Knowledge that Transforms

This paper examines whether climate risk affects the dynamics of banks’ regulatory capital adjustments, based on a large panel of European banks over the 2006–2021 period. Using a dynamic partial adjustment model, we find that climate-exposed banks hold higher capital adequacy ratios and adjust faster toward their optimal capital structure, particularly when exposed to transition risk and post-COP21. Climate risk also induces asymmetric adjustment behaviours. Deleveraging occurs through risk-weighted asset reallocation toward safer exposures, without asset liquidation or lending cuts. While leveraging operates through risk-weighted asset expansion, without reducing equity growth. However, pre-COP21, deleveraging is primarily achieved through lending contraction, whereas leveraging relies mainly on asset expansion. Our findings highlight the policy relevance of climate risk for prudential supervision and bank capital regulation.

This paper applies the Temporal Fusion Transformer (TFT) model to learn dynamic time-series patterns in mutual fund performance and to assess whether these patterns predict future alpha. I summarize the model’s cross-sectional ranking power using diagnostic portfolio spreads: a top-minus-bottom decile exhibits an annualized Carhart four-factor alpha spread of 2.8%, with dispersion persisting for up to four years. In panel regressions controlling for standard predictors and fund and time fixed effects, TFT forecasts improve explanatory power by more than 25% in adjusted R 2 . Leveraging TFT’s interpretable outputs, I show that historical fund returns receive the largest weight (about 29%), their importance displays earnings-cycle seasonality, and attention to past observations rises by 46% during crisis periods. Using fund-by-month variable-importance weights, I define fund-specific informativeness states and construct conditional skill measures that predict and persist precisely when the same signal becomes informative again, beyond coarse macro conditioning. Together, these results provide an alternative explanation for why unconditional performance persistence appears weak: skill is episodic and becomes visible when a manager’s key signals regain relevance. • TFT deep learning model predicts mutual fund alpha from time-series patterns. • Top-minus-bottom decile spread yields 2.8% annualized four-factor alpha. • Model attention to past observations rises 46% during crisis periods. • Conditional skill measures persist when key signals regain informativeness. • Episodic skill explains why unconditional persistence appears weak.

We examine how ETF adoption affects retail investor performance using a comprehensive panel of 524,181 Finnish investors tracked from 2007 to 2022. ETF adopters tend to be older, predominantly male, and more active, holding smaller but better-diversified portfolios. Importantly, first-time use of ETFs yields statistically significant improvements in risk-adjusted returns. ETF users also demonstrate greater portfolio resilience during financial crises. Our findings confirm that ETFs serve as effective tools for enhancing performance and managing risk for retail investors.

Machine learning (ML) could strengthen banks’ resilience through improved credit risk screening and ultimately benefit financial stability. Yet, ML adoption in banking remains limited, with simpler linear models still predominating. We argue that the emergence of highly flexible ML models has created a new challenge for forecasting tasks: ‘model multiplicity’—where equally accurate ML models at the aggregate level produce divergent individual-level predictions (‘predictive multiplicity’) or differ in their decision surfaces (‘procedural multiplicity’). These issues raise fundamental questions: Why should an individual or firm be subject to an adverse credit risk model outcome when there is an equally accurate model that treats them more favorably? Using the world’s largest loss database of corporate defaults, we examine these two phenomena in recovery rate ( RR ) modeling and propose heterogeneous ML ensembles as a natural solution. By combining predictions and decision surfaces from multiple well-performing ML models, ensembles mitigate risks associated with predictive multiplicity by ensuring that borrowers are not subject to the fluctuations of a single model, and reduce procedural multiplicity by providing a robust measure of features that ultimately improve out-of-sample RR predictions. By addressing the ‘multiplicity of good models’ problem, our study emphasizes the importance of model stability and provides new insights for the future development of ML models.