EBA欧洲银行-Additional-guidance-to-the-methodological-note_16页_294kb

2011 EU-Wide Stress Test Methodological Note Summary

Introduction

The European Banking Authority (EBA) conducted a 2011 EU-wide stress test to evaluate the resilience of European banks to severe shocks and to establish a common, conservative benchmark. The methodology, published on 18 March 2011, was designed to ensure consistency and to identify vulnerabilities in the banking sector. The stress test involved 90 banks and included a quality assurance process to review and challenge results, ensuring the methodology was applied correctly and conservatively.

Key Features and Enhancements

• A new consistent capital benchmark of 5% Core Tier 1 capital was introduced.
• Clear assumptions and definitions were established to ensure consistency.
• A common baseline and adverse scenario was developed by the ECB and European Commission.
• A quality assurance process was implemented by EBA staff with support from national and ECB/ESRB experts.

Quality Assurance Process

• The quality assurance process involved bilateral engagement between the EBA, national supervisory authorities, and banks to clarify and address inconsistencies.
• The EBA staff highlighted areas where assumptions appeared optimistic and required adjustments.
• The process was not intended to replace ongoing engagement but to provide additional guidance on thematic issues.

Benchmarks

Benchmark 1: Sovereign and Financial Institution Exposure in the Banking Book

• The EBA decided that market risk haircuts would not apply to sovereign exposures in the banking book.
• Credit risk analysis is still required, using PDs (Probability of Default) and LGDs (Loss Given Default).
• IRB banks calculate both Expected Loss (EL) and Unexpected Loss (UL), while Standardised banks rely on credit ratings for UL, which is zero for domestic government debt in domestic currency.
• Provisions should be based on EL. Some banks already had non-zero PDs, while others had zero, leading to no provisions.
• The EBA proposed a consistent approach using credit ratings as a starting point, with notch downgrades applied based on market conditions as of 1 June 2011.

Notch Downgrades

• AAA / Aaa: no downgrading
• AA / Aa2 to A- / A3: two notch downgrades

Benchmark 2: Cost of Funding

• The macroeconomic scenario implied an increase in interest rates, which affected both assets and liabilities.
• Retail funding costs were expected to rise, and the EBA emphasized that any assumptions of stability were to be challenged.
• Official funding costs should increase by at least 75 basis points in 2011 and remain at that level during the stress test.
• Interest income on defaulted assets should not be computed under the static balance sheet assumption.

Benchmark 3: Haircuts to Sovereign Exposures in the Trading Book

• The EBA updated haircuts for sovereign exposures in the trading book to reflect recent market developments.
• Haircuts were applied based on sovereign credit spread widening up to 24 May 2011.
• Updated haircuts are provided in Table 3, which replaces the original version in the Methodological note.

Sovereign Haircuts (in basis points)

