12 December 2018
The big day has finally arrived and banks are beginning to publish their first financial statements under IFRS 9. First up are the Canadian banks – six have reported over the last few weeks.
As expected, the focus of these banks’ disclosures is on impairment. Classification of financial assets generally has less impact under IFRS 9, so we’ve seen less information on this.
So, what are our first impressions?
As IFRS 9’s impairment model is forward-looking, banks are required to consider future economic scenarios to calculate expected credit losses (ECL).
All six of the Canadian banks have disclosed their assumptions about key macroeconomic variables. The most common variables they disclosed are GDP growth rates, unemployment rates, house prices and oil and commodity prices. All six disclosed this information for Canada, three for Canada and the US separately, and one separately for seven countries.
For the base scenarios, all six banks provided separate assumptions for the next 12 months and for the remaining forecast period. Four banks provided the same information for the alternative scenarios and two disclosed only the remaining forecast period for the alternative scenario.
All six banks have provided a table reconciling, by portfolio, internal risk ratings of exposures to the different stages of IFRS 9’s impairment model. This provides a useful explanation of the relationship between the absolute credit risk at the reporting date (which was the basis of the previous impairment model under IAS 39 Financial Instruments: Recognition and Measurement) and the relative change in credit risk on which the new impairment model is based.
All six banks have disclosed some analysis of loans that are past-due but not credit-impaired – i.e. in Stages 1 and 2 – but two banks excluded loans that are less than 30 days past due. This information was given in aggregate for Stages 1 and 2; it would be interesting to see a similar analysis separately for Stage 2 loans.
What is required?
IAS 1 Presentation of Financial Statements1 requires an entity to disclose information about the assumptions it makes about the future, and other major sources of estimation uncertainty, that have a significant risk of resulting in a material adjustment to the carrying amount of assets and liabilities within the next financial year.
There has been a debate about what type of information banks are required to disclose to satisfy this requirement, and whether it is in addition to the information specifically required by IFRS 7 Financial Instruments: Disclosures.
The European regulator, ESMA, noted in its enforcement priorities for 20182 that ‘the general disclosure requirements on judgement and sources of estimation uncertainty (including in some cases sensitivity of carrying amounts to methods and assumptions) in IAS 1 are also applicable to the new ECL model, when relevant.’
It may be difficult for banks to compile disclosures about how ECL amounts could change if different assumptions – particularly about the future – are made. Time is needed to re-run impairment models and comply with internal review and governance processes. Also, there may be interactions between different assumptions.
In late November, a disclosure group3 set up by UK regulators considered this issue. It noted that ‘the subject is complex and multifaceted, and it believes there are advantages in seeing how firms approach the subject initially before encouraging practice to converge on any particular approach.’
So what have the six Canadian banks disclosed?
We’ve seen two types of information about how the ECL estimate might change under different assumptions:
Impact of staging
All six banks disclosed the impact of staging by showing how ECL would change if all performing loans were in Stage 1, under which ECL is measured on a 12-month basis. On average, the ECL allowance would reduce by 23% (ranging from 11% to 36%).
Conversely, one bank also noted that the ECL allowance would increase by 54% if all performing loans were in Stage 2 under which ECL is measured on a lifetime basis.
Impact of multiple economic scenarios
Five of the Canadian banks provided information about the impact of including multiple economic scenarios.
Similar to the observation we made in our previous blog on UK banks, the range of sensitivities to multiple economic scenarios is sizeable. For the five Canadian banks, the average increase in ECL as a result of using multiple economic scenarios is 8% (ranging from 0.2% to 18%). The chart below compares the impact on the UK banks with the impact on the Canadian banks.
Click to enlarge chart (JPG, 97 KB)
In addition, some Canadian banks also made disclosures about how ECL would change if they only used the upside or downside scenario for measuring ECL for their performing loans.
It will be interesting to see how other banks approach disclosures in this difficult area.
None of the banks reported a large impact from the new classification requirements; however, some detail was given on particular types of portfolios that needed to be reclassified.
In particular, one bank reported that financial assets that are mandatorily measured at fair value through profit or loss include certain precious metal loans (because they did not meet the ‘solely payments of principal and interest’ criterion) and certain residential mortgages (because they were held in a trading business model).
These first full financial statements under IFRS 9 are early examples. It will be interesting to see how European banks compare when they publish their annual financial statements in a few months’ time.
In the meantime, take a look at our publication Guides to financial statements – Illustrative disclosures for banks, which provides a detailed guide to banks’ financial statements under IFRS 9.
1 Paragraph 125 of IAS 1.
2 The European common enforcement priorities for 2018 annual reports, issued in October 2018, page 5.
3 Report issued by The Taskforce on Disclosures about Expected Credit Losses on 29 November 2018, page 29.