A Citizens’ Wānanga on Alcohol Policy in Aotearoa New Zealand: An Essay on Theoretical Underpinnings and Practical Considerations

Abstract

Alcohol use is a major determinant of health at both an individual and societal level. In New Zealand alcohol is legal and highly commercialized. Its harms are pervasive and are inadequately managed by current alcohol policy. However, contemporary methods for policy reform have also proven ineffective, and so effective alcohol regulation has been repeatedly thwarted. This essay first outlines the current state of regulatory failure and establishes the need for reform, before proposing citizens’ assemblies as a solution to this public health conundrum. These are novel democratic institutions designed to secure an informed and representative public voice as the central determinant of policy. This essay contends that an appropriately designed citizens’ assembly is likely to lead to alcohol policy which is more informed, evidence-based, and democratically accountable than current methods. Such an assembly would also be consistent with the Treaty of Waitangi principles and may assist in reducing alcohol-related inequities for Māori. Finally, in discussing such an assembly, this essay presents an internationally applicable model which could be employed for a range of other drug policy issues.

Introduction and Context

Aotearoa New Zealand is an island nation with an estimated population of 5,112,000 (Statistics NZ, 2021). Of this, 70.2% are European, 16.5% are Māori, 15.1% are Asian and 8.1% are Pacific in ethnicity (Statistics NZ, 2020). As of 2019 New Zealand's GDP was US$207 billion (The World Bank, 2021) and expenditure on health as a percentage of GDP was 11.03 percent (New Zealand Government, 2021a; WHO, 2021). In New Zealand alcohol is currently regulated by the Sale and Supply of Alcohol Act 2012 (New Zealand Legislation, 2012).

Central to New Zealand's principles is the Treaty of Waitangi. It governs the relationship between Māori and non-Māori and ensures both have their rights protected (New Zealand Government, 2021b). Despite attempts at rectifying inequity, Māori continue to experience a large disparity when it comes to their population health and healthcare (Hobbs et al., 2019). It is for these reasons that this article makes reference to New Zealand's “Bicultural” context – not to neglect our other cultures, but so as to emphasize the considerable inequity and health disparity present, as well as our social and treaty responsibility to redress this (Ministry of Health, 2020).

The aim of this essay is to explore and discuss the possibility of a citizens’ assembly with the aim of reducing alcohol-related harm and inequities, and overcoming current barriers to policy reform. This is done by presenting the theoretical underpinnings of deliberative democracy, its benefits and limitations, and practical considerations for its application to alcohol policy.

The Case for Reform in New Zealand

Alcohol is the most popular recreational drug in the world, and has become enmeshed in social and cultural life, conferring pleasure, intimacy, cultural belonging and social connection ¹ (Room et al., 2022; Savic et al., 2016). However, when both its physical and societal consequences are taken into account, alcohol is consistently ranked as the most harmful drug (Bonomo et al., 2019; Nutt et al., 2010; van Amsterdam et al., 2015). New Zealand is no exception, and yet here alcohol enjoys a popularity status exceeding that of any other drug (Cobiac & Wilson, 2018; Sellman et al., 2018). At least 25% of New Zealanders are heavy drinkers ² (Sellman et al., 2018) and have a risk of dying from alcohol-related causes which exceeds 1:100 (New Zealand Law Commission, 2010). It is the fourth largest risk factor for health loss in New Zealand, and, for those aged 15–49, alcohol is the largest risk factor for health loss and disability (Cobiac & Wilson, 2018). In 2007, 5.4% of all deaths under 80 were attributable to it (Connor et al., 2015).

It is also important to acknowledge the link between alcohol and suicide, mental health, and co-existing substance abuse problems (Government Inquiry into Mental Health and Addiction, 2018; Office of the Minister of Justice, 2010). For example, evidence from the Christchurch Health and Development Study (CDHS; a longitudinal birth cohort) suggests that rates of major depression and suicidal ideation could be reduced by 14.7% and 30.2%, respectively, if all alcohol misuse could be eliminated (Fergusson et al., 2013).

Finally, alcohol is the only drug where its harm to others is considered higher than its harm to the user (Bonomo et al., 2019; Nutt et al., 2010; van Amsterdam et al., 2015). Approximately 60,000 physical assaults and 10,000 sexual assaults are committed by New Zealanders affected by alcohol each year (Kypri et al., 2013). New Zealand police estimate that 50–70% of their work is associated with alcohol (Power, 2009).

Alcohol's Impacts on Māori

Worldwide, incidence of alcohol misuse is higher among indigenous peoples (Kõlves et al., 2020). Alcohol use is associated with lower socioeconomic status and as a coping strategy for trauma, both present and historical (e.g., trauma from colonization transmitted into adverse health outcomes) (Reid et al., 2014), such that it frequently strengthens historical inequalities within countries (WHO, 2018). This trend extends to New Zealand also. Māori consume more alcohol and experience higher rates of alcohol abuse and dependence than non-Māori (Cobiac & Wilson, 2018; NZ Drug Foundation, 2011). The direct harms of alcohol, including injury, premature death, and cancer are disproportionately suffered by Māori (Connor et al., 2017; Huckle et al., 2019; Kypri et al., 2015). The age-standardized alcohol attributable death rate for Māori is 2.5 times that of non-Māori (Connor et al., 2015; Walker, 2019). Indirect harms of alcohol, such as effects on financial position, work, study, and employment are also more likely to be experienced by Māori (NZ Drug Foundation, 2011). Māori women suffer more adverse effects as a result of other people's drinking than any other population group (New Zealand Law Commission, 2010). Despite all these harms, Māori report higher unmet needs for reducing alcohol consumption (New Zealand Law Commission, 2010). Rather than just reflecting existing inequalities, there is now evidence that alcohol may be actively driving them (New Zealand Law Commission, 2010).

