Electricity price volatility has become a major economic concern for European firms following recent energy-market disruptions and changes in power-market regulation. Sharp movements in electricity prices can affect firms’ costs, profitability, investment decisions and market valuations, yet there is still no widely used, timely, country-specific indicator that captures electricity-market risk and links it to firm-level outcomes. This creates an important gap for researchers, investors and policymakers seeking to understand how electricity-related uncertainty is transmitted through the wider economy.

The PhD project will see the appointed student work with the supervisory team to develop a new set of text-based electricity risk indicators for major European economies by applying large language models and natural language processing techniques to large collections of newspaper articles. The project will focus on leading newspapers in the UK, Germany, France, Italy and Spain, allowing the student to work with a rich multilingual dataset and to gain experience in building comparative cross-country measures of economic risk. The linguistic analysis of text is now widely used in Economics and Finance, but major challenges remain in detecting economically meaningful content across languages, classifying risk-related discussions accurately, and constructing interpretable and robust indicators from unstructured data.

An important contribution of this project is to address these challenges in the context of electricity markets, combining modern NLP tools with rigorous financial econometric methods. The project is particularly timely given the renewed geopolitical tensions affecting global energy markets, which have recently driven sharp movements in oil and gas prices and highlighted the vulnerability of European energy systems to external shocks. The student will assess whether the resulting electricity risk indices provide a meaningful measure of firms’ exposure to energy-related risk, and will then examine whether more exposed firms are valued differently by investors and earn different stock returns. The project therefore offers the opportunity to contribute to a highly topical research agenda at the intersection of energy economics, sustainable finance, text analytics and empirical asset pricing.