Methodology for measuring expected credit losses (ECL) according to IFRS 9
Sparebanken Møre has developed an ECL model based on the Group’s IRB parameters and applies a three-stage approach when assessing ECL on loans to customers and financial guarantees in accordance with IFRS 9.
Stage 1: At initial recognition and if there’s no significant increase in credit risk, the commitment is classified in stage 1 with 12-months ECL.
Stage 2: If a significant increase in credit risk since initial recognition is identified, but without evidence of loss, the commitment is transferred to stage 2 with lifetime ECL measurement.
Stage 3: If the credit risk increases further, including evidence of loss, the commitment is transferred to stage 3 with lifetime ECL measurement. The commitment is considered to be credit-impaired. As opposed to stage 1 and 2, effective interest rate in stage 3 is calculated on net impaired commitment (total commitment less expected credit loss) instead of gross commitment.
Staging is performed at account level and implies that two or more accounts held by the same customer can be placed in different stages. If a customer has one account in stage 3 (risk classes K, M or N), all of the customer’s accounts will migrate to stage 3.
Customers in risk class N have been subject to individual loss assessment with impairment. In connection with individual loss assessment, 3 scenarios based on calculation of the weighted present value of future cash flow after realisation of collateral are prepared. If the weighted present value of cash flow after realisation of collateral is positive, model-based loss provisions according to the ECL model is used.
An increase in credit risk reflects both customer-specific circumstances and development in relevant macro factors for the particular customer segment. The assessment of what is considered to be a significant increase in credit risk is based on a combination of quantitative and qualitative indicators, as well as “backstops” (see separate section regarding “backstops”).
Quantitative criteria
A significant increase in credit risk is determined by comparing the PD at the reporting date with PD at initial recognition. If the actual PD is higher than initial PD, an assessment is made of whether the increase is significant.
Significant increase in credit risk since initial recognition is considered to have occurred when either
- PD has increased by 100 per cent or more and the increase in PD is more than 0.5 percentage points (if initial PD <1 %), or
- PD has increased by 100 % or more or the increase in PD is higher than 2 percentage points (if initial PD was >/= 1 %)
The weighted, macro adjusted PD in year 1 is used for comparison with PD on initial recognition to determine whether the credit risk has increased significantly.
Qualitative criteria
In addition to the quantitative assessment of a changes in the PD, a qualitative assessment is made to determine whether there has been a significant increase in credit risk, for example, if the commitment is subject to special monitoring.
“Backstops”
Credit risk is always considered to have increased significantly if the following events, “backstops”, have occurred:
- the customer’s contractual payments are 30 days past due
- the customer has been granted forbearance measures due to financial distress, though it is not severe enough to be individually assessed in stage 3.
Significant reduction in credit risk – recovery
A customer migrates from stage 2 to stage 1 if:
- The criteria for migration from stage 1 to stage 2 is no longer present, and
- This is satisfied for at least one subsequent month (total 2 months)
A customer migrates from stage 3 to stage 1 or stage 2 if the customer no longer meets the conditions for migration to stage 3:
- The customer migrates to stage 2 if more than 30 days in default.
- Otherwise, the customer migrates to stage 1.
Accounts that are not subject to the migration rules above are not expected to have significant change in credit risk and retain the stage from the previous month.
Scenarios
Three scenarios are developed: Best, Basis and Worst. For each of the scenarios, expected values of different parameters are given, for each of the next five years. The possibility for each of the scenarios to occur is also estimated. After five years, the scenarios are expected to converge to a long-term stable level.
Changes to PD as a result of scenarios, may also affect the staging.
Definition of default, credit-impaired and forbearance
The definition of default has been amended from 1 January 2021 and has been extended to include breaches of special covenants and agreed payment reliefs (forbearance). The new default definition has not changed the Group’s assessment of credit risk associated with individual exposures, and there is therefore no significant effect on the Group’s losses.
