Skip to content

Gangs and Covid

Dr. Carlos Carcach is Professor & Director of the Center for Public Policy at the Escuela Superior de Economía y Negocios (ESEN) in Santa Tecla, El Salvador, and coordinator of ESEN’s post-graduate program in predictive analytics, which offers online instruction in partnership with, using courses from our online certificate program

Dr. Carlos Carcach The focus of Carcach’s scholarly research is crime – in particular, gang activity. I chatted with Carlos about his research into crime in El Salvador as we drove around various neighborhoods in San Salvador last year. Carcach has studied crime statistics in great geographic detail over a decade and a half, and he described how the activities of Salvadoran gangs have spread and evolved through the countryside not uniformly or randomly, but in the form of a growing network of clusters and links.

Gang boundaries San SalvadorA village that has extensive and mature gang activity – protection rackets organized in hierarchical fashion – will penetrate nearby, more innocent areas with tendrils of similar criminal undertakings, tendrils that spread, thicken and mutate. In more technical terms: “A spatio-temporal model of homicide in El Salvador” (the title of his paper in Crime Science in 2015).

I suggested to Carcach that perhaps an epidemiological model might fit the situation – the spread of an infectious disease. The term “epidemiological criminology” was first proposed for this phenomenon in 2009 (see Akers and Lanier, in the American Journal of Public Health). Carcach allowed that there might be merit in this but thought a cancer model was more appropriate. For one thing, cancer tends to mutate as it spreads, changing in form, and the criminal activity he was tracking did likewise. For another, infection tends to produce a defensive response in the body it infects, usually overcoming the infection, while cancer does not excite an immediate counterattack from its host.

Carcach was pessimistic about the potential for an organic evolution of society in which the role of the gangs would wither. He said that their activities at the local level were not new – they had been preceded by local actors whose protection and extortion rackets were similar in character, but not in intensity. The gangs, informed and strengthened by their US experiences and connections, simply took control of these activities, and gave them greater intensity.  They built the local units up into a hierarchical organization resembling a franchise operation, with an average size of 15 members in the bottom-most units. When they got bigger than that, they tended to split up.

Carcach believes that only the extension and strengthening of basic government services (water, sewage, education) into rural areas would provide the external reinforcement of society against gang growth.


I checked back with Carcach recently, curious about the effect that Covid has had on gang activity.  As with so much social science research, it is hard to separate one thing from another. Here’s what he said:

It is not clear whether gang activity has been affected by the Covid emergency or not.  At the beginning of the lockdown, there were some news reports of gangs enforcing it in their territories, and that they declared a sort of temporary suspension of extortion payments. But none of these claims can be verified. Official data indicate a decline in extortions and homicides but it is not possible to separate the Covid impact from the apparent effectiveness of the Plan de Control Territorial. In addition, the El Faro newspaper published a report uncovering an agreement between gangs and the Bukele government, allegedly to reduce homicides and extortions.