国际清算银行-监管政策刺激：来自欧元区股息建议的证据（英）-2023.3-57页_866kb

BIS Working Paper Summary: "Supervisory Policy Stimulus: Evidence from the Euro Area Dividend Recommendation"

Background & Methodology

• Objective: To assess the effectiveness of the European Central Bank's (ECB) 2020 dividend recommendation in supporting lending during the Covid-19 crisis.
• Method: Used a quasi-natural experiment approach exploiting differential compliance with the ECB's dividend recommendation across euro area banks (significant institutions). The key variable was the ratio of planned but non-distributed dividends to Risk-Weighted Assets (RWA) (Dividends/RWA). Estimated its impact on bank lending growth using granular euro area credit registry data (AnaCredit) and bank-level data from ECB supervision. Controlled for other monetary (like TLTRO III) and fiscal support policies (like loan guarantees, moratoria) using firm and bank fixed effects.
• Sample: Covered 99 large banks supervised directly by the ECB. Analyzed bank-firm relationships.

Main Findings

• Overall Impact: The dividend recommendation had a statistically significant positive effect on bank lending. A 1 percentage point increase in the Dividends/RWA ratio was associated with approximately 4.3-4.4 percentage points higher growth in the stock of corporate debt (to non-financial corporations), representing a substantial capital multiplier effect (€141 billion potential lending increase per €11.8 billion dividends retained in 2020).
• Targeted Effect: The impact was stronger for lending to Small and Medium Enterprises (SMEs) compared to large firms and more pronounced for firms operating in Covid-19 vulnerable sectors (e.g., retail, accommodation).
• No Increased Risk-Taking: The study found no significant increase in lending to risky borrowers, "zombie" firms (those heavily impaired but operating due to creditor forbearance), or increased risk-taking by banks with higher existing NPL ratios.
• Interaction with Capital Space: The recommendation's effect was concentrated among banks with higher capital buffers (CET1 MDA Distance). Banks with less capital headroom did not expand lending more, suggesting capital constraints remained binding.
• Temporary Nature: The positive lending impact was significant shortly after the recommendation (e.g., 2020Q3) but faded by later 2020 (e.g., 2020Q4), consistent with the temporary nature of the policy.
• Synergy with Fiscal Policy: The recommendation amplified the effects of government fiscal support measures like loan guarantees.

Policy Implications

• The dividend restriction was deemed an effective policy tool to boost lending during the crisis, particularly for SMEs and vulnerable sectors, without encouraging excessive risk-taking.
• It provided more capital space for banks to issue loans without breaching regulatory requirements, achieving a countercyclical goal.
• From a welfare perspective, restricting dividends without loosening capital requirements is preferable, as capital release mechanically reduces banks' loss absorption capacity, while dividend restriction only postpones payouts.
• Synergies with fiscal guarantees are crucial. Further research is needed on how dividend restrictions interact optimally with other prudential and macroprudential tools.