Harm Minimisation Policies Are Failing Homeless People Now

Marsh Ward met his first rule breaker in the mid 1980s. It changed everything he believed about helping addicts. As clinical director of Clean and Sober Streets in Washington DC, Ward had opened his doors with no rules at all. “They’re all adults, aren’t they?” he reasoned. Feed people, house people, and surely they would pull themselves together.

A year later he changed his mind. “You’re killing them,” he later admitted. “You’re enabling them to stay with their disease.” So he brought in rules: no violence, no stealing, no more chances after repeated relapse. It felt harsh. It also worked better than kindness without structure ever had.

Four decades on, that same argument is playing out at a population level. The numbers are now in.

A Tenderloin Story With a Policy Behind It

In 2023, ten dollars bought a hit of fentanyl on the streets of San Francisco’s Tenderloin. Dealers claimed it was fifty times stronger than heroin. On an average day that year, fentanyl killed two people in the city. Outreach workers pushed wagons of foil, pipes and needles through the crowds. “Harm reduction, need anything?” they called out, like an ice cream van doing the rounds.

San Francisco recorded 810 overdose deaths that year. The toll fell to around 630 over the following two years. That drop did not come from a better harm minimisation approach. It came because the city’s newly elected mayor reversed the approach entirely. “We stopped freely handing out drug supplies and letting people kill themselves on our streets,” he said. “It is not a basic right to use drugs openly in front of our kids.” A city supervisor, himself a former drug user, put it more bluntly. Years of harm minimisation, he said, had produced “perverse incentives that have done more to exacerbate problems than solve them.”

San Francisco is one city. But researchers keep finding the same pattern elsewhere: harm minimisation alone tends to manage a crisis rather than shrink it.

Australia’s Own Experiment, Run in Reverse

Australia effectively ran this experiment on itself. Harm minimisation dominated the country’s approach to youth substance use until 2009. That period tracked closely with high rates of adolescent drug use. Policymakers then introduced abstinence-based prevention programmes and policies from 2009 onward. Adolescent use fell. Those reductions carried through into lower adult use in the years that followed. Life-course researchers would expect exactly this pattern: intervene early enough, and the effect compounds across a generation.

Yet even as prevention worked, funding never followed the evidence. A 2024 review found that Australian governments spent $5.45 billion on illicit drug countermeasures in 2021 to 2022. Just 7 per cent of that, around $363 million, went to prevention, the only pillar actually aimed at stopping people from starting. Law enforcement took 64 per cent. Treatment took 27 per cent. Harm minimisation, in other words, kept its funding while prevention went begging.

The cocaine data shows the consequences of that imbalance. Cocaine use among Australians aged 14 and over grew from 1 per cent in 2004 to 4.5 per cent in 2022 to 23. National monitoring figures, marking thirty years of Australia’s drug tracking programme, recorded the highest wastewater cocaine readings in the country’s history in 2024 to 25. Deaths from cocaine have risen fivefold since 2000. Hospitalisations have tripled since 2011. Cocaine is now the second most commonly used illicit drug in the country. The researchers themselves describe the market as “growing and more established.”

Prevention did not feature once in that bulletin’s policy recommendations.

Britain’s Longer Experience Tells the Same Story

Britain has run harm minimisation for even longer. Its own researchers have already published the audit. Writing in the Journal of Global Drug Policy and Practice, one leading UK addiction sociologist tracked the country’s shift toward harm reduction from the late 1980s onward. The same optimistic claims, examined roughly forty times over, kept failing to survive contact with the data.

Around 40 per cent of injecting drug users in the UK tested Hepatitis C positive. The needle and syringe programmes that supposedly prevented HIV transmission had done little to shift that rate. Heroin and morphine deaths in England and Wales rose from 187 in 1993 to 926 in 2000, the very period when harm minimisation ideas most influenced UK policy.

Homelessness turned out to be deeply entangled with the harm the policy was meant to manage. One London study tracked a thousand homeless young people. It found 88 per cent used at least one drug and 35 per cent used heroin. Hospitals recorded which overdose patients had ever been homeless. Those with a history of homelessness had overdosed far more often than those who had never been homeless. Researchers called it the “combined experience of homelessness and drug use.”

The research reached a blunt conclusion. Prevention, not harm minimisation, should become the central aim of drug policy.

What the Pattern Actually Says

None of this argues that people in crisis should be left to die on a pavement. No serious voice on either side of this debate suggests that. Naloxone saves lives in the moment it is administered. The argument is narrower and harder to dismiss. Policies built almost entirely around harm minimisation, while starving prevention of funding and attention, do not appear to reduce how many people end up using drugs, injecting drugs or dying from drugs in the first place. They manage a crisis. They rarely shrink it.

Marsh Ward’s homeless residents in 1980s Washington were not statistics in a monitoring bulletin. They told him, one withdrawal at a time, that indulgence without structure was not compassion. It was another punch from the addiction they were already losing to. Jessica DiDia made the same point her own way. Her mother begged her to come inside from a San Francisco sidewalk. Jessica told her mother she couldn’t be made to want a way out.

The US, the UK and Australia keep landing on the same uncomfortable finding. Harm minimisation and reducing drug use are not the same policy goal. Treating them as interchangeable has cost more than anyone intended when the idea first caught on.

Sources