Surgical Strikes in the Drug Wars

More than a thousand people die each month in drug-dealing violence in Mexico, and the toll has been rising. In some parts of the country, the police find themselves outgunned by drug traffickers and must rely on the armed forces. Meanwhile, the United States suffers from the widespread abuse of cocaine, heroin, methamphetamines, and cannabis; violence and disorder surrounding retail drug markets; property theft and violent crime committed by drug abusers; and mass incarceration, including half a million people behind bars for drug offenses and at least as many for crimes committed for money to buy drugs.

Current policies, clearly, have unsatisfactory results. But what is to replace them? Neither of the standard alternatives -- a more vigorous pursuit of current antidrug efforts or a system of legal availability for currently proscribed drugs -- offers much hope. Instead, it is time for Mexico and the United States to consider a set of less conventional approaches.

Most of the illicit drugs consumed in the United States come through or from Mexico, and virtually all the revenue of Mexican drug-trafficking organizations comes from sales to the United States. Thinking of this as a single shared drug problem suggests a shared responsibility for controlling that problem. In the conventional telling, Mexico's role is to limit illicit exports, while the United States should act to shrink demand and domestic production, relying on the standard drug-control triad of enforcement, prevention, and treatment.

The conventional alternative to this conventional wisdom holds that the problem is not drugs but drug laws, and that the solution is therefore legal availability. Since prohibition creates illicit markets, the argument goes, only some form of regulated availability can eliminate the illicit market and the resulting problems. Even under legal availability, say the anti-prohibitionists, prevention and treatment efforts can limit the extent of drug abuse and the damage it causes. Last June's report of the self-appointed Global Commission on Drug Policy -- whose signatories included former presidents of Brazil, Colombia, Mexico, and Switzerland, as well as former UN Secretary-General Kofi Annan -- laid out the anti-drug-war view. But the established understanding and the established alternative share an undue faith in the power of prevention and treatment; the established view also embraces an overoptimistic assessment of the power of enforcement.

A more realistic understanding would take into account the limited capacity of the conventional drug-control triad and the enormous power of markets -- licit and illicit -- to shape behavior. Policies based on that understanding would aim at changing the incentives facing both drug dealers and drug users, with the goal of reducing violence and disorder and shrinking the U.S. prison population. Of course, more logical policies cannot guarantee better results, and even a better result would not be a solution to the drug problem. At best, Mexico and the United States could end up with a result that would be, in the words of the economist John Kenneth Galbraith, merely "distasteful" rather than "catastrophic." But a bad result is preferable to a worse one.
 
THE GREAT ASYMMETRY

Mexico and the United States do not occupy symmetrical positions in the binational drug situation. The United States is central to Mexico's drug problem, whereas Mexico is incidental to that of the United States. Before the mid-1980s, when the heavy use of U.S. naval and air power shut off the Caribbean smuggling route from Colombia to the Gulf Coast, Mexico was not the main source or transit country for illicit drugs entering the United States. But the U.S. drug problem was at least as severe then as it is now. By contrast, Mexico's current drug-trafficking problems relate almost entirely to exports to the United States. In other words, if the United States stopped importing drugs, Mexico's drug violence would shrink dramatically. But an end to Mexican exports would, once new routes and sources replaced Mexico in serving the U.S. market, have only a modest impact on the U.S. drug problem.

If stronger Mexican efforts against drug trafficking could substantially reduce drug abuse in the United States, Washington's repeated demands for more vigorous law enforcement in Mexico would have some real basis. But to call on Mexico to make increasing sacrifices for no more potential benefit than redirecting the flow of illicit trade is surely unjustifiable. The upsurge in violence since Mexican President Felipe Calderón began his crackdown against traffickers in 2007 shows how increased enforcement can lead to increased bloodshed.

This line of reasoning seems to support the reply U.S. officials often hear when they demand that Mexico strengthen its antidrug efforts: that the basic problem is not supply from Mexico but demand from the United States, and that it is incumbent on the United States to reduce the quantities of illicit drugs its residents sell, buy, and consume.

