Decriminalise, tolerate or regulate without profit: the Netherlands, Portugal, Uruguay and Germany

Debate about cannabis reform in Britain tends to picture two options: the status quo, or American-style commercial dispensaries. Between those poles lies a range of models tried, some for decades, by countries much closer to home. The Netherlands has tolerated small-scale retail sales since 1976. Portugal stopped treating the possession of any drug as a crime in 2001. Uruguay created the world's first national legal cannabis market in 2013, but deliberately refused to let it become a commercial one. Germany legalised possession and home growing in 2024 and channelled supply through non-profit associations rather than shops.

None of these countries chose a for-profit industry, and each made a different bet about which part of the system to change. The Netherlands relaxed the law at the shop counter but not behind it. Portugal changed the law for users but not for suppliers. Uruguay and Germany legalised supply, but kept it in the hands of the state, of growers' collectives or of individuals tending a few plants at home. Comparing them is one of the best ways to separate two questions that public debate tends to run together: what happens when you stop punishing people who use cannabis, and what happens when you change how cannabis is supplied?

This is the third article in our series on prohibition and legalisation. The first sets out the range of legal regimes, and the second weighs the record of criminal prohibition against its own aims. Here we examine the partial and non-commercial reforms; the next article turns to the commercial markets of the United States and Canada. Like the rest of the series, it expands a critical comparative review and gives the strongest version of each side's case before weighing it.

Four ways of loosening the law

The literature on cannabis policy distinguishes at least six legal regimes, and the differences between them matter more than the labels "legal" and "illegal" suggest (Room et al., 2010; Rolles and Murkin, 2016). The four countries in this article fall into three of those categories.

The table shows why findings cannot simply be carried from one country to another. Findings about Portuguese decriminalisation cannot be transferred to a commercial market such as Colorado's, and the reverse is also true (Kilmer, 2017; Hall and Lynskey, 2020). Each of the four is a different experiment, and each answers a different question.

It also helps to be clear about what "evidence" means here. The most persuasive studies of drug policy try to separate the effect of a law from everything else that was happening at the same time. Cannabis use was rising in many countries before any reform, people become more willing to admit to using a drug once it is legal, the product itself has grown stronger over the decades, and most reforms are recent (Freeman et al., 2021; Caulkins, 2024). These hazards cut both ways: they weaken the claim that a reform was harmless as much as the claim that it was disastrous.

The Netherlands: tolerated retail since 1976

What the Dutch did

The Netherlands has tolerated small-scale retail sales through licensed coffeeshops since the 1976 revision of its Opium Act. Possession for personal use remains formally illegal but is not prosecuted, and the shops' sales are tolerated. Production and wholesale supply, however, remain illegal.

That combination produces the regime's most famous anomaly, usually called the "back-door problem". A coffeeshop sells legally, or at least with official tolerance, at its front door. But the cannabis it sells has to come from somewhere, and since nobody may lawfully grow it or sell it in bulk, it must be bought from criminal suppliers at the back.

What happened to use

One influential early assessment divides the Dutch experience into two phases. MacCoun and Reuter (2001) found that the initial period of depenalisation, in which users were no longer prosecuted, was not followed by a rise in use. The later period, during the 1980s and early 1990s, was different: the coffeeshop sector grew and promoted itself commercially, and that period coincided with increased use among young adults.

The word coincided matters. MacCoun and Reuter's finding is a pattern in time, not a controlled experiment, and other changes in Dutch society could have contributed. But the contrast between the two phases is suggestive. When the law stopped punishing users, use did not rise. When a visible, growing and promotional trade developed, it did. Changing the rules for users is not the same as changing the market.

A decade later, MacCoun (2011) returned to the question and concluded that Dutch prevalence remained moderate by international standards, but that the system's toleration of criminal supply was its central flaw. The international comparison is supported by wider survey work. In the World Health Organization's World Mental Health Surveys, the United States, with some of the most punitive drug laws in the developed world, had higher rates of lifetime cannabis use than the Netherlands (Degenhardt et al., 2008). The authors concluded that drug use does not appear to be related in any simple way to drug policy.

