The Social Impacts of the Cannabis Expiation Notice Scheme in South Australia: Part 3

Results continued..............

Study No. 7: Costs Associated with the Operation of the CEN Scheme in South Australia

The Cannabis Expiation Notice (CEN) scheme in South Australia, as with all law enforcement schemes, has associated costs and benefits. Some of those costs and benefits are less tangible than others, including some of the social impacts which are the focus of the Cannabis Social Impacts Study as a whole. It was the aim of this particular study component to make an attempt to estimate some of the more tangible monetary costs associated with the operation of the CEN scheme.

These costs to the state include the costs of issuing and processing CENs through infringement notice procedures regularly used by police, referral of non-expiated matters for prosecution and court processing, and the costs associated with enforcing court-imposed penalties for convicted minor offenders. Costs to the individual offender relate to the monetary value of the fees, fines and levies associated with offences detected, possible legal advice costs, and potential social impacts of conviction or police intervention upon employment, relationships and other areas.

Expenditures by the state have been used as a proxy measure of the economic costs to society. Fixed costs associated with infrastructure necessary to operate the law (eg. police stations, court facilities, etc.) were not included in the cost estimates for the CEN scheme, because of the difficulty of quantifying the relatively minor allocation which would be for the CEN scheme. Expenditure estimates largely involved personnel time consumed in the tasks of law enforcement and processing of CEN offences. On-costs were added for personnel costs, but no allowances were made for maintenance costs or consumables, because of difficulty in quantifying this factor, and because it was assumed to be relatively minor.

For the present study, unit cost estimates were developed for a number of potential penalty outcome pathways under the CEN scheme. Costs of issuing CENs were estimated from the time a CEN was actually handed to an offender; no allowance was made for the costs associated with detecting the offences, as these costs were considered very difficult to quantify, and subject to wide variation, depending on police resources targeted at offence detection for minor cannabis offences. Furthermore, a substantial but unknown proportion of CEN offences would have been detected opportunistically or in the context of other investigations, making costing of detection more difficult.

The unit cost of issuing a CEN under the CEN scheme, including police time in issuing the notice, entering data onto computer, and other administrative tasks, was estimated to be $32.73. This was also taken to be the unit cost of a finalised expiated CEN, as handling of full expiation payments was taken as adding a negligible amount to this cost. If community service was granted as a payment option before a CEN became a fine default matter, substantial additional cost was added for managing the community service order, raising the unit cost to $257.73.

Additional costs were associated with managing CEN matters which remained unpaid, and had to be prosecuted. For example, a CEN which was not expiated, and resulted in a court hearing and conviction was estimated to cost $51.82 form the time of CEN issue to finalisation by full payment. If such as case was cleared by community service, the unit cost was $276.82. If payment of court fines was not made, costs increased as further criminal justice processing ensued, including the issuing of warrants. Where non-payment of court-imposed fines resulted in a warrant being issued, the unit cost for cases which were eventually payed in full was estimated at $90.75. Where such cases were cleared via community service, the unit cost was $315.74, and if it was cleared by imprisonment, the unit cost was $601.74.

A model for the annual cost of the CEN scheme was generated, based on the figures for the 1995/96 financial year in which 16,321 CENs were issued, with 7,165 being expiated (a 44% expiation rate). It should be highlighted that the total cost estimated for the model is based on approximations of the proportions of offenders within the various final outcome pathways. Police and court data available for this study related to processing and initial court-imposed outcomes, ie. they gave no indication of the proportions of court-imposed fines which were actually paid, or converted to community service at a later date, or required warrants being issued and served for non-payment of court fines. Thus, rough approximations of such outcomes (including imprisonment) were used. For the purposes of the model comparisons carried out in this study, these approximations were adequate.

For the 16,321 CENs issued in 1995/96, with a 44% expiation rate, the total cost was estimated to be $1.24 million. Revenue from CEN fees, fines and costs was estimated to have been $1.68 million for that year.

Further models were calculated, based on estimated costs for a notional CEN scheme in which the expiation rate was 10%, 20% and 30% higher than the 44% observed rate over recent years. With a 10% increase in expiation rate (ie. to 54%), the total cost was estimated at $1.11 million; with a 20% increase, $0.98 million; and for a 30% increase, $0.86 million. These models showed clearly that improving the rate of expiation reduces costs; this is mainly due to the low costs associated with expiated notices.

When likely revenue from fees, fines and levies for these models is calculated, the total annual amount declines slightly: $1.60 million for an expiation rate of 54%; $1.52 million for a 64% rate; and $1.45 million for a 74% rate. However, the total surplus of revenue over expenditure is higher for the higher expiation rates, resulting in greater savings to the state. It remains unclear the extent to which expiation rates can be improved to realise these greater savings. With recent changes to the way expiation offence fees can be handled (with the implementation of the Expiation of Offences Act, 1996), expiation rates may increase for CEN offences, but cost savings may not be realised if community service is more widely used.

