Assessing the impact of police on cannabis markets

Katie Willis and Amanda McAtamney

Key Points

  • The enforcement of laws relating to the production and distribution of illicit drugs is a major investment for the Australian community, with one recent estimate indicating that the annual direct costs of drug law enforcement (DLE) are around $1.7 billion
  • Traditional measures of DLE performance are based on drug seizure and arrest data. While these are simple and well-understood measures of DLE effort, they are ambiguous and imperfect
  • Work undertaken in Australia and overseas to develop more rigorous measurement systems emphasises the use of multiple, cross-sectoral indicators in assessing law enforcement impacts, rather than relying on single indicators of performance
  • While the police and health sectors use very different strategies and interventions to deal with cannabis-related problems, there is important convergence in what both sectors seek to achieve. That is, a community that is less burdened by cannabis-related crime, illness and injury
  • The new measures outlined in this bulletin would not only assist to improve DLE s understanding of the cannabis market, but their impact on that market
  • Work would need to be undertaken to identify or establish suitable data sources for some of the suggested measures
  • Including new questions in existing population surveys or expanding current agency administrative data sets are two low cost ways to improve the types of data available. There may also be scope for development of new data capture methods that focus on populations that often have high levels of cannabis use, such as youths in juvenile justice settings


The enforcement of laws relating to the production and distribution of illicit drugs is a major investment for the Australian community. One recent estimate has put the annual direct cost of drug law enforcement (DLE) in this country at around $1.7 billion (Collins & Lapsley 2008). Just what proportion of this expenditure can be attributed to the policing of cannabis markets is not possible to say based on available data, although the market for cannabis is significantly bigger than for any other drug type. For example, cannabis accounts for around two thirds of all illicit drug arrests nationally every year (ACC 2010). As such, it is likely that the costs would be considerable.

An investment such as this requires that rigorous systems are in place to measure police performance in addressing local drug markets. Traditionally, DLE agencies in Australia and elsewhere have used drug seizure and arrest data to measure the effectiveness of their work performance. These measures are simple and well-understood measures of DLE effort, although they are ambiguous. Essentially, they demonstrate the extent to which DLE engages in certain activities, rather than demonstrating the broader impacts of their work. That is, the more time and effort DLE devote to detecting illicit drugs, the more likely it is that they will seize drugs. Moreover, traditional measures do not say anything about the full range of impacts that DLE deliver, such as making communities feel safer and more secure, as well as contributing to reductions in drug-related crime, illness, injury and death. Yet this is something that Australian law enforcement actually identifies as an important motivating factor for working in DLE (Willis, Homel & Anderson 2010).

Recognising the limitations of traditional measures of law enforcement performance, work has been undertaken in Australia and overseas to develop more rigorous measurement systems (for example, Weatherburn 2000; McDonald et al 2005; McFadden 2006; Castle 2008). Most of this work supports the use of multiple, cross-sectoral indicators in assessing performance, rather than relying on single indicators of performance. The rationale behind this approach is that it helps to overcome the limitations of any particular indicator data source, while also permitting an assessment of subtle changes over time across a range of disparate areas. In the social sciences this is called measurement triangulation. At a very basic level, keeping the approach simple allows operational police to undertake their own timely assessments. That is, it permits police to use internal expertise as opposed to leaving the process to external experts (Willis, Homel & Anderson 2010).

This paper briefly summarises Australian research (Willis, Homel & Gray 2006; Willis, Homel & Anderson 2010) that developed and field-tested a model performance measurement framework for DLE that has application across all major illicit drug types. It then draws upon the results of this work, as well as discussions from a national workshop held in May 2010 (see Box 1), to build a specific module for measuring the impact of police on Australian cannabis markets.

The performance measurement framework

The framework, developed and field-tested by the Australian Institute of Criminology (AIC) between 2004 and 2009, was a product of a number of different stages of development, including extensive stakeholder negotiation, an examination of relevant literature, as well as in-depth field-testing in four law enforcement jurisdictions (representing national through to local-level DLE). It encompasses a suite of performance measures and indicators built around four high level outcomes. It was these outcomes that DLE personnel identified as important impacts of DLE work. The four outcomes are:

  • reducing drug crime and drug-related crime;
  • reducing organised crime;
  • improving public health; and
  • improving public amenity.

The first outcome reducing drug crime and drug-related crime includes measures directed at addressing the importation, supply and distribution of illicit drugs, measures for assessing drug market dynamics, as well as a measure of the incidence of robbery, the crime most reliably associated with illicit drug (mainly heroin) use. To assess changes in drug markets, the framework includes a series of measures related to drug price, purity and availability, as well as measures concerned with drug trafficking practices.