The Economic Impacts of Alcohol-Related Harm

Alcohol, as a legalized commodity, provides a significant economic stimulus, and is responsible for the employment of as many as 70,000 people in hospitality and agricultural sectors (New Zealand Law Commission, 2010). The exact economic value of alcohol in New Zealand is unavailable. However, the economic benefits of alcohol, often accrued by private individuals, should be considered in relation to its economic harms. In New Zealand, the cost of alcohol has been estimated at $7.85 billion dollars per year (Nana, 2018), though this figure may be significantly higher (Sellman et al., 2018).

Current Policy and Policy Reform Methods Are Insufficient

The harms of alcohol in New Zealand are considerable, which suggests that current policy is ineffective when it comes to reducing these. This is especially so considering that the stated objective of the Sale and Supply of Alcohol Act 2012 is that “the harm caused by the excessive or inappropriate consumption of alcohol should be minimized” (New Zealand Legislation, 2012). New Zealand has a long history of attempting to regulate the harms of alcohol (Christoffel, 2006) but in this regard it has failed. Indeed, the 1989 Sale of Alcohol Act and its 1999 amendment have been criticized for perversely increasing the availability and ensuing harms of alcohol (New Zealand Law Commission, 2010; Walker, 2019). In response, the 2010 Law Commission was recruited to conduct the most comprehensive review of alcohol laws in New Zealand's history (Sellman et al., 2017). However, the most important, evidence-based, and WHO-endorsed recommendations which the commission emphasized were all largely rejected (Sellman et al., 2018; Walker, 2019). Excepting the 2014 changes to drink-driving limits, all further investigations into alcohol reform are yet to have a comprehensive government response (Government Inquiry into Mental Health and Addiction, 2018). In the ten years since the 2010 law commission, an estimated 8,000 individuals died prematurely, there were 700,000 cases of physical and sexual assault, and 30,000 children born with brain damage, all alcohol-related and costing New Zealand upward of $70 billion (Alcohol Action NZ, 2019) New Zealand's current alcohol policy remains in what has been described as “a state of crisis” (Walker, 2019).

It is unclear exactly why the Law Commission's priority reforms were all rejected, particularly when the government acknowledged the need for reform (Power, 2009). Alcohol is the most popular drug in our society, but also the most harmful, and yet it is not treated so (van Amsterdam & van den Brink, 2013) Common alcohol industry mantras were frequently voiced in defense of the government's approach, such as that major alcohol harm is an issue of the minority, that the current research is inconclusive, and that the majority of “responsible drinkers” should therefore not be unfairly punished (Sellman et al., 2018). Evidence in favor of alcohol reform was withheld (Goodwin, 2011; Sellman et al., 2017), responsibility for alcohol policy was shifted onto local communities (without adequate national support) (Morrah, 2019; Sellman & Connor, 2014), conscience voting was used to deflect blame despite explicit advice from the Law Commission to dispense with the outdated practice (New Zealand Law Commission, 2009), and the process of reform was delayed by over a year and a half (Sellman et al., 2018). The government at the time went directly against the consensus of both the Law Commission and submissions (Williams, 2010), failed to act on further inquiries they requested (Government Inquiry into Mental Health and Addiction, 2018), and focused on reforms which would be unlikely to have any substantial impact (Kypri et al., 2013; Sellman et al., 2017). Importantly, there is no evidence that effective alcohol policy reform failed because of public opposition. Public submissions made to the Law Commission were mostly in favor of alcohol policy reform (Kypri et al., 2013; Peck, 2010; Sellman et al., 2017) and the 2010 Health Sponsorship Council survey suggested that over 75% of New Zealanders over 15 would support or at least go along with most major WHO-endorsed alcohol reforms. ³ (Devlin, 2011; Kypri et al., 2013; Peck, 2010; Sellman et al., 2017) At the time of the 2012 Act formation, the release of these findings was delayed as a result of the final report being held up on the Associate Minister of Health's desk (Goodwin, 2011; Sellman et al., 2017).

The legalized status of alcohol allows for open interference from economic interests which can proliferate its harmful effects (Barlow et al., 2022; Monteiro & Sanchez, 2022; Slack et al., 2009). The alcohol industry is very well financed (Alcohol Public Policy Group, 2010) and makes up to half of its profit from hazardous drinking (Foster et al., 2003; WHO, 2018). Adequate alcohol reform could entail billions of dollars of loss for the alcohol industry, and so it has a significant incentive to discourage it. But alcohol's legal status also offers an avenue for effective regulation. Current policy-forming methods have delivered too little too late, leading to largely stagnant and ineffective alcohol laws which do not reflect the public will. But other methods are available, which, in conjunction with alcohol's legality, offer the possibility of adequate and more democratically accountable alcohol policy. One such alternative method is a citizens’ assembly. In subsequent sections we present the theoretical underpinnings of deliberative democracy, its benefits and limitations, and practical considerations for its application to alcohol policy.

Deliberative Democracy

Deliberative democracy is a political theory which, in contrast to aggregative democracy, privileges justification and reason-giving above majority votes for decision-making (Gutmann & Thompson, 2004). It is perhaps best understood by considering its constitutive halves. The democratic half is grounded in the idea of popular sovereignty – that the fundamental source of governmental legitimacy is the collective judgment of the people (Gutmann & Thompson, 2004). It requires that decision-making be collective and founded on political equality (Elster, 1998; Gutmann & Thompson, 2004).

The deliberative half affirms the need to justify arguments or decisions (Elster, 1998; Gutmann & Thompson, 2004). It is premised on the belief that the opinions which arise after exposure to impartial evidence, shared experiences, and discussion are superior (more accurate, informed, genuine, considered, and sophisticated) than the “raw” top-of-the-head opinions which preceded them (Denver et al., 1995; Elster, 1998; Setälä, 2017). It places faith in the transformative power of deliberation, maintaining that individual preferences shift when subjected to good quality deliberation, and that these new opinions are better than the preceding “raw” ones (Denver et al., 1995; Setälä, 2017). The result of deliberative democracy is a more authentic approximation of the “public-will” (the combined desires of the public which democracy upholds as a ruling body) than other methods (Elstub, 2008; Gutmann & Thompson, 2004; Isernia & Fishkin, 2014; Landwehr, 2014; Smith, 2011). Proponents of deliberative democracy contend that it fosters tolerance, generates impartiality by highlighting underlying preferences which may be motivating beliefs, encourages collective and public-spirited rather than individualistic decision-making, and does all this in a non-coercive, legitimate, and accountable fashion (Dryzek, 2000; Gargarella, 1998; Gutmann & Thompson, 1998, 2004; Pettit, 2003). Deliberative democracy is a method of reason-based public-will formation (Isernia & Fishkin, 2014). As Lukensmeyer points out “Numerous independent evaluations of deliberative practice … have found these outcomes to be highly reliable. In short, the Deliberative Democracy field no longer needs to demonstrate its capacity and its efficacy” (Lukensmeyer, 2014).

