The content on this page has been converted from PDF to HTML format using an artificial intelligence (AI) tool as part of our ongoing efforts to improve accessibility and usability of our publications. Note:
- No human verification has been conducted of the converted content.
- While we strive for accuracy errors or omissions may exist.
- This content is provided for informational purposes only and should not be relied upon as a definitive or authoritative source.
- For the official and verified version of the publication, refer to the original PDF document.
If you identify any inaccuracies or have concerns about the content, please contact us at [email protected].
IFRS 9 'Financial Instruments' Disclosures in the First Year of Application
- Executive Summary
- Overview of the thematic
- Transition
- Non-banking entities
- Banks - Classification and measurement
- Banks – Impairment: policies and methodology
- Banks - Impairment: staging and credit risk profile
- Banks – Impairment: multiple economic scenarios
- Banks – Impairment: estimation uncertainty
- Banks - Hedging
- Next steps
Executive Summary
Introduction
This report summarises the key findings of our review of disclosures in the first annual report and accounts following the adoption of IFRS 9 'Financial Instruments'. It is a follow-up to our report published in November 2018 which considered the disclosures made in 2018 interim accounts relating to the implementation of IFRS 9.
In our previous review we concluded that, although interim disclosures are less extensive than those for full year accounts, some companies, in particular the smaller banks, did not sufficiently explain the impact of adopting IFRS 9. As a result, whilst this review considers the quality of disclosures made by a number of the larger banks, we have increased the number of smaller banks in our sample. In addition, we also consider the disclosures made by a number of non-banking entities.
Although we believe that it is more valuable to focus on disclosure requirements which will apply to future reporting periods, we have considered the disclosures addressing the impact of the first time adoption of IFRS 9.
Key findings
We have included a number of excerpts from published annual report and accounts to illustrate helpful ways to present and explain how the requirements of IFRS 9 have been applied. These excerpts may enable users to understand the nature and extent of exposures to financial instruments.
The quality of the disclosure was high among the larger banks. We were also pleased to see some good examples of disclosure among some of the smaller banks in a number of areas, including detailed explanations of how forward looking information has been incorporated into the calculation of expected credit losses ('ECLs') and the sensitivity of ECLs to changes in future economic conditions.
However, our review also identified a number of areas where disclosure could be improved and some areas where no disclosure had been provided at all.
Separate topics are addressed throughout this report, but our key findings were that the following disclosures could be improved:
- Analysis of the credit risk profile of financial assets by credit risk ratings grade, days past due or in a provision matrix. This was not done well by non-banking entities in addition to two banks in our sample;
- Disclosure of the qualitative and quantitative factors used to determine whether there has been a significant increase in credit risk;
- Explanation of how the simplified approach has been applied to trade receivables, contract assets and lease receivables; and
- Discussion of the business model in assessing the classification of financial assets, which tended to be boilerplate by most of the banks and not provided by a number of non-banking entities.
We recognise that this is the first full year of adoption and that disclosures will continue to develop over time. We will, however, continue to review compliance with the standard through our normal review work.
We hope that preparers find this review useful and we encourage engagement with external auditors to plan for future reporting periods.
Overview of the thematic
Scope of our review
Our review consisted of a limited scope desktop review of the annual reports and accounts of entities applying IFRS 9 'Financial Instruments' for the first time. Our review considered the comprehensiveness and quality of the disclosures against the requirements of IFRS 9 and IFRS 7 'Financial Instruments: Disclosures', and the judgement and estimates requirements in IAS 1 'Presentation of Financial Statements'.
In addition, in performing our review of the disclosures made by banks, particularly the larger banks in our sample, we considered the additional disclosure recommendations included in the following reports:
- 'Impact of Expected Credit Loss Approaches on Bank Risk Disclosures' published by the Enhanced Disclosure Task Force in November 2015; and
- 'Recommendations on a comprehensive set of IFRS 9 Expected Credit Loss disclosures' published by the Taskforce on Disclosures about Expected Credit Losses in November 2018.
The application of the ECL by banks requires the use of complicated models to determine the level of loan loss provisions. Our review did not consider the reasonableness of the assumptions used in those models nor did we assess the appropriateness of the models applied.
Our sample
We reviewed the full-year accounts of a sample of 20 entities.
Our sample targeted those industries on which we would expect the implementation of IFRS 9 to have the most significant impact. As a result, our sample was skewed towards banking entities, in particular smaller banks 1.
Transition
Generally companies provided sufficient explanation of the impact of adopting IFRS 9, but the quality of the disclosure varied across the sample.
We found that the larger banks reduced the level of disclosure regarding transition compared with that provided in the interim reports and the transition documents published in 2018. However, it was sufficiently granular to understand the impact of adoption. Where the impact was material, the better disclosures disaggregated the adjustment to show individually the impact of applying the classification, measurement and impairment requirements.
Most entities disclosed the accounting policies for financial instruments on both an IFRS 9 and IAS 39 basis. Two entities provided a helpful comparison of the key terms under IFRS 9, IAS 39 and the regulatory framework.
Comparatives
As permitted by IFRS 9, none of the sample restated comparatives, instead the transitional adjustment was charged against opening equity reserves.
The larger banks voluntarily presented the credit risk disclosures at 1 January 2018 on an IFRS 9 basis to enable greater comparability.
We noted that alternative performance measures were generally not restated. In most cases, this was clearly explained. One company, which used a metric based on average net asset value, did adjust the opening net assets in the calculation to include the impact of IFRS 9.
Classification and measurement
Most entities reconciled the changes in the classification categories of financial instruments from IAS 39 to IFRS 9 for affected balance sheet line items. Few provided a qualitative explanation of the reasons for key changes in classification.
Two entities used the adoption of IFRS 9 as an opportunity to change the balance sheet presentation of certain items. The rationale for this change was clearly explained.
Impairment
Other than three entities which had longer term loan balances with customers, most of the non-banking entities reported that the impact was not material. All adopted the simplified approach for the impairment of receivables.
The materiality of the impact of adoption of the impairment requirements for the banks in our sample varied depending on the mix of business. Those with large portfolios of unsecured products and revolving credit facilities experienced more of an impact. On the whole, we found that the disclosure regarding the impact of IFRS 9 on impairment provisions was adequate and commensurate to the size of the entity and loan portfolio.
Hedging
Where non-banking entities engaged in hedging activity, most adopted the hedging requirements under IFRS 9. We found disclosure of the key differences in the requirements compared to IAS 39 was generally adequate.
By contrast, with the exception of one entity which applied IAS 39 for portfolio hedging and IFRS 9 for all other hedging transactions, all of the banks applying hedge accounting continued to apply IAS 39 hedging requirements.
Non-banking entities
Classification and measurement
Almost all non-banking entities in our sample measured cash and trade receivables at amortised cost. Although this was not always well explained, we did not have any particular concerns about this because these financial assets are typically held-to-collect ('HtC') and have contractual terms that meet the 'solely payments of principal and interest' test.
We did observe some examples of cash and trade receivables measured at fair value:
- In one case, an entity had classified an element of trade receivables as fair value through other comprehensive income, which applies when the business model objective is achieved through selling and holding financial assets. The reason for this classification was not, however, very clearly explained in the accounts.
- Another entity had classified investments in money market funds at fair value through profit and loss ('FVPL') because it failed the HtC business model.
Neither change in classification had a material effect.
Equity investments
Most non-banking entities have elected under IFRS 9 to measure their equity investments at fair value through other comprehensive income ('FVOCI'). This classification is similar to the IAS 39 available for sale classification but, under IFRS 9, such equity instruments are not subject to impairment nor are gains or losses recycled.
We did, however, observe one example of a property investment company that did not elect to measure equity investments at FVOCI. Consequently, the changes in fair value were recognised in profit and loss.
Designation as fair value through profit and loss
IFRS 9 retains IAS 39's option to designate financial assets and liabilities at FVPL. Whilst this option is rarely used by non-banking companies, we found one example of an entity that had classified index-linked gilts at fair value through profit and loss to avoid an accounting mismatch on related index-linked derivatives.
Recognition and derecognition
The recognition and derecognition rules under IFRS 9 are the same as IAS 39 except that, under IFRS 9, the carrying amount of a financial liability must be remeasured for all modifications, even those that do not result in derecognition.
