This paper applies collective action and transaction-cost theory to the theoretical debate around the management of common property regimes, with supporting evidence from recent empirical research in Mongolian pastoralism. The use of common grazing in the context of other aspects of pastoral livelihoods is also studied, and it is argued that the more often that groups of herders can cooperate across a range of activities, the more likely it is that they can overcome the transaction costs of controlling the use of the commons.