The case for and against

The strongest case for the Dutch model is that for half a century the Netherlands has tolerated open retail sale while its prevalence has remained moderate by international standards (MacCoun, 2011), and lower on lifetime measures than in the far more punitive United States (Degenhardt et al., 2008). Users are not prosecuted for possession. And buyers can deal with a shop rather than with a dealer who may also sell other drugs.

The strongest case against it is equally serious. First, the back door keeps organised crime at the centre of the supply chain. Second, toleration gives the state little control over the product. A case-control study compares people who have a condition with similar people who do not, and asks how their past exposures differ. One such multicentre study of first-episode psychosis across Europe and Brazil found that, assuming causality, 50.3% of first-episode psychosis cases in Amsterdam would be prevented if high-potency cannabis were unavailable, against 30.3% in London (Di Forti et al., 2019). That is not a verdict on Dutch law as such: high-potency products were also widespread in London under prohibition, and the study cannot by itself prove causation. But tolerating sales evidently did not, on its own, protect buyers from the most harmful products. Third, the commercial phase of the 1980s and early 1990s coincided with rising use among young adults (MacCoun and Reuter, 2001), which suggests that promotion carries a cost.

The controlled supply chain experiment

The government's controlled supply chain experiment, in which licensed growers supply coffeeshops in a limited number of municipalities, is designed to test whether the back-door problem can be removed. The results were not available when the review on which this article is based was written, so it would be premature to say whether the experiment has worked.

Assessment. The Dutch case suggests that removing criminal penalties for users need not increase use, but that commercial promotion probably does. It also shows that half-legalisation, which legalises sales but not supply, leaves organised crime in place.

Portugal: decriminalisation of all drugs since 2001

What Portugal did

On 1 July 2001 Portugal made possession of up to ten days' supply of any drug an administrative rather than criminal matter. The reform did not legalise drugs. Selling them remained a crime, and there was no legal market of any kind. What changed was the way the state treated the person found with a small quantity for their own use. In general terms, an administrative matter is handled outside the criminal courts and does not lead to a criminal conviction. Portugal paired the change with investment in treatment and harm reduction.

Where the Netherlands tolerated a retail trade in one drug, Portugal tolerated no trade at all but stopped prosecuting users of every drug, from cannabis to heroin.

How the evidence was gathered

The best-known evaluation, by Hughes and Stevens (2010), compared Portugal with Spain and Italy. The logic of such a comparison is simple but important. Drug use rises and falls for many reasons, including fashion, prices and the economy. If one simply compared Portugal before and after 2001, any change could be due to those background forces rather than to the law. Comparing Portugal with its neighbours over the same period gives a rough sense of the background trend, and so of what might have happened in Portugal anyway. It does not offer the precision of a formal quasi-experimental design, but it guards against the most obvious error, which is to attribute every post-reform change to the reform.

What they found

Hughes and Stevens (2010) found that, contrary to predictions, decriminalisation did not lead to major increases in drug use. Adult use rose modestly, in line with neighbouring countries. Meanwhile problematic use, drug-related deaths, HIV infections and prison overcrowding fell.

A modest rise in adult use that tracks the rise in Spain and Italy is what one would expect if the reform made little difference to how many people used drugs. The falls, meanwhile, were in the harms that a health-oriented reform was designed to influence.

The case for and against

The strongest case for the Portuguese model is that it achieved the central aim of decriminalisation, keeping people who use drugs out of the criminal justice system, without the surge in use that opponents had feared. It did so for every drug, including heroin. If ending criminal penalties for heroin users did not lead to a major increase in drug use, the argument goes, it is hard to see why ending them for cannabis users should.

The strongest case for caution is not that the reform failed but that its lessons travel less easily than its admirers suggest. Two caveats stand out. First, the Portuguese reform concerned all drugs and was driven mainly by heroin, so its lessons for cannabis are indirect. Several of the outcomes on which Portugal improved, such as HIV infections, belong to a heroin problem rather than a cannabis one. Second, its success depended on sustained health spending. The treatment and harm reduction investment was part of the package, and decriminalisation without that investment is a different policy. A further caution is that adult use did rise, even if modestly and in line with neighbouring countries; the absence of a major increase is not the absence of any increase.

Finally, because supply remained criminal, decriminalisation left the illicit market and the potency of what it sold untouched. The reform never set out to regulate supply, so this is not a failure on its own terms, but it means Portugal answers only half of the policy question.

