Research - 25.03.2025 - 10:45
The New Techniques and Technologies for Statistics (NTTS) is a leading international conference dedicated to advancing official statistics. Organised by Eurostat, the statistical office of the European Union, and the European Commission, this year's conference took place from 11 to 13 March 2025 in Brussels. The research conducted by Despoina Makariou, and her co-authors was selected for presentation, underlining the significance of their work at the intersection of climate finance and machine learning. Their study integrates traditional statistical methods with state-of-the-art machine learning models to more accurately predict carbon prices, a critical step toward enhancing the efficiency and fairness of emissions trading schemes (ETS). Their contribution captured significant interest, particularly from policymakers seeking data-driven solutions for climate regulation.
By examining the divergence between the EU and UK ETS post-Brexit, the researchers gained deep insights into the structural disparities that have emerged between these markets. Originally aligned, the two systems began to diverge in 2021, leading to widening gaps in carbon pricing.
"Our machine learning approach not only improves forecast accuracy but also uncovers key differences in carbon pricing mechanisms across emissions trading schemes," Despoina Makariou explains. These discrepancies hinder the effectiveness of carbon pricing and create opportunities for market participants to exploit regulatory gaps, ultimately weakening the enforcement of pollution costs. For example, businesses may relocate production to regions with lower carbon costs, undermining broader climate goals.
"Our findings underscore the urgent need for coordinated policies to foster fair competition and ensure an efficient allocation of emissions reduction targets across borders." Data-driven solutions, she asserts, are vital for shaping such policies by providing empirical evidence that informs regulatory decisions and enhances market transparency. Her presentation at NTTS was recorded and can be viewed online (starting at 12:44 minutes), while the abstract that was submitted can be downloaded here (from page 6).
The NTTS conference brought together researchers, policymakers, journalists, and representatives from organisations such as the EU, UN, and World Bank. The reaction to her presentation provided valuable insights: "The discussions following my presentation highlighted a critical challenge – carbon pricing mechanisms remain highly fragmented, creating gaps that undermine fair competition and the effectiveness of emissions reduction efforts." The validation of her findings from diverse professionals and the enthusiasm for her research methodology reaffirmed the importance of this work. "It was encouraging to see strong interest from both researchers and policymakers in how data-driven approaches, like our hybrid machine learning methodology, can help bridge divides and foster more coordinated climate strategies," she added. Together with her co-authors, Despoina Makariou is currently preparing to submit their findings to leading academic journals.
Assistant Professor of Risk Management and Insurance
