Citing evidence that a high percentage of arrestees test positive for alcohol and various illegal drugs, advocates of continued criminalization frequently imply, contrary to the implications of the ONDCP cost study, that toxicologically induced crimes are more common or costly than those whose origins are systemic to drug prohibition. Data do show a correlation between crime and illicit drug use that is, upon first consideration, quite distressing: the 2008 Annual Report of the Arrestee Drug Abuse Monitoring Program found that in 2008, among ten major metropolitan areas across the country, the percentage of arrestees testing positive for the presence of some illicit substance ranged from 49 percent in Washington, D.C. to 87 percent in Chicago. However, as we will reiterate in the sections that follow, extrapolating from the ADAM II results to a belief that drug criminalization decreases crime or violence conflicts with a number of theoretical considerations as well as considerable empirical evidence concerning the relatively greater importance of systemic offenses. Three theoretical points should be highlighted. First, as previously noted, the approximately 1.8 million annual arrests for drug abuse violations are more than for any other category of offense . It is neither surprising,grow rack systems nor indicative of a causal relationship between drug use and crime that individuals in this subcategory of arrestees frequently test positive for illegal drugs.

Second, any causal extrapolation from the correlation between drug use and crime runs up against the intractable problem of omitted variables bias: it is quite likely that factors which predispose individuals to frequent use of drugs also push them toward both crime and greater likelihood of apprehension by authorities. This is especially true for marijuana: detectable traces may remain in one’s system for extended periods of time, so one may test positive upon arrest even if the last instance of use occurred days or even weeks before the arrest, and before or after the commission of the offense . Third, the important question is not whether crime systemic to prohibition substantially outweighs toxicologically-induced crime—although, the best evidence supports this hypothesis. Rather, the appropriate inquiry should be into how the marginal decreases in systemic crime would compare to the marginal increases in toxicologically-driven crime given a regime change. Even if lesser penalties, depenalization or legalization would increase use, the new class of users—individuals formerly deterred by criminalization—would constitute a class much less predisposed to commit other crimes than the group of people already using under criminalization. Return to the ONDCP’s aggregate cost study and three of its key insights: roughly forty percent of the current costs of illegal drugs in the United States are crime costs borne by offenders via incarceration and the government via administration of the criminal justice system; these costs dominate the victim-borne costs of drug-related crime and health-related costs of abuse; and the greatest driver of these costs is crime systemic to criminalization, rather than crime motivated by toxicology.

Together, these propositions suggest that a substantial portion of America’s current drug problem is its drug control policy. Since government policies create some of the costliest of all the burdens associated with illegal drugs, a substantial reduction in the social costs of illegal drugs would seem to require a reduction in the costs imposed by the current criminalization regime, not just a restraint of the costs of abuse. While many advocates of legalization and continued criminalization of illegal drugs see sufficient similarities across drug classes to paint with broad strokes, we perceive the nature and extent of the harms associated with each drug to call for careful, individualized analysis.That is not to say that recreational drugs do not share certain similarities or that society’s experience with legal drugs cannot provide insight into the likely impact of legalizing a currently proscribed drug. The gaping disjunction between the law and policy toward cigarettes and alcohol on the one hand, and toward marijuana, cocaine and other currently illegal drugs on the other, appears less the result of thoughtful distinction than of inertia and a self-perpetuating myth that drugs accorded legal status are qualitatively similar to each other and different from drugs that are criminalized.But if a unified approach across certain drugs might be desirable for a variety of reasons, only by meticulously examining each drug’s unique psychopharmacological effects and social attributes can we begin to group together the different drugs that should be treated similarly.In this section, we consider potential changes to America’s policy toward marijuana and cocaine. To oversimplify somewhat, marijuana is the most widely used illegal drug, one of the least dangerous for users across various dimensions, and the frequent subject of debate over policy reform.

Likewise, any decrease in social costs stemming from a change to marijuana policy is likely to be far smaller than would result from a comparable policy change concerning a “harder” drug such as cocaine. On the other hand, because the social costs under America’s current drug regime are highest for cocaine,changes to policy toward cocaine would change the social cost mitigation calculus in a way that would countenance potential risks and rewards of the greatest magnitude. Marijuana use has intruded into mainstream America to a greater degree than any other illegal drug.Moreover, Room et al. observe that because marijuana’s global prevalence so exceeds that of other illegal recreational drugs, the bureaucracies of drug control within individual countries and at the global level depend upon the criminalization of marijuana to broaden the scope of their mission. They note the World Drug Report 2008 estimates that 65 percent of global seizures and 67 percent of “doses” seized were for cannabis and argue, using global use figures, that without cannabis illegal drug use would not be a global population-level issue . Finally, the therapeutic potential of marijuana has given rise to a debate over whether doctors should be allowed to prescribe the drug for medicinal purposes. Perhaps for all these reasons, marijuana has proven an attractive target for advocates of legalization, though many prominent opponents endorse a continued hard-line stand. Growing numbers of commentators in the popular press have advocated the legalization of marijuana , and assessed the revenue boost legalization might provide states facing cash strapped budgets . The debate has been further stimulated in recent months as states have begun reacting to Attorney General Eric Holder’s announcement that the DEA will no longer raid state-approved medical marijuana distributors . Willingness to consider—if not outright endorse—legalization of marijuana has also grown among academics. Over 500 economists,rolling flood tables including three Nobel Laureates,signed an open letter to the President, Congress, Governors, and State Legislatures expressing skepticism about current marijuana policy and calling for open debate over a shift from prohibition to taxation and regulation. The letter highlights Harvard economist Jeffrey Miron’s 2005 report The Budgetary Implications of Marijuana Prohibition, which estimates that legalization would save the federal and state governments a combined $7.7 billion in prohibition enforcement expenditures and yield approximately $2.4 billion in tax revenues if taxed like an ordinary good or as much as $6.2 billion if taxed similarly to alcohol or tobacco .In the wake of the recent economic downturn, old and new reformers have latched on to the “lost revenues” argument for legalization. This section first considers the case for reforming marijuana policy, specifically weighing the costs of depenalization and legalization against those of the current system of prohibition.First, a completely unregulated market for marijuana would lead to undesirably high levels of consumption—either because of negative externalities or internalities .

