18 Sep 2026 8 min read
On 11 June 2026, the European Banking Authority (EBA) published its draft methodology, templates and template guidance for the 2027 EU-wide stress test, opening its industry consultation earlier than for any previous exercise. The earlier release is intended to give banks more time to prepare and follows industry feedback gathered during consultations in May 2026.
The EBA describes the proposed framework as introducing significant simplifications to improve efficiency and risk sensitivity, while preserving the robustness and comparability of results. The changes point to a stress-testing exercise that is, in many respects, structurally familiar, but more closely aligned with banks’ existing supervisory reporting, more selective in its data requirements and broader in its assessment of risk.
The overall architecture of the exercise remains broadly consistent with previous cycles. The 2027 stress test will cover 63 banks from the EU and Norway, including 47 euro-area institutions, representing approximately 75% of the EU banking sector. The EUR 30 billion minimum size threshold remains unchanged.
The core design principles are also retained: a static balance sheet, no defined hurdle rates, a common 30% tax rate and the existing pro-forma treatment of mergers, acquisitions and divestments. As in previous exercises, results will feed into the Supervisory Review and Evaluation Process (SREP).
Credit risk remains the backbone of the exercise. The core engine –projecting transitions between performing and non-performing exposures and assessing the resulting provisioning requirements– remains unchanged. The IFRS 9-based projection of Stage 1, Stage 2 and Stage 3 transitions (S1/S2/S3) is retained, as are the restrictions on releasing provisions for non-performing loans and on curing defaulted loans.
The exercise therefore represents less a fundamental redesign than a roll-forward of the established framework, with a number of targeted methodological changes introduced around it.
One of the most significant proposed changes is the reduction in the amount of data banks will be required to report specifically for the stress test. The draft methodology reduces the number of required data points by 55% compared with the 2025 exercise, largely through greater reliance on information already submitted through regular supervisory reporting, particularly FINREP and COREP. The direction of travel is consistent with the EBA’s broader “Simplifying to strengthen” program and responds to long-standing industry calls to reduce duplication between stress-testing and supervisory reporting.
The change is not simply a reduction in reporting volume. It signals a closer integration between the stress-test framework and banks’ existing data architecture, with a parallel FINREP consultation explicitly intended to allow starting-point credit data to be drawn from information already reported on a quarterly basis rather than through a separate stress-test submission.
For institutions, this creates an opportunity to reduce duplicate submissions and reconciliation effort, but also places greater importance on the quality, consistency and governance of supervisory data that will increasingly serve multiple regulatory purposes.
The simplification agenda is accompanied by several targeted changes to the risk methodology. For securitization exposures, the prescribed fixed risk-weight increase applied in the 2025 exercise is removed. Under the 2027 draft, securitization risk exposure amounts (REA) are instead floored at their 2026 starting-point value, bringing the treatment more closely into line with other portfolios.
At the same time, sector-level credit-risk reporting becomes more granular. The relevant data template is broadened from selected mortgage, real-estate and SME exposures to all non-financial corporate exposures, incorporates the updated NACE Rev. 2.1 classification and introduces a distinction between high- and low-energy-intensive manufacturing. This expansion provides part of the data infrastructure required for the new climate-risk module.
The treatment of counterparty defaults also becomes more risk-sensitive. Instead of assuming the default of three counterparties selected from the ten largest exposures, the 2027 draft considers two to five defaults from the 40 largest counterparties, with the number determined by the riskiness of the bank’s portfolio. While the methodology is simplified in some areas, this change could therefore prove more demanding for institutions with riskier counterparty books.
A further significant methodological development concerns market risk. The 2027 draft moves the description of market-risk capital impacts towards the Fundamental Review of the Trading Book (FRTB) framework, with the Sensitivities-Based Method (SBM), Default Risk Charge (DRC), Credit Valuation Adjustment (CVA) and Residual Risk Add-On (RRAO) forming the principal building blocks. This reflects the expectation that the newer framework will be fully in force by 2027.