We identified one example of a group that remeasured the carrying value of a modified loan between its subsidiaries. Whilst this loan was eliminated at the group level, the change in carrying value at the subsidiary level had a consequential deferred tax effect in the group accounts. This particular example highlights the importance of assessing the effect of IFRS 9 at the subsidiary level and considering possible deferred tax effects.
The main issues identified under classification and measurement were:
- ! As noted here and in our previous thematic, although we would usually expect non-banking entities to have a HtC business model, we would expect the business model to be explained in the accounts.
- ! We noted two instances where an entity provided accounting policies for instruments which were not held by that entity. Accounting policies should be clear, concise and relevant; inclusion of policies which are not relevant only creates unnecessary clutter.
- ↑ Occasionally old IAS 39 terminology has not been updated for the new standard.
Impairment
With the exception of three entities which had longer term loan balances with customers, IFRS 9's new impairment requirements did not have a material effect on most of the non-banking entities in our sample.
Simplified approach
All of the non-banking entities applied the simplified approach, calculating lifetime ECLs for trade receivables and, where relevant contract assets. We found that the quality of the disclosure varied across the sample; in one instance it was not clear that the entity applied the simplified approach.
In every case, IFRS 9 did not result in a materially different loan loss provision for trade receivables. Although this may be justifiable, entities should consider whether the former methodology is consistent with the new requirements, for example, to incorporate forward looking information.
Low credit risk
One company disclosed that it had taken advantage of the low credit risk expedient for investment grade instruments. We identified one entity (a bank) that had applied this expedient to trade receivables, which is not permitted under the standard. Fortunately, the amounts involved were not material.
General model
The three entities which applied the general model included two finance companies and a retailer with a banking subsidiary. The discussion of the ECL methodology was adequate and proportionate to the size of the exposures in each of the entities.
Parent company individual accounts
Parent companies reporting under IFRS should also consider the impact of IFRS 9 on the impairment of financial assets in their individual accounts. In particular, loans to subsidiaries are often material. Only two entities disclosed that they had considered impairment in loans to subsidiaries; this should not be overlooked.
Disclosures
IFRS 7 requires the disclosure of the credit quality of financial assets and movements in loan loss provisions. These disclosures were sometimes inadequate.
Examples of good disclosure...
Centrica plc explained the use of the low credit risk expedient:
> 'In addition, a significant portion of the Group's other financial assets subject to IFRS 9's requirements are in the Group's Treasury function where investment ratings of counterparties result in low credit risk and the calculated loss according to the assessed default rate of these counterparties is not material' > Centrica plc, p130
The main issues identified in this area were:
- ! IFRS 7 requires an entity to disclose the gross carrying amount of financial assets by credit risk rating grades. For trade receivables, contract assets and lease receivables, this may be based on a provision matrix or on days past due. Only one entity applying the simplified approach provided this disclosure. Three entities referred to the use of a provision matrix to determine the loss allowance but did not disclose the matrix. One also provided an analysis of the ageing of receivables.
- ! We identified one instance where an entity held material contract balances but did not appear to apply the impairment requirements.
- ! One entity had an impairment policy which was still on an incurred loss basis.
- ! Where determination of ECLs is identified as a source of estimation uncertainty, we would expect details of the key assumptions and a sensitivity analysis to be provided.
Examples of good disclosure...
Eurasia Mining plc considered ECLs on loans to subsidiaries:
> 'The Group has assessed the estimated credit losses of these loans (to subsidiaries) and given the effective interest rate of the loans is 0%, there would be an immaterial loss expected on these loans.' > Eurasia Mining plc, p42
Examples of good disclosure...
Centrica plc's explanation of the simplified approach and estimation uncertainty was clear and informative:
> 'The IFRS 9 impairment model is applicable to the Group's financial assets including trade receivables, contract assets and other financial assets as described in note S3. As the majority of the relevant balances are trade receivables and contract assets to which the simplified model applies, this disclosure focuses on these balances. > > The provision for credit losses for trade receivables, contract assets and finance lease receivables is based on an expected credit loss model that calculates the expected loss applicable to the receivable balance over its lifetime. Credit losses on receivables due from treasury, trading and energy procurement counterparties are not significant (see note S3 for further analysis of this determination). For residential and business customers default rates are calculated initially by operating segment considering historical loss experience and applied to trade receivables within a provision matrix. The matrix approach allows application of different default rates to different groups of customers with similar characteristics. These groups will be determined by a number of factors including; the nature of the customer, the payment method selected and where relevant, the sector in which they operate. The characteristics used to determine the groupings of receivables are the factors that have the greatest impact on the likelihood of default. The rate of default increases once the balance is 30 days past due and subsequently in 30-day increments. > > Receivables from residential and business customers are generally considered to be credit impaired when the payment is past the contractual due date. The Group applies different definitions of default for different groups of customers, ranging from 60 days past the due date to six to twelve months from the issuance of a final bill. Receivables are generally written off only once a period of time has elapsed since the final bill. Contractual due dates range from falling due upon receipt to falling due in 30 days from receipt........
Sensitivity to changes in assumptions
> The most significant assumption included within the expected credit loss provisioning model that gives rise to estimation uncertainty is that future performance will be reflective of past performance and there will be no significant change in the payment profile or recovery rates within each identified group of receivables. To address this risk, the Group reviews and updates default rates, by group, on a regular basis to ensure they incorporate the most up to date assumptions along with forward-looking information where available and relevant. The Group also considers regulatory changes and customer segment specific factors that may have an impact, now or in the future, on recoverability of the balance. While forward-looking information is usually considered to be immaterial, the exception to this could be the forecast occurrence of a significant one-off event. The Group does not believe that Brexit will have a material impact on the outstanding receivables balance. > > This approach is considered appropriate as the Group's outstanding trade receivable balance is made up of a high volume of individually low value balances relative to the total outstanding debt. As a result, impairment losses on trade receivables are more sensitive to macroeconomic events, rather than customer specific future events, which are unlikely to have a material impact. The Group's receivables are predominantly short term and the rate of default increases significantly when a balance is more than 90 days past due. In order to test the sensitivity to changes in the debt profile, the Group has considered the impact of further credit deterioration of these balances and determined that if all balances were to remain unpaid for a further 30 days, the additional credit loss recognisable by the Group would be up to £20 million.' > Centrica plc, p160
Examples of good disclosure...
NEXT plc impairment policy was found to be clear and comprehensive:
> The directors have taken the simplification available under IFRS 9 5.5.15 which allows the loss amount in relation to a trade receivable to be measured at initial recognition and throughout its life at an amount equal to lifetime ECL. This simplification is permitted where there is either no significant financing component (such as customer receivables where the customer is expected to repay the balance in full prior to interest accruing) or where there is a significant financing component (such as where the customer expects to repay only the minimum amount each month), but the directors make an accounting policy choice to adopt the simplification. Adoption of this approach means that Significant Increase in Credit Risk (SICR) and Date of Initial Recognition (DOIR) concepts are not applicable to the Group's ECL calculations. > > Lifetime ECLs are the ECLs that result from all possible default events over the expected life of a financial instrument. > > ECL is the product of the probability of default (PD), exposure at default (EAD) and loss given default (LGD), discounted at the original Effective Interest Rate (EIR). The assessment of credit risk and the estimation of ECL are required to be unbiased, probability-weighted and should incorporate all available information relevant to the assessment, including information about past events, current conditions and reasonable and supportable forecasts of economic conditions at the reporting date. The forward-looking aspect of IFRS 9 requires considerable judgement as to how changes in economic factors affect ECLs. > > IFRS 9 "Financial instruments" paragraph 5.5.20 ordinarily requires an entity to not only consider a loan, but also the undrawn commitment and the ECL in respect of the undrawn commitment, where its ability to cancel or demand repayment of the facility does not limit its exposure to the credit risk of the undrawn element. However, the guidance in IFRS 9 on commitments relates only to commitments to provide a loan (that is, a commitment to provide financial assets, such as cash) and excludes from its scope rights and obligations from the delivery of goods as a result of a contract with a customer within the scope of IFRS 15 "Revenue from contracts with customers" (that is, a sales commitment). Thus, the sales commitment (unlike a loan commitment) is not a financial instrument, and therefore the impairment requirements in IFRS 9 do not apply until delivery has occurred and a receivable has been recognised. > > Impairment charges in respect of customer receivables are recognised in the Income Statement within cost of sales. > > Delinquency is taken as being in arrears and credit impaired is taken as being the loan has defaulted, which is considered to be the point at which the debt is passed to an internal or external Debt Collection Agency (DCA) and a default registered to a Credit Reference Agency (CRA), or any debt 90 days past due. Delinquency and default are relevant for the estimation of ECL, which segments the book by credit score, banded into very low risk, low risk, medium risk and high risk, by arrears stage. > > Financial assets are written off when there is no reasonable expectation of recovery, such as a customer failing to engage in a repayment plan with the Group. Where receivables have been written off, if recoveries are subsequently made, they are recognised in profit or loss.