The second outcome reducing organised crime includes measures directed specifically at addressing high-level drug crime. It is distinguished from the framework s first outcome because of the other crimes that usually go hand-in-hand with organised criminal groups that traffic illicit drugs (such as money laundering, extortion, corruption of public officers) that have serious and far reaching impacts on the community's safety and welfare. Measures for this outcome are focused on elements of trafficking and reflect a number of the measures within the first outcome area.

The third outcome improving public health includes a range of measures for gauging the impact of illicit drugs on the community's health. For example, trends in illicit drug-related deaths and morbidity and the health services underpinning these are included in the framework. Finally, the fourth outcome improving public amenity incorporates a small number of measures of community safety and well-being.

While separated for the purposes of reporting, in practice these outcomes are interrelated and not discrete. For example, activities directed at reducing aggregate drug consumption and expenditure are likely to impact on all four high level outcomes, while measures specifically targeting crime problems associated with illicit drugs and precursor substances, such as money laundering and extortion, would be likely to have most impact on reducing organised crime, and so on.

The framework has a number of strengths. Aside from providing a means for assessing a wider range of DLE impacts, it:

  • marks a shift in emphasis from viewing drug seizure and arrest measures as the chief means of assessing DLE work performance;
  • is flexible in that it meets the needs of Australian DLE agencies with a brief for border protection and offshore operations, as well as state and territory DLE agencies concerned with a mixture of responsibilities ranging from organised crime suppression to street-level dealing and associated local crime problems;
  • forms the starting point for development of further appropriate performance measures for specific agencies with specific briefs in different settings;
  • incorporates a range of measures that are clear in their purpose, reliable and easy to interpret; and
  • is aligned with the harm minimisation principles outlined in Australia's wider drug policy.

Developing the framework s application to cannabis markets

The framework s full suite of measures and indicators are outlined in Appendix 1. The precise rationale for inclusion of these measures and indicators in the framework can be found in Willis, Homel and Gray (2006) and Willis, Homel and Anderson (2010) and will not be covered again here. These were developed to capture information about all major illicit drug types and so many of the measures and indicators are still suitable measures of the cannabis market. Those measures and indicators that are considered particularly applicable to the cannabis market are asterisked in Appendix 1.

However, in some cases the measures and indicators are not well suited for assessing DLE s impact on the cannabis market. For instance, cannabis is not currently tested for purity and so the measure and indicators reflecting this are not relevant. In addition, unlike heroin use, there is no established relationship between cannabis use and robbery.1 As such, the framework s associated measure and indicator are also irrelevant. Further, as the Australian cannabis market is largely domestic in origin, the measures and indicators relating to importation methods are limited.

A number of the measures that underpin the outcome improved public health are also probably irrelevant, or at least are of quite limited utility. For instance, cannabis has low toxicity and so it is highly unlikely that a cannabis user, even a very heavy user, would die or overdose from cannabis consumption (Hall et al 2001). As such, the measures and indicators reflecting emergency medical intervention from paramedics or in hospital emergency settings are not particularly useful in this context. However, in the future it may be worthwhile including a measure that explores trends in acute cannabis-induced psychosis.

1 However, there is now some suggestion that there may be a link between serious criminal trespass (home invasion) and cannabis cultivation in some jurisdictions.

The next section outlines several additional measures of the cannabis market that could potentially be used by police to measure their impact on local cannabis markets. These measures were generated from discussions held during the national workshop.

Further potential measures of the cannabis market

Age of onset of cannabis use

Frequency and intensity of cannabis use

Workshop participants agreed that the age of onset of cannabis use, combined with frequency and intensity of use, would form useful measures of the cannabis market. Research evidence indicates that the age of initiation into cannabis use is decreasing over time, and that there is a significant relationship between early age of onset of illicit drug use and adverse outcomes, including more regular and heavier use of illicit drugs (McLaren & Mattick 2007). That is, the earlier someone starts using cannabis, the more likely it is that they will become regular and heavier users of cannabis. Evidence also suggests that the earlier someone commences drug use, and the heavier they use drugs, the more likely it is that that person will engage in other offending behaviour (McLaren & Mattick 2007). As the level of inconvenience, time, risk or cost of trying to find a drug seller increases (usually as a result of direct DLE intervention), users of other drugs tend to leave the illicit drug market, while those who stay use drugs less frequently. However, for cannabis, those who stay in the market actually use more frequently (Weatherburn 2000). Further, a recent study found the rate of harmful cannabis use (including daily use) has increased significantly since 1995, even though overall rates of cannabis use is declining. Cannabis users aged 14-19 years old, were found to be significantly more likely to use 10 or more cones or joints a day than older users (aged 40 to 49) (Roxburgh et al 2010). In theory, effective police interventions should assist to increase the age of onset of cannabis use, while at the same time assist to decrease intensity and frequency of use.

