By L. C. Praxis · May 2026
I. The Observation
In April 2026, a man was shot and killed at Lørenbanen in Økern, Oslo. Two Swedish teenagers were charged. The police believe it was a contract killing (NRK, 2026a). In September 2025, two 13-year-olds threw a hand grenade at a nail salon in Pilestredet, linked to the Swedish network Foxtrot (NRK, 2025). In June 2025, a man was shot at Rommen (VG, 2025).
The incidents differ, but share a common thread. The police have mapped approximately 120 criminal networks in Oslo, of which roughly 50 operate within violence, weapons and narcotics (Politiet, 2025). According to the Police Threat Assessment for 2026, criminal networks in all police districts are largely financed by the drugs trade. Drug prices in Norway are among the highest in Europe, making the Norwegian market attractive to transnational networks (Politiet, 2026).
The government's response has primarily been to strengthen the police. Gjengpakke 2 (the second gang crime package) allocates NOK 2.8 billion to combating criminal networks and preventing youth crime (Regjeringen, 2024). The police report that everyone who discharged a firearm in a public place in 2025 and 2026 has been apprehended (Politiet, 2026).
The question this article poses is whether there are measures that could supplement enforcement — measures that target the economic foundation of the violence, not just the perpetrators. It is a question without easy answers.
II. What Do We Know?
The market
To understand the problem, we need to know what people actually use. The Norwegian Institute of Public Health (FHI) estimates that 30 per cent of the population aged 16 to 60 have tried cannabis. Approximately 7 per cent report use in the past year. Among young adults (16-30), cocaine use has more than doubled — from 2.5 per cent in 2020 to 5.2 per cent in 2024. Use of amphetamines and MDMA is also growing (FHI, 2024).
Norway has around 290 overdose deaths per year — a rate of 7.1 per 100,000 inhabitants, among the highest in Europe (FHI, 2023).
Europol and EUDA estimate the European drugs market at over NOK 345 billion annually (EUR 30 billion). But drugs are not the only source of income for organised crime — they are not even the largest. Europol's threat assessment for 2025 (SOCTA) estimates total revenues from organised crime in the EU at around NOK 1,600 billion (EUR 139 billion). Drugs account for roughly a quarter. VAT fraud, counterfeiting, human trafficking and digital fraud are each of comparable size. In Norway, fraud now accounts for 85 per cent of all economic crime, and labour market crime costs an estimated NOK 12-60 billion annually in lost taxes (Skatteetaten, 2025; Europol, 2025).
But there is a crucial difference between these markets: the violence is not evenly distributed.
Fraud and cybercrime are the fastest-growing revenue sources, but they are almost entirely non-violent — Europol describes digital fraud as the fastest-growing form of crime, but characterised by "financial harm rather than physical violence" (Europol, 2025). Labour market crime operates through fictitious companies and systematic tax evasion — not through territorial control (Skatteetaten, 2025). These are largely different environments with different people.
The drugs market is structurally different. The criminologist Paul Goldstein described as early as 1985 three mechanisms for drug-related violence: the pharmacological (the effect of the substance), the economic-compulsive (crime to finance use), and the systemic — violence that is built into the market itself. Territorial disputes, debt collection, competition between suppliers, punishment of informants. In an empirical study of homicides in New York, Goldstein found that 74 per cent of drug-related homicides were systemic — they were caused by the market, not the substances (Goldstein et al., 1992).
This pattern is confirmed by more recent research. Werb and colleagues (2011) reviewed 15 studies of drug enforcement and violence. In 14 out of 15, they found that increased enforcement was associated with increased violence — because it intensifies competition among the remaining actors (Werb et al., 2011). Miron (1999) estimated that the American homicide rate is 25-75 per cent higher than it would be without the drug prohibition (Miron, 1999).
Kripos describes drugs as the "common denominator" for the most violent networks in Norway. "Control over the drugs market has been cited multiple times as the main reason for the serious violence," they write in their 2024 threat assessment (Kripos, 2024). Contract violence has become a revenue source in its own right — a contract killing price in Norway is NOK 160,000-260,000 (SEK 150,000-250,000) according to Kripos (Kripos, 2025). The Swedish gangs operating in Norway — Foxtrot, Shottaz, Dødspatruljen and others — are primarily financed by drugs, although they also engage in extortion and arms dealing.