Criticisms of Deliberative Democracy

This being said, deliberative democracy is no panacea. Criticisms come from several fronts: Firstly, equal participation does not translate into equal deliberation. Some citizens, typically those from majority groups with higher levels of education, will be more persuasive and practiced in making arguments than others, thereby skewing the outcome away from a true “public will” (Sanders, 1997).

Secondly, even when their arguments are persuasive and reasonable, some people are less likely to be listened to because deliberative democracy cannot ensure “equality in epistemological authority” whereby all arguments are treated irrespective of their speaker's background (Sanders, 1997). When people assemble in a deliberative forum, they “do not leave behind the status, power, and privileges that they hold in the outside world” (Sanders, 1997). For example, many deliberative forums begin by choosing a spokesperson. More often than not, this person is white, male, and tertiary-educated (Hans & Vidmar, 1986; Sanders, 1997). Deliberative democracy seems particularly poorly placed to resolve this issue, because any efforts to give a voice to systematically disregarded minorities would “violate the deliberative tenet to attend to the force of argument rather than the interests of particular groups” (Phillips, 1995; Sanders, 1997).

Thirdly, deliberation is resource-intensive, and not all citizens have the time, money, or education to take part (Sanders, 1997). Because of these resource barriers, deliberation may never be entirely representative. Furthermore, some argue that genuine deliberation does not result in any substantially new or useful outcome as compared to the standard opinion. The large expenditure of resources and time therefore risks being a waste (Fishkin & Luskin, 2005; Neblo, 2015).

Fourthly, it has been objected that genuine deliberation is not possible because, even in an organized context, the public does not have the motivation, capacity, or time and attention to adequately engage with the topic (Fishkin & Luskin, 2005; O'Flynn & Sood, 2014; Setälä, 2011). Selective attention and processing of information means that, even when provided with impartial material, participants will limit their learning to that which is attitudinally congenial (O'Flynn & Sood, 2014).

Finally, deliberative democracy's call for consensus and compromise may in fact suppress the perspectives of marginalized groups (Sanders, 1997). Verdict-driven deliberations reduce the chances of a broad consideration of all views on the evidence, reduce the likelihood of a rational discussion, and increase the pressures to conformity (Sanders, 1997).

These are indeed great challenges for deliberative democracy. But it is our view that they do not constitute reason for its abandonment. Instead, acknowledgement of these shortcomings is vital, as it allows for the theoretical model to be reworked into a practical assembly which is more receptive to biases and individual circumstances. In the discussion of our proposed citizens’ assembly, we detail methods for overcoming some of these weaknesses. We also contend that many of these problems are shared by other contemporary methods of policymaking and that, rather than being perfect, deliberative democracy instead represents a preferable alternative in that it is better able to compensate for these issues. It should be kept in mind that citizens’ assemblies are just one potential tool for policymaking, and do not pretend to solve the shortcomings of contemporary democracy on their own. Expecting otherwise misunderstands their role and potential in a larger deliberative system of democracy (Neblo, 2015).

Citizens’ Assemblies

A citizens’ assembly is an instantiation of deliberative democracy. High-quality deliberation is not feasible for an entire citizenry, and so a citizens’ assembly is required to provide a practical format for harnessing the benefits of deliberation. In these, a small, representative group of citizens convene to deliberate, generally in three phases: learning, public consultation, and discussion (Luskin, 2003). Such assemblies have been used to address a wide range of topics all over the world – Canada, America, Ireland, the Netherlands, Australia, France, and Scotland for example. The Irish assemblies, in particular, attested the ability of citizens’ assemblies to address contentious issues, and evidenced their capacity to enact meaningful policy change (Carolan, 2015; Courant, 2018; Elkink et al., 2016; Farrell et al., 2019). The success of citizens’ assemblies across very different political and cultural contexts is an encouraging sign for their success in the New Zealand context also (Hayward, 2014). In New Zealand, citizens’ assemblies have their precedent in a number of methods designed to elucidate the public perspective through smaller, representative groups – for instance citizens’ juries, select committees, and citizens-initiated referenda (Ballance, 2018; Coppell et al., 2020; Palmer & Butler, 2018).

A citizens’ assembly functions by providing a hypothetical model of what the public would think if it were exposed to the same process, if it were given a more adequate chance for considering the issue (Fishkin, 1991). It does this by gathering a representative sample of the nation, with all relevant perspectives present, and exposing them to a process of deliberation. This produces an accurate approximation of the national public-will which can be used to inform policy. When the assembly is properly designed, the conclusion it produces can be treated as the consensus of New Zealanders when their opinions are properly informed, public-spirited, and authentic.

The success of a citizens’ assembly relies on it achieving both genuine deliberation and representation (understood as an ideal to be closely approximated; the securing of inclusivity in a process which cannot otherwise accommodate the entire citizenry). Because a citizens’ assembly involves national decisions being made by a group other than the nation, it must be properly legitimized to the public, so that they accept the decision as their own. Citizens must be able to understand themselves as the authors of the laws to which they are subject (Habermas, 1996).

Why a Citizens’ Assembly Is Suitable for Alcohol Policy

The situation of alcohol policy in New Zealand is characterized in way which makes it particularly appropriate for a citizens’ assembly. Principally, there exists both (a) a widespread desire and need for change; and (b) a recognition that our current approach is not working. Despite repeated evidence of alcohol's harm, the inefficacy of our current policy, and the existence of alternative evidence-based policies, the political system has failed to produce a satisfactory solution (Fournier et al., 2011b). Therefore, there may be value in exploring alternative approaches to policymaking.

Of possible methods, a citizens’ assembly is particularly appropriate because it fosters an impartial, evidence-based approach to alcohol reform which is insulated from bias, false evidence, and misinformation (Gutmann & Thompson, 2004; Isernia & Fishkin, 2014). This is especially important in the context of alcohol reform, which has been frequently hindered by such influences in the past (Sellman et al., 2018; Walker, 2019). A citizens’ assembly could play a role of confutation, exposing and dispelling myths which subvert evidence-based policy formation (Niemeyer, 2014). Furthermore, rather than accentuating tension over such a polarizing and contentious issue, citizens’ assemblies could lead to increased tolerance and understanding by encouraging discussion and shared perspectives (Kavadias, 2014). Deliberation develops empathy, which is crucial on complex and divisive issues such as alcohol policy (Bannock, 2019).

This aims to develop a more authentic approximation of what the New Zealand public desires than that produced by other existing democratic methods, such as pure representative or aggregative democracy (Elstub, 2008; Isernia & Fishkin, 2014; Landwehr, 2014; Smith, 2011). The assembly is designed to ensure political equality and an accurate representation of the amalgamated nation (Parkinson, 2003). This group is then exposed to a thorough process of deliberation to increase the likelihood that their conclusion closely approximates the true public-will – what the nation “actually” wants once its decision-making is unhindered by misinformation, misunderstanding, intolerance, or the constraints of time and energy (Elstub, 2008; Isernia & Fishkin, 2014; Landwehr, 2014; Smith, 2011). Representation is also designed to ensure the inclusion of minority perspectives. In this way a citizens’ assembly can advance justice by giving a voice to groups who are otherwise politically weak, disadvantaged or exclude (Fung, 2003). This is of particular relevance to the New Zealand context, in which Māori only make up 16.5% of the population (Statistics NZ, 2020) and yet are also equal partners in the Treaty of Waitangi. Aggregative democracy therefore risks being at odds with the intent of the Treaty – whereas a citizens’ assembly offers a solution.

Under current policy reform processes, the perceived political cost (both financial and electoral) entailed by assuming responsibility for reform has prevented a sufficient quantity of politicians from doing so (Carolan, 2015; Suiter, 2018). A citizens’ assembly provides a compromise, recognizing that politicians run a considerable risk in addressing certain topics, but that these topics need addressing, and can be successfully addressed by removing them from the direct political arena. This helps to reduce the risk of deliberate political manipulation and obfuscation, such as that which alcohol policy formation has been accused of in the past (Sellman et al., 2018; Walker, 2019). With alcohol, there is a significant financial incentive for certain groups to continue encouraging unhealthy drinking (Sellman et al., 2018). Political interference from the alcohol industry is particularly common for the most effective alcohol policies (WHO, 2018). A citizens’ assembly can shield policy formation from this undue influence to ensure policy reform is more responsive to what the nation needs and guided more accurately by what the public genuinely desires. In doing so, it can help to mend the dislocation between the public and politics, as citizens have their sovereignty restored in decision-making processes (Fung, 2003; Van Reybrouck, 2018).

A citizens’ assembly is preferable than aggregative methods for policy reform, such as referenda, because it provides a better approximation of the informed public-will. Referenda are merely the accumulation of individual preferences, and in no way ensure that the voting public understand or debate the issue (Hayward, 2014). Furthermore, the very nature of referenda undermines the voice of Māori and other minority groups, including those most affected by the policy in question, and can consign them to the tyranny of the majority (Quince, 2020). Citizens’ assemblies are also preferred over other types of deliberative forums because they empower the final consensus to have a tangible influence on political decision-making, whereas other forums, such as citizens’ juries, are less widely advertised, and more liable to being overlooked or deliberately ignored. The purposeful constitution of a citizens’ assembly provides protection against it becoming a merely glorified opinion poll (Renwick & Hazell, 2017; Smith, 2009).

Citizens will not always meet the scientific ideals of objectivity and evidence. Yet, certain models of deliberation encourage and allow for this, out of recognition for the benefits these perspectives bring which can be overlooked in purely theoretical approaches to the issue (Lancaster et al., 2018) - they consider evidence as “all types of knowledges, including that derived from science treated equally with lived experience, personal anecdote and values” (Ritter et al., 2018). Citizens are the ones immersed within the world of alcohol, living within its harms, but also the social opportunity and pleasures, it brings – and they have the political right to manifest their beliefs as public policy. With the support of an appropriately designed forum, they constitute an invaluable policymaking force.

Criticisms of Citizens’ Assemblies

As a novel and inclusive innovation for increasing citizen participation it is unsurprising that citizens’ assemblies are popular. However, alongside the objections to deliberative democracy previously highlighted, there are many criticisms of citizens’ assemblies which have arisen from practical attempts: Deliberative forums are costly, require careful expectation management, and are likely to be subject to selection biases given how difficult sociological representativeness is to attain via self-selection (Font & Blanco, 2007; House of Commons Library, 2007). The most significant bias tends to be education, with lower levels of education clearly underrepresented (Font & Blanco, 2007). Attrition is common, and tends to leave panels “more white, middle class, professional and activist” than the “base populations” (Ritter & McLauchlan, 2022). This leads to difficulties in trying to balance a short assembly – to prevent attrition and fit with the need to react to short political cycles (Breckon et al., 2019) – alongside the need for good quality deliberation, something which tends to take time. The questions asked are sometimes leading, which risks skewing deliberation from the start (Participatory Learning and Action, 2008b), or they can be too broad and complex to allow for useful recommendations (Ritter & McLauchlan, 2022). Facilitators have also been observed to have an influence – steering deliberation away from unbiased perspectives (Participatory Learning and Action, 2008a; Ritter & McLauchlan, 2022). On the other hand, the expectation that participants can and should suspend their personal to begin deliberation with an unbiased perspective may not only be unrealistic, but also suppress valuable insights from testimony (Krinks et al., 2015).

Politicians can be hesitant to cooperate with assemblies if they are nervous about the outcome (Wakeford et al., 2008). Recommendations are often ignored (Wakeford et al., 2008) or are hindered by delays and breakdowns in communication (Participatory Learning and Action, 2008a; Ritter & McLauchlan, 2022). Assemblies have been accused of being “incidental” – created only to conform with a “consultation exercise being perceived as not particularly relevant to the agency's core work” (Kashefi & Mort, 2004). When deliberative forums do succeed, it is often with a limited, irregular, and diffuse impact (Font & Blanco, 2007; Kashefi & Mort, 2004; Ritter & McLauchlan, 2022), and there are concerns that they only achieve their goals where there already existed a great deal of legislative momentum (Ritter & McLauchlan, 2022), rather than instigating this themselves. Similarly, as “isolated experiences in a context of very limited opportunities for participation” (Font & Blanco, 2007) it has also been objected that citizens’ assemblies fail to encourage widespread deliberation amongst the public (Pal, 2012), especially in the absence of adequate media attention (Wakeford et al., 2008).

Finally, such forums are concerningly susceptible to influence by external groups (Pal, 2012). This is especially as a result of their small size making them a high-leverage, low-cost way of influencing public policy, as well as the lack of assembly input in agenda-setting (Neblo, 2015; Setälä, 2011). In particular, if the outcome is not binding then citizens’ assemblies afford a convenient way for policy-makers to cherry-pick recommendations which best suit their political agenda (Grönlund et al., 2014).

What Would a Citizens’ Assembly Look Like?

If, as this essay argues, a citizens’ assembly would be appropriate for alcohol policy reform, consideration must be given to its format. The following recommendations have been developed after a careful analysis of previous assemblies, particularly those of Canada and Ireland, as well as a recognition of New Zealand's unique bicultural context. If an assembly were to take place, modifications to the proposed format should be carefully justified to ensure that they do not jeopardize the legitimacy of the assembly, nor the accuracy of its recommendations.

Agenda-Setting

The agenda-setting is one of the most important aspects of a citizens’ assembly because it fundamentally influences the direction of deliberation (Lang, 2008; O'Flynn & Sood, 2014; Smith, 2009). The purpose of the agenda should be to highlight the issue at stake without influencing its outcome (Pal, 2012; Setälä, 2017; Smith, 2009; The Irish Citizens’ Assembly, 2018). An assembly's recommendation can only be legitimate if the agenda is developed impartially (Lang, 2008; Setälä, 2017). It is therefore recommended that the agenda is established by the government, with advice from political and public health experts (Renwick & Palese, 2019; Smith, 2009; The Irish Citizens’ Assembly, 2018), and then reviewed by an impartial body to determine its appropriateness (The Irish Citizens’ Assembly, 2018). The assembly should ideally avoid being held in conjunction with other topics as this detracts from both assembly member and public attention for each topic. Expanding the agenda only dilutes emphasis and focus and detracts from the potential to enact meaningful change (Renwick & Hazell, 2017; Smith, 2009).

Member Selection

Many previous assemblies have had memberships lying between 100–150 participants as it is thought that this represents an ideal balance between financial, representative, and deliberative quality constraints (Renwick & Hazell, 2017). Too few members compromises accurate representation, whereas too many compromises deliberative quality, and makes the assembly more costly to run. These members must be carefully selected to fulfil the vital criteria of representation. Only when this is sufficiently approximated can the recommendation of the citizens’ assembly be used to indicate the public-will and will the public authorize the assembly as their representatives (Luskin et al., 2002). Representation is closely approximated by random sampling. This also ensures political equality by allowing each citizen the chance to participate (Fishkin & Luskin, 2005). Elections should not be used to select members because they are unlikely to select “ordinary” representative citizens (Renwick & Hazell, 2017).

Stratification further secures political equality by ensuring that certain perspectives, identified as critical to the topic, are not left unrepresented by the chance sweep of randomization (Grönlund et al., 2014; Karpowitz & Raphael, 2016; Street et al., 2014). Moreover, stratification mitigates self-selection bias by evening out inequalities in the willingness of different groups of people to participate (Renwick & Palese, 2019). The assembly cannot feasibly screen for every relevant characteristic in the pursuit of descriptive similarity (James, 2008). Based on the topic of alcohol policy, some important initial strata to include are: gender, geography, age, socio-economic status, employment status, and ethnic background (Renwick & Hazell, 2017). In New Zealand, Māori representation will be especially vital, including considerations of representation across iwi ⁴ .

Self-Selection

Self-selection is an unavoidable element of any citizens’ assembly, because citizens cannot be made to participate (Fung, 2003; Morrell, 2014) so “Those who accept the invitation to participate are, by definition, different from those who do not accept the invitation” (Renwick & Hazell, 2017). But even with self-selection, citizens’ assemblies resemble the public more accurately than other common policymaking institutions (Fournier et al., 2011b; Renwick & Hazell, 2017). Indeed, voluntary participation is necessary and decreases the risk of attrition. Furthermore, as Font and Blanco (2007) argue, the purpose of citizens’ assemblies cannot “be statistical representativeness, but rather an adequate degree of heterogeneity” (Font & Blanco, 2007).

To maximize the range of people who are willing to participate and mitigate the influence of self-selection, affirmative action can be incorporated into recruitment processes (Karpowitz & Raphael, 2016; Renwick & Hazell, 2017). The assembly can be publicized and especially encouraged in communities that would otherwise be underrepresented, or those whose perspectives are deemed to be especially relevant to the matter. Members should be financially compensated for their time (Fung, 2003; Renwick & Palese, 2019) and barriers to attendance should be minimized where possible. To this end, it is recommended that assistance also be provided for transport, lodging, childcare, and so on (Renwick & Palese, 2019).

Exclusion Criteria

Some groups must ideally be excluded from participation in the assembly in order to safeguard its impartiality and legitimacy (Devaney et al., 2020; Renwick & Hazell, 2017). Although this contravenes direct representation, it is a decision made in consideration of the threat their presence would have to the ascertainment of an accurate public-will. Their opinions are nonetheless valuable, but it is preferable to include these at the consultation or submission phase (Renwick & Hazell, 2017). In the context of alcohol reform, consideration should be given toward excluding those with a significant financial interest or working directly in the alcohol industry, public health experts or those identified with substantial involvement in alcohol reform, and politicians as they are directly involved in the legislation of alcohol and may be influenced by their party stance and considerations of voter or sponsor preferences.

Assembly Staff

Chair and Secretariat

The chair of the assembly is involved with the assembly's establishment and coordinating all phases of its proceeding. They are the public face of the assembly, and guide assembly members and staff, exemplifying the assembly spirit. Importantly, the chair must remain impartial and refrain from expressing their view or influencing the expression of that of others (Fournier et al., 2011c; Landwehr, 2014; Renwick & Hazell, 2017). It is therefore recommended that the chair does not have any substantive prior involvement in alcohol policy (Fournier et al., 2011c). The chair is commonly assisted by a secretariat for organizing and administrating the assembly (Renwick & Hazell, 2017). Some of their tasks include booking venues and accommodation, arranging food, supporting the practical needs of members and speakers, facilitating the website, and managing media and public engagement (Renwick & Hazell, 2017).

Academic Staff

An academic panel is required in order to curate material for the learning phase of the assembly. The panel is tasked with developing a curriculum without leading members to a specific conclusion. The information must be accurate, impartial, clear, accessible, and cover the widest possible range of options and views (Renwick & Hazell, 2017).

Design of the Assembly Process

Like previous assemblies, a citizens’ assembly on alcohol reform should be composed of three key phases: Learning, consultation, and discussion (Citizens’ Assembly of Scotland, 2020b; Renwick & Hazell, 2017; The Irish Citizens’ Assembly, 2017).

The learning phase is designed to educate assembly members on alcohol and alcohol reform in New Zealand. Throughout this phase it is vital that the information presented is impartial as it will form a key platform for future debate and discussions (Landwehr, 2014; Renwick & Hazell, 2017; Smith, 2009). The learning phase is typically composed of lectures regularly interspersed with group discussion sessions. Relevant experts can also be invited to briefly discuss their field of research and answer questions (Farrell et al., 2019).

The consultation phase is designed to increase the range of perspectives available to assembly participants. It also strengthens the relationship between the assembly and the public, because the public can make their voice heard and use this to influence the assembly recommendations (Fournier et al., 2011c). The assembly should explicitly aim to hear from certain groups of society which are identified as being important to the debate. If certain groups were excluded from assembly membership (such as politicians, public health experts, and those working in the alcohol industry) it is at this phase that efforts should be made to include their perspectives (Renwick & Hazell, 2017).

Lastly, the discussion phase must be designed to ensure high quality deliberation. This requires a clear task and an adequate timeframe to afford members ample opportunity to discuss and debate the topic (Blais et al., 2008; Smith, 2009). It is recommended that the discussion phase ends with a secret ballot to determine the final assembly recommendations and only large majorities are accepted as grounds for the assembly's acceptance or rejection of its proposals (O'Flynn & Sood, 2014; Ryan & Smith, 2014; Thompson, 2008).

Small group discussion, coordinated by specialist facilitators, is an important component of the assembly structure. They can be used consistently throughout all phases to offer citizens more time to confront differences, correct inaccuracies, break down barriers, express themselves more freely, and work together to find areas of mutual understanding (Ratner, 2008; Smith, 2009). Group membership should ideally be frequently rotated to ensure participants are exposed to many perspectives, and prevent the formation of like-minded bands (Smith, 2009).

Assembly Duration

The assembly duration dictates the extent to which members can engage in high-quality deliberation. Consistent feedback from previous assemblies was that members felt rushed, and certain phases were not allocated adequate time (Rose, 2007). A longer timeframe increases the scope for thoughtful and informed decision-making (Renwick & Hazell, 2017). But a longer assembly costs more and asks a lot more of its members, risking attrition (The Irish Citizens’ Assembly, 2018). The cost should be justified in light of the salience of the issue, and the costs which inadequate policy inflict on society (Renwick & Hazell, 2017). The experience of past assemblies suggests it is a worthwhile investment (Fournier et al., 2011d; Renwick & Palese, 2019). Sessions should ideally be scheduled to run at times that maximize opportunities for participation (e.g., Friday evenings to Sunday afternoon) (Renwick & Hazell, 2017).

Assembly Funding

It is unlikely that any organization other than the government would be able to fund an assembly of sufficient size, duration, and quality in order to ensure its recommendation was representative of the public-will. For its recommendation to have any effect, an assembly requires legitimization in the public eye, and it is therefore vital that it is established by a body perceived as impartial (Kerkin, 2017; Neblo, 2015; Parkinson, 2003). The experience of previous assemblies suggests that the government, seen optimistically as a purveyor of social good, provides the best assurance of this (Citizens’ Assembly of Scotland, 2020b; Renwick & Hazell, 2017; The Irish Citizens’ Assembly, 2017). Other alternatives run the risk of compromising the assembly's impartiality (The Economist, 2020).

As an estimate, based on previous successful assemblies, funding would cost between NZ$3-9 million (Citizens’ Assembly of Scotland, 2020a; Renwick & Hazell, 2017). Although a proper assembly would be expensive, it can be justified by the current social and economic harms of alcohol, which are estimated to cost at least $7.85 billion dollars annually (Nana, 2018). It would also be less expensive and potentially more effective than other available methods of reform, such as a referendum (Renwick & Hazell, 2017).

External Engagement

For the citizens’ assembly to be regarded as legitimate by the public, and to therefore warrant the translation of its recommendations into policy, it must work to secure external engagement (Renwick & Hazell, 2017; Setälä, 2011). Indeed, one of the main reasons past assemblies failed to influence policy was their failure to fully realize publicity (Smith, 2009). This is difficult, because regardless of how well designed the assembly is, external engagement relies on the optional interest of independent agents.