Deterrence through roadside drug testing

All Australian states and territories have now implemented the use of targeted roadside drug testing, generally focused on testing delta-9 THC (the main psychoactive component of cannabis), methamphetamine and MDMA. For example, the NSW Government introduced legislation in 2006 designed as a harm minimisation approach to deter drivers from driving after using drugs including cannabis, methamphetamine and MDMA (Ross et al 2007). Research has found that those who are most likely to be found driving under the influence of drugs are males, professional drivers, dependent or early onset drug users, and people who use drugs but do not believe the drugs affect their driving ability (Butler 2007). Although some studies indicate that there are low levels of public awareness around the dangers and risks of drug driving (Butler 2007), several state and territory roadside drug testing campaigns now involve an element of public education and awareness.

Providing the number of detections made are sufficient for analysis, data on the roadside drug testing of delta-9 THC levels could be used to detect changes in the prevalence of cannabis use and cannabis potency. In some instances it may also be possible to obtain useful baseline information on users, such as if they are social/recreational or occupational, harmful or dependent cannabis and other drug users (Swann 2010). A related measure that may be useful in this context in the future is the perception among road users of the likelihood of apprehension for detection of delta-9 THC while driving. However, this information is not currently collected.

Road trauma

A 2007 study examined the incidence and severity of alcohol and other drug-related trauma among a sample of 2,127 patients in South Australia's Royal Adelaide Hospital (Griggs et al 2007). A key finding from this study in relation to cannabis and driving indicated the presence of cannabis (as the main drug) in 17.5 per cent of injured drivers. A number of positive correlations were found between:

  • the number of injuries recorded and those who tested positive to cannabis;
  • increased injury severity among those who tested positive to cannabis; and
  • increased length of hospital stay (and subsequently increased health costs) among those who tested positive to cannabis and who had also suffered severe injuries.

Research in Victoria has also found that approximately 40 per cent of fatal road incidences can be attributed to illicit drug use (Morris 2010). Moreover, preliminary findings from a Victorian study outlined in Morris (2010) into accident culpability and presence of drugs in persons injured in motor vehicle accidents indicated that 75 per cent of drivers who used one drug were culpable, with the percentage increasing the more drugs the driver had taken (up to 100% of drivers using four drugs found culpable).

Violence linked to cannabis use

The majority of police in remote areas (76%) believe cannabis use contributes to domestic and/or family violence (Putt & Delahunty 2006). This violence can occur either when the cannabis supply runs out and users become violent as a result of severe withdrawal from the drug or where users attempt to obtain money from their family and friends for more cannabis, which is an issue particularly affecting remote Indigenous communities (McAtamney & Willis 2010). Roxburgh et al (2010) found a 30 per cent increase (from 23,826 in 2002 to 31,449 in 2007) in the number of cannabis users presenting to hospital for treatment - either for cannabis dependence issues or for other related problems, including psychosis. Cannabis-induced psychosis was found to be highest among those aged 20 to 29. Encouraging cannabis users to attend treatment earlier (that is, before they begin using at harmful levels) assists to improve successful treatment outcomes. This is not only the case for cannabis-related psychosis, but also increasing awareness of the availability of services for cannabis dependence (Roxburgh et al 2010). Data surveys including the National Hospital Morbidity Database and the Alcohol and Other Drug Treatment Service National Minimum Dataset are both useful tools that provide information on violence linked with cannabis use.

Supply to young people

Workshop participants discussed the lack of strong legislation around the supply of cannabis to young people/juveniles. There is currently very little research and/or data that provides information on where and how young people obtain cannabis. Workshop participants indicated that there needed to be increases in penalties for those caught supplying cannabis to young people. While this measure may form another useful future metric that gauges age of entry into the cannabis market, data are currently lacking.

Regulation of hydroponic equipment

One action outlined in the National Cannabis Strategy 2006-2009 was to assess the feasibility of the regulation of the sale of hydroponic equipment at a national level (MCDS 2006). In South Australia, an increase in hydroponic shops has been detected, even after adjusting for population growth. In 2010 new legislation was introduced into SA, the Hydroponics Industry Control Act, 2009 (SA), which requires hydroponic businesses to become licensed to sell prescribed equipment and any employees to undergo checks by the South Australian Police to determine their suitability to sell prescribed equipment. Further, businesses must keep a record of all hydroponic equipment sold. This legislation provides police with greater powers to regulate the hydroponics industry and to disrupt the hydroponic cultivation of cannabis by organised criminal networks (SAPOL 2009). South Australia is the first state in Australia to implement legislation aimed at regulating the hydroponics industry. If implemented in other states and territories, this regulating legislation and the subsequent development of databases of hydroponic equipment sold, could provide police with indicators of their enforcement impacts on local cannabis markets in the future.