Sweden illustrates where this can lead. With strict prohibitionist policy and high enforcement, Sweden went from one of Europe's lowest shooting death rates to the highest in roughly a decade. In 2024, Sweden recorded 317 bomb attacks — double the previous year — and 337 shootings. Brå (Brottsförebyggande rådet, the Swedish National Council for Crime Prevention) links the development directly to the struggle for control over the drugs market (Brå, 2025).
It is, then, an enormous market. It is not a market that is shrinking. And it is a market where money and violence are more closely linked than in any other form of organised crime.
How harmful are the substances?
The question seems simple, but the answer is surprising.
In 2010, David Nutt and colleagues published a study in The Lancet in which an independent expert panel assessed 20 drugs against 16 harm criteria — nine for the individual, seven for society. The study distinguished between two dimensions: how harmful the substance is to the user, and how harmful the user's behaviour is to others (Nutt et al., 2010).
For harm to the individual — dependence, organ damage, mortality — the ranking looks like this (0-100):
Crack: 37. Heroin: 34. Methamphetamine: 32. Alcohol: 26. Cocaine: 17. Tobacco: 18. Cannabis: 11. MDMA: 7.
Crack, heroin and methamphetamine are in a class of their own. They are intensely addictive and potentially lethal. Cocaine is serious but scores below alcohol for individual harm. Cannabis and MDMA rank far lower.
But the harm score alone does not give the full picture. How addictive are the substances — that is, what proportion of those who try them end up with a dependence problem?
Lopez-Quintero and colleagues (2011) followed over 40,000 Americans in the national NESARC study and calculated the probability of developing dependence after first use (Lopez-Quintero et al., 2011):
Nicotine: 67 per cent. Alcohol: 23 per cent. Cocaine: 21 per cent. Cannabis: 9 per cent.
Anthony and colleagues (1994) found similar figures in an earlier study: heroin 23 per cent, cocaine 17 per cent, alcohol 15 per cent, cannabis 9 per cent (Anthony et al., 1994). For methamphetamine, more recent data shows that over half of those who use it regularly develop a use disorder (SAMHSA, 2023).
Cocaine stands out in that dependence develops rapidly — half of those who become dependent are so within four years of first use (Lopez-Quintero et al., 2011). For alcohol, the corresponding timeframe is 13 years. This means cocaine provides a shorter window for intervention.
Cocaine and alcohol: a particularly dangerous combination
A point rarely discussed publicly: cocaine and alcohol are often used together. Meta-analyses estimate that 74 per cent of cocaine users take the substance together with alcohol (Alves et al., 2024). When the two substances are combined, the liver forms a third substance — cocaethylene — which has a longer duration of action than cocaine alone and is significantly more cardiotoxic.
In a study of emergency patients with overdose, 6.1 per cent of those with cocaethylene in their blood had cardiac arrest, compared with 0.7 per cent of those with cocaine only — a ninefold increase in risk (Ojanperä et al., 2023). Alves and colleagues (2024) found that 76 per cent of sudden cardiac deaths linked to cocaine involved concurrent alcohol use.
This is relevant to the regulation question: in a world where cocaine is available, it will in practice often be used together with alcohol — and the combination is substantially more dangerous than either substance alone.
Withdrawal and detoxification
The substances also differ greatly in how dangerous it is to stop using them.
Alcohol withdrawal can be fatal. Without medical treatment, delirium tremens has a mortality rate of up to 37 per cent (Rahman & Paul, 2023). The symptoms — seizures, hallucinations, autonomic instability — require intensive medical monitoring. Benzodiazepine withdrawal carries similar risks.
Heroin withdrawal is extremely unpleasant — muscle cramps, sweating, nausea, insomnia — but rarely fatal in itself. The great danger is relapse: after detoxification, tolerance is reduced, and a dose that was previously normal can now be lethal. The relapse rate for heroin is 78-95 per cent within one year (NIDA, 2024).
Cocaine withdrawal is not medically dangerous but psychologically serious: intense fatigue, depression, strong cravings. The relapse rate is around 50 per cent within one year and 75 per cent within five (NIDA, 2024).
Cannabis withdrawal is recognised as a diagnosis in DSM-5 (APA, 2013), but is mild: irritability, sleep problems, reduced appetite. It lasts 1-3 weeks and is not medically dangerous.
