IFRS 9 Credit Loss Models Tested by Middle East Conflict & Missile Attacks

When Missiles Fly, Credit Books Bleed

How Escalating Tensions Between Iran, Israel, and the US are Forcing Middle East Banks and Corporates to Rethink Credit Loss Provisioning Under IFRS 9

Executive Summary

The ongoing conflict involving Iran, Israel, and the United States has escalated beyond a security issue. It is quietly becoming a financial problem as well. For banks and companies across the Middle East, the unrest is affecting their financial assets. They are facing higher credit risk, worsening borrower profiles, and increasing uncertainty about collateral recovery. Under IFRS 9, organizations must account for these factors in their Expected Credit Loss (ECL) provisions using a forward-looking approach. This article looks at how the conflict changes the main ECL drivers: Probability of Default (PD), Loss Given Default (LGD), and Exposure at Default (EAD). It also provides practical advice for professionals dealing with this challenge.

1. A Story That Starts in a War Room, Ends in a Boardroom

In the summer of 2024, as Israeli airstrikes increased over Iranian-backed targets in Lebanon and Gaza, and the United States sent more naval assets into the Persian Gulf, something unusual occurred in the boardrooms of banks in the UAE, Saudi Arabia, and Bahrain. Risk committees held special meetings not to discuss military strategy, but to revisit their Expected Credit Loss models.

They asked a seemingly simple question: Are our current loan loss provisions sufficient given the situation around us?

This is a question that Chartered Accountants, finance professionals, and auditors in the region will need to answer more often. The answer is closely tied to IFRS 9 and its ECL framework. The challenge is that most ECL models in the region were created during a time of relative stability. They were never stress-tested against a scenario like this one, where geopolitical risk, sanctions exposure, collateral deterioration, and trade disruption have all come together at once.

2. What Is ECL, and Why Does It Matter Now?

Before the 2008 global financial crisis, banks recognized credit losses only when they actually happened. A borrower defaulted, and only then did the bank record the loss in its income statement. This “incurred loss model” faced a lot of criticism for being too slow and focused on the past.

IFRS 9, which became mandatory on 1 January 2018, changed this significantly. It introduced the Expected Credit Loss model, which requires companies to recognize possible losses before they happen. The idea is straightforward: if you see a storm coming, you should put on your raincoat now, not wait until you are already soaked.

Under IFRS 9, ECL refers to the probability-weighted present value of credit shortfalls. This is the difference between the cash flows that are expected and the actual cash inflows, discounted at the effective interest rate. This standard applies to financial assets evaluated at amortized cost or fair value through other comprehensive income (FVTOCI), lease receivables, loan commitments, and financial guarantee contracts.

3. The Three-Stage Model: A Health Check for Every Loan

Think of it like a hospital classifying patients into three wards. Ward 1 has healthy patients with no signs of illness. Ward 2 has patients whose condition has worsened but who are not yet critical. Ward 3 has critically ill patients who need intensive care. IFRS 9 does something similar by placing every loan, bond, or receivable into one of three stages based on how credit risk has changed since it was first recognized.

• Stage 1 includes assets with no significant credit deterioration. The entity recognizes only a 12-month expected credit loss (ECL), which represents anticipated losses from defaults that could happen within the next year.
• Stage 2 includes assets with a significant increase in credit risk since initial recognition, but that are not yet in default. The entity must recognize lifetime ECL over the entire remaining life of the instrument.
• Stage 3 includes credit-impaired assets where default has occurred or is expected soon. Lifetime ECL continues, but interest revenue is calculated on the net carrying amount instead of the gross balance.

Stages are not permanent. Assets can be upgraded if credit conditions improve, giving the model a dynamic and forward-looking character.

IFRS 9 offers an important safeguard. There is a rebuttable presumption that Significant Increase in Credit Risk (SICR) has happened when a borrower is more than 30 days overdue. Besides being late on payments, the SICR assessment must also look at future qualitative signs. This includes any decline in the borrower’s operating situation or exposure to conflict zones. In the current context of the Middle East, both signs may need to be considered at the same time. A borrower running a hotel in Beirut or a shipping company operating through the Strait of Hormuz might have been clearly in Stage 1 twelve months ago. Today, their situation could be very different.

4. The Mathematics Behind the Worry: PD, LGD, and EAD

ECL is not a gut feeling. It is a calculation resting on three pillars:

ECL = PD x LGD x EAD

4.1 Probability of Default (PD): How Likely Is a Borrower to Fail?

PD is the likelihood that a borrower will fail to repay within a specific time frame. Under IFRS 9, it has to be a Point-in-Time (PiT) estimate that reflects current and future conditions, not just a long-term average.

For banks, estimating PD relies on internal historical data. A bank looks at its loan portfolio over a 5 to 10 year period and tracks how many borrowers in each credit grade actually defaulted. This history is broken down by borrower type, industry, geography, and loan term, and it serves as the foundation of the PD model.

