We offer early evidence on the impact of negative interest rate policy (NIRP) on banks’ risk-taking. Our primary result shows banks in NIRP-adopter countries reduce holdings of risky assets by around 10 percentage points following implementation of NIRP in comparison to banks in non-adopter countries. We augment this result by identifying NIRP’s impact on other aspects of banks’ risk-taking behaviour; NIRP is associated with reductions in banks’ loan growth and average loan price (by 3.7 percentage points and 59 basis points) and a rebalancing of asset portfolios towards safer assets. Secondly, we find the NIRP-effect is heterogeneous; post-NIRP risk-taking increases at strongly capitalised banks and at banks operating in less competitive markets that exploit market power to insulate net interest margins and profitability. Our robust empirical evidence supports the “de-leverage” hypothesis which suggests that banks acquire safer, liquid assets to bolster their capital positions rather than searching for value by acquiring riskier assets. We base our evidence on a sample of 2,584 banks from 33 OECD countries across 2012 to 2016, and from models that employ a difference-in-differences framework.

我们提供了关于负利率政策 (NIRP) 对银行风险承担影响的早期证据。我们的主要结果显示，与未采用 NIRP 的国家的银行相比，采用 NIRP 的国家的银行在实施 NIRP 后将风险资产的持有量减少了约 10 个百分点。我们通过确定 NIRP 对银行冒险行为其他方面的影响来增强这一结果；NIRP 与银行贷款增长和平均贷款价格的下降（分别下降 3.7 个百分点和 59 个基点）以及资产组合向更安全资产的重新平衡有关。其次，我们发现 NIRP 效应是异质的；在资本充足的银行和在竞争较弱的市场中运营的银行，利用市场力量来隔离净息差和盈利能力的 NIRP 后风险承担增加。我们强有力的经验证据支持“去杠杆化”假设，该假设表明银行获取更安全、流动性强的资产以增强其资本头寸，而不是通过获取风险更高的资产来寻找价值。我们的证据基于 2012 年至 2016 年来自 33 个经合组织国家的 2,584 家银行的样本，以及采用差异中差异框架的模型。