The key assumptions in the ECL calculation are:
- PD: The "Probability of Default" is an estimate of the likelihood of default over the expected lifetime of the debt. NEXT has assessed the expected lifetime of customer receivables and other trade receivables to be no more than 36 months, based on historical payment practices. The debt is segmented by arrears stage, Experian's Consumer Indebtedness Index (a measure of consumers' affordability) and expected time of default.
- EAD: The "Exposure at Default" is an estimate of the exposure at that future default date, taking into account expected changes in the exposure after the reporting date, i.e. repayments of principal and interest, whether scheduled by the contract or otherwise and accrued interest from missed payments. This is stratified by arrears stage, Experian's Consumer Indebtedness Index and expected time of default.
- LGD: The "Loss Given Default" is an estimate of the loss arising in the case where a default occurs at a given time. It is based on the difference between the contractual cash flows due and those that NEXT would expect to receive, discounted at the original EIR. It is usually expressed as a percentage of the EAD. NEXT includes all cash collected over five years from the point of default. > > The Group uses probability weighted economic scenarios, in order to evaluate a range of possible outcomes as is required by IFRS 9, that are integrated into the model. The inputs and models used for the ECLs may not always capture all characteristics of the market at the Balance Sheet date. To reflect this, qualitative adjustments or overlays are made, based on external data, historical performance and future expected performance. > NEXT plc, p124
Accounting policy choice to apply the simplified approach is explained
Sales commitments are not financial instruments and, therefore, excluded from IFRS 9's requirements.
Key terms and policies are clearly defined
NEXT plc was also the only non-banking entity in our sample which applied the simplified approach and provided a detailed analysis of trade receivables by credit rating grade:
The following table contains an analysis of the credit risk exposure to customer receivables, using Experian's Consumer Indebtedness Index (a measure of customers' affordability) mapped to NEXT's internal credit grade.
| Credit grade | 2019 Total £m | 2018 Total £m |
|---|---|---|
| Very low risk | 612.7 | 551.5 |
| Low risk | 354.5 | 330.1 |
| Medium risk | 246.1 | 238.3 |
| High risk | 104.2 | 99.4 |
| Gross carrying amount before credit impaired | 1,317.5 | 1,219.3 |
| Credit impaired | 79.0 | 58.0 |
| Gross carrying amount after credit impaired | 1,396.5 | 1,277.3 |
| Loss allowance | ||
| Carrying amount |
Analysis of customer receivables and other trade receivables, stratified by credit grade, is as follows:
| Customer receivables and other trade receivables | Current £m | 1-30 days past due £m | 31-60 days past due £m | 61-90 days past due £m | 91-120 days past due £m | > 120 days past due £m | Total £m |
|---|---|---|---|---|---|---|---|
| Very low risk | 595.2 | 13.3 | 1.2 | 0.2 | 0.3 | 2.5 | 612.7 |
| Low risk | 334.3 | 12.6 | 1.8 | 0.6 | 0.2 | 5.0 | 354.5 |
| Medium risk | 219.0 | 13.4 | 4.0 | 1.6 | 0.9 | 7.2 | 246.1 |
| High risk | 69.5 | 9.0 | 5.2 | 3.7 | 3.8 | 13.0 | 104.2 |
| Otherwise impaired | 79.0 | 79.0 | |||||
| Total | 1,218.0 | 48.3 | 12.2 | 6.1 | 5.2 | 106.7 | 1,396.5 |
| Loss allowance | |||||||
|---|---|---|---|---|---|---|---|
| Very low risk | (3.3) | (0.2) | (0.1) | (1.2) | (4.8) | ||
| Low risk | (10.4) | (1.0) | (0.4) | (0.3) | (0.1) | (3.0) | (15.2) |
| Medium risk | (22.4) | (2.3) | (1.4) | (0.9) | (0.6) | (5.0) | (32.6) |
| High risk | (17.0) | (3.4) | (2.8) | (2.5) | (2.8) | (11.5) | (40.0) |
| Otherwise impaired | (73.4) | (73.4) | |||||
| Total | (53.1) | (6.9) | (4.7) | (3.7) | (3.5) | (94.1) | (166.0) |
| Expected loss rate % | |||||||
|---|---|---|---|---|---|---|---|
| Very low risk | 0.5% | 1.6% | 5.3% | 6.6% | 6.9% | 47.6% | 0.8% |
| Low risk | 3.1% | 7.7% | 21.2% | 53.5% | 70.1% | 61.2% | 4.3% |
| Medium risk | 10.2% | 17.3% | 33.7% | 55.4% | 64.1% | 69.4% | 13.2% |
| High risk | 24.5% | 38.0% | 55.9% | 67.9% | 73.3% | 87.8% | 38.4% |
| Otherwise impaired | 92.9% | 92.9% | |||||
| Total | 4.4% | 14.3% | 38.5% | 60.8% | 67.0% | 88.2% | 11.9% |
NEXT plc, p164-165
Hedging
Although entities may continue to use IAS 39's hedge accounting requirements, all but one non-banking entity adopted IFRS 9's. For those entities, existing hedges were carried over and the effect of IFRS 9 was adequately explained.
When IFRS 9's hedge accounting requirements are adopted, it is necessary to apply the requirements prospectively, including updating relevant hedge documentation.
Disclosures
The disclosure requirements in IFRS 7 have been enhanced on adoption of IFRS 9, requiring more detailed disclosures in respect of hedging arrangements. These additional disclosures apply even if an entity continues using IAS 39's hedging requirements.
Examples of good disclosure...
NEXT plc explained how it had updated its hedge documentation:
> Hedge documentation > > At the inception of a hedge relationship, the Group formally designates and documents the hedge relationship to which it wishes to apply hedge accounting and the risk management objective and strategy for undertaking the hedge. > > Before 28 January 2018 (i.e. under IAS 39 "Financial instruments: recognition and measurement"), the document included identification of the hedging instrument, the hedged item or transaction, the nature of the risk being hedged and how the Group will assess the effectiveness of changes in the hedging instrument's fair value in offsetting the exposure to changes in the hedged item's fair value or cash flows attributable to the hedged risk. Such hedges were expected to be highly effective in achieving offsetting changes in fair value or cash flows and were assessed on an ongoing basis to determine that they actually have been highly effective throughout the financial reporting periods for which they were designated. > > Beginning 28 January 2018 (i.e. under IFRS 9 "Financial instruments"), the documentation includes identification of the hedging instrument, the hedged item, the nature of the risk being hedged and how the Group will assess whether the hedging relationship meets the hedge effectiveness requirements (including the analysis of sources of hedge ineffectiveness and how the hedge ratio is determined). > > A hedging relationship qualifies for hedge accounting if it meets all of the following effectiveness requirements: > > * There is an "economic relationship" between the hedged item and the hedging instrument. > * The effect of the credit risk does not "dominate the value changes" that result from the economic relationship. > * The hedge ratio of the hedging relationship is the same as that resulting from the quantity of the hedged items that the Group actually hedges and the quantity of the hedging instrument that the Group actually uses to hedge that quantity of the hedged item. > NEXT plc, p125
We observed the following areas of good practice in respect of the new requirements:
- All of the banks included clear discussion of the types of hedging applied and explanation of how hedging is used as part of the risk management strategy.
- All provided adequate description of the sources of hedge ineffectiveness for each type of hedge.
- Clear presentation of the effect that hedge accounting has on the financial statements by risk category and type of hedge. IFRS 7 requires disclosure of quantification of the value of:
- the hedging instruments used by the entity;
- the hedged exposures covered by the entity's hedging strategy;
- the effectiveness of the hedging relationships; and
- the impact on the income statement and other comprehensive income.
We identified one bank which did not disclose the balances in the cash flow hedge reserve and foreign currency translation reserve for continuing hedges and the balances remaining in each reserve from hedging relationships for which hedge accounting is no longer applied.
Examples of good disclosure...
