This paper uses an econometric Ricardian model and more than 150 climate change scenarios generated by General and Regional Circulation Models to estimate the impact of climate change on the agricultural sector of the EU15. We find that impact estimates vary greatly depending on the climate models used. Unfortunately, it is not possible to attach probabilities to the climate scenarios. This leaves decision makers with a high degree of uncertainty. There are no obvious ways to deal with this uncertainty. Decision makers ultimately have to choose probabilities that they want to attach to each climate scenario based on their beliefs. This paper explores the implications of using alternative subjective probability distributions on impact estimates.