Cannabis Legalization Overview: Global Legal Status

Why 'cannabis legalization' is the wrong umbrella term for half the world

Many readers hear “cannabis is legal” and assume one thing: you will not be arrested for having some. Law does not work that way. A country can stop jailing people for possession and still criminalize cultivation, trafficking, large-scale supply, and even sharing. It can permit prescription-only medical products while keeping nonmedical use illegal. It can allow home grow but ban retail sale. It can tolerate storefront supply while leaving wholesale production in a gray zone. Calling all of that “legalization” erases the distinctions that actually matter.

That distinction is not academic. It determines who can access cannabis, in what form, under what supervision, from which source, and with what criminal risk. It also shapes taxation, product standards, advertising restrictions, corporate participation, police powers, and public-health oversight. Mark A.R. Kleiman and Beau Kilmer both argued, in different ways, that policy design matters as much as the headline reform. They were right. A possession reform is not a market design. A medical exemption is not adult-use legality. A tolerated retail outlet is not a lawful supply chain.

The baseline for most of the modern era was prohibition, anchored internationally by the 1961 UN Single Convention on Narcotic Drugs, reinforced by the 1971 and 1988 treaties, and reflected in national laws such as the US Controlled Substances Act of 1970. Even the December 2020 vote by the UN Commission on Narcotic Drugs to remove cannabis and cannabis resin from Schedule IV of the 1961 Convention did not legalize cannabis internationally; it left the drug in Schedule I and mainly signaled formal recognition of medical value.

That matters because cannabis reform is happening against mass existing use, not fringe demand. UNODC’s World Drug Report 2024 estimated 228 million users in 2022. Europe alone had an estimated 24 million last-year users, according to the 2024 EMCDDA report, and 4.3 million daily or near-daily users. Any legal framework is governing a large social reality, not creating one from scratch.

Prohibition, decriminalization, medical access, and legalization are different legal categories

Prohibition means production, supply, and possession are criminal offenses, even if enforcement varies. Decriminalization usually removes criminal penalties for possession of small amounts but leaves supply illegal. Portugal since 2001 is the classic example: possession for personal use can trigger administrative sanctions, not prison, while trafficking remains criminal. That is a major reform. It is not legalization.

Medical access is another category entirely. It creates a lawful channel for a defined patient population, often through specialist prescribing, approved products, or tightly controlled imports. California’s Proposition 215 in 1996 launched the modern era of medical reform, but medical systems diverged sharply afterward. Some countries allow only standardized pharmaceuticals such as nabiximols or purified cannabidiol preparations. Others permit herbal cannabis flower. The UK has legal cannabis-based medicinal products since 2018, yet NHS prescribing remains limited in practice. Australia and Israel built large medical systems without adopting general adult-use legalization.

Legalization, properly used, means the law creates a lawful framework for nonmedical adult access. Even then, there is no single model. Uruguay’s Law No. 19.172 of 2013 built a state-controlled system centered on home grow, clubs, and pharmacy supply, with the explicit aim of undercutting illicit trafficking. Canada’s Cannabis Act, in force on 17 October 2018, created a federally regulated adult-use market with licensed production and provincial retail systems; Statistics Canada reported in 2024 that legal channels accounted for about 72% of household cannabis spending in 2023. Germany’s Cannabis Act, effective 1 April 2024, legalized possession within limits and home cultivation, then allowed noncommercial cultivation associations from 1 July 2024. That is legalization of possession and limited access. It is not a Canadian-style retail market.

Malta and Luxembourg make the point even sharper. Malta’s 2021 reform allowed possession up to 7 grams, home cultivation, and later non-profit associations. Luxembourg in 2023 legalized possession and home grow. Neither created broad commercial retail. Describing either as simply “legalized cannabis” leaves out the whole architecture.

Why media shorthand confuses possession rules with supply rules

Most reporting collapses two different questions into one: can an adult possess cannabis, and is there a lawful way to produce and distribute it? Those are separate legal problems. The Netherlands has shown this for decades. Since the 1976 Opium Act revision and later coffee-shop policy, retail sale of small amounts has been tolerated under strict conditions, yet production remained formally illegal for years, creating the famous “back door problem.” A tolerated point of sale is not the same as a fully legal market.

The United States is another case study in bad shorthand. State-level adult-use legalization now covers much of the country, and medical access is wider still, yet cannabis remains Schedule I federally under the Controlled Substances Act. That split affects banking, tax treatment under Internal Revenue Code section 280E, immigration consequences, firearms eligibility, interstate commerce, and research barriers. Saying “marijuana is legal in the US” is false as a statement of federal law and incomplete as a description of lived reality.

Germany is being misreported in exactly this way. Since 1 April 2024, adults may possess up to 25 grams in public and cultivate up to three plants. Noncommercial associations, overseen within a regulated framework that involves authorities including BfArM-linked compliance structures, may supply members under strict limits. No nationwide commercial storefront market opened. Yet many headlines treated the law as if Germany had copied Canada. It did not.

Wayne Hall has long warned that debates framed as pro- versus anti-legalization miss the real issue: what type of system is being built, with what public-health safeguards and what risks.

The article's central claim: global cannabis law is fragmenting, not converging

There is movement away from blanket prohibition. That part is real. But the idea of a single global march toward legalization is lazy and wrong. What the evidence shows is diversification.

Europe is splitting into multiple models at once: Germany’s possession-plus-associations framework, Malta’s non-profit model, Luxembourg’s private-use reform, the Netherlands’ tolerance approach, Spain’s legally unstable club system, and many states that remain prohibitionist. The Americas are no cleaner. Canada and Uruguay have national adult-use laws, but they are built on very different assumptions about state control and market structure. The US has state legalization under federal prohibition. Mexico’s Supreme Court dismantled parts of the prohibition on personal use, yet Congress still has not enacted a complete national regulatory statute. John Walsh’s work on Latin America has tracked this pattern for years: reform often advances through constitutional litigation, prosecutorial discretion, or narrow exemptions rather than tidy legislative legalization.

Africa and Asia-Pacific show the same fragmentation. South Africa’s private-use model is not a broad legal market. Morocco’s 2021 law concerns medical and industrial cultivation, not adult-use. Thailand’s 2022 removal of cannabis from the narcotics list opened a chaotic period of de facto liberalization, followed by repeated attempts to tighten controls. Japan remains restrictive while adjusting some cannabinoid medicine rules. Australia permits a large medical system, while adult-use stays broadly illegal outside the ACT’s limited carve-out.

So this article uses a stricter taxonomy on purpose. Prohibition. Decriminalization. Medical access. Limited adult-use possession. Non-profit supply. Commercial legalization. They are not stages on one inevitable ladder. They are competing legal families. And right now, the world is producing more of them, not fewer.

How cannabis prohibition became global

Cannabis prohibition did not emerge from a single scientific finding or a settled global consensus on public health. It was assembled over decades through colonial-era controls, diplomatic bargaining, US pressure, and treaty design. That history matters because modern reform debates still operate inside legal machinery built in the twentieth century. What now looks like the default position was, in fact, constructed piece by piece.

By the time UNODC estimated in its 2024 World Drug Report that 228 million people used cannabis in 2022, cannabis had already spent generations inside the international narcotics system. The scale of use shows something prohibition often failed to stop. It does not show that prohibition was inevitable.

Early international controls before the UN era

The earliest international drug-control efforts did not begin with cannabis. They centered on opium, morphine, cocaine, and imperial concerns about trade, labor, and social order. Cannabis entered that framework later, and unevenly.

The Hague Opium Convention of 1912 is usually treated as the opening chapter of modern international drug control, but cannabis was not its main target. The shift came with the revised International Opium Convention signed at Geneva in 1925. That convention added controls on “Indian hemp,” especially exports of cannabis resin and international trade in preparations. The treaty did not create a fully modern prohibition system. It did something more foundational: it inserted cannabis into the same diplomatic vocabulary as opiates and cocaine.

Why did that happen? Not because the world had reached a common scientific judgment about cannabis harms. States arrived with very different experiences. In parts of South Asia, North Africa, and the Middle East, cannabis use had long existed in local social and medicinal contexts. Other governments, especially colonial administrations, viewed it through a lens of discipline and order. Egypt pushed hard for stronger international cannabis restrictions in the interwar period. British India was more cautious, partly because the Indian Hemp Drugs Commission of 1894 had not supported absolute suppression and had found moderate use to be less catastrophic than prohibitionists claimed. That contrast alone undercuts the myth that global prohibition was simply the natural result of accumulating evidence.

National laws also moved ahead of or alongside treaties. The United States is the clearest case. Before federal prohibition, cannabis restrictions spread through state laws, often tied to xenophobic narratives about Mexican migrants and racialized fears in the Southwest. Then came the Marihuana Tax Act of 1937. Formally, it was a tax statute rather than an outright ban. In practice, by imposing registration requirements, transfer taxes, and criminal penalties for noncompliance, it made lawful possession and transfer extremely difficult. Harry Anslinger, head of the Federal Bureau of Narcotics, was central to this campaign. The law was not the product of a neutral scientific process; it rested heavily on moral panic, bureaucratic ambition, and selective claims about crime and insanity.

That US model mattered beyond US borders. It helped normalize the idea that cannabis should be governed as a narcotic problem requiring penal control. Long before the UN treaty system solidified, cannabis was being reframed internationally from a diverse set of local practices into a standardized object of transnational criminal law.

The 1961 Single Convention and the architecture of prohibition

The real lock-in came after the Second World War. The League of Nations system gave way to the United Nations, and postwar institutions sought to consolidate the patchwork of earlier agreements into one treaty structure. The result was the 1961 Single Convention on Narcotic Drugs.

This was the decisive legal architecture of global prohibition. The Single Convention did not merely mention cannabis. It placed cannabis and cannabis resin in Schedule I, subjecting them to strict controls, and also in Schedule IV, the category reserved for substances considered especially liable to abuse and to produce ill effects with little therapeutic value. That double placement sent a clear political message: cannabis was to be treated as one of the most tightly controlled drugs in the international system.

The treaty required parties to limit production, manufacture, export, import, distribution, trade, use, and possession of scheduled drugs to medical and scientific purposes. That phrase became the backbone of prohibition. It is easy now to read it as obvious. It was not. It was a legal choice, embedded in an institution-building moment when states were creating permanent international mechanisms for surveillance, reporting, scheduling, and compliance.

