When do armed groups suppress violence rather than perpetrate it? A growing literature finds that economic shocks intensify criminal violence as organizations fight over shrinking rents. But cartels that profit from legal economies also have incentives to contain violence. Instability disrupts supply chains, invites government intervention, and erodes the markets they exploit. I argue that when regulatory institutions make rents durable and contingent on stability, armed groups entering a territory have incentives to choose coexistence over violent competition. I test this argument in Michoacán, Mexico, where access to the U.S. avocado export market requires Mexican authorities to first declare a municipality free of quarantine pests, after which orchards within it can pursue export certification. Pest-free declaration raises the strategic value of territory by shifting production toward export-quality avocados, crops that command higher prices but require years of investment and stable labor. Cartels enter pest-free municipalities to capture these rents, but realizing them requires self-regulation: workers flee violent areas, orchards stop producing, and export certification can be revoked when inspectors face threats. Using a staggered difference-in-differences design that exploits the timing of municipal pest-free declarations, I find that declaration increases the number of active cartels while reducing homicides and increasing overall cartel activity. Positive price shocks raise lethal violence before declaration but have no effect after, consistent with cartels maintaining stable extraction arrangements once territorial value depends on long-term yields and institutional compliance. Cartels enter pest-free municipalities because they are more valuable, but suppress violence upon entry because the rents they seek can only be realized under conditions of order.

How do victorious insurgents build states after civil war? I argue that forced displacement is a key strategy for rebels seeking to consolidate power in areas previously controlled by dissidents: rebel regimes evict politically risky populations from economically valuable areas to destabilize rivals, accumulate resources, and reward loyalists. Crucially, these evictions are tempered where local populations can disrupt vital economic production. Using a novel dataset of weekly district-to-district population flows constructed from cellphone metadata covering more than 7 million users, I provide evidence that after their victory in Afghanistan in 2021, the Taliban used forced displacement strategically to remove politically threatening populations from valuable areas while limiting evictions where local labor and production were economically essential.

Disclosure avoidance techniques distort publicly released U.S. census data in ways that matter for social science research. We reconstruct the confidential U.S. Census Bureau swapping algorithm from public documentation and analyze its statistical effects, comparing the biases it introduces to those introduced by differential privacy. Our tools help researchers identify and adjust for these distortions in applied work.

Territorial control is one of the foundational concepts of conflict research. We trace how the concept has evolved across three generations of scholarship on civil war and identify key conceptual and measurement challenges for future work, with a focus on rigorous causal inference and measurement strategies.

We examine the effects of state-led land redistribution during the post-conflict period on development and conflict reoccurrence. Leveraging quasi-random variation in municipal assignment under Colombia's 2010–2011 Land Restitution Law, we evaluate how returning land, especially to women, shapes outcomes including violence, economic inclusion, and health, drawing on an administrative panel that tracks conflict victims in space and time.

While drug trafficking organizations are often modeled as profit-maximizing firms that compete through violence, empirical evidence reveals significant regional variation: some areas suffer intense conflict while others sustain stable multi-cartel coexistence. Drawing on rationalist theories of war, I develop a formal model linking violence to features of the local economy, emphasizing how demand elasticity to violence shapes cartel strategy. In tourist destinations, for example, cartels often avoid violence to protect the value of an expanding tourism market.

Internal displacement during and after conflict is one of the greatest predictors of long-term poverty and vulnerability, yet high-frequency evidence on how war termination reshapes migration is scarce. Drawing on the universe of user-level cellphone metadata from Afghanistan's largest telecom network, this project measures how the end of large-scale fighting in 2021 immediately changed individual migration decisions, providing the first measure of internal displacement that maintains consistent measurement standards across both wartime and postwar periods.