As it was not possible to reliably cost the prohibition approach which existed in South Australia before the CEN scheme was introduced, and it was not possible to obtain truly comparable costs from another jurisdiction which had a total prohibition approach to minor cannabis offences, a model was generated which estimated the cost of prohibition of minor offences in South Australia, had the CEN scheme not been introduced. In generating this model, allowance was made for net-widening under the CEN scheme, such that it was assumed that around 7,500 minor cannabis offences would have been detected in SA in the 1995/96 financial year, rather than the 16,321 CEN offences detected. Even with this allowance, the total cost of the prohibition approach was estimated to be $2.01 million, while revenue from fines and levies was estimated to be $1.0 million. These models suggest that an expiation scheme such as the CEN scheme, even with a relatively low rate of expiation, has much greater potential for cost savings to the state than does a prohibition scheme for minor cannabis offenders.

Study No. 8: A Review of Law Enforcement and Other Criminal Justice Attitudes, Policies and Practices Regarding Cannabis and Cannabis Laws in South Australia

Through twenty-eight intensive one-to-one interviews and four focus group discussions, the research team obtained intensive feedback from forty-nine people involved in administration of South Australia's cannabis laws. Respondents included the Chief Justice, the Chief Magistrate, a representative from the Office of the Director of Public Prosecutions, the National Crime Authority, police prosecutors, the officer in charge of the Drug Task Force, Drug Task Force and regional detectives and police patrol officers. Discussions also were held with personnel in the Correctional Services and Attorney-General's Departments.

Major findings are as follows.

1. Should South Australia retain the CEN scheme or revert to prosecuting minor cannabis offenders?

Virtually all respondents considered that it would be better for South Australia to continue to issue expiation notices for minor cannabis offences rather than to revert to a system of prosecutions. Reasons for maintaining this view differed, however. Police, who constituted the majority of interviewees, put emphasis on the convenience and cost-effectiveness of CENS. Issuing a CEN eliminated time spent on court attendance, and also significantly reduced administrative burdens associated with storage of court exhibits. Individual users still could be deterred from cannabis use by being given CEN on several different occasions for repeat offences.

By contrast, members of the judiciary and others respondents working outside the enforcement system tended to favour expiation because it provided a way for users to avoid stigma and other adverse social consequences associated with a court conviction.

2. Was the CEN scheme having any unintended consequences?

Respondents agreed that expiation had improved police and court efficiency. However quite a number of police argued that there also had been some unintended consequences. In particular, Drug Task Force and regional detectives argued that individuals and syndicates may be exploiting provisions which specified that cultivation of up to ten plants should be dealt with by means of an expiation notice. In their view, the advent of hydroponics and techniques for cloning female plants meant that it was possible for commercial crops to be grown, while staying within the ten plant limit. Some respondents argued that organised crime had become involved in coordinating small scale cultivations. When requested, the Police Department and the National Crime Authority produced some intelligence evidence that this was occurring. The Director of Public Prosecution's office confirmed that it would be difficult to prosecute successfully individuals or groups conspiring to exploit the CENS system by organising small cultivations in several different locations.

3. Did aspects of the CEN legislation and regulations now needed to be amended?

Respondents who were members of the judiciary or from the Office of the Director of Public Prosecutions could not see any cause for major change to legislation or regulations relating to the CEN scheme (as primarily embodied in the Controlled Substances Act, 1984). Several police - particularly those in the Drug Task Force or working on drug-related special investigations or as regional detectives - argued that the maximum number of plants for which a CEN could be received should be reduced from ten to three or four. They argued that the legislators clearly intended that notices should only be issued in instances where cannabis was being cultivated for personal use. It was argued that, particularly with current growing techniques, people cultivating cannabis for their own personal use should not require more than three or four plants.

4. Did police, other criminal justice personnel and the public adequately understood the CEN scheme in South Australia?

Respondents from the law enforcement sector stated that introduction of the expiation system seemed to have caused some confusion within the general public. A number of people detected for possession, use or cultivation of small quantities of cannabis now were under the impression that this was legal. Police and other justice officials demonstrated good understanding of technical aspects of the expiation laws. However, operational law enforcement officers were reluctant to contemplate exercising greater discretion in the issuing of notices (for example, only issuing a CEN if this was likely to achieve some public benefit), and had not thought about ways of using expiation notices to reshape cannabis markets (for example, to drive out organised crime elements by flooding some locations or groups with notices).

From reviewing law enforcement and other criminal justice attitudes and practices, the general conclusion is that, despite initial opposition from the Police Association and some concerns expressed by the Police Department, the expiation notice approach now enjoys general support. One major source of concern was that, because cultivations of up to ten plants can be dealt with by means of expiation, some individuals and groups may be exploiting the system for commercial purposes.

