May 2, 2016

EU-Wide Stress Test: the ECB Challenge


On February 24th 2016, the ECB launched the EU-wide stress test (ST) and announced the EBA’s methodology and macroeconomic scenarios for this year’s ST Exercise. The ECB (SSM)[1] role is to oversee this exercise by coordinating and consolidating the ST results. However, next to the EU-wide ST, the ECB is also performing, in parallel, another supervisory stress test. These two supervisory stress tests cover approximately one hundred banks in total. One of the supervisory stress tests is the EBA EU-wide ST, covering a sample of 53 banks that represent 70% of the total EU banking assets[2]. The second supervisory ST is the SREP ST, this will cover banks of a smaller size and complexity[3]. Only the EBA EU-wide ST results are going to be published, this is expected by mid-July and is in alliance with the commitment to the overall EU banking sector transparency. 

For the SREP ST Exercise, the ECB has also decided to follow the EBA EU-wide ST methodology, not only to ensure consistency of the treatment of all significant institutions supervised by the SSM and to leverage the synergies created in terms of infrastructure, but also to contribute to the overall SREP[4]. The SREP is one of the major components of Pillar II; its key purpose is to ensure that institutions have adequate arrangements, strategies, processes and mechanisms as well as capital and liquidity to ensure a sound management and coverage of their specific risks, including those revealed by stress testing. In this regard, the outcome of the ST Exercise has a key role within the overall SREP assessment by acting as a challenger model for banks’ capital models and internal scenarios/stress tests approaches.

The quality assurance of this year’s ST Exercise has also been revised. It will now be organised along with 3 submission cycles each including four key phases. The first phase will focus on a thorough analysis of data quality, this is closely followed by a phase focused on the production of benchmarks. The next phase will then review all Quality Assurance issues and the final phase will be the communication of the final report to the banks. EBA banks were set their first deadline on April 15th 2016, at which they had to submit the full data required for the ST Exercise, meaning the 2015 figures and business plan projections for the next three years. Undoubtedly, this is a challenging exercise for the ECB considering the strict timelines and availability of resources as this quality assurance process will cover 100 banks. Looking at the current calendar, the ECB has four months to complete this exercise which implies a higher level of scrutiny in terms of data quality review and a wider coverage in terms of risk compared to 2014 exercise. Therefore, a first conclusion which can be drawn is that this increased effort by the ECB reconfirms what has also been observed in the US with the CCAR; that stress testing is coming to the forefront and it is perceived as a primary tool to set the adequate capital levels to spot emerging risks. 


Improvements and Limitations of the 2016 ST Methodology

The ST Exercise 2016 claims to be more conservative however, it is still considered not as severe as the CCAR in the US. Within the EBA ST only two scenarios are addressed (baseline and adverse) instead of three as in CCAR (baseline, adverse and severely adverse), but also the severity of CCAR scenarios is considered much higher than the ones applied in the EBA ST. Furthermore, CCAR includes a dynamic balance sheet for custom scenarios while the EBA ST remains under the static balance sheet assumption, constraints of zero growth and stable business mix and model (geographical, product strategies and operations)[5]. Although these assumptions are drawn to ensure comparability, they also preclude for incorporating the impact of any mitigation action by the banks. Therefore, the EBA ST will not cover a more forward-looking exploration of how institutions’ business plans might evolve under the stress scenario; also no feedback from ECB to the institutions in terms of mitigation actions will be made.

Another limitation of the current ST methodology refers to the liquidity assessment. The current ST Exercise does not cover the two-way interactions between liquidity and solvency as experienced during the last financial crisis. Hence, there is no inclusion of transferability of capital and liquidity resources under stressed conditions while considering any possible impediment, either legal or operational, that may exist. 

Nevertheless, it is also worth emphasising that despite these limitations, the ECB has eased the application process for banks (for example: detailing the explanation in the methodology or bringing the methodology closer to bank internal approaches) and has proposed simplification when possible. However, some key aspects such as the inclusion of new risks (for example: FX and conduct risk) have been prioritised over simplification.


Challenge of EU-Wide ST: the Quality Assurance Process (QA)  

For this ST Exercise, banks have not been required to perform an Asset Quality Review (AQR) as for the 2014 ST. However, the quality assurance of the banks results is enforced through several iterations and communications between the ECB and the banks. In this regard, the supervisory approach for 2016 has changed as follows:

  • The ST Exercise 2016 runs over three cycles. EBA banks will need to submit their data three times and with supplementary documentation, if required. The ST Exercise 2016 focuses on ensuring a sufficient data quality to generate proper bank projections and therefore fair results.
  • The pre-quality assurance data validation is tougher this year. Not only will banks have to perform their own data quality checks, based on the guidelines provided by the ECB, but also the ECB has put in place stricter tests and controls in terms of data quality at the bank’s submissions stage.
  • The ECB (SSM) challenges banks from three perspectives: the Joint Supervisory Team (JST or also called bank perspective), the Top-Down perspective and the Bottom-Up perspective. In this year’s exercise, the JSTs will play a key role as they will act as the central counterpart for the banks and single point of contact, channelling the communication between the ECB and the banks all along the exercise. The Top-Down perspective represents the Macro-prudential function while the Bottom-Up perspective represents the Micro-prudential function including the National Competent Authorities (NCAs).The Bottom-up perspective is also significant in this year’s exercise as it will provide a peer-benchmark input next to the macroeconomic scenarios.
  • The ECB provides the QA issues in terms of CET 1 impact and not based on the “RAG”[6] thresholds approach as in 2014 ST Exercise.
  • The ECB will publish the ST results by mid-July. However, a final report will be also be communicated to each bank consisting of a list of QA issues meaning a list of deviations ranked by materiality and in terms of capital impact (CET 1).

