The recent jailbreak of Joaquín “El Chapo” Gúzmán, leader of the Sinaloa Cartel, has been a serious embarrassment for the Mexican government. It also draws attention to a policy that the government has regularly employed for almost 10 years now – leadership removal, also known as the kingpin strategy.

Does arresting or killing leaders of violent groups reduce violence? A growing body of research has addressed this question, and as Jason Lyall noted here at the Monkey Cage, results are mixed.

An article I recently published in the Journal of Politics argues that leadership removal might be effective against relatively political groups like terrorists, but is especially prone to backfire against criminal organizations in particular.

Yes, decapitation can help stop terrorist groups.

On the one hand, for groups such as guerrilla or terrorist organizations, there is increasing evidence that leadership removal can lead to reductions in violence.  One article finds leadership removals associated with shorter campaigns by the targeted insurgent groups, as well as reduced violence in the future. Another study finds that terrorist groups are less likely to endure after they lose a leader.

Research by Jenna Jordan shows the effects of leadership removal depend on such factors as how the group is structured. Decapitation is less likely to harm groups that are more bureaucratically structured, for example. Overall, however, the policy has been effective against at least some guerrilla or terrorist organizations.

But taking out a criminal kingpin actually increases violence. A lot. 

On the other hand, research on Mexican criminal organizations consistently finds the opposite: leadership removal on drug-trafficking organizations usually leads to more violence. This unfortunate possibility has been raised by Mexican security experts such as Eduardo Guerrero and Alejandro Hope, and has been confirmed in several peer-reviewed quantitative analyses (here, here, here, and here). Leadership removal might lead to short-term reductions in violence in certain conditions, but the longer-term effect is usually more bloodshed.

Why the difference? 

Why does the kingpin strategy often work for terrorist groups, but usually backfire for criminal groups?

As I argue in my JOP article, distinctions between terrorist groups and criminal groups are not only theoretical, but have important empirical implications. Arresting or killing a leader of a terrorist group can strike a psychological or public relations blow against the group, sapping crucial momentum or public support. How is al Qaeda’s central branch doing without Osama bin Laden? The PKK without Abdullah Ocalan? Sendero Luminoso without Abimael Guzmán?

Organized crime, however, is unlikely to simply go away if a leader is taken out. Criminal groups often replace leaders, or arrested leaders can continue to run the business from prison, as El Chapo apparently did. And instead of a weakened organization plodding along, criminal groups often fragment into competitive factions fighting each other for territory. Even if one group disintegrates, others emerge to battle for valuable turf. This fragmentation and rivalry, caused by leadership removal, has been a substantial source of Mexico’s increased violence.

The Mexican homicide rate was about 8 per 100,000 people in 2007, according to UNODC data. Then the government started capturing high-value targets, leading to group fragmentation and infighting, and the homicide rate almost tripled, jumping up to 23 by 2011. The homicide rate has since started to decrease somewhat, to 17 in 2014, but it is unclear to what extent this decline can be attributed to the dozens of kingpin arrests and kills. One contributing factor is that the government seems to have gotten smarter about leadership removal, focusing on mid-level leaders instead of top leaders, as arresting mid-level leaders is more likely to reduce violence.

Overall, the effects of leadership removal depend on many factors, as Jordan has argued. With criminal groups, however, arresting or killing a kingpin is unlikely to reduce violence in the long term. More often, capturing a kingpin causes violence to spike.

Mexico will try to re-capture El Chapo again, as of course it should. But if he’s caught again – like the last time – it will be mostly a symbolic victory, instead of anything that substantially reduces violence in the country.

Brian J. Phillips received his Ph.D. in political science from the University of Pittsburgh, and is currently a research professor (profesor investigador títular) at the Center for Research and Teaching in Economics (CIDE) in Mexico City.