The referendum to establish an Indigenous Voice to Parliament bears a unique historical burden. This proposal to change the Constitution is the first to come from outside the political system; written and authorised by our nation’s most disadvantaged group. But its success is dependent on the competence of the institution it seeks to reform.
Ideally the question to be put to the Australian people would have been framed by an inspirational government walking alongside a generous opposition and cross bench. Unfortunately, that wasn’t a realistic option given the cast of major and minor party players: a likeable but low-key prime minister, a culture-warrior opposition leader, and a melange of Greens and “teal” independents whose ability to persuade undecided voters stops at the boundaries of their respective electorates.
The next best option was pragmatic bipartisanship between the major parties, with the Greens and teals providing an urban top-up to the national and state-based “Yes” votes to offset the “No” votes coming from supporters of One Nation and the United Australia Party in the regions.
The point made repeatedly to Opposition Leader Peter Dutton in his private discussions with Prime Minister Anthony Albanese, and in the solicited and unsolicited advice from Liberal colleagues, was that bipartisanship was in the Coalition’s interests. Opposing the Voice risked further alienating the former Liberal voters who threw out Scott Morrison’s government at the last federal election. They would hold Dutton personally responsible for the failure of the referendum. And those who vote against the Voice won’t necessarily mark their ballot paper for the Liberals ahead of Labor at the next election, due in 2025. “Yes” won’t forget, while “No” will move on.
“If ‘Yes’ wins, he loses. If ‘No’ wins, he loses anyway,” is how a senior Liberal put it to me. It says something about the opposition leader that his friends and rivals would seek to convince him on the basis of the raw politics rather than any higher moral purpose.
The thesis that Dutton loses either way can be tested by revisiting the 1999 referendum on the republic with the hindsight of the Coalition’s landslide defeat in 2022. Seventeen Liberal electorates, many of which had been Liberal for their entire existence, voted “Yes” for the republic. Six were in Melbourne and six in Sydney, including then prime minister John Howard’s own seat of Bennelong. Adelaide contributed a further three Liberal electorates to the rebel alliance, while Brisbane and Perth had one each.
The lifetime Liberal voters who wanted to be rid of the monarchy became known as “doctors’ wives”. The Liberals worried for a time that they would follow up with a protest vote at a federal election over the asylum seeker issue, the war in Iraq and climate change. But they stuck with Howard’s government in 2001 and 2004, and remained loyal to the Liberal cause throughout the interruption of Rudd–Gillard governments between 2007 and 2013. When the monarchist Tony Abbott yelled the Coalition back into office in 2013, they were still on side. Only one of the original 17 seats – Adelaide – had shifted to Labor, and it stayed there in the first decade of the 21st century. A second electorate had defected by 2019 – Abbott’s own seat of Warringah, on Sydney’s northern beaches. But the 15 that remained allowed the Coalition to win a third term against expectations.
They would have stayed in the Liberal column if Scott Morrison had run a remotely normal government that didn’t pick fights with tertiary educated voters in the cities on issues such as climate change and the treatment of women.
Consider the locations, and recall the swings that turned these Liberal republican electorates Labor red, independent teal and green in May 2022.
In Melbourne, they formed a horseshoe of affluence from bayside Goldstein, through Higgins and Kooyong in the inner east, to Menzies, Deakin and Aston along the city’s eastern fringe. The teals took Goldstein and Kooyong at the federal election, while Labor claimed Higgins and then Aston at a byelection in April this year. This left Menzies and Deakin as the last two Liberal seats in metropolitan Melbourne.
In Sydney, Labor captured Bennelong, while the teals snared North Sydney and Wentworth to go with the previous gain of Warringah. The Liberal Party now has no seats with a view of Sydney Harbour or the northern beaches for the first time in its history.
Adelaide, Brisbane and Perth yielded one defector each: Boothby moved to Labor, Ryan fell to the Greens and Curtin went to the teals.
Only five of the original rebel 17 remain in Liberal hands, four of which – Deakin and Menzies in Melbourne, Bradfield in Sydney and Sturt in Adelaide – are considered marginal and will be on the hit lists of Labor and teals at the next election.