But that goal, too, runs directly into some intractable facts. A small minority of drug users in the United States account for about 80 percent of hard-drug (that is, non-cannabis) consumption and an even larger share of the associated costs of drug abuse, including crime. Among heavy users of hard drugs, about 75 percent have at least one felony arrest in the course of a typical year. Hard drugs account for about 80 percent of the revenue of Mexican drug-trafficking organizations. All this means that reducing the demand for cannabis or the demand for cocaine among casual cocaine users cannot reduce the northbound flow of drugs or the southbound flow of drug money. Shrinking the market would require reducing the hard-drug use of about three million people in the United States who are both heavy users of expensive illicit drugs and also active lawbreakers.

The market forces of replacement and adaptation make the drug-dealing industry resilient even in the face of heavy enforcement: the United States sends five times as many drug dealers to prison today as it did 30 years ago, but this has not prevented the 80-90 percent reductions in the prices of cocaine and heroin over that time, which came as a result of falling dealers' wages and increased efficiency in trafficking. Thus, conventional drug enforcement represents a dead end.

Prevention programs, mostly concentrated on schoolchildren, have only limited efficacy: even the best programs produce only modest gains, perhaps reducing the rate of cannabis initiation by age 13 from 12 percent to nine percent. Reducing initiation is not the same as reducing progression to heavy use; no primary prevention program has ever been shown to prevent addiction. Moreover, such programs have long lag times. Even if a massively successful prevention effort for fifth-grade students began next year, it would not have any substantial impact on hard-drug use until about 2020. Instead of the current mishmash of problem-specific prevention programs, such as drug prevention, bullying prevention, obesity prevention, gang prevention, and so on, the new U.S. National Drug Control Strategy calls for concentrating on building resilient individuals and communities, addressing the common factors underlying a range of personal and social problems. Although such a course correction is welcome, it remains unclear how it will be implemented or how much good it will do in practice.

Treatment offers benefits for some drug abusers; it more than pays for itself by reducing crime and other social costs of drug use. But simply expanding the availability of treatment is only half the battle: most people who need drug treatment (by clinical standards) do not want it. Even supposedly mandatory treatment programs tend to suffer from high dropout rates; the largest, California's Proposition 36, had (even in the early years of this century, when it had full funding) a completion rate of no more than 25 percent and no measurable effect on criminal activity. The most common path out of substance abuse, taken by more than 90 percent of those who recover, is "spontaneous remission": quitting without formal treatment. The one exception is opiate substitution for heroin addicts -- such programs work, and people stay with them. Expanding the availability of substitution could cut into the approximately one-fifth of the U.S.-Mexican drug traffic constituted of heroin. But treatment has little ability to reduce demand in the far larger markets for cocaine and methamphetamines.

Taken together, these facts about drug abuse and the drug trade suggest that the United States' enforcement-prevention-treatment triad has little relief to offer either Mexico or the United States. In fact, some of these efforts, such as eradicating cannabis crops within the United States, actually tend to increase Mexican imports. Moreover, the half million drug dealers now behind bars in the United States constitute a social problem comparable to illicit drug abuse itself. Calls for more intense application of current approaches to drug control meet the philosopher George Santayana's definition of "fanaticism": "redoubling your effort when you have forgotten your aim."
 
LEGALIZE IT?

If cocaine, heroin, methamphetamines, and cannabis were handled in the United States the way alcohol is handled -- in other words, if they were available for sale across the counter in unlimited quantities to any adult, subject only to modest taxes -- the illicit markets would disappear and so would their contribution to violence in Mexico. But at what cost in increased drug abuse?

Alcohol is currently the only addictive intoxicant made available on a commercial basis. There are about four times as many active alcohol abusers in the United States at any one time as there are active abusers of all the illicit drugs combined. The cost, in the form of drunken violence (including domestic abuse), illness, accidents, and incarceration, falls on alcohol abusers, their families, and their neighbors. About half of the 2.4 million people behind bars in the United States were drunk when they committed the crimes that put them there. Thus, alcohol accounts for more incarcerations, as well as more substance abuse, than all the illicit drugs combined.