For harm to others — violence, accidents, family burden, crime, healthcare costs — alcohol dominates: 46 out of 100, more than double heroin (21) (Nutt et al., 2010). This is partly because alcohol is far more prevalent, but also because alcohol itself triggers aggression, risk-taking behaviour and accidents to a degree that most illegal substances do not. Alcohol is involved in 55 per cent of domestic violence cases, illegal drugs in 9 per cent (NCADD, 2015).
The study has been replicated several times: Van Amsterdam and colleagues (2015) conducted a comparable assessment with 40 experts from across the EU and obtained essentially the same ranking (Van Amsterdam et al., 2015). Lachenmeier and Rehm (2015) used a toxicological approach (margin of exposure) and found that cannabis has the greatest safety margin — over 10,000 times the typical use dose — while alcohol has the smallest (Lachenmeier & Rehm, 2015). Crossin, Nutt and colleagues (2023) replicated the study in New Zealand and confirmed the pattern (Crossin et al., 2023).
An important caveat: the total score in the Nutt study — where alcohol scores 72 and heroin 55 — is partly a product of prevalence. Alcohol looks worst because so many people drink. If the same number used heroin, heroin would likely score higher overall. For policy purposes, it is the individual harm and the behavioural harm per user that are most relevant — these are what tell us about the substance's properties, independent of how many use it today.
And here is the point: the legal classification of drugs does not reflect the actual harm profile. Cannabis (individual harm: 11) and MDMA (7) are prohibited, while alcohol (26) and tobacco (18) are legal. Cocaine (17) scores below alcohol for individual harm. This does not mean that drugs are harmless — heroin and crack are among the most dangerous substances known. But it does mean that the line between legal and illegal does not follow a logic based on harm.
The question this raises is not whether drugs are good. It is whether the prohibition does more harm than the substances themselves.
What have other countries done?
There is no simple model to copy. But there are experiences.
Portugal decriminalised possession and use of all drugs in 2001. The first 15 years produced striking results: overdose deaths fell from 76 to 10 per year. New HIV diagnoses among injecting users fell from 1,287 to 16. The share of prisoners incarcerated for drug offences fell from 40 to 16 per cent. But from the 2020s, the picture has changed: use has increased, overdose rates have risen again, and crime has increased. Much of this is due to treatment services not being maintained. Decriminalisation without adequate health and social services is not enough (Transform, 2025).
Canada legalised cannabis in 2018. The share of users purchasing legally went from 23 per cent in 2018 to 71 per cent in 2023 (Statistics Canada, 2024). Youth use did not increase. Tax revenues have surpassed NOK 42 billion (CAD 5.4 billion). But organised criminals did not disappear — most were involved in multiple markets and adapted (Bouchard et al., 2025). The black market has been sharply reduced but still accounts for around 29 per cent.
Oregon decriminalised possession of small amounts of all drugs in 2020, with a 58 per cent majority. Four years later, the state voted to reverse the reform. The reason was not that decriminalisation itself increased overdose deaths — researchers at Portland State University pointed to the fentanyl crisis and the pandemic as the main drivers (Portland State University, 2024) — but that treatment services were not expanded quickly enough, and public frustration over visible drug use became too great. Oregon is a reminder that good policy requires good implementation.
Switzerland chose a different path. From 1994, it offered heroin-assisted treatment (HAT) in clinics — not as a facilitation of drug use, but as medical treatment of a chronic disease. Patients attend the clinic, receive controlled doses under supervision, and simultaneously receive support for health, housing and employment. It is treatment in the same way as insulin treatment for diabetes: the goal is a functioning life, not abstinence at any cost. Over 20 years, overdose deaths fell by 64 per cent. New HIV infections fell 65 per cent. New heroin users fell 80 per cent. Home burglaries fell 98 per cent. In Germany, 40 per cent of HAT patients found employment after four years. The cost-benefit analysis shows a net saving of around NOK 70,000 per patient per year (EUR 6,000), primarily through reduced crime. In 2008, the Swiss electorate voted by a large majority to make the programme permanent (Transform, 2024).
Germany legalised cannabis in April 2024. After one year, drug-related crime has fallen by roughly a third. In Bavaria, cannabis-related crime fell 56 per cent. Youth use has not increased. Only 20 per cent of existing users report the black market as their source (Business of Cannabis, 2025).
The Netherlands has for over 40 years tolerated the sale of cannabis in coffeeshops, but the supply chain remains illegal — the so-called back door problem. The result is that criminal networks supply legal retail outlets. In April 2025, a four-year experiment with a regulated supply chain began in 10 municipalities, precisely to close this back door (Rijksoverheid, 2025).