Illustrative PD Derivation from Internal Data:

These historical probabilities of default (PDs) are converted to a Point-in-Time estimate by applying a macro-overlay or scalar adjustment to reflect current conditions. A bank that had a Grade B PD calibrated during stable times must now increase that estimate to match a live conflict environment.

For Stage 2 assets that need lifetime expected credit loss (ECL), a single 12-month PD is not enough. Banks must create a forward PD term structure, which is a multi-year series of conditional default probabilities modeled using survival functions or hazard rate models. Under conflict stress, this term structure steepens sharply in the near term and must be recalibrated at each reporting date.

Banks also use internal credit scorecards that assign scores based on financial ratios, such as leverage and debt service coverage. They also consider behavioral signals like repayment history and qualitative factors such as management quality and industry outlook. When internal data is limited, especially for sovereign or large corporate exposures, banks use external rating agency benchmarks like Moody’s and S&P as substitutes.

Crucially, IFRS 9 does not allow stopping at historical data. PD estimates must include macroeconomic variables and future-looking overlays. For a UAE bank lending to a Jordanian trading company, current overlays must now consider oil price changes, currency risk from US dollar peg pressures, trade route disruptions through the Strait of Hormuz, and counterparty concentration in Iran-linked corridors. The Through-the-Cycle (TTC) PD used in many older models is too stable for this environment. IFRS 9 requires a Point-in-Time (PiT) PD, and that change alone can significantly raise provisions overnight.

4.2 Loss Given Default (LGD): How Much Will You Actually Lose?

LGD answers the question: if the borrower defaults, what proportion of the exposure will not be recovered?

LGD = 1 – Recovery Rate, where Recovery Rate = Net Recoveries / Exposure at Default

Example: A Bahrain bank extends a USD 1 million mortgage loan to a Beirut developer. The borrower defaults. Collateral is sold for USD 600,000, net of USD 50,000 in legal and selling costs. Recovery rate = USD 550,000 / USD 1,000,000 = 55%. Therefore, LGD = 45%.

The ongoing conflict affects LGD in several ways:

• Collateral devaluation: Values near conflict zones have dropped sharply. Lebanese commercial property is down 30 to 40%. Even assets in the UAE and Saudi Arabia face indirect pressure as investor risk tolerance decreases.
• Legal enforcement delays: In areas where courts are overwhelmed or not functioning due to conflict, recovery timelines are longer. This delays the time value of money and reduces expected recoveries.
• Cross-border recovery complexity: Many banks in the Middle East hold collateral across different jurisdictions. A conflict that disrupts neighboring states makes enforcement in those areas more uncertain.
• Credit enhancements under scrutiny: Guarantees from Iranian-linked entities or sanctioned parties may no longer be legally enforceable. Preparers must carefully check whether such enhancements are still valid and can be used in recovery calculations.

A crucial concept here is Downturn LGD. IFRS 9 does not allow the use of average recovery rates from good periods. LGD must reflect the conditions at the time of default, including the chances that a default happens during a downturn. For Middle East portfolios, that downturn is not theoretical. It is happening now. Assumptions for prior-year LGD in real estate, hospitality, oil services, and shipping are likely much too low.

4.3 Exposure at Default (EAD): How Much Is at Stake?

EAD is the total exposure at the moment of default. For revolving facilities, letters of credit, and undrawn lines, it is estimated using a Credit Conversion Factor (CCF):

EAD = Current Balance + (Undrawn Amount x CCF)

In stressed environments, borrowers use committed lines more quickly than normal, similar to people rushing to ATMs before a storm. We should increase CCF assumptions for Middle East counterparties. For trade finance guarantees, if an Iranian or Lebanese counterparty defaults on a transaction, the regional bank acting as the guarantor faces invocation. That off-balance sheet EAD needs immediate updating.

A less-discussed but important risk is sanctions-driven crystallization. When OFAC or the UN Security Council designates a counterparty, settlement obligations freeze almost instantly. Trade finance positions, SWIFT payments, and cross-border repayments can become unrecoverable overnight, not due to unwillingness to pay, but because of legal prohibition. For banks with Iran-connected counterparty chains, this is a real exposure, not a distant risk.

5. The Forward-Looking Lens: Scenarios Are Not Optional

IFRS 9 requires multiple probability-weighted macroeconomic scenarios in ECL calculations. Most entities use three: a base case, an upside, and a downside.

For entities in the Middle East today, a believable downside scenario would include a full-scale military escalation in the Strait of Hormuz, which would disrupt 20% of global oil flows. This scenario could also involve US secondary sanctions that freeze SWIFT-based trade settlements, along with sovereign rating downgrades for Bahrain, Jordan, or Lebanon due to fiscal issues related to conflict. These events are plausible and can be documented. IFRS 9 requires that each event carries a specific probability weight.

Most ECL models were set up during calmer times. Their links between macroeconomic variables and the relationship between GDP growth and PD do not account for an ongoing regional conflict. Here is where management overlays become crucial. However, each overlay must be backed by “reasonable and supportable” evidence, which may include IMF country assessments, central bank data, or recognized conflict risk indices.

