Schaffer Online Library of Drug Policy Sign the Resolution for a Federal Commission on Drug Policy


Contents | Feedback | Search | DRCNet Home Page | Join DRCNet

DRCNet Library | Schaffer Library | Major Studies | Legislative Options for Cannabis



Chapter 2.


Given the importance which is assigned to setting clear goals for any major social policy, it is surprising how infrequently discussion of drug policy is preceded by a precise specification of what the policy is intended to achieve. Too often, goals are either global (to reduce drug use) or unattainable (to eliminate drug use) and allow many interpretations. Similarly, goals are often stated in such a way that they could be achieved through a number of quite contradictory strategies and at vastly different costs (economic, social, and personal). Only infrequently does the discussion address the issue of the costs we are willing to tolerate in order to achieve stated goals. Rarely is detailed consideration given to how we might go about measuring the costs, both of drug use itself and of attempts to control it.

In examining the range of potential regimes for the control of cannabis use, it is important to have a clear idea of the criteria against which we might evaluate their effects. Yet unambiguous and meaningful statements of the objectives of cannabis policies are difficult to find anywhere. Most official statements about drug policy generally, and most discussions about drug control strategies, suffer from one or more of a number of limitations. They may fail to proffer specific goals. They may state goals which are known to be unattainable. They may fail to recognise the existence of multiple goals or, if they do, they may fail to acknowledge contradictions between them or to provide a means for assigning priorities when the goals clash.

All of these failings contribute to a policy debate which lacks precision and intellectual rigour, and whose outcomes depend more on rhetoric and emotion than on logical argument and empirical data. This report starts from the proposition that it is important to know what are the requirements for rational drug policy in general and what are the objectives of cannabis policy specifically. Only when these issues are articulated clearly and can be subjected to full public scrutiny can it be said that the cannabis policy debate has become a comprehensive and rational one. The issues discussed in this chapter can be applied, with suitable modifications as to specifics, to any area of drug policy. In this chapter we do not take any position on particular policy options. Our intention is to provide a comprehensive framework within which any option can be evaluated.

We intend in this chapter, too, to address the issue of drug policy, rather than drug legislation. As our discussion of legislative options will show, similar-looking laws may have quite different implications for cost and effectiveness depending on the way in which they are administered. In other words, the policy context may be more important than the legislation in understanding the impact of a particular approach to cannabis control. For example, in both the United States and The Netherlands it is illegal to possess and sell personal-use quantities of cannabis. The way in which the legislation is interpreted and enforced, however, puts these two countries at different ends of the cannabis policy spectrum (Wardlaw 1992).

In its turn, the policy framework adopted in any particular country can only be understood fully if the observer is aware of the social and political context within which social problems are viewed, and with a knowledge of the nature and dimensions of the drug problem in that country. Some conceptual clarifications Kleiman and Saiger (1992) suggest that discussion of options for drug control policy should be preceded by agreement on some basic conceptual clarifications. These clarifications are intended to define the nature of a satisfactory debate about drug control options.

The following list is a modification and extension of Kleiman and Saiger's suggestions.

