The Australian Marriage Law Postal Survey: One year on


On 8 December 2017, Australia became the 26th country in the world to change the definition of marriage to include same-sex unions. In so doing, it became only the second advanced democracy to adopt marriage equality through a popular vote – a process largely driven by political necessity for the governing Coalition rather than preference by advocates. The short, yet tumultuous, campaign ended by sending a clear message to the Australian people that ‘love is love.’

In work recently published in the Australian Journal of Political Science however, we show the vote also revealed deep divisions in Australian society and presented important lessons for students of Australian politics. And so, one year onwards, it’s worth looking back at what exactly happened.

The politics of same-sex marriage has a fraught history in Australia. Almost all advanced democracies began to recognise same-sex partnerships in some form from the 1990s. However, despite several minor legislative attempts to broaden the definition of marriage, leaders of both Australia’s major parties continued to oppose change in the twenty-first century. The last such attempt was a private member’s bill in 2012, which – while unsuccessful – revealed deep differences within both major parties.

These differences came to a head after the Coalition’s narrow election win in July 2016, when a series of votes to hold a binding national plebiscite failed in parliament and a Liberal senator, Dean Smith, announced he would introduce legislation to enact same-sex marriage. And so, out of political necessity the government directed the Australian Bureau of Statistics (ABS) to hold a voluntary, non-binding ‘postal survey’ so that it could bypass parliamentary approval.

Only one other country – Ireland – has recognized same-sex marriage through a popular vote. Unlike Ireland however, Australia did not have the same strong tradition of referenda on so-called ‘moral’ issues, such as abortion, divorce and children’s rights. In a country where compulsory voting is the norm, a voluntary vote with no clear direction from the major parties posed a real challenge in predicting the ultimate result. In short, we were heading into the unknown.

Since the vote was conducted by the ABS, and not the Australian Electoral Commission (AEC), it was not subject to the stringent advertising restrictions that apply to federal elections. The rhetoric from both the ‘yes’ and ‘no’ sides was consistent with past arguments for and against same-sex marriage in other countries. Opponents of same-sex marriage argued it would undermine the traditional family structure and be a threat to freedom of religion. Supporters of marriage equality took issue with the zero-sum nature of the arguments from the ‘no’ campaign, arguing it was about the equality and dignity of hundreds of thousands of Australians. As Tiernan Bradly, director of the Equality Campaign, stated – ‘nobody will become less married, nobody will become more gay.’

Support for same-sex marriage remained relatively consistent throughout the campaign, varying between 55% and 63% with no clear upward or downward trend, while opposition rose marginally from 25% to 32%. The final results announced on 15 November 2017 showed a clear majority (61.6%) in favour of same-sex marriage, compared to 38.4% opposed. Concerns about turnout for a citizenry used to compulsory voting too were misplaced, with almost 80% of eligible Australians registering their votes.

Our research set out to explain how people ultimately voted in the postal survey. Specifically, we wanted to know what social/demographic factors, as well as political factors, determined the vote. To do this, we matched the results of the plebiscite with data from the 2016 census for each of the 150 federal electorates. We then analysed 428 census variables, refining them into four scales that represent the major social characteristics of each electorate. Finally, we collected data from Hansard on every vote and parliamentary speech each MP made in the House of Representatives from 2012 to 2016 to create a measure of how supportive the local MP for each electorate was.

Some of what we found was not surprising – for example, the ‘yes’ vote was highest in electorates with large numbers of affluent, well-educated citizens. It was lowest in electorates with large proportions of new immigrants, many of whom come from countries where same-sex marriage is illegal and seen as morally wrong. ‘Traditional’ electorates – with large proportions of older, married, Australian-born residents – were also more likely to vote ‘no’, while ‘suburban’ electorates were more likely to vote ‘yes’.

But, there were some surprises. For example, even though we expected multicultural electorates to be different from more homogenous ones, we underestimated just how different they’d be. These electorates were, by far, the most important predictor of the result in our data. Immigrants as a collective group usually have a small differential impact compared to the Australian-born; as ‘moral’ issues become more salient in advanced democracies, this could mean these groups might be more consequential to the eventual outcome of public policy debates.

We also found evidence of differential turnout made a big difference – each 1 percent increase in turnout increased the ‘yes’ vote by just under 0.8 percent. This too makes sense in international context; younger, more politically interested citizens have higher rates of participation in direct democracies, and also tend to be more likely to support same-sex marriage. This also suggests many opponents of same-sex marriage may have abstained from voting, either because they didn’t feel strongly enough about the issue, or because they thought they saw the writing on the wall.

We also looked at whether the proportion of same-sex couples in an electorate mattered for the outcome, and here too, the findings were contrary to expectations. Our data show that the presence of female couples was highly significant – but the presence of male couples didn’t seem to matter. Why might this be the case? We can only speculate – but some research from the United States may shed some light on this question. In the U.S., women couples are more interested in public ceremonies to confirm their partnerships and are more likely to be carers for children than male couples. By contrast, men are more likely to be interested in the legal arrangements underpinning their relationships, which were already effectively recognised prior to 2017 through domestic partnership legislation.

We also expected urban electorates to be more supportive of marriage equality than rural electorates, since many of these electorates are represented by more conservative MPs and parties. But, the data don’t bear this out – contrary to many of existing stereotypes of rural and regional Australia as very conservative, voters in these areas are no more or less supportive of marriage equality than voters in urban centres, when controlling for other factors.

Perhaps the most surprising finding was the impact of the local MP. While many pundits decry the lack of influence the average MP has, our data show that MPs who supported marriage equality may have influenced some of the voters in their electorates. We think undecided voters may have been more likely to take a cue from their MPs, another finding which should make students of Australian politics think twice about whether parliament still matters in 2018.

For both sides of the same-sex marriage debate, the plebiscite was a major compromise. Simply put, nobody thought a voluntary postal survey was the best way to come to a decision on a question like this. A year later, the issue of same sex marriage has largely faded from public view, but its impacts are still keenly felt. The ABC reports in the six months after parliament enacted marriage equality, about 2,500 couples have been able to celebrate their love through marriage. Last month however, the government-commissioned report on religious freedom was leaked to the press, which recommended public schools in Australia be given the right to turn away gay students. It seems clear while the issue of same-sex marriage is more settled than it was a year ago, the divisions it revealed have been far from healed.

This blog has been adapted from McAllister, Ian, and Feodor Snagovsky. 2018. “Explaining Voting in the 2017 Australian Same-Sex Marriage Plebiscite.” Australian Journal of Political Science 53 (4): 409–27. Read it for free for the next month here.

Feodor Snagovsky is a PhD Candidate in the School of Politics and International Relations, at the Australian National University and Ian McAllister is Professor of Politics in the School of Politics and International Relations, at the Australian National University.