EBA欧洲银行-EBA-BS-2012-78GL-on-Stressed-VaR-_31页_599kb
报告摘要
EBA Guidelines on Stressed Value at Risk (Stressed VaR)
Core Content
The EBA Guidelines on Stressed VaR are part of the regulatory framework under the Capital Requirements Directive (CRD III) and aim to enhance supervisory convergence and transparency in the calculation of market risk capital for trading books. These guidelines are based on the Basel Committee on Banking Supervision (BCBS) proposals and are intended to complement existing rules rather than replace them.
Main Objectives
- Achieve common understanding among EU competent authorities regarding Stressed VaR models.
- Enhance transparency for institutions in their risk management and capital calculation practices.
- Ensure a level playing field across financial institutions.
Key Information
I. Executive Summary
- The amendments to CRD III introduced Stressed VaR requirements to address weaknesses in the regulatory capital framework and reduce pro-cyclicality.
- The EBA is tasked with providing guidance to ensure supervisory practices are consistent across the EU.
- The guidelines are not a comprehensive rulebook but are intended to clarify the application of CRD III provisions.
- Competent authorities must implement the guidelines within six months of their publication and ensure effective compliance by institutions.
II. Background and Rationale
- The 2008 financial crisis highlighted the need for improved risk management and capital adequacy.
- The BCBS proposed Stressed VaR as a supplementary measure to the standard VaR framework.
- The CRD III amendments reflect the BCBS proposals and require institutions to use Stressed VaR for market risk capital calculations.
- The EBA guidelines aim to align supervisory practices and provide clarity on how to apply the rules in practice.
III. EBA Guidelines on Stressed VaR
Title I - Subject Matter, Scope and Definitions
- Subject matter: The guidelines focus on the identification, validation, and methodology of Stressed VaR models.
- Scope: Applies to institutions using the Internal Model Approach (IMA) for market risk capital calculations in the trading book.
- Definitions:
- Institutions refer to credit institutions and investment firms under Directives 2006/48/EC and 2006/49/EC.
- Antithetic data are price movements relevant regardless of direction.
- De-meaning is a process to remove trends from historical data to simulate price variations.
- Proxies are observable variables or prices used to substitute unobservable variables or those with unrealistic hypothetical prices.
Title II - Requirements regarding institutions' Stressed VaR modelling
A. Identification and validation of the stressed period
- Length of the stressed period: Must cover a continuous 12-month period of significant financial stress, even if the institution identifies a shorter period.
- Number of stressed periods: A single group-level stressed period is required unless it is not relevant to a subsidiary's portfolio.
- Approach for identifying the historical period:
- Institutions may use judgement-based or formulaic approaches.
- Formulaic approach is preferred and involves systematic quantitative analysis, such as identifying the most volatile period for risk factors.
- Institutions should consider a range of historical periods and ensure the selected one is relevant to their current portfolio.
- Documentation requirements:
- Robust documentation is necessary to justify the approach used.
- Quantitative assessments and evidence of model soundness must be included.
- For simplified models, justification for simplifications and evidence of consistency with full models must be provided.
B. Review of the stressed period
- Frequency: Institutions must review the stressed period at least yearly, but more frequent reviews may be required under certain conditions.
- Monitoring: Procedures must ensure that the stressed period remains representative of the institution's portfolio, especially after significant market or portfolio changes.
- Ongoing monitoring: Stressed VaR should be monitored relative to VaR to ensure it does not systematically underestimate risk. A ratio below 1 is a warning signal for review.
C. Stressed VaR methodology
- Consistency with VaR: Stressed VaR should align with VaR methodology, but may diverge in certain areas.
- Key areas of divergence:
- Confidence level: Consistent with VaR.
- Weighting scheme: Not required for Stressed VaR.
- Back-testing: Not a requirement for Stressed VaR, but the multiplication factor $m_s$ should be at least 3.
- Periodicity: Should be at least weekly, but institutions may choose more frequent calculations.
- Scaling method: Institutions should demonstrate that assumptions are appropriate.
- Use of proxies:
- Proxies are necessary for new or unobservable risk factors.
- Proxies should be conservative and well-documented.
- Institutions must demonstrate that the proxy is appropriate and that the impact of using it is limited.
Title III - Final Provisions and Implementation
- Implementation date: Competent authorities must incorporate the guidelines into their supervisory procedures within six months of publication.
- Notification requirements: Competent authorities must notify the EBA by 16.07.2012 whether they comply with the guidelines or provide reasons for non-compliance.
- Publication: Notifications must be published on the EBA website.
Summary of Methodological Consistency
| Aspect | Consistency Required | Not Required | Subject to Verification |
|---|---|---|---|
| Confidence level | ✅ Yes | ❌ No | ❌ No |
| Weighting scheme | ❌ No | ✅ Yes | ✅ Yes |
| Changes to models | ✅ Yes | ❌ No | ❌ No |
| Holding period | ✅ Yes | ❌ No | ❌ No |
| Back-testing | ❌ No | ✅ Yes | ✅ Yes |
| Use of Taylor series approximations | ✅ Yes | ❌ No | ❌ No |
| Length of historical observation period | ❌ No | ✅ Yes | ✅ Yes |
| Frequency of computation | ✅ Yes | ❌ No | ❌ No |
These guidelines ensure that Stressed VaR models are robust, transparent, and consistent with the broader regulatory framework.
展开完整摘要
试读结束,高清完整版pdf/doc/ppt,请点下载