The impact of cannabis decriminalisation in Australia and the United States

Introduction and background

In Australia, the United States and other countries, there is a continuing interest in reform of cannabis control laws in order to reduce the high enforcement costs and adverse individual consequences of the criminalisation of cannabis possession for personal use. Each year large numbers of predominantly young citizens are arrested and prosecuted for cannabis possession, many of whom would not otherwise have criminal records, with little or no apparent effect on rates of cannabis use or cannabis-related harm.

Precise estimates of the costs of enforcing cannabis possession laws are lacking but there is evidence that these costs are significant. Drug enforcement has been estimated to cost over AUD$ 404.2 million in Australia in 1992 (13) and more than US$ 17.4 billion in the United States in the same year (24). In addition to the high costs of enforcement, the policy of cannabis prohibition has entailed other social costs. These include the encroachment on individual rights and freedoms in order to facilitate drug enforcement, the adverse effects of a criminal record for the large numbers of convicted offenders, the impact of the penalties (fines, imprisonment) on convicted offenders, and other adverse social consequences on users such as negative impacts on employment and family discord caused by an arrest for cannabis possession.

The presumed benefit of the criminalisation of cannabis possession is the deterrence of cannabis use. There is, however, little evidence of a strong deterrent effect. Substantial increases in marijuana use occurred in the 1960s and 1970s despite the application of criminal penalties for cannabis possession both in the U.S. (7, 23, 44) and in Australia (17). These trends in cannabis use do not constitute conclusive evidence regarding the lack of a deterrent effect, as it is not known whether rates of use might have increased even more if cannabis possession had not been prohibited. Nonetheless, it is noteworthy that nonusers rarely cite fear of legal consequences as a reason for their nonuse (34, 44). Rather, the simple lack of interest or fear of adverse health consequences are the most commonly given reasons for abstention from cannabis use (33, 44). There is no discernable trend in perceived availability of cannabis despite high levels of drug enforcement in the U.S. (30, 31, 52). Criminological research on the deterrence of other forms of deviant behaviour indicates that deterrence does not generally occur if the risk of detection and punishment is low. This is clearly the case with cannabis use (23). In Canada, which until the late 1980s had the highest per capita rates of arrest for cannabis possession in the world, it was estimated that less than one per cent of users and one tenth of one per cent of use incidents were detected by the police (33).

Thus, in both Australia and the United States, laws prohibiting cannabis possession entail considerable enforcement and social costs, yet they appear to have little impact on deterring cannabis use. In the 1970s 11 U.S. states enacted "decriminalisation" laws which reduced the penalties of cannabis possession to a fine only, and more recently, several jurisdictions in Australia have enacted similar measures. In 1987, South Australia introduced a "civil penalty" approach to minor cannabis offences (including personal use and cultivation), which involves the issuing of expiation notices or "on-the-spot" fines to detected offenders. In the 1990’s, the Australian Capital Territory and the Northern Territory have followed with the introduction of similar expiation schemes. More recently, Victoria implemented a system of cautions for minor cannabis offenders in 1998 and Western Australia has since followed with a similar scheme.

These changes in the law regarding cannabis possession have been referred to as "decriminalisation" measures, but this is something of a misnomer, as cannabis possession is still sanctioned by criminal law in these jurisdictions. With respect to cannabis possession, the term "decriminalisation" has generally been used to describe laws which reduce penalties for small amounts for personal use to penalties other than imprisonment. Even under these so-called decriminalisation laws, the possession of cannabis is against the law, albeit subject to a maximum penalty of a fine only. Therefore, in evaluating the impact of such measures it is important to note that we are only dealing with the impact of reducing penalties rather than the impact of eliminating penalties altogether.

The debate about cannabis law, and particularly the decriminalisation of personal use of cannabis, has revolved around a number of arguments for and against reform. Some of these arguments are supported or refuted by empirical evidence, to varying degrees, and others are arguments which appeal to certain values without the need for empirical evidence. One of the key arguments for decriminalisation of personal cannabis use is that the recording of a criminal conviction for such offences is a penalty that is out of proportion to the nature of the offence, and that a conviction has significant negative social impacts upon offenders who are otherwise law abiding. Indeed, it can be argued that reducing penalties could lead to an increased deterrent effect, as law enforcement would be less reluctant to charge offenders when the punishment is more suited to the crime (38). Related to this is the argument that the criminal prosecution approach entails a greater encroachment upon individual rights and freedoms in the name of drug law enforcement. It is further argued that decriminalisation of cannabis use would serve to separate the cannabis market from the market for other illicit drugs, and send a message that cannabis is not regarded by authorities in the same way as illicit drugs such as heroin and amphetamines. Finally, another common argument is that a decriminalisation approach to cannabis use would free up substantial law enforcement and criminal justice resources, and reduce associated costs to the community (11).

Those who argue against cannabis decriminalisation believe that such an approach conveys a message to society, and especially the young, that cannabis is not viewed as a serious issue, and that it carries little risk, which may in turn lead to increased levels of use in the community. Also, it is sometimes argued that a more liberal approach to cannabis use may lead to greater use of other illicit drugs (the "gateway theory"). It should be noted, however, that most young people start using licit drugs (alcohol and tobacco) before using any illicit drug. It is further argued that the principle licit drugs, alcohol and tobacco, cause a substantial burden of illness and social problems in our society, and that a more liberal approach to cannabis use may result in more health problems arising from its use. In the case of the South Australian expiation approach, opponents of its implementation argued that it would make it easier for people to deal in small amounts of cannabis, with a reduced risk of criminal prosecution. Further, police would find it difficult to enforce a system of on-the-spot fines, and they would have their professional image undermined (11).

Regardless of one’s position on these issues, it is clear that the policy debate, both in the United States and Australia, has not been well informed by research, and different legislative approaches have been introduced in different locations at different times, with little or no planning for evaluating the impacts of the changes, and monitoring them on an ongoing basis. Both sides of the debate have presumed that certain impacts will result from changes in policy, with little or no reference to empirical evidence.