For all publicity, though particularly in the case of political agents, it is preferable that they endorse and publicize the assembly itself, without endorsing a particular outcome. Publicity should ideally advertise the assembly's existence and the reasons for its legitimacy, without attempting to influence its recommendations (Smith, 2009). Consideration should be given to the timing of the assembly's recommendations: Where possible, the assembly should avoid delivering its recommendations at a time when the government is distracted, such as during general elections, as this risks the issue being obscured by the election contest (Pal, 2012; Renwick & Hazell, 2017).

The public has not gone through the same process of deliberation as the assembly, so a large gap may exist between their level of understanding and that of the assembly (Smith, 2009). The media can serve to close this gap and, where this is not possible, communicate that the assembly is well-placed to simulate their preferences post-deliberation (Neblo, 2015). Indeed, as Lukensmeyer suggests, media is “the only way” to achieve the necessary scale of information dissemination imperative to deliberative democracy and a healthy democratic society (Lukensmeyer, 2014). This is an area in which all previous assemblies have struggled (Citizens’ Assembly of Scotland, 2020b; Fournier et al., 2011d; Renwick & Hazell, 2017; The Irish Citizens’ Assembly, 2017). Deliberative democracy does not mix well with media attention, as it undermines a pressure-free environment and can spread misinformation or emphasize the wrong factors (Smith, 2009). Given its salience, it is likely that a novel approach to alcohol reform would be granted substantial media attention, but this is not guaranteed, and the media emphasis may be inappropriate.

As such, the media should be directed wherever possible, to reduce attempts to push certain agendas or prioritize viewership over accuracy (Neblo, 2015). This can be achieved by centralizing assembly information through press releases and using the chair as spokesperson. Members should refrain from giving interviews or public commentary without permission, including over social media (The Irish Citizens’ Assembly, 2017, 2020). In efforts to reduce misinformation, the assembly could also subsidize public information with impartial and vetted documents. Of course, there is no guarantee that independent media adheres to these standards, but facilitating them in this way, and encouraging reliable coverage, improves the chances of this occurring.

Role in Decision Making

Consideration needs to be given to the role the assembly's recommendations will have on the government and policy. This needs to be clearly established in advance.

In the past some assemblies have served a purely advisory role – highlighting the public-will to the government so that it can then use this to guide their policy decisions (Citizens’ Assembly of Scotland, 2020b; Fournier et al., 2011a; Renwick & Palese, 2019; The Economist, 2020). An advisory assembly is discouraged, because then the implementation of recommendations is still vulnerable to external influence and cherry-picking (Grönlund et al., 2014). In the context of alcohol reform, one of the primary purposes of a citizens’ assembly is to circumvent the political barriers to change, including the influence of vested interests in the form of lobby groups. When the assembly is designed properly, its recommendations aim to represent the informed public-will. This is not something to be cherry-picked, because the government is meant to be directly representative and responsive to this. It is therefore recommended that some form of contract exist between the assembly and the government, with commitment from the government to consider each recommendation, or translate them directly into policy (Devaney et al., 2020; Goodin, 2008).

Referendum

Owing to concerns for the illegitimacy of assembly-guided policy, past jurisdictions have followed their assemblies with referenda. This is discouraged, because the assembly is designed explicitly to overcome misinformation and uninformed opinions, whereas referenda risk squandering the value of deliberative democracy by placing policymaking back into the hands of the less-informed public of aggregative democracy (Fournier et al., 2011b). This goes against the very reason an assembly is proposed, because high quality deliberation does not often occur naturally. The assembly must aim to achieve the role of a facilitative trustee, wherein the public trusts their recommendations as their own, had they been exposed to the same levels of deliberation (Pearse, 2008; Warren & Gastil, 2015).

Nevertheless, a referendum may be insisted upon because it helps secure reform legitimacy. The perception of self-governance is strongly associated with aggregative democracy; referenda offer an overt and secure way of guaranteeing equality (Fournier et al., 2011b; Smith, 2009). Because of the novelty of a citizens’ assembly compared to the long history and public recognition of referenda by the New Zealand public, the government may be hesitant to host an assembly without an accompanying referendum. A referendum, if it arises, would be out of political necessity rather than best-practice.

The Case for a Māori Citizens’ Assembly

The long history of alcohol-related harm disproportionately suffered by Māori (Cobiac & Wilson, 2018; Kõlves et al., 2020), coupled with the systemic disempowerment of their ability to protect their communities warrants a reinvigoration of the Māori voice and their sovereignty in the formation of alcohol policy. There is a need to acknowledge the colonial roots of Māori alcohol abuse and “to ensure solutions [are] founded on a culturally appropriate model that [does] not treat the individual in isolation but as part of a whānau, ⁵ hapū ⁶ and iwi ⁷ ” (New Zealand Law Commission, 2010).

In this respect, contemporary alcohol policy is failing. Alcohol is the most widely used drug by Māori, and yet they barely feature in the government's current consideration of alcohol control (Walker, 2019). As Māori warden David Ratu states “Every piece of legislation out there has a treaty [of Waitangi] clause in it, except alcohol; except the one that has the biggest impact on and does the most damage to my people. That is simply not good enough” (Thomas, 2017).

A citizens’ assembly would “promote intercultural understanding that aims to erode linguistic, religious, and cultural barriers in the production of consequent and mutually beneficial political dialogues” (Soares et al., 2021) and thereby provide a solution to this problem - one which is superior to other forms of policy-formation because, unlike them, it fundamentally privileges the considered perspectives of all New Zealanders. Having identified the Māori and minority voices as especially relevant to alcohol policy, a citizens’ assembly would accord them the appropriate influence. This is achieved by overrepresenting them where necessary in order to (a) communicate a group's perspective; (b) provide enough members of a group to bolster their confidence to speak their mind; (c) have sufficient members of a group so that their perspectives are distributed throughout break-out groups; and (d) to provide the plurality of viewpoints within a visible minority group perspective (James, 2008; Karpowitz & Raphael, 2016). Other methods, such as referenda and direct democracy, are less suitable for issues that have a disproportionate impact on minorities because they are more prone to misrepresent their needs by failing to adequately weight and accommodate for the complexity and plurality of minority perspectives.