Violence and legalisation
Is there evidence that legalisation reduces violence?
Gavrilova, Kamada and Zoutman (2019) studied American states along the border with Mexico. They found that medical cannabis laws reduced violent crime by 12.5 per cent in border states. Drug-related homicides fell 41 per cent, robberies 19 per cent. The mechanism: local growers replaced the cartels' cannabis, and the loss of revenue reduced cartel activity. The effect was stronger than a doubling of border guards (Gavrilova et al., 2019).
But cannabis is only part of the portfolio. Canadian research shows that criminal groups adapted to legalisation by shifting their operations to other substances (Bouchard et al., 2025). The Swiss model shows that heroin-assisted treatment produced dramatic crime reduction — but for one specific substance and one specific user group.
There is therefore no guarantee that legalisation of one substance removes the violence problem. But there is evidence that it can reduce it.
What happens to the criminals?
A question rarely asked in the legalisation debate: what happens to the people who currently make their living from the drugs trade? They do not disappear because the market is regulated.
History provides a pointer. When the United States repealed alcohol prohibition in 1933, organised crime did not vanish. The actors shifted their operations to gambling, extortion, union control and eventually narcotics. The American Mafia's power reached its peak in the 1950s and 60s — three decades after the end of prohibition. Okrent (2010) documents that none of the major prohibition-era criminals moved into legal work when alcohol was legalised again (Okrent, 2010).
More recent research confirms the pattern. Bouchard and colleagues (2025) found that criminal groups in Canada adapted to cannabis legalisation through three strategies: some undercut the legal market with lower prices, some switched to other substances (cocaine, synthetic opioids), and some diversified into other crime. Europol describes this as a general trend: criminal networks are increasingly "poly-criminal" — they operate in multiple markets simultaneously, and the loss of one revenue source leads to adaptation, not dissolution (Europol, 2025).
Research on criminal careers shows that leaving a criminal lifestyle requires more than the disappearance of a revenue source. Laub and Sampson (2003) identified "turning points" for desistance — that is, ceasing criminal behaviour: stable employment, a partner, and structured environments. But these presuppose social capital — networks, skills, trust — which many in criminal environments lack. The skills developed in the drugs trade — logistics of illegal goods, violence as a business tool, networks based on fear and loyalty — transfer more easily to other criminal enterprises than to regular employment (Laub & Sampson, 2003).
In Norway, the situation is further complicated by the fact that many of the most violent actors are affiliated with Swedish networks. Foxtrot, Shottaz and Dødspatruljen operate across national borders. Even if the Norwegian drugs market were regulated, these networks would still have markets in Sweden and the rest of Europe to serve.
Scandinavian exit programmes — such as the police's EXIT programme in Norway and equivalents in Sweden and Denmark — have shown that it is possible to help individuals leave gang environments. But the programmes have limited capacity, and success rates vary. Research from Brå shows that the most important factor for successful exit is geographical distance from the former environment, combined with a new identity and long-term support — measures that cost significant resources per person (Brå, 2023). In practice, this involves a few dozen exiters per year in Norway, while the police have mapped around 120 criminal networks in Oslo alone.
There is also a question of age composition. Gang crime recruits ever younger — children as young as 13 are used as couriers. For the youngest, prevention and alternative career paths can still work. But for adults who have lived off crime for years, retraining for regular employment is rarely realistic without extensive and sustained effort. Switzerland's experience with heroin-assisted treatment shows that it took four years before 40 per cent of patients found employment — and those were primarily users, not dealers.
The uncomfortable reality is that regulation of the drugs market would probably not eliminate criminal networks. It would change their business model. The question is whether the new model involves less violence — and that depends on which markets the criminals move into.
The Economist's cocaine argument
In October 2022, The Economist argued in a leader article for legalising cocaine. The main argument was that the single most effective measure to reduce violence, corruption and death would be to "legalise and regulate the production and consumption of cocaine". Consumers would get safe dosing, prisons would be emptied, and the justice system could concentrate on lethal synthetic substances (The Economist, 2022).
The article generated considerable debate. Critics pointed out that cocaine is intensely addictive, can cause sudden death, and that no antidote exists for overdose. They warned of "regulatory capture" — that a cocaine industry would gain political influence in the way the tobacco and alcohol industries have — and that the burden would fall hardest on vulnerable groups.
