A significant increase in credit risk (SICR) is the test in IFRS 9 that determines how much loss allowance a lender must hold against a loan or other financial asset. If, at a reporting date, the risk that the borrower will default has increased significantly since the asset was first recognised, the allowance rises from twelve months of expected credit losses to losses expected over the asset's whole remaining life. Banks and other entities reporting under international accounting standards apply the test to loans, debt securities, lease receivables and loan commitments.
The test is the dividing line within the standard's three-stage impairment model, described under staging under IFRS 9. An asset that has not had a significant increase is a Stage 1 credit exposure; one that has, but is not yet credit-impaired, is a Stage 2 credit exposure. The assessment drives the size and timing of provisions for expected credit loss.
What the standard requires
The comparison is relative, not absolute. The entity compares the risk of a default occurring over the expected life of the asset as at the reporting date with the same risk as at initial recognition. A loan made to a weak borrower and priced accordingly does not start in Stage 2 merely because it is risky; it moves only if its risk has become significantly worse than was expected at the outset. Equally, the test looks at the likelihood of default, not at the amount that would be lost, so good collateral does not prevent a transfer.
The assessment must use reasonable and supportable information that is available without undue cost or effort, including forward-looking information such as economic forecasts. The standard does not define "significant"; each entity sets its own criteria and discloses them.
Indicators used in practice
Lenders typically combine three kinds of trigger:
- Quantitative. A comparison of the lifetime probability of default at the reporting date with the value expected at origination, with a transfer when it has risen by more than a set multiple or margin, or when an internal rating has fallen by a set number of grades.
- Qualitative. Events not yet captured in the models, such as placement on a watch list, the grant of forbearance, a covenant breach, or adverse changes in the borrower's sector or region.
- Backstop. A rebuttable presumption in the standard that credit risk has increased significantly when contractual payments are more than 30 days past due.
Information for these triggers comes from routine credit exposure monitoring and loan covenant monitoring. For retail portfolios, where individual information is limited, the assessment is often made on groups of loans with shared characteristics.
Simplifications and exceptions
IFRS 9 offers a low credit risk exemption: an entity may assume there has been no significant increase if the asset has low credit risk at the reporting date, commonly interpreted as equivalent to an investment-grade credit rating. This is used mainly for holdings of high-quality bonds; banking supervisors have discouraged its use for lending.
Trade receivables and contract assets without a significant financing component always carry a lifetime allowance under a simplified approach, so no test is needed. Assets that were already credit-impaired when purchased or originated are also outside the staging test.
Effect on provisions
Moving from a twelve-month to a lifetime allowance can multiply the provision several times, especially for long-dated loans such as mortgages, and the increase is charged to profit or loss in the period of transfer. This step change is often called the cliff effect. The transfer is reversible: if credit risk improves so that the increase is no longer significant, the asset returns to Stage 1, although many lenders require a probation period first. Further deterioration to the point of default or other evidence of impairment moves the asset to a Stage 3 credit exposure, a category that overlaps heavily with the regulatory non-performing exposure.
Challenges and supervisory attention
Because the threshold is set by each lender, similar loans can be staged differently at different banks, which reduces comparability. Supervisors, including the European Central Bank and the European Banking Authority, have reviewed banks' practices and criticised thresholds that are too loose, heavy reliance on the 30-day backstop, and slow recognition of sector-wide stress. During the COVID-19 pandemic, regulators advised that payment holidays offered to all borrowers should not automatically be treated as a significant increase, which tested the judgement the standard requires.
A further criticism is procyclicality: in a downturn, forecasts worsen and many loans transfer at once, raising provisions when banks are already under strain. The equivalent US standard avoids the issue by requiring lifetime losses on all loans from the start. The wider accounting context is covered under impairment accounting.