reports.


If you would like to comment or have a discussion around this article contact the author on This email address is being protected from spambots. You need JavaScript enabled to view it.


Authored by the Claxon Actuaries Team