A

IficiMtbpRatayoldCiccs©

K

I

ihi

0d

International Journal of Drug Policy 18 (2007) 187–193

Research paper

Individual dollar expenditure and earnings fromcannabis in the New Zealand population

Chris Wilkins ∗, Paul SweetsurCentre for Social and Health Outcomes Research and Evaluation (SHORE), Massey University,

P.O. Box 6137, Wellesley Street, Auckland, New Zealand

Received 8 August 2006; received in revised form 17 October 2006; accepted 1 November 2006

bstract

ntroduction: : High spending on illegal drug use can potentially provide insight into a range of drug related harm such as poor health,nancial hardship, loss of opportunity, family neglect and income generating crime. Assessing the impact of high spending on drug use isomplicated by the fact that many heavy drug users support their high expenditure on drug use through selling drugs. The aim of this papers to estimate individual dollar expenditure and dollar earnings from cannabis in New Zealand.

ethods: : As part of the 2003 New Zealand national household drug survey, detailed data were collected on cannabis use and purchasing inhe previous 12 months. Those who had purchased cannabis in the previous year were divided into two groups for further analysis: ‘cannabisuyers’ (i.e. those who only purchased sufficient cannabis for their own personal consumption needs); and ‘cannabis dealers’ (i.e. those whourchased large surpluses of cannabis for re-sale on to others).esults: : Seventy-two percent of those who had purchased cannabis in the previous year were ‘cannabis buyers’ only. Cannabis buyers spentmean of $817 (NZD) each on cannabis in the previous year (median $120). Cannabis dealers spent a mean of $5988 each on cannabis in

he preceding year (median $1250). Once we accounted for projected earnings from selling surplus cannabis, half of the cannabis dealerschieved a mean net annual financial gain of $2739, and the remaining half were left with a mean net annual financial loss of only $350 over aear. Overall, 81% of all those who had purchased cannabis in the previous year had spent less than 5% of their gross annual personal incomen cannabis, with 14% of these making a net financial gain. The proportion of income spent on cannabis was highest among those in the twoowest income earning groups, where approximately 8% spent 20% or more of their income on cannabis. Both cannabis buyers and cannabisealers spending 10% or more of their income on cannabis were four times more likely to be unemployed than the wider population.onclusions: : For the vast majority of those who had purchased cannabis in the previous year, spending on cannabis had only a small

mpact on their total annual income levels. High spending on cannabis can potentially be substantially offset by earnings from selling surplus

annabis. High spending on cannabis had its greatest impact among low income earning groups where approximately one in 13 low incomeannabis spenders were spending high proportions of their income on cannabis. There appeared to be some relationship between high cannabispending and unemployment and this warrants further investigation.

2006 Elsevier B.V. All rights reserved.

tional h

cmc

eywords: Cannabis expenditure; Cannabis dealing; Cannabis earnings; Na

ntroduction

The level of individual spending on illegal drug use is an

mportant, though often neglected, component of drug relatedarm. Money which is spent on illegal drugs is money whichs not available for food, housing, health care, education and

∗ Corresponding author. Tel.: +64 9 366 6136; fax: +64 9 366 5149.E-mail address: c.wilkins@massey.ac.nz (C. Wilkins).

mihtt

955-3959/$ – see front matter © 2006 Elsevier B.V. All rights reserved.oi:10.1016/j.drugpo.2006.11.002

ousehold drug survey; New Zealand

hild support. For drug users on low incomes, even whatight seem like only modest dollar expenditures on drug use

an create financial hardship.National household drug surveys potentially provide the

ethodology and data to estimate individual spending on

llegal drug use in the population. A number of researchersave used national household drug survey data to estimatehe total dollar value of illegal drug markets and, as part ofhese market estimates, have produced one-off estimates of

1 l Journa

iduOBCvatasGNB

c(Mttetuthwhswc

cgFabea&Bctl&domup

eatcaw

aZ

M

wSsaZtlTttdastHTvtcbror

C

tcy‘oo(fmr‘a

D

dow

88 C. Wilkins, P. Sweetsur / Internationa

ndividual expenditure on a drug by simply dividing the totalollar value of the market by the number of users in the pop-lation (Caputo & Ostrom, 1994; Clements & Daryal, 1999;ffice of National Drug Control Policy, 2000, 2001; Wilkins,hatta, & Casswell, 2002a; Wilkins, Reilly, Pledger, &asswell, 2005). These average per head estimates of indi-idual expenditure on drug use are not likely provide a veryccurate picture of individual spending on drug use, due tohe large variation in levels of individual drug consumptionnd drug spending commonly found among populations andub-populations of drug users (see Caputo and Ostrom, 1994;olub and Johnson, 2004; Johnson et al., 1985; Office ofational Drug Control Policy, 2000, 2001; Wilkins, Pledger,hatta, & Casswell, 2004; Wyllie, Zhang, & Casswell, 1993).

