Most public policy projects involve analysing issues that are embedded in our social life and in government activity, likes taxes or schooling or planning. It is rare that governments pass laws that create entirely new public policy horizons, with accompanying challenges and opportunities.

Canada’s decision to legalise cannabis for adult (non-medical) use in 2018 was one rare example. The Cannabis Act was the first time that a major industrialised country had chosen to create a legal, regulated market in what had been a prohibited drug for over ninety years.

Earlier this year, Public First were commissioned by Professor Keith Humphreys at Stanford University to conduct a review of this significant policy change in Canada, now three years after it first came into effect. The results of that project are published today in a new research report that is accompanied by a poll of Canadians and their attitudes to cannabis and to legalisation.

As a company we are agnostic about the merits of legalisation and approached this project with the aim of being impartial and simply assessing what the available data told us. We focused only on the adult-use category, and not the pre-existing medicinal cannabis sector in Canada. And rather than comment on market issues, changing consumer behaviour or industry developments like product availability, or consolidation or over-supply, we focused on the policy goals and quantified the impact of the law change mostly according to the aims and objectives of the Cannabis Act.

These were goals that the Canadian Parliament endorsed when they passed Bill-C45 in summer 2018 and included reducing youth access to cannabis, safeguarding public health, improving public safety and the criminal justice impact, and eliminating the illicit market.

Our research shows that most of these goals have not been met so far, however important steps have been taken and the implementation has largely been successful. With more granular data at provincial level in future it will be possible to draw firmer conclusions.

Some early signs of the policy change having a big societal impact – for better or worse – would include modest but still rising rates of use, including daily or almost daily use, a reduction in the proportion of sales in the illicit market, and a big reduction in the police enforcement of cannabis offences against individuals. However youth use may not be decreasing and there is also some evidence that impaired driving could be a problem and needs closer study.

For a country pioneering such a bold policy reform, it was remarkable how much official data was unavailable, or not released in ways that would allow comparisons over time or across different provinces. Canada has good published survey data but there are other big gaps. We still know very little about the fiscal impact or how revenue from legal sales is being spent. The public were strongly in favour of funds going towards youth education and treatment programmes.

As debate about the merits of legalisation continue in many countries, Canada’s example is instructive. Future reports of this kind will be able to examine longer term trends, including the impact on young people and wider factors like rates of alcohol consumption and opioid prescribing. For now, the poll commissioned for this project is clear on where the public stand: 46% of Canadians supported legalisation on the eve of legalisation in 2018 with 30% opposing it, and now our survey found that 53% of respondents approved of legalisation, and only 18% opposed it.

One key conclusion from this study is that Canada shows how you can legalise cannabis differently from how it has been done in much of the United States, with a stronger emphasis on public health, in a way that should be more appealing to British policy-makers and politicians. What it also shows is how much of the Canadian approach is devolved, with provinces having ownership of many of the key decisions over how cannabis is actually sold and where, and that this can dictate how legalisation is experienced.

Canada actually has ten or more models of legalisation, each with their own variations that directly effect how the local population experience it. If you live in Alberta and can buy legal cannabis in person from over 600 retail stores, then the impact of this law change is very different from living in Quebec where cannabis is only sold in 60 government owned and operated stores. And given marketing restrictions and consumer bias towards buying cannabis in-person rather than online, when it comes to access, stores matter.

The benefit of a devolved model is that local politicians can respond to public priorities and shape the legal market in a way that suits local attitudes and benefits their local economy. It also means that policy-makers and researchers have many different case studies to track within the same country to evaluate how legalisation has worked out in practice.

It was a single bold policy reform but Canada’s approach has given the world many different models to learn from, and this report is one of the first attempts to identify some key lessons.