Cooperative Bank Holdings Limited provided a clear link between the risk management strategy and hedging activities:
> The Group is exposed to interest rate risk arising from changes in market interest rates. A variety of strategies are employed to mitigate interest rate risk with the overall objective of hedging interest rates paid and received, back to three-month LIBOR. To reduce the reporting volatility introduced as a result of entering into derivatives for economic hedging purposes, the Group applies hedge accounting. The hedge accounting strategies applied are as follows:
| Strategy | Hedging instruments and underlying hedged items | Objective of strategy |
|---|---|---|
| 1) Macro fair value hedge | Interest rate risk on fixed rate mortgages and customer loans. | Macro hedge accounting is used to recognise fair value changes related to changes in net interest rate risk in the fixed rate mortgages and customer loans and therefore reduce the profit or loss volatility that would otherwise arise from changes in the fair value of the interest rate swaps alone. |
| 2) Micro fair value hedge | Interest rate risk on certain fixed rate treasury assets and liabilities. | Micro fair value hedge accounting is used to recognise fair value changes related to changes in interest rate risk in certain treasury assets\liabilities and therefore reduce the profit or loss volatility that would otherwise arise from changes in the fair value of the interest rate swaps alone. |
| 3) Macro cash flow hedge | Reset risk on variable rate loans and mortgages. | Macro cash flow accounting is used to mitigate reporting volatility as a result of entering into interest rate swaps to economically hedge market risk on non-interest bearing deposits. The related interest rate swaps are designated in a cash flow hedge accounting relationship with variable rate loans and mortgages of similar maturity. |
> Provided the hedge is effective, changes in the fair value of the interest rate swaps are initially recognised in a hedging reserve in equity via the statement of other comprehensive income. They are transferred (recycled) to the income statement when the hedged transaction affects profit or loss. The ineffective portion of the change in the fair value of the hedging instrument (if any) is recognised directly in profit or loss. > Cooperative Bank Holdings Limited, p97
Examples of good disclosure...
The Royal Bank of Scotland plc clearly identified sources of ineffectiveness:
> 'The main sources of ineffectiveness for interest rate risk hedge accounting relationships are: > > * The effect of the counterparty credit risk on the fair value of the interest rate swap, which is not reflected in the fair value of the hedged item attributable to the change in interest rate (fair value hedge). > * Differences in the repricing basis between the hedging instrument and hedged cash flows (cash flow hedge); and > * Upfront present values on the hedging derivatives where hedge accounting relationships have been designated after the trade date (cash flow hedge and fair value hedge).' > The Royal Bank of Scotland plc, p205
Banks - Classification and measurement
Accounting policies
On the whole, the banks provided comprehensive policies detailing classification and measurement of financial instruments.
With the exception of one small bank, all the banks adequately described the classification categories. However, we continued to find use of boilerplate language taken from the standard in the discussion of the business model and SPPI tests. Whilst this may be appropriate for entities providing basic lending products, it may not be for the larger and more complex entities with potentially different business models for different portfolios.
Four banks designated financial liabilities at FVPL and recognised movements in fair value due to changes in own credit risk in OCI.
Only three of the banks designated equity instruments not held for trading as FVOCI. Two of the other banks included an accounting policy in respect of this designation but did not exercise the accounting policy choice.
We noted that the discussion of the treatment of gains and losses on financial instruments was better than we had previously observed in our review of interim reports.
Other disclosures
All of the banks provided clear presentation of gains and losses on the face of the income statement and statement of other comprehensive income.
Examples of good disclosure...
Standard Chartered plc explained the basis on which financial liabilities are designated at FVPL:
> 'Similarly, to reduce accounting mismatches, the Group has designated certain financial liabilities at fair value through profit or loss where the liabilities either: > > * Have fixed rates of interest and interest rate swaps or other interest rate derivatives have been entered with the intention of significantly reducing interest rate risk; or > * Are exposed to foreign currency risk and derivatives have been acquired with the intention of significantly reducing exposure to market changes; or > * Have been acquired to fund trading asset portfolios or assets. > > Financial liabilities may also be designated at fair value through profit or loss where they are managed on a fair value basis or have a embedded derivative where the Group is not able to bifurcate and separately value the embedded derivative component.' > Standard Chartered plc, p266
The main issues identified were:
- ! Where an entity designates a financial instrument at FVPL or equity instruments at FVOCI, we found that most banks listed the criteria from the standard but did not explain how the criteria had been met.
- ! One entity had significant financial liabilities designated at FVPL recognising changes in fair value attributable to changes in own credit risk through OCI. The accounting policy did not discuss the treatment of gains and losses attributable to changes in own credit risk.
- ! Use of boilerplate language which was generic and quoted directly from the standard.
- ! We identified one bank which included interest from loans at FVPL (therefore not determined on an effective interest rate basis) within interest income. An IFRIC agenda decision in March 2018 clarified that this was not permitted under IAS 1 'Presentation of financial statements'.
Banks – Impairment: policies and methodology
We found that the disclosures regarding the ECL policies and methodologies were generally good.
All of the banks defined the key terms underlying the ECL models and explained the basis on which ECL provisions were determined for major portfolios and product groups.
We observed the following areas of good practice:
- Two entities included information on the material post-model adjustments to reflect factors not included in the model.
- Three entities disclosed that the ECL models were based on regulatory models and provided details of the key adjustments made to regulatory models to determine ECLs. For example, the regulatory 'loss given default' is calculated on a more prudent basis.
- ECLs should be calculated over the maximum contractual period except for revolving credit facilities. Most entities with material credit card, overdraft and other revolving credit products, provided clear explanation of the basis on which the expected life of the facilities were determined.
Examples of good disclosure...
Standard Chartered plc clarified how the lifetime of revolving credit facilities was determined:
> 'Application of lifetime > Expected credit loss is estimated based on the shorter of the expected life and the maximum contractual period for which the Group is exposed to credit risk. For Retail Banking credit cards and Corporate & Institutional Banking overdraft facilities, however, the Group does not typically enforce the contractual period. As a result, for these instruments, the lifetime of the exposure is based on the period the Group is exposed to credit risk. This period has been determined by reference to expected behavioural life of the exposure and the extent to which credit risk management actions curtail the period of exposure. For credit cards, this has resulted in an average life of between 3 and 10 years across our footprint markets. Overdraft facilities have a 22-month lifetime.' > Standard Chartered plc, p137
We expect that, where overlays are used, they should be explained.
Examples of good disclosure...
Barclays plc explained key differences between the regulatory and IFRS 9 models:
> 'ECLs are calculated by multiplying three main components, being the PD, LGD and the EAD, discounted at the original EIR. The regulatory Basel Committee of Banking Supervisors (BCBS) ECL calculations are leveraged for IFRS 9 modelling but adjusted for key differences which include: > > * BCBS requires 12 month through the economic cycle losses whereas IFRS 9 requires 12 months or lifetime point in time losses based on conditions at the reporting date and multiple forecasts of the future economic conditions over the expected lives; > * IFRS 9 models do not include certain conservative BCBS model floors and downturn assessments and require discounting to the reporting date at the original EIR rather than using the cost of capital to the date of default; > * Management adjustments are made to modelled output to account for situations where known or expected risk factors and information have not been considered in the modelling process, for example forecast economic scenarios for uncertain political events; and > * ECL is measured at the individual financial instrument level, however a collective approach where financial instruments with similar risk characteristics are grouped together, with apportionment to individual financial instruments, is used where effects can only be seen at a collective level, for example for forward-looking information.' > Barclays plc, p275
Examples of good disclosure...
Barclays plc provided a helpful analysis of the material post-model adjustments:
> Management adjustments to models for impairment (audited) > > Management adjustments to impairment models are applied in order to factor in certain conditions or changes in policy that are not fully incorporated into the impairment models, or to reflect additional facts and circumstances at the period end. Management adjustments are reviewed and incorporated into future model development where applicable. > > Adjustments in portfolios that have total management adjustments to impairment allowance of greater than £10m are presented by product below. Information as at 31 December 2018 is prepared on an IFRS 9 basis and information as at 31 December 2017 is prepared on an IAS 39 basis. > > During 2018, models have continued to develop and a number of adjustments that were required on IFRS 9 adoption have been incorporated in impairment modelling.
