Alvarez, Giovanny.

University of North Carolina at Chapel Hill (2026)

This dissertation addresses the critical gap between elite peace and community security, exploring how peace interventions interact with armed actor incentives to redistribute violence between political and societal domains. Developing a "Dual-Arena" framework, the research argues that the institutional design of peace policies determines whether violence is reduced, transformed, or displaced. The dissertation comprises three interconnected essays: a cross-national analysis demonstrating how amnesties covering human rights violations pacify elites while endangering communities; a national-level study of Colombia (1980–2023) examining how state strategies interact with hybrid armed groups; and a subnational investigation revealing how consent-based counternarcotics policies inadvertently incentivize "economic spoilers" to escalate local violence. Collectively, these studies challenge single-outcome evaluation frameworks, demonstrating that elite pacification often comes at the hidden cost of everyday community security.

Alvarez, Giovanny.

Studies in Conflict & Terrorism, 2025

This study introduces and analyzes the concept of the “economic spoiler”—armed actors motivated to preserve illicit economies—to explain how developmental peace policies can fail. Focusing on Colombia’s post-2016 shift to a consent-based counternarcotics program (PNIS), the paper argues that this ‘soft’ policy created a strategic incentive for the Revolutionary Armed Forces of Colombia (FARC) dissidents, acting as economic spoilers, to use targeted violence. They sought to sabotage the state’s program and coerce farmers back into the coca economy. Using comparative interrupted time series analysis, quantitative analysis shows homicides increased most severely where the PNIS made civilians vulnerable targets, providing a novel, actor-centric explanation for post-accord violence.

With Patricia Sullivan & Nathan Marx

Journal of Conflict Resolution (JCR), 2025

This article introduces the US Aid to Security Sector Actors (USASSA) dataset, the product of a collaboration between academic researchers and the nonprofit Security Assistance Monitor. In addition to providing the most comprehensive source of data on US security assistance, the USASSA dataset transforms detailed information about how security assistance funds are spent into aid and recipient typologies that can be used to conduct more sophisticated analyses of how this foreign policy tool is employed, its utility, and its limitations. Our data clearly show not only the magnitude and geographic reach of US security assistance, but also its diversity. While some security assistance is akin to humanitarian aid, other types of assistance blur the line between foreign aid and proxy warfare. We demonstrate the utility of the dataset with an exploration of whether the effects of US security assistance on human rights violations and domestic terrorism vary across types of aid.

With Francisco Gutiérrez Sanín

Conflict, Security & Development (CSD), 2008

This paper discusses a complex process of aid oriented towards taming the Colombian conflict. We show that, despite the grand declarations provided generously by all actors involved, there has been much strategic manoeuvring and inconsistency. We describe how Colombian decision makers made a transition from the intent of establishing peace as a nationalist programme to the effort of internationalising war. However, the transition interacted with—and was somewhat dampened by—the set of constraints they faced. A very important part of these constraints was related to the choice, when seeking key international partners, between the United States and Europe. Donors, at the same time, had their own objectives and constraints, and frequently promoted lines of action that were at odds with their stated objectives. At the same time, the analysis suggests that despite—and sometimes even because of—these limits some windows of opportunity for positive developments have been opened.

Alvarez, Giovanny

Manuscript in Progress

Why does violence persist in communities after intrastate wars end? An amnesty for wartime atrocity forgives the past and licenses nothing that follows, so its later effects in war-affected communities must work through belief rather than law. I argue that such provisions are accountability signals: costly public evidence of where a state will draw its enforcement frontier. Whether that signal becomes violence depends on two asymmetric conditions. Organizational capacity is necessary; illicit rents amplify but cannot substitute for it. Using a multi-period difference-in-differences design covering 83 post-conflict countries between 1980 and 2023, I find that rebel-only amnesties cut political violence while raising societal violence, an increase that more than doubles where illicit markets operate. Mutual amnesties reduce violence in both arenas and are not blunted by those markets, which favors the capacity account over an economic-opportunity rival. State-only amnesties produce nulls.

Alvarez, Giovanny

Book project in progress

Why does violence transform rather than terminate following state interventions? I argue this puzzle stems from neglecting the organizational context of violence. My "dual arenas" framework contends that state strategies — accommodation versus coercion — produce different outcomes depending on whether political and societal violence operate through separate actors (separation) or are fused within hybrid armed groups (entanglement). Neither strategy eliminates violence; both redistribute it between arenas. I test this through a structured comparison of Colombia and Mexico, two cases that arrive at entanglement from opposite directions and sequence their strategies in near-mirror order. In Colombia, insurgent organizations entered illicit economies, converting political actors into hybrid ones; the state moved from coercion under separation (1980–1982) to accommodation under separation (1983–1992), accommodation under entanglement (1993–2001), and coercion under entanglement (2002–2021). In Mexico, criminal organizations acquired political functions, converting economic actors into hybrid ones; the state moved from accommodation under separation (pre-2006) to coercion under entanglement (2006–2018), and back to accommodation under entanglement (2018–present). Interrupted time series analysis of Colombia shows that each policy transition transformed violence trajectories rather than ending them. The 1983 shift to accommodation reduced political violence while societal violence rose. The 1993 onset of entanglement reversed this pattern, with political violence accelerating while societal violence declined through market consolidation. Post-2002 coercion reduced political violence but — contrary to the hydra effect prediction — societal violence continued to decline. Applying the same design to Mexico tests whether these effects follow the strategy-context combination or instead the sequence and origin of entanglement. If organizational context is what moderates strategy effectiveness, Mexico's inverted sequence should reproduce Colombia's cell-specific patterns despite its opposite pathway into hybridity.