Assessment. Portugal is the strongest evidence that ending the criminalisation of users does not by itself increase use. Its lessons for cannabis are indirect, because the reform covered all drugs and was driven mainly by heroin, and its success depended on sustained health spending.

Uruguay: a non-commercial state model since 2013

What Uruguay did

Uruguay's Law 19.172 of December 2013 created the first national legal market for cannabis, but deliberately not a commercial one. Adults who register with the state may obtain cannabis through one of three channels:

• home growing;
• membership clubs;
• pharmacy sales of state-licensed product, which began only in July 2017.

Advertising is banned and potency is capped. In the terms of the regulatory toolkit, Uruguay combined registration, product controls, a marketing ban and an ownership model that keeps private profit out of retail (Room et al., 2010; Rolles and Murkin, 2016).

Uruguay's reform also put it at odds with the international drug control system. The International Narcotics Control Board has repeatedly stated that Uruguay's law, like Canada's, is inconsistent with treaty obligations under the 1961 Single Convention, which restricts cannabis to medical and scientific purposes. The international regime is increasingly out of step with national practice, a gap that legal scholars and reform states have not yet resolved (Bewley-Taylor, Blickman and Jelsma, 2014). That tension is discussed in the final article of the series.

Did adolescent use change? Two kinds of study

The commonest fear about legalisation is that it will draw young people into cannabis use. Uruguay has been studied with two strong designs.

The first is the synthetic control method. There is only one Uruguay, and no identical twin that did not legalise, so the method builds one. It takes a pool of comparison countries and combines them, giving each a weight, so that the weighted combination tracks Uruguay's own figures as closely as possible in the years before the reform. That combination, a synthetic Uruguay, is then followed after the reform. If the real Uruguay diverges from its synthetic twin after 2013, the gap is a reasonable estimate of the reform's effect; if the two continue together, the reform probably made little difference.

Using this approach, Laqueur et al. (2020) found no evidence that legalisation affected adolescent cannabis use or the risk young people perceived in using it. One measure did move: perceived availability, the share of young people who thought cannabis would be easy to obtain, rose. It stood at 58% in the real Uruguay against 51% in the synthetic comparison.

The second design is difference-in-differences. Here the comparison is with a single real neighbour. The researchers measure how much an outcome changed in the country that reformed, measure how much it changed over the same period in a comparison country that did not, and treat the difference between those two changes as the effect of the reform. It assumes that, without the reform, the two countries would have followed parallel trends. Rivera-Aguirre et al. (2022) compared Uruguayan and Chilean secondary school students from 2007 to 2018. They found no association between legalisation and past-year, past-month, risky or frequent use.

Two studies with different methods and different comparisons reaching the same answer is about as good as policy evidence gets. It matches US quasi-experimental findings of no average increase in adolescent use (Anderson et al., 2019), discussed in the next article.

Two caveats should be kept in view. First, follow-up is still short. Pharmacy sales, the channel most like an ordinary shop, began only in July 2017, and the Uruguay–Chile comparison ends in 2018, so that study captures little more than a year of the full system. Second, the rise in perceived availability is a reminder that legalisation changes the environment young people grow up in, even when it has not yet changed their behaviour. Whether perceived availability leads, over a longer period, to higher use is not something these studies can answer.

The grey market

The most important unintended effect emerged from qualitative research, which seeks to understand how a system works in practice rather than to count outcomes. A qualitative study ten years after the reform identified a "grey market", in which legally produced cannabis is redistributed illegally (Queirolo et al., 2024). The authors attribute it partly to restrictive access and to an implementation gap between the law as written and the system as it operates.

In plain terms, restrictions meant to prevent commercialisation also made legal access harder, and where legal supply falls short, some legally grown product leaks to people outside the system, beyond the controls the law was designed to impose.

The case for and against

The strongest case for Uruguay is that it built a national legal market that bans advertising, caps potency and keeps profit out of retail, and that two independent evaluations found no effect on adolescent use (Laqueur et al., 2020; Rivera-Aguirre et al., 2022).

The strongest case against is that a regime designed to avoid the harms of commercialisation produced the opposite problem: legal supply that was too limited and too slow to arrive. The pharmacy channel opened more than three years after the law was passed, and ten years on, restrictive access and an implementation gap had left space for a grey market (Queirolo et al., 2024). Critics also point to the rise in perceived availability among adolescents as a possible early warning, and to the tension with the international treaties.

