BY PAUL KILDEA
Prime Minister Malcolm Turnbull has pledged to hold a plebiscite on same-sex marriage early in the next parliamentary term. If his government is re-elected at the upcoming federal election, expected on 2 July, we may see a vote on the issue before the end of the year.
Just months out from a possible vote, we know very little about how the plebiscite will be run. While the conduct of referendums to amend the Constitution is governed by a combination of constitutional and statutory rules – notably those contained in s 128 of the Constitution and the Referendum (Machinery Provisions) Act 1984 – none of these apply to plebiscites. This leaves the federal Parliament enormous discretion in setting ground rules on such basic matters as the franchise, the manner of voting and campaign finance.
The Turnbull government has said that it will announce more details about the plebiscite after the election. This post examines some of the choices it will have to make as it finalises the framework for what is certain to be a high profile and contentious plebiscite campaign.
What is a plebiscite?
In many countries, the terms ‘referendum’ and ‘plebiscite’ are used interchangeably, but in Australia there is a meaningful distinction between the two. A ‘referendum’ generally refers to a popular vote on a proposed amendment to the Commonwealth Constitution or the State constitutions, while a ‘plebiscite’ concerns a vote by the people on other issues.
The legal consequences of each are also different. The result of a federal referendum is binding. A proposal that achieves the threshold specified in s 128 (a national majority of votes plus majorities in four states) must be presented to the Governor-General for royal assent. Conversely, no amendment can be effected in the event that this threshold is not reached. By contrast, a plebiscite generally has no legal effect and is ultimately a mechanism for testing public opinion on an issue. In the event of a Yes vote on same-sex marriage, Parliament would not be required to implement the popular will and, controversially, some conservative parliamentarians have indicated they would vote against marriage reform even if the plebiscite was successful. Anne Twomey has suggested the Parliament could overcome this limitation, to a degree, by passing an Act authorising same-sex marriage in advance of a popular vote and making its commencement contingent upon a successful plebiscite. But, as Twomey acknowledges, Parliament would retain the ability to repeal or amend that Act at a later date.
Turnbull’s proposed vote on same-sex marriage would be only the fourth national plebiscite since Federation. The first two – in 1916 and 1917 – asked voters whether they supported the introduction of compulsory military service during the First World War, and in each case the proposal was defeated by a small margin. In 1977, a third plebiscite asked voters for their preferences on a national song, with Advance Australia Fair prevailing over God Save the Queen, Waltzing Matilda and Song of Australia.
Plebiscites are more common at State and Territory level, where voters have had their say on a wide variety of social and economic issues including hotel closing hours, Sunday trading, prohibition and daylight saving.
Setting the ground rules
The Constitution makes no mention of plebiscites and there is no standing legislation at federal level that governs them. Governments and parliaments therefore have great flexibility in setting plebiscite rules. The same-sex marriage poll could, for instance, be conducted as a voluntary postal vote that requires a supermajority of 60 per cent of votes to succeed. It could include a franchise that extended to sixteen-year-olds.
Despite this flexibility, the most likely scenario is that the plebiscite on same-sex marriage will be run very much like a referendum. The government would risk damaging the credibility of the poll if it departed too much from standard referendum practice. The adoption of voluntary voting, for instance, would raise questions about government motives and would risk a low turnout that could further undermine confidence in the result. The same could be said of any move to conduct the plebiscite entirely by postal vote or to establish a special threshold for the measure to pass.
Even so, the Turnbull government faces some difficult decisions about how it wishes to conduct the plebiscite. The peculiar legal status of plebiscites, and the flexibility that comes with it, presents both challenges and opportunities in a range of areas, including the initiation of the plebiscite, setting the question, the franchise, public education and campaign finance.
Initiating a plebiscite
One of the first decisions the government will need to make is how it intends to initiate the plebiscite. Its strong preference appears to be to pass a plebiscite Act that provides for the holding of the poll and sets down rules about how it will be run (perhaps by adopting much of the process and machinery contained in the Referendum Act). The constitutional authority for this Act would be provided by the marriage power in s 51(xxi). This is the most desirable approach as it imbues the plebiscite with democratic legitimacy and gives parliamentarians a chance to openly and publicly debate the wording of the question, campaign ground rules and other process matters.
However, political circumstances may force the government to consider a different route. Both Labor and the Greens are strongly opposed to a plebiscite on same-sex marriage, in line with their preference for an immediate parliamentary vote on the issue. Depending on the make-up of the Senate after the next election, Turnbull might find that he cannot secure passage of a plebiscite Bill through the upper house. Should this occur, the government might decide to bypass Parliament and ask the Australian Electoral Commission (AEC) to conduct the plebiscite as a fee-for-service election under s 7A of the Commonwealth Electoral Act 1918 (Cth).