Consequently, even those most eager to "end the drug war" are, for the most part, reluctant to propose full commercial availability on the alcohol model. But as the Global Commission's report illustrates, opponents of prohibition are equally reluctant to specify taxes and regulations that would prevent an upsurge in abuse without generating a problematic set of illicit markets. Prohibition, even imperfectly enforced, keeps illicit drug prices an order of magnitude above free-market levels. Any set of taxes and regulations powerful enough to prevent drug abuse from soaring would create financial incentives for evasion, and thus a need for enforcement, comparable to those created by prohibition. Even if total drug legalization were a good idea in policy terms, it would be politically infeasible: at the moment, it has almost zero political support in the United States.

Most of the halfway steps proposed by drug policy reformers would offer little benefit to Mexico and other source and transit countries. Some of them could actually make the Mexican situation worse. For example, decriminalization -- ending arrests of users for simple drug possession -- would, if anything, increase the volume of drugs consumed and thus the size of the export market. The reduced incarceration of users, which would be the main benefit of decriminalization, would be felt in the United States, not in Mexico.

Full commercial legalization of cannabis, or some alternative short of full commercialization, such as lawful production for personal use or by user cooperatives, would shrink the revenue of the Mexican trafficking organizations by approximately one-fifth, according to Beau Kilmer and his colleagues at the RAND Corporation: not a dramatic gain but certainly not trivial. Whether trafficking violence would be reduced by a comparable amount is a question for speculation, with no real evidence either way. Mexican drug traffickers would be left with plenty to fight over and more than enough money to finance their combat.

Cannabis legalization would probably lead to a smaller increase in consumption than would be the case for hard drugs, simply because the current cannabis prohibition is less successfully enforced than the prohibitions against cocaine, heroin, and methamphetamines. The benefits of eliminating $10 billion or more in annual illicit revenue and perhaps ten percent of the drug-related incarcerations in the United States might well outweigh the damage from increased abuse.

National cannabis legalization is still a relatively distant prospect in U.S. political terms; the legalization bill introduced last spring in the House of Representatives found only six sponsors. But public opinion has been changing rapidly. In 2010, more than 46 percent of voters in California cast their ballots for full legalization in the state; 45 percent of those responding to a 2011 nationwide Pew survey, meanwhile, supported the legalization of marijuana use, up from 16 percent in Pew's 1990 survey. Still, any relief Mexico might get from such a move is likely to be a decade or more in the future. And cannabis legalization would leave four-fifths of Mexico's current revenue from illicit drug exports untouched.
 
A NEW HOPE

The futility of conventional drug-control approaches does not mean there are no attractive options. If policymakers are willing to adjust their strategies to reflect both the limits of the possible and the relative importance of various goals, then there could be smaller illicit drug markets, less drug-related violence, and fewer people behind bars. Reducing the number of casual drug users should be a far lower priority than reducing the number of criminally active heavy users of hard drugs; decreasing violence is not only more feasible but also more vital than decreasing the flow of drugs.

Antidrug tactics should be chosen to make the best use of scarce resources, especially the capacity to punish. The sheer volume of the current illicit drug traffic, combined with the market's tendency to adapt to enforcement and replace the drugs, dealers, and even dealing organizations taken out of action by the authorities, makes routine drug law enforcement an exercise in shoveling sand against the tide. But if the United States concentrated its enforcement efforts against dealers selected for their contribution to violence and disorder, it would be possible to issue convincing threats against that subcategory of offenders. When deterrent threats are sufficiently credible and clearly communicated, they do not need to be carried out very often.

Coerced treatment for drug abusers is not very successful, both because drug treatment itself is not very successful and because the coercion is generally more nominal than real. But the idea of focusing on criminally active, chronic high-dose users of expensive illicit drugs makes good sense. Although they constitute a small minority of all users, they account for the bulk of the market in terms of volume and revenue, and they frequently find themselves under the supervision of the criminal justice system. Also, felony probationers and parolees with illicit drug abuse problems make up roughly half the population of active hard-drug abusers in the United States. Once these users come under supervision, there is no need to allow them to continue their drug use.

Those on probation or parole are already forbidden to use illicit drugs. But that mandate is not effectively enforced. The threat of probation or parole revocation is too severe (and expensive) to be carried out often and not swift or certain enough to change behavior dramatically. As a result, most violations go unpunished. By reducing the severity of the punishment for breaking the rules, it is possible to dramatically increase its swiftness and certainty -- and swiftness and certainty matter more than severity in changing behavior.

