Consultation on the future EBA stress test framework

The European Banking Authority (EBA) launched a consultation on the long-term strategy for the EU-wide stress test and on possible changes to the framework. In this discussion paper, the EBA proposed various changes to the framework which aims to improve five key areas in regulatory stress testing as well as to cover some insightful suggestions of the European Court of Auditors.

The European Banking Authority (EBA) is regularly executing EU-wide stress tests focussing on enhancing transparency, restoring trust in the financial sector and improving banks’ resilience after the financial crisis. To improve the stress testing methodology and processes, the EBA has organized numerous workshops and other (in)formal interactions with stakeholders.

Recently, the EBA held a consultation (in the form of a discussion paper) to have a more structural discussion on the long-term strategy for the EU-wide stress test. This two-pager outlines the historical developments of the regulatory stress test over the past decade and the proposed changes to the current framework. On 12 March, the EBA announced that it would postpone the 2020 EU-wide stress test until 2021.

The COVID-19 pandemic created significant immediate challenges to society and risks for the economic outlook. For 2020, the EBA will carry out an additional EU-wide transparency exercise in order to provide updated information on banks’ exposures and asset quality to market participants. Due to COVID-19 the deadline for the submission of comments has been extended by two months to 30 June 2020.

Figure 1: Historical development of supervisory stress testing

Historical developments in supervisory stress testing

Stress testing is used as a tool by supervisors to assess the resilience of banks and the banking sector as a whole through solvency measures.

The first successful effort by supervisors of using stress testing as a tool was implemented by the United States Federal Reserve in 2009 in the Supervisory Capital Assessment Program. This was a one-off exercise during the financial crisis. The supervisory stress test exercise was introduced in the EU as a crisis management tool in 2011. The purpose and methodology were very much driven by the urgent need to address uncertainty on banks’ exposures and capital levels.

The EU-wide stress test has evolved over time, following market developments, the progress of regulatory reforms, and changes in the architecture of EU financial supervision. In 2014, the stress test was used as an instrument to the set-up of the Single Supervisory Mechanism (SSM). The stress test was a component of the so-called Comprehensive Assessment that was executed in parallel to the stress test before moving supervisory responsibilities from the national level to the European Central Bank. In 2016, the focus shifted away from the assessment of capital shortfalls and was directed more towards the identification of potential vulnerabilities.

Prior to the EU-wide stress test conducted in 2018 there was a major update due to the revision of the accounting standards (IFRS 9). This required banks to adopt a forward-looking approach in determining the provisions for credit risk (as banks are required to calculate lifetime expected credit losses for credit impaired loans). Furthermore, the calculation of (net) interest income under various scenarios was affected by IFRS 9.

The EU-wide stress test has served different purposes over the past decade and the focus has changed with different exercises (see also Figure 1). In January 2020, the EBA launched a consultation on the long-term strategy for the EU-wide stress test and on possible changes to the framework. In this discussion paper, the EBA proposes various changes to the framework which also aim to address some of the suggestions of the European Court of Auditors which audited the 2018 EU-wide stress test. The consultation has runned until 30 June 2020.

Reasons for revising of the stress testing framework

So, what are the reasons for the EBA to change the current framework? The EBA feels that there are five areas where improvements can be made:

  1. Current lack of clarity and prioritisation of the EU-wide stress test objectives
  2. Usage of results and their link to the supervisory process
  3. Application of methodological constraints
  4. Ownership of results
  5. Resource-intensive nature of the exercise.

The remainder of this section elaborates on each of the abovementioned points.

First, in the current situation a number of objectives of the exercise are potentially in conflict. An example which is mentioned in the discussion paper is the microprudential purpose of having the stress test results feeding each bank’s Supervisory Review and Evaluation Process (SREP), which conflicts with the macroprudential objective of assessing systemic risks.

The second area of concern is the usage of results and their link to the supervisory process. The EU-wide stress test is far less integrated in the regular supervisory process than in other international frameworks such as the ones in the UK and the US. Some methodological constraints, such as the static balance-sheet assumption, which does not consider management actions, make it difficult to convert the stress test capital depletion into a meaningful supervisory capital add-on (i.e. Pillar 2 Guidance (P2G)) without further adjustments.

The third area of concern is the application of methodological constraints for some risks (as prescribed by the EBA) even though the banks’ internal capabilities for projecting these risks might have improved. For that reason, reassessing the application of some of these constraints could be beneficial for improving the realism of the stress test outcomes.

The fourth area of concern is the ownership of results. In the current approach, banks provide projections that can be overridden by supervisory benchmarks and challenger models. Even though banks are asked to confirm the figures before publication, they do not have to agree with them, and their consent is also not a legal requirement for the publication of the results by the EBA.

The last area of concern is the resource-intensive nature of the stress testing exercise. Often, the stress test exercise demands close cooperation and dedicated resources from multiple departments (for example Risk Management and Finance). The changes in methodology, scenario(s) and templates as compared to the previous exercise need to be assessed and incorporated. Data collection can be very challenging for banks because, often, various risk management systems are used. Furthermore, the quality assurance requirements imply costs for both banks and supervisors.

Proposed changes

The EBA proposal for the new stress testing framework stipulates the objectives and their prioritisation: the EU-wide stress test is primarily a microprudential exercise focused on the assessment of banks’ capital adequacy and the identification of risks. While the outcome from the EU-wide stress test can be used as an input in the assessment of systemic risks and the second-round effects of a crisis, the macroprudential role of the stress test is not considered one of its main objectives.

The framework is based on two so-called legs (see Figure 2 for an illustration): the supervisory leg and the bank leg. The supervisory leg is the basis for supervisory decisions (P2G) and is a constrained approach. The bank leg on the other hand allows for more flexibility and focusses on disclosure and fostering market discipline. Aside from the differences, some comparability between the two legs should be guaranteed.

Therefore, both legs would use the same scenarios and starting points for projecting the stress test results. The supervisory leg would be based on a constrained bottom-up approach similar to the current framework. This means that the banks themselves perform the impact calculations which are subject to a prescriptive methodology and supervisory quality assurance. Supervisors would be the owners of the supervisory leg and would form their view on the capital impact of scenarios (also by using challenger models and benchmarking tools, and even the bank leg results).

Banks would not be required to sign off the results. The bank leg would be obtained by using a flexible bottom-up approach. The bank leg is less prescriptive as compared to its supervisory counterpart. The results are fully owned by the banks and there is no quality assurance by the supervisor. In practice, the flexibility of the bank leg entails that the methodology for the supervisory leg is amended by a list of constraints which potentially can be relaxed or disregarded.

The discussion paper does not state which constraints can possibly be relaxed. This is subject to further analysis. An example of a constraint that could potentially be relaxed is the no cure assumption for stage 3 assets. Both the supervisory leg and bank leg results would be disclosed by EBA (bank by bank and on aggregate basis). The results for the supervisory leg would be limited to capital ratios and other key drivers, whereas for the banking leg more granular data is disclosed.

Figure 2: Proposed framework for an EU-wide stress test with a two-legged approach.

The new proposed framework would be introduced in the 2022 EU-wide stress test at the earliest. In the discussion paper a proposed roadmap can be found (see section 4, pg. 31).

Closing statement

The proposed changes will have a significant impact on the framework and the way banks report the results of the regulatory stress test. In case the EBA adopts the changes, the stress testing framework at bank level needs be revised. Zanders can help banks to prepare for reporting under the new framework.


Zanders can support in assessing the implications of relaxing methodological constraints. Please contact Robbert-Jan Milané for more information on regulatory stress testing

Share this article: