New Zealand's national referendum on whether to support or reject the Cannabis Legalization and Control Bill (CLCB), which would have legalized commercial recreational cannabis supply and sale, has been narrowly rejected (i.e. 51% = ‘no' to 48% = ‘yes'). Our article written in the months preceding the referendum vote critiqued the CLCB, pointing out that while overall it proposed a strictly regulated legal cannabis market, there were real questions about whether it could realistically achieve the aim of reducing cannabis use over time via a commercial market [1]. We also argued that the CLCB could be strengthened with formal minimum pricing, a lower potency cap and greater clarity concerning social benefit operators and the role of local government.

A natural question arises as to whether the defeat of the referendum was due to concerns about the CLCB, or wider concerns about cannabis legalization in general. Our impression is that it was the latter. Most of the public debate centred around the wider health and safety aspects of cannabis legalization, rather than specific provisions of the CLCB. A commonly raised concern was whether age restrictions on use really would be effective at restricting youth access to cannabis, reflecting long experience of fairly ineffectual alcohol purchase age limits in New Zealand [2]. Secondly, there was concern about the potential negative consequences for driving and work‐place safety, with few details about how testing for cannabis impairment and related sanctions would be implemented in practice. Thirdly, New Zealand has a long‐standing public health goal of being smoke‐free by 2025 [3], and it was noted that the legalization of cannabis appeared to be inconsistent with achieving this goal. Fourthly, there were concerns about whether the legal cannabis industry and related retail outlets could be controlled, again based on long experience of the proliferation of alcohol outlets in deprived communities in New Zealand [4]. Fifthly, there was anxiety about whether legalization would further reinforce the normalization of cannabis, fuelling rising use and dependency and adding pressure to an already overstretched mental health and addiction sector [5]. Sixthly, doubts were also raised about whether legalization would actually reduce the power of drug selling gangs, which had been seen to be expanding in recent years, or merely result in gangs refocusing on cannabis sales to underage cannabis users and expanding sales of other drugs, such as methamphetamine [6].

The thoughtful commentaries on our paper focus upon the specific implementation challenges of legalizing cannabis markets. Cousijn highlights the potential for cannabis product labelling detailing tetrahydrocannabinol (THC) and cannabidiol (CBD) levels to contribute to the safety of legal cannabis use by encouraging informed dose administration, but points out that current labelling is poorly understood by consumers. There is only emerging understanding of what constitutes a standard dose, and studies have found widespread mislabelling of the potency and content of cannabis products [7]. The high variability of the cannabinoid profile between and within product batches, and variability in test results from different test services, suggest ‘a difficult road' for the CLCB in terms of taxing cannabis products based on THC potency [7]. The experience in Colorado further reinforces the reality that implementing the testing of cannabis products poses significant technical and capacity challenges [8].

Lenton raises the important issue of ensuring that retailers and their staff, who clearly have a commercial incentive to make sales, adhere to the law concerning selling (e.g. age restrictions) and provide customers with harm minimization advice [9]. As Lenton notes, the CLCB rightly focuses penalties on licence holders and managers rather than retail staff, but asks whether the proposed cannabis authority would actually enforce and prosecute these regulations [9]. Recent experience with the failed Psychoactive Substances Act (PSA) in New Zealand illustrates that regulatory agencies can easily be overwhelmed by compliance failures and legal challenges, particularly when they are not adequately resourced and staffed [10].

Freeman & Lynskey present the evidence for the public health benefits of minimum pricing of alcohol to support calls for a similar approach to legal cannabis [11]. They also note the difficulties of getting minimum alcohol pricing implemented in the face of powerful industry opposition [11]. In the case of cannabis legalization, the industry has been provided a readymade argument to resist price controls, as these supposedly undermine their ability to compete with and eliminate the black market. Recent experience in Canada shows that the black market can respond to legal competition by lowering their prices [12], putting further pressure on the legal market to respond with even lower prices, and undermining regulatory efforts to keep the legal price at a certain level. We have argued that reducing the cannabis black market should not be an over‐riding goal of legalization, as it leads to a weak regulatory regime and low legal prices.

Smart also discusses the real trade‐offs between restrictive regulation that reduces harm and unintentionally creating opportunities for the black market [13]. Smart (2020) cautions that legalization can reduce the harms caused by prohibition but does not eliminate them. Black and grey markets continue, albeit at a reduced level, and juvenile use and racial disparities in arrests potentially persist despite legalization [13]. Smart argues these issues underline the ongoing need for investment in enforcement [13]. We would add they also highlight the need for comprehensive evaluation and a commitment to refine regulatory frameworks to better achieve key goals.