A journey through the IFRS 9 looking glass: Using write-off weighted coverage to assess the adequacy and reasonableness of expected credit losses for banks

04 June 2020

As someone who spends the majority of his time buried deep within IFRS 9’s expected credit losses (ECLs) and related modelling intricacies, it’s easy to focus on details. However, it’s also helpful to step back and understand how one bank’s reserves compare to history and its peers. Of course, banks aren’t identical, and nor should their provisions be. Each has its own unique portfolio, geography, risk appetite, loss experience, expectations about the future, assessments of risks, and techniques for managing them. Together with the fact that each also has unique data and models, it’s easy to lose hope for meaningful comparison. Still, I think it can be done.

While two banks might have very similar allowances in proportion to their loans, that doesn’t necessarily mean they’re equally provided, given that banks don’t lose money at the same rate (for the reasons mentioned above). One approach to taking this into account is to use what we call the write-off weighted coverage (‘WoW coverage’ for short). Consider two banks – Bank A and Bank B. In order to determine the WoW coverage, I first want to know how much each bank has lost historically per dollar loaned (that is, historical net write-offs each year as a percentage of gross loans outstanding during the year, calculated over a reasonable period of, say, 10 years). Let’s assume that it averages 1% for Bank A and 2% for Bank B. I then want to know how much each bank has currently reserved per dollar of performing loan outstanding (that is, for Stages 1 and 2, the allowance for credit losses as a percentage of loans). Call it 3% for each. Notwithstanding that they’re equal (at 3%), using WoW coverage which provides a lens based on historical loss experience, you can quickly see that Bank A looks to be twice as well provided as Bank B:


Is it a perfect measure? No, of course not. For starters, loss experience might have changed considerably over the historical period (for example, think changes in products, portfolio composition, risk management policies, strategy, acquisitions, divestitures and the like). For those reasons and others, the range of write-offs during that period is certainly worthy of examination and, while 10 years might be reasonable as a starting point, for some banks or jurisdictions a shorter period might be necessary to better reflect current portfolios and conditions. Write-off policies might differ between banks too, though our hope is that over a reasonable period they’d even out. On the plus side, using historical losses means that many of the bank-specific attributes (such as portfolio mix, risk appetite, and credit risk management practices) should be captured.

As has been said many times, ECLs are not an estimate of the losses ultimately expected to transpire under a single outcome, but rather a probability-weighted average using a range of possible outcomes. For that reason, WoW coverage is not a surrogate for the details and complexity that banks necessarily apply in developing ECLs. Rather, it’s a useful starting point for asking questions about the adequacy and reasonableness of their result. Put simply, what WoW coverage tells us is (all other things being equal) how conservative or aggressive one bank’s provision is compared to another’s, by taking into account their relative historical loss experience, notwithstanding that there might be valid reasons for differences. For instance, some banks might be more optimistic and have a more favourable economic outlook, which should be reflected in the measurement of their ECLs. In our example, Bank B might believe that the economy is due for a quick rebound, which could explain why it appears to be significantly less provided on the basis of its WoW coverage. That would make sense and the question then is how reasonable is the speed and magnitude of that rebound and its effect on Bank B’s ECL. On the other hand, if Bank B’s outlook is drearier than Bank A’s, there’s surely cause for deeper investigation.

Our guest blogger is Chris Wood, Banking Partner and IFRS 9 specialist, connect with him on LinkedIn here. 


Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been saved. Comments are moderated and will not appear until approved by the author. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Comments are moderated, and will not appear until the author has approved them.