Around the world, significant heterogeneity exists in the approaches countries use to wage the war on drugs. Demand-side approaches include prevention efforts, treatment for abusers, and increases in the cost of use through enforcement efforts and punishment. Supply-side approaches focus on disrupting operations by confiscating drugs and guns, targeting drug precursors, and arresting and punishing traffickers. A crucial question about supply-side policies is whether they generate unwanted side effects, such as violence. Our research examines the impact on violence of a particular supply-side approach that has played a prominent role in Mexico’s drug war: targeting of high-ranked members of criminal organizations, also known as the “kingpin strategy.”
Proponents believe this strategy weakens Drug Trafficking Organizations (DTOs) by disrupting connections, damaging reputations, and creating disarray in the ranks; this may, in turn, reduce criminal activity. Detractors, however, claim the kingpin strategy can increase violence as lower-ranked members battle to succeed the eliminated leader and rival groups attempt to exploit the weakened state of the organization. Both effects are possible, implying a need for empirical research.
Mexico is ideally situated for producing and trafficking drugs. In addition to having a suitable climate, it shares a border with the world’s biggest drug consumer, the United States. Drug trafficking has also been able to flourish in Mexico because of corruption and weak law enforcement. The first DTOs were protected by the government, which designated areas in which each DTO could carry out its illegal activities. In the 1980s, former police officer Miguel Ángel Félix Gallardo — together with Rafael Caro Quintero and Ernesto Fonseca Carrillo — founded the first Mexican cartel in Guadalajara. After incarceration of his partners in 1985, Gallardo kept a low profile and divided up his areas of operation. During this period, the government and DTO leaders had unwritten agreements that DTOs would be allowed to conduct business uninhibited as long as they respected competitors’ territories and refrained from selling drugs in Mexico.
In the 1990s, however, the environment became less stable as Guadalajara’s DTO splintered into four separate DTOs and the Institutional Revolutionary Party lost political power. In the late 1990s and early 2000s, the DTOs became more independent, going from a regimen of political subordination to one of direct confrontation and dispute over control of territory. In late 2005, a new DTO — La Familia — was established in Michoacán followed by a wave of violence. At the beginning of the war on drugs there were five DTOs or alliances of DTOs: Sinaloa/Beltrán-Leyva, Gulf, Tijuana, La Familia, and Juárez. The homicide rate in Michoacán grew dramatically between 2005 and 2006. That said, the national homicide rate continued to be stable at 0.8 per 100,000 residents per month. Nonetheless, on December 11, 2006, newly elected President Felipe Calderón declared war on the DTOs, citing the increased violence in Michoacán as the last straw. While pundits highlighted his desire to have significant reform associated with his presidency, and the fact that he was born and raised in Michoacán, Calderón’s stated reasons for initiating the war was concern about drugs-related violence and criminal groups trying to control entire regions. Calderón’s strategy mainly consisted of a frontal attack led by the army, navy, and federal police seeking the eradication of crops, the confiscation of drugs and guns, and the incarceration or killing of high-ranked drug traffickers (the kingpin strategy). The first operation took place in Michoacán on December 11, 2006, where more than 5,000 army and federal police elements were deployed. Subsequent operations followed in other parts of the country.
Mexico’s war on drugs was initially viewed as a great success: the national homicide rate dropped sharply in January 2007. But the rate jumped back up in March — not quite to its earlier level — and held steady for 9 months. Then, at the beginning of 2008 and, in a clear break from the leveling trend, the homicide rate started to climb, reaching a level 150 percent higher than the pre-drug-war rate by the end of 2010.
This dramatic increase in violence in Mexico has drawn the attention of researchers from different disciplines, and most attribute the increase to Calderón’s war on drugs. Different researchers have focused on the deployment of federal troops across the country, the expiration of the U. S. Federal Assault Weapons Ban in 2004, the increase in cocaine seizures in Colombia, and the increased effort to enforce the law initiated by the National Action Party mayors.
Our research is motivated by the observation that the escalation of violence began in January 2008, the month in which the first cartel leader was captured during the war on drugs (Alfredo Beltrán Leyva). To conduct our analysis, we use newly constructed data on the geographic distribution of DTOs over time — in conjunction with other data sets — to consider the first captures of kingpins associated with each of the five DTOs in operation at the beginning of the war on drugs.
We find that the capture of a DTO leader in a municipality increases its homicide rate by 80 percent, and this effect persists for at least 12 months. Consistent with the notion that the kingpin strategy destabilizes an organization, we also find that these captures significantly increase homicides in other municipalities with the same DTO presence. In particular, we find that homicide rates in neighboring municipalities with the same DTO presence rise 30 percent in the six months after a kingpin capture before returning to expected levels. Further, kingpin captures cause homicide rates to grow over time (to 18 percent above expected levels 12 or more months after a capture) for more-distant municipalities with the same DTO presence. We find little evidence of increased homicide in neighboring municipalities where the captured leader’s DTO did not have a presence.
Several additional pieces of evidence support a causal interpretation of these results. We find no indication that homicides deviate from their expected levels prior to a kingpin’s capture, suggesting that the main results are not driven by efforts that might precede a capture, such as the mobilization of troops into an area. The main results are driven by effects on the individuals most likely to be directly involved in the drug trade: males and, more specifically, working-age males. Domestic violence and infant mortality do not respond to these events. And we present evidence that the drug-war operations did not themselves increase homicides in the first major operations of the war.
This research brief is based on “Kingpin Approaches to Fighting Crime and Community Violence: Evidence from Mexico’s Drug War,” Jason M. Lindo and María Padilla-Romo, National Bureau of Economic Research Working Paper no. 21171, May 2015, http://www.nber.org/papers/w21171. This study was conducted independently of Gabriela Calderón, Gustavo Robles, Alberto Díaz-Cayeros, and Beatriz Magaloni, “The Beheading of Criminal Organizations and the Dynamics of Violence in Mexico,” Journal of Conflict Resolution, forthcoming, http:// jcr.sagepub.com/content/early/2015/06/01/0022002715587053, which also considers the effects of kingpin captures on homicides during Mexico’s war on drugs. The two studies use different methods but arrive at similar conclusions.