This year’s QA process intends to be holistic as it integrates three different perspectives and therefore combines different inputs. More precisely, the bank’s report will contain a list of issues that may come from any of the three different perspectives mentioned.  For example, a Top-Down issue could relate to the capital impact of interest rate shocks or any other macroeconomic variable, a Bottom-up issue may relate to the evolution and deviation of the PD parameters of a bank, compared to a pre-defined peer set and a bank issue which would relate to a specificity of the bank. It is expected that several iterations of reviews will be needed to reach the final list of issues combining the three perspectives and communicated to the bank.

Obviously, the current ST Exercise represents to the ECB a significant challenge in terms of coordination of tasks and information across the multiple stakeholders intervening in it: EBA and ECB (SSM), ECB teams (i.e. Top-Down team, Bottom-Up team, JST team) as well as, in terms of the effort required to execute the assessment and data quality assurance process for 100 banks in parallel. 


Additional challenge: the SREP Stress Testing

The supervisory stress testing is intended to be used as an additional source of information in the SREP process when assessing (a) risks-to-capital or risks-to-liquidity and funding, (b) internal risk governance and controls, (c) severity and plausibility of scenarios used for ICAAP and ILAAP purposes, and (d) recovery plans. As such, the ST outcome represents a forward-looking perspective which complement the supervisory perspective and bank internal perspective run under SREP.

Nevertheless, how exactly the ST quantitative and qualitative outcome is going to be integrated into the SREP remains unclear for banks.

A way to integrate the quantitative outcomes could be through the definition of materiality thresholds. As no minimum capital ratio has been defined for the ST Exercise 2016, materiality thresholds defined for a bank’s peer sets would allow for the assessment of the capital impact and the bank’s overall capital adequacy. In this sense, the ECB would make the ST a real challenger model.

To integrate the ST qualitative outcomes, the ECB would need to define the areas of assessment within a pass/fail criteria. To provide an example within that perspective, one could look at the process, controls and governance required for complying with the data requirements for submission to the ST Exercise. The level of efficacy in compiling this data and the data quality submitted, might be used to assess the maturity of the bank’s risk management practice, which would then be an input to the governance and risk management assessment SREP block. Likewise, the level of responsiveness to issues flagged by the ECB during the exercise requiring additional information or the quality of the additional explanations/comments provided might evidence as well of the bank’s organizational agility and therefore another item to be considered on the SREP.

Furthermore, the integration of the ST results into SREP is not only a key challenge for the ECB as described earlier, but also for banks. Any update to be performed by the banks in their SREP submission considering the ST results has rigid timelines, as SREP is expected by September 2016.  Then, it is important for banks to leverage in the ST Exercise for the SREP through an on-going communication with the ECB prior the final ST results publication to provide more room for manoeuvre.

[1] The EU-wide ST is performed by the ECB (SSM) – called hereafter ECB – under EBA coordination and in collaboration with EU-COM, ESTB, ECB and NCAs.
[2] Although the sample of banks for the ST Exercise 2016 has shrunk compared to the 2014, it still represents a significant coverage in terms of EU banking assets while increases comparability and consistency among banks.
[3] The SREP ST covers 60 significant institutions in SSM approximately.
[4] SREP ensures institutions’ capital and liquidity adequacy, as well as sound risk coverage and internal processes.
[5] More details on the differences can be found in the Blog "How different are EU and US stress scenarios".
[6] The quality assurance process in 2014 was based on RAG thresholds approach (Red/Amber/Green) to assess the alignment with the EBA methodology.


About the Author: 

Maribel Tejada is Senior Manager at Risk Dynamics. Within Risk Dynamics Maribel leads the development of the model risk advisory service line covering model risk management frameworks, methodologies, governance and overall model portfolio management strategy. Click here to email Maribel

Tags: Banking


Maribel Tejada

Maribel Tejada

Maribel is a Senior Manager at Risk Dynamics with more than 10 years of experience working for major institutions in corporate finance, capital management and risk management in the financial sector. She has experience in developing and deploying validation frameworks and risk appetite frameworks across several financial institutions. In the last years, she has been acting as an expert on model risk management topics and leading projects on model risk management diagnostic and transformation for major banks and insurers in the EU and US. At Risk Dynamics, Maribel leads the development of the model risk advisory service line covering model risk management frameworks, methodologies, governance and overall model portfolio management strategy.