Financial institutions

Unlocking supervisory effectiveness in credit risk: insights from the ECB’s supervisory findings and measures

11/09/2026 European Central Bank Central banks

This paper provides an assessment of supervisory effectiveness in credit risk using confidential data from ECB Banking Supervision. The analysis covers 8,106 supervisory measures issued and closed between 2016 and 2025. A central contribution is the construction of a measure level indicator of supervisory effectiveness based on large language model (LLM) analysis of supervisory narratives and closure justifications. The indicator captures whether supervisory measures resulted in full remediation of the underlying deficiencies.

Unlocking supervisory effectiveness in credit risk: insights from the ECB’s supervisory findings and measures

11/09/2026 European Central Bank Central banks

This paper provides an assessment of supervisory effectiveness in credit risk using confidential data from ECB Banking Supervision. The analysis covers 8,106 supervisory measures issued and closed between 2016 and 2025. A central contribution is the construction of a measure level indicator of supervisory effectiveness based on large language model (LLM) analysis of supervisory narratives and closure justifications. The indicator captures whether supervisory measures resulted in full remediation of the underlying deficiencies.

From recognition to integration: the ECB’s progress on climate and nature-related risks and the role of its climate change centre

11/09/2026 European Central Bank Central banks

The ECB climate change centre (CCC) was established in 2021 to design and steer the ECB’s response to climate-related risks. During its first five years, the centre has evolved from a dedicated coordination hub into a facilitator of the integration of climate and nature-related risks into the ECB’s overall analytical and risk framework. It has thereby supported the institution’s progression from “recognition to integration” in addressing climate and nature-related risks.

Deep-π*: a neural network estimate of trend inflation

11/04/2026 European Central Bank Central banks

Most empirical models of trend inflation treat it as a latent random walk driven by unobserved shocks, so they cannot say which economic conditions move it. Deep- π*makes the innovations to the trend a possibly nonlinear function of a large number of macroeconomic indicators, estimated with neural networks, so that Shapley values can be used to attribute every movement of the trend to the indicators behind it. Two complementary loss functions define a current trend, which fits inflation today, and a forward-looking trend, which best predicts inflation over the next year.

Deep-π*: a neural network estimate of trend inflation

11/04/2026 European Central Bank Central banks

Most empirical models of trend inflation treat it as a latent random walk driven by unobserved shocks, so they cannot say which economic conditions move it. Deep- π*makes the innovations to the trend a possibly nonlinear function of a large number of macroeconomic indicators, estimated with neural networks, so that Shapley values can be used to attribute every movement of the trend to the indicators behind it. Two complementary loss functions define a current trend, which fits inflation today, and a forward-looking trend, which best predicts inflation over the next year.

Elastic in cash, inelastic in repo: hedge funds in the treasury and repo markets

11/03/2026 European Central Bank Central banks

The functioning of sovereign bond markets is closely intertwined with repo markets, where investors finance long positions and source bonds for short sales. We show, theoretically and empirically, how conditions in the cash bond market fundamentally shape repo rates: When demand for cash bonds exceeds available supply, arbitrageurs, notably hedge funds, step in to bridge the gap by shorting bonds and borrowing them in the repo market. These arbitrage strategies drive a wedge between monetary policy rates and repo rates, generating “specialness”.

Debt maturity and government spending multipliers

11/03/2026 European Central Bank Central banks

Government spending effectiveness depends critically on how it is financed. Using state dependent SVAR models and local projections on post-war US data, we show that fiscal expansions financed with short-term debt generate significantly larger output multipliers than those financed with long-term debt. This difference mainly stems from private consumption responses: short-term financing crowds in consumption while long-term financing does not. Torationalize this finding, we construct an incomplete markets model in which households invest in short-term and long-term assets.

Pages

Subscribe to Financial institutions