Heavy drug users have also often been found to finan-ially support their drug use with earnings from drug dealingAlder, 1985; Jacobs, 1999; Pearson & Hobbs, 2001; Reuter,

acCoun, & Murphy, 1990). Large purchases of a drug mayherefore represent sources of income as well as expendi-ure (see Johnson et al., 1985). The impact of absolute dollarxpenditures on illegal drug use can also only be evaluated inhe context of drug users’ own personal income levels. Drugsers as a group may have lower incomes than the rest ofhe population and hence may experience greater financialardship for a given absolute dollar expenditure than moreealthy individuals. Much of the focus in the literature to dateas been spending on the more expensive illegal drug types,uch as heroin and cocaine (see Sutton & Maynard, 1992),ith only cursory attention paid to the impact spending on

annabis may have on drug user’s living standards.In New Zealand, as in many countries in the world,

annabis is by far the most widely used and purchased ille-al drug (Black & Casswell, 1993; Caulkins & Pacula, 2006;ield & Casswell, 1999; United Nations Office on Drugsnd Crime, 2005; Wilkins et al., 2002a). There is known toe large scale clandestine cultivation of cannabis in severalconomically depressed rural regions in New Zealand andsubstantial national black market for cannabis (WilkinsCasswell, 2003; Wilkins et al., 2002a, 2005; Wilkins,

hatta, & Casswell, 2002b; Yska, 1990). High spending onannabis has periodically been raised as a possible factor con-ributing to economic and social disadvantage among someow income groups in New Zealand (see Walker, Cocklin,

Blunden, 1998). However, to date, there has been noetailed examination of the extent of individual spendingn cannabis, or the magnitude of earnings which can beade from selling cannabis, among the New Zealand pop-

lation or among lower income groups in the New Zealandopulation.

The aim of this paper is to estimate individual dollarxpenditure and individual dollar earnings from purchasingnd selling cannabis in New Zealand. We distinguish between

wo types of cannabis purchaser: (i) those who purchaseannabis solely for their own personal consumption needs;nd, (ii) those who purchase large surpluses of cannabis,hich are assumed to be for resale on to others. The estimates

ifiw$

l of Drug Policy 18 (2007) 187–193

re calculated using population level data from the 2003 Newealand national household drug survey.

ethod

The 2003 New Zealand national household drug survey,hich is officially known as the 2003 Health Behavioursurvey–Drug Use (2003 HBS-Drug Use), collected a nationalample of 5727 people aged 13–45 years using a computerssisted telephone interviewing (CATI) methodology. Newealand has high levels of telephone ownership by interna-

ional standards, with 96% of households having a connectedandline telephone at the last national population census.o represent the different socio-economic characteristics of

he population the country was divided into 34 strata. Eachelephone number was tried at least ten times at differentates and times of the day in an effort to reach those seldomt home. Within each household one person was randomlyelected for an interview. Respondents were informed thathe study was being conducted on behalf of the Ministry ofealth and that everything they said would be confidential.he questionnaire was designed so that the telephone inter-iew would remain private from anyone who might be inhe room with the respondent when the interview was beingonducted. Questions asked over the telephone could onlye heard by the respondent and responses generally onlyequired one word answers, such as ‘yes/no’, ‘higher/lower’,r ‘all/most/some/hardly any’. An overall weighted surveyesponse rate of 68% was achieved.

annabis purchasing

Those cannabis users who had purchased cannabis inhe previous 12 months were asked whether they had pur-hased any from a list of weights of cannabis in the previousear. The weights asked about were ‘joints’ (approx. 0.5 g);tinnies’ (approx. 1.5 g); ‘$50 bags’ (approx. 4 g); quarterunces (approx. 7.1 g); third ounces (approx. 9.5 g); halfunces (approx. 14.2 g); ounces (approx. 28.4 g); and poundsapprox. 453.6 g). These weights of cannabis were all the dif-erent weights of cannabis known to be sold on the blackarket in New Zealand. For each weight of cannabis that the

espondent indicated they had purchased, they were askedhow many of that weight they had purchased in the last year’nd ‘how much they would expect to pay for that weight’.

