Revisiting the Relationship Between Legislative Power Sharing and Autocratic Survival
Forthcoming, Comparative Political Studies.
Conventional wisdom holds that legislatures stabilize autocratic rule by helping leaders credibly commit to sharing power with regime insiders, but empirical evidence supporting this argument is weak. Data limitations make it difficult to directly measure the extent of legislative power sharing across time and space, leading scholars to use the presence or absence of a legislature as a proxy. I use a dynamic Bayesian latent variable model to develop a time-series cross-sectional measure of legislative power sharing in autocracies from 1946 to 2023. After validating the measure, I use it to replicate previous studies of the relationship between legislative power sharing and autocratic survival. I find that their results do not hold. If anything, legislative power sharing appears to be associated with greater, not less, vulnerability to being removed from office by coups and revolts. I suggest that scholars should devote more attention to the threat-enhancing effects of authoritarian institutions.
A Model of Cooptation in Authoritarian Regimes
Manuscript in preparation for submission in fall 2025
After seizing power, dictators seek to co-opt challengers by integrating them into the central government, offering access to the perquisites of power in exchange for their loyalty. However, co-optation poses a critical challenge: Dictators must not only neutralize immediate threats to regime stability, but also secure the long-term loyalty of challengers, especially in times of crisis when the opportunity costs of revolt are lower. This article presents a formal model to explore why some dictators successfully co-opt their opponents and remain in power, while others falter. In equilibrium, the dictator distributes costly policy concessions and rents to secure the loyalty of the opposition after seizing power and in anticipation of future crises. The model also captures scenarios in which the dictator offers insufficient concessions, failing to prevent an uprising, or makes overly generous concessions that empower the opposition and threaten regime stability. This paper provides a framework for explaining empirical variation in the success of co-optation in promoting autocratic survival across time and place.
Falling into Line: Crises and Opposition Behavior in Autocratic Legislatures
Winner of the Prothro Award for Best MA Thesis in the Department of Political Science, University of North Carolina at Chapel Hill
This paper examines the effect of crises on the behavior of co-opted opposition elites in authoritarian legislatures. Crises are dangerous moments in the lives of autocrats because they create opportunities for elite defection and opposition mobilization. Building on this logic, I argue that autocrats will expect opposition legislators to align with the regime during times of crisis, as this projects an aura of elite unity and regime invincibility. In contrast, they will be more willing to tolerate dissent in normal times, when the costs of opposition are lower and the need to bolster the regime's democratic credentials is greater. However, whether opposition elites comply with the autocrat's demands depends on the severity of the crisis and the extent to which their political survival depends on the regime remaining in power. Evidence for this argument comes from a quantitative textual analysis of all floor speeches delivered in the sixth convocation of the Russian State Duma before and after the 2014 Ukraine crisis (n = 54,827). I find partial support for my theory. While there was significant heterogeneity in how opposition elites and ruling party members spoke about key policy areas in the pre-crisis period, their rhetoric on a number of contentious policy issues---especially those directly related to the situation, such as Ukraine---converged in the post-crisis period.
Cabinet Reshuffles in Authoritarian Regimes: Theory and Cross-National Evidence
Manuscript in preparation for submission in fall 2025
Autocrats often use cabinet appointments as a way to share power with rival elites and to ensure their survival in office. But this strategy is a double-edged sword. While these appointments are intended to foster loyalty by providing elites with access to the perquisites of power, they may inadvertently enhance their ability to challenge the autocrat, turning potential allies into formidable adversaries. In this paper, I suggest that cabinet reshuffles---the strategic movement of elites between ministries---offer a solution to this dilemma. I argue that reshuffles protect autocrats from coup attempts by preventing ministers from developing autonomous power bases and networks that could be used as platforms for mobilizing against the regime. I test this theory using data on cabinet reshuffles and coup attempts in authoritarian regimes from 1967 to 2010. This paper contributes to our understanding of the strategies autocrats use to neutralize elite threats to their rule.
Cabinet Appointments and Self-Enforcing Power Sharing After Conflict (with Victoria Staub)
Manuscript in preparation for presentation at the North American Meeting of the Peace Science Society in November 2025
In post-conflict settings, power-sharing arrangements often integrate former rebels into the cabinet to help prevent conflict recurrence by giving them access to power. However, these arrangements may unintentionally increase the likelihood of coups by ex-rebels who, fearing marginalization by the ruler, leverage their access to the state to seize power. We explore the conditions under which post-conflict power-sharing agreements become self-enforcing and effectively deter coups. Our argument is that such agreements succeed only when ex-rebels are appointed to ministries with control over coercive resources, such as defense and foreign affairs. These appointments increase the ability of ex-rebels to credibly threaten the ruler if he violates the power-sharing agreement, thus creating an incentive for rulers to honor their commitments. To test this argument, we analyze a global dataset of cabinet appointments and coups from 1967 to 2015. Our findings suggest that appointing former combatants to ministries with control over coercive resources significantly reduces the risk of coups. This study provides the most comprehensive analysis to date of cabinet appointments and the credibility of post-conflict power-sharing arrangements.
Just Don’t Talk About It: The Effects of Blocking Discussions About Controversial Political Topics in and Outside Classroom (with Matias Tarillo)
Manuscript under review at Perspectives on Politics
A growing body of research suggests that facilitating discussions between members of opposing political parties on controversial issues can reduce affective polarization and increase support for democratic norms. However, little is known about what happens when such discussions are blocked, a scenario that is becoming increasingly common. We seek to fill this gap by focusing on a setting in which discussion is an increasingly contentious and politically charged activity: the university classroom. We propose that students are less tolerant of professors who block discussion of controversial topics because they perceive universities as appropriate venues for such debate. To demonstrate how universities are uniquely positioned as forums for cross-partisan dialogue, we contrast this with the workplace setting, where we hypothesize that individuals may be more accepting of discussion blocking. We find support for these expectations in two pre-registered vignette experiments. The evidence also suggests that political interest and worldview are important moderators: politically interested individuals are more likely to oppose blocking, and the same is true for those with fluid worldviews. Moreover, we show that blocking discussions in the university setting reduces positive emotions among students.
Social Media and the Construction of Identity During Russia’s War in Ukraine (with Graeme Robertson, Sasha De Vogel, and Samuel Greene)
Initial data collection in progress.