  1. Arguments that apply to the most appropriate control regime for one drug need not - and often do not apply to others. There has been considerable resistance to the idea that use of different drugs may best be controlled by different strategies. Certainly, a dominant theme in the drug control policies of many countries is that it is necessary to have consistent policies towards illicit drug use. It is argued that treating different illicit drugs differently, especially when 'differently' implies that one drug will attract less law enforcement attention or less social disapproval than another, will undermine the general authority and credibility of anti-drug messages. This approach overlooks at least one important consideration. That is that there are a number of approaches to the notion of consistency. Consistency does not necessarily imply that we have to have a monolithic approach to drug problems. We can, for example, have the goal of reducing the harm caused by illicit drug use, but can achieve that goal by a number of avenues. There is no logical reason why these avenues have to be the same for all drugs, as long as we can justify why we are treating different drugs differently, and as long as we can articulate these reasons to the public. The belief that we must maintain a rigid adherence to one set of strategies for all illicit drugs ignores the reality that we already quite successfully accept inconsistencies and contradictions in drug policy across the board. The most obvious example is the fact that we allow a number of drugs, foremost amongst which are tobacco and alcohol, to be used legally despite their causing the same types of damage against which we try to protect individuals and society by banning the use of other drugs. There is general agreement that it is the goal of policy on licit drugs to reduce the harm caused by their use. We seek to achieve this by attempting to persuade people not to take up drug use, to decrease their consumption of these drugs, or to alter their manner or pattern of use. Although the simplest way to reduce harm would be to reduce the number of people using the drugs (assuming for the argument here that remaining users continued their existing consumption levels and patterns), we do not see attempts to get others to move to less dangerous patterns of use (rather than giving up use) as undermining the message that use of drugs is not socially desirable. Rather, we recognise that the overall goal of reducing harm can be achieved by persuading some people not to take up drugs, others to discontinue their use, and yet others to moderate their use or to change to less dangerous drugs or methods of administration. Strategies aimed at moderation are not seen as undermining strategies aimed at abstinence. We simply accept that different strategies are appropriate for different groups in the community and that they can all contribute to the overall goal of harm minimisation. There is no logical reason why we cannot adopt the same approach with illicit drugs.
  2. Drug policies should be crafted to take account of the different patterns and types of harms caused by specific drugs. Attempting to impose across-the-board strategies for illicit drug control in the name of consistency (ignoring for the moment the glaring inconsistency in not treating licit drug use similarly) seems to assume that all drug problems are, in fact, equally susceptible to the same measures. We know, however, that this is not the case. Each drug has its own consumption patterns; each drug harms users and others in different ways and to different degrees; and each has its own demand patterns, supply systems and market characteristics (Kleiman & Saiger 1992). Not only do drugs differ in these ways, but the same drug may exhibit different characteristics in different geographical, socioeconomic or cultural contexts (Haaga & Reuter 1990). Accordingly, it is not logical to expect that there will be uniform solutions to drug problems. 'Consistency for the sake of it should not be able to stifle the ability of local areas, states or nations to deal with drug problems in a manner which is appropriate for their circumstances, cultural context, dimensions of the problem and so forth' (Wardlaw 1992, p154). It follows that decisions about how best to deal with problems in relation to a specific drug should be made on the basis of the circumstances surrounding use of that drug and the practical possibilities for intervening in them.
  3. The details of control regimes are crucial determinants of their outcomes. Such details should not be left undefined. Much of the confusion over what is the most appropriate regime for the control of cannabis use stems from poor definition of terms such as 'decriminalisation' and 'legalisation', and lack of specification of the details of alternative control regimes being debated. As the South Australian Royal Commission into the Non-Medical Use of Drugs ([1]South Australia 1978) commented, people use these terms in a variety of senses and are often unaware of the significant differences that exist between various methods of control that have been proposed under them. We agree with the Commission that these terms are not very helpful. The word 'decriminalisation' sounds as though it refers to the removal of cannabis control from the criminal law. In practice, however, it usually refers to changes to the law which either allow possession of specified amounts of cannabis for personal consumption or reduce the penalties for such possession to minor fines or administrative sanctions involving no criminal record. As such, decriminalisation refers to reducing penalties, rather than removing them altogether ([2]Single 1989). Under most so-called decriminalisation proposals, dealing in cannabis would remain a serious criminal offence. The term 'legalisation' presents similar difficulties. It covers proposals which would literally remove the drug from the criminal law, to those which would limit availability to adults (much as alcohol and tobacco are now regulated), to more specific proposals which would allow access through clinics or other health services only to those with established habits. Discussing 'decriminalisation' or 'legalisation' without specifying the details of the particular model being proposed is not likely to be very productive (Kleiman & Saiger 1990). No sensible discussion of any of the options can take place without specification of such details as to whom proposed changes would apply and under what circumstances, rules about advertising, place and time of consumption, whether or not the changes would apply to all preparations and forms of cannabis (or, for example, be restricted to those falling below a certain potency), and the type of regulatory arrangements which would replace the current law enforcement ones, together with details of their methods of operation and costs. 'Such details should not be dismissed as easily determined or postponed as a problem requiring future thought' (Kleiman & Saiger 1992, p225). This report will discuss policy options under the broad headings suggested by the South Australian Royal Commission into the Non-Medical Use of Drugs, i.e. the total prohibition model; the prohibition/'civil penalty' model; the partial prohibition model; the regulatory model; and the free availability model ([3]South Australia 1978). Any analysis of control regimes should attempt to estimate their effects on both consumption levels and patterns of use. A major concern about any lessening of the current controls on cannabis is that it might lead to increases in consumption. While this is a serious concern, it is not the only dimension which needs to be monitored. It is also important to estimate how patterns of consumption might change under different control options. For example, if numbers of users rose slightly, but consumption per person fell, or less dangerous forms of the drug were used, or less harmful methods of use were employed, the overall amount of harm occasioned might be reduced and the policy might be judged more successful than the status quo. Similarly, if under one option, it seemed that numbers of users might fall, but that heavier use patterns would emerge, or more potent forms of the drug would become more popular, the option might be judged less desirable, even though a decrease in numbers of users had been achieved.