This provision permits the AEC to ‘make arrangements for the supply of goods or services to any person or body’. It enables the Commission to conduct a wide variety of votes, including trade union elections and workplace agreement ballots. In 2007 the Howard government utilised s 7A to allow the AEC to conduct plebiscites in Queensland that asked voters whether they supported the Beattie government’s proposed amalgamations of local councils. Similarly, a Turnbull government could instruct the AEC to enter in an agreement with, say, the Department of the Prime Minister and Cabinet to conduct a plebiscite on same-sex marriage.
While a fee-for-service plebiscite may seem attractive in the face of an oppositional Senate, it would come with a sting in the tail. The plebiscite would be the product of executive agreement and entirely severed from the various rules and protections that apply at referendums. Basic features of a referendum such as ballot secrecy, offence provisions and a process for challenging the result (through a court of disputed returns) would only apply if provided for in the terms of the agreement. Even then, only the parties to the agreement would be bound by these standards. This could give rise to some troubling scenarios. Campaigners could engage in conduct that is unlawful under the Referendum Act – say, the distribution of unauthorised campaign material – without fear of legal consequences. And, in the event that the plebiscite was won or lost by a small margin, there would be no process for challenging the result in a court of disputed returns. For all of these reasons, the Turnbull government will want to initiate the plebiscite through an Act of Parliament.
Setting the question
One of the more important functions of a plebiscite Act will be to set out the question that will appear on ballot papers. This is likely to be hotly debated in Parliament, as how the question is asked could influence the responses of voters and, hence, the plebiscite result.
Parliament has more control over the wording of plebiscite questions, and the manner in which they are asked, than is the case for referendums. The Referendum Act requires that ballot papers present voters with the long title of the Bill that proposed the constitutional alteration, followed by the question ‘Do you approve this proposed alteration?’. By contrast, Parliament can adopt whatever wording it likes for a plebiscite question. This frees legislators to put the question in a more plain and direct fashion, but also provides more scope to ‘load’ the question to nudge voters to answer one way or another.
It is possible to envisage how different wording could influence how voters respond to a question on same-sex marriage. For instance, one approach would be to ask voters if they support allowing same-sex marriage; another would be to ask whether marriage should be restricted to opposite sexes. Voters may respond differently to a question that asks them to change a ‘longstanding’, ‘traditional’ definition of marriage, compared to one that asks them if they support ‘marriage equality’. More neutral options are available. The ballot paper could ask voters if they support Parliament legislating for same-sex marriage; alternatively, if a substantive Bill is passed in advance of the plebiscite (as suggested by Twomey), the question could simply read: ‘Do you approve of the Bill entitled…[name of Bill]?’.
Whatever approach Parliament takes, it should allow its proposed question to be assessed by an independent body for clarity and neutrality. In the United Kingdom, the Electoral Commission has a statutory obligation to perform such assessments. Most recently, the Commission advised the UK government that its proposed question for the referendum on EU membership could be perceived as encouraging voters to favour the status quo. In response, the government altered its initial question, ‘Should the United Kingdom remain a member of the European Union?’ to ‘Should the United Kingdom remain a member of the European Union or leave the European Union?’. The AEC could serve a similar function in Australia.
In the absence of standing legislation, Parliament is free to expand or narrow the plebiscite franchise beyond usual electoral practice. This is in contrast to referendums, where the Referendum Act provides that the franchise is the same as for federal elections.
Parliament has taken advantage of this flexibility at previous national plebiscites, to the benefit of some groups and the detriment of others. Prior to the first vote on conscription, legislators disqualified from voting ‘any naturalised British subject’ who was born in an enemy country, such as Germany. Before the second conscription poll a year later, the Hughes government issued regulations that extended this disqualification to ‘every person whose father was born in an enemy country’. At the same time, Parliament departed from referendum practice to allow residents of federal territories to vote, a precedent that was followed at the 1977 plebiscite on the national song.
Looking ahead to a plebiscite on same-sex marriage, it is probable that Parliament will adopt the same franchise as exists for elections and referendums. However, some legislators might argue in favour of expanding the franchise to include 16- and 17-year olds. George Williams has written that members of this age group should be permitted to vote at the plebiscite given that they are legally allowed to marry (with the permission of a judge or magistrate) and so have a genuine stake in the outcome. Expanding the franchise in this way would be in line with what occurred at the Scottish independence referendum. The Scottish experience is encouraging, with research indicating that young people embraced the opportunity to have a say at the ballot box. It is estimated that turnout among those aged 16 and 17 (75 per cent) was higher than that for 18-24 year olds (54 per cent), albeit lower than that for the 35-54 and 55 and over age groups (85 and 92 per cent, respectively).
Also interesting is the question of whether the Parliament could narrow the franchise for the plebiscite on same-sex marriage. In Roach and Rowe, the High Court ruled that the ss 7 and 24 of the Constitution – which require that the Senate and House of Representatives be ‘directly chosen by the people’ – mandate universal adult suffrage for elections, subject to any exceptions that can be justified by a ‘substantial reason’. This limitation on Parliament’s power arguably applies to referendums, at least in an indirect sense, due to the requirement in s 128 that proposed constitutional amendments be submitted ‘to the electors qualified to vote for the election of members of the House of Representatives’.