emographics

Survey respondents are particularly sensitive to revealingetails about personal income and so, to enhance the levelf response to the income question, respondents were askedhether their gross annual personal income fell into twelve

ncome bands, rather than being asked to reveal a precisegure. The income categories offered to the respondentere ‘Zero/Nil’, ‘$1–$5000’, ‘$5001–$10,000’, ‘$10,001–15,000’, ‘$15,001–$20,000’, ‘$20,001–$25,000’, ‘$25,001

l Journa

–‘mcb

A

vtcheeptdsmhiosofpc‘ostcoAidbTcwg

1tpeear(

R

sacnvgt

cida(qssostctecio

TMa

Gi

0$$$$>A

C. Wilkins, P. Sweetsur / Internationa

$30,000’, ‘$30,001–$40,000’, ‘$40,001–$50,000’,$50,001–$70,000’, ‘$70,001–$100,000’, and ‘$100,001 orore’ (New Zealand Dollars (NZD)). The survey income

ategories were combined into wider income groups toroadly represent different levels of socio economic status.

nalysis

The sample data was weighted by a range of demographicariables to ensure the survey provided an accurate represen-ation of the New Zealand population based on populationensus data. The sample was also weighted by eligible house-old size to adjust for the selection of only one person fromach household. Each household telephone number had anqual chance of selection, and as we interviewed only oneerson in each household, we had to weight respondents byhe number of people in the household to account for theifferent probability of selection in households of differentizes. Respondents were assigned an income value using theidpoint of the gross annual personal income category they

ad chosen. Respondents reporting a gross annual personalncome of ‘$100,001 or more’ were assigned an income valuef $150,000. The New Zealand Misuse of Drugs Act 1975tates that those found with 28 g or more of cannabis at anyne time are legally presumed to be in possession of cannabisor the purposes of supply. Both cannabis use and cannabisurchase data in the survey was only collected for the pre-eding 12 months. Consequently, we defined respondents ascannabis dealers’ if they had purchased a surplus of 28 gr more of cannabis, once their own personal cannabis con-umption was deducted from their cannabis purchases, overhe previous year. Those who had either consumed all theannabis they purchased, or had a yearly surplus of cannabisf less than 28 g, were categorized as ‘cannabis buyers’ only.multiple logistic regression was run with backwards elim-

nation removing non-significant variables to test whetheremographic characteristics were independently related toeing defined as a ‘cannabis buyer’ or a ‘cannabis dealer’.

he analysis was repeated for both cannabis buyers andannabis dealers to test which demographic characteristicsere independently related to having spent 10% or more ofross annual personal income on cannabis in the previous

o

pv

able 1ean dollar (NZD) spending on cannabis by cannabis buyers (unweighted sample n

nd by annual income band

ross annual personalncome ($)

All (unweighted) Percentage of gross annual personal

0–4% 5–9% 10–

–$5,000 49 $52(30) $186(8)5,001–$15,000 39 $89(30) $808(2) $115,001–$25,000 58 $246(40) $1,042(7) $225,001–$40,000 43 $208(36) $2,284(6) $640,001–$70,000 27 $243(27) –$70,001 5 $1,347(4) – $12ll 221 $206(167) $1,135(23) $1,

l of Drug Policy 18 (2007) 187–193 189

2 months. Odds ratios (OR) were calculated and tested athe 95% confidence interval. The demographics tested wereersonal income, gender, ethnicity, employment status, high-st educational qualification, marital status and age. With thexception of age, all of these variables were categorical. Allnalysis was conducted using SAS statistical software envi-onments. All dollar estimates are in New Zealand DollarsNZD) (1$NZ = 0.3317 £ = 0.481D = US$ 0.6179).

esults

One in five (18.4%, unweighted n = 1217) of the totalample had used cannabis in the preceding 12 months,nd one-third (34%) of these last year users had purchasedannabis in the previous year (6.1% of the total sample,= 382). Of those who had purchased cannabis in the pre-ious year, 87.0% had also provided information on theirross annual personal income in the demographic section ofhe survey (5.3% of the total sample, n = 322).

Seventy-two percent (n = 221) of those who had pur-hased cannabis in the previous year and had providednformation on their gross annual personal income wereefined as ‘cannabis buyers’ only. The cannabis buyers spentmean of $817 each on cannabis in the previous year

median $120, range $5–$22,500, lower quartile—$50, upperuartile—$500). Seventy-seven percent of cannabis buyerspent less than 5% of their gross personal income on cannabis,pending a mean of $206 each per year, or $4 each per week,n cannabis (Table 1). Six percent of the cannabis buyerspent 20% or more of their personal income on cannabis inhe previous year. The proportion of total income spent onannabis was highest among the cannabis buyers from thewo lowest income earning groups. Among cannabis buyersarning $0–$5000, 8% spent 20% or more of their income onannabis in the previous year. Among cannabis buyers earn-ng $5001–$15,000, 9% spent 20% or more of their incomen cannabis in the previous year. No cannabis buyers earning

ver $25,000 spent 20% or more of their income on cannabis.