No country has legalised the production and sale of cocaine. We do not know what would happen. That is what makes the question so difficult.
The Norwegian debate
In 2021, the Solberg government proposed decriminalising possession for personal use — 10 grams of cannabis, 2 grams of heroin, cocaine or amphetamine. Use would remain illegal but without punishment. The proposal fell in the Storting (parliament) with 56 per cent against. The Labour Party, the Centre Party and the Progress Party voted against (Stortinget, 2021).
In 2025, a new, milder reform was adopted with 116 of 169 votes. Use and possession remain illegal, but fines are no longer recorded on the criminal record. Minors under 18 avoid fines but meet a municipal counsellor. The reform is described as "down-criminalisation", not decriminalisation (NRK, 2025b).
Norway has thus moved slightly, but not far. At the same time, the same substances are financing the violence we read about in the newspapers.
The cost picture
An honest debate on regulation requires an honest accounting. It has at least four columns: what we spend today, what we can save, what will cost more, and who pays the price.
What the prohibition costs today. Norway spends significant resources on enforcing the drug prohibition. Drug-related offences constitute the largest single category in Norwegian criminal law — 42 per cent of all drug cases led to imprisonment in 2023 (SSB, 2024b). A prison place in Norway costs around NOK 1,050,000 per year (Kriminalomsorgsdirektoratet, 2024). With approximately 900 inmates serving sentences for drug-related offences at any given time, this amounts to nearly one billion kroner annually in prison costs alone. In addition come police investigations, trials, prosecution and correctional services outside prison. Gjengpakke 2 — NOK 2.8 billion — is an addition to the regular effort.
There is also a human cost to the prohibition. Around 290 people die of overdose in Norway every year. Many of these deaths are caused by unknown purity and dosing — something a regulated market would reduce.
What regulation could generate in revenue. Canada has earned NOK 42 billion (CAD 5.4 billion) in tax revenue from cannabis since 2018 (Statistics Canada, 2024). Scaled to Norway's population and price level, this would correspond to NOK 5-7 billion over five years. In addition come reduced costs for police, the justice system and prisons. Cannabis-related crime in Germany fell by a third after legalisation — that represents police resources that can be redeployed.
But these savings are not immediate. In a transitional phase, society must pay for both the old and the new system: maintaining enforcement against the illegal market while building up the regulatory apparatus.
What regulation will cost more. Any reform that makes drugs more accessible must account for increased treatment needs. It is naive to believe otherwise.
In Canada, cannabis use rose from 22 to 27 per cent of the population after legalisation. Daily use went from 5 to 7 per cent (Health Canada, 2024). Cannabis-related emergency admissions more than doubled — from 6.4 to 14 per 100,000. There was a 13-fold increase in cases of cannabinoid hyperemesis syndrome and a marked increase in cannabis-induced psychosis (PHAC, 2025). In American states with legalisation, the population uses cannabis 20 per cent more than in states without, and the proportion with a diagnosed use disorder is 37 per cent higher (Martins et al., 2024).
Substance abuse treatment in Norway costs on average around NOK 120,000 per patient per year, with a daily rate of nearly NOK 12,000 for residential treatment (Helsedirektoratet, 2024). Heroin-assisted treatment is more expensive — NOK 145,000-235,000 per patient per year (EUR 12,700-20,400) — but the Swiss cost-benefit analysis shows that savings in reduced crime, healthcare costs and social services exceed the treatment cost. Six months without treatment costs society around NOK 225,000 per person (USD 21,500); six months with heroin-assisted treatment costs approximately NOK 125,000 (Transform, 2024). But this applies to a mature system that has been built up over decades.
If cannabis legalisation produced a comparable increase in treatment needs as in Canada — around 20 per cent more users and a doubling of emergency admissions — this would mean several thousand new patients in specialist health services annually. At an average treatment cost of NOK 120,000 per patient, we are talking about hundreds of millions of kroner in new health expenditure. For heavier substances, the figures would be substantially higher.
Who pays the price — and who pays it today? This is perhaps the most difficult question. Today, the price is paid by a relatively small group: those who are dependent on illegal substances, their families, and the victims of drug-related violence. This group is already marginalised. They die of overdoses, acquire criminal records, lose housing and employment.
With regulation, parts of the cost shift. Use will likely increase, and more "ordinary people" — individuals who currently do not have substance use problems — will develop them. Canadian data shows that the increased use after cannabis legalisation largely came among adults over 25, not young people. These are people with jobs, families and obligations who begin using cannabis more frequently, and a proportion of them will develop problematic use.
