Research paper
Estimating the production, consumption and export of cannabis: The Dutch case

https://doi.org/10.1016/j.drugpo.2016.02.019Get rights and content

Abstract

Background

Quantifying an illegal phenomenon like a drug market is inherently complex due to its hidden nature and the limited availability of reliable information. This article presents findings from a recent estimate of the production, consumption and export of Dutch cannabis and discusses the opportunities provided by, and limitations of, mathematical models for estimating the illegal cannabis market.

Methods

The data collection consisted of a comprehensive literature study, secondary analyses on data from available registrations (2012–2014) and previous studies, and expert opinion. The cannabis market was quantified with several mathematical models. The data analysis included a Monte Carlo simulation to come to a 95% interval estimate (IE) and a sensitivity analysis to identify the most influential indicators.

Results

The annual production of Dutch cannabis was estimated to be between 171 and 965 tons (95% IE of 271–613 tons). The consumption was estimated to be between 28 and 119 tons, depending on the inclusion or exclusion of non-residents (95% IE of 51–78 tons or 32–49 tons respectively). The export was estimated to be between 53 and 937 tons (95% IE of 206–549 tons or 231–573 tons, respectively).

Conclusion

Mathematical models are valuable tools for the systematic assessment of the size of illegal markets and determining the uncertainty inherent in the estimates. The estimates required the use of many assumptions and the availability of reliable indicators was limited. This uncertainty is reflected in the wide ranges of the estimates. The estimates are sensitive to 10 of the 45 indicators. These 10 account for 86–93% of the variation found. Further research should focus on improving the variables and the independence of the mathematical models.

Introduction

Up until the 1980s, the Netherlands was importing cannabis, mainly in the form of hashish, from countries such as Morocco, Pakistan, Afghanistan, Lebanon, Columbia and India. The majority of cannabis consumed in Europe originated from the Moroccan Rif Mountains and the border regions of Pakistan, Afghanistan and Iran, where cultivation increased explosively since the 1960s to meet the growing European demand (Afsahi, 2011, Weijenburg, 1993). It wasn’t until the 1980s that cannabis cultivation emerged in the Netherlands and this ‘import substitution’ became a topic of particular interest to Dutch scholars.

The emergence and increase of the cultivation of cannabis in the Netherlands is attributed to several factors, including the development of new cultivation techniques and varieties of cannabis (Decorte and Boekhout van Solinge, 2006, Spapens et al., 2014), national and international efforts to counter the smuggling of hashish to the Netherlands (Korf, 2006), and the ‘tolerant’ Dutch attitude toward the sale of cannabis through coffee shops (Jansen, 1993, Fijnaut et al., 1998; Spapens et al., 2014). Consequently, the Netherlands has developed into a significant producer of cannabis over the past three decades (Korf, 2011, EMCDDA, 2013).

Though it is clear that the Netherlands has become a large-scale producer of cannabis, questions continuously arise as to the actual size of the illegal Dutch soft drug market, and how much of domestic production is exported. Estimates of the size of the domestic market and the export also play a central role in the current political debate pertaining to the regulation of the cultivation of cannabis for coffee shops in the Netherlands. The Minister of Security and Justice in part bases his decision to not allow the regulation of cannabis cultivation for Dutch coffee shops on export estimates, stating that if a lot of the cannabis cultivated in the Netherlands is intended for export, regulating the cultivation for Dutch coffee shops will only have a limited effect on the total cultivation. Furthermore, illegal cultivation would continue to exist and regulated cultivation would be ineffective in combatting organized crime related to the cultivation of cannabis (see for example Tweede Kamer, 2014a, Tweede Kamer, 2014b).

However, the information needed to quantify the cultivation, consumption, and export of Dutch cannabis is limited and uncertain due to the clandestine nature of illegal markets. This article presents findings from a recent estimate of the production, consumption and export of cannabis for the case of the Netherlands and discusses the opportunities provided by, and limitations of, mathematical models for estimating the illegal cannabis market (Van der Giessen, Van Ooyen-Houben, & Moolenaar, 2014).