One way of dealing with this problem may be to reduce to three or four the maximum number of plants for which a CEN can be issued. Alternatively, however, police could give consideration to improving intelligence systems, so that individuals or groups exploiting the CEN system for commercial purposes could be 'driven out of the market' by being served with repeated notices. Jurisdictions which took the latter course would still leave scope for the genuine 'amateur' cultivator, who may be unable to obtain sufficient yield from a small number of plants. 

Discussion

As has been observed in other research on the CEN scheme in South Australia, it appears that there has been a degree of net-widening under the scheme since its implementation in 1987. However, the economic analysis carried out for this project suggests that the net-widening has not resulted in greater costs to the state. Furthermore, the expiation approach in South Australia may well be more cost-effective than a prohibition approach, despite net-widening and the fairly low observed rate of expiation.

One of the tasks of this research was to investigate whether there might be a difference between an expiation and a prohibition approach in terms of the observed social costs for offenders and society as a whole. The fairly low rate of expiation of CEN offences observed in South Australia up to 1996 (around 45%) has suggested that financial hardship among a substantial proportion of offenders may be a contributing factor. Sarre, Sutton and Pulsford (1989) noted the low rate of expiation in their initial evaluation of the CEN system. They found that people detected for minor cannabis offences tend to be drawn from lower socio-economic groups, and that they figured disproportionately among those prosecuted for failing to pay expiation fees. The present study provides further support for this view, from interviews with offenders, and from published data (eg. the confirmation that there has been a much greater representation of repeat offenders, who tend to avoid paying fines, amongst non-expiators compared with expiators). While in this study, it was not possible to fully characterise the non-expiator group and compare it with the expiator group because of limited demographic information available on CEN forms, there is suggestive evidence from other research components that financial hardship is a factor for minor cannabis offenders, particularly younger offenders. It remains to be seen whether the recent enactment of the new Expiation of Offences Act, 1996 has an impact on rates of expiation and court loads for minor cannabis offences.

Both the Western Australian and South Australian samples of cannabis offenders interviewed for this project displayed some negative impacts arising from cannabis offence detection and processing, including financial difficulties. The WA group was more likely to report negative employment consequences from offence detection (32% in WA vs 2% in SA). While such consequences are based on the offenders' self-reported perceptions of whether consequences were related to a cannabis offence, the findings indicate an apparent and noteworthy difference between the jurisdictions in terms of a significant area of social impact.

In contrast, from the separate survey of employer groups conducted for this project in WA and SA, it emerged that most employers in both states reported that they were not concerned by employees' prior offences or convictions for minor cannabis offences, but were more concerned about the negative consequences possibly arising from cannabis intoxication in the workplace. There differences between offender and employer perceptions of employment consequences could be due to a number of factors. A bias in the employer study could have been introduced by an over-representation of large employers and under-representation of smaller employers. Furthermore, non-respondents to the survey may have had more conservative attitudes to cannabis use than those employers who were interviewed, thereby biasing the employer study findings towards more accepting views. Nonetheless, the employer study demonstrated that cannabis offences seem not to be especially singled out when employers conduct criminal record checks.

The survey of cannabis offenders also showed significant differences in other areas of social impact, with the SA sample experiencing fewer negative impacts than the WA group in terms of relationship consequences and effects on accommodation status. There was also a difference between the SA and WA groups regarding subsequent criminal justice consequences, with none of the respondents in the SA expiator group identifying any negative episodes of involvement with the criminal justice system which they thought were in some way related to their CEN, while 32% of the WA sample identified at least one such consequence related to their conviction. It may be that this finding is partly related to differences in ease of access to criminal history information on offender records which police have in WA, such that they are able to more readily identify prior offenders than in SA.

In other respects the offender groups in the two states did not differ greatly, with no differences found in terms of the extent to which they saw themselves or others saw them as a criminal, as a result of the incident. There were no differences between the groups regarding the impact of the CEN or conviction on their drug use, with both the CEN and the cannabis conviction groups reporting very little impact of offending on subsequent drug use. There was also no significant difference between the groups in terms of negative travel effects of conviction or CEN.

The majority of both the SA and WA offender groups saw themselves as largely law abiding, and had respect for the role of police as law enforcers and the law in general. The majority of both groups also shared a lack of support for punitive drug laws, had a high level of support for cannabis use being legal, and had positive views regarding cannabis. Most thought that it was a safe drug and that the benefits of cannabis outweighed the harms. Most saw it as much less harmful than a range of other substances including alcohol and tobacco. This finding, combined with the observation of a relatively poor knowledge in the general community in South Australia about the health consequences of heavy or long-term cannabis use, and the relatively heavy cannabis use amongst offenders, suggests a possible avenue for harm reduction through providing CEN offenders with health risk information at the time of a CEN being issued.