  4. Control regimes should not be considered in isolation from the problems of implementation and enforcement. In attempting to predict the outcomes for users of various alternatives to the current methods of cannabis control, it is important to specify the costs and problems of regulation that would accompany them. In all options which have any political reality there remains the need to licence, to tax or to regulate to ensure that restrictions are observed on such things as sale to minors, use in conjunction with the operation of vehicles or machinery, potency levels, amounts purchased etc. The costs and practical problems of implementation should be included in any debate about control options.
  5. Arguments about consequences of drug use should be separated from arguments about morals. Much confusion has been engendered in the drug policy debate because of a failure to separate the negative consequences of drug use from moral considerations. It is quite legitimate for a community to invoke moral arguments in determining the proper course of drug policy (Wilson 1990). But it must be clear that the arguments are moral ones, not morality disguised as concern over consequences. If opposition to cannabis use is based on the view that it is illegitimate to use a drug to alter one's perception, to seek pleasure or to become intoxicated (a moral argument), then it should be clearly stated as such so that the proposition can be debated in those terms. It confuses the issue and injects unproductive emotion into the argument if that concern is disguised by reference to the harms caused by cannabis use. These need also to be debated separately. We need to know whether there would still be opposition to cannabis use on moral grounds even if we could eliminate the health costs.
  6. Options should be evaluated on the basis of evidence of damage. We should require that policies that seek to impose expensive control regimes on any drug be based on evidence of harm occasioned by use of the drug. The cost and restrictiveness of the control regime should be clearly related to the damage caused by use of the drug, and there should be empirical evidence of the type and extent of the damage.
  7. Any policy should recognise the changing nature of the drug problem and be able to change with it. Additionally, all policies should be reviewed periodically to ensure that they are still relevant to current circumstances. A major drawback to current approaches to drug policy is that they are generally poor at adapting to changes in the nature of the drug problem or to changes in our levels of knowledge and understanding of drug use processes and patterns and their interactions with different control strategies. All policies should have in-built capacity for evaluation to enable them to be adjusted in a timely manner to emerging patterns of use or changes in our understanding of the problem.
  8. Policy should be made in the light of the costs of control as well as the benefits. Discussion of any policy option should clearly identify any costs associated with it. An attempt should be made to inventory all of the direct and indirect costs associated with the option, including social costs. It is particularly important to include any unintended consequences predicted or discovered. For example, what are the substitution effects of a particular policy (does enforcement success, for example, result in fewer users or do the users turn to other, more dangerous drugs)?
  9. The goals of drug policy should be realistic. Enunciating unrealistic goals for drug policy can have a number of negative consequences. Foremost amongst these are the undermining of the credibility of the policy and the justification of extreme methods in their pursuit (Wardlaw 1992).
  10. Discussion of policy options should include a specification of which harms they are intended to reduce. Specifying the harms intended to be reduced by the overall policy and the options within it clarifies thinking about goals and provides a necessary basis for evaluation. If possible, effects should also be discussed according to the dimensions of time (short-, medium-, and long-term effects), duration (temporary or permanent), and scale (minor, moderate, or major).
  11. Discussion of cannabis policy (and drug policy generally) should recognise the existence of multiple and sometimes contradictory goals. It may be necessary to choose explicitly between goals, to assign them to a hierarchy, or to accept that different goals (and, therefore, different strategies and programs) may be appropriate for different sub-populations or areas. Examples include accepting that it is appropriate to teach established users safer methods of use while trying to discourage other groups from starting use, or police not arresting cannabis users because it is considered to impose an unwarranted burden on the court and corrections systems.
  12. Policies to discourage cannabis use should be shown to be effective or be changed. Sound social policy should be robust enough to withstand critical scrutiny. The costs of current approaches to cannabis control are considerable and their continuation should be guaranteed only if data can be provided which demonstrate their effectiveness. The same criterion should be applied to any alternative approach.
  13. The harms caused by the control regimes themselves should not outweigh the harms prevented by them. It is very difficult to measure accurately the costs both of cannabis use and of efforts to prevent it. Nevertheless, it is not good enough merely to assert that a policy is the correct one, or to fail to investigate alternatives when there is evidence of ineffectiveness (the large number of users) and cost (the cost of law enforcement, the intrusions on civil liberties, violence and corruption caused by forcing sales onto a black market, and so on). Any social policy should be reviewed when there is reason to believe that the costs of administering it outweigh the harms reduced. The goals of cannabis policy In deciding upon the appropriate model to adopt to control cannabis use, it is necessary to specify the goals of the policy, as they will constitute the criteria against which selection will be made (and against which eventual evaluation of performance will be made). The following list comprises possible goals which have been mentioned in the literature on cannabis policy. It is intended as illustrative of the range of goals which might be considered and is not a set of goals proposed by this report:

