star twitter facebook envelope linkedin instagram youtube alert-red alert home left-quote chevron hamburger minus plus search triangle x

New Ideas for Addressing and Measuring Mexican Cartel Activity

In Mexico, violence and corruption, and their connections to illegal drugs in the United States, are a significant concern. While measuring corruption is notoriously difficult, a study in Science estimated how many people are employed by and flow in and out of Mexican criminal organizations (cartels) responsible for much of the violence and corruption. The study created a model to analyze the policy implications of such information.

In an accompanying article in Science, researchers at Carnegie Mellon University (CMU), the University of Maryland College Park, and the RAND Corporation critically evaluate that study’s contributions and also propose ideas for advancing research on this important topic.

“Their study provides the first systematic estimates of individual cartel sizes and total employment,” notes Jonathan P. Caulkins, professor of operations research and public policy at CMU’s Heinz College, who led the article. “The model is an important contribution because there have been few serious attempts to capture what drives cartels’ size or violence.”

Caulkins and co-authors note that having a sense of scale gives a useful sense of perspective.  For example, dividing the annual number of deaths and disappearances of cartel staff (about 6,500) by the estimate of the number who active at any one time (about 175,000) suggests that cartel staff’s annual death risk is about 1.5 times the death risk for U.S. service members serving in World War II (overall, not just per year).  That underscores just how pervasive violence is in the criminal organizations.

“The study assembled a variety of data that was accessible but scattered and integrated those data through a stocks and flow model,” explains Beau Kilmer, senior policy researcher at RAND, who coauthored the article. “This type of model is common across may scientific disciplines but rare in studies of crime and drug policy.”

In their response, the Caulkins and colleagues analyze the study’s use of this model and discuss its applicability to conditions today. Among the challenges identified, they note that counting the stock of cartel employees is a nuanced task given differences between full-time workers and gig workers.

The authors also suggest that given the figures cited in the study, the risk of incarceration may not be the primary cost to be traded off against the benefits of criminal income, camaraderie, and other advantages of cartel membership. The authors offer three suggestions for work on this topic, including research that:

  • Articulates which principles drive cartel staffing and behavior: Among the questions to address are why cartels’ staffing levels have been growing when drug market trends (e.g., the legalization of cannabis in the United States) might suggest the need for smaller staffs, and whether the size of cartel staff drives violence or vice versa.

  • Improves understanding of drug consumption and prices on both sides of the border: Amid severe deficiencies in relevant data systems, the U.S. government should do more to support analyses of illegal drug markets, especially prices and purity across different market levels, to yield information about U.S. drug consumption and the resulting net revenues for drug trafficking organizations at each market level.

  • Expands the scope beyond the murders of cartel members: Adopting a broader vision of how cartels harm social welfare to include violence against journalists, politicians, and law enforcement may widen the policy options available to address the problems.
“Finding ways to incorporate insights about markets and incentives into the type of model used in the study should be part of the next wave of research in this area,” suggests Peter Reuter, professor of public policy and criminology at University of Maryland, who coauthored the article. “A richer understanding of the underlying dynamics of markets may help inform more effective policy innovations.”

The writing of the article was funded by the National Science Foundation.


Summarized from an article in Science, "Modeling Cartels to Inform Violence Reduction in Mexico" by Caulkins, JP (Carnegie Mellon University), Kilmer, B (RAND), and Reuter, P (University of Maryland College Park). Copyright 2023. All rights reserved.

About Heinz College of Information Systems and Public Policy
The Heinz College of Information Systems and Public Policy is home to two internationally recognized graduate-level institutions at Carnegie Mellon University: the School of Information Systems and Management and the School of Public Policy and Management. This unique colocation combined with its expertise in analytics set Heinz College apart in the areas of cybersecurity, health care, the future of work, smart cities, and arts & entertainment. In 2016, INFORMS named Heinz College the #1 academic program for Analytics Education. For more information, please visit