There were differences between states in reasons for police attendance and location of cannabis offence detection. More offences in SA seemed to result from opportunistic detections, with fewer being detected in private dwellings. This seems to accord with the net-widening phenomenon observed from the published data, where changes to police practices associated with the implementation of an expiation scheme seem to result in a greater number of offences detected in the context of routine police work.

While both the SA and WA groups were equally likely to report that they were friendly, respectful and cooperative toward the police when they were arrested or issued with their CEN, higher proportions of the WA group reported that they had become less trusting and more fearful of police. This is likely to be due in part to the greater number of WA offenders who were apprehended in a private residence, and the greater perceived intrusiveness of such apprehensions.

With regard to the South Australian CEN scheme, the low rate of expiation has been a cause for concern since the first evaluation study conducted by Sarre, Sutton and Pulsford (1989). Besides the inability to pay expiation fines, another suggested reason was that many CENs may be issued in the context of other charges being made, requiring the offender to appear in court, and thereby reducing the incentive to pay expiation fees. The present study was not able to provide a clear answer to this question, owing to the way in which CEN data are stored, and the problems in linking CEN offence data to criminal record data. The concern has also been expressed that this factor could lead to doubling up of court appearances for CEN fee defaulters who were required to appear in court for other offences. Again, the extent of this could not be determined. However, with the recent changes to handling of all expiable offences under the Expiation of Offences Act, 1996, this doubling up effect would not occur, as expiation fee defaulters now receive automatic convictions if they fail to respond to a payment reminder notice, without the need for a court appearance.

Another possible reason for failure to expiate notices in past years might relate to individuals who, at the time of being issued with CENs, give false identifying information. The exact extent of this problem could not be determined, but is probably not as great as was thought. Statistics obtained from SA Police for the present study showed that a fairly small percentage of CENs issued would be withdrawn or cancelled by police before being cleared or forwarded for prosecution (around 1%). After unpaid CENs were forwarded for prosecution, a further proportion would be deemed unsuitable for prosecution and withdrawn. The reasons for such matters being withdrawn would most probably relate to incomplete information being recorded on CEN documentation to enable prosecution to proceed. Among the CEN matters withdrawn at various stages, there would be an unknown proportion that had false identifying information recorded on them. Then, for those cases where a summons was sent out, a proportion would involve offenders who had given false addresses or other identifying information, making it impossible to proceed with the matters any further. Further research is needed, examining final outcomes from court proceedings (ie. after any further defaulting on court-imposed fines, issuing and serving of warrants, and completion of subsequent penalties, including imprisonment) in order to gain a more complete picture of the extent of CEN offences which are not resolved, and the degree to which false information may contribute to follow-up losses.

The Expiation of Offences Act, 1996 (which came into effect from early 1997), besides having a likely effect on the expiation rate for cannabis offences, provides more information to recipients of notices on the consequences of failure to pay expiation fees. This provision alone may contribute to improving expiation rates, as well as improving the perceived fairness of the system. The need for more detailed information on the financial and legal consequences of non-payment of CENs was highlighted by respondents in the SA offender interview study. Furthermore, earlier research on knowledge and attitudes regarding cannabis laws (Bowman & Sanson-Fisher, 1994), as well as the community attitudes survey conducted for this project, have shown that aspects of the CEN scheme are not well understood in the South Australian community. There may be scope for expanding upon information provided to offenders on CEN forms, perhaps by providing offenders with a detailed information sheet at the time of receipt of a CEN.

The observation that the 45 years and over group is the one for which cultivation offences account for the highest proportion of total CENs issued suggests that cannabis cultivation is more of an activity for older people involved in the culture of cannabis and its use. While the majority of cultivators issued with CENs are in the 25 to 34 year age group, it does seem that older people who remain involved with a cannabis using culture (and hence are at risk of detection by police, by whatever means) are more likely to be involved with cannabis cultivation as compared with just using cannabis. It should be emphasised that some of this observed variation in the proportion accounted for by cultivation offences at each age level may be explained by a possible detection bias associated with age; younger people involved with a cannabis using culture may be exposed to greater risk of detection due to the circumstances in which they use the drug, which in turn may mean that the younger groups are more likely to be detected for possession and use offences, rather than cultivation offences.

A related observation was the trend towards higher rates of expiation of cultivation offences among older age groups. As mentioned elsewhere, higher rates of expiation among older offenders generally may reflect a greater awareness of the consequences of failure to pay CEN fees. Furthermore, older offenders are probably less likely to have trouble paying CEN fines. The fact that this trend is mainly evident for cultivation offences, and not possession/use offences, reinforces the notion that cultivation may be perceived by offenders as more serious then possession of cannabis, and therefore results in a greater likelihood of being cleared. Indeed, it could be argued that police also view cultivation offences as more serious than possession offences, as larger total weights of cannabis can be expiated in plant form than as dried cannabis in someone's possession; there is thus greater potential for small commercial amounts to be held by cultivators of cannabis plants than by those who just hold "possession" amounts. This situation may contribute to older cannabis cultivators in particular wishing to clear their offences by paying expiation fees on time.