Assessment. Uruguay offers the strongest evidence that legalisation without commercial promotion can avoid increases in youth use. Its weakness is the reverse of the US problem: supply that is too restricted leaves room for informal markets.

Germany: the "two-pillar" model since 2024

What Germany did

Germany's Cannabis Act (CanG) came into force on 1 April 2024. It permits adults to possess limited quantities, to grow up to three plants at home and, from July 2024, to join non-profit cultivation associations. The reform was described as a "two-pillar" model. The first pillar is the one just described: decriminalised possession, home growing and associations. A planned second pillar of commercial pilot projects has not materialised, so only the non-commercial pillar exists so far.

The law requires independent evaluation, carried out by the EKOCAN consortium of the universities of Hamburg, Düsseldorf and Tübingen. Two interim reports have been published. The final evaluation is not due until 2028.

The first interim report

The first interim report estimated that 5.3 million adults used cannabis in 2024 (EKOCAN, 2025). Three of its findings are central:

• registered cannabis-related offences fell sharply, which is the most direct effect of decriminalising possession;
• the legal medical supply covered only about 12–14% of total demand;
• use among young adults aged 18–25 rose, but use among minors did not.

The general hazards described earlier apply to the last finding. A rise among young adults in the first year could reflect the reform, a pre-existing trend, or a greater willingness to report use now that it is lawful, and a single year cannot settle which. Critics are nonetheless entitled to emphasise it.

The second interim report

The second interim report, published on 1 April 2026, showed how supply was actually reorganising itself (EKOCAN, 2026). The share of users growing their own cannabis as their main source rose from 5.4% in the first half of 2024 to 21.4% in the second half of 2025. Only about 3.5% obtained cannabis from associations.

Home growing thus grew roughly fourfold, while the associations reached only a small fraction of users. And a large share of demand appears to have shifted to medical cannabis obtained through online prescriptions, a channel outside the two pillars the reform was built on. The government moved to restrict it with a draft amendment to the Medical Cannabis Act in October 2025.

The case for and against

The strongest case for the German reform is that registered offences fell sharply without any rise in use among minors (EKOCAN, 2025), that people actually used the legal route it created, with more than a fifth relying mainly on their own plants by late 2025 (EKOCAN, 2026), and that it built independent evaluation into the law from the start.

The strongest case against is that the non-commercial channels were too small to do the job asked of them. Associations supplied only about 3.5% of users (EKOCAN, 2026). Much of the demand that did not go to home growing appears to have gone either to the illicit market or to online medical prescriptions, a route that, critics argue, blurs the line between medical and non-medical use, and that the government itself has moved to restrict. Use rose among young adults (EKOCAN, 2025). And the commercial pillar never arrived, so the model originally proposed has not been tested.

Assessment. Germany has achieved decriminalisation of users quickly, but its non-commercial supply channels are too small to displace the illicit market. Its experience suggests that a legal regime must provide accessible supply if it is to compete with criminal sellers. Because the final evaluation is not due until 2028, conclusions remain provisional.

The cross-cutting lesson: supply that is too restricted

Set side by side, the four countries tell a consistent story with two parts.

The first part concerns users. In each case, removing or relaxing criminal penalties for possession was not followed by a surge in use. The early Dutch period of depenalisation was not followed by a rise in use (MacCoun and Reuter, 2001). Portugal saw only a modest rise in adult use, in line with its neighbours (Hughes and Stevens, 2010). Germany saw a sharp fall in registered offences and no rise among minors, although use among young adults did rise (EKOCAN, 2025). The review on which this series is based concludes that removing criminal penalties from users is the best-supported reform in the literature.

The second part concerns supply, and here the picture is less comfortable. Each model left a gap through which illicit or informal markets continued to operate.

This is the mirror image of the problem in commercial markets. In a for-profit market, the most profitable customers are the heaviest users, and an industry has incentives similar to those of alcohol and tobacco: to increase frequent use, lower prices and lobby against restrictions (Kilmer, 2017). The US experience has been followed by falling prices, rising potency and more daily use among adults (Hall and Lynskey, 2020). The non-commercial models in this article avoided that marketing pressure, but Uruguay's state model and Germany's associations under-supplied the market, leaving space for grey and illicit sellers (Queirolo et al., 2024; EKOCAN, 2026).

