We argue that economic links, such as supply chains, can create a common roof that protects foreign investors in host countries that lack strong institutions to protect property rights. Supply chains link the activities of firms: when a host government breaks a contract with one firm, other firms in the supply chain are harmed. These partner firms therefore have incentive to protect one another’s property rights. This leads to the key implication of our argument: host governments are less likely to violate the property rights of firms that are more tightly linked with other firms in the host economy. We test our argument with cross-national data on investment arbitration, a survey of US multinational subsidiaries in Russia, and case studies from Azerbaijan. Our findings imply that one benefit of outsourcing in developing and transition economies is the creation of a network of partner firms that protect each other’s property rights.

International bureaucrats must often serve multiple principals who collectively choose policy. How does this affect bureaucrats' incentives to truthfully reveal their private information? I construct a cheap talk model in which a bureaucrat possesses private information about how policies translate into outcomes. The bureaucrat can communicate publicly observable messages about this information to two policymakers, who must then bargain over a set of policy choices. I find that both the bureaucrat's willingness to communicate informatively and the choice of an optimal bureaucrat are highly contingent on the bargaining powers of the two policymakers. When each policymaker is bound to adhere to the bargaining outcome, “moderate” bureaucrats are most preferred. In contrast, when at least one policymaker can leave the bargaining table and exercise an outside option, “biased” bureaucrats can be optimal. I illustrate my findings by examining UN weapons inspections in Iraq from 1991 to 2003.I wish to thank Bruce Bueno de Mesquita, Michael Gilligan, Catherine Hafer, Matias Iaryczower, Dimitri Landa, Adam Meirowitz, Alastair Smith, and the editor and two anonymous reviewers for their feedback. Previous drafts of this article were presented in the New York University Politics Department Research Workshop and the 2006 Annual Conference of the Midwest Political Science Association.

Third parties complicate World Trade Organization (WTO) dispute settlement by adding voices and issues to a dispute. However, complainants can limit third parties by filing cases under Article XXIII of the General Agreement on Tariffs and Trade (GATT), rather than Article XXII. We argue that third parties create “insurance” by lowering the benefit of winning and the cost of losing a dispute. We construct a formal model in which third parties make settlement less likely. The weaker the complainant's case, the more likely the complainant is to promote third party participation and to settle. Article XXII cases are therefore more likely to settle, controlling for the realized number of third parties, and a complainant who files under Article XXIII is more likely to win a ruling and less likely to see that ruling appealed by the defendant. We provide empirical support using WTO disputes from 1995 to 2011.

How does variation in the strength of a court’s jurisdiction and enforcement affect strategic behavior by states involved in international disputes? The authors construct a formal model and identify three important ways that legal institutions can have a deleterious effect on international cooperation by magnifying the bargaining problems arising from incomplete information about the quality of the legal claims. First, strong courts create less information revelation in pretrial bargaining. Second, strong courts reduce the likelihood of pretrial settlements between states. Third, strong courts lead to more brinksmanship over high-value assets, which leads to conflict if the court refuses to intervene. The authors argue that a key policy implication of their model is that attempts to strengthen international courts must be accompanied by increased precision of international law to ameliorate the deleterious effects of strong courts.

The past three decades have witnessed the development of a rich literature that applies the formal tools of game theory to understanding international cooperation and international institutions. We divide this literature into three "generations" of scholarship. With a few notable exceptions, the first generation used very simple models-2 × 2 normal form games-to understand why states need to cooperate and why they comply with their cooperative agreements under conditions of anarchy. This first generation unfortunately bogged down in the neorealist-neoliberal debate. Second-generation scholars began to use tailor-made models to address the neorealist-neoliberal debate and to turn to new questions, such as how international agreements are created and how domestic political divisions affect international cooperation. With answers to the key questions of how international agreements are created and complied with, third-generation scholars could turn to increasingly refined models to answer specific questions about international institutions, such as the proper size of multilateral agreements, how the gains of cooperation are distributed, whether flexibility provisions should be built into agreements, and the specific functions of international organizations.