For cannabis, the consequences for most are moderate. But if regulation is extended to cocaine or amphetamine — substances with higher dependence risk — more families will be affected by cycles of dependence, detoxification and relapse. With cocaine's relapse rate of 50-75 per cent (NIDA, 2024), this is not a theoretical concern. The social cost — lost productivity, family burden, sickness absence, children's upbringing conditions — is difficult to quantify, but it is real.
At the same time, society already bears enormous costs from the legal substance alcohol. Alcohol-related illness, accidents, violence and productivity loss cost Norwegian society an estimated NOK 22 billion annually (FHI, 2019). The question is not whether we can avoid all substance-related costs, but whether a different distribution produces a better outcome overall — fewer overdose deaths, less gang violence, more openness about dependence, but more people in treatment.
Workforce: who will do the work? A shift from punishment to healthcare presupposes that the health professionals exist. They do so only to a limited extent. SSB estimates that Norway will lack 28,000 nurses and 4,500 doctors by 2035 (SSB, 2023). Already today, there are over 7,000 vacant health positions (NAV, 2025). Average waiting time for interdisciplinary specialised substance abuse treatment (TSB) was 35 days in 2024, with large regional differences (Helsedirektoratet, 2024).
Regulation requires not only treatment providers, but also a new type of expertise: substance use counsellors, harm reduction workers, quality controllers for legal products, a regulatory authority. In Switzerland, it took over a decade to build up an adequate apparatus around heroin-assisted treatment. Portugal invested heavily in "dissuasion commissions" — interdisciplinary teams of psychologists, social workers and lawyers who meet users instead of judges.
Police resources cannot simply be transferred to health resources — they are different people with different qualifications. Nor is there any guarantee that those who currently investigate drug crime would become redundant: the black market does not disappear overnight, and the criminal networks have other revenue sources. A realistic transition period would therefore mean increased total costs — investment on the health side without a corresponding reduction on the justice side — before any savings materialise.
Experience from Portugal and Oregon shows the same thing: reform without adequate treatment capacity makes the situation worse. If Norway is to move towards regulation of some substances, the treatment apparatus must be strengthened first — not afterwards. It is a prerequisite, not a supplementary appropriation.
What about families?
FHI estimates that around 90,000 Norwegian children — 8 per cent of all under-18s — have at least one parent who meets the criteria for alcohol use disorder. A Norwegian population survey found that nearly 16 per cent of adults reported growing up with problematic alcohol use in the home (Torvik & Rognmo, 2011). The consequences are documented: increased risk of one's own substance problems, mental health issues, behavioural difficulties and problems in education and working life.
Alcohol is involved in 55 per cent of domestic violence cases, illegal drugs in 9 per cent (NCADD, 2015).
The important point here is that the greatest burden on families already comes from a legal substance. The question is whether legalisation of other substances would worsen this picture, or whether regulation and openness would make it easier to identify and help those who are struggling.
The answer depends on the scope. Canadian data suggests that a moderate increase in cannabis use did not produce a corresponding increase in family problems. But we know very little about what would happen to families if access to cocaine or amphetamine became easier. And with relapse rates of 50-75 per cent for cocaine (NIDA, 2024), there are many families who would live with repeated cycles of dependence, detoxification and relapse.
Cocaine, alcohol and the social trap
A point rarely discussed: cocaine and alcohol are closely intertwined in patterns of use. An estimated 74 per cent of cocaine users take the substance together with alcohol, and nearly 60 per cent of those with cocaine use disorder also have an alcohol use disorder (Alves et al., 2024; Pani et al., 2022).
For those who have developed a cocaine dependence, alcohol is often the most important trigger for relapse. Treatment research shows that concurrent alcohol use is the strongest predictor of cocaine relapse, and many treatment protocols require total abstinence from alcohol as a condition for successful cocaine rehabilitation (Pani et al., 2022). In a society where alcohol is deeply embedded in social contexts — work lunches, friendship groups, celebrations — this means that cocaine rehabilitation in practice requires withdrawing from large parts of normal life. This makes detoxification harder than for substances that lack this social connection.
This dynamic reinforces an important point: substances do not exist in isolation. Any policy that addresses cocaine without accounting for the role of alcohol is missing a crucial piece.