Purpose

The purpose of this paper is to summarise and compare the impacts of cannabis decriminalisation measures in Australia and the United States. In each setting, evidence will be presented regarding the impact on rates and patterns of use and indicators of cannabis-related harm, as well as the impacts on the criminal justice system and to individual users. The paper will also discuss the extent to which the impact of decriminalisation can be properly gauged in each setting, including the extent to which data are available and the comprehensiveness of the evidence. The discussion will conclude with the implications to drug policy reform in other countries, including recommendations regarding the development of similar measures in other jurisdictions based on the experience in Australia and the United States.

Cannabis policy in Australia and the United States

There are number of potentially significant similarities and differences between Australia and the United States in terms of cannabis policy, as well as the underlying political and social environment in which this policy operates. Australia and the United States are relatively large countries, with climatic conditions that are favourable to cannabis cultivation in large areas. Australia is relatively near to the opium producing regions of Southeast Asia, while the U.S. is relatively close to the cocaine producing regions of Colombia and other parts of South America. Both countries have well developed market economies. Both have federal systems of government in which state governments have the primary jurisdiction over laws governing the use or trafficking in illicit drugs, and the coordination of federal and state drug policy is an ongoing contentious issue in both countries. Perhaps the key similarity between the two countries is the reliance on criminal law to deter use. Even in those American and Australian states which have "decriminalised" cannabis, possession is against the law and subject to penalties. In both countries, the fundamental response to drug problems is to expend considerable resources in supply interdiction and the enforcement of laws against the possession of illicit drugs.

On the other hand, there are also dissimilarities between the two countries with regard to the social and political environment in which cannabis policy operates. Australia has a much smaller population and it is more geographically isolated. Whereas Australia has no land border with any other country, the U.S. has long borders with Mexico and Canada, rendering the control of drug trafficking more difficult. The United States has a somewhat greater disparity between the highest and lowest income levels, and sharper racial and ethnic divisions. While aboriginal health is certainly a salient and ongoing issue in Australia, illicit drug use is not viewed as a racial issue, as it sometimes portrayed in the U.S. (40). The social welfare system is more highly developed in Australia, and there is universal health care.

There are also important differences in the basic approach to drug policy, the manner in which drug policy is developed and the role played by different socio-political groups on drug issues. In contrast to the U.S. "War on Drugs" policies, the Australian National Drug Strategy has been characterised by two distinctive features. First, whereas American drug policy focuses exclusively on illicit drugs, the National Drug Strategy is a comprehensive approach, including not only illicit drugs but licit drugs such as alcohol. Second, the National Drug Strategy is based on the principle of harm minimisation (14). Priority is placed on reducing the negative consequences of drug use rather than on necessarily eliminating drug use or ensuring abstinence.

American drug policy, on the other hand, explicitly rejects harm minimisation and focuses on the total elimination of drug use. The basic approach to drug prevention is "zero-tolerance" and more punitive. Senior police officials have publicly advocated capital punishment for drug offences (5:38), and William Bennett, Director of the Office of National Drug Control Policy (the so-called "drug czar") at the time, even suggested on a radio talk show that beheadings might be a solution to the drug problem (31). While such rhetoric represents an extreme expression of the War on Drugs mentality, it is indicative of the context in which cannabis decriminalisation measures operate in the U.S. Decriminalisation has not received support from federal authorities and it is generally opposed by law enforcement agencies and non-government organisations concerned with drug issues. The reduction in penalties for any illicit drug is often seen as inconsistent with the generally punitive approach to drug prevention.

In Australia, law enforcement has been much more supportive of harm minimisation measures such as cannabis decriminalisation (46). Australian law enforcement officials are more likely than their American counterparts to view harm minimisation in a favourable light as one means for implementing a new role of law enforcement embodied in the concept of "community policing". Community policing focuses on working with health professionals and community groups to deal with underlying problems. In contrast, American law enforcement agencies tend to view their role in drug enforcement more in terms of the traditional police role of apprehending criminals (39).

Similarly, non-government organisations concerned with drug issues are more likely to support cannabis decriminalisation and other harm minimisation policies in Australia (1). Drug user groups, which receive government funding from communicable disease prevention programmes, are another source of pressure for drug policy reform in Australia. Such user groups are virtually non-existent in the U.S. Indeed, the very same events in the two countries often lead to exactly the opposite policy conclusions. While the premature death of a drug dependent person in the U.S. is generally seen as evidence favouring even more pronounced efforts to enforce strict prohibition, the families of Australian drug users who have died as a result of their drug use often draw the opposite conclusion and see the death of their family member as evidence of the failure of prohibitionist policies and a strong indicator that drug policy should be changed.

Thus, cannabis decriminalisation is more likely to be viewed as inconsistent with general drug policy in the American context, but confluent with the general drug strategy in the Australian context. There has been a great deal more overt public support for measures such as cannabis decriminalisation in Australia than in the United States. Throughout the 1980s, a series of Gallup Polls found that approximately three quarters of Americans oppose the legalisation of marijuana (27). In contrast, a 1993 Australian survey showed that over half of the population indicated that personal use, possession and cultivation of cannabis should be legal (6). The same study found that approximately three quarters of Australians believe that personal use, possession and cultivation should not be regarded as criminal offences.

Australians are more likely to use cannabis, while Americans use other illicit drugs more often than Australians. Table 1 presents the results of the national household surveys regarding self-reported rates of cannabis use in the respondents’ lifetimes, in the previous 12 months, in the prior month and in the past week. Rates of cannabis use were approximately the same in both countries until the late 1980s, but in the past decade cannabis use has become more common among Australians. In the United States, the proportion of respondents reporting that they have ever used marijuana has increased slowly since the late 1970s. Rates of current use (use in the previous 12 months) declined in the U.S. from 14% in 1985 to 8% in 1992, and has remained relatively stable since then. Rates of monthly and weekly use similarly declined in the U.S. from the mid-1980s to the early 1990s. The 1996 and 1997 U.S. national surveys found small but statistically significant increases in cannabis use, particularly among younger persons.