The Single Convention also standardized state obligations. Countries had to establish national control agencies, licensing systems, reporting duties, and penal frameworks. In other words, prohibition was not just a rule against use. It was an administrative project. Cannabis became embedded in customs law, criminal codes, agriculture rules, pharmaceutical regulation, and foreign policy.

The United States reinforced this architecture domestically with the Controlled Substances Act of 1970, which placed marijuana in Schedule I. That classification, still the federal baseline in the US today, echoed the logic of the Single Convention even as US states later moved in sharply different directions. The key point is historical: national prohibition and international prohibition were mutually reinforcing, not separate stories.

The later symbolism of reform shows how durable the 1961 framework remained. In December 2020, the UN Commission on Narcotic Drugs voted to remove cannabis and cannabis resin from Schedule IV of the Single Convention, following recommendations from the WHO Expert Committee on Drug Dependence. But cannabis remained in Schedule I. That did not legalize cannabis under international law. It merely acknowledged medical value while preserving the treaty’s core control structure. Even reform at the UN level moved cautiously, and inside the old architecture.

The 1971 and 1988 treaties and the hardening of enforcement

If the 1961 treaty built the framework, the later conventions made it harder-edged. The 1971 Convention on Psychotropic Substances mainly addressed synthetic drugs such as LSD, amphetamines, and benzodiazepines, not cannabis flower itself. Still, it expanded the global scheduling mentality and strengthened the idea that psychoactive substances should be managed through layered international controls. It also helped entrench a habit of thinking in schedules first, public-health design second.

The sharper turn came with the 1988 Convention against Illicit Traffic in Narcotic Drugs and Psychotropic Substances. This treaty arrived during the late Cold War drug war era, when enforcement, interdiction, and criminalization were intensifying. It pushed states toward criminal penalties for production, distribution, and possession contrary to the 1961 and 1971 conventions, subject to constitutional principles and domestic legal systems. Asset seizure, extradition, precursor controls, and cross-border police cooperation became central tools.

That mattered for cannabis. The drug with the widest user base became tied more tightly to anti-trafficking law, organized-crime policy, and punitive policing. The political story hardened: cannabis was no longer merely a scheduled narcotic under a health treaty, but part of an international enforcement agenda. Scholars such as John Walsh have shown how, in Latin America, this treaty environment constrained reform even where domestic opinion shifted. Mark A.R. Kleiman and Beau Kilmer later argued that if states were going to move away from prohibition, policy design would matter enormously because the inherited legal baseline was punitive and poorly suited to managing real markets. Wayne Hall’s work on legalization debates likewise starts from a plain fact: prohibition has costs, but so does careless commercial liberalization. That framing only makes sense once prohibition is seen as a policy choice rather than a natural state.

So the global story is not a smooth march from ban to freedom. It is a sequence of legal constructions. First cannabis was added to international control. Then it was embedded in the UN treaty order. Then enforcement norms were hardened. Modern legalization, decriminalization, and medical access schemes are all reacting to that inheritance.

The long reform era: from medical exceptions to adult-use legalization

The reform wave that began in the 1990s did not move in one straight line from illegality to free access. It split into different legal families. One path created narrow medical exemptions for patients. Another reduced criminal penalties for possession without permitting lawful supply. A third built adult-use systems, but even those diverged sharply: Uruguay chose state control to weaken illicit trafficking, Canada built a federally regulated market, US states moved under continuing federal prohibition, and Europe has often stopped at possession, home grow, or non-profit associations.

That distinction matters because global use was already widespread before law changed. UNODC estimated in its World Drug Report 2024 that 228 million people used cannabis in 2022, making it the most used drug under international control. Reform did not create cannabis use from scratch. It changed how states responded to a behavior that prohibition had plainly failed to eradicate.

California Proposition 215 and the modern medical cannabis era

The modern era usually starts with California Proposition 215, approved by voters in November 1996. Formally the Compassionate Use Act, it allowed patients and caregivers to possess and cultivate cannabis for medical use with a physician’s recommendation. The political logic was not libertarian legalization. It was compassion access, shaped by AIDS activism, cancer care, chronic pain, and resistance to criminalizing seriously ill people.

That was a major break with the framework built by the 1961 UN Single Convention and, in the United States, the Controlled Substances Act of 1970, which kept marijuana in Schedule I. Proposition 215 did not change federal law. It carved out state protection inside a federal prohibition system. That contradiction became the basic template for US cannabis law over the next three decades.

California’s law was loose by later standards. It did not create a tightly defined prescription-drug model or a centrally planned supply chain. Instead, it opened legal space around patient need and physician recommendation. Other states followed, though often with narrower rules. Some permitted herbal cannabis. Some allowed only low-THC preparations. Some built registries and dispensary licensing; others barely tolerated possession defenses. “Medical cannabis” was never one thing either.

This distinction still gets lost. A country or state may allow cannabis-based medicines while banning herbal cannabis entirely. The United Kingdom after 2018 is a good example: cannabis-based medicinal products were rescheduled, yet NHS access remained very limited in practice. Australia and Israel developed substantial medical frameworks without embracing full adult-use legalization. The WHO Expert Committee on Drug Dependence acknowledged therapeutic evidence for certain cannabis-related preparations, especially cannabidiol and standardized products, but that recognition did not erase the gap between tightly controlled medicine and broader legal access.

The US medical turn had another effect. It normalized cannabis as a policy object that could be regulated rather than simply banned. Once a state accepts that some lawful cultivation, possession, and distribution can exist, the argument shifts. The question is no longer “ban or permit” in absolute terms. It becomes who gets access, under what rules, with what medical evidence, and under whose supervision.

That change in vocabulary mattered politically. By the 2000s, reformers had learned that medical access could attract support from voters who would reject outright adult-use legalization. But medical reform was never just a stepping stone. In many places it remained the endpoint.

Portugal, harm reduction, and the rise of decriminalization

Portugal’s 2001 reform is often misdescribed as cannabis legalization. It was not. Law No. 30/2000 decriminalized possession and use of all drugs for personal use, including cannabis, while leaving trafficking and supply criminal offenses. People found with small quantities are referred to Commissions for the Dissuasion of Drug Addiction, which can impose administrative responses rather than criminal punishment.

The logic here was different from California’s. Portugal’s policy emerged from harm reduction, public health, and a desire to reduce the damage caused by criminalization, especially amid concern over heroin use, HIV transmission, overdose, and social exclusion. Cannabis happened to be included because the reform addressed drug possession generally, not because lawmakers created a legal cannabis market.

This is where many legalization overviews go wrong. Decriminalization changes penalties, usually for possession of small amounts. Legalization creates a lawful framework for supply. Portugal did the first, not the second. No legal retail market followed. No legal nonmedical production system appeared. Trafficking remained criminal. Administrative tolerance is not the same as lawful commercial access.

Portugal still became influential because it showed another route away from punitive prohibition. The policy argument was not “cannabis is harmless.” It was that criminal punishment for personal possession can make public-health problems worse. That approach spread unevenly. Parts of Latin America, the Czech Republic, and some Australian jurisdictions adopted partial decriminalization models. Mexico’s constitutional and Supreme Court developments weakened the legal basis for punishing personal adult use, yet Congress did not complete a full adult-use regulatory statute. Again, not one thing.

Europe today still reflects this patchwork. The EMCDDA’s European Drug Report 2024 estimated that 24 million adults in Europe, or 8.4% of those aged 15 to 64, used cannabis in the last year; 4.3 million were daily or almost daily users. At the same time, cannabis accounted for 36% of new treatment admissions in Europe in 2022. Those figures cut against easy slogans. High prevalence supports the case that blanket prohibition has weak deterrent effect. Treatment data show that public-health burdens do not vanish when penalties are softened.

Colorado, Washington, Uruguay, and the break with prohibition

The real rupture with prohibition came in 2012. Colorado’s Amendment 64 and Washington’s Initiative 502 legalized adult use by ballot measure, making them the first US jurisdictions to create lawful nonmedical systems. Their political logic mixed civil-liberties arguments with tax-and-regulate ideas: if prohibition had failed, the state should replace illicit trade with licensed production, age limits, product rules, and tax collection.

Yet even here the legal picture was unstable. Federal law did not change. Marijuana remained Schedule I under the Controlled Substances Act. So the United States moved into a highly unusual model: state-level legalization under federal prohibition. That creates ongoing problems involving banking, tax treatment under Internal Revenue Code section 280E, immigration consequences, firearm restrictions, and interstate commerce. Calling this simply “legal in America” is wrong.

Policy design mattered. Mark A.R. Kleiman and Beau Kilmer both argued, in different ways, that legalization is not binary; market structure shapes outcomes. Potency rules, pricing, licensing density, home-grow allowances, advertising limits, and enforcement priorities all affect public-health and criminal-market results. Wayne Hall’s work has been especially clear on tradeoffs: legalization can reduce arrests and displace illicit supply, but higher availability and normalized use may also increase heavy use and some health harms. Evidence supports neither panic nor triumphalism.

Uruguay went further than Colorado and Washington in one sense and less far in another. Law No. 19.172, enacted in December 2013, made Uruguay the first country to legalize nonmedical cannabis nationwide. But it did so through a tightly controlled system: home cultivation, membership clubs, and pharmacy sales under state oversight and registration requirements. This was not a broad commercial model. Its central rationale was anti-trafficking and public security. The state sought to undercut illicit markets rather than sponsor a large private industry.

That made Uruguay historically important. It proved that adult-use legalization could be justified not only by personal freedom or tax revenue, but by an anti-organized-crime strategy. John Walsh’s work on Latin American reform has been valuable here: in the region, cannabis policy often developed in conversation with violence, prison overcrowding, and the failures of militarized drug control, not just cultural liberalization.

Canada’s Cannabis Act, in force on 17 October 2018, then added another model: federal nationwide adult-use legalization in a high-income democracy with formal regulation across the country. By early 2024, Statistics Canada reported that 26% of people aged 16 and older had used cannabis in the previous 12 months, and legal channels accounted for about 72% of household cannabis spending in 2023. That is one of the clearest signs that legal regulation can displace illicit supply at scale, though not eliminate it.

