2. estimating the value of illicit drug markets
TRANSCRIPT
123
The illicit drug industry operates outside the law. Its‘companies’ are not listed on any stock exchange, theyare not valued by any private accounting firm, and thedynamics of the drug industry are not regularly poredover by analysts, economists and forecasters. Yet theoverall size of the illicit drug industry is known to belarge and, therefore, a potential threat to a number ofeconomies in terms of the financial power generated.The funds generated can be used to intimidate (includ-ing by means of violence) or corrupt government offi-cials or, in some cases, political systems as a whole, aswell as to crowd out licit economic activities, thus jeop-ardizing a country’s future. If the illicit drug industry isto be successfully controlled, there is a need to come toan understanding of the likely amount of moneyinvolved and where these funds are being generated.
The utility of undertaking such an exercise is clear fromboth a policy and a trend analysis perspective. Knowl-edge of the market’s value is indisputably useful forpolicy formulation. An informed estimate of the size ofthe drug markets also will enable analysts to look at therelative importance of the size of the markets vis a vislocal economies and it will facilitate the comparison ofthe importance of different drugs in economic terms. Inaddition, knowledge of the size of these markets willgive us an element for comparison with other illicit mar-kets – an important issue when it comes to allocatingscarce economic resources to fight various illegal activi-ties.
The obscurity of the global illicit drug market makes theexercise of estimating its size difficult. This is notbecause the drug market does not behave like mostothers in terms of supply and demand - there is a grow-ing acceptance that it does. It is rather because the mostbasic inputs that are needed for such an estimation –data on production, prices, quantities exported,imported and consumed – are themselves often esti-mates and are frequently based on deficient data.
A number of attempts to measure the size of the illicitdrug industry have been made in the past, including bythe Financial Action Task Force and the UnitedNations. The Financial Action Task Force (FATF) esti-mated that in the late 1980s, sales of cocaine, heroinand cannabis amounted to approximately US$124 bil-lion per year in the United States and Europe1, of thistotal some US$85 billion or 70% was considered tohave been available for money laundering and invest-ment.2 Taking inflation into account, the FATF estimateof the size of the illicit drug industry for the late 1980swould be equivalent today to some US$200 billion(expressed in 2005 US dollars).3
Other United Nations estimates, based on cash flowsfrom international banking and capital account statis-tics, suggested that up to US$300 billion per year couldhave been available for money laundering in the late1980s.4
2. Estimating the value of illicit drug markets
2.1 Background
1 The FATF estimated the retail drug sales turnover during the 1980s at $108 billion in the United States and $16.3 billion in Europe, i.e. a total of$124.3 billion. The largest amount was estimated for cannabis ($74.7 billion), followed by cocaine ($28.8 billion), and heroin ($12 billion).Organisation for Economic Co-operation and Development, FATF Working Group on Statistical and Methods, Narcotics Money Laundering -Assessment of Scale of the Problem, 1989, Financial Action Task Force on Money Laundering, report, February 7, 1990.
2 Organisation for Economic Co-operation and Development, Financial Action Task Force on Money Laundering, Paris 1990, p. 6, quoted inUNDCP, Economic and Social Consequences of Drug Abuse and Illicit Trafficking, Vienna 1997, p. 27.
3 The $124 billion referred to estimates for 1988; based on the US Consumer Price Index, this amount would be equivalent to $201 billion in 2005(http://data.bls.gov/cgi-bin/cpicalc.pl).
4 This figure was, however, qualified as "suspect" (probably too high) by the Intergovernmental Expert Group to Study the Economic and Socialconsequences of Illicit Traffic in Drugs (see E/CN.7/1991/25, p. 25).
Based on 1995 drug production estimates, UNDCParrived at a global estimate of $360 billion, with a rangefrom $85 billion to $1,000 billion.5 Given this broadrange and the high degree of uncertainty about thevalidity of some of the assumptions made, UNDCP’s1997 World Drug Report estimated a likely turnover ofthe illicit drug industry at around $400 billion.6 Thisfigure was questioned by some experts in the field aspossibly too high. However, no alternative calculationson the likely size of the global drug industry were pro-vided.
Another attempt as part of a broader exercise to estimatethe total value of money laundered annually (from crim-inal activities) was started by the Financial Action TaskForce in the late 1990s. It was decided to begin thisexercise by looking into the illicit drug market, giventhe fact that it was better studied than most other ille-gal markets. A number of expert meetings were con-vened, bringing together expertise from variousinternational, regional and national organisations.Given the extreme data limitations, existing weaknessesand contradictions of some of the data, the expertscould not agree on the most appropriate methodologi-cal approach. The basic question was whether a top-down approach (starting from global productionestimates) or a bottom-up approach (starting fromcountry estimates based on prevalence rate and esti-mates of expenditure per drug user which would thenhave to be aggregated) offered a better chance to arriveat a realistic estimate of the total value of the drugmarket. Recommendations were made to encouragecountries to improve their drug data collection systemsand to encourage them to undertake drug market esti-mates at the national level.7 Thus far only a limitednumber of country estimates on the value of the illicitdrug market are currently available. These alone wouldbe insufficient for generating global estimates.
Using the valuable lessons learned from these past exer-cises UNODC has continued work in this area. Theorganisation’s objective is to have a reliable idea of thesize of the value of the market, and to stimulate furtherresearch.
Three principles guided the production of these esti-mates: first only readily available data were used; second,the methodology and the model were kept straightfor-ward and the assumptions transparent; and third, it wasensured that by distilling the market down to its mostbasic economic rules, the model would be easilyupdateable. In addition, the methodology chosen triesto combine, as far as possible, the top-down with thebottom-up approach. While UNODC is fully awarethat the results will never have the same level of accu-racy as could be expected from a comparable analysis ofa licit market, and must be thus treated with caution,the new valuation methodology provides the best possi-ble results, based on existing knowledge and data pro-vided by Member States to UNODC. The methodologyused and the results will be discussed in this Chapter.
2.1.1. The model
A global input-output model was developed building onexisting UNODC data collection systems, thus makingit replicable as well as allowing for expert opinion to betaken into account. The model used data published inlast year’s World Drug Report (2002/2003 data), sup-plemented –where data was missing - with dataobtained from Member States over the last year. Themodel was used for the analysis of the main drug mar-kets: opiates, cocaine, cannabis herb, cannabis resin,amphetamines and ecstasy.
Models work on assumptions, but these are madeexplicit so that they can be improved over time. Themain assumption of this model is that what is being pro-duced, less seizures and less losses, is available for con-sumption and is consumed. The amounts available forconsumption in each sub-region are multiplied with theaverage purity adjusted prices of the respective sub-regions to arrive at the sub-regional market values.These values are then added up to arrive at the totalmarket value. The model looks at the market sub-regionally. Data inconsistencies are detected in largepart because the model looks at the market both fromthe supply side and the demand side.
124
World Drug Report 2005 Volume I. Analysis
5 This estimate amounted to US$117 billion for cocaine, US$107 billion for opiates, US$62 billion for cannabis herb, US$13 billion for cannabisresin and US$60 billion for synthetic drugs. UNDCP, Economic and Social Consequences of Drug Abuse and Illicit Trafficking, UNDCP TechnicalSeries, p. 51.
6 United Nations International Drug Control Programme, World Drug Report, (Oxford University Press 1997), p.124.7 Financial Action Task Force, Report of the FATF Ad Hoc Group on Estimating the Magnitude of Money Laundering on Assessing Alternative Methodologies
for Estimating Revenues from Illicit Drugs, FATF-XI/PLEN/45 (2000).
The model starts with global drug production per sub-region and allocates it, less local consumption andpurity adjusted seizures made in the source countries,either according to seizures made in the different sub-regions (for potential ‘supply constrained regions’)8 oraccording to the ‘number of drug users multiplied byper capita drug consumption ratios’ (for potential‘demand constrained regions’).9 The model thus allowsfor different per capita consumption rates for differentsub-regions.10 From the allocated amounts per sub-region, the model deducts purity adjusted seizures andlosses (set at 10%) and then multiplies the remainingamounts that are available for consumption in each sub-region with the purity adjusted prices. It uses the purityadjusted wholesale prices to estimate the wholesale valueand the purity adjusted retail prices to calculate the finalretail value. Adding up these sub-regional values givesthe estimates at the global level.
The drug prices and drug purities of each country areweighted by the number of drug users in that countryin order to calculate the regional average. The ‘typical’drug prices and drug purities, provided by MemberStates were used. If no such typical prices or purity datawere provided, the mid-point estimates of minimumand maximum values were used instead. If for any indi-
vidual country no price or purity data is available, themodel uses the unweighted sub-regional averages as aproxy.
The model allows for a number of calibrations, based onexpert knowledge, to adjust, as far as possible, themodel’s assumptions to reality. Thus, it is possible toadjust for the likely effectiveness of law enforcementbodies in different regions. This affects the calculatedinterception rates and thus the allocation of the drugs tothe various regions. For instance, enforcement effective-ness can be assumed to be higher in North America thanin Africa, thus lower drug seizures in Africa can still gohand in hand with substantial levels of drug consump-tion. The model also has a built-in distribution mecha-nism that assumes that drugs produced in a region are,first of all, used to supply local demand before beingexported. The subsequent distribution of drugs to thedestination markets is then a function of geographicalproximity (i.e. the closer any specific drug producingregion is to another region, the higher the likely pro-portion of total exports going to such a region). Again,these model assumptions can be altered based on expertknowledge. For instance, special ethnic links and estab-lished drug trafficking routes are known to play, in somecases, a far more important role than mere geographic
125
2. Estimating the value of illicit drug markets
8 The main hypothesis for this approach has been that seizures are positively correlated with the size of a drug market. In addition, seizures are, ofcourse, also a function of the effectiveness of law enforcement bodies. This is taken into account by 'rating' the effectiveness of law enforcement ofsome regions versus others. In regions with a weak enforcement infrastructure even small seizures may indicate a sizeable drug market while theopposite can be true in regions with highly effective law enforcement bodies.
9 As a default value, the model assumes that all regions are 'supply constrained', i.e. people would use as much of a drug as they could secure. Fordrug producing and main transit countries, such an assumption is however, not very realistic. Such regions are subsequently set to become 'demandconstrained'. This requires an assumption of the likely per drug capita consumption. If no additional information was available, it was usuallyassumed that average consumption of such regions would be close to the global average, estimated as amounts of drugs available (derived fromproduction estimates less seizures and less losses), divided by the total number of drug users. In order to make the results of the two approaches('supply constrained' and 'demand constrained') comparable, purity adjusted seizures are then added to arrive at the allocated amounts.
10 This is important because, information on per capita consumption rates is still very limited. It is hoped that this will improve over the next fewyears, which should strengthen the 'bottom-up' approach in the model.
Table 1. Drug related data routinely collected by UNODC
Production Trafficking Consumption
Cultivation Drug seizures Annual prevalence
Yields Origin of drugs Trends in drug consumption
Manufacture Transit of drugs
Laboratory seizures Destination of drugs
Prices Prices
Purities PuritiesLargely missing: Information on quantitiesof drugs consumed
proximity. For some specific cases, theoretical traffickinglinks could be completely ruled out (such as exports ofNorth American cannabis herb to Africa or South Asia;differences in price levels would mean that traffickersinvolved in such operations would simply lose money).
One advantage of such a systematic approach withbuilt-in cross-checks is to make explicit to the analyst allpotential data inconsistencies. This systematic analysisof existing data is particularly important given well-known data weaknesses. It enables the identification ofdata that needs to be re-checked and/or indicates newareas of research. Moreover, the model helps to incor-porate new estimates, research findings and intelligencewhenever they should become available.
Key to the outcome of the model are, of course, theinputs used. The main inputs into the model are drugproduction estimates, seizures, drug price data (farm-gate, wholesale and retail prices), drug purity data(wholesale and retail level), estimated number of drugusers and estimates of per capita drug consumption.Most of these data are routinely collected by UNODC.
