Simplified IFRS 9 Staging for Expected Credit Loss

The staging of receivables for determining Expected Credit Loss (ECL) is a key concept under IFRS 9. This process involves complex modeling and can be based on factors such as delinquency, observed probabilities of default (PD), or a combination of both. IFRS 9 outlines two rebuttable presumptions: paragraph B5.5.37 assumes that a receivable is considered defaulted if it is more than 90 days past due, while paragraph 5.5.11 presumes that credit risk has significantly increased if the payment delay exceeds 30 days. However, as these are rebuttable presumptions, management may challenge them and determine their own staging buckets based on internal data.

A simpler approach to staging could involve using PDs. By analyzing PDs over a specific period, we can identify where the rate of increase in PDs changes and use that as a basis for determining stages.

The example below shows the cumulative default rates reported by Fitch in its 2023 study for three years:

From this data, we observe that the PD grows slowly up to a BBB+ rating but increases significantly from BBB+ to B-. Beyond B-, the growth of PD becomes exponential. Based on these patterns, the receivables can be classified into Stage 1, Stage 2, and Stage 3, depending on the observed PD behavior.

A key caveat is that the PDs published by rating agencies are typically through-the-cycle (TTC) PDs, which need to be adjusted to point-in-time (PiT) PDs using models such as the Vasicek model. Additionally, asset correlation should be incorporated to ensure the final PiT PDs represent the true credit risk at a specific point in time, which is critical for accurate staging model.

?