At 11:56 am on December 18, 2015, seven Salvadoran gang leaders got on a conference call. Some of them dialed in from prison, others from the streets. Among the subjects they covered that night was a hit list of eight people. Before they carried out the murders, the gangsters on the outside needed approval from those on the inside.

The deliberations began just 19 minutes into the conversation. The first potential victim, according to the indictment filed by Salvadoran prosecutors, was the father of an alleged cooperating witness. A few minutes later, he was given a “green light,” gang lexicon for condemned to death. Other victims followed: one gang member “for not adapting,” another for “snitching,” a woman for being what the gang called “gossipy.” By the time they had finalized the hit list, one hour, 25 minutes, and 44 seconds had passed since the discussion of the first homicide began.

The conversation, which was captured as part of a sprawling indictment of dozens of gang leaders, is important, not just because eight people may have died as a result of it but because it forms part of the violent mosaic of El Salvador. In 2018, the country had 51 murders per 100,000 people, more than ten times the U.S. rate. Thanks in large part to endemic violence, citizens of El Salvador, together with those of the other countries in Central America’s so-called Northern Triangle, Guatemala and Honduras, have accounted for 75 percent of all undocumented U.S. border apprehensions in 2019.

As debates over Central American migration have grown increasingly divisive in the United States, U.S. policymakers have sought to find ways to address its root causes. Perhaps no cause has loomed larger than endemic gang violence: over the past few years, leading experts on violence have studied whether El Salvador (and, by extension, Honduras and Guatemala) can adopt the types of gang-intervention programs that have proved so successful in U.S. cities such as Boston and Oakland, California. Their response, after a yearlong study funded by the U.S. Agency for International Development (USAID) and made public earlier this year, was a resounding maybe.

It is a frustrating result. As Thomas Abt—a senior research fellow at the Harvard Kennedy School, former prosecutor, and former Obama administration Justice Department official—notes in his recent book, Bleeding Out, the United States has seen an unprecedented drop in homicides since the early 1990s, thanks in part to the effectiveness of violence-prevention programs in a number of U.S. cities. We are, it seems, finally beginning to understand what causes gun violence and how to prevent it. Abt thinks that the United States can go much further, and wrote the book as a kind of memo to local and national policymakers. But applying these lessons in countries such as El Salvador may require more systemic changes—and a more sophisticated approach to gang violence—than what we have seen so far in the United States.


The USAID El Salvador study was inspired by the success of a U.S. gang-intervention strategy known as “focused deterrence.” This strategy builds on the insight that, as Abt writes in Bleeding Out, “murder and mayhem sticks in and around discrete people, places, and things.” In other words, the vast majority of violence is concentrated in a few areas and among a tiny fraction of the population. Authorities should therefore focus on reducing homicides block by block.

Focused deterrence was pioneered in Boston in the early 1990s, when police, communities, and Harvard Kennedy School wonks started working together to reduce gun violence in high-crime areas. By collecting and analyzing data and intelligence, the police and wonks were able to identify the most likely shooters and shooting victims in an area (often they were the same people). Then, police and community members organized face-to-face meetings. The community provided the carrot, offering targeted individuals “assistance and support,” which in the best of circumstances are social and economic pathways out of the violent spirals. The police provided the stick. As Abt writes, they tell “group members that they know they are involved in violence and it must stop. If it continues, law enforcement will come down not just on them as individuals but on their entire group.”

In the United States, this approach has proved remarkably effective. Between 1996 and 1998, Boston saw a 63 percent reduction in youth homicide. And after Oakland adopted a focused deterrence strategy, its homicide rate dropped 32 percent between 2012 and 2017. In a 2016 review of 30 crime- and violence-reduction programs, Abt and Christopher Winship, a Harvard-based sociologist, found that “focused deterrence” had “the largest direct impact on crime and violence.” And with help from the National Network for Safe Communities (NNSC), the John Jay College–based project that headed up the El Salvador study, focused deterrence has been applied in over 30 cities across the globe.