Table 1 About Here

Australia did not experience similar declines in rates of current and weekly cannabis use. Thus, survey results from 1995, the most recent year in which data are available from both countries, show higher rates of use in the prior year (13% in Australia versus 8% in the U.S.). Furthermore, rates of weekly use are higher among Australians in the two years in which comparable data are available. Donnelly and Hall (17) have noted that Australian cannabis users are more likely than American users to report using at least once a week (11% in Australia compared to 8% in the U.S.).

Table 2 presents selected findings from the young adult age group (ages 20-29) in the U.S. Monitoring the Future surveys (30), in comparison with a similar age group (ages 19-28) in the Australian national surveys. While the age groupings are not exactly the same, it would appear that young Australians are more likely than their American counterparts to use cannabis (42.6% versus 26.5% in 1995). This corroborates the conclusion of Donnelly and Hall (17) regarding school-age adolescents in the two countries. They compared the U.S. Monitoring the Future findings on secondary school students to student surveys in New South Wales and Victoria and noted higher rates of cannabis use in the Australian samples. On the other hand, it can also be seen in Table 2 that young Americans are more likely than their Australian counterparts to use other illicit drugs (13.8% versus 10.9%).

Table 2 About Here

In both countries the socio-demographic portrait of a cannabis user is that of a young, unattached male. Rates of use are highest among single men in their late teens or early twenties (16, 52). Current marijuana use is highest among Blacks and lowest among Hispanics in the U.S. (52). Unemployment is related to higher rates of cannabis use and there are significant geographic differences in rates of cannabis use in both countries. In the United States, rates of cannabis use are relatively high in larger cities and in the Western states (52). In Australia, rates of use are highest in the Northern Territory (52% ever used and 21% used in past 12 months in 1995), Australian Capital Territory (42% and 16%, respectively) and Western Australia (37% and 16%, respectively). Use is least common in Queensland, where 26% report ever using cannabis and only 10% used in the past year (16).

The impact of decriminalisation in Australia

South Australia was the first Australian jurisdiction to implement an expiation system for minor cannabis offences. This "prohibition with civil penalties" model (38) does not represent a complete decriminalisation of personal cannabis use, as the possibility remains for offenders to receive criminal convictions if they do not pay expiation fees on time. While similar expiation systems have since been introduced in the Australian Capital Territory and the Northern Territory, this section will focus primarily on the South Australian experience, as it has been the focus of a number of evaluation studies (3, 11, 18, 42).

The South Australian Cannabis Expiation Notice (CEN) system began in 1987, after a period of intense community and political debate in South Australia. The government at that time had pushed strongly for a new approach to cannabis possession, partly on the strength of the experience of US states which had adopted a less punitive approach. The main arguments for an expiation system were the potential cost savings, and the reduction of negative social impacts upon convicted minor cannabis offenders. Implicit in this second view was the belief that the potential harms of using cannabis were outweighed by the harms arising from criminal conviction.

Impacts on patterns of cannabis use: The potential impact of the introduction of the expiation approach upon levels and patterns of cannabis use in South Australia has been assessed in several drug use surveys, with each analysis adding more recent data to the picture (eg, 11, 18, 19). None of the studies have found an increase in cannabis use in the South Australian community which is attributable to the introduction of the Cannabis Expiation Notice scheme. Lifetime use of cannabis did increase significantly in South Australia from 26% in 1985 to 36% in 1995, but similar increases were observed over the same period in jurisdictions with a total prohibition approach to cannabis, such as Victoria and Tasmania. Similarly, there was an Australia-wide increase in rates of weekly cannabis use over the ten year period from 1985 to 1995, and South Australia did not differ from the rest of Australia on this indicator (19). A comparative study of minor cannabis offenders in South Australia and Western Australia concluded that both the CEN scheme and the more punitive prohibition approach had little deterrent effect upon cannabis users (3). Offenders from both jurisdictions reported that an expiation notice or conviction had little or no impact upon subsequent cannabis and other drug use, and most subjects reported that even if they were caught again, they would not stop using the drug (3).

Impacts on law enforcement and criminal justice systems: The CEN scheme in South Australia had a major unanticipated effect on rates of minor cannabis offence detection. The number of offences for which cannabis expiation notices were issued increased from around 6,000 in 1987/88 to approximately 17,000 in 1993/94 and subsequent years (12). This "net-widening" is not related to any change in the pattern of cannabis use, but reflects the greater ease with which police can process minor cannabis offences, and a shift away from the use of police discretion in giving offenders informal cautions, to a process of formally recording all minor offences.

Since the inception of the CEN scheme, the rate of payment of expiation fees has remained low, at around 50%. Substantial numbers of minor cannabis offenders received a criminal conviction for a cannabis offence as a result of their failure to pay expiation fees. The burden which this placed on the courts was not as severe as one might expect because significant numbers of those people who were issued with a summons chose the option of pleading guilty in writing and did not appear in person. The vast majority of those expiation matters dealt with by the courts resulted in convictions for the offenders, with similar fines imposed to the original expiation fees, plus the addition of court fees.

There has been strong support by law enforcement and criminal justice personnel for the CEN scheme, and little support for a return to the former approach among those administering and enforcing the system (50). The CEN scheme has proven to be relatively cost-effective (8). The unit cost of issuing and processing an expiation notice, when paid on time, is estimated at approximately $30 per case. The total costs associated with the CEN scheme in 1995/96 were estimated to be around $1.2 million (not including police time in detecting the offence). Total revenue from fees and fines was estimated to be around $1.7 million. The cost-effectiveness would obviously be greater if the rate of expiation could be increased. It was also estimated that, had a prohibition approach been in place in South Australia in the same year, the total cost would have been around $2 million, with revenue from fines of around $1 million.

Knowledge, attitudes and other social impacts: In 1993 residents of both South Australia and the Australian Capital Territory, the only two jurisdictions with an expiation approach to minor cannabis offences at that time, were significantly more misinformed than in other jurisdictions about the legal status of activities relating to personal cannabis use (6). For example, 34% of South Australians respondents and 43% of those from the Australian Capital Territory incorrectly believed that it was legal to possess cannabis for personal use, compared with less than 10% of respondents from most other jurisdictions. The introduction of the CEN scheme appears to have given rise to misunderstanding in the community regarding the legal status of personal cannabis use, and of the implications of offending.

