Mexican President-elect Andres Manuel Lopez Obrador talks during a news conference after a meeting with the new members of the Senate and lawmakers of his Party MORENA in Mexico City, July 2018.
Mexican President-elect Andres Manuel Lopez Obrador talks during a news conference after a meeting with the new members of the Senate and lawmakers of his Party MORENA in Mexico City, July 2018.
Carlos Jasso / REUTERS

When Andrés Manuel López Obrador takes office as Mexico’s president in December, his biggest challenge will be to stanch the escalating problem of criminal violence. Last year, there were a record-breaking 25,339 murders in Mexico. Armed criminal groups and militias undermine security, particularly in the areas where the illegal drug trade dominates the economy. Local people, alienated from the state, face extortion and abuse but often still look to vigilante groups and criminal organizations for protection.

López Obrador has promised to turn Mexico’s security situation around, largely by focusing on socioeconomic interventions to reduce crime. He has proposed granting conditional amnesty to some accused criminals, and his administration will consider legalizing marijuana and licensing illegal poppy cultivation for the production of medical opiates. The president-elect has also promised to pull the military back from the policing functions it has increasingly assumed.

This program has many promising elements, but will undoubtedly encounter an unyielding reality. Both of López Obrador’s predecessors found the problem of demilitarizing policing fundamentally bewildering, for example. Peña Nieto approached the issue through justice and police reform, but these measures never matured enough to make prosecution and policing effective on their own. In the end his administration deployed almost 53,000 soldiers. Already in August, López Obrador’s officials, too, conceded that withdrawing the military from policing functions would not soon be possible.

Any solution to Mexico’s security problem will require a fundamental recalibration of the state’s capacity to deter crime. To get there, the López Obrador administration must focus on developing a detailed strategy for transitioning to civilian policing and the rule of law. Instead of targeting criminal bosses, law enforcement should build the intelligence capability necessary to go after the middle operational layer of criminal groups. And the new administration must act resolutely against anti-crime militias. Without such serious shifts in policing strategy, the incoming administration’s best laid plans for socioeconomic outreach, amnesty, and poppy licenses will make at best a marginal difference in Mexico’s criminal violence.


The López Obrador team characterizes its proposed amnesty law as transitional justice for Mexico. Called Abrazos, No Balazos (Hugs, Not Bullets), the amnesty is not for top-level traffickers. Rather, it would be open to those who have committed “political” crimes or nonviolent crimes associated with drug trafficking: cultivating poppies, transporting drugs domestically, dealing, and providing nonviolent services to drug trafficking groups, such as working as a lookout.

Any solution to Mexico’s security problem will require a fundamental recalibration of the state’s capacity to deter crime.

No one who committed a serious crime, such as torture or murder, would be granted amnesty. The amnesty would exclude high-ranking members of criminal organizations, public officials, and members of the military and police. Only members of vulnerable groups—youth, farmers, and indigenous groups—would be eligible, and those who request amnesty would have to fully disclose their crimes, turn over all illegally obtained assets, and stop associating with criminal groups. Petitioners would also be required to participate in the judicial and truth-telling processes, including by testifying against those who do not receive amnesty, as well as to contribute to victims’ reparations and refrain from committing new crimes or carrying weapons. The government would create a special tribunal to administer the law.

The incoming administration will have many questions to answer before the amnesty proposal becomes law. How will the government define “youth”? What investigative capacity will the tribunal have to determine whether applicants’ disclosures are accurate and complete? How will it ensure that those granted amnesty are not intimidated or killed, given that they will be required to testify against other criminal group members?

Socioeconomic programs, too, require specificity of purpose when they are intended to fight crime. When such initiatives are well designed, resourced, and implemented, they can be quite effective. But success in this area has been elusive in Mexico. The Peña Nieto administration launched the National Program for Social Prevention of Violence and Delinquency in 57 specially selected zones called polígonos (polygons). The program cost the Mexican government some $190 million annually in 2013 and 2014. But its outcome was difficult to monitor or evaluate, because no one had ever specified exactly how its initiatives were meant to reduce crime. At first the emphasis was on providing cultural programming, extracurricular activities, and psychotherapy to Mexico’s youth. Later the program expanded to include everything from distributing eyeglasses to schoolchildren to providing vocational training. The programs that kept kids off streets seemed to be effective, but just how effective they were depended on how vulnerable an area was to violence or criminal recruitment. The initiatives were rarely well integrated with local policing—and by 2016, their funding ran out.