The present study was not able to shed light on the extent of home cultivation of cannabis which might involve the use of hydroponic techniques. The data on offences issued by month showed a clear cyclic variation in the numbers of CENs issued at different times of the year, particularly for cultivation offences, where such offences were almost six times as numerous in March as in the lowest period around July. It may be that a high proportion of plants in cultivation detected by police at around this time of year are being grown in assisted environments, perhaps with hydroponic techniques, as the winter months are not as conducive to growing mature cannabis plants out of doors. The extent of this cannot be readily gauged from the published data.

It is important to note that none of the data on cannabis offences under the CEN scheme available for the present study gives any indication of the potency (ie. THC content) of dried cannabis seized in possession offences. Furthermore, no data are available on the potency of cannabis plants found in cultivation, including whether plants found might be higher potency strains such as "skunk". An interesting area for research would be to analyse the potency of cannabis plants seized at different times of the year, with comparisons of potency for different growing methods.

The implications of cannabis potency for cannabis markets and the health of cannabis users are somewhat unclear. Offence data do not shed light on what cultivators of expiable numbers of cannabis plants might do with the material they grow, including whether they keep all of what they grow for themselves, give some away to friends, or try to sell some of it. While the interviews conducted with cannabis offenders in SA showed a degree of plant cultivation among the group, with a minority involved in small scale sale of home-grown cannabis, there were insufficient sample numbers to make judgments on average potency of home-grown cannabis in South Australia.

The data presented in this report have shown that a reasonable proportion of CENs issued have been withdrawn at various stages of processing. Among the CENs issued, a small proportion (about 1%) have been cancelled or withdrawn before failure to pay expiation fees became an issue. CENs which have not been expiated are examined by the prosecutions section of the SA Police, for determination to be made on whether a prosecution can successfully proceed. The limited reliable data available showed that 3–4% of CENs forwarded for prosecution (or 2% of CENs issued) would be withdrawn by police at this stage in processing, before charges were laid. The exact reasons for individual CENs being withdrawn at this point are not recorded, but it is likely that many relate to problems in ascertaining the correct identity or contact details of the offenders, or with there being insufficient evidence to allow a successful prosecution to proceed. Such problems are not unexpected in a system where a simple expiation matter can, if left unpaid, convert to a drug offence which requires a greater standard of evidence to support it, and the assumption of guilt on the part of offenders. The percentage of matters withdrawn before charges are laid, it could be argued, is fairly low, given the difficulties faced by police in proving identity of offenders in certain situations where cannabis offences are detected, such as at parties, or when offenders are visitors at the place of detection. Even so, there may be scope for enhancing police training, or revising police operating procedures in this area, in order to maximise the potential for later prosecution of unexpiated CENs. This could include increasing the application of proof of identity requirements in situations where detected offenders do not have proof of identity at the time of offence detection.

The finding that well over 90% of prosecutions relating to failure to pay CEN fees resulted in convictions for the offenders is a cause for concern in terms of adverse social consequences. As discussed, substantial numbers of CEN offenders have failed to pay expiation fees on time since the CEN scheme began operation, and the reasons for the low rate of expiation are probably varied. Whatever the reasons, the low rate of expiation means that large numbers of offenders have exposed themselves to the risk of receiving a conviction for a minor cannabis offence, in a system that, it could be argued, was designed to remove or reduce that risk of conviction. For the five years from 1991/92 to 1995/96, a total of 37,470 convictions for unpaid CEN matters were handed down; these represent 46% of all CENs issued over that period.

In presenting these findings relating to convictions handed down for CEN matters which were not cleared by expiation, it should be emphasised that a small proportion of such cases would be left unresolved because of false or insufficient information being available to adequately identify an offender. This would most likely be a relatively small number in addition to the 5% of CEN offences identified as withdrawn or dismissed for various reasons

As the rate of conviction for offenders who failed to expiate CENs is so high, the differences between the offence types in the proportion of conviction outcomes were minor. Cultivation offences had the lowest proportion of offenders who received no conviction (1.2%), and the highest proportion of convictions being given (95.2%), reinforcing the perception that cultivation offences overall may be viewed by courts as being more serious than cannabis possession offences.

