We provide evidence that institutions strongly inﬂuence where oil and gas exploration takes place. At national borders, exploration companies drill on the side with better institutional quality two times out of three. To identify the effect of institutions, we utilise a global dataset on the location of exploration wells and borders. This allows for a regression discontinuity design, with the key assumption that the position of borders was determined independently of geology. To break potential simultaneity between borders, institutions and activities in the oil sector, we exploit the historical sequence of drilling occurring after the formation of borders and institutions. The results are consistent with the view that institutions shape exploration companies’ incentives to invest in drilling as well as host countries’ supply of drilling opportunities. It follows that the observed distribution of natural capital across countries is endogenous with respect to institutions.