Interviews with cannabis offenders who had been issued with expiation notices showed a range of perceived impacts. A substantial proportion indicated that dealing with the CEN caused them financial hardship, and a majority felt that the fines were unreasonable. Respondents from Western Australia, a prohibition state at the time of the study, were found to be more fearful and less trusting of police compared with the South Australian sample after their offence detection. In addition, Western Australian offenders were more likely than South Australian offenders to report difficulties in obtaining or staying in employment following their offence detection, and to report relationship and accommodation difficulties stemming from their offence. Offenders from the two jurisdictions did not differ with respect to their experience of overseas travel difficulties, their perceptions of themselves as criminals, or their self-reported drug use after their offence (34).

Implementation issues and problems: As noted earlier, there have been an ongoing implementation problems regarding the "net-widening" and the low rate of expiation (42). Large numbers of expiation notices are being served by the police and most offenders who do not pay the expiation fee receive a criminal conviction (12). Thus, an unintended consequence of the CEN scheme has been that the number of persons criminalised is as large or perhaps even greater than before the measure was implemented. A common reason cited by offenders for the failure to pay expiation fees is financial hardship (26). Another reason is that a small number of cannabis expiation notices are issued in the context of other charges being laid, for which a court appearance is required anyway. In an unknown and presumably small number of cases, there is insufficient or false identifying information provided by offenders at the time of detection, so that the case cannot be prosecuted at a later time.

Perhaps the most important reason for the low rate of expiation is the level of misunderstanding of the cannabis laws and of the consequences of failure to pay expiation fines. Of a group of offenders who did not expiate minor cannabis offences, three quarters indicated that they were not aware that they would have a criminal record as a result of not paying expiation fees (26). The failure of many persons to realise that cannabis possession is not legal, noted earlier, and the lack of understanding of the legal consequences of an offender’s expiation options highlights the importance of public education on the details and impact of legislative changes.

In recognition that large numbers of people continue to receive criminal convictions for minor cannabis offences, the South Australian government has introduced changes to the payment provisions for expiable cannabis offences in order to increase the number of offenders who opt to pay their expiation fees. These include an option for payment of expiation fees by instalments and an option to substitute community service for the payment of fees. It remains to be seen how these new options will affect the outcomes for CEN offenders.

Other implementation issues include the importance of having efficient data recording and management systems for offence data, particularly where expiation-type offence data are kept separately from criminal justice data. It is important to be able to track outcomes, including final disposition and penalties, for those expiation matters which progress to formal prosecution. In practice, this is not an easy task for many jurisdictions, because police, court and/or penalty information are often maintained separately. To identify repeat offenders, a reliable system of keeping track of expiable offences issued to the same person needs to be in place. Improved procedures are also required to help police to correctly identify offenders when issuing offence notices, and to follow up these offenders if necessary. This may not necessitate extreme surveillance measures, but only the more rigorous application of existing procedures for demanding proof of identity from offenders.

A further implementation problem in the South Australia CEN scheme concerns the ten plant limit for the expiable cannabis cultivation offence. This was initially justified as a reasonable number of plants for personal cannabis users to have in cultivation, as growers would probably lose some plants in the course of the growing cycle, and regular users could supply their own needs by harvesting only what they required from the growing plants. In reality, whole cannabis plants are usually harvested when mature. When even one cannabis plant is harvested, the grower faces the risk of being detected with a non-expiable amount of perhaps a kilogram or more of dried cannabis, the limit for personal possession being 100 grams. With improvements in hydroponic techniques as well as the cloning and hybridisation of cannabis plants, the loss of plants has been reduced and yield increased. Therefore, it has been argued that three plants in cultivation would be a more appropriate limit for expiation.

The impact of decriminalisation in the U.S.

In the United States, there was a movement in favour of reforming marijuana laws throughout the 1970s. Widespread and increasing use of marijuana despite its criminalisation had led to substantial enforcement and court costs, and public attitudes were becoming more accepting of marijuana use. In October 1973, Oregon reduced the offence of possession of less than 1 oz. of marijuana to a civil violation, with a maximum penalty of a $100 fine. From 1973 to 1978, 10 other states had enacted legislation which reduced the maximum penalties for cannabis possession to a fine (45).

Impact on cannabis use: The available data indicate that these decriminalisation measures substantially reduced enforcement costs, yet had little or no impact on rates of use. Statewide evaluations of decriminalisation measures were conducted in Oregon (20), Maine (36), Ohio (48) and California (10). In Oregon, a series of studies conducted after the change in law (20) indicated no major change in rates of use, but given the lack of data prior to the change (other than retrospective questions of dubious reliability) no conclusive interpretation of the results is possible. In Maine, a similar ex post facto study by the Maine Office of Alcoholism and Drug Abuse Prevention (36) resulted in a very positive assessment of a decriminalisation measure, despite the lack of data from before the change in law. In Ohio and California, surveys were conducted before and after changes in the law regarding cannabis possession. In Ohio, a decriminalisation measure went into effect in November 1975. Among those aged 18-24, use increased from 27% in 1974 to 33% in 1978 (48), while among those aged 25-34 use increased from 6% in 1974 to 19% in 1978. In California (10), the proportion of adults reporting that they had ever used marijuana increased from 28% in February 1975 to 35% in November 1976, eleven months before and after "decriminalisation." Thus, all four state-wide evaluations of decriminalisation measures found modest increases in rates of self-reported use. The authors of all of these studies chose to interpret the increases as insignificant, deeming the decriminalisation measures to be successful, although the very same data might well have been used to support the opposite conclusion.

