2012年-BIS国际清算银行_Loan_loss_provisioning_practices_of_Asian_banks_31页_236kb

Loan Loss Provisioning Practices of Asian Banks: Summary

Core Content

This paper by Frank Packer and Haibin Zhu examines the loan loss provisioning (LLP) practices of Asian banks in the context of financial system procyclicality, which refers to the tendency of banks to increase provisions during economic downturns and decrease them during upturns. The study focuses on the period from 2000 to 2009 and analyzes data from 240 banks across 12 Asian economies.

Main Findings

• Countercyclical Provisioning Dominates: Overall, there is mixed evidence of countercyclical LLP in Asia, but countercyclical practices are more pronounced in emerging Asia, especially in India.
• Procyclical Provisioning in Japan: Japanese banks show procyclical LLP, which is largely attributed to higher provisions during the 2007-09 financial crisis.
• Regulatory Influence: Many Asian jurisdictions have implemented stricter LLP requirements post-Asian financial crisis, aiming to ensure timely and adequate reserves for risk management.
• International Accounting Standards (IAS 39): The adoption of IAS 39 has influenced LLP practices in several countries, leading to more forward-looking provisions.
• Earnings Smoothing: There is evidence that LLP is used to smooth earnings, which can contribute to financial system procyclicality. However, this is less common in emerging markets compared to industrialized countries.
• Macroeconomic and Credit Quality Factors: LLP is influenced by macroeconomic conditions (GDP growth), credit quality (NPL ratio, loan/asset ratio, loan growth), and capital adequacy.

Key Information

Regulatory Changes Post-Asian Financial Crisis

• China: Required general reserves of at least 1% of loans since 2005. In 2009, the CBRC encouraged banks to maintain a reserve-to-NPL ratio of 150%.
• Hong Kong SAR: Implemented IAS 39 in 2005. Maintains a regulatory reserve of 0.5-1% of total loans to cover future credit losses.
• India: Increased general provision levels for standard loans from 0.25% to 0.40% in 2005. Further raised requirements for certain sectors in 2006 and 2007.
• Indonesia: Adopted a five-grade loan classification system in 1998 and tightened definitions in 2005. General provisions of at least 1% are required for loans without heightened credit risk.
• Japan: Has long had general and specific provisions. Required provisions are based on past three-year loss experience. No discretionary changes in LLP requirements have been made in response to macrofinancial conditions.
• Korea: Tightened LLP norms multiple times, increasing general reserve requirements for corporate loans. Incorporated expected loss considerations into provisioning guidance.
• Malaysia: Increased reserve requirements for prudential loan grades post-Asian crisis. Planned to implement IAS 39 by 2010.
• Philippines: Adopted IAS 39 in 2005. Maintains higher reserve levels based on IAS 39 or BSP guidelines.
• Singapore: Implemented IAS 39 in 2005. Maintains a minimum general provision of 1% of loans net of collateral.
• Thailand: Increased minimum LLP requirements in 1998. Further tightened provisions in 2006 and 2007 to align with IAS 39. Not fully applied to performing loans yet.

Methodology

• The paper uses a baseline econometric model (Equation 1) to estimate the determinants of LLP.
• The model includes variables such as GDP growth, bank earnings, NPL ratios, loan/asset ratios, and capital adequacy ratios.
• The coefficients on GDP growth and earnings are interpreted as indicators of countercyclical or procyclical behavior.
• The paper distinguishes between micro-oriented (bank-specific) and macro-oriented (system-wide) factors influencing LLP.

Empirical Analysis

• Mixed Evidence: While LLP is negatively related to economic growth, there is also evidence of earnings smoothing.
• Country-Specific Differences: Significant differences exist in LLP behavior across countries, with India and other emerging markets showing more countercyclical tendencies.
• Data Cleaning: The authors cleaned the data by removing banks with outlier observations and those with incomplete financial records.

Conclusion

• The study suggests that while LLP practices in Asia have become more conservative, they have not uniformly become countercyclical.
• The regulatory environment has played a crucial role in shaping LLP practices, with some countries adopting more forward-looking approaches.
• The paper highlights the importance of analyzing LLP in the context of macroprudential policy and financial stability.