IFRS quantitative finance series
Macroeconomic Scenarios in IFRS 9: From Forecasts to ECL
Forward-looking information is central to IFRS 9. The challenge is to convert economic narratives into coherent paths for unemployment, rates, inflation, property prices and output, then connect those paths to credit losses without double counting stress.
Scenario design
A useful scenario set spans materially different outcomes and remains internally consistent. Scenario severity should be judged by portfolio impact, not by whether every variable sits at the same percentile. The bank should document forecast sources, cut-off dates and why the selected range captures relevant nonlinearities.
Weights and nonlinearity
Applying one model to a weighted-average macro path can differ from calculating ECL in each scenario and then weighting the losses. The latter preserves nonlinear relationships between economic conditions, defaults, collateral and exposure. Weights should reflect reporting-date information rather than a desire to reach a target allowance.
Forecast horizon and reversion
Beyond the reasonable and supportable forecast horizon, assumptions should move toward a documented long-run basis. Abrupt reversion can create artificial cliffs in lifetime ECL, so transition mechanics and sensitivity to the horizon should be tested.
Quantitative expression
Implementation controls
- Approve narratives, variables and weights as one package.
- Test nonlinear portfolio response.
- Prevent double counting between scenarios and overlays.
- Archive every vintage used in reporting.