Seizure, price and purity data are collected annuallyfrom countries through UNODC’s Annual ReportsQuestionnaires and are supplemented by informationcollected from other international or regional bodies(such as INCB, Interpol, WCO, Europol, OAS etc.).Seizure data is thus the most complete data set. In addi-tion, countries report typical drug trafficking patterns toUNODC, including the most typical trafficking routes.This information entered the model in the form of ex-post calibrations.
Prevalence data is basically collected through UNODC’sAnnual Reports Questionnaires. However, this data setis not as complete as seizure data as many governmentsstill do not have appropriate monitoring systems inplace. Thus, UNODC developed over the years a spe-cial methodology to estimate annual prevalence datafrom partially available data sets (e.g. extrapolatingannual prevalence data from life-time prevalence data,from student surveys or from treatment data usingannual prevalence data from other countries in theregion as benchmark figures).
Largely missing – and not part of any routine data col-lection – is information on the per capita consumption
of drugs by drug users. The lack of this information hasbeen one of the biggest constraints to market analysis onthe demand side and thus a main stumbling block toalmost every attempt to gain greater insight into themarket from the consumption side. There is almost nosystematic and comparable data on the quantities ofindividual substances consumed per users in differentregions. The information which does exist is limited andoften contradictory. More research efforts in this areaare clearly needed.
UNODC’s strongest data sets are is for the cultivationof coca and opium poppy. Through its InternationalCrop Monitoring Programme, UNODC, in coopera-tion with the respective national governments, usesground and satellite based survey methods to measurethe extent of cultivation (for coca, opium poppy andcannabis resin11). In combination with yield surveys,drug production estimates can thus confidently beestablished.
Production estimates on cannabis herb have been takenfrom replies to UNODC’s Annual Reports Question-naire as well as from other Government reports. Theproblem here is that most of these estimates are notbased on rigorous scientific studies. In addition, formany countries the information is missing altogether. Anumber of countries in Africa, Asia and Europe, forinstance, have been frequently identified by other coun-tries as important source countries, but they did notprovide any cannabis production estimates toUNODC. In such cases, it was assumed that the coun-tries cover their local demand and use a certain per-centage for export purposes. The total cannabis herbproduction estimate thus increased from otherwise35,000 mt to 42,000 mt for the year 2003. However, asimilar amount (5,000 mt) was subsequently deductedagain as ‘extraordinary losses’ from one sub-region(North America) as available production estimates inthis sub-region, reported to UNODC by variousnational authorities, seemed to exceed realistic con-sumption estimates.12
In the case of ATS indirect estimation methods wereused, as described in other parts of this report, based onATS consumption, ATS seizures and ATS precursorseizures.
126
World Drug Report 2005 Volume I. Analysis
11 For Morocco. 12 This had to be done as a possible alternative explanation - exports - does not apply in this case; no information is available to UNODC that cannabis
herb produced in North America is being exported to any other region in significant quantities.
2.2 Results
Based on the inputs and the calculations explainedabove, the value of the global illicit drug market for theyear 2003 was estimated at US$13 bn at the productionlevel, at $94 bn at the wholesale level (taking seizuresinto account), and at US$322bn based on retail pricesand taking seizures and other losses into account. Thisindicates that despite seizures and losses, the value of thedrugs increase substantially as they move from producerto consumer.
The largest market, according to these estimates, iscannabis herb (with a retail market size of $113 bn), fol-lowed by cocaine (US$71 bn), the opiates (US$65bn)and cannabis resin (US$29 bn). The ATS marketstogether (methamphetamine, amphetamine andecstasy) amount to US$44 bn. The valuation does nottake into account the value of other drugs.
While UNODC is reasonably confident with its esti-mations on opiates, cocaine and the ATS, the degree ofcertainty is far lower for cannabis, notably for cannabisherb, as information for production and consumption
of this substance is highly contradictory. If better infor-mation becomes available, a major revision cannot beruled out.
If compared to global licit exports (US$7,503 bn in2003)13 or compared to global GDP (US$35,765 bn in2003)14 the estimated size the global illicit drug marketmay not appear to be particularly high (0.9% of globalGDP at retail level or 1.3% of global exports measuresat wholesale level).15
Nonetheless, the size of the global illicit drug market issubstantial. The value, measured at retail prices, ishigher than the GDP of 88% of the countries in theworld (163 out of 184 for which the World Bank hasGDP data) and equivalent to about three quarters ofSub-Saharan Africa’s combined GDP (US$439 bn in2003). The sale of drugs, measured at wholesale prices,was equivalent to 12% of global export of chemicals(US$794 bn), 14% of global agricultural exports(US$674 bn) and exceeded global exports of ores andother minerals (US$79 bn) in 2003. Such sales of drugswere also higher than the combined total licit agricul-tural exports from Latin America (US$75 bn) and theMiddle East (US$10 bn) in 2003.16
127
2. Estimating the value of illicit drug markets
Fig. 1: Size of the global illicit drug market in 2003
12.8
94.0
321.6
0
50
100
150
200
250
300
350
Producer level Wholesalelevel
Retail level
bill
ion
US&
Fig. 2: Size of the global illicit drug market in 2003by substances
1.2 0.58.8
0.7 0.6 1.0
20.6 18.829.7
10.4 6.8 7.7
64.8 70.5
113.1
28.8 28.316.1
0
20
40
60
80
100
120
Opi
ates
Coc
aine
Can
nabi
s he
rb
Can
nabi
s re
sin
Am
phet
amin
es
Ecst
asy
bill
ion
US$
producer-level wholesale-level retail-level
Source: Illicit Drug Market Estimation Model, based on UNODC,Annual Reports Questionnaire Data, Govt. reports and UNODCproduction and consumption estimates.
Source: Illicit Drug Market Estimation Model, based on UNODC,Annual Reports Questionnaire Data, Govt. reports and UNODCproduction and consumption estimates.
13 World Trade Organisation, International Trade Statistics 2004, p. 19. 14 World Bank, World Development Indicators 2005 Report, http://www.worldbank.org/data/wdi2005/. 15 The comparison with wholesale prices is more appropriate as export prices are usually closer to wholesale than to retail prices. 16 World Bank, World Development Indicators database, April 2005.
The relative importance of the size of illicit drugsmarket becomes more pronounced if compared to theexports of individual products. Exports of wine(US$17.4 bn) and beer (US$6.7 bn) are equivalent tojust a quarter of the wholesale value of illicit drugs.17
Coffee, one of the world’s most ubiquitous beverages,used to generate some US$15bn in export revenue inthe 1990s,18 falling to less than US$6 bn in 2003.19
Global exports of tobacco products (including ciga-rettes) are equivalent to about a fifth of the globalwholesale value of illicit drugs. Wheat, a staple of a largeportion of the global population, generated US$16bn inexport revenue in 2003. All cereal exports togetherresulted in export revenue of $41 bn,20 less than half thewholesale value of the global illicit drugs market.
In terms of the regional distribution, the world’s largestdrug market – in economic terms – was identified to beNorth America,21 accounting for 44% of the world’stotal drug sales at the retail level, followed by Europe(33%). Within Europe, West and Central Europe22 isthe dominant drug market (27% of total). The nextlargest retail drug markets are Asia (11%) followed byOceania (5%) and Africa (4%).
128
World Drug Report 2005 Volume I. Analysis
17 Food and Agriculture Organization of the United Nations, FAOSTAT, http://faostat.fao.org/faostat/collections?version=ext&hasbulk=0. 18 Aksoy, M.A. and Beghin, J.C. eds., Global Agriculture and Trade in Developing Countries, World Bank, Washington DC, 2005, p 297 (evaluated
at 1997-98 average prices and volumes.)19 International Coffee Organization, Annual Review 2003/04, p. 6.20 Food and Agriculture Organization of the United Nations, FAOSTAT, http://faostat.fao.org/faostat/collections?version=ext&hasbulk=021 North America is defined to include: Canada, Mexico and the United States of America. 22 West & Central Europe includes the 25 EU countries, the EFTA countries and small countries such as Monaco, Andorra and San Marino.
Fig. 3. Value of illicit drugs at wholesale level (in billion US$) compared to the export values of selectedagricultural commodities in 2003
$2.6
$5.7
$6.7
$9.9
$16.0
$17.3
$21.6
$40.7
$52.5
$94.0
0 10 20 30 40 50 60 70 80 90 100
Tea
Coffee
Beer
Chocolate products
Wheat
Wine
Tobacco products
All cereals
Meat
Illicit drugs*
billion US$
* illicit drugs measured at the wholesale level, used as a proxy for the export price
Sources: UNODC, Illicit Drug Market Estimation Model, FAO, FAOSTAT and ICO, Annual Review 2003/04.
Fig. 4: Regional breakdown of the global illicit drugmarket in billion US $ (N = $322 bn)
* Including Caribbean and Central America.
Sources: UNODC, Illicit Drug Market Estimation Model.
$142North
America44%
$106Europe33%
$35Asia11%
4%
$16Oceania
5%
$9South-
America*3%
On a per capita basis, the results of the model suggestthat the highest expenditures on drugs per year(expressed in current US-dollars) are found in the Ocea-nia region, followed by North America and West andCentral Europe. Below average expenditures on drugsare seen in Asia, Africa and South America. This ismainly the result of far lower drug prices in these coun-tries. Global expenditures on drugs amount to aboutUS$50 per person per year.
Expressed as a percentage of GDP, drug sales (at theretail level) seem to be most important in the Oceaniaregion, followed by East and South-East Europe23 andAfrica. The lowest importance of retail sales of drugs ascompared to the size of the overall economy is in Asia. Though only about a third of the world’s drug users arelocated in OECD countries, about three quarters of theworld’s retail drug market – in economic terms - isfound in the industrialized world (some US$245 bil-
129
2. Estimating the value of illicit drug markets
Fig. 5: Per capita expenditure on drugs (in current US$)
* Including Caribbean and Central America.
Sources: UNODC, Illicit Drug Market Estimation Model, United Nations, Population Division, Department of Economic and Social Affairs,World Population Prospects: The 2002 Revision (POP/DB/WPP/Rev.2002/4/F1).
502
331
186
58 5122 15 9
$0
$100
$200
$300
$400
$500
Oceania NorthAmerica
West andCentralEurope
East & South-East Europe
GLOBALAVERAGE
South-America*
Africa Asia
US$
Fig. 6: Expenditure on drugs in % of GDP (2003)
* Including Caribbean and Central America.
Sources: UNODC, Illicit Drug Market Estimation Model, World Bank,World Development Indicators 2005.
2.6%
2.2%2.1%
1.1%0.9% 0.9% 0.8%
0.4%
0.0%
0.5%
1.0%
1.5%
2.0%
2.5%
3.0%
Oceania East &South-East
Europe
Africa NorthAmerica
GLOBALAVERAGE
South-America*
West andCentralEurope
Asia
23 East Europe is defined to include the European countries of the C.I.S. (Russian Federation, Ukraine, Belarus, Rep. Of Moldavia); South-East Europeis defined to include Turkey and the (non EU-25) Balkan countries.
lion).
The calculations also show that, in absolute terms, thehighest profits are made between the wholesale and theretail level. Given the concentration of the retail marketsin the industrialized countries, the results suggest thatmost of the ‘value-added’ (gross profits) of the illicitdrug industry actually takes place in the industrializedworld. Of the total ‘value-added’ of the illicit drugindustry, 76% is generated in the industrialized coun-tries, 19% in developing countries and the rest in tran-sition countries. Total producer income is, on average,4% of the final retail value. For heroin and cocaine, it isclose to 1% of the final retail value.
2.3 Results of the individualmarkets
2.3.1 The cocaine trade - valued at overUS$70 bn per year (retail level)
Table 2 presents an analysis of levels of production insource countries and distribution to consumer coun-tries. For a variety of reasons, it was necessary to base thecocaine production estimates on a three-year average(2001-2003), resulting in a figure of 761 mt Thisamount, however, does not reach the consumers. After
deducting seizures in the source countries (Colombia,Peru, Bolivia), the amount available for shipment toconsumers was 653 mt Based on these data and averagecocaine base prices of US$808 per kilogram, the localincome from cocaine base production in South Americawas estimated at US$527 million.
