Guideline 2: Methodology for DGS stress tests

2.1 In order to ensure a comprehensive approach, stress tests should be programmed over a medium-term perspective. Each stress-test exercise should follow a number of key phases as described in Guideline 2.

2.2 To safeguard compliance with Article 4(11) of Directive 2014/49/EU, designated authorities should ensure that DGSs obtain and use the information necessary to perform stress tests of their systems only for the performance of those tests and keep it no longer than is necessary for that purpose. To safeguard compliance with Article 4(9) of that Directive and in particular where the test involves the processing of data pertaining to depositors ’ accounts, designated authorities should ensure that DGSs preserve confidentiality, process that data pertaining to depositors’ accounts in accordance with Directive 95/46/EC⁴ and fully protect those data including by means of applying anonymisation methods as appropriate.

Programming of stress tests

2.3 Over a period that is not more than 3 years, DGSs should develop a stress-testing programme including exercises covering all core tests defined in Guideline 3. In any event, a stress-testing cycle reaches completion when all the core tests have been conducted at least once (since the enforcement of the guidelines or the adoption of the last cycle). 2.4 The programme should set out the estimated timeframe of programmed exercises and define the planned scope of each exercise in terms of formats and assumptions. The programme can include either one or several stress-testing cycles.

2.5 The programme should be updated on a regular basis, taking into account the results of previous stress tests (e.g. results that would highlight a need for a deeper assessment), actual DGS interventions or regulatory developments (e.g. a shortening in repayment deadlines).

2.6 Where a real-life intervention occurred during the stress-testing cycle and enabled a DGS to assess its resilience related to some or all of the core tests and/or the indicators under the programme, the DGS should consider amending the programme to reflect that the real-life-based test will substitute the originally envisaged test. In such a case, instead of completing all the key phases below, the DGS can focus on the reporting and corrective action phases.

Key phases of a stress-test exercise

2.7 DGSs should complete the following phases when conducting a stress-test exercise.

Planning phase

2.8 DGSs should appoint, among their internal staff, a steering team or steering officer (hereinafter referred to as ‘steering team’) in charge of planning and coordinating the various tasks involved in a stress-test exercise. For this purpose, internal staff means the staff employed or subcontracted by the DGS and staff from another public authority entrusted with the DGS’s functions provided by the DGSD and the BRRD. Without being part of the steering team, external participants and observers can be involved in the various phases of the stress tests. Senior management should ensure that the steering team is provided with all the information necessary and receives full support from the rest of the DGS staff.

2.9 Ahead of each exercise, the steering team should define the timeframe for executing the test and identify the internal and/or external participants and observers involved.

2.10 Based on the stress-testing programme, the steering team should define in more detail the focus of the test, formats, indicators to be measured and assumptions underpinning the exercise (e.g. amount to finance DGS’s intervention, level of payout in liquidation or credit institutions to be covered by quality checks of SCV files).

2.11 The DGS can use assumptions from previous intervention cases and assess the way in which the DGSs’ systems performed. They can also simulate the way in which, under the current conditions, their scheme would behave if it were faced with a similar situation.

2.12 The DGS should allocate the necessary resources for the test in terms of supporting staff, budget and infrastructure. The adequacy of these means should be continuously reviewed during the development of the exercise. 2.13 DGSs should make arrangements to ensure objectivity in the definition of assumptions for the stress test, the running of the test and the elaboration of unbiased conclusions. DGSs are asked to report on such arrangements in the reporting template. Such arrangements should be documented by the scheme and ensure that objectivity requirements apply to all participants and observers to the test and in all phases. As part of those arrangements, DGSs should establish a clear separation between the steering team and other participants and observers who, within the DGS, also take part in the exercise. DGSs should also report on elements the DGSs took into account when making such arrangements, such as the particular setup/governance of the DGS, costs, conflicts of interest, added value, national provisions on professional secrecy and DGS oversight.

2.14 Those arrangements can provide for the involvement of external observers of the process. Observers might be the designated authorities where they are not themselves administering the schemes, other public authorities, consultancy firms or other DGSs. Observers should strive to verify that the process is being led objectively and, in case of doubt, express their concerns to the steering team. Observers should have access to the relevant information regarding all phases of the process. Any information shared in this context should be subject to strict professional secrecy requirements. The requirement to establish separation or alternatively involve observers should be considered as met with regard to the tests of SCV files.

2.15 The steering team should contact the internal and external participants and observers who will be involved in the various stages of the exercise and ensure mutual understanding as to the role expected by everyone in the exercise.