A subtler but important technical point is that ECL is non-linear under scenario weighting. A downside scenario with a 20% probability can greatly influence the final provision because PD responds in a non-linear way to macroeconomic stress. Management cannot give a minimal weight to an escalation scenario and think it is managed. The model’s mechanics will guarantee that a well-calibrated downside is reflected in the final number, which is exactly what the standard intends.

6. Sector-Specific Flashpoints in the Middle East

Banking and Financial Institutions: Cross-border exposures to Levantine borrowers, trade finance corridors near Iran, and sovereign bond holdings in conflict-affected states face multiple SICR triggers. Revolving credit facilities are particularly at risk as drawdown chances increase with borrower stress.

Real Estate and Construction: Property values in Lebanon and parts of Iraq have dropped significantly. Mortgage portfolios face rising PD due to income disruption and increasing LGD because of collateral decline at the same time.

Energy and Commodities Trading: Gulf producers benefit from price increases driven by conflict. Downstream importers sourcing from areas near Iran deal with supply uncertainty, cost pressures, and the risk of breaching covenants.

Tourism and Hospitality: Revenues in Jordan, Lebanon, and Egypt have fallen sharply due to decreasing inflows. Most regional banks need to reassess SICR for hotel and hospitality loans.

Shipping and Trade Finance: Houthi attacks on Red Sea routes in 2024 raised shipping insurance costs and required expensive rerouting. We need to reassess PD and EAD assumptions on related trade finance exposures immediately.

7. Practical Checklist for CA Professionals

8. The Disclosure Obligation: Tell the Story, Not Just the Number

IFRS 7 requires entities to disclose not only the ECL amount but also the context behind it. In light of a geopolitical crisis, disclosures must cover four key areas. First, they need to explain the nature of conflict-related risks included in ECL models and how those risks have been assessed. Second, they should detail the macroeconomic scenarios used and the probability weights for each. Third, they must discuss how sensitive the overall ECL is to changes in key assumptions, especially in downside scenarios. Fourth, they need to identify significant concentrations of credit risk in conflict-affected areas and among certain counterparties.

Transparency serves two main purposes. It meets the expectations of auditors and regulators, who are increasingly focused on the quality of forward-looking assumptions. It also gives investors and other stakeholders the information they need to evaluate an entity’s exposure to the ongoing crisis. Standard disclosures copied from last year’s annual report without meaningful updates will not hold up under scrutiny. If the ECL note in the 2025 financial statements looks the same as in 2023, there should be questions about why that is the case.

9. Conclusion: The Accountant in the War Room

There is an old saying that wars are won by generals but paid for by accountants. There is more truth in this than it seems.

When missiles are fired and borders are disputed, the immediate human cost is enormous. But the financial impact spreads outward, affecting credit portfolios, trade finance books, real estate values, and the credit ratings of entire nations. It is the job of preparers, auditors, and Chartered Accountants to make sure those effects are accurately and clearly shown in financial statements. Overlooking them, or assuming that last year’s provisioning model still works, is not neutral. It is a failure of professional duty.

IFRS 9 was created for situations like this. Its forward-looking principles, scenario-weighted methods, and requirement for “reasonable and supportable information” are not just bureaucratic rules. They are tools designed for uncertainty. The Middle East conflict tests how seriously entities take those tools.

For CA professionals, the message is clear. This is not just a newsworthy geopolitical event. It is a professional responsibility to understand, model, and disclose. Every loan book with regional exposure and every trade finance portfolio connected to the Persian Gulf or the Levant must now be viewed through the lens of IFRS 9, with updated assumptions, documented judgment, and full transparency. In the world of financial reporting, silence, even in the face of war, is not neutrality. It is a misstatement.

References

1. International Accounting Standards Board (IASB). (2014). IFRS 9 Financial Instruments. IFRS Foundation, London. https://www.ifrs.org/issued-standards/list-of-standards/ifrs-9-financial-instruments/

2. International Accounting Standards Board (IASB). (2010). IFRS 7 Financial Instruments: Disclosures. IFRS Foundation, London. https://www.ifrs.org/issued-standards/list-of-standards/ifrs-7-financial-instruments-disclosures/

3. International Accounting Standards Board (IASB). (2014). Basis for Conclusions on IFRS 9 Financial Instruments. IFRS Foundation, London. https://www.ifrs.org/content/dam/ifrs/publications/pdf-standards/english/2022/issued/part-a/ifrs-9-financial-instruments.pdf

4. Basel Committee on Banking Supervision (BCBS). (2006). International Convergence of Capital Measurement and Capital Standards: A Revised Framework (Basel II). Bank for International Settlements, Basel. https://www.bis.org/publ/bcbs128.htm

****

Disclaimer: The author is a Chartered Accountant with experience in IFRS implementation and financial reporting advice. The views expressed are personal and do not reflect the position of any organization.