In order to determine the destination of this produc-tion, the number of consumers in each region was firstconsidered. In addition, the cycle of the epidemic playsan important role. Countries or regions in an early stageof a drug epidemic can be expected to have many recre-ational users but only a limited number of hard-coreaddicts, while the opposite is true in more advanced sit-uations. Based on a limited number of studies on theper capita consumption patterns of drug users, it wasestimated that the average cocaine user in North Amer-ica consumes 44 grams of pure cocaine per year whilethe average cocaine user in Western and Central Europeand in South America consumes some 35 grams peryear.
Taking the information on the estimated number ofcocaine users and the estimated number of per capitaconsumption rates into account, the model calculatesthe amount of drugs consumed in these sub-regions.Factoring in the purity adjusted seizures made in thesesub-regions, the model arrives at the likely amounts ofcocaine being imported. Based on these calculations, itwould appear that the bulk of the cocaine produced inthe Andean region goes to North America (352 mt),with lesser amounts being received in West and CentralEurope (134 mt), the Caribbean (17 mt) and CentralAmerica (16 mt). About 101 mtare retained in SouthAmerica for domestic consumption. Between them,these regions account for the bulk of the cocaine traf-ficked (96%). Deducting purity adjusted seizures andlosses (set at 10%), the model calculates the amountsactually available for consumption24 in North America(280 mt of pure cocaine), West and Central Europe(107 mt), and South America (69 mt). For otherregions, see Table 2.
Multiplying these amounts with the purity adjustedaverage cocaine prices (i.e. prices calculated for 100%pure cocaine) gives a wholesale value for the region.Adding up the wholesale-values from all regions gives atotal market value of US$18.8 bn, including the large
130
World Drug Report 2005 Volume I. Analysis
Fig. 7: Distribution of the ' value added' of the illicitdrug industry (N = $322 bn)
Sources: UNODC, Illicit Drug Market Estimation Model, WorldBank.
Others5%
Industrialized countries
76%
Developing countries
19%
24 The model does not differentiate between seizures made at the wholesale level and those made at the retail level. The implicit assumption here isthat most of the seizures and losses take place in the shipment of cocaine from the Andean region to the destination countries; seizures at a laterstage, i.e. at the retail distribution level, are considered to be rather small. Such seizures are already included in the overall seizures figures at thewholesale level.
131
2. Estimating the value of illicit drug markets
Tab
le 2
. Co
cain
e: P
rod
uct
ion
an
d d
istr
ibu
tio
n f
rom
so
urc
eco
un
trie
s to
des
tin
atio
nco
un
trie
s
East
Afr
ica
00
083
09
00
00
00
00
00
00
00
00
0N
orth
Afr
ica
00
01,
194
130
00
00
00
00
00
00
00
00
Sout
hern
Afr
ica
00
05,
456
298
00
00
00
00
00
00
00
00
0W
est
and
Cen
tral
Afr
ica
00
07,
054
260
00
00
00
00
00
00
00
00
Car
ibbe
an0
00
17,1
1011
,350
00
00
00
00
00
00
00
00
0C
entr
al A
mer
ica
00
015
,599
8,88
00
00
00
00
00
00
00
00
00
Nor
th A
mer
ica
00
035
2,76
672
,776
00
00
00
00
00
00
00
00
0So
uth
Am
eric
a76
1,00
010
8,38
165
2,61
910
0,63
031
,796
830
1,19
45,
456
7,05
417
,110
15,5
9935
2,76
610
0,63
04
2,99
91,
403
755
3,19
013
3,80
62,
373
7,45
165
2,61
9C
entr
al A
sia
& T
rans
cauc
asus
00
04
00
00
00
00
00
00
00
00
00
East
and
Sou
th-E
ast
Asi
a0
00
2,99
944
00
00
00
00
00
00
00
00
0N
ear
& M
iddl
e Ea
st /S
W A
sia
00
01,
403
102
00
00
00
00
00
00
00
00
0So
uth
Asi
a0
00
755
110
00
00
00
00
00
00
00
00
E ast
ern
Euro
pe0
00
3,19
047
00
00
00
00
00
00
00
00
0W
este
rn &
Cen
tral
Eur
ope
00
013
3,80
626
,444
00
00
00
00
00
00
00
00
0So
uth
East
Eur
ope
00
02,
373
430
00
00
00
00
00
00
00
00
Oce
ania
00
07,
451
580
00
00
00
00
00
00
00
00
All
Co
un
trie
s76
1,00
010
8,38
165
2,61
965
2,61
915
1,89
883
01,
194
5,45
67,
054
17,1
1015
,599
352,
766
100,
630
42,
999
1,40
375
53,
190
133,
806
2,37
37,
451
652,
619
Tab
le 3
. Co
cain
e: S
up
ply
an
d d
eman
d in
des
tin
atio
nco
un
trie
s
00
00
00
076
1,00
00
00
00
00
076
1,00
00
00
00
00
108,
381
00
00
00
00
108,
381
00
00
00
065
2,61
90
00
00
00
065
2,61
90
00
00
00
808
00
00
00
00
808
00
00
00
052
70
00
00
00
052
7
830
1,19
45,
456
7,05
417
,110
15,5
9935
2,76
610
0,63
04
2,99
91,
403
755
3,19
013
3,80
62,
373
7,45
165
2,61
99
1329
826
11,3
508,
880
72,7
7631
,796
044
102
1147
26,4
4443
5815
1,89
882
01,
181
5,15
87,
028
5,76
06,
719
279,
990
68,8
344
2,95
51,
301
744
3,14
410
7,36
12,
329
7,39
350
0,72
113
315
025
3311
1332
460
4365
7010
164
9214
038
109
177
128
229
6488
9,08
924
70
128
8552
319
6,83
121
41,
037
18,7
97
6566
265
540
164
191
6,38
21,
961
594
3026
108
3,05
917
221
113
,339
738
1,06
34,
642
6,32
55,
760
6,71
927
9,99
068
,834
42,
660
1,17
066
92,
829
107,
361
2,09
67,
393
498,
255
1116
1812
3535
4435
128
3926
2635
1235
3734
852
047
6049
2015
646
183
253
121
227
124
161
208
380
141
257
553
220
378
281
132
43,5
703,
172
167
314
115
235
117
,328
437
2,80
970
,455
Oceania
All Countries
South Asia
East Europe
West & Central Europe
SouthEast Europe
South America
Central Asia & Transcaucasus
East & SouthEast Asia
Near & Middle East /SW Asia
West & Central Africa
Caribbean
Central America
North AmericaA
vera
ge R
etai
l Pric
e U
S$ /g
m
Ret
aile
r In
com
e (U
S$m
ill)
Dem
and
:Es
timat
ed U
ser
Popu
latio
n (T
hous
ands
)Es
timat
ed A
ctua
l Con
sum
ptio
n pe
r ye
ar (K
g C
ocai
ne E
quiv
)Im
plie
d C
onsu
mpt
ion
per
user
(gm
s C
ocai
ne E
quiv
)
Tota
l Ava
ilabl
e fo
r C
onsu
mpt
ion
(Kg
Coc
aine
Equ
iv)
Who
lesa
le p
rice
at D
estin
atio
n U
S$ /g
mW
ho
lesa
ler
Inco
me
(US$
mill
)
Pro
du
ctio
n:
Tota
l Pro
duct
ion
in S
ourc
e C
ount
ry (K
g C
ocai
ne E
quiv
)To
tal S
eize
d/Lo
st in
Sou
rce
Cou
ntry
(Kg
Coc
aine
Equ
iv)
Tota
l Ava
ilabl
e fo
r Sa
le (K
g C
ocai
ne E
quiv
)Fa
rmga
te P
rice
at O
rigin
(US$
/Kg
Coc
aine
Equ
iv)
Pro
du
cer
Inco
me
(US$
mill
)
Sup
ply
:To
tal I
nten
ded
for
Con
sum
ptio
n (K
g C
ocai
ne E
quiv
)
Central Asia & Transcaucasus
North Africa
Near & Middle East /SW Asia
Tota
l Sei
zed/
Lost
at
Des
tinat
ion
(Kg
Coc
aine
Equ
iv)
South America
East & SouthEast Asia
Yo
u n
eed
to
Re-
Cal
ibra
te!
==
==
==
=>
>>
South Africa
West & Central Africa
Caribbean
Tran
sfer
red
to
M
arke
ts (
Kg
C
oca
ine
Equ
iv)
Tota
l Sei
zed
/ lo
st in
Tra
nsi
t (K
g C
oca
ine
Equ
iv)
South Asia
East Europe
West & Central Europe
SouthEast Europe
Oceania
Pro
du
cer
Reg
ion
sTo
tal P
rod
uct
ion
in
So
urc
e C
ou
ntr
y(K
g C
oca
ine
Equ
iv)
Tota
l Sei
zed
/Lo
st
in S
ou
rce
Co
un
try
(Kg
C
oca
ine
Equ
iv)
Tota
l Ava
ilab
le
for
Sale
(K
g
Co
cain
e Eq
uiv
)
Reg
ion
s
Co
nsu
mer
Reg
ion
s
All Countries
Central America
North America
East Africa East Africa
North Africa
South Africa
markets of North America (US$ 9.1 bn), West andCentral Europe (US$6.8 bn), and South America(US$0.3 bn). The South American figure reflects, how-ever, only the gross income of wholesalers supplying thedomestic market. The total wholesale income in SouthAmerica, where much of the cocaine is not destined forlocal consumption but for exports, is much larger. Thetotal gross income of wholesalers in South Americawould be equivalent to about US$2.6 bn.
Retail values were calculated by multiplying theamounts available for consumption by the purityadjusted retail prices, resulting in remarkably high fig-ures in North America (US$44 bn), West and CentralEurope (US$17 bn), and South America (US$3 bn).The global retail market for cocaine adds up to US$70.5bn. The results of the model suggest that North Amer-ica (62%), followed by Europe (26%) are, in economicterms, the largest cocaine markets.
2.3.2 The opiates trade - valued atUS$65 bn per year(retail level)
Global production of opiates is estimated at 476.5 mt(in heroin equivalents) in 2003, most of which is pro-duced in the Near and Middle East/South-West Asiasub-region (365 mt), which includes Afghanistan. Incontrast to cocaine, however, opiate production takesplace in more than one region. The second most impor-tant production region is East and South-East Asia (94mt), mainly Myanmar and Laos. Other productionareas of importance are in North America (reflectingproduction in Mexico) and in South America (mainlyreflecting production in Colombia).
For each of these production areas, distinct distributionpatterns can be identified. Most of the opiates producedin the Near and Middle East/South-West Asia sub-region are either consumed locally (more than a fifth) orexported to Europe (about half ). The rest goes to otherregions. In the case of East and South-East Asia, twothirds are for consumption within the region. All of theopiates produced in North America remain within thisregion (mainly destined for the US market) and opiatesproduced in South America are for the local market andfor the market in North America.
According to the results of the model, close to 100 mtof heroin are destined for the markets of West and Cen-tral Europe, about 90 mt for East Europe and 10 mt forSouth-East Europe. Deducting seizures and losses(assumed to amount to 10%), 84 mt are actually avail-able for consumption in West and Central Europe,equivalent to 58 grams per heroin user per year. This ishigher than the average at the global level (28 grams).However, one internal study, commissioned byUNODC, found that average heroin consumptionamong heroin users in the three months prior to under-going drug treatment was close to 68 grams of pureheroin per year.25 According to reports from the Swissheroin maintenance program, which covers a group ofhard-core heroin addicts, 135 grams per addict are con-sumed annually.26 Against this background, a per capitaconsumption of 58 grams of heroin per year in Westand Central Europe appears to be feasible.
132
World Drug Report 2005 Volume I. Analysis
Fig. 8: Regional distribution of cocaine retail sales in2003 in billion US$ (N = US$70.5 bn)
* including Caribbean and Central AmericaSources: UNODC, Illicit Drug Market Estimation Model.
