Most authoritarian elections are non-competitive affairs, manipulated by state actors to guarantee seats for regime- backed candidates. However, on rare occasions, the state provides an unexpected electoral opening for the political opposition, leading to a defeat for regime-backed candidates. This dissertation traces these unexpected electoral openings to factional rivalries within the state's constitutional structure. Actors within an authoritarian state can be grouped broadly into conservative and reformist camps. Satisfied with the political status quo, conservatives oppose any meaningful steps toward political liberalization. In contrast, reformists aim to increase their share of the state by opening the political system to societal groups. The ability of reformists to change the electoral process and provide an electoral opening depends on the country's constitutional framework. The configuration of accountability relationships within the state, as well as the jurisdiction over the different parts of the electoral process, determines the leeway reformists have in steering the elections to their favor. This dissertation tests the above theory through comparative analyses of parliamentary elections in Egypt, Algeria, and Iran. The parliamentary elections are presented as qualitative stories. These stories identify relevant political agents, their preferences, and their powers, and then discuss political strategies adopted, constraints on strategies, and paths taken and paths rejected. By doing so, the narratives highlight the importance of institutional variation in determining electoral openings and outcomes in authoritarian systems