The major limitation with these early evaluations was the lack of control groups or comparative data to provide an estimate of what one might have expected the rates of use to be if there had been no change in the law. Three studies conducted in the mid-1970s did include some form of control group. Stuart and his colleagues (49) concluded that marijuana use was not affected by several changes in municipal cannabis laws, including a decriminalisation measure providing for a maximum penalty of a $5 fine, in the community of Ann Arbor when compared to three neighbouring communities in Michigan. A secondary analysis of four national surveys conducted in the U.S. between 1972 and 1977 found no changes in marijuana use that could be attributed to decriminalisation (43, 45). States which moderated penalties after 1974 (essentially a group of decriminalisation states) experienced an increase in marijuana use, among both adolescents (age 12-17) and adults (age 18 or older), but marijuana use increased even more in states which had the most severe penalties. The third controlled study involves the Monitoring the Future national surveys of high school students. Johnston (28) oversampled seniors in the decriminalisation states and concluded that "Decriminalization has had virtually no effect either on the marijuana use or on related attitudes and beliefs about marijuana use among American young people in this age group" (29:27, italics in original).

The underlying reason why decriminalisation did not lead to increased cannabis use may well be that the reduction in penalties under decriminalisation laws did not appear to have any appreciable impact on availability. (52). There is no apparent connection between the trends in use and the trends in perceived availability. While cannabis use remained relatively constant in the first half of the decade and then increased slightly from 1995 to 1997, there was relatively little change in perceived availability with approximately 60% of Americans reporting that it is "fairly easy" or "very easy" to obtain marijuana throughout the 1990s (52).

Changes in rates of use appear to be more strongly connected with changing perceptions of health risks rather than availability or any changes in the legal status of the drug. The upward trend in cannabis use during the 1970s in both the decriminalisation and non-decriminalisation states did not continue into the next decade. In the 1980s and early 1990s cannabis use declined throughout the U.S. (52). Attitudes became less accepting of cannabis use, particularly among young persons, in both "decriminalised" and non-decriminalised states. For example, annual surveys of students in California, a decriminalisation state, found a trend toward less accepting attitudes toward marijuana use (48). Nationally, only 22% of college freshmen supported legalisation of marijuana in 1985, compared with 53% in 1977 (4). In 1997, 58% of high school seniors thought that there is great risk of harming oneself from smoking marijuana regularly, compared with only 35% in 1978 (30).

As seen in Figure 1, there is a clear relationship in the U.S. National Household Survey on Drug Abuse between perceived health risk and cannabis use from 1985-1997 among those aged 12-17. One cannot be certain of the causal relationship without longitudinal data. Nonetheless, it is noteworthy that as perceived health risk increased in the late 1980s, rates of cannabis use declined. As perceived health risk diminished in the mid-1990s, rates of cannabis use began to rise again.

Figure 1 About Here

Impacts on public use and the health care system: Impressionistic evidence indicates there was no increase in the use of marijuana in public places. The National Governors' Conference evaluation of marijuana policy in eight states, five of which had "decriminalised" marijuana, concluded: "There is also concern that increased public display and use may occur as a result of decriminalisation. Our interviews have indicated that this has not occurred to any substantial extent" (41:1). There was little evidence of any significant impacts on the health care system (45). Given the difficulties in detecting intoxication from cannabis consumption, there was also virtually no evidence concerning impact on the incidence of driving under the influence of cannabis.

Impacts on drug enforcement costs and priorities: The beneficial consequences of decriminalisation consist mainly of savings to law enforcement and the criminal justice system. In all of the decriminalisation states, there were reductions in the number and nature of marijuana cases processed through the law enforcement system (45). There is some evidence that law enforcement in the decriminalisation states redirected efforts toward the detection and arrest of more serious offences and other illicit drugs. In California, marijuana possession charges declined by 36% in the first year but charges other than simple possession (e.g., possession with intent to sell, trafficking, etc.) did not decline, as indicated by an increase in felony arrests (2). As the result, there were declines in the processing of cannabis possession offenders, decreases in incarceration and increases in revenues from fines, and the total cost of marijuana enforcement declined substantially. For example, in California the total cost of marijuana enforcement declined from $17 million in the first half of 1975 to under $4.4 million in the first half of 1976 (10).

Conclusions and recommendations

Decriminalisation measures in the U.S. and Australia were much less radical than their name implies. The new laws involved a change in penalties whereby cannabis possession offenders were no longer subject to potential jail terms, which had already been an uncommon sentence in most jurisdictions, as well as providing the opportunity for possession offenders to avoid a criminal conviction and the resultant problems. In both countries, these so-called decriminalisation laws did not appear to have had a major impact on rates of use, as many feared that it might have.

In Australia, the expiation model of decriminalisation succeeded in avoiding the imposition of criminal convictions for many offenders. However, substantial numbers of offenders still received convictions because of a general "net-widening" in cannabis offence detections, and the failure of a substantial proportion of offenders to pay expiation fees on time, in large part due to a poor understanding by cannabis users of the legal consequences of not clearing expiation offences. Despite these problems, the expiation approach has been viewed as a success by a majority of law enforcement and criminal justice personnel, although some implementation and ongoing operational difficulties have been raised. The expiation approach appears to be more cost-effective compared to a system of prohibition and prosecution of minor cannabis offences. In terms of deterrence, both the expiation and total prohibition approaches to cannabis use seem to be equally ineffective in lessening cannabis use by offenders and in the general population. No increase in use could be ascribed to the decriminalisation measure.

In the United States, decriminalisation resulted in substantial savings to drug enforcement due to lower numbers of cannabis possession cases and increases in fine revenues. These enforcement resources were generally redirected toward the enforcement of trafficking offences and laws regarding other drugs. It cannot be claimed that decriminalisation laws eliminated the social costs and adverse individual consequences associated with cannabis prohibition, but it would appear that decriminalisation succeeded in reducing enforcement costs without increasing the health and safety hazards associated with use. Despite the weight of evidence, decriminalisation has not been universally viewed as a success in the United States. This may be in part due to lack of conclusive data--although these measures received considerable attention by the public and in the media, no national study on the impact of decriminalisation has ever been commissioned. There is also an ongoing fear that cannabis decriminalisation might lead to the liberalisation of policies regarding other illicit drugs.

In order to assess and compare the impact of cannabis decriminalisation measures, it is important to consider the goals of cannabis policy. In the United States, the express purpose of cannabis policy is the prevention or cessation of use, while in Australia the focus is placed on minimising the harms associated with use. While this represents a significant difference in strategies to achieve the goals of drug policy, the explicit aim in both countries is essentially the same: to minimise the health and safety hazards associated with drug use.

