Despite growing engagements between firms and nonmarket stakeholders—such as local communities and nongovernmental organizations —research has yet to examine the emergence of formal contracts between them. Given that a very large number of such contracts are theoretically possible but only a small number exist, we seek to understand what factors explain the use of contracts to govern some relationships between firms and nonmarket stakeholders but not others. We draw on transaction cost economics to study transactions wherein a nonmarket stakeholder provides a firm access to a valuable resource and to understand when these transactions are governed by formal contracts. We propose that, when a firm makes site-specific investments, a stakeholder’s use rights to the resource sought by the firm, the negative externalities generated by its use, and the stakeholder’s capacity for collective mobilization increase holdup risk for the firm and therefore the probability of a contract. We collect novel data on the location of indigenous communities and mines in Canada to identify a plausible exhaustive set of indigenous communities “at risk” of signing a contract with a mining firm. We test our hypotheses by relying, respectively, on historically assigned property rights over lands, the mine-community colocation in a watershed and proximity on transportation routes, and archival records of community mobilization events. We find support for our propositions by examining which of the 5,342 dyads formed by 459 indigenous communities and 98 firms signed 259 contracts between 1999 and 2013.