Key takeaways
- The provision matrix applies historical loss rates by ageing bucket, adjusted for current and forward-looking conditions.
- Loss rates should come from actual migration of balances through the ageing buckets — not from judgement alone.
- Segment customers with different risk profiles; one matrix rarely fits all.
- Forward-looking adjustments must be supported and documented.
The simplified approach in one paragraph
For trade receivables and contract assets without a significant financing component, IFRS 9 and EAS 47 require lifetime expected credit losses to be recognised from day one. A provision matrix is the most practical way to do this: balances are grouped by days past due, and a loss rate is applied to each bucket.
Step 1 — Segment your receivables
Group customers that share similar credit risk characteristics: government vs private, domestic vs export, distributors vs end-customers, or by product line. A matrix that mixes very different customers produces averages that fit none of them.
Step 2 — Build historical loss rates
Using monthly ageing reports over a representative period (typically two to five years), track how balances roll from one bucket to the next — the roll-rate or migration approach. Multiplying roll rates through to write-off gives a historical loss rate for each bucket.
| Ageing bucket | Balance | Historical loss rate | Adjusted loss rate | ECL |
|---|---|---|---|---|
| Current | 1,000,000 | 0.8% | 1.0% | 10,000 |
| 1–30 days | 400,000 | 2.5% | 3.0% | 12,000 |
| 31–90 days | 200,000 | 9.0% | 10.5% | 21,000 |
| 91–180 days | 80,000 | 35% | 38% | 30,400 |
| Over 180 days | 50,000 | 80% | 85% | 42,500 |
The figures above are illustrative only.
Step 3 — Adjust for forward-looking information
Historical loss rates reflect past conditions. Adjust them for current and expected conditions — for example, inflation, interest rates, sector performance or customer-specific developments. The adjustment should be linked to evidence, and its direction must make sense.
Step 4 — Consider individually significant balances
Customers in dispute, in financial difficulty, or with specific information indicating default should be assessed individually rather than through the matrix.
Common errors
- Using the ageing at a single date instead of migration over time
- Ignoring recoveries after write-off
- Applying the same loss rates year after year without recalibration
- Excluding related-party or government balances without analysis
- No reconciliation between the ageing report and the general ledger
How ASA can help
We build provision matrices in transparent workbooks, recalibrate existing models and validate management's ECL for audit purposes.
This article is for general information only and does not constitute professional advice. Thresholds, rates and procedures change — please contact us or refer to the latest official sources before acting.