The Voice Referendum: Australia’s Brexit?
Sign up here to receive our regular Election Note! Read the full analysis and receive the full version of each note as soon as it is published.
Published 12 October 2023
Key takeaways
The ‘No’ campaign looks set for a comprehensive victory. With just days to go until the referendum, ‘No’ is the strong favourite in the Voice referendum
The voting coalitions on both sides resemble the two Brexit camps. Generational, educational and geographic fault lines divide the Australian electorate in a way that is similar to the Leave/Remain camps of the 2016 referendum
Nevertheless, it is unlikely that the Voice will trigger ‘Brexit-style’ political realignment. Realignment is an ongoing trend in Australian politics. The Voice can contribute to it, but it is unlikely that the referendum is salient enough to have a lasting effect
What is the state of play?
Figure 1: Trends of voting intention polls of the Voice referendum
On October 14th, Australians will head to the ballot box to vote in a historic ‘Indigenous Voice’ referendum. The proposed change would create a constitutional mechanism for Indigenous people to advise the Australian parliament on policies that specifically affect their communities. This new advisory body would include Indigenous Australians from across the country, with members elected by local Indigenous people.
So far, the referendum has divided the nation’s dinner tables. Supporters have argued that this would be an important step towards the recognition of the historical suffering of Indigenous communities, while also providing this group with greater self-determination in the future. Indigenous people make up nearly 4% of the Australian population and are a strongly disadvantaged group suffering from higher rates of suicide, domestic violence, alcohol abuse, and imprisonment than other Australians. By enshrining Indigenous rights in the constitution, supporters hope to create a better way to safeguard Indigenous rights in government policy-making.
Opponents, however, have held that the plans for this advisory mechanism are too vague, leading to legal uncertainty. Some have argued that this body would become too powerful and act as a ‘Third Chamber’ of the Australian government. Others have opposed on the exact opposite grounds, claiming that the current proposal does not go far enough. Specifically in Liberal circles, opponents have also made the case that creating special constitutional provisions for Indigenous communities risks embedding narratives of ‘victimhood’ in Australian policy-making, encroaching on the principle of equality.
‘No’ is the strong favourite going into the referendum. As Figure 1 illustrates, after an early lead, the ‘Yes’ campaign is now estimated to win just 42% of the national vote. The ‘No’ camp, currently at 58%, holds a comfortable lead of 16 percentage points. This is well outside the pollsters’ margin of error.
To make matters worse for the ‘Yes’ campaign, history is not on their side. Just 8 out of Australia’s 44 referendums have succeeded. To be successful, the ‘Yes’ camp in the referendum needs the support of a majority of Australians and a majority in 4 out of Australia’s 6 states (Northern Territory and the Australian Capital Territory are not states and are therefore not included). Past campaigns have struggled to meet that threshold. With current polling numbers, it is difficult to see how this one can.
How Brexit-y is the Voice referendum?
A divisive referendum, rifts in both major parties, and a campaign characterised by uncertainty about what a ‘Yes’ would mean in practice. The parallels between the ‘Voice’ and the UK Brexit referendum are striking. It’s no surprise, then, that the list of media articles comparing the two is endless (see, for example, here, here, and here). But will the ‘Voice’ realign Australian politics in the same way?
Brexit forced realignment because it put the official Conservative and Labour party positions at odds with a substantial proportion of their voters. We do not see splits to the same extent in the Voice referendum. Nationally, our polling shows ‘No’ winning the referendum with 55% of the vote. Just 21% of Liberal-National Coalition voters are intending to vote ‘Yes’, with the other 79% following the party line of voting ‘No’. 61% of past Labor voters are planning to stick with the party line and vote ‘Yes’ in the referendum, with the other 39% saying they will vote ‘No’ (Figure 2).
Figure 2: Stacked column chart of vote intention in the referendum, grouped by voting behaviour in the 2022 Australian federal elections
So Australian Labor is in a similar bind to UK Labour, with a large minority in the party disagreeing with the party line. The official Liberal position (‘No’), however, is only at odds with a relatively small minority within the party. Labor may face problems in holding onto their ‘No’ voters, but the Liberal vote is much more unified.
Zooming in on past Labor voters, we do find similarities between the Voice and Brexit. Evidence from our polling suggests that the “No” campaign draws from the older and lower educated parts of the Labor vote (see Figure 3). We also find that Labor voters in major cities (with the exception of Perth) are less likely to vote ‘No’ than Labor voters in rural areas. These findings appear to echo Leave-Remain lines in the Labour camp during the Brexit referendum, when older, lower educated, and voters outside of major cities were more likely to vote Leave.
By contrast, Figure 4 illustrates the factors that increase the probability of a Liberal voter intending to vote ‘Yes’. Potential Liberal ‘Yes’ voters are more likely to be young and hold a degree. Again, these generational and educational fault lines are something that the Voice and Brexit referendums have in common.
Figure 3: Coefficient plot of a regression model showing the factors that make 2022 Labor voters more likely to vote ‘No’
Figure 4: Coefficient plot of a regression model showing the factors that make 2022 Liberal voters more likely to vote ‘Yes’
Where is the Australian Red Wall?
Figure 5: Map of Australian constituencies by 2022 two-party preference and projected ‘Yes’ vs ‘No’
The impact of Brexit on British politics was not limited to the 2016 referendum itself. The vote created rifts in the major parties, leading to a (possibly temporary) realignment of British voters and their parties that became visible during the 2019 election. Leave voting ex-Labour supporters jumped ship to join a more socially conservative Tory party under Johnson, a trend most visible in the notorious Red Wall. This cluster of industrial Labour strongholds falling to the Conservatives was the game-changer of the 2019 general election. Will we see a similar trend in Australian politics?
