Mexico has become known for drug violence, which escalated dramatically from 2006 to 2012, including a ten-fold intensification of cartel-state conflict. However, large DTOs operated peacefully in Mexico for more than a century. Since its advent in the 1930s, the hegemonic Partido Revolucionario Institucional (PRI) actively managed cartels through a highly conditional state-sponsored protection racket. This system began to erode in the late 1980s and eventually collapsed with Mexico's democratization in the 1990s. Though cartels grew stronger from the 1990s onward due to market forces, the triggering factor for cartel–state conflict—and possibly inter-cartel turf war—was shifts in state policy. This chapter makes four key claims: (1) Under single-party PRI rule (1930s–1989), conditionality was high, leading cartels to eschew anti-state violence. (2) Democratization (1990–2000) weakened conditionality, pushing cartels toward violent strategies, though actual anti-state violence remained rare. (3) Fox's limited, unconditional crackdown (2003–2005) pushed cartels to fight back, triggering the onset of cartel–state conflict. (4) Calderón's massive crackdown (2006–2012) severely reduced conditionality, leading to full-blown cartel–state conflict. I also conjecture that attempts to reform policy in a more conditional direction in 2010 and 2011 mostly failed, but modestly increased conditionality in some repressive agencies. This may have induced some cartels to adopt less confrontational strategies. Under Enrique Peña Nieto, state policy seems to have remained largely unconditional, and cartel–state conflict abides.

This chapter introduces the central logics that drive cartels to use violence against the state. Purely defensive violence to physically reduce losses from state repression is one important logic of violence, but cannot explain cartels’ use of threats, retribution, and terror tactics, all common and clearly coercive in nature. I distinguish two coercive logics by which anti-state violence can influence policy outcomes. In “violent corruption”, cartels use threats of violence against police and other enforcers to induce lax enforcement and more advantageous bribe agreements; in “violent lobbying”, cartels use high-profile, terroristic violence to pressure state leaders into making changes to de jure policy. I summarize the factors favoring each and their predicted empirical footprints. Violent lobbying is subject to collective action problems among cartels, and thus unlikely under conditions of turf war; I present quantitative evidence that it was more salient in Colombia (where cartelswere united at onset in a mutual-protection organization) than Rio de Janeiro and Mexico. Overall, violent corruption is the more prominent pathway. I then discuss several logics of anti-state violence that could arise from inter-cartel turf war, potential pathways of causal interaction between turf war and cartel–state conflict, and “general equilibrium” effects of repression when criminal markets are fragmented.

This chapter delineates the concept of “cartel–state conflict” and its relationship to other conflict types and the theories scholars have advanced to explain them. The militarized drug wars of Colombia, Mexico, and Rio de Janeiro have caused destruction on par with some of the most violent civil wars; the peaks of violence involved both intense fighting among cartels for turf and sustained armed confrontation between cartels and state forces. Nonetheless, cartel–state conflict and inter-cartel turf war are logically and causally distinct conflict types: each can exacerbate the other, but each can also occur in the absence of the other, or in widely differing proportions. Cartel–state conflict is also distinct from civil war, a fact which the popular “criminal insurgency” concept has obfuscated. I present a framework for analysis that distinguishes conflict types by central goals and battle aims; introduces the key logics of violence that operate within each conflict type; and discusses the (imperfect) mapping between conflict types, logics, and observed patterns of violent events, i.e., the “microdynamics of conflict.”