The remaining 28% (n = 101) of those who hadurchased cannabis in the previous year and had pro-ided information on their gross annual personal income

umbers) by percentage of gross annual personal income spent on cannabis

income

19% 20–39% 40–59% 60–99% >100%

$369(5) $827(2) $1,360(1) $2,095(2) $22,500(1),000(1) $2,503(5) – $8,400(1) –,508(7) $4,972(3) $9,600(1) – –,000(1) – – – –

– – – – –,600(1) – – – –670(15) $2,631(10) $6,494(2) $4,470(3) $22,500(1)

190 C. Wilkins, P. Sweetsur / International Journal of Drug Policy 18 (2007) 187–193

Table 2Mean dollar spending (NZD) on cannabis by cannabis dealers (unweighted sample numbers) by percentage of gross annual personal income spent on cannabisand by annual income band

Gross annual personalincome ($)

All (unweighted) Percentage of gross annual personal income

0–4% 5–9% 10–19% 20–39% 40–59% 60–99% >100%

0–$5,000 18 – – $297(4) $584(4) $1,033(7) $1,932(2) $2,500(1)$5,001–$15,000 19 $400(1) $602(3) $1,655(4) $2,653(3) $4,610(4) $7,281(2) $15,578(2)$15,001–$25,000 27 $576(10) $1,537(13) $3,000(1) $5,000(1) $10,000(1) – $40,000(1)$25,001–$40,000 22 $744(13) $2,489(6) $4,161(3) – – – –$40,001–$70,000 11 $997(8) $4,473(2) – – – – $90,000(1)>A $2,

we$qspodioA2at

dc2wipicmo(at$

foidyli$looficeiOte

catspcs

TMn

Gi

0$$$$>A

$70,001 4 $1,546(4) –ll 101 $896(36) $2,023(24)

ere defined as ‘cannabis dealers’. The cannabis deal-rs spent a mean of $5988 each on cannabis (median1250, range $100–$90,000, lower quartile—$600, upperuartile—$3000). Thirty-one percent of cannabis dealerspent 20% or more of their total income on cannabis in therevious year, and 23% spent 40% or more of their incomen cannabis in the previous year (Table 2). Among cannabisealers earning $0–$5000, 79% spent 20% or more of theirncome on cannabis in the previous year, and 48% spent 40%r more of their income on cannabis in the last 12 months.mong cannabis dealers earning $5001–$15,000, 60% spent0% or more of their income on cannabis in the previous year,nd 52% spent 40% or more of their income on cannabis inhe last 12 months.

The mean surplus of cannabis purchased by the cannabisealers was 539 g, or 19 ounces, of cannabis once personalannabis consumption was deducted (median 91 g, range1–9050 g, lower quartile—42 g, upper quartile—196 g). Weere able to make some projections of the likely dollar

ncome earned by the cannabis dealers from selling this sur-lus cannabis by multiplying the surplus quantity of cannabisn grams held by each dealer by the average price paid forannabis at the retail level of the market. Cannabis is com-only sold at retail level in New Zealand in weights of 4 g

r less [i.e. $50 bags (4 g), ‘tinnies’ (1.5 g) and joints (0.5 g)]see Wilkins & Sweetsur, in press; Wilkins et al., 2005). The

verage price paid per gram of cannabis for all the transac-ions reported by the survey respondents of 4 g or less was13.32. Half of the cannabis dealers made a net financial gain

tr(

able 3ean dollar spending (NZD) on cannabis by last year cannabis dealers adjusted

umbers) by percentage of gross annual personal income spent on cannabis and by

ross annual personalncome ($)

All (unweighted) Profit Percentage of gros

0–4% 5–9

–$5,000 18 $43(4) $89(7) $5,001–$15,000 19 $702(7) $142(7) $15,001–$25,000 27 $3,231(14) $257(12) $1,25,001–$40,000 22 $602(12) $269(8) $1,40,001–$70,000 11 $8,857(8) $229(3)$70,001 4 $210(3) $33(1)ll 101 $2,739(48) $186(38) $

