This Commentary sets out four lines of evidence to argue both that emission forecasts are intrinsically uncertain, and that there is clear evidence of projection inflation in the forecasts of sector emissions used to underpin the setting of sector caps in emission trading systems. From a limited evidence base, the authors conclude that uncertainty is at least 2%/year, overlaying an upward bias (projection inflation) on the order of 1%/year, cumulative. The Commentary concludes that this has important implications both for allocation approaches, and for some other design elements in the EU ETS. Forecasting uncertainty is not an inconvenience which is best ignored, but a fundamental fact that must be accommodated in the future design of the EU ETS and other CO 2 emission trading schemes.

Climate Policy 6 (2006) 495501