Tinbergen Seminar – Climate shift uncertainty and economic damages

Focusing on global annual averages of climatic variables, as in the standard damage function approach, can bias estimates of the economic impacts of climate change. Here we empirically estimate global and regional dose-response functions of GDP growth rates to daily mean temperature levels and combine them with regional climate projections. We disentangle how much of the missing impacts are due to differences in warming versus heterogeneous damage patterns over space and time. Global damages in 2050 are around 20% higher, when accounting for the shift in the entire distribution of daily mean temperatures at the regional scale. Differences in the shape of daily temperature distributions between climate models transform standard risk rankings based on temperature anomaly, and increase uncertainty across climate models. Joint paper with Romain Fillon and Manuel Linsenmeier.