Most consumers use the drug relatively infrequently and for relatively short periods of time

Reuter characterizes cocaine as a “career” rather than an “event,” because as they come to appreciate the harmful consequences of the drug, the casual users quit, leaving in place a core of more serious users. Marijuana consumption is concentrated among individuals in their late teens and their twenties.Taken as a whole, these drug use distribution patterns suggest that the most severe problems stemming from drug use are concentrated within a relatively small percentage of users. The National Council on Alcoholism and Drug Dependence relies on a 2001 study by the Schneider Institute for Health Policy at Brandeis University, Substance Abuse: The Nation’s Number One Health Problem, for the claim that about 18 million Americans have alcohol problems and 5 to 6 million Americans have drug problems . A similar set of estimates—not of “alcoholics” and problem users but of abusive or “dependent” users—comes from the National Survey on Drug Use and Health from 2007 . The study found that in 2007, approximately 22.3 million people aged 12 or older had, in the past year, abused or experienced dependence on alcohol, illegal drugs, or both: 15.5 million abused or depended upon alcohol, 3.2 million on alcohol and illegal drugs, and 3.7 million on illegal drugs but not alcohol .The plight of drug users who are deemed to be in the thrall of addiction, abuse, or dependence is central to understanding illegal drug policy. Those who advocate maintaining severe criminalization frequently raise the specter of ballooning addiction to make depenalization or legalization seem intolerably reckless. They argue, consistent with the dominant understanding of addiction today as a disease,rolling bench that once addiction sets in, individuals find themselves caught in a pattern of self-destructive behavior that is nearly impossible to escape. Severe sanctions for use, on this account, offer a strong incentive to avoid initiating the addiction cycle and to get out of it once it begins.

Yet there are fundamental tensions within the viewpoint just described. First, if punitive treatment of users reduces the number of people trying illegal drugs , it may conflict with the aim of providing problem users with the therapeutic treatment they need to wean themselves from drugs. It is not simply that governments with punitive dispositions divert resources that could be used on treatment programs. Punitive criminalization may create fears of punishment and demonization that directly discourage users from seeking treatment. The Economist recently noted that in Portugal, which decriminalized possession of illegal drugs in 2001, “[t]he number of addicts registered in drug-substitution programmes has risen from 6,000 in 1999 to over 24,000 in 2008, reflecting a big rise in treatment ” .The United States has tried a different approach, using criminalization as a vehicle to promote treatment by sometimes offering it as an alternative to, or a means of reducing, criminal penalties.The greater deterrence value of more immediate and likely sanctions seems especially important given the apparent risk and time preferences of drug users—individuals whose behavior suggests a present-moment orientation and a heavy discounting of future burdens. An experiment with offenders on probation in Honolulu, for example, which tested the effect of a program oriented around imminent but short incarceration for violators, found that program participants were 55 percent less likely to be arrested for a new crime and 72 percent less likely to test positive for drug use . Third, to the extent that addiction means a lack of voluntariness on the part of the addict, sanctioning addicts with the full brunt of criminal law is in tension with the core American criminal law requirements of mens rea and actus reus. If drug addiction is characterized by involuntariness, then addicted users appear neither culpable to a degree meriting criminal sanctions nor likely to be deterred by such punishment.

On the other hand, if drug addiction is a disorder of choice, as some have recently and compellingly argued on the basis of strong evidence that most addicts recover,then an internalities-based justification of criminalization is weakened: addiction begins to look less like an irreversible step into self-destruction and more like a habit that individuals will struggle over, but quite likely eventually overcome. The greater the degree of choice involved, the less catastrophic is initiation into use and even addiction, and the less justifiable are the costly sanctions designed in large part to keep individuals from ever experimenting with illegal drugs. The significance of addicts in aggregate marijuana and cocaine consumption is important to bear in mind when considering the effect of changes in price—and policy shifts that would affect a drug’s price, such as changes to criminal sanctions, depenalization, or legalization—on the prevalence and intensity of use. Initially, one might assume that non-addicts and prospective dabblers would be more responsive to changes in price than addicts, whose compulsive behavior is often equated with an inability to quit, rising costs notwithstanding. According to this line of thinking, marginal increases in price—through, say, more severe criminal penalties—would affect casual users much more than heavy users, thus decreasing aggregate harms of use only by changing the behavior of marginal users, without substantially diminishing the core of problem users. A lower price resulting from more lenient policies would induce some new users whose intensity of use would be harder to predict—though a reasonable assumption might be that the new group would contain no greater percentage of addicts than the initial population of users.However, it is also possible—and consistent with the economic model of rational addiction put forth by Becker and Murphy —that addicts will be responsive to price changes over longer time spans. While neither we nor Becker and Murphy believe that all addictive behavior can be explained adequately as the rational pursuit of welfare maximization,their model highlights an important theoretical consideration in attempting to assess the impacts of actual and hypothetical policy changes on use: time horizon.