Policy design therefore faces a genuine trade-off. A regime that wants to displace criminal markets needs legal supply that is accessible, affordable and sufficient. A regime that wants to avoid promoting heavy use needs to restrain the commercial forces that make supply accessible, affordable and plentiful. Too much commerce, and the industry grows heavy use. Too little supply, and the criminal or grey market keeps its customers. The article on commercial legalisation examines the first half of that trade-off; this article has documented the second.

None of the four countries has shown where the point of balance lies. The review's own suggested model combines legal supply sufficient to compete with illicit sellers, strict marketing restrictions and public or non-profit control of at least part of the supply chain, and it notes that each element has some empirical support, though no jurisdiction has yet combined all of them. The lesson of the non-commercial models is that a design admirable on paper but too small or slow in practice will be undercut by sellers who face none of its rules.

What this means for the United Kingdom

The United Kingdom has not adopted any of these models. It remains a prohibition state for non-medical use, with a medical scheme since November 2018 under which almost all prescribing takes place in private clinics. In the UK, cultivation and possession of cannabis remain criminal offences, and growing the plant requires a Home Office licence even for patients with a prescription. This article is analytical and gives no guidance on obtaining, growing or using cannabis.

Still, the comparisons matter for British debate. The finding that removing criminal penalties from users does not by itself increase use comes from Europe rather than North America. The review's suggested priorities for the UK, including ending the criminalisation of personal possession, would remain consistent with the international conventions. And the German final evaluation, due in 2028, and the Dutch controlled supply experiment test non-US models in European settings close to the UK; neither has yet reported in full.

What we know, and what we don't

What is reasonably well established

• Removing criminal penalties from people who use drugs did not, in Portugal or in the early Dutch period, lead to major increases in use (Hughes and Stevens, 2010; MacCoun and Reuter, 2001). In Germany, registered offences fell sharply and use among minors did not rise (EKOCAN, 2025).
• In Uruguay, two quasi-experimental studies using different methods found no effect of non-commercial legalisation on adolescent use (Laqueur et al., 2020; Rivera-Aguirre et al., 2022).
• Tolerating retail sale without legalising production leaves organised crime in the supply chain (MacCoun, 2011).
• Non-commercial supply channels in Uruguay and Germany have been too restricted to displace informal and illicit markets (Queirolo et al., 2024; EKOCAN, 2026).

What is plausible but less certain

• That commercial promotion, rather than the relaxation of penalties, explains the rise in Dutch young-adult use in the 1980s and early 1990s. The finding is a coincidence in time, and other factors may have contributed (MacCoun and Reuter, 2001).
• That Portugal's results would carry over to a cannabis-only reform, or to a country that did not match the change with sustained health spending.

What we do not yet know

• Whether the Dutch controlled supply chain experiment has removed the back-door problem; its results were not available at the time of writing.
• Whether the rise in perceived availability among Uruguayan adolescents will eventually translate into higher use. Follow-up is short, especially for the pharmacy channel that opened in July 2017.
• Whether the rise in use among German young adults reflects the reform, a pre-existing trend or greater willingness to report use. The German evaluation is incomplete, and its final report is not due until 2028.
• Where the point of balance lies between supply generous enough to compete with criminal sellers and restraint strict enough to avoid promoting heavy use. No jurisdiction has yet shown it.

This article is based on the critical review it expands; figures are reported as that review gives them and were not re-checked against every original study.

Sources

Anderson, D.M., Hansen, B., Rees, D.I. and Sabia, J.J. (2019) 'Association of marijuana laws with teen marijuana use: new estimates from the Youth Risk Behavior Surveys', JAMA Pediatrics, 173(9), pp. 879–881. doi:10.1001/jamapediatrics.2019.1720.

Bewley-Taylor, D., Blickman, T. and Jelsma, M. (2014) The Rise and Decline of Cannabis Prohibition: The History of Cannabis in the UN Drug Control System and Options for Reform. Amsterdam and Swansea: Transnational Institute and Global Drug Policy Observatory.