– – – – –293(12) $1,118(8) $2,794(12) $5,685(4) $40,913(5)

rom selling surplus cannabis once their dollar expendituresn cannabis were deducted from their projected dollar earn-ngs (Table 3). The mean net financial gain of the cannabisealers who achieved a net financial gain was $2739 each perear, or $52 each per week (median $333, range $7–$80,544,ower quartile—$45, upper quartile—$1214). The remain-ng cannabis dealers experienced a mean net financial loss of350 dollars each per year (median $142, range $18–$3934,ower quartile—$54, upper quartile—$364). Ninety percentf the cannabis dealers spent less than 5% of their incomen cannabis, with 50% of the cannabis dealers making a netnancial gain once projected earnings from selling surplusannabis were taken into account. Only 2% of cannabis deal-rs spent 10% or more of their income on cannabis oncencome from selling surplus cannabis was taken into account.nly cannabis dealers earning $15,000 or less spent more

han 10% of their income on cannabis once we adjusted forarnings from selling surplus cannabis (Table 3).

Table 4 combines the expenditures on cannabis of theannabis buyers and the cannabis dealers to provide an over-ll picture of spending on cannabis, with the expenditures ofhe cannabis dealers offset by their projected earnings fromelling surplus cannabis. Fourteen percent of those who hadurchased cannabis in the previous year made a net finan-ial gain from cannabis, once projected income from sellingurplus cannabis were taken into account. The remainder of

hose who had purchased cannabis in the previous year expe-ienced a mean net financial loss of $742 each over a yearmedian $120, range $5–$22,500, lower quartile—$52, upper

for projected earnings from selling surplus cannabis (unweighted sampleannual income band

s annual personal income

% 10–19% 20–39% 40–59% 60–99% >100%

157(4) $282(1) $580(2) – – –633(2) – $3,130(3) – – –471(1) – – – – –712(2) – – – – –

– – – – – –– – – – – –

858(9) $282(1) $1,751(5) – – –

C. Wilkins, P. Sweetsur / International Journal of Drug Policy 18 (2007) 187–193 191

Table 4Mean dollar spending (NZD) on cannabis by all those who had purchased cannabis in the previous year adjusted for projected earnings from selling cannabis(unweighted sample numbers) by percentage of gross annual personal income spent on cannabis and by gross annual personal income band

Gross annual personalincome ($)

All (unweighted) Profit Percentage of gross annual personal income

0–4% 5–9% 10–19% 20–39% 40–59% 60–99% >100%

0–$5,000 67 $43(4) $59(37) $177(12) $365(6) $694(4) $1,360(1) $2,095(2) $22,500(1)$5,001–$15,000 58 $702(7) $97(37) $703(4) $1,000(1) $2,616(8) – $8,400(1) –$15,001–$25,000 85 $3,231(14) $249(52) $1,050(8) $2,508(7) $4,972(3) $9,600(1) – –$25,001–$40,000 65 $602(12) $220(44) $2,125(8) $6,000(1) – – – –$40,001–$70,000 38 $8,857(8) $241(30) – – – – – –>A 1,076(3

qcotctctphte2Aa9pto

D

9dtetttl1619bs(icC

81

D

Ibtsgludhaitittplrtc(P

pspwstuc

$70,001 9 $210(3) $1,073(5)ll 322 $2,739(48) $202(205) $

uartile—$454). Eighty-one percent of those who had pur-hased cannabis in the previous year had spent less than 5%f their total gross annual personal income on cannabis inhe previous year, with 14% of these making a net finan-ial gain once projected income from selling cannabis wasaken into account. Only 4% of those who had purchasedannabis in the last 12 months had spent 20% or more ofheir income on cannabis in the previous year. The pro-ortion of total income spent on cannabis continued to beigher among the lowest income earning groups. Amonghose who had purchased cannabis in the previous year andarned $0–$5000 gross annual personal income, 8% spent0% or more of their income on cannabis in the last year.mong those who had purchased cannabis in the last year

nd earned $5001–$15,000 gross annual personal income,% spent 20% or more of their income on cannabis in therevious year. None of those who had purchased cannabis inhe last year and who earned over $25,000 spent 20% or moref their income on cannabis.

emographics

Cannabis buyers were more likely to male (OR = 2.09,5% CI: 1.54, 2.84), to be employed rather than a stu-ent (OR = 1.78, 95% CI: 1.09, 2.84), to be single/separatedhan with a partner (OR = 2.56, 95% CI: 1.75, 3.75) toarn $5001–$15,000 (OR = 1.74, 95% CI: 1.05, 2.86) ando earn $15,001–$25,000 (OR = 2.08, 95% CI: 1.17, 3.68)han $0–$5000. Cannabis buyers spending 10% or more ofheir gross annual personal income on cannabis were moreikely to be male (OR = 6.17, 95% CI: 2.11, 17.85), aged3–17 years old than 25–45 years old (OR = 29.41, 95% CI:.36, 142.85), to be unemployed (OR = 4.36, 95% CI: 1.56,2.19), and have no educational qualifications (OR = 2.49,5% CI: 1.00, 6.17). Cannabis dealers were more likely toe male (OR = 2.68, 95% CI: 1.64, 4.36) and to have a highchool qualification rather than no educational qualification

