How do elections and post-election protest shape political trust in a competitive autocracy? Taking advantage of largely exogenous variation in the timing of a survey conducted in Moscow in 2011, we find that an election had little systematic effect on political trust, perhaps because vote improprieties were not new information. In contrast, the unexpected protest that followed increased trust in government. We argue that when autocrats permit protest unexpectedly, citizens may update their beliefs about the trustworthiness of the government. In this case, heightened trust arises largely from opposition voters - those most likely to be surprised by permission to hold the protest - who update their beliefs. Our results suggest that citizens may cue not off the content of a protest, but off the government's decision to permit it. In addition, autocrats can increase trust in government by allowing protest when it is unexpected.
Does decentralization affect how voters attribute blame for poor economic performance? The question of whether political centralization ties regime leaders to local economic outcomes is particularly important in authoritarian regimes, where economic performance legitimacy is a key source of regime stability. Using political and economic data from large Russian cities for the period 2003-2012, we investigate whether replacing direct mayoral elections with appointments affects the way voters attribute blame for economic outcomes. Using a difference-in-differences design, we find that the ruling party is more likely to be punished for poor economic performance in cities with appointed mayors than it is in cities with elected mayors.