Replication Data for: From Protest to Rebellion? Institutions and Protest Escalation in Autocracies.

Why does political contention induced by major exogenous shocks escalate into broad-based rebellion in some autocratic states but remain scattered and tame in others? The question has important implications for our understanding of contention under authoritarianism and, ultimately, the politics of authoritarian survival and change. While a growing body of literature has greatly advanced our knowledge in this area, existing accounts leave important unanswered puzzles such as why strongly oppositional frames come to quickly resonate with would-be demonstrators in some countries but not others, why some autocrats shoot at unarmed protesters but others use restraint, and why some rulers succeed in calming or dividing demonstrators with distributive concessions while others fail to de-escalate collective action in their territories with the same goods. My dissertation offers a new explanation for variation in protest escalation that is able to address these and other puzzles. I contend that in a given autocracy the likelihood that shock-induced protests, once begun, will escalate into mass rebellion hinges on a key aspect of autocratic design: the extent to which the hegemon institutionalizes opportunities for his popular base – the “minimal winning coalition” necessary for an uprising – to bargain with him and his intermediaries over non-strategic everyday matters, such as local status positions or distributive goods. All things equal, I argue that autocrats who do not root their power in some form of bargaining and who seek to rule by decree are most likely to face protest escalation because the top-down approach endogenously creates incentives for predatory forms of governance which alienate the popular base. I demonstrate my theory through a structured comparison of Syria and Jordan, which saw puzzlingly different protest dynamics in response to the wave of regional unrest in early 2011. Using a natural experiment and range of historical, geographical, and individual-level survey data, I show how Syrian and Jordanian state-builders' choice of dissimilar institutional-bargaining approaches in the 1960s -- a top-down non-bargaining model in Syria and a bargaining-centric strategy in Jordan -- led, over time, to systematically different relationships between the states and their rural mass constituencies prior to 2011 and how this explains the systematically different protest dynamics observed in 2011. Despite similar starting points in the 1960s, by late 2010 rural Syrians were not only more politically disaffected with their state than rural Jordanians, but also more economically independent and more unified by visceral solidarities (partly around their shared grievances against the state). The combination of political and economic alienation and strong visceral solidarities, I demonstrate, made escalation of protests in Syria more likely than in Jordan for several reasons: they rendered the Syrian mass constituency predictably more receptive to the \"rebellious\" Tunisian and Egyptian protest frames in early 2011 and more able to reach across geographic divides to mobilize a broad oppositional movement; they made it more difficult for the Syrian regime to calm protests by \"buying off\" and dividing the demonstrators with selective distributive concessions since the state's distributive goods were intrinsically less valued; and, given that they neutralized the regime's non-violent distributive policy tools, they pushed the Syrian state to fall back on violence, which added to public outrage and sped up protest mobilization and calls for regime change. I supplement these analyses with a range of rich qualitative and quantitative evidence to show why my institutions- and governance focused explanation provides a better account than popular competing explanations including, among other, the generosity of the autocrat's distributive bargain, the breadth of the autocrat's mass coalition, the strength of pre-existing civil society, supposed monarchical legitimacy, access to rents, and colonial legacies. The datasets attached here can be used to replicate the main quantitative analyses in Chapters 3-6 of the dissertation.