Now consider the Labor electorates that voted “Yes” in the 1999 referendum across Melbourne, Sydney, Newcastle, Wollongong, Canberra, Brisbane and Hobart. While the Liberals had ceded 12 of their 17 by 2022, Labor lost just four of its 25. The Greens took Melbourne in 2010 and have held it ever since, while independent Andrew Wilkie won the Hobart-based Denison at the same election (the seat was renamed Clark in 2017). The Greens also took Brisbane in 2022, which had previously moved from Labor to Liberal. The most intriguing of the four Labor defectors was Fowler; the only electorate to vote “Yes” in Sydney’s western suburbs in 1999 was taken by independent Dai Le in 2022.
The cities have shifted decisively to the left, and the voters driving this trend are educated women. The Liberal Party knows this, and it’s why two sharp sentences in its post-election review underline the potential for self-sabotage in Dutton’s approach to the Voice. First: “The Liberal Party now holds only four of the 44 inner metropolitan seats.” And second: “Of particular concern in the results is that in seats with high numbers of female professional voters, the Liberal Party only holds three of the top 30 seats where previously it held 15.”
The Coalition requires a net gain of 20 seats if it is to form a majority government. It won’t get there without the return of electorates likely to vote “Yes” for the Voice. But Dutton isn’t playing the long game of rebuilding the conservative side of politics in the cities. He wants to kill the Voice and shift the blame to Albanese for its defeat. His target audience is Labor’s blue-collar base in the outer suburbs and provincial towns, beyond the “Yes” zones of the capitals. A net loss of just three seats at the next election would see Albanese’s government slip into minority.
The historical burden that Labor carries on behalf of the Voice is that it has just one previous referendum victory to its name, and the primary-vote loyalty of just one third of the electorate at the moment.
Only eight of the 44 referendum questions put to the Australian people between 1906 and 1999 were carried with a majority of the national vote and a majority of the states, while a further five claimed the first majority but not the second.
Labor governments wrote 25 of those 44 questions, and succeeded just once: under Ben Chifley in 1946 “to give the Commonwealth power to legislate on a wide range of social services”. Labor has also been responsible for seven of the 10 least popular referendums, including Bob Hawke’s 1988 “rights and freedoms” question, which was rejected with a “No” vote of 69.2 per cent.
Conservative governments, by contrast, secured six of the 19 questions they wrote (seven if you count the defeat of the republic as a “win” for Howard).
The lessons of two failed referendums in particular are worth recalling for what they tell us about our democratic character, and what a “No” vote in the Voice referendum might mean for the conduct of federal politics in the future.
Labor’s John Curtin – arguably Australia’s greatest leader – sought to amend the Constitution in 1944 to extend the Commonwealth’s wartime powers for a period of five years to manage the peacetime reconstruction of the economy. Among the so-called 14 powers were measures addressing the rehabilitation of ex-servicemen, national health and family allowances, and to allow the Commonwealth to make laws on behalf of “the people of Aboriginal race”, rather than remaining the domain of the states. Guarantees of freedom of speech and religion were also to be inserted into the Constitution.
Robert Menzies’ United Australia Party, then in its death throes, opposed the referendum on the grounds that it would give too much power to bureaucrats, and that it was a Trojan horse for Labor’s socialist agenda.
The referendum was duly defeated with a national “No” vote of 54 per cent, and in four of the six states. The only states to vote “Yes” were South Australia and Curtin’s home state of Western Australia.
The most fascinating detail, though, was in the difference between the four million voters on the home front, and the 400,000 who were serving overseas in the armed forces. The diggers weren’t exposed to a “No” campaign and supported the referendum with a “Yes” vote of 53 per cent, while the people they were fighting for back home returned a “No” vote of 54.8 per cent.
Menzies, now leading the new Liberal Party, was responsible for one of the more spectacular losses on the Coalition side with his 1951 referendum to ban the Communist Party of Australia. The Australian Gallup Poll found 73 per cent of Australians intended to vote “Yes” one month out from the referendum, while only 17 per cent intended to vote “No” with the remainder undecided. The “Yes” vote had collapsed by polling day, and Australians rejected the referendum with a “No” vote of 50.6 per cent, and in three of the six states, including Menzies’ Victoria.
Peter Dutton can clutch at these straws, if he wishes, as examples of oppositions in the political wilderness that foiled a mighty incumbent’s cultural agenda. But in both cases the government of the day easily won the next federal election. Menzies managed to turn his setback into a two-decade scare campaign against Labor as being soft on communism. The warning here for Dutton, and conservatives more generally, is that Labor, the teals and the Greens can play a version of this card whatever happens to the Voice. The perception that the Liberals are soft on racism will not help them win back multicultural electorates in the middle and outer suburbs of Melbourne and Sydney.