Probably not. There are definitely some geographic similarities. Figure 5 categorises all Australian constituencies by which major party secured the most votes in 2022 and by predicted referendum results based on our polling data. Due to Australian geography, we do not see one clear ‘Red Wall’. Instead, the Labor seats that are leaning towards the ‘No’ camp are concentrated primarily in the outer suburbs of major population centres. In New South Wales, for example, the ‘Yes’ camp would hold its ground in Sydney, but would see the blue-collar seats of Hunter, Werriwa, and Fowler all vote ‘No’. Similarly, in the Melbourne outer ring, Gorton, Holt and Lalor are all traditional Labor seats with higher than average levels of inhabitants working in manual trades. Our poll projects all three as ‘No’ wins.
To draw out this point in greater detail, Figure 6 plots all Australian SA1’s (a low-level statistical geography) by our modelled estimate of the share of ‘No’ voters in that area (y-axis) and the proportion of constituents who share certain key demographics (x-axis). This analysis shows a clear trend for the ‘No’ vote to be higher in areas where a larger share of the population have long-term health conditions (often a proxy for deprivation). Conversely, ‘No’ wins fewer votes in areas where large shares of the population hold degrees. Like the Leave campaign in the UK in 2016, the ‘No’ camp appears significantly more successful in more working-class areas. This could help to drive a wedge through the Labor coalition, as the Brexit vote did for the UK Labour Party, dividing well-educated metropolitan voters from traditional working-class heartlands.
Figure 6: Scatterplot of SA1, a low-level Australian statistical geography, by projected ‘No’ vote and proportion of constituents who have long-term health conditions and who hold a Bachelor's degree or higher qualification
The Voice referendum definitely resembles the Brexit referendum to some extent. However, this does not mean that the Voice will reshape Australian politics in the way that Brexit hit UK politics. There are three main reasons for this.
The Voice referendum has not unearthed a ‘new’ fault line in Australian politics. In fact, the coalitions backing either side resemble the two camps on any ‘culture war’ issue: an urban, higher-educated segment with progressive values and a more suburban or rural, lower-educated camp with more conservative views. Whilst perhaps a new ‘type’ of political conflict in the UK in 2016, culture war issues have become increasingly common across the Western world, and in Australia. A recent example is the transgender rights debate in the run-up to the 2022 Australian federal election. With culture war issues already occupying a spot on the national political stage, the introductionof a new cultural issue, like the ‘Voice referendum’, will have less of an impact on how voters perceive their parties.
Relatedly, a realignment can only occur when an issue splits the existing bases of political parties. When a party’s voters hold contrasting views on this new issue, some of them may start looking for a new political home. In the case of the Voice, however, it is only Labor that is split. The Liberal coalition only sees +-20% of their voters leaning towards ‘Yes’, a proportion too small for realignment on a Brexit scale.
Finally, it does not appear that the Voice referendum is as salient to the Australian public in general as Brexit was to the British electorate. Our polling suggests that “the Voice Referendum” is a top priority for only 2% of the Australian public while Yougov tracking polling indicated 19% felt “Britain leaving the EU” was the most important issue facing the UK in June 2016. Naturally, it is extremely important to Australia’s Indigenous population in particular. However, for other Australians, it is unlikely to be as fundamental as Brexit, which entailed a complete renegotiation of the UK’s economic and political position on the world stage. This would suggest that, even if the Voice pushes some voters away from their parties, it is unlikely to have as lasting an effect as Brexit.
The political landscape in Australia also limits the extent to which the Voice Referendum can trigger a Brexit-like realignment. In the UK the political upheaval following Brexit really came to fruition under a popular, and populist, leader on the Right in the shape of Boris Johnson. For the realignment to take hold it needed not only the “push” of the divisive referendum but also the “pull” of a charismatic leader, who was popular among disaffected Labour voters. In Australia however, the roles are somewhat reversed. Anthony Albanese rode to victory championing a relatively populist Left-wing movement and despite a slump in approval ratings remains significantly more popular than Peter Dutton, the leader of the Liberal party. Furthermore, Dutton’s personal favourability ratings are particularly poor among Labor voters, according to our polling only 6% have a favourable view of him. In the absence of a significant “pull” factor on the Right and with a relatively popular leader, it is possible that Labor may yet paper over any cracks in their coalition caused by the referendum.
We are also keeping an eye on:
Polish parliamentary elections - Polls are narrowing in the most important EU election of the year. A recent poll now even predicted an opposition victory. Read our previous election note here to learn more about how polling error in Poland could affect the final results. Voting 15th October.
Argentinian presidential elections - With high inflation, government scandals, and low approval ratings, the incumbent centre-left government faces an uphill battle in this year’s autumn elections. Our own polling suggests libertarian Javier Milei could win, whilst public polls suggest a neck-and-neck race between him, centre-left candidate Massa, and the established opposition's alternative Bullrich. First round 22nd October.
Dutch parliamentary elections - Without long-serving Prime Minister Mark Rutte, the November elections could cause major turmoil in the otherwise relatively stable politics of the Netherlands. With a united left, two new protest parties and a strong far-right flank, the race is chaotic, unpredictable, and - for the political geeks among us - incredibly fascinating. Voting 22nd November.