Non-standard errors in carbon premia

This research investigates the influence of methodological choices in portfolio sorts on the size of the carbon premium. By analyzing more than 100,000 methodological paths, we find that variations in the construction of brown-minus-green portfolios create substantial non-standard errors. From 2009 to 2022, the mean carbon premium is -0.16% per month, with a non-standard error of 0.26%. Additionally, there is significant time-series variation in non-standard errors, which correlates with climate media attention. Controlling for unexpected changes in climate concerns substantially reduces methodology-induced uncertainty and helps explain the absence of a consistently positive carbon premium.