This list is not intended to be comprehensive. It is merely indicative of the range and detail which needs to be considered when deciding on policy directions and on the resulting legislation and implementation strategies. Examination of such a list makes a number of things apparent. One of these is that to make sense of it requires the assignment of priorities. Priorities may be of two types. The first is simply to assign a ranking in order of importance as judged by some defined criteria. This is useful when attempting to assign resources to different parts of a total strategy. A consideration which may also play a part here is the impact that a particular strategy is estimated to be capable of having. The second involves deciding between two or more conflicting goals. Considerations here are likely to include harms caused or prevented, as well as considerations such as cost and effectiveness. Laying out goals in this fashion also alerts us to the need for high quality data for decision-making, and to the need for sound evaluation methodologies to be able to measure the impact of our policies and their resulting implementation strategies. Examination of the list above makes it immediately apparent that extensive empirical data are required to be able to assign priorities and measure impacts. The list also provides a practical example of the sorts of considerations which policy makers should include in their deliberations as they attempt to apply the conceptual tools discussed earlier in this chapter.


Haaga, J. & Reuter, P. 1990, 'The limits of the Czar's ukase: drug policy at the local level', Yale Law and Policy Review, vol. 8, no. 1, pp36-74.

Kleiman, M. & Saiger, A. 1992, 'Taxes, regulations and prohibitions: Reformulating the legalization debate', in Drug Policy in the Americas, ed. P. Smith, Westview Press, Boulder.

Kleiman, M. & Saiger, A. 1990, 'Drug legalization: the importance of asking the right question', Hofstra Law Review, vol. 18, pp527-565.

Single, E. 1989, 'The impact of marijuana decriminalization: an update', Journal of Public Health Policy, vol. 10, no. 4, pp456-466.

South Australia. Royal Commission into the Non-Medical Use of Drugs 1978, Cannabis: A Discussion Paper, South Australian Government, Adelaide.

Wardlaw, G. 1992, 'Discussion', in M. Bull, D. McDowell, J. Norberry, H. Strang & G. Wardlaw, Comparative Analysis of Drug Strategy, NCADA Monograph Series No. 18, AGPS, Canberra.

Wilson, J.Q. 1990, 'Drugs and crime', in Drugs and Crime: Crime and Justice: A Review of Research, vol. 13, eds M. Tonry & J.Q. Wilson, University of Chicago Press, Chicago.

Contents | Feedback | Search | DRCNet Home Page | Join DRCNet

DRCNet Library | Schaffer Library | Major Studies | Legislative Options for Cannabis