The literature on escalation situations and audience costs suggests that democratic executives tend to increase commitment to a foreign policy in response to negative feedback. However, real-world cases from international politics suggest otherwise. Specifically, executives do not appear to respond uniformly to failing situations. While scholars have begun to unravel the audience cost mechanism, up until know, we know little about reasons for the variation in how executives use policy feedback to update commitment to a foreign policy. In this dissertation, I adopt an integrative approach and present a model of sequential decision-making that explains the conditions under which leaders escalate and de-escalate commitment in response to feedback. I attempt to break down the audience cost mechanism to explain why democratic executives do not respond uniformly to negative feedback. While the literature on the escalation of commitment suggests decision-makers tend to increase investment in the face of negative feedback, my theory suggests that under certain conditions, executives may find it politically advantageous to back down from a failing policy. My theory emphasizes the relationship between citizens, executives, and foreign policy effectiveness. Next, I suggest that the foreign policy tool of military intervention provides a suitable test case for a theory of sequential decision-making. I first test hypotheses derived from the theory regarding the preference formation process of democratic citizens during the course of such an episode. Understanding the response of citizens to feedback is an important first step to understanding the updating decisions of democratic executives. While previous work has relied on aggregate survey data, experimentation provides me with the ability to analyze how an individual citizen?s preference over commitment is impacted by policy feedback. The results of the experimental analyses suggest that citizens act as investors: they favor increasing commitment to military interventions when viewing negative feedback, up to a point. I then test the main hypotheses derived from the theory regarding executive decision-making on a dataset of major power military interventions from 1960-2000. Overall, the results support the hypotheses: public approval conditions the manner in which executives use feedback to update intervention commitments. In the conclusion, I summarize the study by highlighting key results, present the broad implications for the study of democratic foreign policy making, and discuss avenues for future research.

My dissertation contributes to the accountability literature in international relations by examining the role constituents' preferences can potentially play in fomenting or constraining coercive foreign policies in democracies. In times of international crises, domestic audiences have specific coercive foreign policy preferences and will support executives who represent them when selecting coercive foreign policies. Executive actions will increase popular support or generate audience costs depending on whether these actions are consistent with the specific policy preferences that domestic audiences have given the threat a crisis poses to national security. To determine when audiences prefer economic or military coercion and how these preferences affect their evaluation of the executive I conduct three experiments, including a survey experiment conducted with a representative sample of Americans and an experiment conducted with a convenience sample in the United Kingdom. The results show interesting similarities and differences between the cross-national samples regarding foreign policy preferences and the public's propensity to support and punish leaders during times of international conflict. Mainly, I find that (1) the concept of audience costs can be expanded to cases of economic coercion, (2) under certain circumstances audience costs operate even in crises that are not very salient and (3) when there is a mismatch between public preferences and threats issued by the executive, audience costs do not operate at all.

This dissertation explores the interactive effects of domestic audience costs and strategic interests on state behavior in international crises. I argue that the magnitude of a leader’s audience costs is influenced by the level of strategic interests, which leads to several predictions of crisis behavior in terms of (1) decisions to issue threats, including bluffs, (2) the credibility of these threats and the willingness of opponents to resist, and (3) crisis outcomes, including war. In the theoretical chapters, a formal model of crisis bargaining is stylized under conditions of complete and incomplete information. Based on this model, several novel predictions are derived regarding crisis behavior. These predictions are quantitatively tested through a series of monadic and dyadic probit and multinomial logit models using a dataset of deterrence crises for the period 1895-1985. The results lend strong validity to the approach advanced here that does not consider endogenous and exogenous factors in isolation, but rather models their interplay to predict the dynamics of crisis behavior. With respect to dispute initiation, the results show that strategic interests have a much stronger influence on authoritarian leaders’ willingness to initiate disputes than they do for democracies. Moreover, the formal stylization and empirical analyses show that democracies can and do bluff, which is in contrast to the conventional expectations from audience cost research. Relatedly, this study specifies if and when democratic threats are credible and how the interplay between variable domestic costs and strategic interests can lead to deterrence success, failure, or war. While there is little difference between the credibility of democratic and authoritarian threats at the lower level of interests, democratic threats become more credible and less likely to be resisted as the interests at stake increase. As for crisis outcomes, among others, war is more likely between opponents with vital interests involved; yet even here, the predictions are not straightforward but rather the probability of war is increasing at a differential rate for democratic and authoritarian initiators. Whereas the formal models in this study provide the logical rationale for these and other expectations, the quantitative findings demonstrate their empirical validity as well.

"Audience costs" models of international relations suggest a purely informational role for domestic politics in conflict settings. Here, domestic politics serve as a rich signal of belligerents' true intentions, allowing them to more quickly resolve disagreements, decreasing the likelihood and duration of war. But if belligerents can have different beliefs about publicly available information, then domestic politics might confuse rather than clarify conflict situations, increasing the likelihood and duration of war. I present empirical evidence of conventional "audience costs" models' shortcomings in explaining the dynamics of the US counterinsurgency efforts in Iraq and the response of Iraqi insurgents to those efforts. I then develop a formal model to show how differences in beliefs between insurgents and counterinsurgents about domestic political audiences in Iraq may have contributed to the prolonged nature of the conflict. I argue that the underlying cause of the conflict's duration is disagreement between belligerents about whether and how Iraqi civilians contribute to a successful counterinsurgency, leading belligerents to disagree not only before fighting about who is likely to win, but during fighting about who is actually winning. / text

This thesis investigates how religious holidays affect the chance for ceasefire success. It does so while engaging in the topic of explanation and prediction, with combined methods consisting of regression analysis and forecasting using random forests. The theoretical framework argues that religious holidays impose higher audience costs for violence on leaders, increasing the chance for successful agreements. This would manifest as ceasefires connected to religious holidays being more successful than those that are not. The findings from the regression analysis find no support for the hypothesis, and rather indicate that conflict intensity, ceasefire duration, as well as monitoring and verification and enforcement mechanisms better explain the apparent variation in success. The forecasting indicates a minor difference in predictive power between the model including holiday and religiosity and the one excluding them, as well as minor effects of the independent variables on ceasefire success from partial dependence plots. These findings do not oppose the rejection of the hypothesis but rather indicate the need for an increased prevalence of forecasting methods within peace and conflict research and robustness tests using other definitions of success.

This thesis examines the role of mediators in WTO trade negotiations and the impact of individual mediation strategies on the outcome of these negotiations. It seeks to answer the question how mediators can help negotiators to overcome the bargaining problem and reach an agreement. The thesis first presumes that mediators who frequently apply more interventionist and inclusive tactics can increase the likelihood of a successful outcome of negotiations. Secondly, it presupposes that particular mediation tactics enable negotiators to save their face not only in front of their opponents, but more importantly also in front of domestic constituencies. Using the case of Geneva-based negotiations and the Bali Ministerial Conference, the thesis tracks individual mediation strategies which were employed by WTO chairs or the Director-General to determine their impact on the final adoption of the Bali package, representing the first multilateral trade agreement concluded under the auspices of the WTO. The analysis indicates that mediators who actively intervene in the bargaining process and assist negotiators in decreasing their audience costs may positively influence the chances of reaching an agreement. Contrarily, it shows at the same time that the effects of inclusive mediation are varied and...