Frequent or random drug testing, with a guaranteed short jail stay (as little as two days) for each incident of detected use, can have remarkable efficacy in reducing offenders' drug use: Hawaii's now-famous HOPE project manages to get 80 percent of its long-term methamphetamine users clean and out of confinement after one year. The program more than pays for itself by reducing the incarceration rate in that group to less than half that of a randomly selected control group under probation as usual. HOPE participants are not forced to receive drug treatment; instead, they are required to stop using. About 15 percent fail repeatedly, and that small group is ordered into treatment, but most succeed without it. Fewer than ten percent wind up back in prison.

These impressive results have led to similar efforts in Alaska, Arizona, California, and Washington State; where the HOPE model is faithfully followed, the outcomes are as consistent and positive as those in Hawaii. The U.S. federal government is set to sponsor four new attempts to reproduce those results. If HOPE were to be successfully implemented as part of routine probation and parole supervision, the resulting reduction in drug use could shrink the market -- and thus the revenue of Mexico's drug-trafficking organizations -- by as much as 40 percent. The potential gains on both sides of the border justify the attempt, despite the daunting managerial challenges.
 
ALL ILLICIT MARKETS ARE NOT CREATED EQUAL

Although antidrug law enforcement has little capacity to shrink the volume of drugs sold, it has a great, but largely latent, capacity to reduce the damage done in the process. Assigning priority to violence reduction would reshape U.S. and Mexican enforcement efforts, albeit in different ways: the United States would have to change its approach to retail dealing, whereas Mexico would need a new strategy for handling the country's six large-scale trafficking organizations. Yet the basic goal would be the same on both sides of the border: to reduce the levels of violence caused by the trafficking and sale of illicit drugs.

All illicit markets pose the threat of violence and disorder, but they are not all equally violent and disorderly. Flagrant retail dealing -- stranger-to-stranger transactions in public places or sole-use trafficking locations such as crack houses -- is especially troublesome compared to discreet dealing in private settings or multiuse locations. When an area is taken over by flagrant dealing, the conduct of the dealers, the buyers, and the robbers who come to prey on them can force ordinary residents off the streets and drive legitimate businesses away. Violence among and against dealers is intertwined with the feuding of street gangs; in neighborhoods plagued by such problems, routine drug law enforcement imposes enormous costs but has nothing to offer in the way of either reducing violence or shrinking the market.

Yet the majority of drug dealers are not violent -- and this variability creates a strategic opportunity. By focusing drug-dealing arrests, prosecutions, and prison terms on the most violent individuals and groups, governments can achieve the double benefit of incapacitating the worst actors and deterring the rest -- not from drug dealing (an incarcerated or deterred dealer will merely be replaced) but from violence, or from the flagrant dealing practices that give rise to violence and disorder. The goals are to suppress violence and to force the market into a more orderly mode, perhaps ultimately to home delivery arranged by telephone or the Internet, and thus make the affected neighborhoods once again fit places to live and do business. The experience of the last several decades has shown the futility of arrests and incarceration in shrinking drug volumes; demand-reduction programs such as HOPE are the best means to that end. More creative law enforcement can at least shrink violence and disorder. The same methods that characterize routine law enforcement -- gathering intelligence, developing informants, and deploying undercover agents -- can be used to identify and build cases against the most violent dealers. In fact, other market participants may even prove eager to give them up simply for self-protection. The main challenge here is not operational but managerial: violence-minimizing enforcement leads to fewer arrests and smaller drug seizures, so law enforcement officials who want to encourage such a strategy will need to adjust their performance metrics accordingly.

Shutting down an entire retail drug market ridden by violence will require a different process. Stranger-to-stranger drug dealing is what the economist and game theory analyst Thomas Schelling calls a "focal point" activity: buyers will come only where they expect to find dealers, and dealers will locate only where they expect to find buyers. Both sides benefit when their numbers overwhelm the police. Once established, such a market is stable, but if the pattern of expectations that holds it together is disrupted, there is no natural mechanism to bring it back into existence.