Community perceptions of risk of detection

Perceived risk of detection is another potential measure that could be used to assess the likelihood that individuals would engage in the cannabis market. At present, each state and territory in Australia has different cannabis legislation. The ACT, SA, WA and the NT have decriminalised minor cannabis offences, such as possessing small quantities of cannabis for personal use (NCPIC 2009). However, in the remaining states (NSW, QLD VIC and TAS) where any cannabis offence is illegal, first time offenders and/or juveniles often receive a caution rather than receiving a criminal conviction (with the exception of Queensland, where cautioning is mandatory for a minor cannabis offender) (NCPIC 2009). It is also at the discretion of police officers to issue a caution in these instances, which can lead to inconsistency in the application of cannabis legislation.

Lenton (2000) has suggested that being arrested for cannabis use does not lead to a decrease in subsequent use (particularly among juveniles), but can have other adverse affects, such as affecting future employment prospects. Generally, the Australian public appears to be unaware of the differences in cannabis laws within each state, with many misunderstanding what decriminalisation and legalisation of cannabis actually means (Fetherston & Lenton 2005). This is exacerbated by the disparity in the application of cautions (in those states where cautions apply). Participants from the workshop suggested that police awareness and education campaigns were needed to send consistent messages to cannabis users (and the community) about the risk and penalties of getting caught possessing, selling or growing cannabis to dispel confusion about the legal status of cannabis in different jurisdictions.

Prevalence of cannabis use among other populations

This measure could be used to detect changes in cannabis use in high-risk communities. For example, cannabis use among Aboriginal and Torres Strait Islander (ATSI) groups is generally much higher than among non-Indigenous people, in both urban and rural or remote locations, with trends indicating increasing use (Delahunty & Putt 2006). Data from the 2004-2005 National Aboriginal and Torres Strait Islander Health Survey indicated cannabis use among Indigenous males in remote communities was double that of Indigenous males living in non-remote communities (28%), and cannabis use among Indigenous females in remote communities was nearly one third higher than Indigenous females living in non-remote communities (17%) (ABS 2006). Accurate data on illicit substance use in remote communities is generally lacking. Most studies that are conducted in remote communities on illicit substance use and its effects are location-specific and have small sample sizes, reducing data robustness and the ability to extrapolate to other settings (Delahunty & Putt 2006). Further work into understanding or creating ways of better capturing data in these communities should be a priority.

Summary and conclusions

Use of multidisciplinary measures can assist police to better understand and describe the benefits their work delivers to the community. Moreover, such measures provide police with important contextual information in which to view cannabis arrest and seizure data. The additional measures outlined above provide scope for further improving the type and range of measures available to police for assessing their work effort on local cannabis markets, although it is acknowledged that work would need to be undertaken to identify or establish suitable data sources for some of the measures. Including new questions in existing population surveys or expanding current agency administrative data sets are proposed as two low cost ways that could improve the types of data available. There may also be scope for the development of new data capture methods that focus on populations that often have high levels of cannabis use, such as youths in juvenile justice settings.

While the value of adopting a measurement system such as that outlined in this paper may be evident to many, the reality is that changes to how police (and others) go about their business takes time and can be especially difficult to do when old systems are entrenched, even if these systems are viewed by the people that use them as imperfect. The success of a new system, such as that outlined here, hinges on (among other things) not just, whether it is a good idea , but on whether there is support for the system at different organisational levels (executive, middle management and frontline) and whether it is incorporated within an agency s business function. Where there is genuine support and commitment to a new system, it is much more likely that it will be adopted. Similarly, when a new system is embedded in core organisational processes (for example, regular monitoring or formal policing review processes) then it will help to ensure that the new system operates effectively in the long term.


The authors would like to express their thanks to the workshop participants who provided advice and input into this bulletin.


