Abstract It is frequently suggested that human foragers occupy ‘marginal’ habitats that are poor for human subsistence because the more productive habitats they used to occupy have been taken over by more powerful agriculturalists. This would make ethnographically described foragers a biased sample of the foragers who existed before agriculture and thus poor analogs of earlier foragers. Here, we test that assertion using global remote sensing data to estimate habitat productivity for a representative sample of societies worldwide, as well as a warm-climate subsample more relevant for earlier periods of human evolution. Our results show that foraging societies worldwide do not inhabit significantly more marginal habitats than agriculturalists. In addition, when the warm-climate subsample is used, foragers occupy habitats that are slightly, though not significantly, more productive than agriculturalists. Our results call into question the marginal habitat criticism so often made about foragers in the ethnographic record.