This dissertation examines the robustness of systems of cooperation—the ability to maintain levels of cooperation in the presence of a potentially disruptive force. I examine rankings as a potentially disruptive force that is commonplace in organizations. A ranking is the ordering of individuals according to their performance on a specific dimension. Systems of cooperation often operate in contexts that feature rankings (e.g., the ride-sharing company Uber uses a ;;rank and yank” performance evaluation system, yet still expects cooperation on complex cooperative coding tasks) and some explicitly use rankings to motivate cooperative contributions toward a collective goal (e.g., the character improvement App ;;Peeple” consists of members’ public evaluations of each other’s character and uses a public ;;positivity rating” to motivate members to maintain a more collegial environment). Yet, a growing body of research is highlighting potential downsides to rankings that could undermine the maintenance of systems of cooperation. This research suggests that rankings may unexpectedly introduce new dynamics into a system of cooperation that drive actors toward uncooperative behaviors and undermine the system as a whole. This dissertation aims to address this tension by exploring how systems of cooperation interact with rankings. Specifically, it explores how rankings can both enrich and perturb a system of cooperation and how systems can achieve robust cooperation in the presence of rankings.Chapter 1 introduces the dual role of rankings for systems of cooperation, reflects on the importance of identifying characteristics that make these systems robust, and discusses how the changing nature of work creates a new urgency for understanding how rankings affect cooperation. This introductory chapter is followed by two empirical chapters that examine distinct pieces of the puzzle for how rankings affect the maintenance of cooperation over time. Chapter 2 examines how the introduction of a performance ranking affects established systems of cooperation. Using a between-groups, no-deception experimental design that includes 74 groups, 594 participants, and over 11,000 cooperation decisions, it examines 1) whether the self-sustaining properties of systems of cooperation are naturally able to overcome the potentially disruptive effects of rankings, and 2) in the case of disruption how managers may be able to restore cooperation in the presence of rankings—making these systems of cooperation more robust. Chapter 3 examines an online community that explicitly uses a ranking to promote cooperation. Using over 1.2 million observations of members’ weekly behaviors, this chapter examines how potential losses and gains in rank inspire individuals to perform both cooperative and uncooperative behaviors and explores how the system-level implications of these behaviors may affect the robustness of systems of cooperation. Chapter 4 concludes the dissertation by synthesizing findings from the empirical chapters, discussing their joint implications for building robust systems of cooperation, and detailing areas of future research.