Current Projects

My Research

Indiscriminate Violence, Civilian Preferences and Insurgent Activity

I test empirical implications of indiscriminate violence against civilians in Afghanistan in attempt to gauge the effectiveness of security as a selective incentive for insurgent recruitment.

Using a differences in differences design and synthetic controls, I explore the effect of a deadly wave of suicide bombings orchestrated by Pakistan-based Lashkar-e-Jhangvi in Afghanistan's three main cities on civilian's perception of Taliban strength, preferences for the Taliban, and incidences of direct confrontation between insurgent and allied forces.

Post-Conflict Transitions to Autocracy

In this on going project, I investigate how different conflict resolution pathways following mass mobilization affect the outcomes of institution building and survival after revolutions.

Post-revolutionary regimes are amongst the most durable in the world, and they're also overwhelmingly composed of single party autocracies. Using the comparative method, I argue that the resilience of post-revolutionary autocracies is due to the fact that a significant portion of the population has got to be demobilized in order for peace to be achieved, which in turn leads to the creation of institutions that allow for a larger percentage of the population to participate in the Selectorate and Winning Coalition. These institution building processes are characterized by the need for elite cooperation and the consequential revamping of the Wining Coalition, the introduction of national ideologies and mass-based hegemonic parties, and the simulation of political rights.



Projects I'm Assisting On

Battlefield Losses and Combat Tactics

In an ongoing project, Professor Austin Wright examines the impact of battlefield losses on combat tactics in Afghanistan. This study utilizes formally declassified records of combat operations previously housed in the CIDNE system.

We utilize data on lightning strikes to identify ‘as if random’ variation in the occurrence of these battlefield losses. In particular, lightning strikes and storm formation activity influence the ability of coalition forces to engage in MEDEVAC operations, reducing the probability of offensive patrols by coalition/allied forces.

My contributions include data acquisition and processing, as well as participating in creating an instrument with storm data.

Randomized Combat

Austin Wright, Jarnikae Wilson and Konstantin Sonin "test empirical implications of a simple model of combat and information-gathering using highly detailed information about Afghan rebel attacks, military base infiltration, insurgent-led spy networks, and counterinsurgent operations. As rebels gather more resources, their attacks become temporally concentrated: a one standard deviation increase in opium revenue leads to a .3 standard deviation increase in temporal clustering of rebel attacks. In contrast, following abnormal battlefield losses (labor scarcity), the timing of insurgent attacks becomes less concentrated: a one standard deviation increase in labor scarcity increases randomization of attack timing by .12 standard deviations."


My contributions include creating in-depth literature reviews on the creating of synthetic instrumental variables and the use of machine learning in causal inference.