Peña Nieto’s program followed Todos Somos Juárez, an expensive and similar effort from the Felipe Calderón administration. Though Calderón’s initiative had some good outcomes, these were only loosely connected to crime. Deficiencies plagued the program, which was eventually discontinued. Before launching a new anti-crime socioeconomic initiative, the López Obrador administration would do well to study what has worked, what has not, and why.


López Obrador and his advisers have been rightly critical of Mexico’s long-standing approach to criminal interdiction, which relies on a strategy known as high-value targeting, in which police focus their fire on the top leaders of criminal groups. Members of the new administration point out that despite the capture of 120 top criminal leaders since 2006, violence has only grown.

High-value targeting is not only ineffective but often counterproductive. Arresting or killing the leaders of one drug trafficking organization in Mexico’s multipolar criminal market does not strengthen the relative power of the state, nor does it leave another drug trafficking organization to peacefully dominate the criminal market. Instead, this kind of intervention undermines the preexisting balance of power among such groups. This creates new temptations for rivals to increase their territories and internal challengers to secede from existing organizations.

Drug trafficking groups, even more than terrorist groups, have little problem producing new leaders. When their leaders are captured or killed, such groups often splinter, and though they are sometimes weakened they rarely disappear. Constant hits against bosses also disrupt the lines of succession, and younger narcos often have to prove their claims to leadership through acts of violence.

The new presidential administration has yet to explain how it would replace high-value targeting, or with what. The Peña Nieto administration, too, entered office intending to dispense with the practice, but, for lack of alternative ideas, it fell back on high-value targeting within months.

If López Obrador hopes to do better, his officials should consider targeting the middle operational layer of criminal groups. This can more effectively disable criminal groups, particularly if most members are captured at once. But the strategy requires more tactical and strategic intelligence than Mexican authorities currently possess. Building that intelligence picture of the middle layer can be risky, because it takes time, and retaining any single piece of intelligence for too long can become a temptation for corrupt law enforcement officers. In the meantime, officials will have few visible successes to display to the public. Still, a reformed Mexican law enforcement strategy would emphasize intelligence, because knowledge of the criminal landscape is at least as important as the density of police deployments.

But shifting to middle-layer interdiction and strengthening intelligence gathering are not the only law enforcement reforms that the new administration should pursue. Corruption and criminal infiltration are serious problems within local police forces. Mexico’s police need uniform federal training standards and stronger vetting. National and state attorneys general, as well as special federal police internal affairs units, should be able to review local police departments at any time, putting them into receivership if they commit human rights abuses or fail to reduce corruption and infiltration.

Such law enforcement reforms would allow the Mexican government gradually to pull back military units. At first, the units would overlap with reformed police forces. Then the transition from military to police could take place in concentric circles of increasing security, analogous to a counterinsurgency ink-spot strategy.


The López Obrador administration intends to explore legalizing the consumption, production, and sale of marijuana, as well as licensing poppy production for medical purposes. Although the proposal sounds radical, it is not surprising, because farmers in Guerrero and Michoacán, where poppy cultivation is intense, have long been among López Obrador’s key supporters.

Legalization risks provoking the ire of the Trump administration, which strongly favors forced eradication, having sharply demanded it of Colombia, for example. But even bigger challenges lie closer at hand. Tens of thousands of Mexican farmers are likely employed in cultivating illicit drug crops, among which poppy cultivation is especially labor intensive. For this reason, generating jobs in licensed poppy cultivation or finding alternative livelihoods for those involved will be a particular challenge. If all existing poppy farmers were given licenses, what would prevent other poor farmers from switching to opium poppy cultivation? Would these newcomers, too, receive licenses, or would the state try to eradicate the new poppy fields? For these and other reasons, rule of law in the poppy cultivation areas will be both crucial and elusive.

If Mexico is to shift to producing poppies for medicinal purposes, it must comply with the 1961 Single Convention on Narcotic Drugs. This convention mandates that the state prevent licensed opium from being diverted into the illegal trade. The condition will be hard for Mexico to meet, given the country’s weak state institutions, corruption, and pervasive insecurity. In the violent Michoacán and Guerrero areas, even basic monitoring of cultivation and diversion will be difficult. Neither the Calderón nor the Peña Nieto administrations managed to robustly police these areas, despite deploying thousands of military and federal police there. But without a significantly stronger state and law enforcement presence, organized crime can easily penetrate, extort, and dominate the legal drug trade. Already, in Michoacán and Guerrero, Mexican criminal groups extort all kinds of legal businesses, including mining and logging companies, avocado farmers, gas stations, and retail shops.

