How Certain Can We Be About NGFS Climate Scenarios?, by Alex Pui, Sebastian EV Werner, Gireesh Shrimali, and Motoshi Tomita
This paper investigates uncertainty in NGFS climate scenario projections across scenario, model, and vintage dimensions, with a particular focus on how revisions to initial conditions and recalibration across successive releases affect projected outcomes.
Drawing on 254 variables across four NGFS vintages, five climate pathways, and three integrated assessment models, the analysis finds that vintage-related uncertainty is a first-order feature of NGFS outputs and can, in many cases, rival or exceed uncertainty arising from scenario design or model choice.
These results have important implications for the use of climate scenarios in financial risk assessment, stress testing, and supervisory exercises, particularly over the shorter horizons most relevant to decision-making