• Austria: 0.2% (3M), 0.5% (1Y), 1.1% (2Y), 1.9% (3Y), 3.4% (5Y), 5.5% (10Y), 8.4% (15Y)
• Belgium: 0.2% (3M), 1.2% (1Y), 2.1% (2Y), 3.7% (3Y), 5.9% (5Y), 9.8% (10Y), 15.3% (15Y)
• Bulgaria: 0.3% (3M), 1.1% (1Y), 2.1% (2Y), 3.3% (3Y), 5.4% (5Y), 8.7% (10Y)
• Cyprus: 0.4% (3M), 2.1% (1Y), 3.4% (2Y), 5.0% (3Y), 7.7% (5Y), 12.3% (10Y)
• Czech Republic: 0.2% (3M), 0.5% (1Y), 1.2% (2Y), 2.3% (3Y), 3.2% (5Y), 5.8% (10Y), 11.1% (15Y)
• Denmark: 0.1% (3M), 0.5% (1Y), 1.0% (2Y), 1.6% (3Y), 2.6% (5Y), 6.3% (10Y), 9.9% (15Y)
• Finland: 0.1% (3M), 0.4% (1Y), 0.9% (2Y), 1.9% (3Y), 2.7% (5Y), 4.2% (10Y)
• France: 0.2% (3M), 0.8% (1Y), 1.6% (2Y), 2.6% (3Y), 4.1% (5Y), 7.3% (10Y), 13.1% (15Y)
• Germany: 0.1% (3M), 0.5% (1Y), 0.9% (2Y), 1.3% (3Y), 2.1% (5Y), 3.5% (10Y), 6.2% (15Y)
• Greece: 0.5% (3M), 5.1% (1Y), 25.6% (2Y), 29.5% (3Y), 18.1% (5Y), 25.2% (10Y), 26.2% (15Y)
• Hungary: 0.2% (3M), 0.9% (1Y), 1.8% (2Y), 2.9% (3Y), 5.1% (5Y), 8.0% (10Y)
• Ireland: 0.9% (3M), 3.8% (1Y), 12.7% (2Y), 16.9% (3Y), 19.4% (5Y), 19.1% (10Y), 22.7% (15Y)
• Italy: 0.3% (3M), 1.5% (1Y), 3.0% (2Y), 5.0% (3Y), 8.4% (5Y), 13.1% (10Y), 20.1% (15Y)
• Latvia: 0.2% (3M), 0.8% (1Y), 1.8% (2Y), 2.8% (3Y), 4.3% (5Y)
• Lithuania: 0.2% (3M), 1.2% (1Y), 1.8% (2Y), 3.1% (3Y), 4.8% (5Y)
• Luxembourg: 0.0% (3M), 3.1% (1Y), 5.6% (5Y), 9.3% (10Y)
• Malta: 0.3% (3M), 1.3% (1Y), 2.6% (2Y), 4.8% (3Y), 7.4% (5Y), 13.2% (10Y)
• Netherlands: 0.1% (3M), 0.6% (1Y), 1.2% (2Y), 2.1% (3Y), 3.2% (5Y), 5.2% (10Y), 9.5% (15Y)
• Poland: 0.1% (3M), 0.6% (1Y), 1.2% (2Y), 1.7% (3Y), 2.8% (5Y), 5.0% (10Y), 7.7% (15Y)
• Portugal: 0.5% (3M), 6.4% (1Y), 12.6% (2Y), 18.9% (3Y), 21.7% (5Y), 22.3% (10Y), 33.2% (15Y)
• Romania: 0.3% (3M), 1.0% (1Y), 2.0% (2Y), 3.4% (3Y), 8.7% (10Y)
• Slovakia: 0.2% (3M), 0.8% (1Y), 1.4% (2Y), 2.2% (3Y), 3.5% (5Y), 5.9% (10Y), 8.0% (15Y)
• Slovenia: 0.2% (3M), 0.8% (1Y), 1.4% (2Y), 2.3% (3Y), 3.7% (5Y), 6.0% (10Y)
• Spain: 0.5% (3M), 1.6% (1Y), 3.4% (2Y), 5.5% (3Y), 9.0% (5Y), 14.6% (10Y), 23.2% (15Y)
• Sweden: 0.1% (3M), 0.3% (1Y), 0.6% (2Y), 1.2% (3Y), 1.9% (5Y), 3.1% (10Y), 6.5% (15Y)
• United Kingdom: 0.2% (3M), 1.1% (1Y), 1.9% (2Y), 3.1% (3Y), 4.7% (5Y), 7.6% (10Y), 14.1% (15Y)
• Iceland: 0.0% (3M), 0.3% (1Y), 1.0% (2Y), 1.6% (3Y), 3.6% (5Y), 5.2% (10Y), 7.4% (15Y)
• Norway: 0.1% (3M), 0.4% (1Y), 0.8% (2Y), 1.1% (3Y), 1.5% (5Y), 3.3% (10Y)

Benchmark 4: Trading Book RWAs in the Context of CRD 3

• The EBA requested banks to forecast the increase in trading book RWA due to the introduction of CRD 3.
• A scaling factor of 40% was applied to 2010 figures for banks unable to provide internal forecasts.
• Most banks provided their own internal forecasts, but significant differences were found between these and the scaling factor.
• The EBA expects consistency in the treatment of RWAs and encourages banks to use their internal models where possible.

Additional Guidance

• The EBA provides clarifications and re-statements of the existing methodology to ensure uniform application.
• The average implied PDs are based on 2-year cumulative default rates from Fitch, Moody's, and S&P.
• The LGD is set at 40% based on the EBA's review of banks' own estimates.
• Provisions should be built up by 50% in 2011 and another 50% in 2012.
• The EBA encourages the use of benchmarks as a floor in the absence of compelling evidence to the contrary.

Conclusion

The 2011 EU-wide stress test aims to ensure a common and conservative benchmark for European banks. It includes four key benchmarks covering sovereign and financial institution exposure, cost of funding, trading book haircuts, and trading book RWAs. The EBA emphasizes consistency, clarity, and quality assurance to ensure that all banks apply the methodology correctly and transparently. The guidance provided is intended to support the process and not to replace it, ensuring that the stress test remains a supervisory tool to enhance the resilience of the EU banking sector.