With half a million drug dealers already crowding U.S. prisons, market-disruption strategies must avoid making dozens or hundreds of arrests and adding to the problem of mass incarceration. David Kennedy and his colleagues at the National Network for Safe Communities have developed a promising tactic. The Drug Market Intervention strategy, first employed to great success in 2004 in High Point, North Carolina, identifies all the active dealers in a flagrant market area, builds cases against them, arrests and prosecutes a handful of the most violent players, and warns the rest, publicly and simultaneously, that anyone who does not stop dealing is headed to prison. By forcing all the dealers to stop at the same time, the intervention causes the market to disappear literally overnight, and the police make only a handful of arrests in the process. (This is another application of the principle that a credible threat rarely needs to be carried out.) Focusing U.S. domestic drug law enforcement on reducing violence and disorder could allow a very substantial cutback in the overall level of drug arrests, prosecutions, and incarcerations. Neither the availability of drugs nor the level of drug abuse in the United States would change dramatically if the current number of drug dealers behind bars were cut in half.
 
KEEPING SCORE

Mexico's different problem calls for a different strategy: creating disincentives for violence at the level of the largest trafficking organizations. Those six organizations vary in their use of violence; total violence would shrink if market shares changed in favor of the currently least violent groups or if any group reduced its violence level. Announcing and carrying out a strategy of violence-targeted enforcement could achieve both ends.

The Mexican government could craft and announce a set of violence-related metrics to be applied to each organization over a period of weeks or months. Such a scoring system could consider a group's total number of killings, the distribution of its targets (among other dealers, enforcement agents, ordinary citizens, journalists, community leaders, and elected officials), its use or threat of terrorism, and its nonfatal shootings and kidnappings. Mexican officials have no difficulty attributing each killing to a specific trafficking organization, in part because the organizations boast of their violence rather than trying to hide it. At the end of the scoring period, or once it became clear that one organization ranked first, the police would designate the most violent organization for destruction. That might not require the arrest of the kingpins, as long as the targeted organization came under sufficiently heavy enforcement pressure to make it uncompetitive.

The points of maximum vulnerability for the Mexican trafficking organizations might not even be within Mexico. U.S. law enforcement agencies believe that for every major domestic distribution organization in the United States, they can identify one or more of the six dominant Mexican trafficking organizations as the primary source or sources. If the U.S. Drug Enforcement Administration were to announce that its domestic target-selection process would give high priority to distributors supplied by Mexico's designated "most violent organization," the result would likely be a scramble to find new sources.

Removing an organization would not reduce total smuggling capacity; someone would pick up the slack. But the leaders of the targeted trafficking group would, if the program were successful, find themselves out of business. The result might be the replacement of more violent trafficking activity by less violent trafficking activity. Less happily, it could lead to a temporary upsurge in violence due to the disruption of existing processes and relationships. But in either case, if the destruction of the first designated target was followed by an announcement that a new target-selection process was under way using the same scoring system, there would be great pressure for each of the remaining trafficking groups to reduce its violence level to escape becoming the next target.

The process could continue until none of the remaining groups was notably more violent than the rest. In effect, such a strategy would condition the traffickers' ability to remain in business on their willingness to conduct their affairs in a relatively nonviolent fashion. This does not mean any sort of explicit negotiation or "treaty" with Mexico's trafficking organizations. Trafficking, even nonviolently, would remain subject to enforcement. But highly violent trafficking would be the target of differential enforcement.

Of course, such an approach would face many challenges: agreeing on a set of metrics, collecting accurate data (especially if some organizations tried to carry out violent actions intended to implicate their rivals), keeping tabs on sourcing relationships, and maintaining sufficient publicity and transparency to avert accusations of corruption. But unlike the conventional approach of enforcement, prevention, and treatment, targeting violence at least has logic behind it. And unlike legalization, it would not cause a huge increase in drug abuse and has a political chance of being adopted. In the absence of another plausible way out of the current situation, it might be worth trying.

In 2010, as Antonio Maria Costa was leaving his post as executive director of the UN Office on Drugs and Crime, he remarked on the implausibility of either "a drug-free world" or "a world of free drugs." Liberating the debate over drug policy from the grip of those twin chimeras might allow the makers of drug policy on both sides of the Rio Grande, and throughout the rest of the world, to seek a set of policies less destructive than those of the past 40 years.