The Information Costs of Climate Misinformation: Evidence from Global Policy and Financial Data

Climate misinformation has emerged as a major systemic risk at the intersection of environmental governance, digital communication, and global financial markets. The rapid diffusion of misleading or selectively framed climate narratives - amplified by elite actors, platform algorithms, and politically polarized discourse - distorts public understanding of climate science and alters expectations about transition pathways. These distortions reshape the informational environment within which firms, investors, and policymakers operate, generating elevated uncertainty and weakening the credibility of environmental disclosures. This paper develops an integrated theoretical and empirical framework to measure how climate misinformation affects information asymmetry in financial markets. It contributes to our understanding of current political fragmentation surrounding climate policy. The study advances three core innovations. First, it constructs new multidimensional measures of climate misinformation and related climate policy-related features using large-scale text classification applied to social-media communication and legislative documents in a global context. These indicators capture misinformation intensity, advocacy, skepticism, scientific alignment, and ideological polarization, generating continuous time-series that reflect the evolving climate information ecosystem. Second, my paper links these indicators to financial microstructure outcomes by estimating daily firm-level information asymmetry for approx. 5000 firms using the standardized Probability of Informed Trading (PIN). Because PIN is sensitive to shifts in the quality and availability of public information, it provides a tractable analogue of how misinformation alters trading environments. I suggest misinformation increases the costs of price discovery, and widens informational gaps between informed and uninformed market participants. Using a daily high-frequency panel, I estimate multiple regression and machine learning models that isolate within-firm responses to day-to-day variation in climate misinformation over time. The results show that temporal increases in misinformation are consistently associated with higher PIN - indicating greater information asymmetry - while scientifically credible climate communication reduces it. As a test, I compare these effects to CAPM betas, Fama–French factor loadings, and credit spreads - workhorse financial indicators - and I will see whether the analysis reveals that misinformation primarily operates through the information-friction channel or also through traditional risk-based pricing based on fundamental values. My theoretical contribution supports a transaction-cost mechanism in which misinformation raises costs of communication and exchange, distorts expectations about climate policy trajectories, and slows the rate at which markets incorporate environmental information.