Section snippets

Data collection

Because limited information is available on the Dutch cannabis market due to its clandestine nature, the models are data driven. Where data for a certain variable were not available, substituting in other variables or combinations of variables approximated the value. Unfortunately, this means that a fairly large number of variables are required for the estimates. To validate the data, we not only searched for the most recent data, we also assessed the uncertainty inherent in the data.

The

Data analysis

We recorded the lowest and the highest reported values we could find for each variable. With these, we calculate the lowest and the highest possible outcomes for the production, consumption and export models. These constitute the lower and upper bounds of our interval estimates. Furthermore, we perform a Monte-Carlo simulation (MC) by drawing 100,000 random values between the lowest and highest value of each input variable. This yields a range in which 95% of the estimates lie. This approach

Modelling the export of Dutch cannabis

The mathematical model to estimate the production, consumption, and export of cannabis developed by Jansen (2012) was used as a starting point and expanded upon with the models used in other state of the art, national and international estimates (Bouchard, 2008, Carpentier et al., 2012, Hakkarainen et al., 2011Kilmer and Pacula, 2009, Van der Heijden, 2006, Van Laar et al., 2013a, Van Laar et al., 2013b, Van Laar et al., 2013c, Van Laar et al., 2013d). This resulted in several alternative

Production

The annual production of Dutch cannabis is estimated to be between 171 and 965 tons depending on which model is used (see Fig. 1). This is a very wide range; the upper limit is roughly 5.6 times the lower limit. The width of the range is indicative of the great uncertainty inherent in available information. The MC simulation revealed that 95% of the estimates lie between 271 and 613 tons, depending on the model used.

Consumption

The annual consumption of Dutch cannabis in the Netherlands is estimated to be

Possibilities and limitations of the research

The availability and reliability of the data required for the cannabis production, consumption and export estimates proved to be limited and, as a consequence, we had to rely on many assumptions and inter/extrapolations. This uncertainty is reflected in the wide ranges of the estimates. The estimates were also limited at the model level; it was not possible to triangulate outcomes using only independent models as a number of variables – in particular the ones regarding electricity loss and

Acknowledgements

This work is based on a study performed at the request of the Dutch Minister of Security and Justice by the Scientific Research and Documentation Centre. A special thanks to P. Jans-Rat, J.A.F. Jespers, E.W. Kruisbergen, N. Maalsté, R.F. Meijer, A.W.M. van der Heijden, W. Vanhove and M. Wouters who have contributed to the data collection and have partaken in the expert panel to review and, where necessary, adjust variables. We are also grateful to the guidance committee, D. van de Mheen

References (91)

  • B. Bieleman et al.

    Coffeeshopbezoekers in Venlo 2009. Tellingen en enquête coffeeshopbezoekers Venlo

    (2009)
  • BOOM, Openbaar Ministerie

    Wederrechtelijk verkregen voordeel hennepkwekerij bij binnenteelt onder kunstlicht. Standaardberekening en normen

    (2005)
  • BOOM, Openbaar Ministerie

    Wederrechtelijk verkregen voordeel hennepkwekerij bij binnenteelt onder kunstlicht. Standaardberekening en normen. Update 1 November, 2010

    (2010)
  • M. Bouchard

    A capture–recapture derived method to estimate cannabis production in industrialized countries

    (2007)
  • M. Bouchard

    Toward a realistic method to estimate cannabis production in industrialized countries

    Contemporary Drug Problems

    (2008)
  • J. Cervantes

    Marijuana horticulture. The indoor/outdoor medical grower's bible

    (2006)
  • COT

    Ervaren coffeeshopgerelateerde overlast in Maastricht & coffeeshopbezoek in Maastricht, Heerlen, Kerkrade en Sittard-Geleen. Gebundelde resultaten nulmeting 2011

    (2011)
  • T. Decorte

    Small scale domestic cannabis cultivation: An anonymous web survey among 659 cannabis cultivators in Belgium

    (2008)
  • T. Decorte et al.

    Het aanbod van cannabis in Nederland en België

    Tijdschrift voor Criminologie

    (2006)
  • T. Decorte et al.