The 2020 UN Commission on Narcotic Drugs vote to remove cannabis and cannabis resin from Schedule IV of the 1961 Convention belongs in this story, but only as a symbol. Cannabis remained in Schedule I. International prohibition was not dismantled. The vote mattered because it acknowledged medical value at the treaty level. It did not create global legalization. If anything, the period since then has confirmed fragmentation: Malta’s 2021 non-profit association model, Luxembourg’s 2023 possession and home-grow reform, and Germany’s 2024 Cannabis Act with possession limits and cultivation associations all moved away from prohibition without embracing a Canadian-style retail market. The long reform era did not replace one orthodoxy with another. It broke the old consensus into competing systems.

The legal distinction that matters most: decriminalization versus legalization

Popular coverage often treats any softening of cannabis penalties as “legalization.” That is wrong, and the error matters. A country can stop arresting people for small-scale possession while still treating the plant’s production and supply as a criminal market. It can allow patients access through doctors and pharmacies while keeping nonmedical adult use illegal. It can legalize possession and home growing without permitting commercial retail. These are not semantic differences. They produce different arrest patterns, different public-health outcomes, different state capacities, and different relationships to illicit supply.

That distinction has become more important as cannabis reform has accelerated against a very large baseline of use. UNODC’s World Drug Report 2024 estimates that 228 million people used cannabis in 2022, making it the world’s most used internationally controlled drug. Reform is not unfolding at the margins. It is happening in response to an already widespread market shaped by the prohibition architecture built through the 1961 Single Convention on Narcotic Drugs, later reinforced by the 1971 and 1988 treaties. Even the December 2020 UN Commission on Narcotic Drugs vote removing cannabis and cannabis resin from Schedule IV of the 1961 Convention did not legalize cannabis internationally; it left them in Schedule I and changed symbolism more than baseline legality.

What decriminalization removes and what it leaves criminal

Decriminalization usually means that possession of a small amount for personal use no longer triggers criminal prosecution or jail. It does not mean cannabis becomes lawful in any general sense. The substance remains prohibited, supply remains unauthorized, and trafficking remains criminal.

Portugal is the classic example because it is so often misdescribed. Since 2001, possession of small quantities of any drug for personal use has been decriminalized there, but not legalized. If a person is found with an amount under the statutory threshold, the case may go to a regional “Commission for the Dissuasion of Drug Addiction” rather than a criminal court. That is an administrative process, not a free-pass system. Possible outcomes include warnings, fines, suspension of professional licenses in some cases, referral to treatment, or other administrative measures. If quantities exceed the threshold, criminal law can still come back into play. Supply and trafficking never left the criminal sphere.

That pattern repeats elsewhere with local variations. Some systems replace arrest with civil fines. Some use diversion schemes, meaning police or prosecutors send the person to education, assessment, or treatment instead of pursuing a criminal conviction. Some formally retain the offense but downgrade it from criminal to administrative. In all of these models, threshold quantities do heavy legal work. A legislature might say that up to a defined amount is presumed personal possession; above that amount, intent to supply may be inferred or at least more easily alleged. The threshold is not just a number. It is the line between “handled by ticket, commission, or warning” and “handled by the criminal law.”

This is why decriminalization should not be confused with legalization. If there is no lawful source of supply, the market remains illicit by design. A person may avoid a criminal record for carrying a few grams while the people who grew, transported, or sold that cannabis remain exposed to trafficking charges. Beau Kilmer and Mark A.R. Kleiman both stressed, in different ways, that market structure is the policy question. Decriminalization changes penalties at the user end; it does not answer who may produce cannabis, under what rules, with what quality controls, and with what sanctions for noncompliance.

The Netherlands shows how easy it is for public discussion to blur these categories. Dutch coffee shops have operated for decades under a tolerance policy tied to the 1976 Opium Act revision and later guidance, but that was never full legalization. Retail sale in tolerated settings coexisted with the famous “back door problem”: stock for those outlets was not fully legalized upstream. A tolerated retail endpoint without fully lawful production is not the same thing as a legalized supply chain. It is a hybrid.

What legalization requires: a lawful production and distribution framework

Legalization begins where decriminalization stops: it creates a lawful route to produce and supply cannabis. That route can be narrow or broad, state-run or private, commercial or non-profit. But there must be some authorized framework for cultivation, processing, distribution, and access.

Uruguay’s Law No. 19.172 of 2013 is one clear model. It legalized nonmedical cannabis nationwide, but in a tightly controlled form aimed at displacing illicit trafficking rather than building a large consumer market. Adults can access cannabis through home cultivation, membership clubs, or pharmacy sales under state supervision. That is legalization, yet it is not the same system as Canada’s.

Canada’s Cannabis Act, in force from 17 October 2018, created a federally regulated adult-use market with licensed cultivation, processing, and provincial retail systems. It is the clearest example of full national legalization in a G7 country. The measurable effect matters: Statistics Canada reported in 2024 that legal channels accounted for about 72% of household cannabis spending in 2023. Not complete displacement of illicit supply. Still real displacement. Wayne Hall’s work on legalization has long argued that the public-health impact depends less on the word “legal” than on the details of price, potency, promotion limits, product rules, and enforcement.

Germany’s 2024 Cannabis Act, or KCanG, makes the opposite point. Adults may possess up to 25 grams in public and grow up to three plants from 1 April 2024, and non-commercial cultivation associations became possible from 1 July 2024 under regulatory oversight involving BfArM. Many headlines called this “legal cannabis in Germany,” but that shorthand skipped the legal design. Germany did not create a general commercial retail market like Canada. It legalized possession within limits, home cultivation, and supply through regulated non-profit associations. That is legalization, but of a bounded and non-commercial kind.

Malta’s 2021 reform and Luxembourg’s 2023 possession/home-grow reform sit in the same family. Both moved beyond decriminalization because they created lawful adult possession and home cultivation. Yet neither opened a broad commercial retail system. Malta also moved toward regulated non-profit associations. Again: legalization, yes. Commercial legalization, no.

The United States is messier. Since Colorado and Washington voted for adult-use legalization in 2012, two dozen states plus DC have built adult-use systems, while many more allow medical access. But federal law still classifies marijuana as Schedule I under the Controlled Substances Act of 1970. So a person may be fully compliant under state law and still face federal consequences tied to banking, tax under Internal Revenue Code section 280E, immigration status, firearms eligibility, federal employment, or interstate commerce. That is not a contradiction in journalism terms; it is a dual-sovereignty legal conflict.

Where medical legalization fits and why it is neither of the above

Medical cannabis occupies a third category. It is neither decriminalization nor general adult-use legalization. It creates a lawful channel for a defined patient population under medical criteria.

That channel may be extremely narrow. Some countries allow only pharmaceutical cannabinoids such as nabiximols or purified cannabidiol products. Others allow herbal cannabis but only for specified conditions and only on specialist prescription. The United Kingdom is a good example of law on paper diverging from access in practice: cannabis-based medicinal products were rescheduled in 2018, yet NHS prescribing has remained very limited. Australia and Israel have built substantial medical frameworks without adopting full adult-use legalization. Those systems are legal access models, but only for eligible patients and only through approved pathways.

The 1996 passage of California Proposition 215 opened the modern medical era, and many jurisdictions followed that sequence: first medical access, then later debates over adult use. But the first step does not guarantee the second. WHO’s Expert Committee on Drug Dependence has recognized therapeutic evidence for some cannabis-related preparations, especially CBD and standardized products. That medical recognition supports rescheduling debates and patient access. It does not amount to a mandate for nonmedical legalization.

So the line is simple, even if the world is not. Decriminalization reduces or removes criminal penalties for possession of small amounts, often replacing them with administrative penalties, civil fines, or diversion, while leaving supply criminal. Legalization creates a lawful supply framework, whether commercial, state-controlled, home-grow based, or non-profit. Medical legalization opens legal access only for patients who meet defined medical rules. Once those categories are separated, the global map makes more sense. It is not a march from prohibition to freedom. It is a patchwork of different legal families moving at different speeds.

North America: the most commercially developed cannabis law region

North America is where the legal architecture is most fully built out, and also where the phrase legal cannabis causes the most confusion. Canada has a federal adult-use statute backed by national rules and provincial distribution systems. The United States has large state markets operating inside a federal criminal ban. Mexico moved through constitutional litigation that weakened prohibition on personal use, yet still does not have a completed nationwide adult-use retail framework. Those are three different legal families, not versions of the same reform.

That makes the region the leading testing ground for regulatory design. It is where policymakers have had to answer hard operational questions rather than abstract ones: who may produce, who may possess, what taxes do, how product forms are limited, how advertising is restricted, whether home cultivation is allowed, and how far legal supply can displace illicit supply. Scholars such as Beau Kilmer and Mark A.R. Kleiman have long argued that these design choices matter as much as the yes-or-no question of legalization. North America proves the point.

Canada's Cannabis Act and the first G7 national adult-use market

Canada became the first G7 country to legalize nonmedical cannabis nationwide when the Cannabis Act came into force on 17 October 2018. Legally, this was a clean break from prohibition in a way many other reforms were not. Adults gained lawful access within a regulated national framework, while criminal penalties remained for conduct outside that framework.

The federal law set the broad terms. Adults may generally possess up to 30 grams of legal dried cannabis or equivalent in public. Home cultivation of up to four plants per household is allowed under federal law, though Quebec and Manitoba moved to restrict home growing and triggered constitutional disputes. Packaging, promotion, product standards, youth access rules, and criminal offenses for illicit distribution were all addressed at the national level.

But Canada did not create a single uniform market in practice. Provinces and territories control distribution and retail models, and those choices produced visible differences. Ontario moved from a state-retail concept to a private-store model. Quebec relied on a government monopoly retailer. Alberta allowed a dense private retail network early. Minimum age also differs: 18 in Alberta, 19 in many provinces, 21 in Quebec. Online access, store density, and wholesale arrangements vary too. So even under federal legalization, implementation is patchy.

That distinction matters because Canada is often described as if one statute solved the entire issue. It did not. It created a lawful national market, then left much of the actual consumer-facing system to provinces. Wayne Hall’s work on legalization and public health is relevant here: the social effects depend heavily on price, potency rules, marketing limits, outlet density, and enforcement priorities, not just legal status on paper.

Canada is also the strongest case that regulated legal supply can pull consumers away from illegal channels, though not erase them. Statistics Canada reported in 2024 that the legal share of household final consumption expenditure on cannabis was 72% in 2023, up from 69% in 2022. That is major displacement. It is not total displacement. Untaxed supply, lower-price illicit sellers, and unlicensed online sources still exist. Yet compared with early post-2018 doubts, the legal market has captured most measured spending.