Māori long ago acknowledged the value of deliberation, indeed participatory democracy is entrenched in Māori culture - as exemplified in the ad hoc rūnanga assemblies which employ discussion and debate in the resolution of issues (Palmer & Butler, 2018; Smith et al., 2021). In light of the disproportionate harm of alcohol experienced by Māori it is particularly appropriate to actively engage with these Māori legal traditions in an effort to strengthen these deliberative practices and empower Māori communities (Palmer & Butler, 2018). In recognition of Māori traditional practices, the value of a Māori voice in alcohol policy, Te Tiriti o Waitangi, and the spirit of bicultural Aotearoa, it is recommended that Māori leadership, including the involvement of kaumatua, should be engaged at all stages of the assembly.

The Citizens’ Assembly Format

There are two formats which a citizens’ assembly could take in order to best incorporate the Māori voice.

Firstly, there is the option of instituting a single assembly with Māori overrepresentation. This is the approach advocated for by deliberative theorists and used by previous assemblies because it preserves the unity of voice offered by the recommendations. In New Zealand, the extent and degree of Māori alcohol-related harm would warrant significant Māori overrepresentation. A tentative proposal would be to represent Māori at a ratio 150%-200% that of their population proportion. Overrepresentation could also be accompanied by a number of other techniques designed to adequately secure a Māori voice, such as the inclusion of Māori leadership and perspectives at the planning phase, and the use of active recruitment and community outreach groups at the member selection phase.

Such an assembly would be more in line with the philosophy and practicality of deliberative institutions. Members learn from one another and come to understand and respect alternative points of view. This process of discussion, at the heart of deliberative democracy, allows minority voices to be judged by their reason, not numerical strength. It therefore fosters tolerance and public-spiritedness, and shields minority voices from the tyranny of the majority. The assembly would only require a single forum and staff team. The final recommendation carries with it the weight of a single, united, and unambiguous voice, which makes it easier to establish a clear contract between the government and the assembly.

Secondly, there is the more novel approach of hosting two citizens’ assemblies, both identical in format, but one to represent the amalgamated nation and one to represent the Māori perspective alone. As with the single assembly option, significant emphasis would be placed on incorporating Māori leadership and perspectives at the planning phases, but the second assembly goes further by allowing Māori to independently mould the assembly's establishment and proceedings in line with their values and priorities. The rationale for this is grounded in the magnitude of alcohol-related harm experienced by Māori and an understanding that this is best redressed by appealing to their own independent perspectives. Such an approach does not neglect the nation's voice, because this is ascertained in the national assembly. The idea is to establish both the Māori and the national public-will and have these both enter into policymaking processes. This is out of recognition that they are both pertinent and that a singular assembly would not elicit them to the same extent. Such an approach is more in line with the political reality of a bicultural Aotearoa New Zealand. In recognition of the importance of language, we also propose that members of this Māori assembly are given the opportunity to decide on a Te Reo name for the citizens’ assemblies on alcohol policy themselves.

At the end of each assembly, both assemblies would publish their recommendations. Where these concur, they can be treated as an amalgamated national public-will, and there are very strong grounds for reform. This is one main advantage of such a format – the government has fewer grounds for rejecting reform proposed both by the national and the Māori public-will. Following this, a further assembly meeting would take place, where members from each assembly join to share the rationale for their recommendations where these diverge. Through this consensus-building process, the two assemblies may further concur on additional recommendations. Finally, conflicting recommendations which the assemblies cannot reconcile are nevertheless useful to elicit because they offer valuable information to the government.

This two-assembly format exposes the difficulty of reconciling the benefits of deliberation with those of biculturalism. Firstly, it undermines deliberation because it compromises the sharing of perspectives and the learning which ensues from their amalgamation. Instead, the Māori perspective is somewhat isolated. Secondly, the two-assembly format may threaten the overall legitimacy of the process. By having two separate assemblies, the public may identify less with the process, and become disillusioned with the accuracy of its representation. But legitimacy should not be a concern with recommendations that the assemblies concur on, nor those reached following a consensus-building process, as the public can be confident it was out of recognition of a superior argument and not because they were obliged or outvoted by aggregative measures. If a consensus does not eventuate on a particular recommendation, it would not be implemented. Thus, at least in theory, there is no a priori reason to delegitimize the two-assembly format.

International Applicability

While these recommendations have been framed in a New Zealand context they may also be of international interest, given alcohol is a vexed problem globally, and other jurisdictions are therefore also seeking to reduce alcohol-related harm. The citizens’ assembly model may also be an appealing prospect for intervention in other contentious policy domains, such as other drug policy issues, which have suffered from similar obstacles to progress. Here too, a deliberative approach can help to overcome misinformation, bias, complexity, competing financial interests, and entrenched political resistance. If such an assembly does occur, any deviation from the above recommendations must be justified following a careful consideration of the assembly's new context, and the cost any assembly-design modification might pose to the assembly's legitimacy. Principally, organizers must remain vigilant against attempts to subvert the assembly through any unwarranted alterations.

Conclusion

There exists strong evidence that alcohol is harmful, that New Zealanders consume too much, and harm reduction opportunities are well-known. The case for policy change is strong, and there is evidence that New Zealander's support alcohol reform; the question, then, is why hasn't this occurred? A citizens’ assembly provides an approach to this policy issue which may be less clouded by misinformation or external influence and delivered in a format which can readily guide policymaking. By recruiting a representative sample of participants, such an assembly models what the New Zealand nation would think, hypothetically, if it were exposed to a thorough process of education, public consultation, and deliberation. The policies which then ensue are likely to be epistemically better, more ethically robust, and politically more legitimate than decisions made using other pre-existing methods (Calvert & Warren, 2014). They also have the potential to better incorporate New Zealand's unique bicultural voices, in accord with Treaty of Waitangi obligations. This essay therefore concludes that a citizens’ assembly, appropriately designed, is one of the most promising methods for reinvigorating the stalled alcohol policy reform in New Zealand.