Why is the current EU-wide stress testing framework losing relevance?

The current EU-wide stress test consists of a solvency stress test, mostly based on a constrained bottom‐up approach, designed with 4 main goals in mind:

1.     Identifying the risks and vulnerabilities of EU banks and contributing to the overall assessment of systemic risk in the EU financial sector

2.     Assessing bank’s resilience, challenging bank’s capital position and supporting the determination of Pillar 2 guidance (P2G)

3.     Fostering bank’s own stress-testing and risk management capabilities

4.     Strengthening market discipline through the publication of consistent and granular data

Banks and supervisors acknowledged that the exercises conducted since 2011 significantly contributed to the evolution of the financial system’s resilience and safety. One of the key features, that is decisive in strengthening of the financial system, was the use of bottom-up approach in stress testing since it pushed banks to invest in and to improve their data management and stress testing capabilities. Banks dedicate a significant amount of effort and resources, not only to be compliant with regulatory requirements but also to leverage the tools and methodologies developed to improve their own stress testing and risk management tools. While this benefit was true for the first exercises, since 2018 the EU-wide stress test methodology has stabilised and the focus of banks has shifted to execution rather than modelling.

While banks have been investing in more mature and sophisticated stress testing frameworks, they have increasingly become disengaged with the regulatory exercise as its limitations became more evident. For supervisors too, the current framework is losing relevance. In a discussion paper on the future changes to the EU-wide stress test (published in Jan. 2020), the EBA presented the following areas of concerns that contribute to this loss of relevance of the exercise:

1.     The lack of clarity and prioritisation of the current objectives which are conflicting. Particularly, whether the exercise aims to identify bank specific or system-wide risks in the EU banking sector

2.     The limited usage of the results and their link to the supervisory process. The rigidity of the exercise is the root for this concern as the methodological constraints and assumptions are not closely aligned to the risk perspective, which makes the outcomes of limited value for internal decision making and for supervisors to impose capital add-ons

From our experience with stress testing, we would also add the need of a complementary top-down approach that would assess how a particular event would impact the overall risk exposure and identifying emerging risks and vulnerabilities, both system-wide and bank specific.

How would the new framework improve the relevance of the exercise?

In the discussion paper, the EBA also proved open to review the modalities of the EU-wide stress test framework to improve its relevance. While disrupted by the COVID-2019 pandemic, that reflection remains ongoing and banks can expect to see a more meaningful and integrated stress testing framework emerging.

At the time, the EBA had proposed a framework with a “two-legged approach”, the bank leg and the supervisory leg.

On the supervisory leg, by using a combination of a constrained bottom-up approach and potential top-down features, the supervisor aims at increasing the reliability of the results. Additionally, the controlled relaxation of some methodological constraints is meant to make the results more realistic than in the existing framework.

On the other hand, the idea is to allow banks to use their own models for producing the bank leg results and thereby to enable a closer alignment with their internal processes starting with the ICAAP, which would support more effectively its usefulness for internal decision-making. 

Overall, the proposal by the EBA to move towards a “two-legged approach” is meant to strike a balance between risk sensitivity (focus of the bank leg) and comparability (focus of the supervisory leg) and could contribute to improving the relevance and use of the stress test outcomes.

Is the EBA moving towards a more risk sensitive framework?

Although the two-legged framework is not yet confirmed, the EBA is showing signs to move in the direction of a more risk-sensitive approach, as indicated by the changes implemented in the 2023 EU-wide stress methodology:

·       A first step of revising the EU-wide stress test framework towards a hybrid one was taken by introducing a top-down approach for the NFCI projections using prescribed growth rate parameters

·       Additionally, the EBA also hints at the removal of some of the constraints such as: the cap on NII under the adverse scenario on country/currency level and for fixed and floating portfolios, the 10.000 threshold for the conduct and operational risk losses, and requesting banks to use sector-specific risk parameters

·       The introduction of the new template for banks to measure and report credit risk exposure by economic sector (most likely to gauge the uneven impact of higher energy prices and interest rates across industries). This can be perceived as a sign that the EBA is preparing the data granularity for the bank leg with the goal of increasing the relevance and transparency. This will however require participating banks to develop new capabilities as few have developed sectoral credit risk models

With the proposed changes, banks can expect some additional internal benefits in the interpretation and application of the EU-wide stress test. However, in order to fully benefit from the proposed “two-legged approach”, banks will need to leverage strong stress testing expertise (e.g. EBA methodology and internal stress testing approaches) in order to “connect” the two exercises. Having assisted other banks in Luxembourg and overseas, Avantage Reply will continue to monitor the steps taken by the EBA towards the new framework, especially to which extent they would benefit from applying internal models rather than the methodologies imposed by the EBA.