    Cannabisteelt in Vlaanderen. Patronen en motieven van 748 telers

    (2007)
  • District Breda

    Convenant districtelijke aanpak hennepkwekerijen. Resultaten 2009–2011

    (2011)
  • Domeinen Roerende Zaken

    Beleidsdoorlichting Domeinen Roerende Zaken

    (2013)
  • N. Dorn

    Performance management, indicators and drug enforcement, in the crossfire or at the crossroads?

  • EMCDDA

    A cannabis reader: Global issues and local experiences. Perspectives on cannabis controversies, treatment and regulation in Europe. EMCDDA Monographs 8(1)

    (2008)
  • EMCDDA

    Cannabis production and markets in Europe. EMCDDA Insights 12

    (2012)
  • EMCDDA

    Case study 1. Cannabis: Changing demand and an increase in domestic production

    (2013)
  • EMCDDA

    New developments in Europe's cannabis market. Perspectives on drugs

    (2014)
  • I. Emmett et al.

    Het groene goud. Verslag van een onderzoek naar de cannabissector voor het Nationaal dreigingsbeeld criminaliteit met een georganiseerd karakter

    (2008)
  • M. Fendrich et al.

    The utility of drug testing in epidemiological research: Results from a general population survey

    Addiction

    (2004)
  • C. Fijnaut

    Een drugsbeleid in de revisie: een nuchtere kijk op de Hollandse coffeeshop

    (2009)
  • C. Fijnaut et al.

    Organized crime in the Netherlands

    (1998)
  • C. Fijnaut et al.

    Voor een gezamenlijke beheersing van de drugsgerelateerde criminaliteit in de Euregio Maas-Rijn

    (2008)
  • G. Green

    The Cannabis Grow Bible. The definitive guide to growing marijuana for recreational and medicinal use

    (2001)
  • P. Hakkarainen et al.

    Small-scale cannabis growing in Denmark and Finland

    European Addiction Research

    (2011)
  • Inspectie Openbare Orde en Veiligheid, & Rijksauditdienst

    Afhandeling van in beslag genomen drugs. Fase 1 – Randvoorwaarden voor verbetering

    (2009)
  • A.C.M. Jansen

    Hasj-coffee shops als experiment

    Justitiële Verkenningen

    (1993)
  • F. Jansen

    Criminaliteitsbeeldanalyse 2012: Georganiseerde cannabisteelt

    (2012)
  • A. Kepper et al.

    Middelengebruik bij jongens in Justitiële Jeugdinrichtingen. Het gebruik van tabak, alcohol, cannabis en harddrugs bij jongens met en zonder PIJ-maatregel

    (2009)
  • B. Kilmer et al.

    Estimating the size of the global drug market. A demand-side approach. Report 2

    (2009)
  • KLPD, Dienst Landelijke Informatie Organisatie

    Hennepkwekerijen 2013

    (2014)
  • D.J. Korf

    De normalisering van cannabisgebruik

    Justitiële Verkenningen

    (2006)
  • D.J. Korf

    Marihuana behind and beyond coffee shops

  • D.J. Korf et al.

    De illegale gebruikersmarkt

  • D.J. Korf et al.

    De illegale gebruikersmarkt van cannabis

  • Cited by (4)

    • Triangulating web & general population surveys: Do results match legal cannabis market sales?

      2019, International Journal of Drug Policy
      Citation Excerpt :

      The first, sometimes referred to as “demand-side estimates”, starts with estimates of numbers of users (e.g., Wilkins, Bhatta, & Casswell, 2002; Legleye et al., 2008; Kilmer et al., 2013; van Laar et al., 2013; Giommoni, 2014). The second starts with measures pertaining to supply, such as acreage under cultivation (e.g., Bouchard, 2008; Kalacska & Bouchard, 2011; van der Giessen, van Ooyen-Houben, & Moolenaar, 2016). Since the supply-side estimates were so strong, being rooted in Washington’s regulatory tracking system, contrasting them shed some light on the effectiveness of efforts to combine GPS and web-based surveys.

    • A digital twin reference architecture for pharmaceutical cannabis production

      2023, International Journal of Computer Integrated Manufacturing
    View full text