Running phase

2.16 When running the exercise, the steering team should request and collect from test participants and observers the information necessary to assess the performance of the DGS’s systems in relation to the test areas and indicators described in Guidelines 3 and 4.

2.17 Exercises can be run in various formats, including live role-playing sessions where internal and external participants simulate the actions and decisions they would take in a given core test as defined in Guideline 3, or back-office exchanges (e.g. where the steering team requests SCV files from an institution and measures the accuracy of the information). DGSs should report on the type of format chosen for each exercise in the reporting template, using the following categories:

• • Desk-based reviews, which include reviews (of the quality) of existing procedures and arrangements in place, for example to outline and go through the processes of a (fictional) payout event from beginning to end, to assess a number of areas;
• • On-site inspections, for example visits by DGSs or their service providers to credit institutions in order to assess the quality of their SCV files. These visits can be provided and specified through the applicable national framework;
• • Simulations, for example an end-to-end simulation of a certain core test, or simulations of parts of the process such as the transmission of a payment instruction file from a home to a host DGS or the transfer of an amount of ex ante funding, and drawing on a credit line;
• • Real-life cases that occurred during the stress-testing cycle and that allowed for the assessment of the DGSs ’ capabilities included in these guidelines; and
• • Other types of exercises, to be used only when the exercise does not fit in any of the categories above and should be explained by the DGS when reporting.

2.18 Distinguished from the steering team, participants in the running phase should represent those authorities, entities or even internal departments, including within the DGS, which would have to take the necessary actions or decisions or provide the necessary information in a real-life scenario. This can include internal participants (for example, the internal department in charge of funding issues within the DGS) or external participants (for example resolution authorities which would determine, after consulting the DGS, its contribution to resolution).

Reporting and corrective action phase

2.19 The steering team should process and interpret the results of the test with a view to making an objective assessment of the resilience of the DGS in the achievement of its legal functions.

2.20 The steering team should record results in a consistent manner over time, using a standard template such as the template developed by the European Forum of Deposit Insurers. DGSs should report stress-test results to the designated authorities at least annually.

2.21 Stress tests should be part of a continuous improvement process. Accordingly, where weaknesses in a DGS’s systems are identified in the context of a stress test, that DGS should take corrective measures. Where weaknesses have been identified that are attributable to credit institutions, for example flaws in the quality of SCV files, the DGS should seek corrective measures, if necessary via the competent authority in charge of the supervision of those institutions. The DGS should then seek to ascertain, in subsequent tests, that the weaknesses have been remedied.

Cooperation with relevant administrative authorities

2.22 DGSs should keep designated authorities fully informed when planning and conducting stress tests, unless the DGS is also the designated authority. For this purpose, within 3 months after finalising it, DGSs should submit their stress-testing programme to the designated authorities. This information sharing could initiate a constructive dialogue and lead to refinements of the programme. The designated authorities should provide their input within 6 months of receiving the programme developed by the DGS. Any material update should be promptly reported to the designated authorities. 2.23 Subsequently, when planning each exercise, they should inform designated authorities of the scope of the test in terms of participating credit institutions, test formats, assumptions and any other relevant information.

2.24 In addition, before conducting a core test defined in Guideline 3, DGSs should inform the public authorities that would be involved in the kind of legal function being tested. At a minimum, the ‘ relevant administrative authority ’ identified pursuant to Article 3(1) of Directive 2014/49/EU, as well as the competent authority referred to in Article 2(1)(17) of the same Directive, should be informed when testing a repayment scenario. The competent and resolution authorities should be informed when testing a resolution scenario.

2.25 DGSs should seek the opinion of those authorities on assumptions for the test and offer them participation in the running phase. Where a DGS is separate from the designated authority, such participation or consultation can be organised via the designated authority.

2.26 Competent and resolution authorities should cooperate, directly or via designated authorities, with DGSs in defining scenarios and running tests.

2.27 DGSs should share the outcomes of the stress tests in the form of the reporting template provided in Annex 1 with the designated authorities. They should also share the outcome of the stress tests, in the form of the reporting template or in another form, with relevant authorities⁵ upon their request and subject to applicable confidentiality provisions.

⁴ Replaced by Regulation (EU) 2016/679 of the European Parliament and the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data.

⁵ Including (but not limited to) resolution authorities or national competent authorities.

1. Overview

1.1 References

1.2 Identified Requirements

1.3 Related Standards

2. Identified Requirements

3. Related Standards

• EBA - Published by the European Banking Authority at https://www.eba.europa.eu/