A Forward-Looking IFRS 9 Methodology, Focussing on the Incorporation of Macroeconomic and Macroprudential Information into Expected Credit Loss Calculation

The International Financial Reporting Standard (IFRS) 9 relates to the recognition of an entity’s financial asset/liability in its statement, and includes expected credit loss (ECL) framework for recognising impairment. quantification ECL is often broken down into three components, namely, probability default (PD), given (LGD), exposure at (EAD). IFRS standard requires that model accommodates influence current forecasted macroeconomic conditions on loss. This enables a determination forward-looking estimates impairments. paper proposes methodology based principal component regression (PCR) adjust PD term structures forecasts. We propose risk index (CRI) derived from historic defaults approximate behaviour portfolio. PCR used CRI with variables as set explanatory variables. A novice all-subset variable selection proposed, incorporating business decisions. demonstrate method’s advantages real-world banking data set, compare it several other techniques. proposed portfolio-level recommendation derive scalar each different segment intended loan-level PDs information.