Do drug policies affect cannabis markets? A natural experiment in Switzerland, 2000–10
Effectiveness bank home page. Opens new windowResearch analysis

This entry is our analysis of a study added to the Effectiveness Bank. The original study was not published by Findings; click Title to order a copy. Free reprints may be available from the authors – click prepared e-mail. The summary conveys the findings and views expressed in the study.

Links to other documents. Hover over for notes. Click to highlight passage referred to. Unfold extra text Unfold supplementary text
Copy title and link | Comment/query |

Do drug policies affect cannabis markets? A natural experiment in Switzerland, 2000–10.

Killias M., Isenring G.L., Gilliéron G. et al.
European Journal of Criminology: 2011, 8(3), p. 171186.
Unable to obtain a copy by clicking title? Try asking the author for a reprint by adapting this prepared e-mail or by writing to Dr Killias at

Studies of a 'natural experiment' in Switzerland in the 2000s suggested that the effective re-criminalisation of cannabis production and distribution did diminish availability and use of the drug. The results contradict other findings suggesting that national policies have little effect on cannabis use.

Summary A 'natural experiment' in Switzerland in the 2000s revealed the impacts of changes in the enforcement of cannabis production and distribution laws. By 2001, in response to public sentiment Switzerland had already relaxed its enforcement of laws against the use and distribution of cannabis. At this time the government prepared reforms to enshrine this in law by officially tolerating the sale, possession and use of small amounts of cannabis (usually below 5g), and the production and sale of larger quantities as long as producers and retailers agreed to act under strict control by police and the Department of Agriculture. Though this change had yet to be implemented, in anticipation over the following years visible and quasi-official structures of production, distribution and sale emerged. Concerned over some of the consequences, in 2003 and again in 2004 the Swiss parliament rejected the proposed changes. Over the following months, police and prosecutors resumed former more repressive policies, especially in respect of production and distribution. As a result, shops and production centres were closed during 2005 and 2006. It was this reversal which offered the opportunity to evaluate the impact of tolerance of legal production and distribution versus lack of tolerance.

Main findings

Early in 2004 shortly before most of their shops were closed, a survey of cannabis retailers suggested that competition between shops was quite stiff, particularly in respect of price. Nearly all felt they had to to provide excellent products and service to keep their customers. Though many said they had never sold high strength and/or smokable cannabis, this conflicted with the number of prosecutions for selling cannabis whose main active ingredient (THC) was above the legal limit.

In summer 2004 when many cannabis shops were still operating, two young men aged around 18 conducted 'test purchase' operations at 50 shops. Of these, 29 sold cannabis without reservation and 26 did so regardless of the young men's age. Usually, the fake clients asked for 5g or the quantity available for about 50 Swiss francs. The quantities actually sold generally varied between 3.8g and 6.5g and THC levels between 8% and 28%, averaging 16%. Overall, the study confirmed that minors easily obtained high-strength cannabis. Most samples contained THC close to the average of 16% and prices varied little around 11 Swiss francs per gram. In short, quality and prices were fairly well standardised.

In 2009 when all known cannabis shops had closed, a second 'test purchase' operation was conducted, but this time to test the availability of supplies on the now fully illicit market. Two young men walked through inner-city areas where police said cannabis was most available, looking for potential dealers. Over 15 afternoons they made 29 relevant contacts; during 27 they were able to obtain cannabis. All the sales took place in streets and parks. Usually the fake clients were able to spot a dealer in under 20 minutes. The quantity purchased varied far more than in 2004, ranging from 0.38 to nearly 13 grams. Equally inconsistent were prices, varying greatly between 8 and 200 francs per gram. A typical price was 28 francs. The THC content varied between 4% and 18% and averaged 12%, lower than in 2004. At every transaction, the fake clients asked whether the dealer might be able or willing to supply other substances. Only one said they could.

Compared to 2004, typical prices paid per gram had increased from 11 to 28 francs and the variability in price and quantity was much greater and THC content lower. From the relatively standardised market of 2004, by 2009 the price structure was, from the clients' point of view, relatively obscure and bore little relation to the origin or strength of the product.

The authors' conclusions

The results of our studies suggest that legal policies can strongly affect production, supply, distribution and sale of cannabis. The switch from a liberal to a more repressive policy meant that large-scale agricultural was partly replaced by small-scale production on private premises, and sales moved back from shops to the streets. Formerly an export country, illegal import of cannabis in to Switzerland resumed, though probably not enough to compensate for lost local production. For users without links to home-based production networks, availability of cannabis may have decreased substantially, probably prompting decreased consumption. However, the market and its price structure became far more variable and obscure. Prices soared, possibly reflecting reduced supply and more marginal and criminal suppliers. Street sales favoured cheating because quantities cannot be accurately weighed and suppliers had little interest in repeat sales to unknown customers, feeling little need to gain their trust. On the other hand, and contrary to a widely held view, markets for cannabis and other substances seem to have remained separated.

Surveys in Switzerland and abroad suggest that policies making cannabis more easily available were followed by increasing rates of use, whereas Switzerland's opposite policy after 2004 was associated with a drop in both the prevalence and frequency of cannabis use. Establishing to what extent policy changes caused changes in use is for the moment impossible, but data is consistent with the assumption that policies affect the availability and (indirectly) use of cannabis.

Thanks for their comments on this entry in draft to Martin Killias of the Institute of Criminology at the University of Zurich. Commentators bear no responsibility for the text including the interpretations and any remaining errors.

Last revised 06 October 2011

Open Effectiveness Bank home page

Top 10 most closely related documents on this site. For more try a subject or free text search

STUDY 2014 Drugs: international comparators

REVIEW 2011 Effect of drug law enforcement on drug-related violence: a systematic review

REVIEW 2010 Effect of drug law enforcement on drug-related violence: evidence from a scientific review

OFFCUT 2003 Criminalising drug possession in the Czech Republic did not reduce drug problems

STUDY 2005 US study finds penalties for drug dealing near schools fail to shift dealers

STUDY 2001 No increase in use after cannabis possession made a civil offence

DOCUMENT 2016 Modern Crime Prevention Strategy

REVIEW 2013 Quitting drugs: quantitative and qualitative features

STUDY 2015 Changing patterns of substance misuse in adult prisons and service responses

REVIEW 2010 Polarized drinking patterns and alcohol deregulation. Trends in alcohol consumption, harms and policy: United Kingdom 1990–2010