It is also of concern that many convictions for CEN offences may have resulted from individuals who did not pay CEN fees on time because they faced financial difficulty, or who had let CEN matters lapse because they misunderstood the consequences of failure to expiate CEN fees (ie. that an unexpiated CEN offence effectively converts to a drug offence, with the attendant risk of conviction). As discussed earlier, the new Expiation of Offences Act, 1996 may to a large extent address the problems of CEN offenders facing difficulty paying expiation fees, through the provision of more payment options which can be chosen if financial hardship can be demonstrated. Another suggestion that has been raised is that there be a greater range of expiation fee amounts, so that offences involving very small amounts might incur a lower fee, thereby improving the likelihood of expiation. Furthermore, the new CEN forms may improve the level of understanding the consequences of failing to pay. However, the new Expiation Act has not removed the possibility of conviction for unpaid expiation fines. The introduction of automatic conviction processes for CEN offenders who do not pay fines or specify an alternative method for clearing the matter may mean that a small group of CEN offenders will receive convictions more readily than under the older Expiation Act (eg. those who receive CENs in the context of other offences). It is too early as yet to comment on what effect the new Expiation Act will have on rates of expiation of minor cannabis offences, and the risk of receiving a conviction after being issued with a CEN.

The analyses conducted for this study using SA Police data showed that community service orders were infrequently imposed as the main penalty accompanying conviction for CEN offences which remained unexpiated. Despite this, there would be a greater number of CEN offenders who at some time after their court matter has been dealt with apply to have their court-imposed fines converted to community service orders. Unfortunately, it was beyond the scope and resources of this study to obtain figures on these later outcomes following on from court processing of offenders. Such an analysis could be carried out by the Courts Administration Authority, but would require additional funds and more time than was available to the researchers for the present study. Similarly, a reliable analysis of CEN-related fee defaulters receiving imprisonment as a result of failure to pay court-imposed fines could not be undertaken within the present study, as the available data sets would require considerably more work to enable thorough tracking of expiation offenders through to imprisonment.

The data provided by SA Police for this study allowed a fairly rudimentary picture to be gained of the extent of repeat offending for expiable cannabis offences. There are methodological difficulties that need to be acknowledged in assessing the extent of repeat offending under the CEN scheme, and these largely relate to the difficulties in identifying separate CEN offence incidents for the same unique individuals. Identifying separate CEN offences from separate occasions involving the same person involves carrying out computer matching on name and date of birth data. It is not known how many repeat offenders might have been missed through this process. It can be said that those repeat offenders identified in the study will be an under-representation of the actual numbers of such offenders; furthermore, the figures generated will under-estimate the true numbers of repeat offences for each repeat offender.

It should also be recognised that the numbers of repeat offenders identified, and the numbers of separate offences found for each repeat offender, will depend on the period over which data records have been analysed. For this research, analyses were conducted on the extent of repeat offending within each financial year, and also within the total five year period for which complete data were available, from 1991/92 to 1995/96. Had adequate records been available for a longer period, perhaps going back as far as the beginning of the CEN system in 1987, the extent of repeat offending found within the data would have no doubt been higher.

It was noteworthy that close to 25% of all CEN offences issued over the five year analysis period were verified as being issued to repeat offenders (offenders who had been issued with more than one CEN over the five year analysis period, from 1991/92 to 1995/96). This is a substantial proportion, and as mentioned above, will be an underestimate of the full extent of repeat offending. Most repeat offenders had been issued with two CENs on different occasions, and within any given year, around 8 – 9% of all CENs issued were to people who had two CEN offences within the year. For the entire five year study period, around 13% of all notices issued went to individuals identified as having two separate offences. Also, a substantial number of repeat offenders had had three CENs issued within the five year period, accounting for a further 5 – 6% of all CENs issued. As would be expected, fewer individuals had larger numbers of repeat offences for which CENs were issued. It could be argued that heavier and longer-term users of cannabis, those who choose to cultivate cannabis plants for their own use, and those who are more involved with a cannabis-related subculture, will be more likely to be detected repeatedly over time. Age of offenders will be a factor in the risk of repeat offence detection, in that older users of cannabis (who have maintained some degree of ongoing use) will have had a longer period of exposure to police detection.

When repeat offenders are divided into those for whom repeat offences were expiated, and those for who repeat offences were prosecuted, it emerges that the majority of repeat offenders tend not to expiate CENs. While precise reasons for this cannot be ascertained from the data, repeat offenders are probably more likely to be known to police, and have other social problems which make their payment of expiation fees on time less likely.