Management adjustments to models for impairment (audited)
| As at 31 December | 2018 Management adjustments to impairment allowances, including forbearance £m | 2018 Proportion of total impairment allowances % | 2017 Management adjustments to impairment allowances, including forbearance £m | 2017 Proportion of total impairment allowances % |
|---|---|---|---|---|
| Home loans | 54 | 11.6 | 71 | 15.5 |
| Credit cards, unsecured loans and other retail lending | 370 | 6.9 | 80 | 2.6 |
| Corporate loans | (7) | (0.7) | 138 | 12.1 |
Note a Positive values relate to an increase in impairment allowance.
Home loans: Due to the high quality nature of the UK Home Loans portfolio, ECL estimates are low in all but the most severe scenarios. An adjustment is held to maintain an appropriate level of ECL.
Credit cards, unsecured loans and other retail lending: Model related adjustments to maintain adequacy of Loss Given Default estimates and retail staging criteria updates were applied during the year. This also includes a £100m ECL adjustment held in UK Cards for the anticipated impact of economic uncertainty in the UK.
Corporate loans: Includes a £50m ECL adjustment held in Corporate Bank for the anticipated economic uncertainty in the UK, offset by a release in the Investment Bank to reduce inappropriate ECL sensitivity to a macroeconomic variable. Barclays plc, p158
Banks - Impairment: staging and credit risk profile
We observed more comprehensive disclosures about the credit risk profile of significant portfolios and the allocation of assets to each of the three stages compared with that seen in our review of interim accounts. However, disclosure by some companies was much clearer and more helpful than others.
We found that most of the banks clearly explained the difference between the stages and the measurement of ECL at each stage.
Nearly all of the banks outlined both the qualitative and quantitative triggers used to assess whether there has been a significant increase in credit risk. The better disclosures provided details of the relative and absolute changes in probability of default ('PD') for key portfolios or product groups. None of the banks rebutted the 30 days past due backstop.
All but two of the banks discussed the link between forbearance and staging, including the application of cure periods to stage 2 and stage 3 balances.
All of the banks provided analyses of the gross and net loans to customers and the related ECLs by stage for significant portfolios. The level of detail and disaggregation was generally proportionate to the size of the entity. Two banks provided a clear analysis of stage 2 balances split between the qualitative and quantitative triggers of significant increases in credit risk.
Not all entities mapped internal to external credit risk ratings bands. The larger banks provided an analysis of loans and advances to customers by credit risk rating grade mapped to PD bands.
The main issues identified were:
- ! It was not always clear if an entity had applied the low credit risk expedient. Where it was applied, there was occasionally no explanation of how extensive it was.
- ! Two of the smaller banks referred to both qualitative and quantitative indicators of whether there has been a significant increase in credit risk but did not provide details of what these indicators were.
- ! Whilst most of the banks outlined the factors considered in determining whether a loan is in default, two banks stated that loans are in default if 90 days past due but did not elaborate if there were other factors which result in an asset being classified as in default.
- ! The analyses of the movement in ECL balances could be improved. Most banks did not disaggregate movements between stages to show how much moved in and out of each stage. The better disclosures also provided narrative which explained the movements.
Assessment of whether there has been a significant increase in credit risk is a key judgement and we would expect banks to provide details of all factors considered in making this assessment.
IFRS 7 requires disclosure of the gross carrying amount of financial assets and the exposure to credit risk on loan commitments and financial guarantee contracts by credit risk rating grades. Where ECLs are measured on a collective basis and cannot be individually allocated to a grade, these amounts should be separately disclosed. Given the sophisticated models used by banks, we would not expect the information provided to be based solely on days past due.
- ! Two entities provided an analysis of the exposure to credit risk by credit risk rating grades for loans and advances to banks and debt securities but not for loans and advances to customers, loan commitments or financial guarantee contracts. Whilst all the other banks provided this disclosure for debt securities and loans and advances, it was not always given for loan commitments and financial guarantee contracts.
Disaggregation of movements between the stages enables users to see if there is a deterioration or improvement in the loan portfolio.
Examples of good disclosure...
Skipton Building Society provided a clear analysis of movements in ECLs:
| Loss allowance as at 1 January | Group Stage 1 £m | Group Stage 2 £m | Group Stage 3 £m | Group Total £m | Society Stage 1 £m | Society Stage 2 £m | Society Stage 3 £m | Society Total £m |
|---|---|---|---|---|---|---|---|---|
| 2018 | 3.3 | 3.3 | 3.4 | 10.0 | 2.3 | 0.6 | 1.8 | 4.7 |
| Transfers due to changes in credit risk: | ||||||||
| From stage 1 to stage 2 | 0.9 | 0.9 | 0.4 | 0.4 | ||||
| From stage 1 to stage 3 | 0.5 | 0.5 | 0.2 | 0.2 | ||||
| From stage 2 to stage 1 | (0.4) | (0.4) | (0.1) | (0.1) | ||||
| From stage 2 to stage 3 | (0.2) | 0.6 | 0.4 | 0.2 | 0.2 | |||
| From stage 3 to stage 2 | 0.2 | (0.2) | 0.1 | (0.1) | ||||
| Remeasurements within existing stage | (0.6) | 0.4 | 0.1 | (0.1) | (0.7) | 0.1 | (0.6) | |
| Transfer of engagements | 0.1 | 0.1 | 0.1 | 0.1 | ||||
| Increases due to origination | 0.2 | 0.2 | 0.2 | 0.2 | ||||
| Decrease due to derecognition and repayments | (0.1) | (0.1) | (0.2) | (0.1) | (0.1) | (0.2) | ||
| Write-offs | (0.4) | (1.1) | (1.5) | (0.5) | (0.5) | |||
| Loss allowance as at 31 December | 2.8 | 3.8 | 3.3 | 9.9 | 1.7 | 1.1 | 1.6 | 4.4 |
Skipton Building Society, p189
Examples of good disclosure...
Barclays plc explained how cure periods apply to the stages and the link to forbearance:
> 'Both performing and non-performing forbearance assets are classified as Stage 3 except where it is established that the concession granted has not resulted in diminished financial obligation and that no other regulatory definition of default criteria has been triggered, in which case the asset is classified as Stage 2. The minimum probationary period for non-performing forbearance is 12 months and for performing forbearance, 24 months. Hence, a minimum of 36 months is required for non-performing forbearance to move out of a forborne state. > > No financial instrument in forbearance can transfer back to Stage 1 until all of the Stage 2 thresholds are no longer met and can only move out of Stage 3 when no longer credit impaired.' > Barclays plc, p128
Examples of good disclosure...
Cooperative Bank Holdings explained the use of the low credit risk expedient:
> 'The 'low credit risk exemption' available within IFRS 9 applies to certain highly rated treasury assets. Accordingly, at each balance sheet date, it is assumed that credit risk on all such financial assets has not increased significantly since initial recognition. The 'low credit risk exemption' has not been applied to any other type of financial asset.' > Cooperative Bank Holdings Limited, p128
Examples of good disclosure...
TSB Banking Group plc mapped each stage to internal credit risk gradings for all portfolios:
Internal rating scales In assessing the credit quality of the loan portfolio TSB uses an internal rating scale based on a customer's 12 month expected default probability.
| Credit quality | Internal grading |
|---|---|
| Strong | 1-4 |
| Satisfactory | 5-6 |
| Higher Risk | 7-9 |
| Credit Impaired | 10 |
| 11-13 |
Secured (retail)
| Credit quality | Stage 1 £ million | Stage 2 £ million | Stage 3 £ million | Credit impaired £ million | 2018 total £ million |
|---|---|---|---|---|---|
| Excellent quality | 25,200.2 | 13.1 | 27,201.4 | ||
| Good quality | 27.8 | 150.8 | 11.8 | 190.4 | |
| Satisfactory quality | 2.1 | 97.1 | 24.9 | 124.1 | |
| Lower quality | 0.3 | 8.3 | 6.7 | 15.3 | |
| Below standard | 1.3 | 6.9 | 230.9 | 190.0 | 429.1 |
| Gross carrying amount | 25,231.7 | 2,251.2 | 287.4 | 190.0 | 27,960.3 |
TSB Banking Group plc, p97
Examples of good disclosure...