Second, regulation and taxation will not adequately correct for these market failures. Third, severe criminal sanctions for users and sellers are cost-justified deterrence mechanisms for reducing use. A more sophisticated version of this third argument would make explicit an important hypothesis frequently left implicit but nevertheless underlying much thought about drug policy: criminalization may not only raise the price for the user but also, through the normgenerating or socializing effect of the law, actually alter individuals’ preferences such that for any given price, use and distribution hold less appeal.54 Though some libertarians argue that the value of individual autonomy dictates allowing marijuana use and simply sanctioning user behavior when it directly infringes upon the liberty of others, a cost-minimization approach demands consideration of the magnitude of social costs of use before accepting the notion that autonomy can trump all such social costs not generated directly from physical force or fraud. Few dispute that marijuana creates at least some externalities and also internalities—certainly at least in the case of minors not yet capable of adequately processing the risks but also perhaps for the 1 in 11 who becomes dependent on the drug merely from trying it. The great contention is over which policies can most efficiently mitigate the total costs associated with marijuana use—in other words, which policies will yield the lowest total social costs, combining the costs of use and control. The crux of the argument in favor of retaining the prohibitions on use, possession, and sale of marijuana is that eliminating any of these sanctions would increase marijuana use by reducing the cost and decreasing the risk. Full legalization might also stimulate demand by enabling advertisement and brand development. Increased use—either in terms of intensity and frequency or number of users—would in turn increase the costs of use borne by users themselves and society. There are also two related, subsidiary arguments, worth addressing. First, it is often contended that marijuana is a “gateway drug” that renders its users more likely to begin using other, more dangerous drugs, and, therefore, an increase in marijuana users as a result of depenalization or legalization would in turn increase the number of users of other illegal drugs.55 Second, it is sometimes argued that marijuana use induces crime. Before turning in subsequent sections to the evidence regarding expected increases in marijuana use under depenalization and legalization, it is helpful to consider briefly the insightful analysis of the gateway issues offered by MacCoun and Reuter. Though they believe that “there is little evidence that expanding marijuana use does increase the use of other, more harmful drugs,” MacCoun and Reuter present a taxonomy of seven possible meanings of the gateway concept: the first step; the spurious correlation; the early warning; the trap; the tantalizer; the toe in the water; and the foot in the door . The basic problems for an econometrician attempting to identify whether—and if so which—gateway hypotheses reflect actual experience are omitted variables bias and endogeneity. At the level of the individual, it is difficult to pinpoint a gateway mechanism because it is quite likely that underlying characteristics that predispose individuals to use marijuana also increase the likelihood of using other drugs. At the population level, it is difficult to assess the effect of marijuana use on the use of other drugs for an additional reason: causality likely runs in both directions. However, even without precisely estimating the impact of marijuana use on the likelihood of trying other drugs, one may place a rough upper bound on the extent of such an effect by noting how commonly individuals use marijuana without going on to other, more harmful drugs. In their recent cannabis report, Room et al. write: “Few [marijuana users] go on to use more dangerous illicit drugs; the 1995 US National Household Survey on Drug Abuse found that only 23 percent of 26-34 year olds who had used marijuana at some time had also used cocaine during their lives.” Similarly, the 2007 NSDUH found that those who used marijuana exclusively constituted 53.3 percent of illegal drug users and 73.2 percent of marijuana users . As MacCoun and Reuter remind us, it is also important to understand the mechanism of any gateway effect, assuming one exists at all. If the “gateway” is a matter of individuals becoming comfortable with illegal behavior and black market consumption, then legalization could undermine this gateway effect, even as it increased consumption directly via lower prices to users. The most-cited evidence in support of the hypothesis that marijuana users are driven to crime while under the influence is undoubtedly the ADAM II data indicating that in 8 of 10major metropolitan areas studied in 2008, over 40 percent of arrestees tested positive for marijuana at the time of arrest .