But there is implicitly a second goal to drug policy that is not often expressed: to minimise the social costs and adverse individual consequences that result from attempts to control use. There are always limits, financial and otherwise, regarding the extent to which efforts may be placed on reducing the harms associated with drug use. Thus, policy considerations always entail balancing the benefits of public policy against the costs involved in carrying out that policy. Decriminalisation measures should therefore be judged not only terms of their impacts on cannabis use and associated consequences to public health and safety, but also in terms of their impacts on reducing enforcement and other social costs involved in controlling cannabis use. When this dualistic standard is applied to the impact of decriminalisation measures in Australia and the United, we arrive at the following conclusions:

• The reduction of maximum penalties for cannabis possession to exclude the possibility of receiving a jail sentence has had no discernible impact on rates of cannabis use or problems associated with cannabis use. This was true in all of the jurisdictions in Australia and the United States that enacted decriminalisation measures.

• Decriminalisation has led to substantial savings in drug enforcement and other social costs. The extent to which these savings are realised, however, is strongly moderated by the manner in which these measures are implemented. In the case of the Cannabis Expiation Notice system of South Australia, cost savings have also been realised, despite "net-widening" and an unexpectedly low rate of payment of expiation fines. Clearly, even greater savings would be realised if the rate of expiation had been greater.

• The conceptualisation of cannabis-related problems and associated policy responses are more important than empirical evidence in both the public and policy maker perceptions of the impact of decriminalisation. Although the empirical evidence indicates that decriminalisation in the U.S. has been a model of successful legal reform, reducing costs without affecting rates of use, decriminalisation is more often viewed as a success in Australia where implementation problems rendered similar measures to be somewhat less successful. One cannot help but wonder if public opinion on decriminalisation in both countries is based more on preconceived notions of appropriate drug policy rather than empirical evidence.

• Part of the reason why ideology plays a key role in the perceived impact of cannabis decriminalisation is that the empirical evidence is generally inadequate. Decriminalisation measures have been typically enacted without concern for evaluation or monitoring of impact. In the U.S., no study of the impact of cannabis decriminalisation at the national level has ever been commissioned. Those evaluations which have been commissioned at the state level tend to be restricted to data collected after the change in policy and without the development of a consensus concerning appropriate success indicators.

• The manner in which decriminalisation measures are implemented by law enforcement and criminal justice officials can influence the health and social impacts of these measures. In South Australia, the apparent lack of education for the public, and for cannabis offenders in particular, about the status of cannabis use offences under the expiation model, seems to have led to a general under-estimation of the seriousness of personal use offences, and consequently an unexpectedly high proportion of unpaid expiation fees and resultant convictions for fee defaulting.

In light of these conclusions, the following recommendations are offered for those jurisdictions considering the enactment of decriminalisation measures:

• Decriminalisation and similar law reform measures should be subject to systematic evaluation of impact. These could include factors such as: collection of baseline data before new legislative measures are implemented; establishment of ongoing data collection and monitoring systems for offence statistics; regular updates regarding the impact on patterns of use, as measured by population surveys of drug use; establishment of systems for monitoring costs of enforcement; and surveys of offenders and key informants to gauge unintended consequences.

• Issues and potential problems of implementation should be fully considered and, in so far as possible, dealt with. Consideration should be given to alternative procedures that might avoid implementation problems, e.g. the provision of delayed payment or payment in instalments in an expiation system. The South Australian experience with the expiation approach has highlighted the importance of being prepared to "fine tune" the operational parameters of the system to achieve greater efficiency, and the need for public awareness to be enhanced regarding any changes to legislative approach for such a widespread activity as cannabis use.

• The relevant impacts to be considered and the specific success indicators should be developed in close consultation with all relevant parties, including public health agencies, law enforcement agencies, social welfare agencies, users and non-government organisations. Impacts should include not only rates of use but also patterns of use likely to lead to harm, harm indicators, impacts on law enforcement practices and priorities, impacts on the courts and penal system, economic impacts and the extent to which users are marginalised by a particular policy.

• The enactment of decriminalisation and similar legal reforms should be timed such that relevant data can be collected prior to as well as after the change in policy.

• The evaluation of impact and monitoring of success indicators should not only consider impacts over a relatively short period of time such as one or two years, but also longer term impacts.

The reduction of penalties for cannabis possession to exclude incarceration as a possible sentence is in many ways a model of successful drug law reform, with clear benefits and little, if any, adverse consequences. The key aspect of cannabis decriminalisation is that it does not appear to lead to increases in availability, use or problems associated with cannabis use. There are several reasons for this. Controls over cannabis availability probably have had less impact on cannabis use and cannabis problems than they do with regard to other illicit drugs. Cannabis is a drug with limited addictive potential. Its effects are often described as pleasant, but it is not as strongly reinforcing as other illicit drugs such as heroin or cocaine, the use of which is more likely to lead to compulsive or dependent use.

Cannabis use is very popular, being by far the most widely used illicit drug in both countries. The cannabis market has been described in the U.S. as being "near-saturation" (23) and this would probably also apply to Australia as well. Most people know someone who could either obtain cannabis for them or put them in touch with someone who could. While thousands of young persons have been arrested and criminally processed for cannabis possession, the risk of arrest is probably much smaller than it is for cocaine, heroin or other illicit drugs. The major constraint on cannabis use is concern about adverse health effects rather than fear of legal consequences.

Under these circumstances, it should not be surprising that the reduction of penalties for cannabis possession to a fine only did not lead to significant changes in rates of cannabis use. The success of decriminalisation should not be taken to indicate that the use of cannabis is without risk or that legalisation would also have little or no impact. Cannabis use can become compulsive, problems may result even from casual use, and a sudden expansion of availability under a legalisation system could well lead to significant increases in rates of use and consequently on the number of persons who experience problems as a result of their use. But decriminalisation has not changed the availability or use of cannabis in a significant manner. Thus, decriminalisation succeeded in reducing social costs without increasing the problems associated with use in the U.S. and Australia, but this success should not be interpreted to indicate that the same impacts would necessarily apply either to the legalisation of cannabis or to the decriminalisation of other illicit drugs.