MS-13 gang members arrive at a prison in Zacatecoluca, El Salvador, October 2017
Jose Cabezas / Reuters


The goal of the El Salvador study was to evaluate whether focused deterrence would work in one of the most homicidal places on earth outside of a war zone. Yet, as its lead researchers noted, focused deterrence depends on three pillars: outside experts, the local community (which must establish strong norms around nonviolence), and law enforcement. All of these actors must have legitimacy—especially the police, who need to carefully select and approach their targets so as not to provoke a backlash from the community.

In El Salvador, all three pillars are weak or absent. For nearly two decades, Salvadoran law enforcement has employed a so-called Mano Dura (Hard Line) security policy, rounding up and incarcerating suspected gang members en masse. This has eroded police-community relations and public confidence in the system. What’s more, in recent years police have been implicated in dozens of extrajudicial executions of suspected gang members, destroying their claim to be fair arbiters of justice.

There are, similarly, few community norms around nonviolence. Although nearly 30 years have passed since the end of El Salvador’s brutal civil war, conflicts are still regularly resolved through violence. Where gang violence is not prevalent, vigilante justice is. And domestic violence is a perennial problem: on average, 18 criminal complaints about violence against women are filed every day, but only five percent of them are ever resolved.  

Religious communities—some Catholic, but mostly evangelical—are the only stable institutions that offer an escape from gang life. Gangs typically kill members who try to leave but make an exception for those who find religion. Yet even this exit strategy is fraught with danger: as I have found in my own research, gangs monitor their former members, killing those who commit to religion but slip back into alcohol and drug use. And they never forgive members of rival gangs, regardless of whether they join a church.

Can focused deterrence work in one of the most homicidal places on earth?

Finally, while El Salvador has high-quality gang investigators, it lacks innovative violence-reduction researchers. There are, quite simply, no local scholars who delve into what is politically perilous territory. In the land of Mano Dura, those who talk about an alternative risk getting burned—not just by a vengeful state but by the bloodthirsty gangs themselves.      


In Bleeding Out, Abt argues that violence-reduction strategies should steer clear of working with gangs. “By focusing on group-oriented violence, not the group itself,” he writes, “the strategy can pacify gangs while draining them of their status, power, and reason for existence.” In the United States, this approach makes sense. Gangs are less organized, violence more diffuse. In El Salvador, however, the gangs are a grade more hierarchical, a level more sophisticated, and a lot more violent.

But might it still be possible to focus on the worst actors? The authors of the USAID report determined that a small part of the population—between 0.34 percent and 0.9 percent—was responsible for a disproportionate amount of the violence. They recommended that any program bring “the capacities of law enforcement, community, and service providers to bear on the behavior of [these] several hundred people.”

This would presumably include the gang members who participated in the December 18, 2015, conference call. One of them was eventually arrested in Guatemala and returned to El Salvador to face trial for those and other crimes. But it’s unclear if there were any repercussions for the other six gang leaders, several of whom were already in jail. What’s more, the December 18 call was just one of 22 similar calls chronicled by authorities during a three-month period in which the gang authorized 180 homicides from prison. In another phone call, gang leaders in prison greenlit 24 homicides. Narrowing the focus under these circumstances, where so many people are so lethal, will be a much greater challenge. And changing the group dynamic of gangs such as these might be near impossible.  

“The strategies in the book are designed to address disorganized or loosely organized violence,” Abt told me during a telephone interview. “[But] a gang conflict in El Salvador tends to be much more structured than conflicts here in the United States.” Still, despite these differences, the USAID report concluded that focused deterrence could make “a positive contribution” in El Salvador, likening it to the other locales where the NNSC has worked. “This will take time and effort, including re-imagining the analysis of groups, invigorating improved approaches to coordination, and integration of partners previously excluded from the conversation,” they wrote. “Yet all of these are tasks that are typical of any city choosing to adapt a focused deterrence approach.” In other words: Just stay focused.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now