$18.1Europe26%
$3.6South-
America*5%
$1.4Africa2%
$2.8Oceania
4%
$1.0Asia1%
$43.6North
America62%
25 These results were based on the results of a UK study on people entering treatment in 1997 (Gossop et al., "National Treatment Outcome ResearchStudy in the United Kingdom", Psychol. Addictive Behaviours, 1997). The study showed an average consumption of 0.6 grams per day, and aconsumption of, on average, 22 days per month. Average consumption per month was thus 14.9 grams of heroin (at street purity), which amountsto 179 grams per year. Applying the average purity of around 38 % reported by forensic laboratories in the UK in 1997 (The Forensic ScienceService, "Drug Abuse Trends", various issues), average annual consumption would be 68 grams of pure heroin per problem drug user.
26 Institut für Suchtforschung, Universität Zürich, Institut für Sozial und Präventivmedizin, Versuche für eine ärztliche Verschreibung vonBetäubungsmitteln, Synthesebericht, (Ambros Uchtenhagen), June 1997.
133
2. Estimating the value of illicit drug markets
Yo
u n
eed
to
Re-
Cal
ibra
te!
==
==
==
=>
>>
Tab
le 4
. Op
iate
s: P
rod
uct
ion
an
d d
istr
ibu
tio
n f
rom
so
urc
eco
un
trie
s to
des
tin
atio
nco
un
trie
s
East
Afr
ica
00
0 2,
293
47
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0N
orth
Afr
ica
714
0 71
4 3,
410
35
0 71
4 0
0 0
0 0
0 0
0 0
0 0
0 0
0 So
uthe
rn A
fric
a 0
0 0
1,09
97
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0W
est
and
Cen
tral
Afr
ica
00
0 9,
954
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0C
arib
bean
00
0 36
40
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
Cen
tral
Am
eric
a 0
0 0
783
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0N
orth
Am
eric
a 8,
400
209
8,19
1 28
,735
6,62
1 0
0 0
0 0
0 8,
191
0 0
0 0
0 0
0 0
0 8,
191
Sout
h A
mer
ica
5,26
8 44
0 4,
829
3,25
539
9 0
0 0
0 0
0 1,
574
3,25
5 0
0 0
0 0
0 0
0 4,
829
C. A
sia
& T
rans
cauc
asus
77
00
770
12,0
762,
748
0 0
0 0
0 0
0 0
770
0 0
0 0
0 0
0 77
0Ea
st a
nd S
outh
-Eas
t A
sia
94,0
50
582
93,4
68
59,9
286,
885
240
267
119
1,07
6 39
85
2,
050
0 1,
222
59,9
28
0 6,
663
9,48
3 10
,508
1,
028
762
93,4
68N
r &
M.E
ast
/SW
Asi
a 36
5,15
0 9,
551
355,
599
78,3
5227
,723
2,
053
2,42
9 98
0 8,
878
325
698
16,9
20
0 10
,084
0
78,3
52
55,0
03
78,2
83
86,7
40
9,18
0 5,
673
355,
599
Sout
h A
sia
00
0 61
,666
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0Ea
ster
n Eu
rope
2,
147
0 2,
147
89,9
1392
3 0
0 0
0 0
0 0
0 0
0 0
0 2,
147
0 0
0 2,
147
Wes
tern
& C
entr
al E
urop
e 0
0 0
97,2
483,
994
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0So
uth
East
Eur
ope
00
0 10
,208
4,71
3 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
Oce
ania
00
0 6,
436
198
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0A
ll C
ou
ntr
ies
476,
500
10,7
81
465,
719
465,
719
54,8
56
2,29
3 3,
410
1,09
9 9,
954
364
783
28,7
35
3,25
5 12
,076
59
,928
78
,352
61
,666
89
,913
97
,248
10
,208
6,
436
465,
719
Tab
le 5
. Op
iate
s: S
up
ply
an
d d
eman
d in
des
tin
atio
nco
un
trie
s
0 71
4 0
0 0
0 84
00
5268
77
0 94
050
3651
50
0 21
47
0 0
0 47
6500
0
0 0
0 0
0 20
9 44
0 0
582
9551
0
0 0
0 0
1078
1 0
714
0 0
0 0
8191
48
29
770
9346
8 35
5599
0
2147
0
0 0
4657
19
0 0
0 0
0 0
0 0
0 13
39
2830
0
0 0
0 0
2520
0 2
0 0
0 0
21
12
2 12
5 10
06
0 5
0 0
0 11
73
2293
34
10
1099
99
54
364
783
2873
5 32
55
1207
6 59
928
7835
2 61
666
8991
3 97
248
1020
8 64
36
4657
19
47
35
7 51
80
18
0 66
21
399
2748
68
85
2772
3 25
2 92
3 39
94
4713
19
8 54
856
2246
33
75
1092
99
03
285
603
2211
3 28
56
9328
53
042
5062
9 61
413
8899
0 93
254
5496
62
37
4108
63
28
42
28
15
38
73
98
25
21
38
7 71
36
73
26
14
0 50
63
142
30
153
11
44
2175
71
19
4 20
22
379
4355
31
98
6779
14
5 87
4 20
635
84
108
85
524
21
24
1301
28
6 37
0 21
05
2009
31
02
2406
14
50
180
99
1415
4.3
2021
30
38
983
8913
28
5 60
3 22
113
2856
93
28
5304
2 50
629
6141
3 80
091
8392
9 54
96
5614
39
0353
24
.2
28.0
11
.5
17.0
13
.5
25.2
17
.0
10.0
25
.2
25.2
25
.2
19.8
33
.3
57.9
30
.6
56.5
27
.6
40
124
81
57
41
83
402
101
51
108
12
123
133
302
183
561
166
82
377
80
508
12
50
8886
28
8 48
0 57
25
596
7567
10
681
2534
0 10
06
3148
64
825
&
Tota
lA
vaila
ble
fo
r Sa
le(K
g H
ero
in
Equ
iv)
Tota
lSe
ized
/Lo
st in
So
urc
e C
ou
ntr
y (K
g H
ero
in
Equ
iv)
Tota
l Pro
du
ctio
n
in S
ou
rce
Co
un
try
(Kg
Her
oin
Eq
uiv
)
Pro
du
cer
Reg
ion
s
Co
nsu
mer
Reg
ion
s
Reg
ion
s
&
Tota
l Sei
zed
/ lo
st
in T
ran
sit
(Kg
Her
oin
Eq
uiv
)
Tran
sfer
red
to
M
arke
ts(K
g H
ero
in
Equ
iv)
Tota
l Sei
zed/
Lost
at
Des
tinat
ion
(Kg
Her
oin
Equi
v)
Tota
l Ava
ilabl
e fo
r C
onsu
mpt
ion
(Kg
Her
oin
Equi
v)
Pro
du
ctio
n:
Tota
l Pro
duct
ion
in S
ourc
e C
ount
ry (K
g H
eroi
n Eq
uiv)
To
tal S
eize
d/Lo
st in
Sou
rce
Cou
ntry
(Kg
Her
oin
Equi
v)
Tota
l Ava
ilabl
e fo
r Sa
le (K
g H
eroi
n Eq
uiv)
Fa
rmga
te P
rice
at O
rigin
(US$
/Kg
Her
oin
Equi
v)
Pro
du
cer
Inco
me
(US$
mill
)
Sup
ply
:To
tal I
nten
ded
for
Con
sum
ptio
n (K
g H
eroi
n Eq
uiv)
Who
lesa
le p
rice
at D
estin
atio
n U
S$ /g
m
Wh
ole
sale
r In
com
e (U
S$m
ill)
Dem
and
:Es
timat
ed U
ser
Popu
latio
n (T
hous
ands
) Es
timat
ed A
ctua
l Con
sum
ptio
n pe
r ye
ar (K
g H
eroi
n Eq
uiv)
Im
plie
d C
onsu
mpt
ion
per
user
(gm
s H
eroi
n Eq
uiv)
A
vera
ge R
etai
l Pric
e U
S$ /g
m
Ret
aile
r In
com
e (U
S$m
ill)
714
51
80
180
252
West & Central Europe
SouthEast Europe
Oceania
All Countries
East & SouthEast Asia
Near & Middle East /SW Asia
South Asia
East Europe
Central America
North America
South America
Central AsiaTranscaucasus
East Africa
North Africa
South Africa
West & Central Africa
All Countries
South Asia
South America
East & SouthEast Asia
Caribbean
Near & Middle East /SW Asia
Caribbean
East Europe
West & Central Europe
SouthEast Europe
Oceania
Central AsiaTranscaucasus
Central America
North America
East Africa
South Africa
West & Central Africa
North Africa
Multiplied with purity adjusted retail prices, weightedby the number of consumers in each country, the valueof the opiate market in West and Central Europe isestimated at US$25 bn. This is in line with previousUNODC estimates on the size of West Europe’s heroinmarket. The total retail market value of Europe’s opiatemarket (including those of East and South-East Europe)is estimated at US$37 bn. Europe accounts thus for56% of the global opiates retail market, valued ataround US$65 bn. The next largest retail market – ineconomic terms – is Asia, accounting for 22% of thetotal. The third largest market is North America, whichconsumes US$9 bn worth of the drug or 14% of thetotal.27
2.3.3. The cannabis trade, valued at overUS$140 bn per year (retail level)
There are two distinctly different cannabis markets:herbal cannabis, valued at US$113 bn and cannabisresin, valued at US$28 bn.
A great deal of effort has gone into modelling thecannabis markets. Nonetheless, a word of caution isneeded. The potential error between the estimationsshown below and the ‘true value’ of the cannabis marketcould be significant, much higher than the potentialerrors that could be expected from the calculation of theheroin or the cocaine market. This is due to apparentdata inconsistencies that make it difficult to reconcilesupply-based estimates with demand-based estimates.Nonetheless, as far as possible, such an attempt wasmade, based on the assumption that the ‘truth’ is some-where in the middle. The resulting estimates are the bestthat could be made, given the current level of informa-tion. This does not rule out the possibility that sub-stantial changes could occur (notably for cannabisherb), once better, scientifically generated informationbecomes available.
Valuation of cannabis herb
Production estimates were taken from Member States’replies to UNODC’s Annual Reports Questionnairesand official Government reports. Very strong year-to-year changes, particularly with regard to yields, suggestthat these estimates were based on limited informationand are not always reliable. One example of the dataweakness in this area is the lack of credible productionestimates for Africa. A number of African countries arefrequently reported as important source countries forcannabis herb imported into Europe, but these coun-tries do not provide production estimates to UNODC.Based strictly on available data, the model would predictthat North America should be exporting cannabis toAfrica, a trafficking route that does not, in fact, exist.The same applied to a significant number of countriesfrom other regions as well.
Against this background, a systematic review was under-taken of all the countries that, over the last decade, hadbeen reported by other countries as a source of cannabisor themselves reported the seizure of whole cannabisplants. The seizure of whole plants is indicative ofdomestic cultivation, because only a portion of the plantis used as a drug, and so whole plants are rarely traf-ficked across borders. For these countries, productionwas estimated to cover domestic demand, multiplying
134
World Drug Report 2005 Volume I. Analysis
27 Estimates for North America, however, highlighted a problem that still needs to be resolved in future. There are some apparent contradictions as tothe origin of heroin and its reported availability. According to US Government reports, heroin produced in Colombia and Mexico account for thebulk of illegal heroin imports in the USA. However, current production estimates available for these countries are not sufficient to cover the bulkof the North American demand for heroin.
Fig. 9: Regional distribution of opiate retail sales in2003 in billion US$ (N = US$64.8 bn)
* including Caribbean and Central AmericaSources: UNODC, Illicit Drug Market Estimation Model.
$14.4Asia22%
$8.9North
America14%
$1.0Africa2%$3.1
Oceania5%
$0.3South-
America*1%
$37.0Europe56%
the number of estimated cannabis users by the averageglobal cannabis herb consumption rate, derived fromthe initial calculations. For countries that were identi-fied as cannabis producing countries but were not iden-tified as major cannabis exporting countries, a certainpercentage of domestic demand was used to estimatelocal production. The percentages chosen depended onquantitative and qualitative information available fordifferent regions. For instance, based on estimates pro-vided by the authorities of some European countries,local cannabis herb production from European coun-tries, which (i) apparently had domestic production but(ii) had not provided a production estimate toUNODC, was set at 25% of calculated domesticdemand. Clearly, this is not an ideal estimation tech-nique but, in a number of cases, subsequent indicationsof likely orders of magnitude of cannabis production,referred in scientific literature, came rather close to theseresults.