One reason among many for caution in extrapolating from the results of short-lived policy experiences is that a policy affecting the use rate of a drug through the price mechanism may have a substantially greater impact if retained over a long period of time. Provocatively, the model put forth by Becker and Murphy suggests that in the long run, consumption of addictive goods may be even more responsive to price changes than consumption of nonaddictive goods.In the long run at least, price changes may indeed expand or contract the core of problem users. Another factor to take into account when considering consumption across users is the age at which users are most likely to become addicted to illicit substances. In surveys of individuals in the United States, psychiatric researchers have found that drug abuse disorders, excluding alcohol, have a lifetime prevalence of 8.5 percent, and that age is the median age-of-onset for such disorders . More importantly, these same surveys indicate that drug abuse disorders have a narrow age-of-onset range, with an interquartile range of 17-23 years . This suggests that efforts aimed at curtailing drug use among young people can play a key role in preventing drug addiction. Individuals who do not develop a disorder by their late twenties are much less likely ever to develop such a disorder.Therefore, finding ways to limit access to drugs among children and teens should be central to any regime—and any depenalization or legalization proposal. Simply identifying the trends in illegal drug use over time is a difficult task,dry rack cannabis given the obvious obstacles to securing accurate information about illegal behavior over extended periods of time. A number of surveys of illegal drug use include the percentages of individuals reporting to have used in the past month, past year, and ever. These are often broken down by drug and user characteristics—most notably age group. While the percentage of a population using a drug during a given time period is a valuable measure, all such statistics are limited in that they do not capture other important variables such as quantities and potencies used by individuals, much less the severity of harms associated with the instances of use.Even if one accepts the accuracy of the data, one must also use caution in analyzing historical data regarding drug use trends. There is a natural—but potentially misguided— tendency to equate periods of low prevalence with successful policy and to attribute spikes in the percentage of users with policy failings. Even if prevalence of use were the sole criterion by which to measure the success of drug policy, it would remain extremely difficult to attribute causation to specific policies given the myriad other social factors that influence use. With those provisos in mind, it is worth taking a cursory look at historical trends in the use of marijuana and cocaine in the United States. In Figure 5 we report data from the Monitoring the Future surveys on the percentage of high school seniors reporting use of marijuana, cocaine, any illegal drug, and any illegal drug other than marijuana, within the past 30 days.

While these are relatively narrow measures, we present them not just for the intrinsic significance of use prevalence among late adolescents, but also because they are broadly consistent with overall prevalence and have the important advantage of consistent tracking over a long period of time. As Figure 5 suggests, the percentage of high school seniors who recently used marijuana reached a peak during the late 1970s, declined until the early 1990s, rose during the mid-to-late part of that decade and has since leveled and begun to decline in the 2000s. The percentage of high school seniors who recently used cocaine rose through the late 1970s, stayed high in the early 1980s, rose again in the middle of the decade, then declined by its end, falling until the early 1990s, after which time it rose fairly modestly by historical standards, then leveled. The tight correlation between the percentage using marijuana and the percentage using any illegal drug is broadly consistent with drug use trends—not just in the United States, but globally as well. This correlation is quite common across populations because the percentage of individuals around the world using marijuana dwarfs the percentage using all other illegal drugs. Figure 5 might be taken to suggest that the “Just Say No” campaign of the Reagan years led to a major decline in consumption that was reversed during the more permissive Clinton years, although one must consider whether the Reagan campaign influenced reporting behavior as well as drug use. In addition, scholars have offered two reasons to doubt that policy changes in the U.S. can explain the declines in cocaine and marijuana use from the mid-eighties through the early nineties, the subsequent rise in use during the 1990s or the leveling off in the new century. Current U.S. drug control policy is largely punitive in nature. In 2007, law enforcement agencies nationwide made over 1.8 million arrests for drug abuse violations, more arrests than for any other category of offense . Of these arrests, approximately four-fifths were for possession, with 42.1 percent resulting from marijuana possession and 21.5 percent from heroin or cocaine possession . The Office of National Drug Control Policy has found that the largest cost increases in the war on drugs from 1992 to 2002 came as a result of increased incarceration rates for drug offenses and drug-related offenses and from the law enforcement and judicial proceedings needed to put offenders in prison . Now imprisoning a greater percentage of its population than any other country , the U.S. has less than five percent of the world’s population but nearly 25 percent of its prisoners . The punitive focus of U.S. drug policy is a major component of our country’s record-sized prison populations. The American incarceration rate has increased greatly since President Ronald Reagan’s emphasis on the war on drugs in the early 1980s. The number incarcerated in prison or jail on drug charges is estimated have risen from about 40,000 in 1980 to about 500,000 today—more than the total number incarcerated for all offenses thirty years ago . As of 2004, drug offenders constituted an estimated 55 percent of the federal prison population and 21 percent of the state prison population .Though most arrests involving drug offenses are for possession, most individuals serving prison sentences for drug offenses are behind bars for trafficking offenses, not just possession. In 1999, the most recent year for which the Bureau of Justice Statistics did a comprehensive report on federal drug offenders, simple possession was the most serious offense for only 2.1 percent of drug offense suspects referred to U.S.