Abstract In marine and terrestrial ecosystems, organisms are affected by environmental variations that cause fluctuations in population size. The harvest–interaction hypothesis predicts that environmentally induced fluctuations in population size are magnified by harvesting. Empirical evidence is urgently needed in the context of global change because greater fluctuations will increase extinction risk. Here, we review theoretical and empirical work that has addressed the harvest–interaction hypothesis in fish, birds and mammals. We identify the mechanisms by which harvesting might make population size more variable over time and thereby increase the risk of extinction. Theoretical models show that harvest can modify population structure in time and space, and that changes in the amplitude and synchrony of population dynamics both increase extinction risk. Empirical evidence indicates that fishing amplifies the effects of environmental changes on the population variability, but no empirical study of terrestrial species has tested for amplified environmentally induced fluctuations due to hunting. Synthesis and applications. In terrestrial species, theoretical studies have evaluated how environmentally induced fluctuations in population size are magnified by different harvest strategies, but there is now an urgent need for an empirical evaluation of this hypothesis. Future research is needed to explore how hunting and climate interact and to test whether hunting enhances environmentally induced fluctuations in population numbers of terrestrial species.