Understanding IFRS 9 expected credit losses in banking

Posted by

admin

December 7, 2025

On December 7, 2025

Category: IFRS 9 & Compliance | Section: Knowledge Base | Publish date: 2025-12-01

Financial institutions and companies that apply IFRS 9 and need accurate, fully compliant models and reports for Expected Credit Loss (ECL) calculations face practical challenges: integrating credit risk models with accounting, meeting disclosure requirements, and proving robust governance to auditors and supervisors. This guide explains IFRS 9 expected credit losses, why IFRS 9 replaced IAS 39, the IFRS 9 impairment model and staging approach, and gives step-by-step, actionable advice to build compliant ECL processes and reports.

Why this topic matters for financial institutions and companies

IFRS 9 changed the provisioning landscape: it moved accounting from an incurred-loss model to an expected-loss model, which requires forward-looking estimates and earlier recognition of credit losses. For banks and corporates the consequences are material: balance sheet volatility, changes in provisioning and capital planning, enhanced disclosure requirements, and heavier demands on credit risk modelling and data pipelines.

Understanding the background and Definition of IFRS 9 is essential: IFRS 9 was issued to improve on IAS 39 and provide a more timely, transparent reflection of credit risk. The standard’s objectives are also instructive for practitioners, see the guidance on the Objectives of IFRS 9 to align your implementation with policy intent.

Core concepts: IFRS 9 expected credit losses, impairment model and staging approach

What is the expected loss model in accounting?

Under IFRS 9 expected credit losses (ECL) must be recognized for financial assets measured at amortized cost or fair value through other comprehensive income. ECL = Probability of Default (PD) × Loss Given Default (LGD) × Exposure at Default (EAD), discounted to present value. The core shift is: estimate expected lifetime losses when credit risk has significantly increased; otherwise recognize 12-month ECL.

Staging approach explained (Stage 1, Stage 2, Stage 3)

Assets are classified in three stages:

Stage 1: No significant increase in credit risk since initial recognition — recognize 12-month ECL.
Stage 2: Significant increase in credit risk — recognize lifetime ECL (but asset not credit-impaired).
Stage 3: Credit-impaired — recognize lifetime ECL and interest revenue on net carrying amount.

Significant increases in credit risk are determined using quantitative and qualitative indicators: e.g., a relative PD increase, >30 days past due, or borrower-specific information. The standard does not prescribe a single test; documentation and governance are required to justify the trigger.

Simple ECL calculation example

Example — retail loan of initial carrying amount 100,000 with remaining term 2 years:

Estimate PDs: Year 1 PD = 2%, Year 2 conditional PD = 4%.
Estimate LGD = 40%, EAD = 100,000 for both years.
Year 1 expected loss = 100,000 × 2% × 40% = 800.
Year 2 expected loss = 100,000 × 4% × 40% = 1,600; lifetime ECL = 2,400.
If in Stage 1, recognize 12-month ECL = 800; if in Stage 2, recognize lifetime ECL = 2,400.

Discounting: unless immaterial, discount these cash flows using the effective interest rate.

Credit risk modelling under IFRS 9

Practical IFRS 9 models must link to your credit risk frameworks — PD curve generation, LGD segmentation, forward-looking macro scenarios and overlays. For deeper discussion about how modeling practices changed under the standard, see our discussion of IFRS 9 ECL modeling and the implications for model governance.

Practical use cases and recurring scenarios

Monthly provisioning and reporting

Most banks run monthly ECL calculations to support financial reporting close. Typical pipeline: nightly exposure roll-forward → run PD/LGD models → generate base-case and alternative macro scenarios → compute ECL per instrument → aggregate for financial reporting. Expect run times of minutes to hours depending on portfolio size; optimize by precomputing lifetime PDs for stable segments.

Portfolio reviews and staging migrations

At quarter-end, credit risk teams must review staging decisions. A common scenario: a corporate borrower shows early signs of stress — downgrade triggers lifetime ECL even if not yet >30 days past due. Documentation should record the indicators and governance approvals for migrating the exposure to Stage 2.

Model recalibration and validation

Frequent recalibration is required after material economic changes. Model validation teams should run backtests comparing realized default rates to PDs and test sensitivity to macroeconomic assumptions. Use independent validation reports before sign-off to the audit committee.

Supervision and regulatory interaction

Supervisors expect transparent methodologies and evidence. To prepare, align your ECL governance with supervisory expectations and review guides such as the IFRS 9 ECL supervision material to anticipate common supervisory questions and documentation needs.

Impact on decisions, performance and outcomes

IFRS 9 affects profitability, capital planning, pricing and risk appetite:

Profit & Loss: Earlier recognition of expected losses increases P&L volatility — initial adoption often causes a one-off increase in provisions.
Capital: Higher provisions lower CET1 temporarily; banks must plan transitional arrangements and capital buffers.
Pricing: ECL expectations feed into loan pricing and credit limits — pricing models should incorporate expected lifetime credit losses for longer-term products.
Operational: Increased data, scenario management, and reporting needs create IT and process upgrades.

