There is a good way to conduct government-initiated referendum, and there are bad ways to conduct them.
During the course of this Parliament, New Zealand will conduct one or perhaps two, referendums – one of the legalisation of cannabis use (a result of the Green Party’s confidence and supply agreement with the Labour Party), and perhaps another on euthanasia. Unfortunately, indications are not promising that the process for either these referendums will be good.
During September, October and November last year Australia conducted a referendum. We should learn from its mistake. Australia’s nationwide “plebiscite” on the legalisation of same sex marriage, in an exceedingly useful example of how not to conduct a public referendum.
How it went is no longer breaking news: a sizeable majority of the voting public indicated support, and then the Federal Parliament passed a law providing for it.
The marriage vote asked Australia voters to give a yes or no answer to the question:
“Should the law be changed to allow same-sex couples to marry?”
This is a perfectly reasonable question to ask to get a general sense of public feeling about a general issue, but is a stupid question to ask when wanting guidance on what a law should say.
How should you vote if you wanted politicians yet-to-consider a same-sex marriage bill to know that you would support changing the law to allow people of the same sex to marry, but would object to a law which might require churches to marry people in breach of church doctrine?
How should you vote if you wanted politicians to know that you thought people of the same sex should be able to marry, but would find offensive a system where a whole new law was created, setting up a separate same-sex marriage register, with separate same-sex marriage celebrants wholly separate from marriage celebrants?
You couldn’t. Holding a vote in advance of a bill being written, like they did in Australia, makes that impossible. You are being asked to vote in a vacuum – discussing the general idea of something that, in its final form, you may find you oppose.
This is a trap we should aim to avoid in both the cannabis referendum, and the potential referendum on euthanasia that New Zealand First is pushing for (and which David Seymour, the sponsor of the euthanasia bill, says he supports holding).
Because the intricacies of a same-sex marriage law are simple compared to either the regulation of cannabis or the regulation of euthanasia. These topics are much more wide-ranging, with vastly different possible systems of regulation.
And when people being asked to vote on something like this, people should know what trade-offs will be made.
Will euthanasia be limited to the terminally ill? Will it need a judge to sign off a decision to offer aid in dying? What will a doctor who refuses to take part have to do, if anything? Will advance directives be able to be enforced, or will applications have to be made by people who are conscious?
Will it be legal to sell cannabis, or just to possess and use it? Will people be able to grow their own, or will they have to buy it from specific government licenced dealers? Will you be able to smoke it outside in public places, like tobacco, or will that be prohibited, like the public consumption of alcohol often is? How will it be taxed? Will cannabis advertising and sponsorship be banned? Will councils have a role in regulating where it can be sold, or used? Will they be able to set up enforceable non-cannabis zones (like they can with alcohol), or only unenforced zones (like non-smoking areas)?
These are important questions, and for a lot of people, the answers to some of these questions will be decisive in how they cast their votes. So voters should know what the proposal actually involves.
A bill should be introduced. The House should debate it, it should receive full select committee consideration, advice should be sought, and submissions received. If MPs consider that broader public input into the detailed policy questions is needed, they, or the Government, should conduct market research: legitimate opinion polls and/or focus groups about specifics. They should then form their view as to exactly the regulatory framework they think is necessary, and put it in legislative form. And if it’s considered important for voters to have a direct say, at that point voters should get their opportunity to say no.
This legislative process is happening with the euthanasia law (select committee submissions are due by 20 February). But the early indications are that it will not happen on the question of recreational cannabis use. There, to date anyway, the suggestion seems to be there should be a general non-binding referendum, asking a general question.
If that is the plan, MPs should just not waste our time. If they want a general insight into public views on the general topic of cannabis legislation, they should simply engage Colmar Brunton to undertake a large opinion poll on the topic.
There is a legitimate debate about what is, and what is not, an appropriate topic to hold a public referendum over. The basic answer – those things where we do not MPs making the decision for us – only restates the question. New Zealand First has the view that certain issues, those commonly called conscience votes, should be for voters. This is far from a universally held opinion, but there is a broad consistency to it, and it fits within the mandate theory of democracy. In an MMP voting system, we vote for political parties that put forward policies. They don’t get to enact them all unless they get a majority, but they are broadly expected not to act inconsistently with their promises. But conscience votes are matters on which political parties tend not to have policies. For such policies to have a mandate, it must come from elsewhere.
There are answers to this concern, but the objection I usually hear to holding referendums on such issues – that they are blunt instruments, reducing complex policy questions to yes/no questions – doesn’t apply to a referendum held after the process I describe above.
When you get to the end of the legislative process, legislators face the same basic yes/no question: given what this bill contains, should it become law?
In the New Zealand House of Representatives, it’s asked in a form like “the question is that the motion be agreed to. Those of that opinion will say ‘aye’. The contrary, ‘no’.” There isn’t any nuance in this. If the motion that the bill be read a third time is agreed to, the bill is read a third time, and will become law. If the motion is rejected, the bill fails. It is highly unusual for a bill to get this far and fail. A bill that won’t become law is usually rejected at an earlier stage, but the principle is the same.
This is the approach we used in the 1993 referendum on the electoral system. It is the system we used in the 1997 referendum on compulsory retirement savings, and the system we used for the 2015 and 2016 flag referendums. It contrasts with the system we use in citizens-initiated referendums.
Unfortunately, while the legislative process underway for the euthanasia legislation should work through the detail of the scheme, any referendum seems likely to be an afterthought. The bill itself doesn’t provide for a referendum, and New Zealand First, and bill sponsor David Seymour say the proposal for a referendum will not be introduced until the committee of the whole. This is the second to last stage, well after any opportunity for public input. This is a problem. Electoral legislation should not be passed without the possibility for public input.
The obvious solution would a Referendum Act, a permanent law that, like the Electoral Act, would provide for the basic rules for a referendum any time Parliament wanted to hold one, but we don’t have one, so every time Parliament wants to hold a referendum they have to pass a law not only setting out the question, and the consequences, but also the rules around advertising, and counting the votes, and laws providing for secrecy, and prohibiting bribery, etc. They might get this right on the first attempt, but draft laws are easy things to stuff up, and public scrutiny of legislation is important.
Fortunately, it’s not too late. The euthanasia law is still early in its early legislative process, and the cannabis referendum isn’t set up yet. Hopefully, we can avoid not only the mistakes Australia made, but also new ones of our own.