A commitment is defined to be in default and credit-impaired (non-performing) if a claim is more than 90 days overdue and the overdue amount exceeds the highest of 1 per cent of the exposure (loans and undrawn credits) and NOK 1,000 for the retail market and NOK 2,000 for the corporate market. Breaches of covenants can also trigger default.
A commitment is also defined to be credit-impaired (non-performing) if the commitment, as a result of a weakening of the debtor's creditworthiness, has been subject to an individual assessment, resulting in a lifetime ECL in stage 3.
A commitment is defined to be subject to forbearance (payment relief due to payment difficulties) if the bank agrees to changes in the terms and conditions as a result of the debtor having problems meeting payment obligations. Performing forbearance (not in default) is placed in stage 2 whereas non-performing (defaulted) forbearance is placed in stage 3.
As part of the process of granting payment relief, a specific, individual assessment is made of whether the application for payment relief is ‘forbearance’ and whether the loan should thus migrate to stage 2 (performing) or stage 3 (non-performing) in the Group’s ECL model.
Management override
Quarterly review meetings evaluate the basis for the accounting of ECL losses. If there are significant events that will affect an estimated loss which the model has not taken into account, relevant factors in the ECL model will be overridden. An assessment is made of the level of long-term PD and LGD in stage 2 and stage 3 under different scenarios, as well as an assessment of macro factors and weighting of scenarios.
Consequences of increased macroeconomic uncertainty and measurement of expected credit loss (ECL) for loans and guarantees
Pursuant to the accounting rules (IAS 34), interim financial reports must provide an explanation of events and transactions that are significant to an understanding of the changes in financial position and performance of an entity since the last annual report. The information related to these events and transactions must take into account relevant information presented in the most recent annual report.
The bank’s loss provisions reflect expected credit loss (ECL) pursuant to IFRS 9. When assessing ECL, the relevant conditions at the time of reporting and expected economic developments are taken into account.
Price inflation has risen rapidly through 2022 and in the first quarter of 2023 and has been significantly higher than estimated by Norges Bank. Inflation is clearly above Norges Bank’s target, and it is anticipated that it will remain high for longer than previously estimated. The job market is tight, but there are clear indications of a turnaround in the Norwegian economy. Less pressure in the economy will contribute to curbing price inflation. Capacity problems in production as a result of the reopening of the economy in combination with increased energy prices and raw material prices have led to rising inflation. Increased uncertainty about economic development and interest rate hikes have led to a sharp rise in market interest rates internationally.
There are prospects of lower commercial property prices, but there may be large geographical variations. While the required rate of return for some commercial properties in Oslo has been at a record low level, the required rate of return on properties in Møre og Romsdal has not changed appreciably. Sparebanken Møre has not changed the lower required rate of return on commercial property in its credit policy during the period of record low interest rates. This has contributed to a relatively solid equity ratio for commercial properties.
Projections for rental price inflation and required rate of return are expected to result in a fall in selling prices on commercial property in the years ahead.
Low required rates of return make commercial property prices particularly vulnerable to higher interest rates or risk premiums. An abrupt increase in the required rate of return may lead to a marked fall in selling prices. Many commercial real estate companies have high debt-to-income ratios, and higher interest rates will lead to a larger portion of the income being spent on servicing debt.
In the Group’s calculations of expected credit loss (ECL), the macroeconomic scenarios and the weightings have been impacted by the changes in economic conditions in 2022. The probability of a pessimistic scenario is increased from 10 per cent to 20 per cent, the base case scenario is 70 per cent and the best case scenario is reduced from 20 per cent to 10 per cent.
The model-based provisions have increased in the quarter, which is attributed to increased uncertainty in the retail market due to increased energy prices, interest expenses and general price increases in society. Overall, this will increase household expenses, reduce purchasing power and potentially increase default somewhat in the future. Overall, the level of model-based provisions is assessed as robust.
So far, no significant increase in arrears and forbearance has been observed as a result of increased interest expenses and higher inflation. In the 1st quarter of 2023, there has been an increase in applications for payment holidays and reduced term payments.