EBA欧洲银行-Guidelines-on-institutions-stress-testing-28EBA-GL-2018-0429_COR_HR_43页_679kb

Summary of EBA/GL/2018/04: Guidelines on Stress Testing of Institutions

1. Compliance Obligations and Reporting Requirements

• Status of the Guidelines: These guidelines are issued under Article 16 of Regulation (EU) No 1093/2010. Supervisory authorities and financial institutions must align with them, either through legal framework changes or supervisory procedures. Institutions must report to EBA whether they are compliant or intend to become compliant by a specific date, or provide reasons for non-compliance.
• Reporting Process: Institutions must submit a completed form to EBA via the email address compliance@eba.europa.eu with the subject line "EBA/GL/201x/xx". Any changes in compliance status must also be reported.
• Publication: Reports will be published on EBA's website in accordance with Article 16(3) of the Regulation.

2. Scope, Application, and Definitions

• Purpose: The guidelines aim to ensure common organisational conditions, methodologies, and processes for conducting stress tests, considering capital adequacy and risk management, within the institution's risk management framework.
• Applicability: The guidelines apply to institutions participating in a specific stress test implementation, in line with the scope of that test and the application level as described in Articles 108 and 109 of Directive 2013/36/EU.
• Definitions:
  • Stress Testing of Solvency: Evaluation of the impact of specific scenarios (macroeconomic or microeconomic) on the institution’s capital position, including minimum or additional capital requirements.
  • Stress Testing of Liquidity: Assessment of the impact of scenarios on the institution’s liquidity position, including funding and liquidity shocks.
  • Bottom-Up Stress Testing: Conducted by institutions using internal models, based on internal assumptions or scenarios, and using internal or external data. It focuses on specific portfolios or the institution as a whole, including concentration effects and risk spillovers.
  • Top-Down Stress Testing: Conducted by supervisory authorities or macroprudential bodies using general or systemic assumptions and scenarios. It is based on aggregated data and provides a common framework for comparing stress test results across institutions.
  • Static Balance Assumption: Assumes constant balance and stable business model during the projection period, enhancing comparability of results.
  • Dynamic Balance Assumption: Allows for changes in balance and business model, reflecting more realistic conditions but reducing comparability.
  • Portfolio Stress Testing: Focuses on the implications of shocks from individual or multiple risk factors on specific portfolios or the institution as a whole.
  • Sensitivity Analysis: Measures the impact of individual or multiple risk factors on capital or liquidity, using reference values or expert judgment.
  • Scenario Analysis: Involves assessing the impact of a set of risk factors in a specific scenario, with all relevant risk factors occurring simultaneously. It aims to identify risk interactions and feedback effects.
  • Reverse Stress Testing: Starts with a predefined outcome (e.g., insolvency) and identifies scenarios and conditions that could lead to it. It includes the assessment of management actions and strategies.
  • Second-Order Feedback Effects: Refers to the effects of the institution's responses to stress scenarios, which may amplify or mitigate the initial shock, leading to additional negative feedback loops.
  • Scenario Severity: Indicates the degree of stress in a scenario, from basic to adverse, expressed through macroeconomic and financial variables.
  • Scenario Likelihood: Refers to the probability of a scenario occurring, based on consistency with current macroeconomic and financial variables, expert judgment, and simulation methods.
  • Reference Scenario: A scenario typically developed by supervisory authorities to determine the severity level for a specific stress test.
  • Risk Data Aggregation: Involves collecting, processing, and analyzing risk data to measure an institution's risk tolerance and appetite.
  • Data Infrastructure: Refers to the physical and organisational structure and tools used to support risk data aggregation and reporting.

3. Implementation

• Effective Date: These guidelines apply from 1 January 2019.
• Withdrawal of Previous Guidelines: The CEBS guidelines (GL32) are withdrawn from 1 January 2019 upon publication in all official EU languages.

4. Stress Testing of Institutions

4.1 Stress Testing Program

• Institutions must establish a stress testing program that includes:
  • Types of stress tests and their objectives and applications.
  • Frequency of different stress tests.
  • Internal risk management solutions, including clear lines of responsibility and procedures.
  • Scope of subjects and coverage (e.g., types of risks and portfolios) in the case of groups.
  • Relevant data infrastructure.
  • Methodological details, including models and linkages between solvency and liquidity stress tests.
  • Range of assumptions and management measures related to each stress test.
• The program must be documented and reviewed regularly to ensure effectiveness and robustness.
• The program must be integrated into the institution's risk management framework, including ICAAP and ILAAP. It should support various business decisions and strategic planning, including capital and liquidity planning.
• Stress testing results (quantitative and qualitative) should be used as inputs for determining risk limits and appetite. They should also serve as a tool for planning and assessing the effectiveness of existing and new business strategies.

4.2 Governance Aspects

• The board of directors must approve the stress testing program and oversee its implementation and execution.
• The board must ensure that it has sufficient knowledge, skills, and experience to understand the impact of stress events on the institution’s risk profile.
• The board should engage in discussions with risk committees or external consultants, and review key assumptions, scenario selection, and risk tolerance levels.
• The program must be implemented in line with the institution’s internal policies and procedures, with clear assignment of responsibilities and sufficient resources (human, financial, and technical).

4.3 Data Infrastructure

• Institutions must ensure that their stress testing program is supported by an adequate data infrastructure.
• They should apply the principles of effective risk data aggregation and reporting as outlined by the Basel Committee.
• The data infrastructure must be capable of handling extensive data needs and provide mechanisms for continuous stress testing.
• It should offer flexibility, appropriate data quality, and control mechanisms.
• The data infrastructure must be proportionate to the institution's size, complexity, and risk profile, enabling the execution of stress tests covering all relevant risks.
• Institutions should allocate sufficient human, financial, and material resources for the development and maintenance of their data infrastructure, including IT systems.
• They should also consider stress testing of their data infrastructure as part of their overall IT infrastructure planning, including business continuity planning and long-term investments.

4.4 Scope and Coverage of Stress Testing

4.4.1 General Requirements

• Stress tests must consider all types of significant risks, including on-balance sheet and off-balance sheet exposures.
• The scope of stress testing can vary from simple portfolio sensitivity analysis to comprehensive institution-wide stress testing, depending on the institution's size, complexity, and risk profile.
• Institutions must consider changes in risk correlation at both the individual subject and group levels, especially during periods of economic or financial difficulties.

4.4.2 Stress Testing at the Level of Individual Risk and Portfolio

• Institutions must conduct stress tests on individual portfolios, considering all relevant risk types and using sensitivity analysis and scenario analysis.
• They should identify risk factors and appropriate stress levels at the portfolio level.
• Institutions must assess concentration risks within and across risk types, including the effects of risk spillovers.
• Special attention should be given to aggregating risk data across all risk types, particularly market and credit risks, to better understand potential risk concentrations under stress conditions.
• Institutions must review assumptions about dependencies and correlations of risk types under stress scenarios.

4.4.3 Stress Testing at the Level of the Entire Institution

• Stress testing at the institutional level provides a comprehensive and holistic view of the institution's risk profile.
• It includes the analysis of potential impacts of stress scenarios on the institution’s overall capital and liquidity positions, including feedback effects and interactions within the financial system.