When population survey data on levels of cannabis use are considered, it appears that the CEN scheme has had little impact. Data from national drug use surveys of the general community indicate that there has been a greater increase in self-reported lifetime cannabis use in South Australia between 1985 and 1995 than in the average of the other Australian states and territories. However, this increase is unlikely to be due to the CEN system because: (1) similar increases occurred in Tasmania and Victoria, where there was no change in the legal status of cannabis use; (2) there was no differential change in weekly cannabis use in South Australia as compared with the rest of Australia, and (3) there was no greater increase in cannabis use among young adults aged 14 to 29 years in South Australia. It is also possible that part of the observed increase in self-reported lifetime use in South Australia can be attributed to a greater willingness of people to admit to cannabis use in a population household survey, compared to jurisdictions with a total prohibition approach to minor cannabis offences

A final area of significant enquiry in the present study involved the opinions about the CEN scheme of law enforcement and criminal justice personnel working in South Australia. Senior officials involved in dealing with the CEN scheme in the SA Police and other departments have generally agreed that the CEN scheme should remain in place, and provides an efficient way of dealing with minor cannabis offences. It was also highlighted as provided advantages for offenders, through avoiding the stigma associated with criminal conviction. A significant concern was raised by some senior police, regarding the potential for exploitation of the expiable cultivation offence limit of 10 plants. It was argued that some commercial cannabis cultivation enterprises were spreading their operations, in order to keep individual site plantations at 10 plants or fewer, while maximising the yields through sophisticated cultivation techniques. It has been proposed that this problem can be readily dealt with by reducing the maximum number of plants expiable under the CEN scheme from ten to three or four. An alternative approach might be to maintain the ten plant limit, and try to impact on organised cultivation groups in other ways, such as police utilising the provision under the law, allowing commercial cultivation charges to be laid if the quantity of cannabis under cultivation or the circumstances lead them to suspect a commercial operation. Another suggestion that has been put forward is that, where police suspect that a commercial operation is taking advantage of the expiation provisions, police use the expiation scheme to repeatedly issue CENs for cultivation offences. This would require additional surveillance efforts, but continued seizure of plants and growing equipment may have a significant impact on these groups, and may be easier for police to sustain as a strategy for putting an end to their operations.

From the foregoing discussion, it has emerged that the CEN scheme in South Australia appears to have numerous benefits for the community, not the least of which are cost savings for the community as a whole, reduced negative social impacts for offenders, and greater efficiency and ease in having minor cannabis offences dealt with, associated with less negative views held by offenders of police. While there have been problems identified with the administration of the CEN scheme over time, it appears that the purported adverse effects associated with some unintended consequences of the CEN scheme are less problematic than previously thought. A good example is that, while net-widening has occurred under the system, it does not appear to have adversely affected court loads and costs. However, the effect of net-widening on offenders is less clear, as there is likely to be a small group of repeat offenders for whom repeated detection and prosecution may place them in greater financial hardship. One of the more telling findings from this study is that a significant number of CEN offenders will ultimately still receive a conviction as a result of an expiable offence, with the associated stigma and potential consequences. It remains to be seen whether modification of the CEN scheme, as has been already done to some extent via the changes to the administrative and payment procedures for expiable offences in South Australia, will result in fewer minor cannabis offenders receiving convictions.