Barclays plc analysed loan commitments and financial guarantees by credit risk rating grade:
Credit exposures by internal PD grade
Credit risk profile by internal PD grade for loans and advances at amortised cost (audited)
| Credit quality description | PD range | Stage 1 £m | Stage 2 £m | Stage 3 £m | Gross carrying amount Total £m | Allowance for ECL Stage 1 £m | Allowance for ECL Stage 2 £m | Allowance for ECL Stage 3 £m | Allowance for ECL Total £m | Net exposure £m | ECL coverage % |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Strong | 0.0 to <0.60% | 232,163 | 13,556 | 245,719 | 146 | 67 | 213 | 245,506 | 0.1 | ||
| Satisfactory | 0.60 to <11.35% | 48,730 | 24,768 | 73,498 | 508 | 1,517 | 2,025 | 71,473 | 2.8 | ||
| Higher Risk | 11.35 to <100% | 333 | 5,123 | 5,456 | 34 | 1,131 | 1,165 | 4,291 | 21.4 | ||
| Credit Impaired | 100% | 8,503 | 8,503 | 3,367 | 3,367 | 5,136 | 39.6 | ||||
| Total | 281,226 | 43,447 | 8,503 | 333,176 | 688 | 2,715 | 3,367 | 6,770 | 326,406 | 2.0 |
Credit risk profile by internal PD grade for contingent liabilities (audited)
| Credit quality description | PD range | Stage 1 £m | Stage 2 £m | Stage 3 £m | Gross carrying amount Total £m | Allowance for ECL Stage 1 £m | Allowance for ECL Stage 2 £m | Allowance for ECL Stage 3 £m | Allowance for ECL Total £m | Net exposure £m | ECL coverage % |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Strong | 0.0 to <0.60% | 15,000 | 443 | 15,443 | 6 | 3 | 9 | 15,434 | 0.1 | ||
| Satisfactory | 0.60 to <11.35% | 3,541 | 964 | 4,505 | 10 | 14 | 24 | 4,481 | 0.5 | ||
| Higher Risk | 11.35 to <100% | 49 | 228 | 277 | 10 | 10 | 267 | 3.6 | |||
| Credit Impaired | 100% | 74 | 74 | 2 | 2 | 72 | 2.7 | ||||
| Total | 18,590 | 1,635 | 74 | 20,299 | 16 | 27 | 2 | 45 | 20,254 | 0.2 |
Credit risk profile by internal PD grade for loan commitments (audited)
| Credit quality description | PD range | Stage 1 £m | Stage 2 £m | Stage 3 £m | Gross carrying amount Total £m | Allowance for ECL Stage 1 £m | Allowance for ECL Stage 2 £m | Allowance for ECL Stage 3 £m | Allowance for ECL Total £m | Net exposure £m | ECL coverage % |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Strong | 0.0 to <0.60% | 206,511 | 5,440 | 211,951 | 21 | 5 | 26 | 211,925 | |||
| Satisfactory | 0.60 to <11.35% | 84,141 | 11,806 | 95,947 | 59 | 80 | 139 | 95,808 | 0.1 | ||
| Higher Risk | 11.35 to <100% | 747 | 3,245 | 3,992 | 3 | 38 | 41 | 3,951 | 1.0 | ||
| Credit Impaired | 100% | 610 | 610 | 20 | 20 | 590 | 3.3 | ||||
| Total | 291,399 | 20,491 | 610 | 312,500 | 83 | 123 | 20 | 226 | 312,274 | 0.1 |
Barclays plc, p169
Examples of good disclosure...
Nationwide Building Society provided a helpful analysis of stage 2:
Reason for residential mortgages being included in stage 2 4 April 2019
| Prime £m | Prime % | Specialist £m | Specialist % | Total £m | Total % | |
|---|---|---|---|---|---|---|
| Quantitative criteria: | ||||||
| Payment status (greater than 30 DPD) (note i) | 267 | 13 | 213 | 3 | 480 | 6 |
| Increase in PD since origination (less than 30 DPD) | 1,613 | 79 | 2,186 | 34 | 3,799 | 45 |
| Qualitative criteria: | ||||||
| Forbearance (less than 30 DPD) | 148 | 7 | 7 | 155 | 2 | |
| Interest only-significant risk of inability to refinance at maturity (less than 30 DPD) | 4,018 | 63 | 4,018 | 47 | ||
| Other qualitative criteria | 20 | 1 | 7 | 27 | ||
| Total Stage 2 gross balances | 2,048 | 100 | 6,431 | 100 | 8,479 | 100 |
Nationwide Building Society, p115
Banks – Impairment: multiple economic scenarios
All of the banks explained the methodology used to determine the impact of alternative economic scenarios, including disclosure of the number of scenarios and the weightings applied to each scenario. However, there is still scope for improvement, especially in respect of disclosure of the assumptions underpinning the scenarios.
IFRS 7 requires disclosure of the key inputs into the calculation of ECLs. Whilst we believe it is useful for users to understand the key underlying assumptions used for the discrete scenarios, the basis on which these assumptions were presented varied and was sometimes confusing. One entity presented the assumptions for each scenario on a different basis using average rates for the central scenario, cumulative growth for the upside scenarios and peak-to-trough fall for the downside.
Some of the smaller banks listed, but did not quantify, the key economic variables considered when developing the forecasts.
The more helpful disclosures allowed users to see how the assumptions change through the projection period and how they compared year on year.
Multiple economic scenarios are used to model non-linearities and are not the same as stress testing. The disclosures regarding the application of multiple economic scenarios and the methodology were generally good. Two banks applied a Monte Carlo approach to all or some of the loan portfolios, all other entities calculated a number of discrete scenarios.
We identified the following:
- Four entities quantified the impact on the ECL provision as a result of the alternative scenarios.
- Most of the banks disclosed the significant assumptions underlying the central scenario. Nearly all of these banks, quantified the assumptions underlying the determination of each scenario, the larger banks provided these for key territories or portfolio.
- Four entities also noted when the assumptions were expected to revert to long-term averages.
- Two entities explained significant changes in the forecast assumptions relative to those applied at the start of the year.
- Where the Monte Carlo approach was used, it was generally well explained.
Examples of good disclosure...
Nationwide included narrative which discussed the assumptions through the projection period:
> '....In this severe downside scenario, real GDP growth over a five year period is slightly negative. In the first two years unemployment rises sharply by 4.8%, and house prices fall by 33% from peak to trough, before gradual recovery from year 3 onwards. Due to the way in which the additional provision has been calculated, the results of this scenario have not been used in determining the reported stage allocation of loans, although in this scenario an increased proportion of loans are assumed to migrate to stage 2 and stage 3 over the projection period.' > Nationwide Building Society, p201
All of the banks using a discrete scenario approach disclosed the number of scenarios, the nature of the scenario (for example, severe stress scenario reflecting the impact of Brexit) and the weightings applied. All used between three and five scenarios.
Examples of good disclosure...