When Menzies led the Coalition into office in 1949, at just the second federal election the modern Liberal Party contested, the Australian population was split 50/50 between the capitals and the regions. Today, two thirds live in the capitals.
The past is not on the side of the Voice. Yet the Constitution as it stands, and which the Liberal and National parties are defending, will remain dead-letter law in a cultural sense if the Voice is defeated. The document will speak to who we were, without the effective power to stop who we are becoming. Australia has already crossed the twin thresholds of identity that the founding fathers of federation had hoped to avoid by legislating for a “White Australia”: we are a majority migration nation now with a Eurasian face and a booming Indigenous population.
Imagine the Australian people represented in a family tree, with First Australian (Aboriginal and Torres Strait Islander) roots, an Old Australian trunk comprising non-Indigenous people who were born in this country as were their parents and grandparents, and New Australian branches comprising migrants and their locally born children. At the last census, conducted in 2021, the New Australian branches extended to almost 51 per cent of the population, the Old Australian trunk comprised just over 45 per cent, and the First Australian roots were almost 4 per cent.
The most vibrant part of our family tree are the roots: they grew by 48 per cent, or around 275,000 people, between the 2011 and 2021 census. First Australians now number almost one million – a figure that puts them above the English-born population in Australia for the first time since the 1820s.
The New Australian branches expanded by 31 per cent, or around three million people over the same period, to 12.2 million. The rub for a Coalition trying to take back the country from outside the big cities is that the Old Australian trunk is approaching the point of demographic no return, when deaths and overseas departures exceed births. Sydney is the first capital to reach this milestone – its Old Australian population fell by 7000, or 0.4 per cent in absolute terms, over the past decade. Nationally, the trunk increased by just 5 per cent, or around 600,000 people, to 11 million.
Old Australia commanded around 55 per cent of population when the republic was rejected in 1999 by a national vote of 54.9 per cent to 45.1 per cent, and in every state. New Australia had 43 per cent and First Australia just over 2 per cent. I am not suggesting that white Australians voted to retain the monarchy, while the republic was the losing cause of migrant and Indigenous Australia. The result was more complex than that.
Three-quarters of the electorate were in favour of a republic, based on a comprehensive exit poll undertaken by researchers at the Australian National University. What split the republican vote, and cruelled the referendum, was the method of choosing the president. A majority of all Australians – 55 per cent – wanted a president directly elected by the people. Only 21 per cent preferred the parliamentary appointment model they were being asked to vote on. Retaining the monarchy with Queen Elizabeth II as Australia’s head of state and the governor-general as her representative was supported by just 24 per cent of voters. In other words, republicans delivered more than half of the “No” vote.
The republic referendum answered a question that the political system didn’t realise it was asking, about the cultural divides in Australia as we approached our centenary of federation. The most prominent political rift it revealed was between Melbourne and Queensland. Seventeen of the 20 electorates in metropolitan Melbourne voted “Yes” compared to just two of the 27 in Queensland. The inner-city electorate of Melbourne recorded the highest “Yes” vote in the nation – 70.9 per cent. The lowest was in the southern Queensland outback electorate of Maranoa, where just 22.8 per cent of voters wanted the republic on offer. Melburnians of a certain age would know the monarchy’s heartland from its most famous racehorse, Gunsynd, “the Goondiwindi Grey”. Today, Maranoa is the electorate of National Party leader David Littleproud, the first one out of the gate to oppose the Voice.
Canberra was the nation’s republican capital in 1999, with 63.3 per cent voting “Yes” followed by Melbourne with 57.1 per cent, Sydney 52.6 per cent and Hobart 52.4 per cent. Adelaide and even Brisbane both recorded city-wide “Yes” votes just above 50 per cent. That left Perth as the sole capital to vote against the republic. The “No” vote here was 52.5 per cent. The “No” vote outside the capitals was large enough to ensure that the referendum was defeated in all six states.
