1 Introduction

The size of a market is based on factors influencing both demand and supply. Changes in market size, therefore, provide
valuable information about the net effects of movements in both parts of the market. For example, while the number of
users may decrease in response to a prevention policy targeting initiation, total expenditures may simultaneously increase due
to factors shifting more light users into heavy use or an increase in supply. Therefore, estimating the size of the market, in
terms of both participants and expenditures, is critical to fully understanding the impact of interventions intended to influence
demand and/or supply.
Further, understanding the size of the market for specific illicit drugs is critical for improving government decision-making and
evaluating alternative policy approaches. On one hand, knowing how much revenue is generated for different substances
within a country can help decision makers target enforcement resources. On the other hand, knowing the size of a market
is necessary but not sufficient for projecting the revenue from a legalization and tax regime. Information about drug markets
may also be used to guide decisions in other policy areas. For example, Reuter & Greenfield (2001) suggest that before
September 11, 2001, the focus on international money laundering controls was largely based on what was known about the
size of the international drug trade. Additionally, understanding the magnitude of the opium trade in Afghanistan and how
it has changed may improve military strategies for addressing opium-funded insurgents.
The goal of this report is to generate country-level consumption and retail expenditure estimates for cannabis, heroin, cocaine,
and amphetamine-type substances. Unfortunately, most of the information required for such an effort is unavailable and the
data that do exist are often not comparable across countries and time. This confines researchers to simplifying assumptions
that make it easy and appropriate to question the validity of the results. It also means that most of the focus is on countries
with well-developed data collection systems.
There are a variety of methods for calculating the size of an illicit drug market. The supply-side approach uses estimates about
production and how much is seized or lost on the way to its final destination. Combining these figures with information about
prices generates estimates of the total size of the market. There are at least two different methods on the demand side. One
is based on self-reported information about what individuals spend on illicit drugs, and the other uses prevalence estimates
and combines them with assumptions about quantity consumed and retail prices to generate expenditure estimates. Each
method has its own advantages and drawbacks, but in most cases the decision regarding which approach to use is a practical
one determined by the available data for the market considered. It is important to note that the methods are not mutually
exclusive and ideally multiple methods could be used to try to triangulate available information from each, as has been done
in previous attempts to measure the size of the drug market (e.g., Abt, 2001; UNODC, 2005).
Given the objective to estimate the size of the drug market for individual countries, this report adopts the prevalence-based
approach for calculating country-specific consumption and retail expenditure estimates. The focus is on a handful of readily
available parameters and evidence-based assumptions about quantity consumed to generate estimates that are reasonably
close to what is available in the peer–reviewed and grey literatures. This approach may prove most insightful for developed
countries for which drug data are relatively scarce or where efforts are currently under way to collect information, as it could
guide them on what type of information is necessary for constructing a similar estimate. This approach remains hampered by
the lack of information about typical quantities consumed, so it is necessary to draw on a broad array of sources about drug
user behaviour and evidence-based assumption to fill in gaps using this method.
This report contributes to the literature on sizing drug markets in at least four different ways. First, it presents country-specific
estimates for countries which account for the major share of consumption and/or retail expenditures for cannabis, heroin,
cocaine, and amphetamine-type substances (ATS). Previous studies either provide expenditure estimates for different regions
of the world or for a specific country. With respect to the latter, many of these studies only include cannabis. Second, it presents
most results in term of ranges, not just point estimates. In doing so, it enables readers to better understand the uncertainty
associated with generating any point estimate for these markets. Third, it presents statistics from a variety of international
data sources (published and unpublished) that should be useful to other researchers in this field. Fourth, throughout the text
insights are given regarding data elements that could be improved to generate a better understanding of global consumption
and retail expenditures. As better data are collected, there should be less reliance on controversial assumptions.

Given the popularity of cannabis across the globe, there is relatively more information available about cannabis prevalence
and consumption patterns. Thus, more confidence can be placed in these estimates than those for the other drug markets.
Furthermore, direct comparisons of results for cannabis can be made to those by other researchers given growing number
of studies which focus on the size of the retail cannabis market in specific countries (e.g., Bramley-Harker 2001; ABT, 2001;
Wilkins et al. 2002; Wilkins et al. 2005; Clements & Zhao, 2005; Pudney et al. 2006; Gettman, 2007; Hakkarainen et al.,
2007; Legleye et al., 2008). The general comparability of findings across studies provides additional confidence that the results
generated for cannabis here are indeed reasonable. Because of the lack of data and the stigma associated with self-identifying
as a cocaine user in surveys, less confidence can be placed in our best cocaine estimates for Europe. The large differences in
our low and high estimates reflect this uncertainty, and it is imperative that efforts be made to improve the available data
given the growth in European cocaine use in recent years (EMCDDA, 2007b).
It is also important to note that this report does not provide country-specific estimates for every country in the world. Such an
effort would be impossible given the relatively poor data collection in some countries. Nonetheless, estimates are generated
for those countries that represent the major share of consumption and/or retail expenditures for each substance. Since retail
prices are larger in developed countries (and hence the currency value of the market is larger), most of the focus is on Europe,
North America, and Oceania.
The report proceeds as follows. The second section discusses some of the methodological issues associated with using
demand-side estimates to generate consumption and expenditure figures. The third section focuses on the retail market for
cannabis and is followed by the markets for heroin, cocaine, and amphetamine-type substances. The final section discusses
some of the results and ideas for obtaining information that would improve these estimates.1
1 We also include a brief section on farm-gate and international trade values for cocaine and opiates in Annex 1.