Metro Bank plc presented the economic variable assumptions for the initial four year projection period as at 31 December 2018 and 1 January 2018:
Economic variable assumptions The period-end assumptions used for the ECL estimate as at 31 December 2018 are as follows:
| 2019 Base: | 2020 Base: | 2021 Base: | 2022 Base: | |
|---|---|---|---|---|
| Interest rates (%) | 2.2% | 2.6% | 2.8% | 3.2% |
| Upside: 2.1% | Upside: 3.1% | Upside: 3.1% | Upside: 3.5% | |
| Downside: 0.9% | Downside: 1.2% | Downside: 1.4% | Downside: 1.6% | |
| Brexit:0.5% | Brexit: 0.8% | Brexit: 0.9% | Brexit: 1.3% | |
| UK unemployment (%) | 4.6% | 4.8% | 5.0% | 5.0% |
| Upside: 3.3% | Upside: 3.4% | Upside: 3.6% | Upside: 3.0% | |
| Downside: 6.2% | Downside: 7.2% | Downside: 7.3% | Downside: 6.9% | |
| Brexit: 6.7% | Brexit: 8.4% | Brexit: 8.5% | Brexit: 8.1% | |
| UK house price index | 1.9% | 0.5% | 1.2% | 1.9% |
| - % change year-on-year | Upside: 7.6% | Upside: 4.5% | Upside: 1.9% | Upside: 0.9% |
| Downside: (5.3)% | Downside: (6.4)% | Downside: 0.0% | Downside: (3.7)% | |
| Brexit: (8.5)% | Brexit: (11.1)% | Brexit: (1.7)% | Brexit: (4.3)% | |
| UK GDP - % change | 1.6% | 1.4% | 1.9% | 1.8% |
| year-on-year | Upside: 4.0% | Upside: 2.1% | Upside: 1.9% | Upside: 1.6% |
| Downside: (1.9)% | Downside: 0.8% | Downside: 2.6% | Downside: 2.0% | |
| Brexit: (3.6)% | Brexit: (0.2)% | Brexit: 2.6% | Brexit: 2.3% |
The assumptions used for the ECL estimate as at 1 January 2018 are as follows:
| 2018 Base: | 2019 Base: | 2020 Base: | 2021 Base: | 2022 Base: | |
|---|---|---|---|---|---|
| Interest rates (%) | 1.7% | 2.3% | 2.7% | 2.6% | 3.0% |
| Upside: 1.8% | Upside: 2.6% | Upside: 2.9% | Upside: 3.0% | Upside: 3.3% | |
| Downside: 1.5% | Downside: 1.0% | Downside: 1.0% | Downside: 1.3% | Downside: 1.8% | |
| Brexit: n/a | Brexit: n/a | Brexit: n/a | Brexit: n/a | Brexit: n/a | |
| UK unemployment (%) | 4.6% | 4.8% | 5.0% | 5.1% | 5.1% |
| Upside: 4.0% | Upside: 3.5% | Upside: 3.6% | Upside: 3.9% | Upside: 4.1% | |
| Downside: 5.7% | Downside: 7.1% | Downside: 7.5% | Downside: 7.3% | Downside: 6.9% | |
| Brexit: n/a | Brexit: n/a | Brexit: n/a | Brexit: n/a | Brexit: n/a | |
| UK house price index | 2.9% | 1.3% | 0.9% | 1.8% | 2.4% |
| % change year-on-year | Upside: 5.8% | Upside: 7.2% | Upside: 3.2% | Upside: 1.6% | Upside: 1.0% |
| Downside: (0.9)% | Downside: (7.3)% | Downside: (2.7)% | Downside: 1.8% | Downside: 4.3% | |
| Brexit: n/a | Brexit: n/a | Brexit: n/a | Brexit: n/a | Brexit: n/a | |
| UK GDP - % change | 1.6% | 1.6% | 1.8% | 1.9% | 1.8% |
| year-on-year | Upside: 3.4% | Upside: 3.2% | Upside: 2.1% | Upside: 1.7% | Upside: 1.6% |
| Downside: (1.1)% | Downside: (0.8)% | Downside: 1.9% | Downside: 2.5% | Downside: 2.0% | |
| Brexit: n/a | Brexit: n/a | Brexit: n/a | Brexit: n/a | Brexit: n/a |
Examples of good disclosure...
Standard Chartered plc explained the use of a Monte Carlo approach:
> 'To take account of the potential non-linearity in expected credit loss, the Group simulates a set of 50 scenarios around the Base Forecast and calculates the expected credit loss under each of them. These scenarios are generated by a Monte Carlo simulation, which considers the degree of uncertainty (or volatility) around economic outcomes and how these outcomes have tended to move in relation to one another (or correlation). The use of Monte Carlo simulation is motivated by the number and spread of countries in which the Group operates. This implies that the number of countries' macroeconomic variables to forecast is large, but more importantly the observation that a downturn in one part of the world is never perfectly synchronised with downturns everywhere else means that the Group may be challenged to capture a full range of scenarios with a handful of manually tuned scenarios. > Standard Chartered plc, p176
Metro Bank plc, p139
Banks – Impairment: estimation uncertainty
We were pleased to note that most of the banks made a reasonable attempt to show the sensitivity of ECLs to changes in economic variables. However, there is still scope for improvement.
Examples of good disclosure...
Barclays plc included the impact of a univariate sensitivity:
> 'Staging sensitivity (audited) > An increase of 1% (£3,332m) of total gross exposure into Stage 2 (from Stage 1), would result in an increase in ECL impairment allowance of £200m based on applying the difference in Stage 2 and Stage 1 average impairment coverage ratios to the movement in gross exposure' > Barclays plc, p161
Examples of good disclosure...
Standard Chartered plc disclosed the sensitivity of ECLs to a multivariate change in assumptions:
> 'As the Group has two principal uncertainties related to the macroeconomic outlook, a sensitivity analysis of ECL was undertaken to explore the combined effect of these: extended trade tensions that could lead to a China slowdown with spillovers to emerging markets. In this scenario, current trade policy tensions between the US and China increase dramatically.... > > Applying this scenario, modelled stage 1 and 2 expected credit loss provisions would be approximately $362 million higher than the reported base case expected credit loss provision (excluding the impact of non-linearity).' > Standard Chartered plc, p177
All of the banks identified the determination of ECLs as a key source of estimation uncertainty.
The most common sensitivity performed by the banks was the application of a 100 per cent weighting to the alternative economic scenarios. The manner in which the banks performed this sensitivity was not consistent across the sample with some disclosing the impact for material portfolios whereas others disclosed the impact on the total ECL.
It was also not always clear if the sensitivity included movements between stages and most of the banks excluded stage 3 from the analysis. We believe, banks should not assume that sensitivities will not affect stage 3 provisions.
We observed the following areas of good practice:
- Explanation of the reason for the choice in the sensitivities used.
- The better disclosures had considered both multivariate and univariate sensitivities.
In addition to the application of the 100 per cent weighting, a number of banks disclosed the impact on ECLs of a change in a single variable, such as HPI, collateral values or a shift of balances from stage 1 to stage 2.
Whilst the use of multivariate sensitivities was generally well done, univariate analysis should not be overlooked, particularly where there are individual assumptions which are more dominant than others.
In some instances, the most severe downside scenario was selected to demonstrate the sensitivity of the ECL balances.
Whilst this is helpful disclosure, we would remind preparers that IAS 1 requires consideration of a reasonably possible change in assumptions which might not necessarily include the most extreme change.
Examples of good disclosure...
- Basis of preparation and significant accounting policies continued The weightings applied to each scenario are considered to represent a significant accounting estimate. We have performed an assessment of the impact on the ECL if each of the Baseline, Upside, Downside and Hard Brexit scenarios were applied to the ECL calculation using a 100% weighting (that is, ignoring all other scenarios in each case):
| Scenario | ECL (£million) | Variance to reported weighted ECL at 31 December 2018 |
|---|---|---|
| Weighted | 33.8 | |
| Baseline | 31.6 | (7%) |
| Upside | 26.5 | (22%) |
| Downside | 40.4 | 19% |
| Hard Brexit | 49.4 | 46% |
We note that the sensitivities disclosed above represent example scenarios and may not represent actual scenarios which occur in the future. If one of these scenarios did arise then at that time the ECL would not equal the amount disclosed above, as the amounts disclosed do not take account of the alternative possible scenarios which would be considered at that time. We also note that the sensitivities disclosed above do not take into account movements in impairment stage allocations that would result under the different scenarios. Metro Bank plc, p117
Examples of good disclosure...
To give an indication of the sensitivity of ECLs to different economic scenarios, the table below shows the ECL if 100% weighting is applied to the upside, central and downside scenarios.
Sensitivity analysis impact of multiple economic scenarios
| Upside scenario ECL £m | Central scenario ECL £m | Downside scenario ECL £m | |
|---|---|---|---|
| 4 April 2019 | |||
| Residential mortgages | 99 | 112 | 242 |
| Consumer banking | 381 | 383 | 400 |
| Commercial and other lending | 37 | 37 | 37 |
| Total | 517 | 532 | 679 |
| 5 April 2018 | |||
| Residential mortgages | 128 | 143 | 279 |
| Consumer banking | 350 | 352 | 374 |
| Commercial and other lending | 24 | 24 | 24 |
| Total | 502 | 519 | 677 |
The ECL for each scenario multiplied by the scenario probability will not reconcile to the final probability weighted ECL, since the stage allocation of loans varies in each scenario. In the probability weighted ECL, each loan is allocated to a discrete stage based on the weighted average PD under the economic scenarios. The impact of the severe downside scenario on impairment provisions is explained above. Nationwide Building Society, p202
Banks - Hedging
IFRS 7 hedging disclosure requirements were expanded significantly as a result of IFRS 9. All of the banking entities which apply hedge accounting, with the exception of one, continued to apply IAS 39 for hedge accounting. Disclosures by the banks were generally good in this area.
Not all of the banks used hedge accounting. Of those that did, all expanded their disclosures in order to meet the new requirements of IFRS 7.