Proceeding along these lines on a country-by-countrybasis, global cannabis production estimates increasedfrom 35,000 mt to 42,000 mt Looking at the seizurefigures, this would suggest an interdiction rate ofaround 14%, which is not unrealistic. After the modelwas run with these ‘adjusted’ production figures, the dis-tribution pattern with regard to importing and export-ing regions fell into line with what is known aboutactual trafficking patterns. The basic pattern reflected inthis model is that, for most countries, local productionis destined for domestic demand and only relativelysmall amounts are destined for export. The most impor-tant importer is West and Central Europe, while thelargest market is North America.
One problem remained with regard to reconciling theseproduction estimates with consumption figures: NorthAmerica. Cannabis production estimates in NorthAmerica exceed estimated consumption levels. Thisproblem has been highlighted by US authorities else-where28, but no solution has been found to overcomethis data discrepancy. One potential explanation – thatcannabis herb is being exported from North America –
can be also ruled out, as cannabis prices are high inNorth America and exports to most markets wouldresult in losses for the traffickers.
Both demand side estimates and supply side estimatesseem to be based on scientific research, and this makesit difficult to simply ignore one or the other. Assumingthat the truth is probably somewhere in the middle,UNODC tried to find a compromise solution. Theapproach was to choose the lowest available productionestimates (14,370 mt for Mexico, the USA andCanada,29 instead of production estimates of around25,000 mt for the region30) and to subsequently deductanother 5,000 mt (about a third of the lower productionestimates) as ‘extraordinary losses’. After deductingseizures made in the region, this resulted in an estimateof 5.9 mt of cannabis herb available for consumption inNorth America, equivalent to a per capita consumption
135
2. Estimating the value of illicit drug markets
28 Drug Availability Steering Committee, Drug Availability Estimates in the United States, December 2002,http://www.whitehousedrugpolicy.gov/publications/pdf/drugavailability.pdf; see also UNODC, World Drug Report 2004.
29 Estimate for cannabis herb production in North America: Mexico: 7,900 tons in 2002 (US. Dept. of State, International Narcotics Control StrategyReport, 2004), USA: 5,670 tons in 2003 (UNODC, ARQ), Canada: 800 tons in 2003 (UNODC, ARQ).
30 According to the National Drug Intelligence Center, National Drug Threat Assessment 2005, cannabis herb production increased in 2003 in Mexicoto 13,500 tons; US cannabis production, according to ONDCP, may have amounted to more than 10,000 tons (ONDCP, National Drug ControlStrategy 2003) and the upper estimate of production in Canada was reported at 2000 tons. (National Drug Intelligence Center, National DrugThreat Assessment 2005).
Fig. 10: Regional distribution of cannabis herb retailsales in 2003 in billion US$ (N = US$113.1 bn)
* including Caribbean and Central AmericaSources: UNODC, Illicit Drug Market Estimation Model.
$4.2South-
America*4%
$23.9Europe21%
$7.9Africa7%
$8.6Asia8%
$6.1Oceania
5%
$62.5North
America55%
136
World Drug Report 2005 Volume I. Analysis
Tab
le 6
. Can
nab
is h
erb
: Pro
du
ctio
n a
nd
dis
trib
uti
on
fro
m s
ou
rce
cou
ntr
ies
to d
esti
nat
ion
cou
ntr
ies
East
Afr
ica
632,
513
063
2,51
31,
261,
816
774,
031
632,
513
00
00
00
00
00
00
00
063
2,51
3N
orth
Afr
ica
4,15
5,14
10
4,15
5,14
11,
793,
604
69,0
6039
3,87
31,
793,
604
064
0,92
00
73,8
130
00
00
047
9,35
576
8,57
10
5,00
44,
155,
141
Sout
hern
Afr
ica
2,96
9,52
00
2,96
9,52
01,
811,
648
62,7
7919
3,11
80
1,81
1,64
831
4,24
60
36,1
910
00
00
023
5,03
037
6,83
40
2,45
42,
969,
520
Wes
t an
d C
entr
al A
fric
a4,
152,
362
04,
152,
362
5,17
6,37
956
2,16
10
00
4,15
2,36
20
00
00
00
00
00
04,
152,
362
Car
ibbe
an55
8,88
10
558,
881
473,
486
133,
443
00
00
473,
486
7,43
80
00
00
00
77,4
520
504
558,
881
Cen
tral
Am
eric
a17
3,35
10
173,
351
331,
427
7,65
70
00
00
173,
351
00
00
00
00
00
173,
351
Nor
th A
mer
ica
14,3
70,0
008,
440,
349
5,92
9,65
15,
929,
651
9,88
90
00
00
05,
929,
651
00
00
00
00
05,
929,
651
Sout
h A
mer
ica
3,20
9,50
20
3,20
9,50
22,
153,
740
482,
186
00
00
040
,633
02,
153,
740
00
032
5,40
126
3,88
142
3,09
20
2,75
53,
209,
502
Cen
tral
Asi
a &
Tr
ansc
auca
sus
2,25
0,98
10
2,25
0,98
162
4,05
026
,431
00
00
00
00
624,
050
555,
337
034
3,50
127
8,55
944
6,62
60
2,90
82,
250,
981
East
and
Sou
th-E
ast
1,76
7,68
80
1,76
7,68
82,
425,
686
46,6
990
00
00
00
00
1,76
7,68
80
00
00
01,
767,
688
Nea
r &
Mid
dle
East
/SW
A
sia
1,93
8,23
60
1,93
8,23
61,
526,
315
88,1
5342
,312
00
68,8
510
00
00
102,
661
1,52
6,31
563
,500
51,4
9582
,564
053
81,
938,
236
Sout
h A
sia
3,53
9,37
80
3,53
9,37
84,
271,
781
82,2
390
00
00
00
00
00
3,53
9,37
80
00
03,
539,
378
East
ern
Euro
pe87
,520
087
,520
1,44
9,03
441
,845
00
00
00
00
00
00
87,5
200
00
87,5
20
Wes
tern
& C
entr
al
Euro
pe96
6,30
70
966,
307
3,22
6,73
266
,368
00
00
00
00
00
00
096
6,30
70
096
6,30
7
Sout
h Ea
st E
urop
e54
0,47
60
540,
476
401,
439
46,5
130
00
00
00
00
00
053
,194
85,2
8840
1,43
955
554
0,47
6
Oce
ania
680,
099
068
0,09
969
4,81
713
,606
00
00
00
00
00
00
00
068
0,09
968
0,09
9
All
Co
un
trie
s41
,991
,953
8,44
0,34
933
,551
,605
33,5
51,6
052,
513,
060
1,26
1,81
61,
793,
604
1,81
1,64
85,
176,
379
473,
486
331,
427
5,92
9,65
12,
153,
740
624,
050
2,42
5,68
61,
526,
315
4,27
1,78
11,
449,
034
3,22
6,73
240
1,43
969
4,81
733
,551
,605
Tab
le 7
. Can
nab
is h
erb
: Su
pp
ly a
nd
dem
and
in d
esti
nat
ion
cou
ntr
ies
632,
513
4,15
5,14
12,
969,
520
4,15
2,36
255
8,88
117
3,35
114
,370
,000
3,20
9,50
22,
250,
981
1,76
7,68
81,
938,
236
3,53
9,37
887
,520
966,
307
540,
476
680,
099
41,9
91,9
530
00
00
08,
440,
349
00
00
00
00
08,
440,
349
632,
513
4,15
5,14
12,
969,
520
4,15
2,36
255
8,88
117
3,35
15,
929,
651
3,20
9,50
22,
250,
981
1,76
7,68
81,
938,
236
3,53
9,37
887
,520
966,
307
540,
476
680,
099
33,5
51,6
051
44
148
311
393
616
2730
5032
665
1,38
533
1,47
625
8
115
134
270
205,
552
5262
5497
114
581,
338
181,
004
8,67
1
1,26
1,81
61,
793,
604
1,81
1,64
85,
176,
379
473,
486
331,
427
5,92
9,65
12,
153,
740
624,
050
2,42
5,68
61,
526,
315
4,27
1,78
11,
449,
034
3,22
6,73
240
1,43
969
4,81
733
,551
,605
774,
031
69,0
6062
,779
562,
161
133,
443
7,65
79,
889
482,
186
26,4
3146
,699
88,1
5382
,239
41,8
4566
,368
46,5
1313
,606
2,51
3,06
048
7,78
51,
724,
544
1,74
8,86
94,
614,
218
340,
042
323,
770
5,91
9,76
21,
671,
554
597,
619
2,37
8,98
71,
438,
162
4,18
9,54
21,
407,
189
3,16
0,36
435
4,92
668
1,21
131
,038
,545
10
00
20
20
01
00
13
03
1
488
318
120
299
549
6111
,759
376
194
2,17
852
228
01,
914
8,53
715
92,
106
29,8
60
2,43
95,
159
5,48
520
,974
1,36
093
535
,877
8,35
81,
827
11,3
434,
198
24,2
074,
900
15,8
021,
775
3,40
614
8,04
348
7,78
51,
552,
090
1,57
3,98
24,
614,
218
340,
042
291,
393
5,91
9,76
21,
671,
554
537,
857
2,14
1,08
91,
294,
346
3,77
0,58
81,
266,
471
3,16
0,36
435
4,92
668
1,21
129
,657
,677
200
301
287
220
250
312
165
200
294
189
308
156
258
200
200
200
200
21
01
72
111
13
10
26
59
4
1,16
081
555
34,
651
2,21
744
362
,488
1,54
535
26,
228
1,80
858
13,
147
19,1
161,
726
6,12
811
2,95
9
West & Central Europe
SouthEast Europe
Oceania
All Countries
East & SouthEast Asia
Near & Middle East /SW Asia
South Asia
East Europe
Central America
North America
South America
Central Asia & Transcaucasus
North Africa
South Africa
West & Central Africa
Caribbean
Reg
ion
s
East Africa
Impl
ied
Con
sum
ptio
n pe
r us
er (g
ms
Can
nabi
s Eq
uiv)
Ave
rage
Ret
ail P
rice
US$
/gm
Ret
aile
r In
com
e (U
S$m
ill)
Dem
and
:Es
timat
ed U
ser
Popu
latio
n (T
hous
ands
)Es
timat
ed A
ctua
l Con
sum
ptio
n pe
r ye
ar (K
g C
anna
bis
Equi
v)
Tota
l Sei
zed/
Lost
at
Des
tinat
ion
(Kg
Can
nabi
s Eq
uiv)
Tota
l Ava
ilabl
e fo
r C
onsu
mpt
ion
(Kg
Can
nabi
s Eq
uiv)
Who
lesa
le p
rice
at D
estin
atio
n U
S$ /g
m
Wh
ole
sale
r In
com
e (U
S$m
ill)
Pro
du
cer
Inco
me
(US$
mill
)
Sup
ply
:To
tal I
nten
ded
for
Con
sum
ptio
n (K
g C
anna
bis
Equi
v)
Tota
l Pro
duct
ion
in S
ourc
e C
ount
ry (K
g C
anna
bis
Equi
v)To
tal S
eize
d/Lo
st in
Sou
rce
Cou
ntry
(Kg
Can
nabi
s Eq
uiv)
Tota
l Ava
ilabl
e fo
r Sa
le (K
g C
anna
bis
Equi
v)Fa
rmga
te P
rice
at O
rigin
(US$
/Kg
Can
nabi
s Eq
uiv)
Yo
u n
eed
to
Re-
Cal
ibra
te!
