Cross‑default clauses can create powerful cliff effects in IFRS 9 provisioning, particularly for smaller, concentrated banks in emerging markets that rely on a limited number of large borrowers for a significant share of their loan portfolios. When default on one facility automatically triggers default status on all facilities of an obligor or group, all related exposures may migrate to stage 3 in the same reporting period, generating a sudden spike in expected credit losses (ECL) and a sharp depletion of regulatory capital. This non‑linearity in provisioning can exacerbate procyclicality in downturns, especially where capital headroom is thin and credit‑risk models are less sophisticated.
Cross‑Default, IFRS 9 and Cliff Effects
IFRS 9 requires banks to define default in a way that is consistent with internal credit‑risk management and regulatory practice, which often embeds cross‑default through “unlikeliness to pay” and group‑level default notions. Under the three‑stage ECL model, most loss recognition still tends to occur at or around the time of default, as banks often delay transfers from stage 1 to stage 2 due to challenges in identifying a “significant increase in credit risk” at an early stage. This means that when a cross‑default trigger is activated, loans that were previously in stage 1 or 2 can be reclassified to stage 3 simultaneously, creating a provisioning cliff that reflects lifetime ECL on all exposures at once.
Vulnerabilities of Smaller Concentrated Banks in EMDEs
Smaller banks in emerging markets and developing economies (EMDEs) are particularly exposed to such cliff effects because their portfolios are typically more concentrated, both by obligor and sector, and their risk‑management infrastructure is often less developed. A single corporate group with multiple facilities can therefore represent a large share of risk‑weighted assets, so that cross‑default‑driven stage 3 migration has an outsized impact on provisions and capital ratios relative to more diversified institutions. Empirical evidence under IFRS 9 suggests that banks with less capital headroom provision less ahead of default and show larger jumps in provisions at default, implying that thinner‑capitalised, often smaller banks are more prone to under‑recognise risk until a cliff event occurs.
Procyclicality and Systemic Implications
The interaction between cross‑default and IFRS 9 can amplify the procyclical behaviour of bank lending in EMDEs, where macroeconomic volatility and weaker legal frameworks already pose challenges for credit‑risk management. Front‑loaded loss recognition induced by ECL, when concentrated in a small set of large obligors, may absorb scarce capital during downturns and force banks to cut new lending or tighten terms, thereby transmitting and magnifying real‑sector stress. Where provisioning systems remain relatively formulaic and rely on simplified staging thresholds, as documented in cross‑country reviews of loan classification and provisioning, the risk increases that deterioration in borrower quality is recognised late and then in a highly discontinuous fashion when cross‑default is triggered.
Supervisory and Risk‑Management Responses
Mitigating cross‑default‑induced cliff effects in smaller EMDE banks requires an integrated response combining accounting policy, risk management and prudential oversight.
On the bank side, clearer articulation of default definitions, including how cross‑default operates across products and group entities, should be accompanied by more granular early‑warning indicators and staging criteria that capture deterioration before legal default is reached.
Supervisors can complement this by imposing concentration risk limits, using Pillar 2 capital add‑ons or buffers to cover model risk and under‑provisioning, and by setting expectations for management overlays and scenario design that explicitly consider name concentration and cross‑default channels in stress testing.
Enhanced disclosures on ECL sensitivities, default definitions and the impact of alternative staging assumptions help investors and other stakeholders assess the magnitude of potential cliff effects more transparently.
References
Behn, M., et al. “Same Same but Different: Credit Risk Provisioning under IFRS 9.” ECB Working Paper Series, No. 2841, 2023.ecb.europa
Caruso, E. “IFRS 9 in Emerging Markets and Developing Economies.” World Bank, 2021.documents1.worldbank
Cohen, B., and J. Edwards. “The New Era of Expected Credit Loss Provisioning.” BIS Quarterly Review, March 2017.bis
International Accounting Standards Board. “IFRS 9: Financial Instruments.” 2021 (consolidated text).ifrs
Malovaná, S. “Banks’ Credit Losses and Provisioning over the Business Cycle.” Review of Economic Perspectives, 2022.sciendo
World Bank. “Loan Classification and Provisioning: Current Practices in 26 Countries.” FinSAC Technical Paper, World Bank, 2015.thedocs.worldbank
Leave a Reply