  1. Australian Bureau of Statistics. (2006). National Aboriginal and Torres Strait Islander Health Survey: Australia, 2004-05. Canberra: Australian Bureau of Statistics.
  2. Australian Crime Commission (ACC). (2010). Illicit drug data report 2008-09. Canberra: Australian Crime Commission.
  3. Australian Institute of Health and Welfare. (2007). Aboriginal and Torres Strait Islander health performance framework, 2006 report: Detailed analyses. Canberra: Australian Institute of Health and Welfare.
  4. Butler, M. (2007). Australia's approach to drugs and driving. Of Substance 5, 25-26.
  5. Castle, A. (2008). Measuring the performance of drug law enforcement. Policing 2, 311-321.
  6. Collins, D. & Lapsley, H. (2008). The costs of tobacco, alcohol and illicit drug abuse to Australian society in 2004/05. Australia: Australian Government Department of Health and Ageing.
  7. Delahunty, B. & Putt, J. (2006). The policing implications of cannabis, amphetamine and other illicit drug use in Aboriginal and Torres Strait Islander communities. Canberra: Australian Institute of Aboriginal and Torres Strait Islander Studies and Australian Government Department of Health and Ageing.
  8. Griggs, W., Caldicott, D., Pfeiffer, J., Edwards, N., Pearce, A., & Davey, M. (2007). The impact of drugs on road crashes, assaults and other trauma a prospective trauma toxicology study. National Drug Law Enforcement Research Fund Monograph Series No. 20. Payneham: National Drug Law Enforcement Research Fund.
  9. Hall, W., Degenhardt, L. & Lynskey, M. (2001). The health and psychological effects of cannabis use. Canberra: Commonwealth Department of Health and Ageing.
  10. Fetherston, J. & Lenton, S. (2005). Community attitudes towards cannabis law and the proposed Cannabis Infringement Notice scheme in Western Australia. Drug and Alcohol Review 24, 301-309.
  11. Lenton, S. (2000). Cannabis policy and the burden of proof: Is it now beyond reasonable doubt that cannabis prohibition is not working? Drug and Alcohol Review 19, 95-100.
  12. MacDonald, Z., Tinsley, L., Collingwood, J., Jamieson, P., & Pudney, S. (2005). Measuring the harm from illegal drugs using the Drug Harm Index. Home Office Online Report 24/05. London: Home Office.
  13. McAtamney, A. & Willis, K. (2010). Policing cannabis and other illicit substances in remote Indigenous Australian communities. Criminal Justice Bulletin Series No. 6, National Cannabis Prevention and Information Centre Bulletin:
  14. McFadden, M. (2006). The Australian Federal Police Drug Harm Index: A new methodology for quantifying success in combating drug use. Australian Journal of Public Administration 65, 68-81.
  15. McLaren, J. & Mattick, R. (2007). Cannabis in Australia. Use, supply, harms and responses. Monograph No. 57. Sydney: National Drug and Alcohol Research Centre.
  16. Ministerial Council on Drug Strategy (MCDS). (2006). National Cannabis Strategy 2006-2009. Canberra: Commonwealth of Australia.
  17. Morris, C. (2010). The drugs and driving project: Examining the effects of alcohol and other drugs on vehicle collisions. Drug Info 8, 4.
  18. National Cannabis Prevention and Information Centre (NCPIC). (2009). Cannabis and the law. Factsheet 2:
  19. Putt J., & Delahunty, B. (2006). Illicit drug use in rural and remote Indigenous communities. Trends and Issues in Crime and Criminal Justice 322. Canberra: Australian Institute of Criminology.
  20. Ross, J., Campbell, G., Degenhardt, L., & Dillon, P. (2007). Driving and clubbing in Sydney: A study of drug use and risk among nightclub attendees. Technical Report No. 289. Sydney: National Drug and Alcohol Research Centre.
  21. Roxburgh, A., Hall, W., Degenhardt, L., McLaren, J., Black, E., Copeland, J., & Mattick, R. (2010). The epidemiology of cannabis use and cannabis-related harm in Australia 1993-2007. Addiction 105, 1071-1079.
  22. South Australia Police (SAPOL). (2009). Hydroponics Industry Control Bill 2009 Report. Parliamentary Second Reading Speech:
  23. Swann, P. (2010). Drugs and driving: The challenges for the future. Guest editorial. Drug Info 8, 2.
  24. Weatherburn, D. (2000). Performance indicators for drug law enforcement. Contemporary Issues in Crime and Justice No. 48. Sydney: Bureau of Crime Statistics and Research.
  25. Willis, K., Homel, P. & Anderson, J. (2010). Developing the capacity and skills for national implementation of a drug law enforcement performance measurement framework: Final report. National Drug Law Enforcement Research Fund Monograph Series No. 34. Hobart: National Drug Law Enforcement Research Fund.
  26. Willis, K., Homel, P. & Gray, K. (2006). Developing and implementing a performance measurement framework for drug law enforcement in Australia. National Drug Law Enforcement Research Fund Monograph Series No. 18. Marsden, SA