==
==
==
=>
>>
Pro
du
ctio
n:
Central America
North America
Tota
l Sei
zed
/ lo
st in
Tr
ansi
t (K
g
Can
nab
isEq
uiv
)
Tran
sfer
red
to
M
arke
ts (
Kg
C
ann
abis
Equ
iv)
Tota
l Ava
ilab
le
for
Sale
(K
g
Can
nab
is E
qu
iv)
Tota
lSe
ized
/Lo
st in
Sou
rce
Co
un
try
(Kg
C
ann
abis
Equ
iv)
Tota
lPr
od
uct
ion
in
Sou
rce
Co
un
try
(Kg
Can
nab
is
Equ
iv)
Pro
du
cer
Reg
ion
s
North Africa
East Africa
Co
nsu
mer
Reg
ion
s
South America
East & SouthEast Asia
All Countries
South Africa
West & Central Africa
Caribbean
Near & Middle East /SW Asia
Central Asia & Transcaucasus
South Asia
East Europe
Oceania
West & Central Europe
SouthEast Europe
rate of 165 grams. This is about twice the rate indicatedby some previous US studies,31 but it is in line with theorders of magnitude shown in studies or reports from afew other countries. It also seems to be a feasible orderof magnitude, taking the distribution pattern betweeninfrequent and intensive cannabis users as well as infor-mation about the amounts of cannabis herb taken bysuch groups32 in North America into account.
Multiplying these consumption estimates by reportedprices (US$10.6 per gram at the retail level), the NorthAmerican cannabis herb market was calculated toamount to some US$63 bn. This is far more than pre-vious estimates, starting from the demand side, had sug-gested, but it is the lowest estimate UNODC couldcome up with without completely disregarding NorthAmerican cannabis production estimates. The nextlargest market, using similar per capita consumptionrates of around 200 grams per year, are Europe (US$24bn), followed by Asia (US$9bn) and Africa (US$8 bn).
In short, there are existing data weaknesses on both thesupply and the demand side with regard to cannabisherb. An attempt was made, based on the triangulationof existing data and information, to reconcile, as far aspossible, the data discrepancies. This resulted in an over-all estimate of the amounts available for consumption of30,000 mt of cannabis herb, giving a global farmgatevalue of cannabis production of U$9 bn, a wholesalevalue of U$30 bn and a retail value of US$113 bn.
Valuation of cannabis resin
An evaluation of global cannabis resin production wasdone by UNODC, for the first time, for last year’sWorld Drug Report. In co-operation with the Govern-ment of Morocco using modern remote sensing tech-nology, ground verification and a yield survey,UNODC estimated resin production in that country at3,070 mt in 2003. This led to a minimum globalcannabis resin production estimate of 5,100 mt Basedon a slightly different approach, analysing cannabis herband cannabis resin seizures, a final global productionestimate of 7,400 mt was established.33
Making use of existing cannabis resin estimates from theprevious year and information from the main sourcecountries, the remaining 4,330 mt were allocated to thedifferent regions.34 This resulted in an estimate of closeto 2,000 mt for the Near and Middle East/South-WestAsia region, mainly reflecting production inAfghanistan, Pakistan and Lebanon, and an estimate ofaround 600 mt for the Central Asia and Caucasus sub-region, reflecting, in particular, important levels of pro-duction in Kazakhstan and Kyrzystan.
The model assumes that the main destination for thecannabis resin produced in North Africa is Europe,notably West and Central Europe, while the bulk ofcannabis resin produced in Near and MiddleEast/South-West Asia region is for local consumptionand only smaller amounts are destined for markets inWestern Europe. The bulk of cannabis resin consumedin East Europe is assumed to originate in Central Asia.Cannabis resin produced in the Caribbean (mainly
137
2. Estimating the value of illicit drug markets
Fig. 11: Regional distribution of cannabis resin retailsales in 2003 in billion US$ (N = US$28.8 bn)
* including Caribbean and Central AmericaSources: UNODC, Illicit Drug Market Estimation Model.
$22.3Europe78%
$0.0South-
America*0%
$1.5North
America5%
$0.1Oceania
0%
$2.4Asia8%
$2.6Africa9%
31 Abt Associates, What America's Users Spend on Illegal Drugs, 1988-1998, December 2000. 32 The 1998 US household survey distinguished between three groups of cannabis users: those consuming it on 51 days or more (36% of all cannabis
users), those consuming it on 12 to 51 days (20%) and those consuming it on less than 12 days (44%). Assuming that a group of 'hard-core' cannabisusers smokes up to 4 grams (8 joints) for, on average, 107 days a year (equivalent to 1.2 grams of cannabis herb or 2.4 joints per day, every day),that the second group uses a daily cannabis dose of 1.5 grams for 31.5 days a year, and the third group uses a dose of 0.5 grams for 6 days a year,and applying the cannabis prevalence data from the 2003 Survey on Drug Use and Health to this distribution pattern, the average cannabisconsumption per user (annual prevalence) would be equivalent to 165 grams.
33 UNODC, 2004 World Drug Report, Volume 1: Analysis, p. 129.34 The allocation is of cannabis production according to regions is intended to show the production pattern, but is not critical for the final outcome
of the wholesale or retail values of the cannabis resin market.
138
World Drug Report 2005 Volume I. Analysis
Tab
le 8
. Can
nab
is r
esin
: Pro
du
ctio
n a
nd
dis
trib
uti
on
fro
m s
ou
rce
cou
ntr
ies
to d
esti
nat
ion
cou
ntr
ies
East
Afr
ica
121,
932
012
1,93
271
,099
1,88
671
,099
00
1,52
20
01,
357
00
00
068
47,8
510
3412
1,93
2N
orth
Afr
ica
3,08
0,00
066
,394
3,01
3,60
677
8,60
04,
206
077
8,60
00
87,9
040
078
,401
00
00
03,
941
2,06
2,78
60
1,97
43,
013,
606
Sout
hern
Afr
ica
142,
041
014
2,04
112
5,49
01,
744
00
125,
490
496
00
442
00
00
022
15,5
800
1114
2,04
1W
est
and
Cen
tral
Afr
ica
36,0
270
36,0
2714
7,66
52
00
036
,027
00
00
00
00
00
00
36,0
27C
arib
bean
256,
777
025
6,77
72,
106
560
00
00
2,10
60
6,57
40
00
00
353
247,
568
017
725
6,77
7C
entr
al A
mer
ica
00
00
00
00
00
00
00
00
00
00
00
Nor
th A
mer
ica
00
011
4,75
878
00
00
00
00
00
00
00
00
00
Sout
h A
mer
ica
20,0
000
20,0
002,
470
1,34
10
00
00
048
32,
470
00
00
2417
,011
012
20,0
00C
entr
al A
sia
& T
rans
cauc
asus
638,
014
063
8,01
467
,551
238
00
00
00
00
67,5
510
00
268,
738
301,
510
021
563
8,01
4Ea
st a
nd S
outh
-Eas
t A
sia
60,0
000
60,0
0017
,059
752
00
00
00
00
017
,059
00
6142
,850
031
60,0
00N
ear
& M
iddl
e Ea
st /S
W A
sia
1,99
1,73
385
,126
1,90
6,60
61,
226,
378
151,
089
00
021
,716
00
18,1
260
00
1,22
6,37
80
911
639,
018
045
61,
906,
606
Sout
h A
sia
679,
584
067
9,58
433
8,96
225
,322
00
00
00
9,37
60
00
033
8,96
247
133
0,53
90
236
679,
584
East
ern
Euro
pe20
,000
020
,000
294,
894
1,43
40
00
00
00
00
00
020
,000
00
020
,000
Wes
tern
& C
entr
al E
urop
e0
00
3,94
2,65
073
0,65
00
00
00
00
00
00
00
00
00
Sout
h Ea
st E
urop
e35
3,89
10
353,
891
115,
482
1,61
50
00
00
00
00
00
030
323
7,93
711
5,48
217
035
3,89
1O
cean
ia0
00
3,31
656
00
00
00
00
00
00
00
00
0
All
Co
un
trie
s7,
400,
000
151,
520
7,24
8,48
07,
248,
480
921,
676
71,0
9977
8,60
012
5,49
014
7,66
52,
106
011
4,75
82,
470
67,5
5117
,059
1,22
6,37
833
8,96
229
4,89
43,
942,
650
115,
482
3,31
67,
248,
480
Tab
le 9
. Can
nab
is r
esin
: Su
pp
ly a
nd
dem
and
in d
esti
nat
ion
cou
ntr
ies
121,
932
3,08
0,00
014
2,04
136
,027
256,
777
00
20,0
0063
8,01
460
,000
1,99
1,73
367
9,58
420
,000
035
3,89
10
7,40
0,00
00
66,3
940
00
00
00
085
,126
00
00
015
1,52
012
1,93
23,
013,
606
142,
041
36,0
2725
6,77
70
020
,000
638,
014
60,0
001,
906,
606
679,
584
20,0
000
353,
891
07,
248,
480
095
00
00
00
00
00
00
00
95
1228
513
324
00
260
618
164
20
340
687
71,0
9977
8,60
012
5,49
014
7,66
52,
106
011
4,75
82,
470
67,5
5117
,059
1,22
6,37
833
8,96
229
4,89
43,
942,
650
115,
482
3,31
67,
248,
480
1,88
64,
206
1,74
42
560
078
01,
341
238
752
151,
089
25,3
221,
434
730,
650
1,61
556
921,
676
69,2
1477
4,39
412
3,74
614
7,66
31,
546
011
3,97
91,
129
67,3
1316
,306
1,07
5,29
031
3,64
029
3,46
03,
212,
000
113,
867
3,26
06,
326,
804
12
32
13
61
14
00
42
17
2
561,
168
412
269
10
635
152
6733
612
21,
128
5,96
812
821
10,3
6 4
461
5,16
382
598
410
076
08
449
109
7,16
92,
091
1,95
617
,829
759
2238
,594
69,2
1477
4,39
412
3,74
614
7,66
31,
546
011
3,97
91,
129
67,3
1316
,306
1,07
5,29
031
3,64
029
3,46
02,
890,
800
113,
867
3,26
06,
005,
604
150
150
150
150
150
015
015
015
015
015
015
015
016
215
015
015
61
25
310
813
21
222
14
73
245
621,
455
604
443
150
1,45
92
9336
41,
711
226
1,27
920
,723
329
7928
,845
Impl
ied
Con
sum
ptio
n pe
r us
er (g
ms
Can
nabi
s Eq
uiv)
Ave
rage
Ret
ail P
rice
US$
/gm
Ret
aile
r In
com
e (U
S$m
ill)
Dem
and
:
Estim
ated
Use
r Po
pula
tion
(Tho
usan
ds)
Estim
ated
Act
ual C
onsu
mpt
ion
per
year
(Kg
Can
nabi
s Eq
uiv)
Tota
l Int
ende
d fo
r C
onsu
mpt
ion
(Kg
Can
nabi
s Eq
uiv)
Tota
l Ava
ilabl
e fo
r C
onsu
mpt
ion
(Kg
Can
nabi
s Eq
uiv)
Who
lesa
le p
rice
at D
estin
atio
n U
S$ /g
m
Wh
ole
sale
r In
com
e (U
S$m
ill)
Tota
l Sei
zed/
Lost
at
Des
tinat
ion
(Kg
Can
nabi
s Eq
uiv)
Pro
du
ctio
n:
Tota
l Pro
duct
ion
in S
ourc
e C
ount
ry (K
g C
anna
bis
Equi
v)To
tal S
eize
d/Lo
st in
Sou
rce
Cou
ntry
(Kg
Can
nabi
s Eq
uiv)
Tota
l Ava
ilabl
e fo
r Sa
le (K
g C
anna
bis
Equi
v)Fa
rmga
te P
rice
at O
rigin
(US$
/Kg
Can
nabi
s Eq
uiv)
Pro
du
cer
Inco
me
(US$
mill
)
Sup
ply
:
Yo
u n
eed
to
Re-
Cal
ibra
te!
