This paper analyzes the statistical relationship between climatic factors and the global distribution of population and economic activity. Building on this analysis, a new method is developed for assessing geographically explicit impacts of climate change on the suitability of regions for human habitation and economic activity. This method combines information about differences in the conditional distributions of population density and economic activity across climate categories with climate change projections from an ensemble of general circulation models. In contrast to other cross-sectional analyses of the economic impacts of climate change, the method applied here does not require specific assumptions about the functional form of the relationship between climatic and non-climatic factors on the one hand, and population density and economic activity on the other. The results indicate that climate change will improve the habitability of some scarcely populated regions, in particular in Canada, Scandinavia, Russia, Mongolia, northern China, Tibet, and parts of Central Asia, but it will impair the habitability of many densely populated regions in the eastern USA, southern Europe, northern and southern Africa, eastern China, and parts of Australia. Most parts of India, South-East Asia and Oceania, Central America and northern South America, the Sahara and the Sahel are projected to experience climatic conditions during this century that have no geographical analogue in the present climate. Hence, a large majority of the world’s population is living in regions whose habitability is either projected to decrease or that are projected to experience globally unprecedented climate conditions within this century under a business-as-usual emissions scenario.
