The increasing number of credible campaigns from independents in the federal election is making for some intriguing contests, and in some seats may raise questions for voters about where to place their preferences, as well as their primary vote.
The role of preferences in Australia’s election may at times seem complicated, but what voters need to consider is really quite simple (in the lower house, at least).
On election night, votes are tallied up according to the first preference, or primary vote. Once all the votes have been counted, preferences are distributed. The candidate with the fewest first-preference votes is eliminated, and their votes are distributed to the next preference, and so on until only two candidates remain.
In most cases it is clear on election night who those two candidates will be, so a special count is conducted just between these two candidates, without having to wait for all the votes to be counted before a full distribution of preferences.
But in some races the identity of the top two candidates is not clear – in particular where there is a close contest for second place. These races are becoming more common as more independent and minor party candidates mount serious challenges. They are likely to cause delays in calling races, as some seats will require a full count before we know which candidates will make the final two.
In some seats, the order of elimination can determine whether the second-placed candidate gains enough preferences to catch the leading candidate, and thus can change the ultimate outcome.
This sometimes happens when Labor and the Greens are competing for second place and attempting to chase down a leading Coalition candidate. Labor voters tend to give their preferences to a Greens candidate, and Greens voters to Labor, at very high and even rates. At the 2019 election Labor polled 25.4% of primary votes in the Victorian seat of Higgins, to the Greens’ 22.5%. Similarly in the seat of Brisbane, Labor won 24.5% to the Greens’ 22.4%. In this kind of seat, whichever party comes second will normally have a roughly equal chance of beating the Coalition candidate (in both cases they failed to do so in 2019).
But there are also races where the preference flows are not symmetrical. This is most common when there is an independent competing with a Labor candidate, both vying to make it to the final count against a Coalition candidate.
A recent example was in the seat of Warringah in 2019, where the independent candidate Zali Steggall won 57.2% of the two-candidate-preferred count against Tony Abbott. The Australian Electoral Commission also conducted a count between the Liberal and Labor candidates, with the Labor candidate receiving just 47.9%. That means 9.3% of voters in Warringah preferred Steggall over Abbott, but preferred Abbott over Labor.
In this case it didn’t matter, because Steggall had no trouble making the final two (in fact she topped the primary vote count), but in a closer race those voters could make a difference.
The charts below show a simplified example. There are three candidates, and the vote splits 41%-30%-29%. Candidate 1 is the clear leader, but the other two candidates’ preferences tend to favour each other over Candidate 1.
In the first case, Candidate 2 comes second, and Candidate 3’s preferences decide the outcome. While Candidate 3 voters do favour Candidate 2, they don’t do so overwhelmingly, and Candidate 1 wins narrowly.
But in the second case, Candidate 3 comes second, and Candidate 2’s preferences flow overwhelmingly towards them, allowing them to narrowly defeat Candidate 1.
If you knew this was how the primary votes would fall, and your goal was to defeat Candidate 1, you would want to vote for Candidate 3 in the first instance. But it’s also worth noting that in the first scenario, Candidate 2 comes within 2% of winning. They are not an unelectable candidate, and might have won in slightly different circumstances. If Candidate 2 is the one you really want to win, do you sacrifice their chance to try to ensure defeat for your least favoured candidate?
Some have called for voters to consider casting a strategic vote in this way.
I think voters should instead stick to a principled vote: number your ballot paper in the order of your genuine preference.
To continue with the example of two chasing candidates competing to defeat a leading candidate: there will be seats where one candidate is well clear of the other and isn’t troubled by the order of elimination (like Warringah in 2019). There will be other seats where both candidates are strong enough to beat the leader on preferences, so a vote for either will achieve that outcome. There may be races where the two candidates are in a close race for second place and only one will receive a strong enough flow of preferences to go on to win. And there may even be seats where there are more than three plausible contenders for the final two spots (Hughes in Sydney may be among those at this election).
How do you know which scenario may play out in your seat? You don’t, I don’t. I could make a list of seats that I think fall into each category, but I’m sure I would get some wrong. None of us have a crystal ball, and seat-level opinion polling is not accurate enough to be confident. I expect many people who previously voted Labor or Greens in conservative electorates will switch and vote for a viable independent, either because they prefer that candidate or would like to vote for someone they think can win. But voters should not feel pressured to switch from their preferred candidate because of a possible outcome which is quite unlikely, particularly if their preferred candidate has a genuine chance of winning.
We are lucky to live in a country where voters don’t have to weigh up which candidates are likely to win when deciding how they vote, as they do in the UK. Unintended outcomes can happen, but are rare enough and unpredictable enough not to change the advice I would give every voter: number the boxes in the order that you genuinely prefer.