==
==
==
=>
>>
South America
Co
nsu
mer
Reg
ion
s
North Africa
East Africa
Tran
sfer
red
to
M
arke
ts (
Kg
C
ann
abis
Equ
iv)
All Countries
East & SouthEast Asia
Central Asia & Transcaucasus
All Countries
South Asia
Near & Middle East /SW Asia
East Europe
West & Central Europe
SouthEast Europe
Oceania
Pro
du
cer
Reg
ion
s
Tota
lPr
od
uct
ion
in
Sou
rce
Co
un
try
(Kg
C
ann
abis
Equ
iv)
Tota
lSe
ized
/Lo
st in
So
urc
eC
ou
ntr
y (K
g
Can
nab
isEq
uiv
)
Tota
l Ava
ilab
le
for
Sale
(K
g
Can
nab
is E
qu
iv)
Tota
l Sei
zed
/ lo
st in
Tr
ansi
t (K
g
Can
nab
isEq
uiv
)
East Africa
North Africa
South Africa
Reg
ion
s
North America
Central America
Caribbean
West & Central Africa
South Africa
West & Central Africa
Caribbean
Central America
North America
South America
Central Asia & Transcaucasus
East & SouthEast Asia
Near & Middle East /SW Asia
Oceania
South Asia
East Europe
West & Central Europe
SouthEast Europe
Jamaica) is destined for North America. Cannabis resinproduced in South Asia (mainly Nepal) is destined forconsumption within the region and for export to Westand Central Europe.
After seizures and losses, the model assumes that about6,000 mt of cannabis resin are available for consump-tion. The number of cannabis resin users was deducedfrom the total number of cannabis users, based on thesplit between cannabis resin and total cannabis seizuresover a ten-year period, and taking the possibility ofsome overlap between cannabis herb and resin con-sumption into account. This estimate resulted in a percapita estimate of 150 grams of cannabis resin per user.This is in line with some estimates on per capita con-sumption of cannabis resin obtained from countries inEurope. It is also in line with reports that the averagepotency of cannabis resin is still higher than the averagepotency of cannabis herb (even though there are impor-tant exceptions when it comes to hydroponically pro-duced cannabis), which means that per capitaconsumption of cannabis resin is usually lower than percapita consumption of cannabis herb.
Based on prevalence data and per capita consumptionfigures, the largest cannabis market resin market is thatof West and Central Europe (2,900 mt), which, whenmultiplied with average retail prices, gives a marketvalue of US$21 bn. Europe thus accounts for 78% ofthe global cannabis resin market, followed by Africa(9%) and Asia (8%). The main cannabis resin market inAsia is the Near and Middle East; the main market inAfrica is North Africa.
2.3.4 Amphetamine-type stimulantstrade - valued at US$44 bn peryear (retail level)
The ATS market consists of three main products:methamphetamine, amphetamine and ecstasy. Metham-phetamine, amphetamine and related stimulants arecombined under the category of ‘amphetamines’ . Theglobal amphetamines retail market was valued at US$28 bn. The global ecstasy retail market, includingMDMA and related substances, was valued at US$16bn. Taken together, the ATS retail markets add up toUS$44 bn. The largest ATS retail markets in economicterms are North America (57%), followed by Asia(20%), Europe (11%) and Oceania (9%).
Valuation of the amphetamines market
The valuation of the amphetamines market started froma global production estimate of 332 mt (range: 278 –401 mt), derived from production estimates based onextrapolation from seizures of amphetamines, seizures ofprecursors and consumption estimates. This produc-tion was ‘allocated’ to countries based on identificationsas a source country by other countries; the number ofdismantled laboratories; and seizures made in countrieswith dismantled laboratories that were identified byother countries as significant source countries. In addi-tion, information from production estimates fromNorth America was used to adjust the weights given tothe different indicators. According to ONDCP,methamphetamine production in North America is esti-mated to range between 106 and 144 metric mt35
The results of these calculations suggests that the largestshare of the world’s production of amphetamines is inEast and South-East Asia (162 tons), followed by NorthAmerica (114 mt) and West and Central Europe (39mt). While most of the production in East and South-East Asia and in North America concerns methamphet-amine, European production is mainly focused onamphetamine.
139
2. Estimating the value of illicit drug markets
Fig. 12: Regional distribution of ATS retail sales in2003 in billion US$ (N = US$44.3 bn)
* including Caribbean and Central AmericaSources: UNODC, Illicit Drug Market Estimation Model.
$8.7Asia20%
$4.8Europe11%
$1.4South-
America*3%
$3.8Oceania
9%
$0.2Africa0%
$25.4North
America57%
35 The White House, The National Drug Control Strategy, February 2003.
140
World Drug Report 2005 Volume I. Analysis
Tab
le 1
0. A
TS (
excl
ud
ing
ecs
tasy
): P
rod
uct
ion
an
d d
istr
ibu
tio
n f
rom
so
urc
eco
un
trie
s to
des
tin
atio
nco
un
trie
s
East
Afr
ica
00
01,
105
00
00
00
00
00
00
00
00
00
Nor
th A
fric
a0
00
705
120
00
00
00
00
00
00
00
00
Sout
hern
Afr
ica
20
22,
019
10
02
00
00
00
00
00
00
02
Wes
t an
d C
entr
al A
fric
a0
00
6,26
43
00
00
00
00
00
00
00
00
0C
arib
bean
00
053
30
00
00
00
00
00
00
00
00
0C
entr
al A
mer
ica
00
02,
229
00
00
00
00
00
00
00
00
00
Nor
th A
mer
ica
113,
713
1,11
411
2,59
911
1,19
549
10
00
079
331
111,
195
913
00
00
00
082
112,
599
Sout
h A
mer
ica
00
06,
148
00
00
00
00
00
00
00
00
00
Cen
tral
Asi
a &
Tra
nsca
ucas
us0
00
00
00
00
00
00
00
00
00
00
0Ea
st a
nd S
outh
-Eas
t A
sia
161,
528
4,81
815
6,71
015
2,10
59,
312
282
051
40
116
484
01,
334
015
2,10
51,
080
00
00
796
156,
710
Nea
r &
Mid
dle
East
/SW
Asi
a0
00
2,32
846
60
00
00
00
00
00
00
00
00
Sout
h A
sia
00
00
00
00
00
00
00
00
00
00
00
East
ern
Euro
pe2,
282
02,
282
1,01
056
6657
121
505
2711
40
314
00
400
1,01
00
028
2,28
2W
este
rn &
Cen
tral
Eur
ope
39,0
130
39,0
1328
,666
3,97
953
946
198
44,
102
221
926
02,
555
00
328
00
28,6
660
229
39,0
13So
uth
East
Eur
ope
7,31
10
7,31
12,
475
547
218
186
398
1,65
889
374
01,
032
00
880
00
02,
475
07,
311
Oce
ania
8,15
10
8,15
19,
286
569
00
00
00
00
00
00
00
08,
151
8,15
1
All
Co
un
trie
s33
2,00
05,
931
326,
069
326,
069
15,4
371,
105
705
2,01
96,
264
533
2,22
911
1,19
56,
148
015
2,10
52,
328
01,
010
28,6
662,
475
9,28
632
6,06
9
Tab
le 1
1. A
TS (
excl
ud
ing
ecs
tasy
): S
up
ply
an
d d
eman
d in
des
tin
atio
nco
un
trie
s
00
20
00
113,
713
00
161,
528
00
2,28
239
,013
7,31
18,
151
332,
000
00
00
00
1,11
40
04,
818
00
00
00
5,93
10
02
00
011
2,59
90
015
6,71
00
02,
282
39,0
137,
311
8,15
132
6,06
984
840
41,
120
2511
198
23,
028
639
2585
589
764
1,01
42,
065
1,68
67,
099
1,92
6
00
00
00
341
00
134
00
281
1258
628
1,10
570
52,
019
6,26
453
32,
229
111,
195
6,14
80
152,
105
2,32
80
1,01
028
,666
2,47
59,
286
326,
069
012
13
00
491
00
9,31
246
60
563,
979
547
569
15,4
371,
105
693
2,01
76,
261
533
2,22
911
0,70
46,
148
014
2,79
31,
862
095
324
,687
1,92
88,
718
310,
632
84
40
110
310
017
91
1413
1263
22
93
73
122
3,42
63
02,
416
160
1432
122
550
6,81
3
221
249
403
1,25
289
372
3,45
91,
025
317
,855
143
152
156
2,05
716
179
428
,391
1,10
562
42,
017
6,26
153
32,
229
110,
704
6,14
80
128,
514
1,86
20
858
24,6
871,
928
7,8 4
629
5,31
75
35
56
632
60
713
06
1212
1010
4311
111
156
153
50
5349
140
7626
293
96
477
228
112
516
,896
330
6,76
191
034
1,88
649
2,29
628
,256
All Countries
Tran
sfer
red
to
M
arke
ts (
Kg
A
TS (
excl
. ec
stas
y) E
qu
iv)
Tota
lSe
ized
/ lo
st
in T
ran
sit
(Kg
ATS
(e
xcl.
ecst
asy)
Equ
iv)
Pro
du
cer
Reg
ion
s
Tota
lPr
od
uct
ion
in
Sou
rce
Co
un
try
(Kg
A
TS (
excl
. ec
stas
y)Eq
uiv
)
Tota
lSe
ized
/Lo
st in
So
urc
eC
ou
ntr
y (K
g
ATS
(ex
cl.
ecst
asy)
Equ
iv)
Tota
lA
vaila
ble
fo
r Sa
le (
Kg
ATS
(e
xcl.
ecst
asy)
Eq
uiv
)
Co
nsu
mer
Reg
ion
s
All Countries
Reg
ion
s
East Africa East Africa
North Africa
South Africa
West & Central Africa
Caribbean
Impl
ied
Con
sum
ptio
n pe
r us
er (g
ms
ATS
(exc
l. ec
stas
y) E
quiv
)A
vera
ge R
etai
l Pric
e U
S$ /
pure
gm
Ret
aile
r In
com
e (U
S$m
ill)
Dem
and
:Es
timat
ed U
ser
Popu
latio
n (T
hous
ands
)Es
timat
ed A
ctua
l Con
sum
ptio
n pe
r ye
ar (K
g A
TS (e
xcl.
ecst
asy)
Equ
iv)
Tota
l Sei
zed/
Lost
at
Des
tinat
ion
(Kg
ATS
(exc
l. ec
stas
y) E
quiv
)To
tal A
vaila
ble
for
Con
sum
ptio
n (K
g A
TS (e
xcl.
ecst
asy)
Equ
iv)
Who
lesa
le p
rice
at D
estin
atio
n U
S$ /p
ure
m
Wh
ole
sale
r In
com
e (U
S$m
ill)
Pro
du
cer
Inco
me
(US$
mill
)
Sup
ply
:To
tal I
nten
ded
for
Con
sum
ptio
n (K
g A
TS (e
xcl.
ecst
asy)
Equ
iv)
Tota
l Pro
duct
ion
in S
ourc
e C
ount
ry (K
g A
TS (e
xcl.
ecst
asy)
Equ
iv)
Tota
l Sei
zed/
Lost
in S
ourc
e C
ount
ry (K
g A
TS (e
xcl.
ecst
asy)
Equ
iv)
Tota
l Ava
ilabl
e fo
r Sa
le (K
g A
TS (e
xcl.
ecst
asy)
Equ
iv)
Labo
rato
ry P
rice
at O
rigin
(US$
/Kg
ATS
(exc
l. ec
stas
y) E
quiv
)
North Africa
Yo
u n
eed
to
Re-
Cal
ibra
te!
==
==
==
=>
>>
Pro
du
ctio
n:
Central America
North America
South Africa
West & Central Africa
Caribbean
Near & Middle East /SW Asia
Oceania
Central Asia & Transcaucasus
West & Central Europe
SouthEast Europe
South America
East & SouthEast Asia
South Asia
East Europe
Central America
North America
South America
Central Asia & Transcaucasus
West & Central Europe
SouthEast Europe
Oceania
East & SouthEast Asia
Near & Middle East /SW Asia
South Asia
East Europe
The model also reflects the general perception thatamphetamines are mainly traded intra-regionally. Thus,most of the production of East and South-East Asia isfor consumption within the region, and the sameapplies to North America and to West and CentralEurope. Out of the total of 332 mt, 295 mt are esti-mated to be available for consumption after seizures andlosses are deducted. The model assumes that 129 mt areavailable for consumption in East and South-East Asia,111 mt in North America, and 25 mt in West and Cen-tral Europe. The implied per capita consumption is highfor North America (32 grams per user per year)36 andmuch lower in West and Central Europe (12 grams) andin East and South-East Asia (7 grams). This is a conse-quence of the rather high production levels estimated bythe authorities in North America, and the fact that thereis no information of methamphetamine or ampheta-mine produced in North America being shipped toother regions. Thus, all of the amphetamines producedin North America, less seizures and losses, are presum-ably consumed there. Using these consumption levels,the amphetamines market in North America was esti-mated at US$17 bn, in East and South-East Asia atUS$7 bn and in Oceania and in Europe at US$2 bneach. The total market was valued at US$28 bn.
