Domestic politics and conflict among rivals: Domestic incentives for conflict or cooperation

This thesis develops a game theoretic model of how the foreign policy preferences of the executive and the government in the legislature interact over the electoral cycle to affect democratic leaders' foreign policy choices and the foreign policy choices of their rivals. When the foreign policy that satisfies the legislature is not the same policy that maximizes support among the executive's constituency, the executive is cross-pressured by the conflicting demands of remaining in a productive government and obtaining votes in the next election. This forces democratic leaders to weigh these competing preferences over the course of the electoral cycle, which leads to trends in foreign policy where the executive chooses policies that please the legislature early in the term and later revert to the policies that please the executive's constituency when elections draw near. These trends also have implications for their foes' foreign policy choices. These trends are empirically assessed with large-n analysis of KEDS and COPDAB event data. The central findings are that democratic institutions and processes have a predictable influence on foreign policy over time, and some configurations of preferences and election timing are not conducive to peace. Executives tend to 'rally around their party' when elections are near, which leads to partisan differentiation: when the preferences of the legislature and the executive's constituency do not match, leftist executives become more peaceful and rightist executives become more hostile as elections approach.