Use remains common. Statistics Canada’s National Cannabis Survey found that 26% of people aged 16 and older reported cannabis use in the previous 12 months in the first quarter of 2024. A legal market did not produce mass novelty uptake from nowhere; it formalized and redirected a pre-existing consumer market. That is one reason Canada is such an important case study. It is not a story of creating demand. It is a story of governing demand that already existed.

The public-health debate is real and unresolved. Supporters of the Cannabis Act point to age-gated sales, mandatory testing, labeling, and reduced criminalization. Critics point to high-THC products, accidental pediatric exposures from edibles, normalization of use, and concern over cannabis-impaired driving. Hall has consistently argued that legalization trades some harms for others: fewer arrests and better product oversight can coexist with increased commercial pressure and heavier use among some groups. Canada illustrates exactly that tradeoff.

So Canada is plainly legalization. But it is not laissez-faire legalization. It is a federal regulatory state model, with strict promotion limits, controlled packaging, criminal backstops, and major provincial variation layered on top.

The United States: state legalization under federal Schedule I prohibition

The United States is the most commercially expansive cannabis region in the world, but legally it is still contradictory. Cannabis remains a Schedule I substance under the Controlled Substances Act of 1970, the federal category reserved for drugs the government says have high abuse potential and no accepted medical use under federal law. That remains true even after Colorado, Washington, California, and many other states legalized adult use.

So when Colorado legalized in 2012, cannabis did not become lawful in the United States as a whole. It became lawful under Colorado state law while remaining federally prohibited. That conflict is the defining fact of the US system.

The modern reform path began with medicine, not adult use. California Proposition 215 in 1996 opened the contemporary medical era by allowing medical marijuana despite federal prohibition. Many states followed with physician-certification systems, patient registries, dispensary structures, and cultivation rules. That medical-first sequence mattered politically and legally. It built institutions, normalized therapeutic claims, and created constituencies before adult-use legalization arrived. Adult-use markets in Colorado and Washington in 2012 did not appear from a standing start; they grew out of a medical reform lineage.

Today, adult-use legalization spans roughly half the country, and medical access is broader still. Yet none of that changes the federal baseline. Federal law still affects banking, tax, immigration, firearms possession, employment in federal sectors, public housing, interstate transport, and research. Internal Revenue Code section 280E, for example, has long blocked ordinary business deductions for enterprises trafficking in Schedule I or II substances under federal law, producing tax burdens that would be unthinkable in ordinary commerce. A person can follow state law and still face federal collateral consequences.

Interstate commerce is where the legal contradiction becomes especially obvious. A product grown legally in one state cannot simply move to another legal state because crossing state lines triggers federal jurisdiction. Every state market is, in effect, an island. That fragments supply, protects inefficient local production, and keeps prices and quality systems uneven. Beau Kilmer has argued that market structure is not a side issue but the policy itself. The United States shows why: the federal-state split prevents anything like a coherent national regime.

Federal enforcement has often been selective rather than maximal. The Obama administration’s 2013 Cole Memorandum signaled a hands-off approach to well-regulated state systems, though it did not legalize them and was later rescinded in 2018 by Attorney General Jeff Sessions. Congress has also repeatedly adopted appropriations riders restricting the Department of Justice from interfering with state medical marijuana programs, but those protections are limited and do not cover all adult-use activity. Tolerance is not legality.

Scale is what makes the US case so influential. SAMHSA’s 2023 National Survey on Drug Use and Health reported that 61.8 million people aged 12 or older used marijuana in the past year. The market is huge, culturally visible, and administratively dense. But the law remains split-level. State legalization changes arrest patterns and creates licensed systems within those states. It does not remove cannabis from federal Schedule I, rewrite immigration law, or authorize a national retail system.

Calling the US “legalized” without qualification is simply wrong. It is a federation where prohibition and legalization coexist in the same territory, at the same time, with different sovereigns claiming authority.

Mexico: constitutional change without a completed national retail system

Mexico is often placed in the legalization column, but that overstates what has happened. The better description is that constitutional litigation broke down the legal basis for blanket prohibition of personal adult use, while Congress failed to complete a full regulatory statute.

The turning point came through the Supreme Court of Justice of the Nation. Beginning with a 2015 decision involving members of the group SMART, and followed by later rulings, the Court held that absolute prohibitions on personal cultivation and consumption violated the constitutional right to the free development of personality. By 2021, the Court had invalidated key provisions that blocked adults from obtaining permits for personal use. This was a serious legal shift. It weakened prohibition at its constitutional core.

But it did not by itself create a nationwide commercial adult-use system. That is the distinction many summaries miss. Court rulings can strike down prohibitions or compel administrative accommodation; they do not automatically produce a licensing code for cultivation, processing, distribution, retail, taxation, packaging, advertising controls, or impaired-driving standards. Mexico still lacks that completed national framework.

In practice, adults have had more room to seek permits and challenge enforcement, and criminal law has become harder to apply in the old blanket way. Yet the absence of a final statutory regime leaves uncertainty. What is lawful for private use is not the same thing as what is lawfully produced and distributed at national scale. John Walsh’s work on Latin American drug reform has repeatedly stressed this point: constitutional doctrine and criminal-code softening can move faster than administrative state-building.

Mexico therefore sits between prohibition and full legalization. It is not Portugal-style decriminalization, because the constitutional dimension is stronger and more rights-based. It is also not Canada-style legal regulation, because there is no completed nationwide adult-use retail architecture. It is a partial post-prohibition space shaped by jurisprudence more than by a settled market statute.

That ambiguity matters beyond Mexico itself. It shows that “legalization” can fail to materialize even after major court victories. Legislatures still have to write the operating system. North America, taken as a whole, makes the larger point of this article with unusual clarity: legal reform is not a single road. Canada built a federal regulated market. The United States built state markets under federal criminal law. Mexico loosened prohibition through constitutional adjudication without finishing the legislative job. Same region. Three models.

Europe: reform is real, but it is narrower than headlines suggest

Europe is often described as if it were marching toward “legalization” in the same way Canada did in 2018. That is wrong. What Europe has instead is a patchwork of legal families: tolerated retail in one country, decriminalization in another, home-grow allowances somewhere else, medical access in many places, and outright prohibition still common across much of the continent. The labels get blurred, and that blur matters.

The scale of use helps explain why reform keeps returning to the agenda. The EMCDDA reported in its 2024 European Drug Report that 24 million adults in Europe, or 8.4% of those aged 15 to 64, used cannabis in the last year. An estimated 4.3 million were daily or almost daily users. At the same time, cannabis was involved in 36% of all drug treatment admissions in Europe in 2022. So the policy debate is not about a marginal substance. It sits at the intersection of enforcement limits, public health, and political symbolism.

Still, Europe is not building a single continental model. If Canada is the clearest example of a federally regulated adult-use market, Europe is the clearest example that legalization is not one thing. Mark A.R. Kleiman and Beau Kilmer both argued, in different ways, that cannabis policy design matters as much as the legalization decision itself. Europe proves the point. Most reforms there are bounded, partial, and defensive. They aim to reduce criminal penalties or move use into supervised channels, not to create a broad lawful retail market.

The Netherlands: tolerated retail without fully legal production

No European case is more misunderstood than the Netherlands. For decades, headlines have treated Dutch cannabis policy as proof that the country “legalized” cannabis long ago. It did not.

The modern Dutch framework grew out of the 1976 revision of the Opium Act, which formalized a distinction between drugs deemed to pose “unacceptable” risk and those considered less dangerous. This created the legal and political setting for the coffee-shop system. Retail sale of small quantities of cannabis in licensed coffee shops was tolerated under prosecutorial guidelines, even though the underlying conduct remained unlawful in formal statutory terms. Possession of small amounts was deprioritized. Retail storefronts became visible. Tourists noticed. The rest of Europe called it legalization.

But the supply side remained the system’s central contradiction. Coffee shops could sell cannabis at the “front door,” while large-scale cultivation and wholesale supply at the “back door” remained criminal offenses. That mismatch is the Dutch coffee-shop paradox: tolerated retail without a fully lawful production chain. A customer could buy cannabis in a regulated-looking environment, yet the cannabis entering that environment historically came through illicit production and distribution channels.

That is not a technicality. It shaped the entire Dutch experience. The state partially normalized access while leaving organized supply in criminal hands. Critics argued this undercut claims that the model would displace illicit trafficking. Defenders responded that the policy still reduced harms associated with arrest and separated cannabis retail from markets for higher-risk drugs. Both points can be true.

Recent Dutch policy has tried, cautiously, to address the problem. The closed coffee shop chain experiment, launched in selected municipalities, allows regulated cultivation to supply participating coffee shops. It is an attempt to close the back door and test whether a lawful supply chain can replace the tolerated-illicit hybrid. Yet even this remains an experiment, not a full national commercial legalization system. The Netherlands is historically important not because it legalized cannabis in the Canadian sense, but because it showed how far tolerance can go without resolving the legal status of production.

Malta, Luxembourg, and Germany: home grow and noncommercial access models

The newer European reforms are narrower than the phrase “adult-use legalization” suggests. Malta, Luxembourg, and Germany each moved away from simple prohibition, but none created a broad commercial retail market.

Malta’s 2021 reform is often described as making it the first European Union country to legalize adult-use cannabis. That shorthand hides the actual structure. Adults were allowed to possess up to 7 grams, keep up to 50 grams at home, and cultivate up to four plants for personal use. The model also allowed non-profit cannabis associations, later subject to regulation by the Authority on the Responsible Use of Cannabis. This was not a free-market retail system. It was a tightly bounded possession, home-grow, and non-profit collective access model. Malta’s law matters because it broke with prohibition, but it did so through controlled private and associative access, not through commercial storefront sales.

Luxembourg followed a similarly limited path. Its 2023 reform legalized possession in private, within limits, and allowed home cultivation, usually summarized as up to four plants per household. Public possession and public use remained restricted, and no general commercial supply chain was created. So Luxembourg also shifted away from criminal prohibition for private adult conduct, while stopping well short of a Canadian-style market. It is better understood as private-use legalization than as retail legalization.

