Dissertation: “​Mark Dictators’ Calendars: An Informational Theory of Election Schedules in Autocracies”

Why do some dictators hold elections on time, while others manipulate the election date or cancel elections? How does the manipulation of election schedules influence regime stability and citizens' political participation? This dissertation answers these questions using game theory, cross-national analysis, and a survey experiment. Chapter 2 presents an informational framework of election schedules: they reveal regime strength to ruling allies, opposition elites, and ordinary citizens. Dictators' scheduling decisions hinge on their expected strength: they manipulate the date when they are concerned about their performance in a nearing election or their declining popularity in a future one. Therefore, the manipulation sends a public signal of regime weakness. This chapter also introduces an original dataset of the schedules of 1,390 elections in 280 post-World War II autocracies. Chapter 3 finds that weak dictators are likely to manipulate the election date, and the chapter builds a game model between the dictator and his challenger to explain why. Chapter 4 shows that autocracies that observe election calendars are more likely to experience peaceful, negotiated transitions than autocracies that manipulate schedules, especially where the ruling party is weak. Chapter 5 presents a survey experiment in Hong Kong and argues that manipulated election date motivates voters – both opposition and government supporters – to turn out. Together, this dissertation brings a new perspective into the discussion of autocratic elections: their timing. Using novel global panel data and an original survey, this dissertation applies a combination of game-theoretic, statistical, and experimental methods.