OR = 2.48, 95% CI: 1.17, 5.27). Cannabis dealers spend-ng 10% or more of their gross annual personal income onannabis were more likely to be unemployed (OR = 3.77, 95%I: 1.33, 10.64), to be students (OR = 3.33, 95% CI: 1.37,

cr2S

– $12,600(1) – – – –2) $1,649(16) $2,412(15) $6,494(2) $4,470(3) $22,500(1)

.12) and be of single martial status (OR = 4.18, 95% CI:

.48, 11.82).

iscussion

Several limitations with our analysis are acknowledged.t is likely that our dollar estimates of cannabis expenditure,ased on national household drug survey data, understate therue extent of heavy cannabis spending in New Zealand toome extent. While national household drug surveys provideood measures of occasional moderate drug use, they areikely to underestimate the level of heavy drug use in the pop-lation (MacCoun & Reuter, 2001). This is because heavyrug users are more likely to be non-contactable throughousehold surveying, as they tend to live erratic lifestyles, andre more likely to be homeless, have no permanent address, bencarcerated, and be in drug treatment. There is good reasono believe that there is much lower levels of under-reportingn the case of cannabis than other ‘harder’ drug types as hard-o-reach populations (i.e. prisoners, homeless, those in drugreatment and addicted users) account for a much smaller pro-ortion of the total cannabis using population. Furthermore,egal penalties for cannabis use are often fairly modest whicheduces the propensity of respondents to withhold informa-ion out of fear of prosecution, and social disapproval ofannabis use is much lower than for many other ‘hard’ drugsCaulkins & Pacula, 2006; Office of National Drug Contrololicy, 2001).

Second, our projections of the earnings from selling sur-lus cannabis detail some of the most disciplined sellingcenarios where a cannabis dealer manages to sell all their sur-lus cannabis rather than consuming it themselves, sharing itith friends, or having it seized by law enforcement. Analy-

is of illegal markets for cannabis in both New Zealand and inhe United States suggest a fairly large proportion of cannabissers obtain their cannabis for ‘free’, most likely during grouponsumption sessions at social gatherings, rather than pur-

hasing it, and that even those involved in selling cannabisegularly give cannabis away for free (Caulkins & Pacula,006; Taylor & Costa, 2003; Wilkins et al., 2005; Wilkins &weetsur, in press).

1 l Journa

fcocg

gcctsnnpghffBatTfilactmWnetrtibcT(mtw

fcew$orBofub

toaiCaasclhsiqg

pnorporatbddtlauwidRbss&ftssbs

A

f

92 C. Wilkins, P. Sweetsur / Internationa

Three conclusions with policy implications can be drawnrom this paper. Firstly, the vast majority of those who pur-hase cannabis spend only a small proportion of their incomesn cannabis. Eight out of 10 of those who had purchasedannabis in the previous year had spent less than 5% of theirross annual income on cannabis.

Secondly, high spending on cannabis can potentially bereatly offset by the earnings made from selling surplusannabis. Cannabis dealers spent an average of $5988 each onannabis in the previous year but, after we took into accounthe earnings which could potentially be made from sellingurplus cannabis, half of the cannabis dealers made a meanet financial gain and the remaining half suffered a meanet financial loss of only $350 per year (median $142). Therofits earned from selling surplus cannabis, however, wereenerally not of the magnitude to afford sports cars and luxuryouses. In New Zealand, the average gross annual earningsor a sales worker is $21,372, for an office clerk $29,328, andor a professional $48,984 (Statistics New Zealand, 2005).ased on our projections, only 50% of the cannabis dealersctually made a net financial gain from selling cannabis onceheir own cannabis consumption expenses were included.he median earnings among cannabis dealers making a netnancial gain from cannabis were $333 per year, with the

ower quartile of dealers only earning $45 per year. As notedbove, our projections represent some of the most optimisticannabis earning scenarios. These projected earnings suggesthat the majority of cannabis sellers are earning just enough

oney to cover their own cannabis consumption expenses.hile income earned from cannabis dealing may be a conve-