Where the state is absent, as in Michoacán and Guerrero, militias and self-defense forces have filled the vacuum, and these are vulnerable to cooptation by criminal groups. López Obrador’s transition team has remained mostly silent about the future of these militias, but the new administration will have to address the problem. In Michoacán and Guerrero, self-defense forces arrest people, hold extrajudicial trials, and mete out sentences. In some towns, they have expelled local police forces and government officials. Many militias have been implicated in murders as well as extortion and illegal taxation. Yet the Mexican government has shied away fromdisarming them for fear of violent confrontations and because these self-declared people’s defenders are sometimes locally popular.

Finding an effective way to handle these militias has posed a challenge for successive Mexican administrations. The Peña Nieto government tried rolling some militia groups into something it called the Rural Corps, a sort of supplemental, lightly vetted, deputized police force. Many units refused to join.They continue to engage in criminal activities, human rights abuses, and armed protests. López Obrador will likely face a similar problem with his amnesty proposal. Some militia members, particularly those from indigenous groups, may qualify for amnesty – but what if they show no more interest in it than they did in joining the Rural Corps? What about those who accept the deal, but violate its terms and continue to act illegally?

Some self-defense groups and militias, including those in Chiapas, engage in land theft and ethnic rivalries. To effectively curtail such behavior would mean dismantling the groups and punishing serious crimes with lengthy prison terms. But López Obrador defines himself as an insurgent politician, fighting for the rights of the indigenous and the downtrodden. Having personally engaged in boycotts and protests and advocated extralegal behavior, such as not paying utility bills, he may be sympathetic to such groups and their methods. Mexico needs strong oversight institutions and an empowered civil society if it is to replace vigilante justice with the rule of law. But with his disdain for institutions, the president-elect himself may prove an obstacle to such improvement.


Mexican citizens are rightly exasperated with the unrelenting violence in their country. But the incoming administration has raised unrealistic expectations about just how much security will improve under its watch: it has promised to reduce the nationwide homicide rate from the current 25 homicides per 100,000 people to four per 100,000 by 2021. None of the policies that the administration has proposed offer a clear path to this reduction. Amnesty would not be available to murderers; socioeconomic policies take much longer than three years to bear fruit; and so does police reform. The transition team has yet to articulate a targeting or policing strategy.

There is no quick fix for eradicating violence in Mexico. At the moment, the government lacks the required deterrence capacity to do more than secure a temporary advantage for one among the country’s many clashing criminal gangs. The key to lasting security is to change that.

If the Mexican state is to credibly deter violent criminal activity, it needs a comprehensive law enforcement strategy that goes well beyond the decapitation of criminal groups. Socioeconomic policies for fighting crime are a good idea, but they must be refined, sustainably financed, and integrated with policing for a decade and more. The federal and state governments should give all militias that do not qualify as indigenous police a grace period to disarm and demobilize. After that, the government must confront and arrest any militias that haven’t complied, while moving resolutely against new ones if they arise. Militia members who have not been found guilty of murder, kidnapping, or extortion should be considered eligible for the proposed amnesty.

Mexico’s federal government urgently needs a plan to recover territories ruled or extorted by criminal groups, regardless of whether or not poppy and marijuana cultivation are to become legal in these areas. Without such a plan, legalization will fall flat. The state must supply security forces capable of protecting newly legalized farmers and other businesses from extortion and of preventing crops from being diverted to the illegal drug trade. And licensing drug crops should not substitute for a robust, decades-long investment in rural development and alternative livelihoods.

For the last twelve years, Mexico has seen violence decline only when one criminal group temporarily wins enough turf and advantage to discourage its challengers. These lulls are ephemeral and easily disrupted by changes in the balance of power among criminal groups. A lasting security will come to Mexico only when the state has the capacity to effectively deter criminals and secure the allegiance of local populations. That’s the objective for which the incoming administration must strive.

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
  • VANDA FELBAB-BROWN is a Senior Fellow at the Brookings Institution and the author of the forthcoming Narco Noir: Mexico’s Cartels, Cops, and Corruption.
  • More By Vanda Felbab-Brown