Germany’s 2024 Cannabis Act, the KCanG, is the largest and politically most important example of this European pattern. As of 1 April 2024, adults may possess up to 25 grams in public and cultivate up to three plants for personal consumption. From 1 July 2024, non-commercial cultivation associations became permissible under regulation and oversight, with BfArM involved in the supervisory framework. That is a substantial legal change in Europe’s biggest economy. It removes many adults from the criminal system for possession and home growing. It creates lawful access through associations. It also imposes limits on age, quantity, geography, and organizational form.

What Germany did not do is just as important. It did not establish a nationwide adult-use retail market. It did not authorize broad commercial production and sale akin to Canada’s Cannabis Act. Media coverage often implied otherwise, partly because Germany had earlier discussed a more expansive plan. The enacted law is narrower. It is best described as legalization of possession and home cultivation, plus regulated non-commercial associations.

That distinction is not pedantic. Wayne Hall’s public-health work has long stressed that outcomes depend on policy architecture: price, potency controls, promotion limits, access rules, and enforcement priorities all shape use patterns and harm. Germany’s model reflects that logic. It is an effort to reduce criminalization while avoiding a full commercial market. Whether it succeeds is a separate question. But legally, it belongs in a different category from Canadian federal legalization.

Spain, Portugal, Czechia, and the broader European grey zone

Much of Europe sits in a grey zone that produces confusion in international reporting. Spain is the clearest example. Cannabis social clubs there are often presented as if they were plainly legal. They are not.

Spanish law historically created space around private consumption and cultivation for personal use, while trafficking remained criminal. In that space, cannabis social clubs emerged as member-based associations claiming to cultivate collectively for private consumption. Some regional authorities, especially in parts of Catalonia and the Basque Country, attempted to regulate them. Courts pushed back. The Spanish Supreme Court issued decisions against broader club operations, and the Constitutional Court struck down aspects of regional regulation. The result is neither straightforward legality nor simple prohibition. Clubs exist, but under legal contest, uneven local practice, and continuing risk. Spain is not a stable legal-access model in the way Uruguay’s club system is under Law No. 19.172.

Portugal is misunderstood in a different way. Since 2001, Portugal has decriminalized possession of small amounts of all drugs for personal use, including cannabis. That reform replaced criminal penalties for possession below threshold levels with administrative responses, usually handled by Commissions for the Dissuasion of Drug Addiction. Trafficking remains criminal. Supply remains unlawful. Decriminalization is not legalization. Portugal removed many users from the criminal courts; it did not create a lawful nonmedical cannabis market.

Czechia also belongs in this category of partial reform. Small-quantity possession has long been treated more leniently than trafficking, and debates over broader legalization recur regularly. Yet a reduced penalty or tolerated possession threshold does not amount to legal regulation of supply. Across Europe, this pattern repeats: less punishment for users, no fully lawful chain of production and sale.

That is the real European story. Reform is real. Prohibition is no longer the only model on the table. But Europe is not rapidly converging on Canada. It is fragmenting into tolerated retail, decriminalization, private-use legalization, home-grow permissions, non-profit associations, medical programs, and legally disputed collective schemes. Headlines flatten those differences because “legalization” is catchy. Law does not work that way.

Latin America and the Caribbean: constitutional rights, anti-trafficking politics, and uneven implementation

Latin America did not move on cannabis through a single reform script. Courts, congresses, health ministries, and security agencies have pushed in different directions at different times. That matters because the region is often described as “progressive” on cannabis when the legal reality is much messier: some countries reduced penalties for possession, some built medical systems, one created a national adult-use framework, and many still leave users and patients exposed to police discretion despite formal reform.

The region’s legal politics also differ from North America’s retail-led story. As John Walsh has argued in work on drug policy in the Americas, reform debates in Latin America have often been tied to the failures of militarized anti-trafficking policy, prison overcrowding, and constitutional protections for privacy and personal autonomy. Those drivers produce laws that look very different from a commercial legalization model.

Uruguay's Law 19.172 and the world's first national adult-use system

Uruguay is the clearest example of why “legalization” needs definition. Law No. 19.172, enacted in December 2013, made Uruguay the first country to legalize nonmedical cannabis nationwide, but it did not set out to create a broad consumer market. President José Mujica’s government framed the reform as a public-security intervention aimed at weakening illicit trafficking, taking cannabis out of the hands of criminal networks, and bringing users into a monitored legal system.

That design choice shaped everything that followed. Adults were allowed access through three tightly bounded channels: home cultivation, membership in cannabis clubs, and purchases through pharmacies. Registration with the state became part of the system. Potency, supply, and distribution were all subject to state oversight. This is much closer to Mark A.R. Kleiman’s idea of regulated legalization as a control strategy than to a commercial model built around broad product choice and market expansion.

Uruguay therefore stands apart from Canada’s federally regulated commercial framework and even more so from the fragmented state-level systems in the United States. Beau Kilmer’s work on cannabis policy design is useful here: who may produce, how much competition is allowed, whether branding is permitted, and how supply is tracked are not side details. They define the regime. Uruguay chose restraint.

The country’s model also exposed the limits of legal change by statute alone. Pharmacy participation was slow. Banking barriers linked to international financial compliance reportedly discouraged some actors from involvement. Registered access took time to scale. So even the world’s first national adult-use law did not produce instant, frictionless legal access. It produced a state-managed alternative to prohibition, built gradually and with visible bottlenecks.

Medical and decriminalization models across Latin America

Outside Uruguay, most Latin American reforms fit into two other categories: decriminalization of personal possession, and medical cannabis frameworks. Neither is the same as adult-use legalization.

Colombia is a good example of layered reform. The Constitutional Court decriminalized possession of a personal dose in 1994, grounding that decision in rights-based reasoning. Later reforms addressed medical cannabis, especially Decree 2467 of 2015 and Law 1787 of 2016, which created a licensing framework for cultivation, production, and medical use. Colombia has often been portrayed as near-legalization. That overstates the case. Personal possession protections and a medical licensing regime do not create a lawful adult-use supply chain.

Argentina followed a different path. Law 27.350 of 2017 established a medical cannabis framework focused at first on research and limited patient access. Regulation broadened under Decree 883/2020, which expanded conditions for access and recognized home cultivation for registered medical users through the REPROCANN system. Again, this is not adult-use legalization. It is a patient-centered exception to prohibition, and access still depends heavily on administration.

Mexico sits in an even more legally unstable category. A series of Supreme Court rulings culminated in 2021 with the court invalidating the absolute prohibition on personal adult use, effectively forcing recognition of a right to consume and cultivate for personal purposes under permit. But Congress did not pass a full regulatory law. The result is not a settled legal market. It is constitutional deprohibition without a complete statutory supply framework.

Elsewhere in the region, partial decriminalization has long coexisted with criminal enforcement. Brazil’s 2006 drug law removed prison terms for possession for personal use, yet it left police and courts wide latitude to distinguish users from traffickers. That distinction has been applied unevenly, often with class and racial bias. Jamaica amended its Dangerous Drugs Act in 2015 to decriminalize possession of small amounts and create a medical, therapeutic, and sacramental framework, particularly relevant to Rastafarian religious use. Chile permits some forms of private personal use and has a medical market, but cultivation and supply remain legally contested in practice. Peru and Ecuador also provide examples of possession thresholds or reduced penalties without full legalization.

Why legal text and practical access often diverge

This gap between formal reform and lived reality is one of the region’s defining features. On paper, a law may decriminalize small amounts, authorize prescriptions, or create licensing channels. In practice, patients wait months for approvals, domestic production lags, imported products remain expensive, and police continue to stop, search, and arrest people under older enforcement habits.

Brazil shows the problem sharply. Even where courts have moved toward tolerance for personal possession, ambiguous quantity standards have let enforcement fall back on officer judgment. That means legal status is filtered through street-level discretion. A middle-class patient with a lawyer may obtain a court order for medical access. A poor young person carrying cannabis may still be treated as a trafficker.

Medical systems face another set of implementation failures. Ministries may legalize cannabis-derived products but authorize only a narrow list, often imported and costly. WHO’s Expert Committee on Drug Dependence has recognized therapeutic value in certain cannabis-related preparations, especially CBD and standardized cannabinoid medicines, yet many Latin American systems still do not translate that recognition into broad, affordable access. Argentina’s REPROCANN expanded lawful home cultivation for patients, but bureaucracy and inconsistent enforcement have remained problems. Colombia built a sophisticated licensing structure, but patient access has not always matched the ambition of the legal text.

That is the larger regional lesson. Latin America and the Caribbean do not present a simple march from prohibition to legalization. They show a fragmented field of constitutional rulings, anti-trafficking reforms, medical exceptions, partial decriminalization, and one distinctive state-controlled adult-use system in Uruguay. Calling all of that “legalization” erases the point. The law on the books, the institutions that administer it, and the behavior of police and courts are three separate questions. In this region, they often point in different directions.

Africa and Asia-Pacific: reform at the margins, prohibition still dominant

English-language legalization roundups often spend pages on Canada, the United States, Germany, and Uruguay, then reduce Africa and Asia-Pacific to a few vague lines about “emerging reform.” That misses the legal reality. These regions matter precisely because they show how little the word legalization explains on its own. South Africa is not Canada. Morocco is not Uruguay. Australia is not moving like Germany. Thailand’s turn after 2022 did not settle anything; it exposed how fast cannabis policy can swing when lawmakers remove criminal controls before building a stable regulatory structure.

In both regions, prohibition still sets the baseline. Reforms exist, sometimes significant ones, but they are usually narrow, conditional, and politically fragile.

South Africa's private-use framework and Morocco's medical-industrial cultivation law

South Africa is one of the clearest examples of a model that many summaries mislabel. The key turning point was not the creation of a legal adult-use market. It was a constitutional ruling on privacy. In Minister of Justice and Constitutional Development v Prince (Constitutional Court, 2018), the court held that criminal prohibitions on private possession, use, and cultivation of cannabis by adults were unconstitutional to the extent they intruded on the right to privacy. That was a major break with prohibition. It was not a green light for ordinary retail sales.

Parliament later enacted the Cannabis for Private Purposes Act, 2024, which gives statutory form to that private-use approach. Adults may possess cannabis in private and cultivate it for personal use, within legal limits. The structure matters. This is possession-and-home-grow reform, anchored in privacy rights, not a broad commercial supply system. Mark A.R. Kleiman and Beau Kilmer both argued, in different ways, that design choices define outcomes; South Africa proves the point. A system can relax criminal penalties without authorizing a legal market.

