International Studies Quarterly(1998)42,709-731 Conflict Expectations and the Paradox of Economic Coercion DANIEL W.DREZNER University of Colorado at Boulder Despite their increasing importance,there is little theoretical under- standing of why nation-states initiate economic sanctions or what deter- mines their success.These events are often explained away as "symbolic politics"driven completely by domestic-level factors.This article develops a simple game-theoretic model of economic coercion to show that both "senders"and "targets"of economic coercion incorporate expectations of future conflict as well as the short-run opportunity costs of coercion into their behavior.Conflict expectations have a paradoxical effect on'coercion events.First,senders that anticipate frequent conflicts will be more willing to initiate economic coercion,even if such attempts are costly.Senders that anticipate few conflicts will not threaten sanctions unless they incur mini- mal costs and the target would suffer significantly.While a robust antici- pation of future disputes might make the sender prefer a coercive strategy, it also reduces its ability to obtain concessions.Target states that anticipate frequent conflict with the sender will make fewer concessions.Ironically, a sender will obtain the most favorable distribution of payoffs when it cares the least about its reputation or the distribution of gains.These hypotheses are tested statistically,with the results strongly supporting the conflict expectations model. In 1990,the United States convinced the United Nations Security Council to impose stringent economic sanctions against Iraq in response to Saddam Hussein's invasion of Kuwait.These sanctions achieved the greatest degree of international coop- eration in modern history and cost Iraq half of its GNP in lost trade.The Iraqi regime refused to back down,however,and force was needed to restore Kuwaiti independence. In 1991,the United States threatened to withhold loan guarantees from Israel unless that country agreed to halt the construction of new housing in the West Bank and attend a multilateral peace conference in Madrid.Unlike the Iraqi case,in this episode the United States acted alone,without much fanfare,and against the wishes of an active domestic lobby.The costs of the sanctions to Israel were significant but far lower than in the Iraqi case.Yet in the end the Israelis acquiesced. Why were the outcomes so different?More generally,under what conditions will a sanctioning country (called the sender)attempt economic coercion?What determines the magnitude of the concessions made by the sanctioned country(called the target)? Author's nate:I thank Bruce Bueno de Mesquita,John Ferejohn,Kurt Taylor Gaubatz,Judy Goldstein,Stephen D. Krasner,and Paul Krugman for their comments on earlier drafts.Previous versions of this article were presented at Stanford University,the University of Colorado at Boulder,and the University of California at Santa Barbara.I am grateful to the seminar participants at these places for their valuable contributions.All remaining errors are my own. 1998 International Studies Association. Published by Blackwell Publishers,350 Main Street,Malden,MA 02148,USA,and 108 Cowley Road,Oxford OX4 IJF,UK.International Studies QuaTterly (1998) 42, 709-73 1 Conflict Expectations and the Paradox of Economic Coercion Uniuersity of Colorado at Boulder Despite their increasing importance, there is little theoretical under￾standing of why nation-states initiate economic sanctions or what deter￾mines their success. These events are often explained away as "symbolic politics" driven completely by domestic-level factors. This article develops a simple game-theoretic model of economic coercion to show that both "senders" and "targets" of economic coercion incorporate expectations of future conflict as well as the short-run opportunity costs of coercion into their behavior. Conflict expectations have a paradoxical effect oncoercion events. First, senders that anticipate frequent conflicts will be more willing to initiate economic coercion, even if such attempts are costly. Senders that anticipate few conflicts will not threaten sanctions unless they incur mini￾mal costs and the target would suffer significantly. While a robust antici￾pation of future disputes might make the sender prefer a coercive strategy, it also reduces its ability to obtain concessions. Target states that anticipate frequent conflict with the sender will make fewer concessions. Ironically, a sender will obtain the most favorable distribution of payoffs when it cares the least about its reputation or the distribution of gains. These hypotheses are tested statistically, with the results strongly supporting the conflict expectations model. In 1990, the United States convinced the United Nations Security Coun'cil to impose stringent economic sanctions against Iraq in response to Saddam Hussein's invasion of Kuwait. These sanctions achieved the greatest degree of international coop￾eration in modern history and cost Iraq half of its GNP in lost trade. The Iraqi regime refused to back down, however, and force was needed to restore Kuwaiti independence. In 199 1, the United States threatened to withhold loan guarantees from Israel unless that country agreed to halt the construction of new housing in the West Bank and attend a multilateral peace conference in Madrid. Unlike the Iraqi case, in this episode the United States acted alone, without much fanfare, and against the wishes of an active domestic lobby. The costs of the sanctions to Israel were significant but far lower than in the Iraqi case. Yet in the end the Israelis acquiesced. Why were the outcomes so ddferent? More generally, under what conditions will a sanctioning country (called the sender) attempt economic coercion? What determines the magnitude of the concessions made by the sanctioned country (called the target)? Author's note: I thank Bruce Bueno de Mesquita, John Ferejohn, Kurt Taylor Gaubatz, Judy Goldstein, Stephen D. Krasner, and Paul Krugman for their comments on earlier drafts. Previous versions of this article were presented at Stanford University, the University of Colorado at Boulder, and the University of California at Santa Barbara. I am grateful to the seminar participants at these places for their valuable contributions. All remaining errors are my own. 01998 International Studies Association. Published by Black~~~ell Publishers, 350 Main Street, Malden, MA 02148, USA, and 108 Cowley Road, Oxford OX4 lJF, UK