[1] Beyond the headline: Measuring Monetary Policy Uncertainty from Bank Earnings Calls

Solo-authored

The role of financial institutions in the impact of monetary policy uncertainty on the economy is not fully understood. I construct an index of bank-level monetary policy uncertainty from U.S. bank earnings calls since 2002 and validate the measure with its correlation with past interest rate forecast errors and aggregate disagreement in the Survey of Professional Forecasters. SVAR evidence reveals that monetary policy uncertainty lowers real GDP and increases credit spreads. Looking at the cross section, banks with high uncertainty charge higher interest rates in syndicated loans. The findings stress that banks beliefs impacts both lending conditions and business cycle fluctuations.

Presentations (* for scheduled): IFABS 2025 | Ghent Empirical Macro Workshop | EFMA 2025 | 2nd Frankfurt Summer School  | Wolpertinger Annual Conference | FDIC Annual Bank Research Conference*

Links: SSRN (last update September, 2025)

[2] Do actions follow words? How bank sentiment predicts credit growth (R&R at Journal of Financial Intermediation)

With Martien Lamers

This paper constructs a bank sentiment index using earnings call transcripts of US banks from 2001 to 2023 to analyze its impact on future credit growth. A one standard deviation increase in sentiment leads to a 1.908% rise in credit growth over the next year, lasting up to three years. Optimistic banks overextend credit and take on excessive risk, leading to lower future profitability and higher future provisioning. Using syndicated loans, we find sentiment affects both credit supply and loan pricing. Our results highlight the importance of bank sentiment as a regulatory early-warning indicator and risk-monitoring tool for investors.

Presentations (* for scheduled): 49th Eastern Economic Association Conference | 16th Annual Conference of the Behavioural Finance Working Group | 5th Behavioral Macroeconomics Workshop | Belgian Financial Forum

Links: SSRN 

[3] Narrative Forecasts (R&R at Journal of Economic Psychology)

With Yuting Chen, Maurizio Montone, Valerio Potì

We propose a novel methodology to identify managerial beliefs from earnings call transcripts using lexicon-based sentiment analysis with machine-learning guided topic modeling. Unlike traditional survey-based approaches, our method extracts managerial beliefs with near-universal coverage. We show that managerial sentiment significantly predicts analyst forecast revisions, with presentation sentiment showing stronger effects than question and answer (Q&A) interactions. The influence of sentiment varies by topic salience and media attention, suggesting that narrative content shapes analyst expectations beyond fundamental information. Our approach offers a scalable alternative to surveys for measuring economic beliefs, providing high-frequency, comprehensive coverage of managerial perspectives across firms and time.

Work in Progress