About Me
I am a PhD candidate in my 4th year at Frankfurt School of Finance & Management. I have a Master’s degree in Finance from Frankfurt School and a Bachelor of Science in Economics from Justus-Liebig-University in Giessen. My interests are in the field of financial intermediation and the use of AI to generate new data.
Research Interests
CovenantAI – New Insights into Covenant Violations (with Anthony Saunders and Sascha Steffen) [Paper]
Best PhD Presentation Award, Generative AI Conference Concordia University
Conferences: Sydney Banking and Financial Stability Conference, 38th Australasian Finance and Banking Conference, 14th MoFiR Workshop on Banking in Singapore, 2nd Workshop on the Latest Advances in NLP and GenerativeAI in Finance and Management, 13th EFI Workshop in Brussels, BIS-CEPR-Gerzensee-SFI Conference on Financial Intermediation 2025, 1st Lapland Financial Institutions Summit, Finance Research Seminar Krakow University of Economics, 3rd Bonn/Mannheim Workshop on Digital Finance, Generative AI in Finance Conference in Montreal, Harvard-Wharton Insolvency and Restructuring Conference, 8th Household Finance Workshop, 6th Future of Financial Information Conference, UCLA Conference on Financial Markets, Harvard Corporate Restructuring & Insolvency Seminar, 2023 NBER Big Data and Securities Markets Conference, Bonn-Frankfurt-Mannheim PhD Conference 2023, 11th Workshop Banks and Financial Markets in Halle, 4th Vaasa Banking Research Workshop and New (and Old) challenges to Financial Intermediaries Conference in Bayes Business School
Abstract
We introduce CovenantAI, an advanced AI-based approach to accurately identify loan covenant violations from SEC filings. CovenantAI outperforms traditional keyword-based and Dealscan approaches by precisely classifying complex renegotiation outcomes such as amendments, waivers, and technical defaults. Our analyses validate CovenantAI’s higher accuracy and consistency against existing methods. By capturing nuanced creditor-borrower interactions, CovenantAI enables new research into the economic implications of covenant resolution outcomes and investor behaviors surrounding violations. This novel dataset thus significantly enhances empirical research capabilities in corporate finance by providing comprehensive, granular, and reliable data on covenant violations and resolutions.
Do Institutional Investors Trade on Covenant Violations? (with Anthony Saunders and Sascha Steffen) [Paper]
Conferences (* scheduled): 2026 FIRS Conference (*), SFS Cavalcade North America 2026 (*), Sydney Banking and Financial Stability Conference, 38th Australasian Finance and Banking Conference, Brownbag University of Technology Sydney, Rising Scholar Conference in Finance 2025, Bonn-Frankfurt-Mannheim PhD Conference 2024
Abstract
We develop CovenantAI, an artificial intelligence-powered covenant monitoring methodology, to examine whether institutional investors strategically trade around covenant violations in leveraged loan markets. We document a persistent decline in loan prices during the 100 days preceding violations, with a most pronounced drop 80 days prior to the violation: for example, we find cumulative abnormal returns of -6.39% during the [-80,-60] event window. Price effects are most severe for loans amended post-violation or those that remain in technical default. Covenant violations significantly increase the likelihood that the firm’s credit rating is downgraded or the company enters bankruptcy, particularly among non-investment-grade loans held by Collateralized Loan Obligations (CLOs). We document substantial cross-sectional heterogeneity in CLO constraints driven by overcollateralization ratios and CCC-rated loan holdings. Loans predominantly owned by constrained CLOs exhibit steeper pre-violation price declines and significantly more negative abnormal returns. Our evidence demonstrates that constrained institutional investors preemptively divest loan positions in anticipation of covenant violations, with trading intensity reflecting both violation severity and investor-specific portfolio constraints.