From prescription to dependence
It is documented that 80 per cent of heroin users in the US first misused prescription painkillers (NIDA, 2024). The path from a legitimate prescription — after surgery, an injury, chronic pain — to dependence and onward to illegal opioids is well described. It is one of the mechanisms behind the American opioid crisis, in which over 80,000 people die of opioid overdoses every year (CDC, 2025).
The question is relevant to the regulation discussion: if more substances become legally available, does the threshold for transitioning from one substance to another decrease? Will some who develop tolerance for a weaker, regulated substance seek something stronger — and have we then inadvertently built a staircase?
The evidence is mixed. Canadian and American research finds no increase in use of cocaine or heroin following cannabis legalisation — that is, no "gateway effect" in that direction (Cato Institute, 2021). But the American opioid crisis shows that such a staircase can emerge within the legal system: doctors over-prescribed, patients became dependent, and when prescriptions were tightened, many transitioned to heroin and fentanyl (NIDA, 2024).
A possible lesson is that the problem is not legality itself, but poor regulation: insufficient follow-up, failure to map risk profiles, and profit-motivated actors with an interest in increasing sales. If we regulate new substances, the system must be designed to avoid the mistakes the American pharmaceutical industry made.
III. The Questions We Don't Have Good Answers To
In the party programme, Pragmatikerne write that we are "open to decriminalising and regulated legalisation of substances that are on a par with alcohol in terms of harm". This is a principled position. But what does it mean in practice?
According to Nutt and colleagues' harm scale, cannabis scores 20 and MDMA 9 — both lower than alcohol (72) and tobacco (26). Cocaine scores 27, roughly on a par with tobacco. Amphetamine scores 23.
Does this mean that cannabis and MDMA are clearly within scope? That cocaine is a borderline case? That heroin (55) and crack (54) are clearly outside? The scale provides a framework, but it does not provide an answer. Harm depends on context: purity, dosage, access to help, social conditions.
Here are some questions we do not have good answers to:
Is cannabis legalisation enough? Cannabis is the most widely used illegal substance. But cocaine use is growing fastest, and cocaine and amphetamine probably yield higher margins for criminal networks. If we legalise cannabis but not the other substances, have we made a real difference to the level of violence — or have we primarily made cannabis safer for users?
Can we regulate cocaine? No country has tried. The Economist argued it was possible, modelled on alcohol and tobacco. But cocaine is more addictive, more unpredictable in overdose, and completely without an antidote. The political influence of the tobacco and alcohol industries is hardly a model either. At the same time: the prohibition is clearly failing to limit access. Cocaine use among young Norwegians has more than doubled in four years.
What happens to the criminal networks? Canadian experience shows that organised groups adapt. They sell other substances, engage in other crime, or undercut the legal market with lower prices. Legalisation of one substance does not necessarily remove the networks — it changes their business model. And drugs are only a quarter of organised crime's total revenue in the EU. Fraud, labour market crime and counterfeiting are at least equally large markets. If we remove the drug revenues, the networks still have other income sources to fall back on.
Is Oregon a warning — or just poor implementation? Oregon reversed its decriminalisation after four years. But the researchers who evaluated it point to the fentanyl crisis and insufficient treatment capacity as the real causes, not the reform itself. Switzerland, with its more gradual and holistic approach, succeeded over 20 years. Norway has strong institutions and a good healthcare system. Does this mean we would manage better? Or is it naive to think so?
What about children and young people? Canadian and German experience shows that youth use did not increase after cannabis legalisation. But the concern is real: children as young as 13 are already being recruited as couriers. Would regulation reduce recruitment by removing the criminal market — or would it normalise use and increase demand?
What about treatment? Portugal and Oregon show the same thing: reform without treatment capacity produces worse outcomes. Switzerland invested in all four pillars — prevention, treatment, harm reduction and enforcement — over decades. Norway's 2025 drug reform is a step towards help rather than punishment, but it does not touch the economic foundation that drives the violence.
The review above shows that we know more than the public debate typically suggests. The prohibition does not prevent access — use is rising. Criminal networks earn billions from the substances we ban. Some countries have found models that deliver better outcomes for certain substances. Alcohol, which is legal, does more documented harm than several of the illegal substances.