Morocco’s 2021 reform is different again. Law No. 13-21, adopted in 2021, legalized and regulated cannabis cultivation for medical, cosmetic, and industrial purposes under state supervision. It did not legalize adult nonmedical use. The law is often linked to Morocco’s long history of cannabis cultivation in the Rif region, but the modern framework is tightly bounded: licensed production, specific end uses, and regulatory oversight through the National Agency for the Regulation of Cannabis-Related Activities.

That distinction matters politically and legally. Morocco’s reform was framed less as a civil-liberties measure than as an agricultural, industrial, and public-administration project. It also sits more comfortably with the post-2020 international climate, after the UN Commission on Narcotic Drugs removed cannabis and cannabis resin from Schedule IV of the 1961 Single Convention while keeping them in Schedule I. Symbolic, yes. Full international legalization, no.

Australia and New Zealand: medical expansion without broad adult-use legalization

Australia has built one of the larger medical cannabis systems outside Europe and North America, yet adult-use remains mostly prohibited. That split is easy to miss if every softening of policy gets filed under legalization.

At the federal level, Australia legalized cultivation and manufacture for medical and scientific purposes through the Narcotic Drugs Amendment Act 2016. Patient access then expanded through the Therapeutic Goods Administration pathways, especially the Special Access Scheme and Authorised Prescriber system. Prescribing numbers climbed sharply in the early 2020s. Australia now looks, in practical terms, like a major medical-cannabis jurisdiction.

But medical access does not equal legal adult use. Outside one territory, possession and supply for nonmedical purposes remain criminal offenses under state and territory law, though penalties differ. The outlier is the Australian Capital Territory, where from 31 January 2020 adults were allowed limited possession and home cultivation for personal use. Even there, the reform is narrow and legally awkward because Commonwealth law still criminalizes possession. John Walsh’s broader point about fragmented reform applies here too: subnational tolerance can coexist with higher-level prohibition.

New Zealand took a different route and then stopped. The country established a medical cannabis scheme under regulations that took effect in 2020, allowing licensed manufacture and prescribing within a controlled framework. Yet adult-use legalization was rejected in the 2020 cannabis referendum, with the “no” side prevailing by a small margin. So New Zealand is not a case of failed prohibition giving way to full legalization. It is a case of medical reform proceeding while nonmedical legalization remains politically contested and legally blocked.

Wayne Hall’s work on cannabis policy has long stressed public-health tradeoffs rather than ideological absolutes. Australia and New Zealand fit that pattern: cautious medical liberalization, persistent criminal controls, and little appetite for a national commercial adult-use model.

Thailand, Japan, and the volatility of reform in Asia

Asia is often portrayed as uniformly prohibitionist. That is too simple, but the region is still much closer to prohibition than to stable legalization.

Thailand became the headline case because it moved fast. In June 2022, the government removed cannabis from the national narcotics list. That change, combined with weak transitional controls, opened space for widespread nonmedical availability in practice. International coverage often called this legalization. Legally, the picture was messier. Thailand had loosened prohibition dramatically, especially around possession and plant status, without first locking in a settled adult-use law comparable to Canada’s Cannabis Act or Uruguay’s Law No. 19.172.

The result was instability. Successive governments proposed tighter rules, including restricting use to medical purposes and re-criminalizing parts of the system. That attempted retrenchment is not a side note; it is the story. Thailand shows what happens when reform outruns institutional design. Beau Kilmer has repeatedly argued that market architecture determines policy effects. Thailand’s post-2022 experience is evidence for that claim. Remove a drug from the narcotics list without a durable framework, and politics rushes back in.

Japan sits at the opposite end of the spectrum. The country has long enforced some of the strictest cannabis controls in the industrialized world under the Cannabis Control Act of 1948. Possession remains criminal. So does unauthorized cultivation. Yet even Japan is changing, though cautiously and mostly on the medical-scientific side. In 2023, lawmakers passed revisions allowing the use of cannabis-derived pharmaceuticals, including products containing cannabinoids, while also moving toward clearer criminalization of cannabis use itself. That is not liberalization in the ordinary sense. It is selective medical accommodation inside a still-punitive system.

Taken together, Africa and Asia-Pacific do not support a simple story of global legalization spreading outward from a few pioneer states. They show fragmentation instead: private-use constitutionalism in South Africa, licensed medical-industrial cultivation in Morocco, medical expansion without adult-use reform in Australia and New Zealand, volatile liberalization in Thailand, and highly selective pharmaceutical change in Japan. Prohibition is no longer absolute everywhere. It is still dominant.

Medical cannabis frameworks: access, evidence, and the gap between law and reality

Medical cannabis law is often presented as a halfway house between prohibition and adult-use legalization. That framing hides more than it explains. A country can authorize cannabis for patients while keeping nonmedical possession criminal, and the medical scheme itself can range from a narrow pharmaceutical exception to a broad system allowing herbal flower, oils, extracts, and long-term prescribing. “Medical cannabis is legal” may describe a real statute. It does not automatically describe meaningful access.

The distinction matters because medical reform usually turns on evidence standards, prescribing rules, and health-system finance, not on the political arguments that drive adult-use legislation. The World Health Organization’s Expert Committee on Drug Dependence has recognized therapeutic value in some cannabis-related preparations, especially cannabidiol and standardized cannabinoid medicines, and that recognition fed into the UN Commission on Narcotic Drugs decision in December 2020 to remove cannabis and cannabis resin from Schedule IV of the 1961 Single Convention while leaving them in Schedule I. Symbolically important, yes. It did not create a common medical model. States still decide what counts as acceptable evidence, who may prescribe, and who pays.

Pharmaceutical cannabinoids versus herbal cannabis programs

Some countries allow only pharmaceutical cannabinoids. Others permit magistral preparations or imported herbal cannabis. Others again let physicians prescribe dried flower within a regulated medical framework. These are not minor administrative variations; they create different patient populations and very different levels of actual use.

The classic pharmaceutical route relies on licensed medicines that passed the ordinary drug-approval process for specific indications. Examples include nabiximols, marketed as Sativex, a standardized oromucosal spray containing THC and CBD, and purified cannabidiol as Epidiolex or Epidyolex. This model fits most cleanly with conventional evidence-based medicine because dose, composition, and indication are defined. Health authorities tend to be more comfortable here. The drawback is obvious: approved indications are narrow, prices can be high, and many patients who seek symptom relief from pain, spasticity, chemotherapy-related nausea, or refractory anxiety fall outside the label.

Herbal cannabis programs are broader but more politically and clinically contested. Germany before its 2024 adult-use reform is a good example of a large herbal medical system inside a still-prohibitionist nonmedical framework. Since the 2017 “Cannabis as Medicine” law, physicians could prescribe cannabis flowers and extracts for serious illness in limited circumstances, and statutory insurers could reimburse after prior approval. That looked liberal on paper. In practice, it depended on physician willingness, insurer approval, and product availability. It was not adult-use by another name.

Israel has long operated one of the world’s most developed medical cannabis systems without full nonmedical legalization. Its framework historically combined licensed producers, defined indications, and specialist oversight, while allowing substantial patient numbers and a wider therapeutic culture than seen in many European systems. Australia also illustrates the split. It has a large and growing national medical pathway, mainly through the Therapeutic Goods Administration’s Special Access Scheme and Authorised Prescriber scheme, yet adult-use cannabis remains broadly illegal outside the Australian Capital Territory’s limited personal-use carve-out. Again, medical access did not mean general legalization.

The UK sits closer to the restrictive end. In November 2018 it rescheduled cannabis-based medicinal products so specialist doctors could prescribe them legally. That headline suggested a major opening. The actual model remained narrow, with strong preference for licensed products or specialist justification, and very limited National Health Service prescribing.

Prescription standards, specialist gatekeeping, and reimbursement barriers

The central access problem in medical cannabis is not always criminal law. Often it is clinical gatekeeping. Many systems require specialist initiation rather than ordinary primary-care prescribing. That sharply reduces access, especially for chronic conditions managed mostly by general practitioners.

The UK shows this with unusual clarity. Legal since 2018, yes. But routine NHS access remains tiny because National Institute for Health and Care Excellence guidance has been cautious, largely limiting recommendations to a few indications such as severe treatment-resistant epilepsy, chemotherapy-induced nausea and vomiting, and multiple-sclerosis spasticity for certain products. For chronic pain, NICE did not recommend cannabis-based medicinal products in its 2019 guideline, citing insufficient evidence of cost-effective benefit. That position shaped prescriber behavior. Families and patients often turned to private clinics because legality alone did not produce mainstream care.

Germany’s pre-2024 medical system was more permissive than the UK’s but still burdened by reimbursement friction. Statutory insurers could reject applications, and physicians had to justify use where standard therapies were unavailable or unsuitable. This mattered because reimbursement determines whether medical law is a real treatment route or a formal privilege for patients who can absorb sustained out-of-pocket costs. Once coverage becomes uncertain, prescribing becomes administratively risky and slower.

Australia offers another version of the same problem. Access expanded through special schemes rather than ordinary medicine registration for most products. Prescriptions increased rapidly, yet the pathway often sat outside standard subsidy arrangements under the Pharmaceutical Benefits Scheme. A doctor might be legally able to prescribe. The patient might still face high recurring costs. Legal access without affordable access is a thin form of legality.

Selected EU states reveal the same pattern. Some allow magistral preparations or tightly bounded prescriptions but make doctors reluctant through paperwork, weak clinical guidance, or fear of professional scrutiny. Others formally permit use while limiting approved indications so narrowly that patient numbers stay small. Beau Kilmer’s work on policy design helps here: the architecture of a cannabis regime matters as much as the formal permission. Mark Kleiman made a similar point in a different register—law on the books and incentives in practice are rarely the same thing.

Why “medical cannabis is legal” can be technically true and practically misleading

This phrase is misleading because it compresses at least four separate questions into one. Is there a lawful product? Who can prescribe it? For which conditions? Who can realistically obtain and pay for it?

Take the UK. Medical cannabis is legal, but most patients will not receive a routine NHS prescription. Take Germany before April 2024. Medical cannabis was legal and nationally regulated, yet access still depended on doctor willingness, insurer approval, and documentation burdens. Take Australia. Legal pathways exist at scale, but cost and prescriber practices shape who actually benefits. Take Israel. A comparatively mature medical framework exists, but it remains a medical framework, not evidence that the state has embraced adult-use legalization.