Parliament would seem to face no such limitations with respect to the franchise for plebiscites. The constraint enshrined in ss 7 and 24 does not apply: when Australians cast a vote at a plebiscite, they are not engaged in a ‘choice’ of the kind contemplated by those sections. This raises the prospect that Parliament could exclude men, or left-handed people, from voting at the plebiscite and, subject to the High Court discovering and applying an implied constitutional guarantee (of equality, say, or of participation in democratic processes) there would appear to be no constitutional objection. Less facetiously, legislators could exclude overseas voters or prisoners from voting, or close the rolls to prevent new enrolments once the plebiscite writs are issued, without fear of censure from the courts.
Public education and campaign finance
There has already been speculation that the government will provide public funding to promote public education and support advocacy during the plebiscite campaign. Here, the absence of standing rules is a blessing as current referendum practices are flawed. The plebiscite presents an opportunity for Parliament to experiment with some new approaches.
One question is whether public funding should support the production and distribution of an official information pamphlet. This is a familiar feature of referendum campaigns. Under rules set out in the Referendum Act, the pamphlet contains arguments for and against the proposed constitutional amendment (authorised by a majority of the MPs who voted for or against the amendment in Parliament) plus a statement showing the proposed textual changes to the Constitution, and must be sent to each household. The official pamphlet was introduced over a century ago to help electors make an informed choice on complex issues of constitutional reform, but it has never really lived up to its high ideals. In recent decades the pamphlet has been criticised for being confusing, adversarial, and generally unhelpful to voters seeking to learn something about an issue. The slogan ‘Don’t Know? – Vote “NO”’, which appeared in the 1999 republican referendum pamphlet, is indicative of the sort of material that finds its way into this document.
The government should use the plebiscite to trial a pamphlet that provides more impartial information to voters. Practice in overseas jurisdictions shows that this can be effective. In Ireland, for instance, an independent Referendum Commission publishes a short booklet that describes the current state of the law and the legal consequences of a Yes vote. In 2015, Irish voters received an information leaflet of this kind prior to the referendum on same-sex marriage. There is no reason why the same could not occur in Australia. An official plebiscite pamphlet could, for instance, set out the current definition of marriage under the Marriage Act 1961 (Cth), and explain the effect of a Yes vote on religious and marriage celebrants, recognition of international marriages, and so on. The pamphlet could also include ‘Yes’ and ‘No’ cases authorised by MPs but, given their questionable value to public education, there would be no harm in leaving them out as part of the trial.
Another question is whether the government should provide public funding to help organisations promote arguments for and against change. Legislation prevents this practice at referendums, but this has not proved an insuperable barrier: in 1999, Parliament suspended the prohibition and the Howard government allocated $15 million to support the campaign activities of Yes and No committees.
While public funding to support advocacy has its merits, it should not be provided for the plebiscite on same-sex marriage. The arguments for and against same-sex marriage are so widely known that any benefit to public awareness would be minimal. There is also a risk that advocates would use the funding to publicise messages that many would find offensive and may cause harm to members of the LBGTI community. To take just one example, former Liberal MP Chris Miles recently prepared and funded an anti-gay marriage pamphlet that claimed that adult children raised by same-sex parents might experience a range of ‘social outcomes’ including unemployment, sexual victimisation, sexually transmitted disease and drug use and abuse. It would be problematic for public money to be used to disseminate material of this nature.
Whatever approach is taken to public funding, the plebiscite should be used as an opportunity to experiment with new controls on private money. The intense interest in same-sex marriage means that we are likely to see significant spending on advocacy by individuals and groups. Chris Miles, for instance, has said that he would be prepared to print and distribute ‘millions’ of copies of his pamphlet; more widely, it would be surprising if political parties, churches, and interest groups (such as Australian Marriage Equality) did not also spend money on campaigning. To ensure a level playing field, Parliament should impose expenditure limits on individuals and organisations that wish to participate in the campaign. This would be in line with laws in the United Kingdom and New Zealand. The plebiscite experience could then inform future debates about whether similar constraints should be introduced for referendums.
As we get closer to a possible plebiscite on same-sex marriage, public attention is naturally focussed on the merits of the arguments for and against change. But, before we each get our chance to have a say at the ballot box, critical issues of process need to be resolved. While the Turnbull government faces difficult choices, it should also seize the opportunity to experiment with innovations to the franchise, public education and campaign finance, with an eye to improving practices at future referendums.
Dr Paul Kildea is a lecturer at UNSW Law and the Director of the Referendums Project at the Gilbert + Tobin Centre of Public Law. This blog post draws on research the author is undertaking into the conduct of plebiscites in Australia. The author thanks Graeme Orr for helpful comments on an earlier draft.
Suggested citation: Paul Kildea, ‘Setting the ground rules for the same-sex marriage plebiscite’ on AUSPUBLAW (27 April 2016) <https://auspublaw.org/2016/04/same-sex-marriage-plebiscite/>