ient way for heavy cannabis users to alleviate the financialxpense of high cannabis use, it is unlikely to be viewed byhe authorities as a positive strategy to ensure financial secu-ity. Our projections of the modest earnings available fromhe retail level selling of cannabis in New Zealand are notnconsistent with the often fairly low earnings which haveeen previously found for street level cocaine and street levelrack dealing in the United States (see Caulkins, Johnson,aylor, & Taylor, 1999; Reuter et al., 1990). Reuter et al.1990) found that street level cocaine sellers earned fairlyodest incomes with many dealers merely supplementing

heir earnings from full and part time employment in lowage jobs with income from occasional cocaine selling.Thirdly, even when we accounted for possible earnings

rom selling surplus cannabis, the burden of spending onannabis remained highest among those cannabis spendersarning the lowest incomes. Approximately one in 13 of thoseho had purchased cannabis in the previous year and earned15,000 or less were spending 20% or more of their incomen cannabis. Many of those earning $15,000 or less will beecipients of government provided social welfare payments.oth cannabis buyers and cannabis dealers spending 10%

r more of their income on cannabis were approximatelyour times more likely to unemployed than the wider pop-lation. This suggests that there may be some relationshipetween high cannabis spending and economic disadvan-

GIZT

l of Drug Policy 18 (2007) 187–193

age. It is not possible to tell from our analysis the directionf any casual relationship between high cannabis spendingnd unemployment. High spending on cannabis may wellncrease the likelihood that a person will be unemployed.onversely, unemployment may create the emotional stressnd extended leisure time which results in increased usend spending on cannabis. Alternatively, both high cannabispending and unemployment may be explained by some otherausal factor such as low educational achievement. Regard-ess of the nature or direction of any relationship betweenigh spending on cannabis and unemployment, our analysishows that high spending on cannabis has a much greatermpact on low income earning cannabis spenders and conse-uently contributes to lower standards of living among theseroups.

Further research of the role cannabis use and spendinglays in economic disadvantage could be undertaken in aumber of areas. Qualitative research of heavy cannabis usersn low incomes may provide a clearer understanding of theole heavy cannabis use and high spending on cannabis maylay in economic disadvantage. Research into the wider issuef drug use and economic disadvantage would of courseequire the inclusion of expenditure on legal drugs, such aslcohol and tobacco, and expenditure on other illegal drugypes, such as amphetamine and ecstasy. There may alsoe specific interest in the impact drug expenditure has onisadvantaged ethnic minorities or those in economicallyepressed regions of a country. In New Zealand, achievinghese research objectives would require a commitment to aarger sample size in future national household drug surveysnd the inclusion of additional economic questions on these of other drug types. An important issue when decidinghat policy measures should be enacted on the basis of find-

ngs concerning drug spending is the extent that differentrug types are substitutes or complements in consumption.ecent studies have suggested that alcohol and cannabis maye consumption complements and hence policy measuresuch as increasing the level of taxation on alcohol will reducepending on cannabis as well alcohol (Pacula, 1998; Williams

Mahmoudi, 2004). Alternatively, if two drug types wereound to be substitutes, policy measures which increasedhe cost of one drug type may merely result in drug usersubstituting the more expensive drug type for the less expen-ive drug type. An important issue in this scenario woulde whether this product substitution would bring about a netocial health gain.

cknowledgements

The funding for the analysis in this paper was awardedrom the New Zealand National Drug Policy Discretionary

rant Fund which is jointly managed by the New Zealand

nter-Agency Committee on Drugs (IACD) and the Newealand Ministerial Committee on Drug Policy (MCDP).he 2003 HBS-Drug Use was funded by the New Zealand

l Journa

MHtbfetcJH

R

A

B

C

C

C

C

F

G

J

J

M

O

O

P

R

S

S

T

T

U

W

W

W

W

W

W

W

Wyllie, A., Zhang, J., & Casswell, S. (1993). Drinking patterns and prob-

C. Wilkins, P. Sweetsur / Internationa

inistry of Health and carried out as part of the Publicealth Intelligence (PHI) Health Behaviours Survey Moni-

or. The statistical design for the 2003 survey was completedy Dr. Megan Pledger at SHORE, and by James L. Reillyrom Statistical-Insights. The questionnaire was adapted andxtended from the National Drug Survey questionnaire forhe Health Behaviours programme by Dr. Chris Wilkins. Dataollection was managed by Dr. Chris Wilkins, Rachael Lane,oe Morley and Mary Blade. The statistical analysis of 2003BS-Drug Use was completed by Paul Sweetsur.

eferences

lder, P. (1985). Wheeling and dealing: An ethnography of an upper-leveldrug dealing and smuggling community. New York: Columbia Univer-sity Press.

