2013年-IMF国际货币组织全球_Risk_Exposures_and_Financial_Spillovers_in_Tranquil_and_Crisis_Times_Bank_46页_1mb

Summary of "Risk Exposures and Financial Spillovers in Tranquil and Crisis Times: Bank-Level Evidence"

Core Content

This paper investigates the evolution of bank-level risk exposures to global and country-specific factors, as well as financial spillovers between banks, using data from 83 financial institutions across 21 countries from 2003 to 2011. The study addresses three key questions: (1) how have banks' stock returns been affected by global and country-level risk factors over time? (2) What bank-specific characteristics influence these risk exposures? (3) Are there significant spillover effects between banks beyond common risk factors?

Main Findings

1. Evolution of Risk Exposures

• Global and Country Risk Sensitivity: Bank sensitivities (betas) to both global and country risk factors increased significantly in 2011, although still below the levels observed during the 2008–09 subprime crisis.
• European Risk Sensitivity: The average sensitivity of European banks to European risk factors has steadily increased since 2008, reaching historical highs in 2011.
• Regional Variations: French and German banks showed higher sensitivity to global and country risk factors in 2011 compared to U.K., U.S., and rest-of-world banks.
• Volatility Decomposition: The share of stock return volatility explained by global risk factors rose from around 30% in 2003 to about 50% by the end of 2011. Country risk factors accounted for roughly 20% of volatility throughout the period. The idiosyncratic component accounted for the remaining 30%.

2. Bank-Specific Characteristics Influencing Risk Exposure

• Wholesale Funding Reliance: Banks with higher dependence on wholesale funding are more vulnerable to global and country shocks.
• Capital Levels: Lower capital ratios are associated with higher risk exposure, particularly in crisis times.
• Profitability: Lower profitability is linked to higher risk exposure, suggesting that less profitable banks are more sensitive to market fluctuations.
• Size: Initially, larger banks were less sensitive to global shocks due to diversification and "too-big-to-fail" perceptions. However, during the 2011 EA crisis, larger banks were perceived as more vulnerable.
• Exposure to Crisis Countries: Banks with higher exposure to European crisis countries (e.g., Greece, Italy, Spain) showed greater sensitivity to regional risk factors.

3. Financial Spillovers

• Regional Linkages: Spillovers between European banks are strong and region-specific. German banks are more connected to other European banks, particularly French and U.K. banks. French banks are also closely linked to U.K. and Southern European banks.
• Non-European Banks: There is limited evidence of spillovers between European and non-European banks, except for specific U.S. institutions.
• U.K. Banks: U.K. banks show stronger spillover connections with both Asia (pre-crisis) and the U.S. (in both periods).
• Control Variables: Including control variables for financial conditions does not significantly alter the results, suggesting that the spillover effects reflect actual risk transmission rather than shared macroeconomic sensitivities.

Key Insights

• The paper emphasizes the importance of time-varying risk exposures, which reflect structural changes and crisis-induced spillovers.
• Globalization of the EA crisis is mainly channeled through U.K. and U.S. banks, with limited direct spillover effects to other regions.
• Factor models are used to decompose bank stock returns into common and idiosyncratic components, with the world-local CAPM framework allowing for partial capital market integration.
• Bank betas (sensitivities to risk factors) are influenced by several characteristics, including capital, profitability, size, and exposure to crisis countries.
• The regression results suggest that wholesale funding reliance, low capital ratios, and low valuations are key indicators of vulnerability to global and country risk factors.

Policy Implications

• The increasing sensitivity of European banks to global and country risk factors underscores the importance of monitoring and managing cross-border financial risks.
• Systemic risk in the EA is more pronounced among larger banks, which may be more vulnerable due to their exposure to peripheral European economies.
• The findings support the idea that financial integration has increased, leading to more interconnected banking systems, particularly within Europe.
• Regulatory and supervisory frameworks should consider the evolving nature of risk exposures and the potential for spillovers, especially during periods of financial stress.

Methodology

• The study uses a factor model approach to estimate bank betas for global and country risk factors.
• The model is re-estimated every six months to allow for time-varying factor loadings.
• Country-specific factors are derived by orthogonalizing local market returns against the global market.
• Spillover effects are tested using alternative models and control variables to ensure they reflect true contagion rather than shared macroeconomic influences.

Conclusion

The paper highlights the increasing interconnectedness of the financial sector, particularly in Europe, and the evolving risk exposures of banks over time. It also identifies key characteristics that influence these exposures and provides insights into the mechanisms of financial contagion during both tranquil and crisis periods. The results suggest that while the EA crisis had a significant impact on European banks, the effects on other regions were more limited, with U.K. and U.S. banks playing a central role in the transmission of risk.