WHO and national health authorities have been more conservative than campaign rhetoric. They generally accept evidence for certain cannabinoid medicines in certain indications. They are much less willing to say that all cannabis forms have established efficacy across broad symptom categories. That caution is not simply bureaucratic delay. It reflects ordinary evidence thresholds in medicine: randomized controlled trials, product standardization, known dose, and pharmacovigilance. The evidence base is strongest for some specific products and weaker for whole-plant claims made across heterogeneous conditions. Wayne Hall has argued for years that cannabis policy should be judged by public-health outcomes rather than symbolism. Medical law is where that argument bites hardest.

So medical reform should be treated as its own legal family. It often precedes adult-use reform, but it does not predict it. It can function as a narrow pharmaceutical exception, a broad herbal access system, or a highly restrictive specialist program that exists mostly in statute books and ministerial press releases. That is why “medical cannabis is legal” is a starting point for analysis, not an endpoint.

Country-by-country legal status: a practical classification system

“Cannabis is legal” usually tells you far too little to be useful. A country may allow adult possession but no lawful retail supply. It may permit only specialist-prescribed medicines. It may tolerate storefront sales while keeping wholesale production formally illegal. Those are different legal worlds, with different risks, rights, and enforcement patterns.

That distinction matters because the global baseline is still prohibition under the 1961 UN Single Convention on Narcotic Drugs, reinforced by the 1971 and 1988 treaties. The December 2020 vote by the UN Commission on Narcotic Drugs to remove cannabis and cannabis resin from Schedule IV of the 1961 Convention did not create international legality; it only recognized that cannabis has medical value and should not sit in the treaty’s most restrictive category. Against that backdrop, national systems have splintered into separate regulatory families rather than marching in one direction.

UNODC estimated in its World Drug Report 2024 that 228 million people used cannabis in 2022. Reform is happening in a high-use environment, not at the margins. Yet, as Beau Kilmer and Mark A.R. Kleiman argued in different ways, policy design matters as much as the yes-or-no question of legalization. The practical map below groups countries by model, because that is how the law is actually lived.

Adult-use legal with national framework

Canada — Adult-use cannabis became legal nationwide on 17 October 2018 under the Cannabis Act. This is the clearest example of a federally regulated nonmedical system among major economies: licensed production, provincial retail rules, product standards, packaging controls, and national criminal-law amendments. It is not prohibition-lite. It is full legal regulation, even though provinces differ on retail structure and public-consumption rules. Statistics Canada reported in 2024 that 26% of people aged 16+ used cannabis in the previous 12 months in Q1 2024, and legal channels accounted for about 72% of household cannabis expenditure in 2023. Canada shows that federal legalization can displace illicit supply significantly, though not erase it.

Uruguay — The first country to legalize nonmedical cannabis nationwide, through Law No. 19.172 in 2013. Uruguay’s model is tightly controlled: home cultivation, membership cannabis clubs, and pharmacy dispensing under state supervision. It was designed to weaken illicit trafficking, not to create a broad private retail sector. Calling Uruguay “just like Canada” is wrong. Wayne Hall’s public-health framing fits here: legalization can be structured to limit commercialization, and Uruguay is the clearest case.

Germany — Since 1 April 2024, the Cannabis Act (KCanG) has allowed adults to possess up to 25 grams in public, store limited amounts at home, and cultivate up to three plants for personal use. From 1 July 2024, non-commercial cultivation associations became possible under regulation, with oversight involving authorities including BfArM. Germany did legalize adult use in a bounded sense. It did not create a Canadian-style retail market. That difference is central.

Malta — Malta’s 2021 reform is often compressed into “the first EU country to legalize cannabis.” That shorthand hides the structure. Adults may possess up to 7 grams, cultivate at home within limits, and access cannabis through regulated non-profit associations. There is no broad commercial adult-use retail model. Malta belongs in the legal adult-use family, but in its own subcategory: possession, home grow, and association access, not open-market legalization.

Luxembourg — Since 2023, adults may possess limited amounts in private and grow a small number of plants at home. Public possession remains restricted and there is no national commercial retail chain. Luxembourg therefore sits closer to Germany and Malta than to Canada. It legalized defined adult conduct. It did not legalize a consumer market.

South Africa — South Africa is often mislabeled “legal” in broad terms after the Constitutional Court’s 2018 Prince judgment and later legislation including the Cannabis for Private Purposes Act. The reality is narrower. Adult private possession and cultivation for personal use are protected within limits, but this is not a general adult-use retail framework. It is best understood as a constitutional private-use model.

Medical legal only or highly restricted cannabinoid access

Australia — Australia has a large and expanding medical cannabis system under the Therapeutic Goods Administration pathway, especially through the Special Access Scheme and Authorized Prescriber route. Patients can lawfully access prescribed cannabis products, including flower in many cases, but adult nonmedical use remains mostly illegal nationwide. The Australian Capital Territory has a limited personal-use carve-out for adults, which complicates the picture, but Australia as a whole is still medical-only rather than adult-use legal.

Israel — Israel is one of the oldest and most scientifically influential medical cannabis jurisdictions, with a structured patient program and a long research tradition. It is often treated as more liberal than it is. Adult nonmedical use remains illegal, though enforcement priorities have shifted at times. Israel belongs firmly in the medical-access category, not the legal adult-use category.

United Kingdom — Since 2018, cannabis-based medicinal products can be prescribed under rescheduling changes, but access through the NHS remains narrow in practice. This is a classic example of legality on paper combined with limited real-world availability. Most patients do not have broad routine access, and adult nonmedical use remains prohibited.

Morocco — Morocco legalized cannabis cultivation in 2021 for medical and industrial purposes under a licensing system. That reform did not legalize adult nonmedical use. Given Morocco’s historic role in cannabis production, this is a major policy shift, but it is sector-specific and tightly bounded.

Thailand — Thailand removed cannabis from the narcotics list in 2022, producing one of the most confusing legal environments in the world. For a period, access expanded dramatically and many observers described the country as having legalized cannabis. That overstated the legal certainty. The Thai system has been marked by ministerial rules, contested enforcement, and repeated efforts to restrict nonmedical use again. In practical classification terms, Thailand has had broad liberalization but unstable legal footing; it is better placed in a medically framed or highly restricted-access category than treated as settled full legalization.

Mexico — Mexico does not fit neatly anywhere, but it is not yet a fully legalized national adult-use system. Supreme Court rulings found the absolute ban on personal use unconstitutional in key respects, and administrative pathways for personal cultivation permits emerged, yet Congress has not enacted a complete national regulatory framework for nonmedical supply. Medical cannabis exists in law. Adult-use legality exists in fragments. John Walsh has long argued that Latin American reform often moves through constitutional and judicial routes before legislatures catch up; Mexico proves the point.

Decriminalized, tolerated, or legally ambiguous systems

Netherlands — The Dutch model dates to the 1976 Opium Act revision and the soft/hard drug distinction. Coffee shops may sell small amounts under a policy of tolerance, but this has long coexisted with the famous “back door problem”: retail sales tolerated, supply to those outlets not fully legalized. The Netherlands is not a full legalization state. It is the classic tolerated-retail system, now slowly experimenting with regulated supply in pilot form.

Spain — Spain’s cannabis social clubs operate in a legally ambiguous space shaped by private-consumption doctrine, regional practice, and periodic court crackdowns. Private use in private spaces is treated differently from public possession, and supply remains legally precarious. Spain is not legal adult use in any clean national sense.

Portugal — Portugal decriminalized possession of small quantities of all drugs in 2001. That did not legalize cannabis. Possession for personal use below threshold amounts is generally handled as an administrative matter rather than a criminal one, while trafficking and supply remain criminal offenses. Portugal is the textbook example of why decriminalization and legalization are not synonyms.

United States — The US is the most important example of legal fragmentation. At the federal level, cannabis remains Schedule I under the Controlled Substances Act of 1970. At the state level, adult-use legalization now covers roughly half the country, and medical cannabis reaches more states still. SAMHSA reported in 2023 that 61.8 million people aged 12 or older used marijuana in the past year. But state legality does not cancel federal consequences involving immigration, firearms possession, banking access, tax treatment under Internal Revenue Code 280E, or interstate transport. The US is neither prohibition in the old uniform sense nor national legalization. It is a federal conflict system.

Strict-prohibition states — Many countries still maintain criminal prohibition with very limited or no recognized lawful access outside tiny pharmaceutical exceptions. Japan remains highly restrictive, even as rules around cannabinoid medicines have begun to shift. Singapore, Indonesia, Philippines (despite periodic reform proposals), and many Middle Eastern states continue to treat possession, use, and supply as serious criminal matters, often with severe penalties. In these jurisdictions, “medical cannabis is being discussed” should not be mistaken for legal access on the ground.

This classification is more useful than a simple legal/illegal map because it matches the real policy split now taking shape worldwide. The dominant movement is not from prohibition straight to commercial legalization. It is from absolute ban toward mixed systems: medical pathways, private-use rights, non-profit associations, administrative decriminalization, tolerated retail, and only in a smaller set of countries, fully lawful adult-use regulation.

The policy arguments now shaping reform

The live argument is no longer whether prohibition has “worked.” In many places, the baseline facts have already answered that. UNODC estimated 228 million cannabis users worldwide in 2022, and Europe alone had about 24 million adults reporting past-year use in 2024. The harder question is what kind of post-prohibition or semi-post-prohibition system produces fewer harms. That is why “legalization” as a single label obscures more than it explains. Uruguay’s state-controlled model, Canada’s federally regulated adult-use market, Germany’s 2024 possession-and-association law, Malta’s non-profit association system, and US state legalization under federal Schedule I are not variations on one template. They embody different priorities, and they are judged by different outcomes.

Wayne Hall has been one of the clearest voices on this point for years: legal access may reduce criminal penalties and improve product control, but it can also increase heavy use if policy encourages a large, high-potency market. Beau Kilmer has pushed the same debate in a more institutional direction. The key policy choice is not simply legal versus illegal; it is how production, potency, price, promotion, retail access, home growing, and enforcement are structured. On the evidence so far, that is the right frame. Policy design matters at least as much as the yes/no legalization question.

