IFRS 9 expected credit loss in SAP Analytics Cloud: staging, ECL and model structure

IFRS 9 Expected Credit Loss (ECL) provisioning is one of the most computationally intensive requirements in bank financial reporting — and one of the most consequential for the P&L. The shift from the IAS 39 incurred-loss model to the IFRS 9 expected-loss model means banks must now recognise credit losses earlier, using forward-looking probability estimates rather than waiting for a default event. This guide explains the three-stage model, the key parameters (PD, LGD, EAD), and how to structure ECL reporting and scenario analysis in SAP Analytics Cloud.

The three-stage model

IFRS 9 classifies financial instruments into three stages based on the change in credit risk since initial recognition. Stage 1: instruments with no significant increase in credit risk — provision equals 12-month ECL (the expected loss from default events in the next 12 months). Stage 2: instruments with a significant increase in credit risk but not yet credit-impaired — provision equals lifetime ECL (the expected loss over the remaining life of the instrument). Stage 3: credit-impaired instruments (effectively defaulted) — provision also equals lifetime ECL, but calculated on an individual basis. The stage determination is the key judgment call: moving an exposure from Stage 1 to Stage 2 typically multiplies the provision by a factor of five to ten, making stage boundaries a significant driver of P&L volatility.

The key parameters: PD, LGD and EAD

ECL is the product of three components. Probability of Default (PD): the likelihood that the borrower defaults within a given horizon (12-month for Stage 1, lifetime for Stages 2 and 3). Loss Given Default (LGD): the proportion of the exposure that would be lost if the borrower defaults, after recoveries and collateral. Exposure at Default (EAD): the amount outstanding at the point of default, including undrawn commitments for revolving facilities. ECL = PD × LGD × EAD × discount factor. All three parameters must be forward-looking (incorporating macroeconomic scenarios) and point-in-time (not through-the-cycle, as used for regulatory capital). This distinction — between IFRS 9 accounting provisions and Basel III regulatory capital — is a common source of confusion.

Multiple economic scenarios

IFRS 9 requires that the ECL estimate reflects a probability-weighted range of possible future economic conditions. In practice, banks typically define three scenarios (base, upside, downside) and weight them (e.g. 50% / 20% / 30%), with each scenario producing a different set of PD curves. The weighted average ECL across scenarios is the provision. SAC's Version dimension handles this naturally: each scenario is a version, the weighted average is a calculated measure, and the sensitivity of the provision to scenario weights can be explored interactively in a Story.

Structuring ECL reporting in SAC

An ECL model in SAC needs: Instrument or portfolio segment (at whatever granularity the bank models — individual exposure for Stage 3, portfolio cohort for Stages 1 and 2), Product type (mortgage, corporate loan, credit card, etc.), Stage (1, 2, 3 — and the movements between stages, which drive the P&L charge), Scenario (base, upside, downside), Version (Actual provision vs Forecast vs Prior period), and Period. The key output measures are: opening provision, new ECL charge, releases, write-offs, closing provision, and provision coverage ratio (provisions divided by gross exposure). Stage migration — the movement of exposures between Stage 1 and Stage 2 — is the most important driver to track, as it is where the largest P&L surprises come from.

The link to capital adequacy

IFRS 9 provisions interact with Basel III capital through the accounting-regulatory interface. Under the standardised approach, IFRS 9 provisions reduce the exposure subject to risk-weighting; under the IRB approach, there are complex rules around excess or shortfall of provisions relative to regulatory expected loss. Getting both calculations right in the same SAC environment — IFRS 9 for the P&L, Basel III for the capital ratio — is the primary use case for an integrated bank finance and risk model. See our Basel III in SAC guide for the capital ratio calculation.

Where to start

Our banking credit risk template provides the exposure and provision structure for ECL reporting, with PD/LGD/EAD dimensions and stage classification pre-built. You configure the parameters and validate the outputs — the template eliminates the blank-sheet setup. Not sure which template fits your regulatory scope? Let the assistant recommend one.

Sources

IFRS 9 Financial Instruments: IASB / IFRS Foundation. Basel III — IRB and standardised approaches: Bank for International Settlements.