This project is led by Lancaster University’s Management School with collaborators from the University of Exeter. It uses prediction markets, with expert participants, to forecast climate-related risks.
Duration of the Project: 2025-2027.
Background
Accurate predictions of climate are needed by governments designing policies to mitigate climate change, decision makers planning adaptations to climate change, and financial institutions assessing their exposure to climate-related risks. Forecasting climate requires knowledge from a wide range of disciplines, such as meteorology and oceanography to predict climate in light of future concentrations of greenhouse gases (GHGs), as well as economics and social sciences to help predict what those concentrations might be. Some of this disparate expertise is codified in coupled general circulation models (CGCMs) but much is not. Also, CGCMs are now being joined by other modelling approaches, such as AI-based techniques. Synthesizing all this disparate interdisciplinary activity into actionable forecasts is a major challenge which CRUCIAL intends to tackle using purpose-built prediction markets with expert participants.
Objectives
- To recruit collaborating research groups with expertise relevant to climate forecasting, both in physical and social sciences.
- To establish preliminary prediction markets for climate-related risks to introduce the research groups to how the markets work.
- To form a steering group of potential forecast users to advise on the topics of markets.
- To create joint-outcome prediction markets (e.g., to simultaneously predict GHG concentrations and global temperature anomalies).
- To develop methods to translate prices in the joint-outcome markets to probabilities of different climate/emission scenarios widely used for climate adaptation planning.
- To establish a platform for running long-range prediction markets, with time horizons of years to decades, sponsored by organizations who need the long-range climate predictions that the markets can produce.
Fields of Research
Climate science, prediction markets, behavioural economics.
Kim Kaivanto
Kim Kaivanto is a Senior Lecturer in Economics and the Director of the MSc in Money, Banking and Finance at Lancaster University’s Management School (LUMS). Before joining LUMS he held fellowships at the Eitan Berglas School of Economics and Warwick Business School, from where he received his PhD. Kim's research interests are theoretical and descriptive models of decision making and behaviour under risk and uncertainty. He has applied his expertise to problems such investor sentiment, security behaviour, civil aerospace R&D support schemes, aviation slot allocation and CO₂ emissions, venture capital, the exploitation of social science research. He has also advised the banking sector on climate risk exposure.
Mark Roulston
Mark led the development of the AGORA prediction market platform while at investment firm Winton Group, where he worked for a decade, and at Hivemind, a technology company spun-out from Winton in 2018. He has a PhD in planetary science from Caltech for a thesis on the predictability of El Niño and he continued research on climate predictability at Oxford and Pennsylvania State Universities before working at the UK Met Office, prior to joining Winton. At Winton he led the integration of weather and climate information into quantitative trading strategies.
