EBA欧洲银行-11-Ignatowski-Korte-Slides_31页_1mb

Summary: Tightening Bank Resolution Regimes and Bank Risk-Taking

Core Content

This paper examines the impact of tightening bank resolution regimes on bank risk-taking behavior, focusing on the introduction of the Orderly Liquidation Authority (OLA) in the U.S. as a case study. The authors investigate whether the implementation of a more robust resolution framework leads to a decrease in risk-taking by financial institutions, particularly Bank Holding Companies (BHCs), and whether this effect varies with bank size and resolution credibility.

Main Hypotheses

• Main Hypothesis: Affected banks (those with a high share of non-FDIA-regulated assets) will alter their behavior towards less risk-taking and safer business models after the introduction of the OLA.
• Extended Hypothesis: If the application of the new resolution regime is not credible due to bank-specific characteristics (e.g., size), the effect on risk-taking will be lower or non-existent.

Theoretical Model

The paper presents a theoretical model that views the choice between bank closure and bailout as a trade-off between liquidity and discipline:

• Option 1: Resolution → increases discipline but reduces liquidity
• Option 2: Bailout → increases liquidity but reduces discipline

The time discount rate of the regulator is crucial in determining the optimal resolution strategy, with liquidity effects being short-run and discipline effects long-run. Improvements in resolution technology (e.g., OLA) change the level of this trade-off, potentially leading to reduced risk-taking.

Identification Strategy

The authors use a difference-in-differences (DiD) methodology to analyze the effect of the OLA as a quasi-natural experiment. The setup includes:

• Treatment: The introduction of the OLA in July 2010
• Control Group: Banks not significantly affected by the OLA
• Timing: Pre-OLA (before 2010) and post-OLA (after 2010) periods

The treatment group consists of BHCs and their associated banks with a high share of non-FDIA-regulated assets, which are most affected by the change in resolution regime.

Key Findings

Baseline Results

• Affected banks show a significant decline in risk measures (e.g., z-score, asset risk) after the introduction of the OLA, compared to non-affected banks.
• The effect is robust across different levels of aggregation (bank and BHC level) and different proxies (treatment dummy and continuous treatment intensity).

Robustness Tests

• Placebo treatment: No significant difference in risk measures between treatment and control groups, supporting the validity of the DiD approach.
• Sample attrition: Results remain consistent even when excluding failed or exited banks.
• Solvency constraint: The effect is more pronounced when the solvency constraint is more binding.
• Alternative explanations (e.g., Volcker Rule, stress tests): The effect of the OLA is not overshadowed by these other regulatory actions.

Business Model and Investment Changes

• Affected banks significantly decrease risky activities, including:
  • Loan-to-income ratios of new mortgage loans
  • Probability of loan approval for riskier loan applications
  • Total number of loan applications (especially for higher-risk loan-to-income ranges)

Credibility of the Resolution Threat

• The credibility of the OLA varies with bank size:
  • Larger banks (e.g., those with assets over USD 50 billion) show less significant reductions in risk measures, suggesting the resolution threat may not be as credible for them.
  • The triple interaction term (affected bank x after OLA x total assets) indicates that the risk measures may increase with total assets for affected banks, suggesting diminishing effectiveness of the resolution regime for larger institutions.

Policy Implications

• The introduction of the OLA is a technological improvement in the U.S. resolution regime, offering legal and financial empowerment to regulators.
• The results suggest that tightening resolution regimes can lead to reduced risk-taking by banks, especially those with a high share of non-FDIA-regulated assets.
• However, larger banks may not respond as strongly due to lower credibility of the resolution threat, which raises questions about the effectiveness of resolution regimes for "too-big-to-fail" institutions.

Conclusion

The paper concludes that the OLA significantly reduces risk-taking by banks, particularly those with a high exposure to non-FDIA-regulated assets. The findings support the effectiveness of resolution regimes in promoting disciplined behavior, but also highlight the importance of credibility in ensuring the desired impact, especially for larger banks.