There was an intriguing contrast between Melbourne and Sydney in the location of their respective monarchist electorates. Melbourne’s were minor outposts on the city’s southern boundary and the south-eastern end of Port Phillip, in the safe Labor seats of Holt and Isaacs, and the then Liberal seat of Dunkley. Sydney’s monarchist seats loomed like an invading army, encircling the cosmopolitan suburbs from Labor’s working class and multicultural west and the deep white south of the Sutherland Shire, where the electorate of Cook awaited Scott Morrison’s entry into parliament.
Melbourne was the more socially cohesive city with little difference between the strong “Yes” votes in the city’s eastern and western suburbs. Sydney fractured along the lines of class. The “No” voters in the west and south were united in their antipathy of the north and east. It meant that New Australians in Melbourne and Sydney found themselves on opposite sides of the great identity contest of the 1990s between Paul Keating and John Howard.
The dichotomy is best seen in the diverse seats that Bob Hawke and Gough Whitlam once held. In Hawke’s former seat of Wills, in Melbourne’s old manufacturing north, the “Yes” vote was 58.7 per cent. In Whitlam’s former seat of Werriwa, in Sydney’s outer south-west, the “No” vote was 58.1 per cent. Well may we say that Gough’s electorate saved the Queen.
Werriwa happened to have a larger overseas-born population than Wills in 1999. The two split once again on the question of marriage equality in 2017; the “No” vote in Werriwa was 63.7 per cent and the “Yes” vote in Wills was 70 per cent.
Referendums are elections about our system of government, not who forms government. An idea appears on the ballot paper without a candidate or political party attached to it. Voters can choose to engage with the idea, carefully weighing the risk and return of changing the foundational document of our federation. Or they can vote tribally and demonstrate their identity, as cosmopolitan or parochial.
This is what the Voice ultimately has in common with the republic: the “No” vote is expected to rise with age, and with distance from the city centre. And the groups most likely to vote “Yes”, regardless of age or location, are First Australians and New Australians. Importantly for the “Yes” case, Sydney’s west is likely to be more aligned with Melbourne’s migrant heartlands in the north and west in 2023 than it was in 1999.
The roots and branches of our family tree now account for almost 55 per cent of the national population: 56 per cent in Victoria, 57 per cent in NSW and 63 per cent in Western Australia. But Old Australia retains a majority in South Australia (51 per cent), Dutton’s Queensland (53 per cent) and Tasmania (65 per cent).
There is a path for a successful “Yes” campaign through the open doors of diversity in Melbourne and Sydney, which could carry Victoria and NSW over the line and tip the national vote above 50 per cent. South Australia and Tasmania would prove pivotal in that scenario, as Queensland and Western Australia are expected to vote “No”. But there is no precedent for a narrow “Yes” win. In seven of the eight successful referendums, all six states voted “Yes”; in the other, in 1910, five of the six states voted “Yes”.
Dutton has ensured a divided result because no referendum has succeeded without bipartisan support. What he can’t stop, however, is the long march of diversity. The danger for his constituents in the regions is a “No” vote will deny First Australians the right to tell a unifying story on behalf of Old and New Australia. Elections are increasingly decided in the cities, and New Australia won’t have the incentive to engage with Old Australia as its majority grows. Why bother if Old Australia still treats national identity as a game of winner takes all?
The political cliché, that demography is destiny, is misplaced. Political parties do not need to be overwhelmed by the rise of voters beyond their tribe. They can choose to embrace them by engaging with them as fellow Australians. Otherwise, demography does indeed become fate. In choosing to campaign against the Voice on behalf of a shrinking part of the electorate, Dutton risks turning the party of Menzies, which governed Australia for 51 of the past 74 years, into a protest party of permanent opposition.
There is nowhere quite like The Monthly. We are told that we live in a time of diminished attention spans; a time where the 24-hour-news-cycle has produced a collective desire for hot takes and brief summaries of the news and ideas that effect us. But we don’t believe it. The need for considered, reflective, long-form journalism has never been greater, and for almost 20 years, that’s what The Monthly has offered, from some of our finest writers.
That kind of quality writing costs money, and requires the support of our readers. Your subscription to The Monthly allows us to be the home for the best, most considered, most substantial perspectives on the state of the world. It’s Australia’s only current affairs magazine, an indispensable home for cultural commentary, criticism and reviews, and home to personal and reflective essays that celebrate and elevate our humanity.
The Monthly doesn’t just comment on our culture, our society and our politics: it shapes it. And your subscription makes you part of that.
Select your digital subscription