At the same time, we do not know enough to say with certainty what would happen if Norway legalised cannabis, regulated cocaine or introduced heroin-assisted treatment along Swiss lines. Each country has its context. Each substance has its risk profile. Legalisation entails increased treatment needs, potentially increased use and consequences for families that are difficult to foresee. Criminal networks do not disappear — they adapt. And the healthcare system that would take over is already short of staff. There is no reform without risk.
What is possible is to conduct the debate on the basis of what we actually know — about harm profiles, about what other countries have experienced, and about what current policy achieves and does not achieve. This article is intended as a contribution to that.
Amounts in foreign currency are converted to Norwegian kroner using approximate exchange rates as of May 2026 (1 EUR ≈ 11.5 NOK, 1 USD ≈ 10.5 NOK, 1 CAD ≈ 7.7 NOK, 1 SEK ≈ 1.05 NOK). Original amounts are given in parentheses.
References
Alves, P. et al. (2024). Cardiovascular risks of simultaneous use of alcohol and cocaine. Journal of Clinical Medicine, 13(5), 1475. → Full reference
Anthony, J. C., Warner, L. A. & Kessler, R. C. (1994). Comparative epidemiology of dependence. Experimental and Clinical Psychopharmacology, 2(3). → Full reference
Bouchard, M. et al. (2025). How organized crime adapts to cannabis legalization. Sociological Inquiry. → Full reference
Brå. (2025). Increased gun violence in Sweden. → Full reference
Brå. (2023). Avhoppare från kriminella nätverk [Exiters from criminal networks]. → Full reference
Crossin, R., Nutt, D. et al. (2023). The relative harms of new psychoactive substances. Journal of Psychopharmacology. → Full reference
Europol. (2025). EU-SOCTA 2025. → Full reference
Folkehelseinstituttet [Norwegian Institute of Public Health]. (2024). Nye tall for narkotikabruk [New figures for drug use]. → Full reference
Folkehelseinstituttet [Norwegian Institute of Public Health]. (2019). Alkohol i Norge: Forbruk, skader og kostnader [Alcohol in Norway: Consumption, harm and costs]. → Full reference
Gavrilova, E., Kamada, T. & Zoutman, F. (2019). Is legal pot crippling Mexican drug trafficking organisations? The Economic Journal, 129(617). → Full reference
Goldstein, P. J. et al. (1992). Drug-related homicide in New York. Crime & Delinquency, 38(4). → Full reference
Kriminalomsorgsdirektoratet [Norwegian Directorate of Correctional Services]. (2024). Kriminalomsorgens årsstatistikk 2023 [Annual statistics of the correctional services 2023]. → Full reference
Kripos [National Criminal Investigation Service]. (2025). Vold som handelsvare [Violence as a commodity]. → Full reference
Laub, J. H. & Sampson, R. J. (2003). Shared beginnings, divergent lives. Harvard University Press. → Full reference
Lopez-Quintero, C. et al. (2011). Probability and predictors of transition from first use to dependence. Drug and Alcohol Dependence, 115(1–2). → Full reference
Miron, J. A. (1999). Violence and the U.S. prohibitions of drugs and alcohol. American Law and Economics Review, 1(1). → Full reference
NAV [Norwegian Labour and Welfare Administration]. (2025). Bedriftsundersøkelsen 2025: Mangel på arbeidskraft [Enterprise survey 2025: Labour shortages]. → Full reference
Nutt, D. J., King, L. A. & Phillips, L. D. (2010). Drug harms in the UK. The Lancet, 376(9752). → Full reference
Ojanperä, I. et al. (2023). Cocaethylene cardiotoxicity. Eur. J. Emergency Medicine. → Full reference
Okrent, D. (2010). Last call: The rise and fall of Prohibition. Scribner. → Full reference
Statistisk sentralbyrå [Statistics Norway]. (2024b). Straffereaksjoner: Type reaksjon og lovbruddsgruppe [Criminal sanctions: Type of sanction and offence category]. → Full reference
Statistisk sentralbyrå [Statistics Norway]. (2023). Framskrivinger av arbeidsstyrken og sysselsettingen [Labour force and employment projections]. → Full reference
Transform Drug Policy Foundation. (2025). Drug decriminalisation in Portugal. → Full reference
Transform Drug Policy Foundation. (2024). Heroin-assisted treatment in Switzerland. → Full reference
Werb, D. et al. (2011). Effect of drug law enforcement on drug market violence. Int. J. Drug Policy, 22(2). → Full reference
About the author: L. C. Praxis writes about politics, economics and everyday life for Pragmatikerne.