References:

1. Alcohol and other Drugs Council of Australia (ADCA). Discussion paper on cannabis. Canberra: Alcohol and other Drugs Council of Australia, 1993.
2. Aldrich, M. and Mikuriya, T., Savings in California marijuana law enforcement costs attributable to the Moscone Act of l976--a summary, J. Psychoactive Drugs, 20, 75-81, 1988.
3. Ali, R., Christie, P., Lenton, S., Hawks, D., Sutton, A., Hall, W. & Allsop, S. The social impacts of the Cannabis Expiation Notice Scheme in South Australia. Canberra: Department of Health and Aged Care, 1999.
4. Astin, A., Green, K. and Korn, W., The American Freshman: Twenty-year Trends, l966-l985, Higher Education Research Institute, UCLA, Los Angeles, 1987.
5. Beers, D. Just say whoa! Mother Jones July/August 1991: 36-43 & 52-56.
6. Bowman, J. and Sanson-Fisher, R. Public perceptions of cannabis legislation. National Drug Strategy Monograph Series, No. 28. Canberra: Department of Human Services and Health, 1994.
7. Brecher, E.M. and the editors of Consumer Reports, Marijuana: the health questions/ the legal question, Contemporary Drug Problems 4:ll5, 1975.
8. Brooks, A., Stothard, C., Moss, J., Christie, P. and Ali, R. Costs associated with the operation of the Cannabis Expiation Notice scheme in South Australia. National Drug Strategy Monograph Series. Canberra: Department of Health and Aged Care, 1999.
9. Brownell, G., Marijuana and the law in California: a historical and political overview, J. Psychoactive Drugs, 20, 7l-74, 1988.
10. Budman, K., A First Report of the Impact of California's New Marijuana Law (SB95), State Office of Narcotics and Drug Abuse, Sacramento, California, 1977.
11. Christie. P. The Effects of Cannabis Legislation in South Australia on Levels of Cannabis Use. Adelaide: Drug and Alcohol Services Council, 1991.
12. Christie, P. Cannabis Offences Under the Cannabis Expiation Notice Scheme in South Australia. National Drug Strategy Monograph Series, No. xx. Canberra: Department of Health and Aged Care, 1999.
13. Collins, D. and Lapsley, H., The Social costs of Drug Abuse in Australia in 1988 and 1992, Canberra: Commonwealth Department of Human Services and Health, National Drug Strategy Monograph No. 30, 1996
14. Commonwealth Department of Health, Housing, Local Government and Community Services, National Strategic Drug Plan 1993-97, Canberra: Australian Government Publishing Service, 1993.
15. Commonwealth Department of Health and Family Services, Statistics on Drug Abuse in Australia 1994, Canberra: Department of Health and Family Services, 1994.
16. Commonwealth Department of Health and Family Services, National Drug Strategy Household Survey: Survey Report 1995, Canberra: Department of Health and Family Services, 1996.
17. Donnelly, N. and Hall, W., Patterns of cannabis use in Australia, NDS Monograph No. 27,Canberra: Australian Government Printing Service, 1994.
18. Donnelly, N., Hall, W. & Christie, P. The effects of decriminalisation on cannabis use in South Australia 1985–1993. Australian Journal of Public Health 19 (1995): 281–287.
19. Donnelly, N., Hall, W. & Christie, P. Effects of the Cannabis Expiation Notice Scheme on levels and patterns of cannabis use in South Australia: evidence from the national Drug Strategy Household surveys 1985–1995. Canberra: Department of Health and Aged Care, 1999.
20. Drug Abuse Council, Marijuana survey -- State of Oregon, News releases, Washington, D.C., l974-l977.
21. Dupont, R., Two prongs in a winnable assault on drugs. The New York Times, September 26, 1984: A22.
22. Fulton, M., Clark, R., and Robinson, T., The Decriminalization of Marijuana and the Maine Criminal Justice System: A Time/Cost Analysis, November, Maine Office of Alcoholism and Drug Abuse Prevention, Augusta, Me., 1979.
23. Goode, E. Drugs in American Society, Fourth Edition, New York: McGraw-Hill, 1993.
24. Harwood, H., Fountain, D., Livermore, G. The Economic Costs of Alcohol and Drug Abuse in the United States 1992, Rockville, Maryland: U.S. Department of Health and Human Services, 1998
25. Heale, P., Hawks, D. and Lenton, S. Public awareness, knowledge and attitudes regarding the CEN system in South Australia. Perth: National Centre for Research into the Prevention of Drug Abuse, 1999.
26. Humeniuk, R., Brooks, A., Christie, P., Ali, R. & Lenton, S. Social impacts and characterisation of offenders under the Cannabis Expiation Notice scheme in South Australia. Adelaide: Drug and Alcohol Services Council, 1999.
27. Inciardi, J. and McBride, D. The case against legalization. Pp. 45-79 in Inciardi, J. (ed.) The Drug Legalization Debate. Newbury Park, California: 1991.
28. Johnston, L., Marijuana use and the effects of marijuana decriminalization, Testimony presented at hearings on the effects of marijuana held by Subcommittee on Criminal Justice, U.S. Senate Judiciary Committee, January l6, 1980; available from Institute for Social Research, University of Michigan.
29. Johnston, L., O'Mally, P. and Bachman, J., Marijuana decriminalization: the impact on youth l975-l980, Monitoring the Future Occasional Paper Series, paper l3, Institute for Social Research, University of Michigan, 1981.
30. Johnson, L., O’Malley, P. and Bachman, J., National Survey Results on Drug Use from the Monitoring the Future Study, 1975-1977, Volume I: Secondary School Students, Washington: National Institutes of Health, 1998.
31. Johnson, L., O’Malley, P. and Bachman, J., National Survey Results on Drug Use from the Monitoring the Future Study, 1975-1977, Volume II: College Students and Young Adults, Washington: National Institutes of Health, 1998.
32. Lazare, D. The drug war is killing us. The Village Voice January 23, 1990: 22-29.
33. LeDain, G. Final Report of the Commission of Inquiry into the Non-medical Use of Drugs, Ottawa: Canadian Government Publishing Service, 1972.
34. Lenton, S., Christie, P., Humeniuk, R., Brooks, A., Bennett, M. and Heale, P. Infringement versus conviction: the social impact of a minor cannabis offence under a civil penalties system and strict prohibition in two Australian states. Canberra: Department of Health and Aged Care, 1999.
35. Lenton, S. and Single, E. The definition of harm reduction, Drug and Alcohol Review 17 (1998): 213-220.
36. Maine Office of Alcoholism and Drug Abuse Prevention, An Evaluation of the Decriminalization of Marijuana in Maine, Office of Alcoholism and Drug Abuse Prevention, Augusta, Me., 1979.
37. Makkai, T. and McAllister, I., Marijuana in Australia: patterns and attitudes, NDS Monograph No. 31,Canberra: Australian Government Printing Service, 1997.
38. McDonald, D., Moore, R., Norberry, J., Wardlaw, G. & Ballenden, N. Legislative Options for Cannabis in Australia. National Drug Strategy Monograph Series, No. 26. Canberra: Department of Human Services and Health, 1994.
39. McDowell, D. Law enforcement. Pp. 94-109 in: Australian Institute of Criminology, Comparative Analysis of Illicit Drug Strategy, National Campaign Against Drug Abuse Monograph No. 18, Canberra: Australian Government Printing Service, 1992.
40. Musto, D. The American Disease: Origins of Narcotic Control, 2^nd ed., New Haven: Yale University Press, 1987.
41. National Governors' Conference, Marijuana: A Study of State Policies and Penalties, 3 Vols., Center for Policy Research and Analysis, Washington, D.C., l977.
42. Sarre, R., Sutton, A. & Pulsford, T. Cannabis - The Expiation Approach. Adelaide: South Australian Attorney-General’s Department, 1989.
43. Saveland, W., and Bray, D., American trends in cannabis use among states with different and changing legal regimes, Bureau of Tobacco Control and Biometrics, Health and Welfare, Canada, Ottawa, 1980.
44. Shafer Commission, Marijuana: Signal of Misunderstanding, First Report of the National Commission on Marijuana and Drug Abuse, U.S. Government Printing Office, Washington, D.C., 1972.
45. Single, E. The impact of marijuana decriminalization: an update, Journal of Public Health Policy 10:4 (1989), 456-466.
46. Single, E. and Rohl, T. The National Drug Strategy: Mapping the Future, report commissioned by the Ministerial Council on Drug Strategy, Canberra: Australian Government Publishing Service, Publication No. 2014, 1997.
47. Sommer, R., Two decades of marijuana attitudes: the more it changes the more it is the same, J. of Psychoactive Drugs, 20(1), 67-70, 1980.
48. Spitzner, J., Drug Use in Ohio: l978, Ohio Bureau of Drug Abuse, Columbus, Ohio. 1979.
49. Stuart, R., Guire, K, and Krell, M., Penalty for the possession of marijuana: an analysis of some of its concomitant, Contemporary Drug Problems 5:553, 1976.
50. Sutton, A. and McMillan, E. A review of law enforcement and other criminal justice attitudes, policies and practices regarding cannabis laws in South Australia. National Drug Strategy Monograph Series. Canberra: Department of Health and Aged Care, 1999.
51. U. S. Department of Health and Human Services, National Household Survey on Drug Abuse: Main Findings 1996, Washington: Department of Health and Human Services, 1998.
52. U. S. Department of Health and Human Services, Preliminary Results from the 1997 National Household Survey on Drug Abuse, Washington: Department of Health and Human Services, 1998.
53. Williams, P., Progress of the National Drug Strategy: Key National Indicators, Canberra: Dept of Health and Family Services, Publication Number 2116, 1997.