Summary and Conclusions

1. The establishment of the Cannabis Expiation Notice (CEN) scheme in South Australia in 1987 has resulted in some degree of "net-widening", in that the number of minor cannabis offences detected under the scheme increased about two and a half times between 1987 and 1996. This increase appears to be mainly due the greater ease with which a CEN can be issued under the scheme, compared to the procedures for an arrest and charge that would required for a prosecution.
2. The majority of offences for which CENs have been issued in South Australia relate to the minor offences of possession of less than 25 grams of cannabis, possession of equipment for consuming cannabis, and cultivation of no more than 10 plants. Most CENs for equipment offences are issued in conjunction with another minor cannabis offence.
3. Since the introduction of the CEN scheme, the rate of expiation of notices has remained low compared with other types of infringement notices, and has remained fairly stable at approximately 45% for the last few years of operation of the scheme. The reasons for the low rate of expiation of cannabis offences are likely to relate to financial difficulty experienced by a substantial proportion of those detected for minor cannabis offences, as well as poor understanding amongst this group of the actual legal status of minor cannabis offences and the consequences of failure to pay expiation fees. In particular, there has been a fair degree of misunderstanding of the fact that failure to expiate a CEN can result in a criminal conviction being recorded against the offender.
4. Around 90%of those CENs which were forwarded for prosecution between 1991 and 1996 resulted in a conviction being recorded against the offender, because expiation fee payments were not made. This represents about 45% of all CENs issued over that period, and a large number of offenders for whom the conviction would have been avoided had they payed expiation fees on time.
5. The rate of expiation of CEN offences may improve following recent changes to the way in which all expiable offences are administered under the Expiation of Offences Act, 1996. The provision to offenders of a range of payment options (eg. instalment payments, community service) which can be specified before an unpaid CEN matter is forwarded for prosecution may result in a higher proportion of CENs being expiated. In addition, the provision of clearer and more detailed information on the consequences of failure to pay expiation fees may help to improve expiation rates. Furthermore, these changes may bring about a corresponding reduction in the proportion of CEN matters which result in a conviction for the offenders. Ongoing monitoring will be required to determine whether the new CEN forms and payment provisions bring about such changes.
6. Around 5% of CEN matters are withdrawn before payment is made or prosecution is completed, most likely due to inadequate information being available to ensure a successful prosecution in the event of failure to pay expiation fees. A further proportion would be withdrawn or dismissed after court proceedings have been completed, involving cases where the offenders could not be located for follow-up regarding payment of fines. The exact proportion of CEN cases involving false or incomplete information could not be ascertained in the present study. There may be scope for addressing this issue by providing supplementary training of operational police in the issuing of CENs.
7. National population survey data indicate there has been a national increase in self-reported lifetime cannabis use between 1985 and 1995, with a greater degree of increase in South Australia than in the average of the other Australian states and territories. However, the South Australian increase is unlikely to be due to the CEN system because: (1) similar increases occurred in Tasmania and Victoria, where there was no change in the legal status of cannabis use; (2) there was no differential change in weekly cannabis use in South Australia as compared with the rest of Australia, and (3) there was no greater increase in cannabis use among young adults aged 14 to 29 years in South Australia.
8. Many minor cannabis offenders in both South Australia and Western Australia appear to be people who are otherwise law-abiding. Surveys of samples of cannabis offenders in both states found that the majority in both states had respect for police and the law in general. It was also found that their offence apprehension and subsequent arrest (WA) or issuing of a CEN (SA) had no impact on their patterns of cannabis or other drug use.
9. Interviews with cannabis offenders found that negative employment consequences arising from a cannabis offence apprehension were more likely to be experienced by offenders in Western Australia compared to South Australia (eg. loss of job, missing out on a job opportunity). Those in the WA system were also more likely to report relationship problems, accommodation problems, and further involvement with the criminal justice system related to their first minor cannabis offence. In terms of impacts upon drug use and travel opportunities, no differences were found between offenders in both states. However, offenders in Western Australia were more likely than those in South Australia to have more less favourable attitudes towards police following their cannabis offence detection.
10. No differences were found in the self-reported attitudes of employers in both South Australia and Western Australia towards employing people with prior cannabis offences, there being a general lack of discrimination expressed against such offenders. This is somewhat at odds with the reported experiences of cannabis offenders in the two states. It was clear that cannabis offending is not an important part of employer screening in many employment areas, although employers in both states were concerned about the potential risks associated with cannabis intoxication in the workplace, and the long term effects of cannabis use on work performance.
11. While there is a level of acceptance in the South Australian community of personal cannabis use among adults, activities relating to the commercial sale or supply of cannabis are not viewed favourably by the public.
12. The general public in South Australia had a reasonable awareness of the CEN scheme in general, but retained a fair degree of confusion about the details of the CEN scheme and the consequences of being detected for various minor cannabis offences. There was also incomplete recognition of the possible health risks associated with long-term or heavy cannabis use. As many people issued with expiation notices are heavy consumers of cannabis, there is an opportunity to deliver health messages with the notice at the time of issuing.
13. Despite the fairly low rate of expiation and the apparent "net-widening" observed under the CEN scheme since its implementation in 1987, it would seem that the scheme is more cost effective for dealing with minor cannabis offences than a prohibition approach based predominantly on prosecution and conviction. It might be expected that greater efficiencies could be achieved if the rate of expiation can be increased in the future, with a corresponding reduction in the number of CEN fee defaulters who receive convictions.
14. There is clear and widespread support for the CEN scheme amongst South Australia law enforcement and criminal justice personnel. This support is based largely on the perception that the expiation approach provides a fair and cost-effective way of dealing with minor cannabis offences.
15. Concern has been expressed by some South Australian police officers about the potential for exploitation of the CEN scheme by organised criminal syndicates who grow commercial quantities of cannabis in separate locations while operating within the expiable cultivation limit of 10 plants. In order to address this issue, it has been suggested that the CEN scheme be modified to reduce to maximum expiable number of plants under cultivation from ten to three or four.
16. In addition to the provision of more payment options for offenders and more detailed information on the financial and legal consequences of non-payment, other suggestions have been put forward for possible changes to the CEN scheme in South Australia, which may be of interest to other jurisdictions considering the adoption of expiation systems for minor cannabis offences. A system involving a more graduated scale of expiation fees, including lesser fees for offences involving very small amounts of cannabis, could result in higher rates of expiation, especially where offences involve young people. Other suggestions which may reduce the extent of net-widening under an expiation approach, should that be deemed desirable, are: inclusion of a provision for some form of cautioning for certain categories of minor cannabis offence (although such use of police discretion in the issuing of cautions may not be deemed desirable where full accountability in relation to all minor offence detections is required); dropping the offence of possession of equipment for using cannabis, as it is a very common offence under the CEN scheme, and is mostly detected in the context of CENs being issued for other cannabis offences.