lack, S., & Casswell, S. (1993). Drugs in New Zealand: A survey 1990.University of Auckland: Alcohol & Public Health Research Unit.

aputo, M., & Ostrom, B. (1994). Potential tax revenue from a regulatedmarijuana market: A meaningful revenue source. American Journal ofEconomics and Sociology, 53(4), 475–490.

aulkins, J., Johnson, B., Taylor, A., & Taylor, L. (1999). What drug dealerstell us about their costs of doing business. Journal of Drug Issues, 29(2),323–340.

aulkins, J., & Pacula, R. (2006). Marijuana markets: Inferences fromreports by the household population. Journal of Drug Issues, 36(1),173–200.

lements, K., & Daryal, M. (1999). The economics of marijuana consump-tion. Perth: Economic Research Centre, University of Western Australia.

ield, A., & Casswell, S. (1999). Drugs in New Zealand: A national sur-vey 1998. University of Auckland: Alcohol and Public Health ResearchUnit.

olub, A., & Johnson, B. (2004). How much do Manhattan arrestess spendon drugs? Drug and Alcohol Dependence, 76, 235–246.

acobs, B. (1999). Dealing crack: The social world of street corner selling.Boston: NorthEastern University Press.

ohnson, B., Goldstein, P., Preble, E., Schmeidler, J., Lipton, D., Spunt, B.,et al. (1985). Taking care of business: The economics of crime by heroin

users. Lexington, MA: Lexington Books.

acCoun, R., & Reuter, P. (2001). Drug war heresies: Learning from othervices, times, and places. New York: Cambridge University Press.

ffice of National Drug Control Policy. (2000). What America’s usersspend on illegal drugs 1988–1998. Rockville, MD: ONDCP, Execu-

Y

l of Drug Policy 18 (2007) 187–193 193

tive Office of the President of the United States. http://www.whitehou-sedrugpolicy.gov/index.html.

ffice of National Drug Control Policy. (2001). What America’s users spendon illegal drugs 1988–2000. ONDCP, Executive Office of the Presidentof the United States. http://www.whitehousedrugpolicy.gov/index.html.

earson, G., & Hobbs, D. (2001). Middle market drug distribution. London:Home Office Research Development and Statistics Directorate

euter, P., MacCoun, P., & Murphy, P. (1990). Money from crime: A studyof the economics of drug dealing in Washington, DC. California: RANDCorporation.

tatistics New Zealand. (2005). Labour market statistics 2004. Wellington.http://www.statistics.govt.nz.

utton, M., & Maynard, A. (1992). What is the size and nature of the “DrugProblem” in the UK? University of York: Centre for Health Economics

aylor, B., & Costa, M. (2003). Drug markets, In 2000 Arrestee drug abusemonitoring: Annual report. Washington, DC: National Institute of Jus-tice, Office of Justice, Programs, US Department of Justice.

e Runanga O Te Rarawa. (1995). Cannabis Project Report. Te Runanga OTe Rarawa.

nited Nations Office on Drugs and Crime. (2005). World drug report2005. Vienna: UNODC. http://www.unodc.org/unodc/en/world drugreport.html.

alker, L., Cocklin, C., & Blunden, G. (1998). Cannabis highs and lows:Sustaining and dislocating rural communities in Northland. Universityof Auckland: Department of Geography

ilkins, C., Bhatta, K., & Casswell, S. (2002a). A demand side estimateof the financial turnover of the cannabis black market in New Zealand.Drug and Alcohol Review, 21, 145–151.

ilkins, C., Bhatta, K., & Casswell, S. (2002b). The effectiveness ofcannabis crop eradication operations in New Zealand. Drug and AlcoholReview, 21, 369–374.

ilkins, C., & Casswell, S. (2003). Cannabis cultivation and organised crimein New Zealand. Contemporary Drug Problems, 30, 757–777.

ilkins, C., Pledger, M., Bhatta, K., & Casswell, S. (2004). Patterns ofamphetamine use in New Zealand: Findings from the 2001 NationalDrug Survey. New Zealand Medical Journal, 117(1190), 1–12.

ilkins, C., Reilly, J., Pledger, M., & Casswell, S. (2005). Estimating thedollar value of the illicit market for cannabis in New Zealand. Drug andAlcohol Review, 24(3), 227–234.

ilkins, C., & Sweetsur, P. (In press). Exploring the structure of the illegalmarket for cannabis. De Economist.

lems: Auckland survey data 1990–1992. Auckland: Alcohol & PublicHealth Research Unit.

ska, R. (1990). New Zealand green: The story of marijuana in New Zealand.Auckland: David Bateman.