Interest Rate Hedging Strategies of FTSE MIB Non-Financial Firms: A Comparative Analysis Across Rising and Declining Rate Environments

This work combines innovative AI-driven techniques for data extraction with traditional economic analysis, highlighting the complexity and effort required in building a reliable dataset as the foundation of the research. The study builds on the work of G. Bison, L. Pelizzon, D. Sartore - “La Copertura dei rischi finanziari nelle imprese non finanziarie italiane attraverso gli strumenti derivati”- , extending and updating the analysis by relating it to interest rates fluctuations while integrating recent data and advanced methodologies. The research investigates whether non-financial companies listed on the FTSE MIB intensify their use of hedging strategies during periods of rising or falling interest rates. To address this question, a time frame was selected that encompasses three distinct phases of monetary policy: expansionary (2021), transitional (2022), and restrictive (2023). Data from the financial statements of the selected firms were gathered through a structured methodology based on Large Language Models (LLMs), implemented within the PTCF framework and supported by the chain-of-thought technique. This approach enabled systematic extraction and processing of information, ensuring consistency and replicability despite the inherent complexity of the task. The dataset was then compared with market expectations on interest rates, measured by the daily average spread between the 3-year OIS and the ECB Deposit Facility Rate. This indicator captured both the official policy stance and the expectations embedded in market pricing. The empirical analysis, conducted using Python and statistical techniques, identified two significant Pearson correlations between the number of hedging operations and rate expectations, as well as a stronger correlation between the latter and the total amount of debt hedged. Finally, the thesis also examines endogenous determinants, showing through a logistic regression that firm size and floating-rate exposure significantly increase the likelihood of hedging, while leverage and profitability do not appear to be decisive factors.