More than a feeling

Climate impact models are forced to make sweeping assumptions when estimating social and economic welfare damages to countries around the world, because of a lack of data and understanding of local causal mechanisms. In this paper, we estimate the effects of rising temperatures on countries around the world using an experienced utility approach, based on subjective well-being survey data collected in 160 countries for 13 years. We take advantage of 40 years of variation in daily land surface temperature data, to find that one exceptionally hot day significantly lowers well-being. Furthermore, the effect size varies substantially between and within countries. Identifying this high degree of heterogeneity is important because it illustrates the shortcomings of many current models which are geographically coarse. Moreover, we compare the marginal utility of income and non-income effects and find that income accounts for only a small proportion of the damages caused by extreme temperature s. This demonstrates the adverse effect on non-market goods is dramatically higher than previously assumed, which indicates current models are missing a fundamental source of climate-related damages.