We observed the following areas of good practice in respect of the new requirements:
- All of the banks included clear discussion of the types of hedging applied and explanation of how hedging is used as part of the risk management strategy.
- All provided adequate description of the sources of hedge ineffectiveness for each type of hedge.
- Clear presentation of the effect that hedge accounting has on the financial statements by risk category and type of hedge. IFRS 7 requires disclosure of quantification of the value of:
- the hedging instruments used by the entity;
- the hedged exposures covered by the entity's hedging strategy;
- the effectiveness of the hedging relationships; and
- the impact on the income statement and other comprehensive income.
We identified one bank which did not disclose the balances in the cash flow hedge reserve and foreign currency translation reserve for continuing hedges and the balances remaining in each reserve from hedging relationships for which hedge accounting is no longer applied.
Examples of good disclosure...
Cooperative Bank Holdings Limited provided a clear link between the risk management strategy and hedging activities:
> The Group is exposed to interest rate risk arising from changes in market interest rates. A variety of strategies are employed to mitigate interest rate risk with the overall objective of hedging interest rates paid and received, back to three-month LIBOR. To reduce the reporting volatility introduced as a result of entering into derivatives for economic hedging purposes, the Group applies hedge accounting. The hedge accounting strategies applied are as follows:
| Strategy | Hedging instruments and underlying hedged items | Objective of strategy |
|---|---|---|
| 1) Macro fair value hedge | Interest rate risk on fixed rate mortgages and customer loans. | Macro hedge accounting is used to recognise fair value changes related to changes in net interest rate risk in the fixed rate mortgages and customer loans and therefore reduce the profit or loss volatility that would otherwise arise from changes in the fair value of the interest rate swaps alone. |
| 2) Micro fair value hedge | Interest rate risk on certain fixed rate treasury assets and liabilities. | Micro fair value hedge accounting is used to recognise fair value changes related to changes in interest rate risk in certain treasury assets\liabilities and therefore reduce the profit or loss volatility that would otherwise arise from changes in the fair value of the interest rate swaps alone. |
| 3) Macro cash flow hedge | Reset risk on variable rate loans and mortgages. | Macro cash flow accounting is used to mitigate reporting volatility as a result of entering into interest rate swaps to economically hedge market risk on non-interest bearing deposits. The related interest rate swaps are designated in a cash flow hedge accounting relationship with variable rate loans and mortgages of similar maturity. |
> Provided the hedge is effective, changes in the fair value of the interest rate swaps are initially recognised in a hedging reserve in equity via the statement of other comprehensive income. They are transferred (recycled) to the income statement when the hedged transaction affects profit or loss. The ineffective portion of the change in the fair value of the hedging instrument (if any) is recognised directly in profit or loss. > Cooperative Bank Holdings Limited, p97
Examples of good disclosure...
The Royal Bank of Scotland plc clearly identified sources of ineffectiveness:
> 'The main sources of ineffectiveness for interest rate risk hedge accounting relationships are: > > * The effect of the counterparty credit risk on the fair value of the interest rate swap, which is not reflected in the fair value of the hedged item attributable to the change in interest rate (fair value hedge). > * Differences in the repricing basis between the hedging instrument and hedged cash flows (cash flow hedge); and > * Upfront present values on the hedging derivatives where hedge accounting relationships have been designated after the trade date (cash flow hedge and fair value hedge).' > The Royal Bank of Scotland plc, p205
Examples of good disclosure...
Nationwide Building Society clearly presented the effects of hedging on the financial statements in a tabular format for each type of hedge:
Fair value hedge accounting 2019
| Hedged item balance sheet classification | Hedging instrument | Risk category | Hedged item £m | Instrument £m | Hedge ineffectiveness recognised in the income statement £m | Carrying amount of the hedged item £m | Of which: accumulated fair value adjustment (note i) £m |
|---|---|---|---|---|---|---|---|
| Assets: | |||||||
| Loans and advances to customers (note ii) | Interest rate swaps | Interest rate | 396 | (406) | (10) | 94,635 | 1,294 |
| Investment securities | Interest rate swaps, bond forwards | Interest rate | 230 | (193) | 37 | 13,292 | 323 |
| Investment securities | Inflation swaps | Interest rate and inflation | 16 | (15) | 1 | 1,439 | 19 |
| Total assets | 642 | (614) | 28 | 109,366 | 1,636 | ||
| Liabilities: | |||||||
| Shares (note iii) | Interest rate swaps | Interest rate | (37) | 38 | 1 | 4,131 | (17) |
| Debt securities in issue | Interest rate swaps | Interest rate | (31) | 28 | (3) | 4,662 | 642 |
| Subordinated liabilities | Interest rate swaps | Interest rate | (3) | 4 | 1 | 2,407 | 37 |
| Subscribed capital | Interest rate swaps | Interest rate | (3) | (3) | (0) | 240 | 40 |
| Total liabilities | (71) | 67 | (4) | 11,440 | 702 | ||
| Total fair value hedges | 571 | (547) | 24 |
Cash flow hedge accounting 2019
| Hedged item balance sheet classification | Hedging instrument | Risk category | Hedged item £m | Instrument £m | Changes in instrument fair value reported as: Foreign exchange retranslation recognised in the income statement £m | Changes in instrument fair value reported as: Net amounts deferred to other comprehensive income £m | Amounts accumulated in the cash flow hedge reserve (excluding deferred taxation): Continuing hedges £m | Amounts accumulated in the cash flow hedge reserve (excluding deferred taxation): Discontinued hedges £m |
|---|---|---|---|---|---|---|---|---|
| Assets: | ||||||||
| Loans and advances to customers | Interest rate swaps | Interest rate | (2) | 2 | 2 | 2 | ||
| Total assets | (2) | 2 | 2 | 2 | ||||
| Liabilities: | ||||||||
| Debt securities in issue | Inflation swaps | Interest rate and inflation | (10) | 10 | 10 | 6 | ||
| Debt securities in issue | Interest rate swaps, cross currency interest rate swaps | Interest rate and foreign exchange | (49) | 85 | 10 | (190) | 357 | 19 |
| Subordinated liabilities | Interest rate swaps, cross currency interest rate swaps | Interest rate and foreign exchange | (308) | 335 | 13 | 159 | 44 | |
| Total liabilities | (367) | 430 | 23 | (31) | 407 | 19 | ||
| Total cash flow hedges | (369) | 432 | 23 | (31) | 409 | 19 |
Nationwide Building Society plc, p196-197
Next steps
This report sets out the findings from our thematic review of the implementation of IFRS 9. Overall, we were very pleased to see the level of improvements made by most of the entities in our sample. Where we did identify specific areas for improvement, we have engaged directly with the respective entity.
We will continue to challenge companies during our routine reviews when we do not see:
- Entity specific classification policies with linkage to the business model for more complex portfolios. For non-banking entities, discussion of a hold-to-collect model, where relevant.
- Explanation of policy choices where assets and liabilities are designated at FVOCI or FVPL, in particular, how the criteria for designation has been met.
- Clear explanation of the judgements applied, particularly to determine if there has been a significant increase in credit risk.
- Analysis of the gross carrying amount of financial assets by credit risk ratings grade, days past due or in a provision matrix.
- Application of the ECL requirements to material contract balances or lease receivables.
- Discussion of the basis on which forward looking information has been factored into the calculation of ECLs.
- Sensitivity analysis where ECLs are identified as a source of estimation uncertainty.
Information about the Financial Reporting Council can be found at: https://www.frc.org.uk
Follow us on Twitter @FRCnews or Linked
The FRC's mission is to promote transparency and integrity in business. The FRC sets the UK Corporate Governance and Stewardship Codes and UK standards for accounting and actuarial work; monitors and takes action to promote the quality of corporate reporting; and operates independent enforcement arrangements for accountants and actuaries. As the Competent Authority for audit in the UK the FRC sets auditing and ethical standards and monitors and enforces audit quality.
The FRC does not accept any liability to any party for any loss, damage or costs howsoever arising, whether directly or indirectly, whether in contract, tort or otherwise from any action or decision taken (or not taken) as a result of any person relying on or otherwise using this document or arising from any omission from it.
The Financial Reporting Council Limited 2019 The Financial Reporting Council Limited is a company limited by guarantee. Registered in England number 2486368. Registered Office: 8th Floor, 125 London Wall, London EC2Y 5AS
-
Reference to banks and banking entities throughout this report include the building societies selected for review. ↩