Handling of climatic and economic uncertainties in simple integrated assessment models of climate change

Hans-Martin Fussel
Center for Environmental Science and Policy (CESP), Stanford University, Stanford, CA
and
Potsdam Institute for Climate Impact Research (PIK), Potsdam, Germany

Absract

This presentation discusses the flawed treatment of several uncertainties in climate policy analyses with welfare-optimizing climate-economy models. My motivation is to raise awareness of some challenges faced by integrated assessment models of climate change, and more specifically to improve the representation of uncertainties of the climate system in these models. I start my presentation with a brief introduction to the concept of welfare-maximization in climate policy analysis. I then discuss how the flawed representation of uncertainties of the climate system in two probabilistic climate policy analyses has lead to overconfidence in transient climate change projections and, in turn, to a significant underestimation of “dangerous” climate change in the 21st century. In the remainder of my talk, I address the representation of selected uncertainties related to key socio-economic processes. I show how contradicting assumptions regarding the evolution of carbon mitigation costs over time in two implementations of DICE-99 and inconsistencies between different social welfare functions applied in past climate policy analyses can lead to radically different policy recommendations. My findings indicate that much more caution is needed in the development, application, and modification of simple climate-economy models, and in the interpretation of their results. I provide recommendations on how to address the problems identified here in future climate policy analyses.