For a concise summary of the broader business effects, including observed industry impacts, see our analysis of IFRS 9 impact.

Common mistakes and how to avoid them

Below are frequent errors and practical mitigations learned from implementations.

Pitfall: Loose definition of “significant increase in credit risk”

Fix: Set quantitative thresholds (e.g., relative PD increase of X% or absolute increase of Y basis points) and complement with qualitative override rules. Log approvals and maintain a migration dashboard for auditors.

Pitfall: Ignoring forward-looking information

Fix: Use at least three macro scenarios (base, upside, downside) with assigned probabilities. Tie scenario design to your ICAAP and stress-testing processes to avoid inconsistent assumptions.

Pitfall: Data gaps and weak EAD inputs

Fix: Map critical fields (origination date, contractual cash flows, collateral valuations) and run data completeness checks. Where data are missing, apply validated proxies and document them in model governance.

Pitfall: Overcomplicated calculations early on

Fix: Start with pragmatic segmentation and expand granularity as data quality and model maturity improve. Use simplified PD/LGD buckets initially but maintain an explicit roadmap to full segmentation.

Practical, actionable tips and a checklist

Implementing or improving IFRS 9 expected credit losses calculations requires coordinated actions across credit risk, finance, IT and compliance. Use this checklist to prioritize tasks.

Pre-implementation checklist (or review checklist)

Governance: Establish an IFRS 9 steering committee with monthly sign-offs.
Data: Inventory critical fields and measure completeness; target >98% completeness for origination and performance data.
Models: Document PD, LGD, EAD methodologies and validation evidence.
Scenarios: Define at least three forward-looking macro scenarios with probabilities.
Controls: Implement reconciliations between risk and finance systems and an automated roll-forward for balances.
Disclosures: Prepare disclosure templates and narrative notes aligned with IFRS 9 disclosures.

Operational tips

Automate repeatable steps (data ingestion, model runs, aggregations) to compress monthly close to days rather than weeks.
Version-control model code and scenario assumptions for audit trails.
Maintain an “ECL dashboard” for senior management showing provisions by segment, drivers, and scenario sensitivities.
Use specialized vendor solutions as needed; evaluate IFRS 9 ECL tools for functionality like lifetime PD curves and scenario management.

Model governance and validation

Ensure independent validation, clear model owners, and a remediation plan for model deficiencies identified during internal audit or external inspection.

KPIs / success metrics for IFRS 9 expected credit loss programs

Coverage ratio (provisions / gross loans) — trend and variability.
Provision volatility (quarterly standard deviation) — monitor against peers.
Model backtesting error (realized defaults vs PD forecast) — target stable calibration within tolerance bands.
Data completeness rate for critical fields — target >98%.
Time-to-close for monthly ECL run — target within reporting calendar (e.g., 5 business days).
Number of audit or supervisory findings related to ECL — aim for zero critical findings.
Disclosure completeness score — internal checklist coverage of required items.

FAQ

How do I decide between 12-month and lifetime ECL?

Assess whether credit risk has increased significantly since initial recognition. Use objective triggers (PD changes, forborne status, 30+ DPD) and qualitative information (borrower-specific credit deterioration). Document your policy and maintain evidence trails for each migration.

Can I use regulatory PDs for IFRS 9 ECL calculations?

You can use regulatory PDs provided they meet IFRS 9 requirements for time horizon and forward-looking adjustments. Validate that the PDs are appropriate for accounting purposes or apply mapping and overlays where necessary.

How many macro scenarios should we include?

At minimum three scenarios (base, upside, downside) with assigned probabilities. Many institutions add a stress scenario aligned with ICAAP. Document scenario drivers and link to economic forecasts used across the bank.

What disclosure items typically attract auditor attention?

Auditors focus on methodologies for PD/LGD/EAD, scenario probability assignment, staging policies, and governance evidence. Prepare reconciliations and sensitivity analyses to support your disclosures.

Next steps — accelerate IFRS 9 compliance and accuracy

Ready to improve your ECL calculations and reporting? Start with a short action plan:

Run a 90-day diagnostic covering data gaps, model maturity and governance.
Prioritize fixes: critical data fields, scenario definition, and model validation.
Automate repetitive steps and build an ECL dashboard for senior management.
Engage an independent reviewer before your next reporting date.

If you prefer hands-on support or a platform tuned to IFRS 9 workflows, consider trying eclreport’s services — we provide validated tooling, templates and implementation expertise to reduce time-to-compliance and improve disclosure quality. For guidance on selecting vendor tools, review our evaluation of IFRS 9 ECL tools.