Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Back to the Futures: A Machine Learning Analysis of European Hedging and Cost of Capital

Welléus, Melker LU and Murat, Jacob LU (2025) BUSN79 20251
Department of Business Administration
Abstract
This thesis investigates whether firms that hedge, specifically against interest rate, currency, or commodity risks, experience a lower cost of capital compared to non-hedging firms. It further explores the firm-specific conditions under which hedging is most effective in reducing financing costs. We develop a custom Python-based model that integrates regular expressions, Natural Language Processing, and BART zero-shot classification to extract hedging disclosures from over 2,400 annual reports of European firms. These disclosures form the basis of a panel dataset covering 2,022 firm-year observations of non-financial STOXX600 constituents from 2019 to 2024. The data is analysed using pooled OLS regressions with industry, year, and country... (More)
This thesis investigates whether firms that hedge, specifically against interest rate, currency, or commodity risks, experience a lower cost of capital compared to non-hedging firms. It further explores the firm-specific conditions under which hedging is most effective in reducing financing costs. We develop a custom Python-based model that integrates regular expressions, Natural Language Processing, and BART zero-shot classification to extract hedging disclosures from over 2,400 annual reports of European firms. These disclosures form the basis of a panel dataset covering 2,022 firm-year observations of non-financial STOXX600 constituents from 2019 to 2024. The data is analysed using pooled OLS regressions with industry, year, and country fixed effects, and standard errors clustered at the firm level. The theoretical framework draws on financial distress theory, agency theory, information asymmetry, and the Modigliani-Miller irrelevance proposition to explain how hedging may lower the required return in imperfect capital markets. Empirical results show that interest rate and commodity hedging are significantly associated with lower cost of capital, while currency hedging has no overall effect unless conditioned on Eurozone status. Hedging appears particularly beneficial for firms experiencing greater financial distress, supporting the view that it serves a risk-reducing function under market frictions. (Less)
Please use this url to cite or link to this publication:
author
Welléus, Melker LU and Murat, Jacob LU
supervisor
organization
course
BUSN79 20251
year
type
H1 - Master's Degree (One Year)
subject
keywords
Hedging, Cost of Capital, Natural Language Processing, Corporate Risk Management, Europe
language
English
id
9192226
date added to LUP
2025-06-26 14:21:35
date last changed
2025-06-26 14:21:35
@misc{9192226,
  abstract     = {{This thesis investigates whether firms that hedge, specifically against interest rate, currency, or commodity risks, experience a lower cost of capital compared to non-hedging firms. It further explores the firm-specific conditions under which hedging is most effective in reducing financing costs. We develop a custom Python-based model that integrates regular expressions, Natural Language Processing, and BART zero-shot classification to extract hedging disclosures from over 2,400 annual reports of European firms. These disclosures form the basis of a panel dataset covering 2,022 firm-year observations of non-financial STOXX600 constituents from 2019 to 2024. The data is analysed using pooled OLS regressions with industry, year, and country fixed effects, and standard errors clustered at the firm level. The theoretical framework draws on financial distress theory, agency theory, information asymmetry, and the Modigliani-Miller irrelevance proposition to explain how hedging may lower the required return in imperfect capital markets. Empirical results show that interest rate and commodity hedging are significantly associated with lower cost of capital, while currency hedging has no overall effect unless conditioned on Eurozone status. Hedging appears particularly beneficial for firms experiencing greater financial distress, supporting the view that it serves a risk-reducing function under market frictions.}},
  author       = {{Welléus, Melker and Murat, Jacob}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Back to the Futures: A Machine Learning Analysis of European Hedging and Cost of Capital}},
  year         = {{2025}},
}