In a recent paper in this journal, Azar and Schneider present carbon abatement costs as a few years delay in an increasing path of welfare and they suggest the adoption of stricter abatement policies to avoid climate change. We show that the same argument can be used to derive the opposite conclusion. In most studies, not only abatement costs, but also climate change costs represent a small loss in future welfare. The reason for this is the implicit assumption of perfect long-term substitutability between man-made and environmental goods that many of these models adopt. We argue that, instead of downplaying the costs of abatement measures, we should try to better understand the substitutability possibilities between man-made and environmental goods, that is, the character of long-term costs of climate change.