Public health: youth use, dependence, impaired driving, and product potency

Public-health debates are often reduced to one slogan from each side. That misses the real disputes. Most serious analysts are asking narrower questions: Do youth-use rates rise after reform? Does daily or near-daily use increase among adults? Do emergency presentations change? Are products becoming stronger? Does impaired driving enforcement improve or get murkier?

Youth use is still the headline concern, but the evidence is mixed rather than cleanly polarizing. In North America, some post-legalization studies have found little clear increase in adolescent use, which undercuts one standard anti-reform claim. That does not settle the matter. Stable prevalence can coexist with changes in frequency, modes of use, or potency exposure. A teenager who uses less often but consumes high-THC extracts is not captured well by crude “past year use” metrics.

Dependence and heavy use matter more than occasional use for health burden. EMCDDA’s 2024 report estimated 4.3 million Europeans are daily or almost daily users, and cannabis accounted for 36% of treatment admissions in Europe in 2022. Those figures should not be waved away. They are a reminder that reform debates happen in the presence of a real treatment burden. Hall has argued that the likely public-health cost of wider legal access is not mass addiction in the caricatured sense, but a larger population of frequent users exposed to higher risks of dependence, psychotic symptoms in susceptible people, and poorer mental-health outcomes.

Potency is now central. Legal regulation can improve labeling and contaminant testing, which is a public-health gain. It can also normalize very strong products unless governments cap THC, tax by potency, or impose package and serving restrictions. Jurisdictions that legalize without addressing concentrates, edible dosing, and high-potency flower are not practicing harm reduction; they are improvising.

Impaired driving is another unresolved area. Per se blood-THC limits are controversial because THC does not map onto impairment as neatly as blood alcohol concentration. Legalization can reduce one harm while making another harder to police if roadside testing and public education lag behind. This is why the strongest public-health position is not “keep prohibition” or “legalize and relax.” It is regulate tightly where legal, collect data fast, and revise rules when product strength or patterns of harm shift.

Germany’s 2024 Cannabis Act shows how this logic differs from a commercial market model. Adults may possess limited amounts and grow plants at home, and non-commercial cultivation associations are allowed under regulation, but there is no Canadian-style retail chain. That design appears aimed, at least in part, at reducing criminalization without fully opening the door to aggressive market expansion. Whether it succeeds will depend on enforcement, product access, and substitution away from illicit supply.

Criminal justice and social equity arguments

The criminal-justice case for reform is stronger than many opponents admit. If a drug is widely used across society, criminal enforcement does not fall evenly. It concentrates. That means arrests, records, searches, and collateral penalties for conduct that many governments no longer regard as warranting the full force of criminal law.

This is where decriminalization and legalization diverge sharply. Portugal’s 2001 decriminalization reduced criminal penalties for possession, but it did not create lawful supply. That can lower arrests without displacing trafficking. Adult-use legalization can go further by replacing illegal supply channels with lawful ones, but only if the legal channel is broad enough and accessible enough to compete. A possession-only reform may ease one injustice while leaving the underlying market untouched.

Social-equity claims are harder. Expungement, reduced arrests, and fewer police encounters are measurable benefits. Promises that legalization will repair the long afterlife of prohibition are much less certain. In the United States, state systems have often struggled to turn equity language into durable outcomes, especially where licensing rules, local bans, capital requirements, and federal illegality distort participation. The lesson is not that equity goals are misguided. It is that they cannot be achieved by rhetoric alone.

Mark A.R. Kleiman long argued for regulated legalization designed to minimize both punishment and commercialization. That remains a useful benchmark. Reform should be judged first on whether it cuts criminal justice harm: fewer arrests, fewer prosecutions, fewer racial disparities in enforcement, fewer lifelong records for low-level possession. On that metric, even narrow reforms can matter a great deal.

Tax revenue, illicit-market displacement, and regulatory realism

Tax revenue gets attention because it is easy to count, but it is often the least important metric. A government can collect revenue and still fail on public health or illicit-market displacement. Canada offers a better benchmark than the raw tax story: by 2023, legal channels accounted for about 72% of household cannabis spending. That is a substantial shift away from illicit supply. Not total victory. Still substantial.

This is where Beau Kilmer’s work is especially useful. If legal product is too expensive, too weak, too restricted, or too inconvenient, many users stay with illegal suppliers. If it is too cheap, too potent, and too widely available, consumption may rise more than policymakers intended. There is no magic point. There is only market design.

Uruguay recognized this early. Law No. 19.172 was built to undercut illicit trafficking, not to create a large commercial sector. The Netherlands shows the opposite regulatory contradiction: tolerated retail sales without full production legalization created the famous “back door problem,” where supply remained in a gray zone. Germany’s 2024 model avoids full commercialization, but that may limit illicit-market displacement if association access proves narrow. Malta and Luxembourg have made similar tradeoffs.

Regulatory realism means accepting that governments are choosing among imperfect systems, not between order and chaos. A well-designed reform can reduce arrests, test products, weaken criminal groups, and keep youth use stable. A poorly designed one can legalize on paper while preserving illicit supply and creating new public-health risks. That is the real debate now. Not whether cannabis policy is liberalizing in some abstract sense, but which legal architecture can actually move the numbers that matter.

What international law still allows, constrains, and leaves unresolved

The modern legal argument over cannabis still runs through treaties written for a prohibitionist age. That matters because many headlines imply that once a country changes domestic law, the international question disappears. It does not. The core UN drug conventions still set a baseline that is much stricter than Canada’s 2018 adult-use market or Uruguay’s 2013 Law No. 19.172. They also leave enough room, or at least enough ambiguity, for governments to move anyway.

UNODC’s World Drug Report 2024 estimated that 228 million people used cannabis in 2022. So this is not a marginal issue in treaty law. It is a stress test for the whole system.

Treaty obligations under the UN drug control conventions

The legal starting point is the 1961 Single Convention on Narcotic Drugs, later reinforced by the 1971 Convention on Psychotropic Substances and the 1988 Convention Against Illicit Traffic in Narcotic Drugs and Psychotropic Substances. Cannabis and cannabis resin remain under international control. The December 2020 vote by the UN Commission on Narcotic Drugs removed cannabis and cannabis resin from Schedule IV of the 1961 Convention, but left them in Schedule I. That was symbolically important because Schedule IV was reserved for especially dangerous substances with little or no therapeutic value. It was not an international legalization event.

The treaty architecture still expects states to limit cannabis to medical and scientific purposes. That phrase is the constraint that keeps appearing in legal analysis. A fully regulated nonmedical retail market sits uneasily with it, and in many readings flatly conflicts with it. This is why “legalization” is too blunt a term. Germany’s 2024 Cannabis Act, Malta’s non-profit association model, Luxembourg’s possession and home-grow reform, the Netherlands’ retail tolerance policy, and US state legalization under continuing federal prohibition all relate to the treaties differently because they legalize different things.

The conventions are stricter on supply than on possession. Decriminalizing possession or replacing criminal penalties with administrative sanctions has long been easier to defend than authorizing a nationwide commercial adult-use chain. Portugal’s 2001 decriminalization is a classic example: less punishment for use and possession, but no legal nonmedical supply. By contrast, Canada’s federal Cannabis Act created a lawful national market for nonmedical production and distribution. That is much harder to square with the treaty text.

How reforming states justify legal changes anyway

States that depart from the old model rarely admit they are simply ignoring international law, even when that is functionally close to what is happening. Instead they use several lines of argument.

One is constitutional law. South Africa’s private-use reform grew out of Constitutional Court reasoning about privacy. Mexico’s Supreme Court decisions struck at the constitutionality of banning personal use, even though Congress still has not enacted a complete regulatory framework. These moves do not solve the treaty issue, but they shift the hierarchy of norms at home.

Another is human-rights reasoning. Uruguay defended its 2013 law as a security and public-policy measure aimed at reducing harms linked to illicit trafficking. That did not make the treaty conflict vanish. It did show how states can frame reform as protection of health, safety, and rights rather than simple permissiveness. John Walsh’s work on Latin American reform has tracked this shift well: the debate is often less about “freeing” cannabis than about limiting the damage done by punitive enforcement.

A third route is narrow design. Germany’s 2024 framework did not create a Canadian-style retail market. Adults may possess up to 25 grams in public and grow up to three plants, with non-commercial cultivation associations allowed from 1 July 2024 under regulatory oversight. Malta and Luxembourg took a similar path. These models appear calibrated to reduce criminalization while avoiding a large commercial supply system that would more directly collide with treaty expectations.

Then there is simple policy divergence backed by enforcement discretion. The Netherlands tolerated coffee-shop retail for decades without fully legalizing production, producing the famous “back door” contradiction. In the United States, federal law still classifies cannabis as Schedule I under the Controlled Substances Act of 1970, yet two dozen states plus DC allow adult use under state law. That split survives through politics and prosecutorial choice, not doctrinal elegance. It leaves real consequences in banking, tax, immigration, firearms, and interstate commerce.

Researchers such as Beau Kilmer and Mark A.R. Kleiman have been right to insist that market structure is the real legal question. Not whether a place is “legal,” but legal for whom, for what amount, through what supply route, and under what enforcement logic.

The next decade: more adult-use markets or more limited home-grow models

The evidence points away from one dominant endpoint. Canada remains the clearest example of a mature federally regulated market, with Statistics Canada reporting that legal channels accounted for about 72% of household cannabis spending in 2023. Uruguay remains tightly state-shaped. Germany, Malta, and Luxembourg point in another direction: possession legal, home grow allowed, non-profit or association supply tightly bounded, no broad retail system. The Netherlands still occupies its own category. The United States is a federal-state contradiction that may persist for years.

Wayne Hall has long argued that legalization debates must be judged against public-health tradeoffs, not slogans. Europe’s 2024 EMCDDA report underscores why: 24 million adults used cannabis in the last year, 4.3 million were daily or almost daily users, and cannabis accounted for 36% of treatment admissions in 2022. Governments see demand. They also see risk.

So the next decade is unlikely to produce one global march toward commercial adult-use markets. More probable is a spread of limited legalization models that permit possession, home cultivation, and perhaps collective non-profit supply while stopping short of full retail commercialization. Some countries will still choose regulated sales. Many will not.

That is the unresolved treaty question and the likely future direction. International law still constrains full nonmedical commercialization, but states have shown they can move through constitutional rulings, human-rights claims, narrow statutory design, discretionary enforcement, or open divergence. The result is not convergence. It is the formation of several durable legal families.