Caulkins, J.P. (2024) 'Changes in self-reported cannabis use in the United States from 1979 to 2022', Addiction. doi:10.1111/add.16519.

Degenhardt, L., Chiu, W.T., Sampson, N., Kessler, R.C., Anthony, J.C. et al. (2008) 'Toward a global view of alcohol, tobacco, cannabis, and cocaine use: findings from the WHO World Mental Health Surveys', PLoS Medicine, 5(7), e141. doi:10.1371/journal.pmed.0050141.

Di Forti, M., Quattrone, D., Freeman, T.P., Tripoli, G., Gayer-Anderson, C., Quigley, H. et al. (2019) 'The contribution of cannabis use to variation in the incidence of psychotic disorder across Europe (EU-GEI): a multicentre case-control study', The Lancet Psychiatry, 6(5), pp. 427–436. doi:10.1016/S2215-0366(19)30048-3.

EKOCAN (2025) Evaluation des Konsumcannabisgesetzes (EKOCAN): 1. Zwischenbericht. Hamburg, Düsseldorf and Tübingen: EKOCAN consortium. Available at: https://www.uke.de/landingpage/ekocan/.

EKOCAN (2026) Evaluation des Konsumcannabisgesetzes (EKOCAN): 2. Zwischenbericht, 1 April 2026. Hamburg, Düsseldorf and Tübingen: EKOCAN consortium. Available at: https://www.uke.de/landingpage/ekocan/.

Freeman, T.P., Craft, S., Wilson, J., Stylianou, S., ElSohly, M., Di Forti, M. and Lynskey, M.T. (2021) 'Changes in delta-9-tetrahydrocannabinol (THC) and cannabidiol (CBD) concentrations in cannabis over time: systematic review and meta-analysis', Addiction, 116(5), pp. 1000–1010. doi:10.1111/add.15253.

Hall, W. and Lynskey, M. (2020) 'Assessing the public health impacts of legalizing recreational cannabis use: the US experience', World Psychiatry, 19(2), pp. 179–186. doi:10.1002/wps.20735.

Hughes, C.E. and Stevens, A. (2010) 'What can we learn from the Portuguese decriminalization of illicit drugs?', British Journal of Criminology, 50(6), pp. 999–1022. doi:10.1093/bjc/azq038.

Kilmer, B. (2017) 'Recreational cannabis — minimizing the health risks from legalization', New England Journal of Medicine, 376(8), pp. 705–707. doi:10.1056/NEJMp1614783.

Laqueur, H., Rivera-Aguirre, A., Shev, A., Castillo-Carniglia, A., Rudolph, K.E., Ramirez, J., Martins, S.S. and Cerdá, M. (2020) 'The impact of cannabis legalization in Uruguay on adolescent cannabis use', International Journal of Drug Policy, 80, 102748. doi:10.1016/j.drugpo.2020.102748.

MacCoun, R.J. (2011) 'What can we learn from the Dutch cannabis coffeeshop system?', Addiction, 106(11), pp. 1899–1910. doi:10.1111/j.1360-0443.2011.03572.x.

MacCoun, R. and Reuter, P. (2001) 'Evaluating alternative cannabis regimes', British Journal of Psychiatry, 178(2), pp. 123–128. doi:10.1192/bjp.178.2.123.

Queirolo, R., Repetto, L., Alonso, J., Álvarez, E., Sotto, B., Pardal, M. and Kilmer, B. (2024) 'Intended and unintended effects of cannabis regulation in Uruguay'. Available at: https://repositorio.ausjal.org/handle/20.500.12032/179430.

Rivera-Aguirre, A., Castillo-Carniglia, A., Laqueur, H.S., Rudolph, K.E., Martins, S.S., Ramirez, J., Queirolo, R. and Cerdá, M. (2022) 'Does recreational cannabis legalization change cannabis use patterns? Evidence from secondary school students in Uruguay', Addiction, 117(11), pp. 2866–2877. doi:10.1111/add.15913.

Rolles, S. and Murkin, G. (2016) How to Regulate Cannabis: A Practical Guide. 2nd edn. Bristol: Transform Drug Policy Foundation.

Room, R., Fischer, B., Hall, W., Lenton, S. and Reuter, P. (2010) Cannabis Policy: Moving Beyond Stalemate. Oxford: Oxford University Press.