Other elements of the market-risk methodology remain broadly unchanged, including the treatment of valuation reserves and the cap on client revenue projections. The shift is therefore principally one of methodological modernization, bringing the stress test into closer alignment with the evolving regulatory capital framework.
The most consequential new dimension of the 2027 exercise is the introduction of a dedicated climate-risk module. For the first time, the EU-wide stress test will assess both transition risk and physical risk, including flood risk, in a structured manner. The module will run alongside the standard macro-financial adverse scenario but, at this stage, will not affect headline stress-test results or capital requirements (REA).
The transition-risk component examines how a sudden or disorderly shift in climate policy could affect corporate borrowers in carbon-intensive sectors over the 2027-2029 horizon. The physical-risk component focuses on the impact of a one-in-100-year river flood event, applied as a one-off shock in 2027.
The scope initially focuses on lending to non-financial corporations and real-estate-secured exposures, including a breakdown of mortgages according to the energy efficiency of the underlying property. Climate shocks are layered onto, rather than substituted for, the conventional adverse macro-financial scenario.
The EBA explicitly positions this as a first step in a gradual, multi-year integration of climate risk into stress testing and supervision. The underlying climate scenario parameters, including carbon-price paths, flood maps and emission pathways, were not yet published with the consultation draft.
The 2027 draft presents a dual dynamic. On one side, banks can expect a leaner and more data-efficient exercise, with substantially fewer bespoke data requirements and greater use of existing supervisory reporting. On the other, the methodology introduces new analytical demands through climate risk, more risk-sensitive counterparty assumptions, expanded sector data and the FRTB-based market-risk framework.
The practical implications extend beyond the headline 55% reduction in data points. Institutions may need to assess the consistency of their FINREP, COREP and ESG data; review the mapping between supervisory reporting and stress-test requirements; assess the methodological implications of the revised credit, market and counterparty-risk treatments; and begin preparing the data and modelling capabilities required for climate-risk assessment.
The timing is also important. The consultation has been launched earlier than in previous exercises, providing banks with a longer preparation window while the methodology remains open to industry feedback. At the same time, several elements that will ultimately determine the severity and capital impact of the exercise remain unpublished, including the 2027 macro-financial adverse scenario and the detailed climate scenario.
The proposed 2027 EU-wide stress test can therefore be viewed as an evolution in two directions: simplification of the reporting infrastructure and expansion of the risk perspective. The framework retains the established architecture of the exercise while reducing duplication, bringing stress-test data closer to banks’ regular supervisory reporting and modernizing selected methodological components. At the same time, the introduction of climate risk marks a broader shift in the scope of stress testing, signaling the EBA’s intention to progressively incorporate environmental risks into the prudential assessment of banks.
For institutions, the priority is consequently not simply to prepare for a new stress-test cycle, but to understand how the revised methodology will interact with existing models, data structures, reporting processes, capital planning and risk-management frameworks.
Early assessment can help banks identify methodological gaps, determine where existing supervisory data can support the new reporting architecture, and prepare the modelling and governance capabilities required as the framework moves towards finalization.
Grant Thornton Financial Services supports institutions across the stress-test cycle, from methodology interpretation and data readiness to model validation and capital-impact analysis. This includes assessing the proposed methodology against existing stress-testing approaches, aligning stress-test requirements with FINREP, COREP and ESG reporting structures, supporting the design of the new climate-risk module, and assessing the implications of methodological changes for capital projections and regulatory processes.
Subscribe here to receive our newsletters
Targeted amendments following CRR3 preserve the core prudential framework while introducing important clarifications and operational changes.
The ECB’s July 2026 clarification reinforces existing supervisory expectations, with greater emphasis on how ICAAP and ILAAP are embedded, governed and evidenced in practice.
The adoption of the Crypto-Asset Reporting Framework (CARF) and the updated Common Reporting Standard (CRS 2.0) through the EU Directive DAC8 marks a significant evolution in the international tax reporting landscape.