The model shows some trafficking of amphetamines tocountries in Africa and South America. This, however,is only partially correct. Both Africa and South Americahave, in terms of ATS users, quite substantial ATS mar-kets, but much of these markets are sourced from legallyproduced ATS which are subsequently diverted, ratherthan from illicitly produced ATS. This is a problem forthe current model, as such diverted drugs were not con-sidered in the initial phase when the model wasdesigned. This means that the overall markets foramphetamines (licit and illicit) in Africa and SouthAmerica are larger than what is reflected in this model.
Valuation of the ecstasy market
Global production of ecstasy – extrapolated fromseizures of ecstasy, from seizures of ecstasy precursorsand from consumption estimates was estimated at 90mt (range: 45 – 141 mt). The allocation of productionto countries/regions was based on dismantled laborato-ries, citations as countries of origin by other countries,and seizures (for countries that had laboratories andwhich were cited as countries of origin).
Using this approach, data suggest that the bulk ofecstasy production (69 mt out of 90 mt or 77%) con-tinues to take place in West and Central Europe. The
141
2. Estimating the value of illicit drug markets
36 Per capita consumption of amphetamines, according to these estimates, is still lower in North America than the corresponding estimates for cocaine,another stimulant.
Fig. 13: Regional distribution of amphetamines retailsales in 2003 in billion US$ (N = US$28.3 bn)
* including Caribbean and Central AmericaSources: UNODC, Illicit Drug Market Estimation Model.
$6.9Asia24%
$2.3Oceania
8%
$0.2South-
America*1%
$2.0Europe
7%
$0.1Africa0%
$16.9North
America60%
Fig. 14: Regional distribution of ecstasy retail sales in2003 in billion US$ (N = US$16.1 bn)
* including Caribbean and Central AmericaSources: UNODC, Illicit Drug Market Estimation Model.
$8.5North
America52%
$0.1Africa1%
$1.5Oceania
10%
$1.2South-
America*7%
$1.9Asia12%
$2.8Europe18%
142
World Drug Report 2005 Volume I. Analysis
Tab
le 1
2. E
csta
sy:P
rod
uct
ion
an
d d
istr
ibu
tio
n f
rom
so
urc
eco
un
trie
s to
des
tin
atio
nco
un
trie
s
East
Afr
ica
00
00
00
00
00
00
00
00
00
00
00
Nor
th A
fric
a0
00
00
00
00
00
00
00
00
00
00
0So
uthe
rn A
fric
a93
10
931
1,13
559
00
931
00
00
00
00
00
00
093
1W
est
and
Cen
tral
Afr
ica
00
022
00
00
00
00
00
00
00
00
00
Car
ibbe
an0
00
612
170
00
00
00
00
00
00
00
00
Cen
tral
Am
eric
a0
00
568
20
00
00
00
00
00
00
00
00
Nor
th A
mer
ica
12,2
6982
411
,445
36,3
2732
00
00
00
11,4
450
00
00
00
00
11,4
45So
uth
Am
eric
a1,
179
01,
179
4,07
520
60
00
00
00
1,17
90
00
00
00
01,
179
Cen
tral
Asi
a &
Tra
nsca
ucas
us0
00
00
00
00
00
00
00
00
00
00
0Ea
st a
nd S
outh
-Eas
t A
sia
4,08
70
4,08
75,
920
367
00
00
00
00
04,
087
00
00
00
4,08
7N
ear
& M
iddl
e Ea
st /S
W A
sia
207
020
745
610
00
00
00
00
00
020
70
00
00
207
Sout
h A
sia
00
00
00
00
00
00
00
00
00
00
00
East
ern
Euro
pe0
00
762
10
00
00
00
00
00
00
00
00
Wes
tern
& C
entr
al E
urop
e69
,150
3,09
266
,058
28,6
011,
623
00
204
2261
256
824
,882
2,89
60
1,83
324
90
762
28,6
0127
45,
156
66,0
58So
uth
East
Eur
ope
1,59
40
1,59
41,
867
280
00
00
00
00
00
00
01,
594
01,
594
Oce
ania
784
078
45,
940
748
00
00
00
00
00
00
00
078
478
4
All
Co
un
trie
s90
,200
3,91
686
,284
86,2
843,
182
00
1,13
522
612
568
36,3
274,
075
05,
920
456
076
228
,601
1,86
75,
940
86,2
84
Tab
le 1
3. E
csta
sy:S
up
ply
an
d d
eman
d in
des
tin
atio
nco
un
trie
s
00
931
00
012
,269
1,17
90
4,08
720
70
069
,150
1,59
478
490
,200
00
00
00
824
00
00
00
3,09
20
03,
916
00
931
00
011
,445
1,17
90
4,08
720
70
066
,058
1,59
478
486
,284
00
6,96
10
00
24,4
8339
,421
021
,803
23,0
100
07,
887
23,0
0147
,014
11,8
39
00
60
00
280
460
895
00
521
3737
1,02
2
00
1,13
522
612
568
36,3
274,
075
05,
920
456
076
228
,601
1,86
75,
940
86,2
840
059
017
232
206
036
710
00
11,
623
2874
83,
182
00
1,07
622
595
565
36,2
953,
869
05,
552
356
076
126
,977
1,84
05,
192
83,1
0293
9375
4923
315
510
316
493
9377
9385
5797
143
93
00
811
138
883,
723
633
051
827
065
1,53
017
974
17,
724
00
108
259
573,
561
387
055
536
013
92,
698
184
519
8,30
40
01,
076
2259
556
532
,665
3,86
90
5,55
235
60
685
26,9
771,
840
5,19
279
,396
00
1010
1010
910
010
100
510
1010
1020
420
487
6028
015
726
024
420
433
012
520
419
391
139
298
204
00
941
166
898,
503
944
01,
835
450
132
2,45
825
71,
549
16,0
72
Co
nsu
mer
Reg
ion
s
All Countries
Tran
sfer
red
to M
arke
ts
(Kg
Ecs
tasy
Eq
uiv
)
Tota
lSe
ized
/lo
st in
Tr
ansi
t (K
g
Ecst
asy
Equ
iv)
West & Central Europe
East Europe
South Asia
Pro
du
cer
Reg
ion
s
Tota
lPr
od
uct
ion
in S
ou
rce
Co
un
try
(Kg
Ec
stas
yEq
uiv
)
Tota
lSe
ized
/Lo
stin
So
urc
e C
ou
ntr
y (K
g
Ecst
asy
Equ
iv)
Tota
lA
vaila
ble
for
Sale
(K
g E
csta
sy
Equ
iv)
SouthEast Europe
North America
South AmericaSouth America
East & SouthEast Asia
Near & Middle East /SW Asia
South Asia
North Africa
North America
East Africa
Tota
l Sei
zed/
Lost
at
Des
tinat
ion
(Kg
Ecst
asy
Equi
v)To
tal A
vaila
ble
for
Con
sum
ptio
n (K
g Ec
stas
y Eq
uiv)
Who
lesa
le p
rice
at D
estin
atio
n U
S$ /g
m
Oceania
Central America
Near & Middle East /SW Asia
Central Asia & Transcaucasus
Tota
l Int
ende
d fo
r C
onsu
mpt
ion
(Kg
Ecst
asy
Equi
v)
South Africa
West & Central Africa
Caribbean
Farm
gate
Pric
e at
Orig
in (U
S$/K
g Ec
stas
y Eq
uiv)
Pro
du
cer
Inco
me
(US$
mill
)
Sup
ply
:
Pro
du
ctio
n:
Tota
l Pro
duct
ion
in S
ourc
e C
ount
ry (K
g Ec
stas
y Eq
uiv)
Tota
l Sei
zed/
Lost
in S
ourc
e C
ount
ry (K
g Ec
stas
y Eq
uiv)
Tota
l Ava
ilabl
e fo
r Sa
le (K
g Ec
stas
y Eq
uiv)
East Europe
West & Central Europe
East Africa
North Africa
South Africa
West & Central Africa
Reg
ion
s
SouthEast Europe
Oceania
All Countries
Central Asia & Transcaucasus
East & SouthEast Asia
Caribbean
Central America
Estim
ated
Use
r Po
pula
tion
(Tho
usan
ds)
Estim
ated
Act
ual C
onsu
mpt
ion
per
year
(Kg
Ecst
asy
Equi
v)
Wh
ole
sale
r In
com
e (U
S$m
ill)
Impl
ied
Con
sum
ptio
n pe
r us
er (g
ms
Ecst
asy
Equi
v)A
vera
ge R
etai
l Pric
e U
S$ /g
m
Ret
aile
r In
com
e (U
S$m
ill)
Dem
and
:
next largest ecstasy producing region is North America(12 mt), followed by East and South-East Asia (4 mt).The model results also suggest that Europe is the onlyregion with important ecstasy exports. More than half ofthe ecstasy produced in West and Central Europe is des-tined for export to other regions. With ecstasy pricesalmost three times the level seen in West and CentralEurope, North America seems to be a particularly lucra-tive market, but European ecstasy exports go to mostother regions as well.
Deducting seizures and losses, about 80 mt remainavailable for consumption. Using existing prevalenceestimates and applying an average rate of 10 grams perperson per year (equivalent to some 100 pills a year, ortwo pills per weekend), the largest ecstasy marketappears to be North America (33 mt), followed by Westand Central Europe (27 mt). Multiplying theseamounts with reported prices, the North Americanecstasy market appears to be substantially larger(US$8.5 bn) than the European market (less than US$3bn). However, this may change, as there are strong indi-cations that the North American ecstasy market isshrinking. As outlined in the beginning of this chapter,data used for the market calculations were those pub-lished in last year’s World Drug Report (2.7 millionecstasy users for North America). The numbers pub-lished in this year’s World Drug Report are already 15%less (2.3 million ecstasy users in North America), and –using school surveys as an early indicator for subsequenttrends in the general population - one can still expectfurther declines to take place. While the bulk of theecstasy market is in North America and Europe, 30% ofthe global ecstasy market is in other parts of the world,notably in Asia (12%), Oceania (10%) and SouthAmerica (7%).
2.4 Conclusions
This review of UNODC’s global drug market valuationhas highlighted some of the complexities involved inmaking such estimations. The technical details of themodel have not been discussed in this review. Clearlythere are still areas where estimates can be improved.Asnew information emerges, it will be incorporated intothe model. The overall figure of US$322 bn should beseen as representing reasonable order of magnitude. Asstated previously, some market estimates can be madewith more precision than others. The estimates for theopiates market (US$65 bn) and the cocaine market(US$70 bn), for example, are quite strong – becausethere is rigorous data at least on the production side.
The estimates for the ATS (US$44 bn) and the cannabisresin (US$28bn) markets are also reasonably wellgrounded; but the cannabis herb market estimate(US$113 bn), remains rather weak due to the paucity ofunderlying data.
Ideally, results from the top-down and the bottom-upapproaches should match, simply because there is nodrug consumption without production and there willbe, most probably, no drug production without ademand for drugs. This does not preclude the possibil-ity that stocks can be built-up or depleted, thus distort-ing this relationship in the short-term.
One key parameter for analysing the market from bothsides is still largely missing: the average consumptionper user. Only some vague and often contradictoryinformation is currently available, often from case stud-ies which may or may not be representative of a locality,a country or a region. This put a severe constraint onthis exercise. More systematic research on quantitiesconsumed could greatly improve the rigour of theresults.
In presenting this work in progress, UNODC shares itsunderstanding of the illicit drug markets, as well as lackof it in some areas, in order to improve the commonlevel of understanding, stimulate discussion and promptnew research to overcome existing gaps in information.
143
2. Estimating the value of illicit drug markets