Economic sanctions, increasingly used as instruments of foreign policy in recent decades, have been the focus of numerous academic studies. Recent theoretical advances in our understanding of sanctions cannot be tested adequately with existing data. This article presents a newly developed dataset that contains information on 888 cases in which sanctions were threatened and/or implemented in the 1971—2000 period. We describe the dataset, present descriptive statistics for some of the key variables included, and offer comparisons with the Hufbauer, Schott, and Elliot dataset on sanctions that has been frequently used in previous research. We also present simple statistical relationships between sanctions outcomes and some of the variables commonly believed to affect sanctions success.

Research on when economic sanctions end has emphasized either the international bargaining game played by the sender and the target or the redistributive politics and ruling coalition changes in each state. We contend that neither approach offers a fully satisfactory explanation for economic coercion termination. Bargaining is inconsistent with long coercion episodes while ruling coalition changes cannot inform our understanding of very short episodes. We argue that both bargaining factors and domestic realignments matter, but the influence of bargaining factors declines as a sanctions episode continues while the relevance of domestic realignments increases over time. Our empirical tests, which utilize the new Threat and Imposition of Economic Sanctions data set, provide support for both the bargaining and domestic realignment approaches, suggesting that unifying them is beneficial.

The crisis bargaining literature sees demands as endogenous to crises. However, despite the parallels between military and economic coercion, sanctions researchers have preferred to analyze economic coercion after demands have been issued, and have not explored sufficiently the possibility that when senders formulate their policy objectives, they consider the international constraints imposed by the capabilities and interests of target states. I complement the sanctions literature by deriving the implications of strategic goal formulation in a game theoretic model of economic coercion that assumes endogenous demands. The model explains the inconsistent empirical relationship between sanctions costs and outcomes as well as the paradoxical tendency of senders to select into difficult disputes. I find that threats are not always more effective than sanctions and suggest what an optimal sanctions policy might look like.

Many researchers have reported empirical support for the liberal proposition that increased trade between states reduces their propensity to engage in militarized conflict. However, the literature has been less vocal on the effects of interdependence once actual conflict has started. The author, who builds on the opportunity-cost explanation for the commercial peace, first shows how that explanation can fit a bargaining framework of conflict. It is then argued that if trade reduces conflict by raising its opportunity costs prior to its onset, the prospective opportunity costs should remain high during conflict as well and decrease its observed duration, a prediction that can also be derived from a simple war-of-attrition game. The proposed inverse relationship between interdependence and conflict duration is evaluated empirically, using all militarized interstate disputes between 1950 and 1992. The author utilizes both continuous and discrete-time duration models and considers a number of alternative model specifications and concept measurements, and the results are generally supportive of the tested hypothesis. In addition, a two-stage model of dispute onset and duration is estimated, which reveals that more probable disputes are also longer, but the observed selection pattern does not seem to influence the positive effect of interdependence on the hazards of dispute termination.