Table 1: Trends in Cannabis and Other Illicit Drug Use in Australia and the United States 

	1979	1982	1985	1988	1991	1992	1993	1994	1995	1996	1997
Ever used cannabis in lifetime
--Australia	--	--	28	28	32	34	34	--	31	--	--
--United States	28	29	29	31	31	30	31	31	31	32	33
Use of cannabis in past 12 months
--Australia	--	--	--	12	12	--	12	--	13	--	--
--United States	17	16	14	10	9	8	9	9	8	9	9
Use of cannabis in past month
--Australia	--	--	--	--	--	--	--	--	--	--	--
--United States	13	12	10	6	5	5	5	5	5	5	5
Use of cannabis in past week
--Australia	--	--	--	5	5	--	4	--	5	--	--
--United States	--	--	--	5	3	--	2	--	--	--	--

Sources:

Commonwealth Department of Health and Family Services, National Drug Strategy Household Survey: Survey Report 1995, Canberra: Department of Health and Family Services, 1996, p. 30.

Commonwealth Department of Health and Family Services, Statistics on Drug Abuse in Australia 1994, Canberra: Department of Health and Family Services, 1994, Table 5.1.

Donnelly, N. and Hall, W., Patterns of cannabis use in Australia, NDS Monograph No. 27,Canberra: Australian Government Printing Service, 1994, Table 7.

Makkai, T. and McAllister, I., Marijuana in Australia: patterns and attitudes, NDS Monograph No. 31,Canberra: Australian Government Printing Service, 1997, Figure 1.1, Figure 5.1.

U.S. Department of Health and Human Services, Preliminary Results from the 1997 National Household Survey on Drug Abuse, Washington: Department of Health and Human Services, 1998, Tables 3B, 4B, 5B.

Table 2: Recent Trends in Marijuana and Other Illicit Drug Use Among Young Adults in Australia and the United States

	1988	1991	1993	1995
Use of marijuana in past 12 months
--Australia (aged 20-29)	35.4	37.5	43.3	42.6
--United States (aged 19-28)	31.8	23.8	25.1	26.5
Use of other illicit drug in past 12 months
--Australia (aged 20-29)	7.8	10.6	9.2	10.9
--United States (aged 19-28)	21.3	14.3	13.0	13.8