Does tactical voting happen in Australia?

61

A common talking point when discussing Australia’s electoral system is that it eliminates the need for tactical voting. This is usually true, but there is a narrow set of cases in which tactical voting can be a relevant factor, and there may be more of these cases in 2022.

When a centrist independent is positioned between Labor and Liberal in a traditionally safe Liberal seat, the question of whether the independent, Labor or Greens makes it to the final two could make a difference as to whether the Liberal holds on. This could come into play in some seats in 2022, with the proliferation of the ‘Voices’ independents.

A recent tweet by Voices for Goldstein caused a stir, with a bunch of replies calling it “disinformation” or patronisingly explaining to them that, under preferential voting, you simply need to order the candidates in order of preference to ensure a candidate receives your vote.

I do think the tweet is a bit simplistic and overstates its case by broadening the scenarios in which this is true to include all “seats with a Voices candidate” (which is quite a few seats), but it may be true in some seats, potentially including Goldstein and Wentworth.

We use a few tools when counting votes in Australia which simplify the preferential system, and in most cases it works well. On election night, we count up the primary votes for each candidate and then divide up those votes between two candidates – usually these candidates are from Labor and the Coalition, but in some cases other candidates make this “notional two-candidate-preferred count”.

In most seats, we don’t need to know any more than this information. If it’s close, we need to see how every seat breaks down, but we need not concern ourselves with a voters’ other preferences.

We only need to worry about the detailed procedures of counting preferential ballots when it’s not clear which candidates make the top two. In very rare cases this is because a candidate may gain preferences from third place to make the top two, but usually it’s a matter of which candidate is in second place on primary votes.

In some cases, two candidates are competing for second place, and whoever achieves that outcome will win the seat on the others’ preferences. The most common cases of these are where Labor and Greens are neck-and-neck for second place. Examples include Balmain in 2011, Ballina, Lismore, Maiwar and Prahran. Some of the better prospects for the Greens to pick up extra House of Representatives seats would require this sort of result: Griffith and Macnamara amongst others.

This dilemma of tactical voting doesn’t particularly apply in these races because Labor and Greens preferences are mostly symmetrical – Labor voters and Greens voters tend to favour the other party strongly over the Liberal Party, such that if one is in a winning position on a two-candidate-preferred count, the other will be too. So you can just vote your personal preference. If you favour Labor over Greens, you can help Labor win, but if they lose your vote will elect the Green, and vice versa. Traditionally Labor preferences usually flow slightly stronger to the Greens than vice versa, but usually the preference flows are similar enough so that both candidates are in with a chance, so you shouldn’t consider casting a tactical vote against your preferred candidate.

But when the preference flows are asymmetrical, you can have a situation where two candidates are in a close contest for second place: one of them will be in an election-winning position, and the other will not.

This is more common under compulsory preferential voting than under optional preferential voting, where the primary vote position matters more as less preferences flow.

The most common situation where this has emerged in recent years, and the scenario likely to occur in 2022, involves a centrist independent challenging in a traditionally safe Coalition seat. This centrist independent can appeal to Labor and Greens voters (these seats often have a relatively high Greens vote) to switch to the independent in the hope of flipping a seat where Labor and the Greens are seen to have no chance. This independent is also able to peel away a few Coalition voters, enough to win the seat.

There were four electorates in 2019 which demonstrate this asymmetry in preference flows: Indi, Mayo, Warringah and Wentworth. The Mayo case involves a Centre Alliance candidate, but she generally resembles an independent.

The AEC conducts both a two-candidate-preferred count and a two-party-preferred count where the final two candidates are not Labor and Coalition, which allows us to compare these counts in these four electorates in 2019. In each seat, the Liberal Party made both counts, so this table shows the Liberal percentage in each count:

Electorate LIB 2CP LIB 2PP Gap
Indi 48.61 62.73 14.12
Mayo 44.86 52.54 7.68
Warringah 42.76 52.12 9.36
Wentworth 51.31 59.85 8.54

There are differences between these seats (the Liberals won Wentworth, for a start), but this pattern is similar. The independent outperformed Labor in a head-to-head race against the Liberal Party by 7.7%-14.1%.

Now this particular strategic voting dilemma wouldn’t apply in these seats in 2019 because the independent (or CA candidate) polled a primary vote far above any other candidate apart from the Liberal. There’s no doubt who made the top two. But in cases where this independent is closer to the pack, it could come into play.

Kevin Bonham wrote an excellent blog post on this topic in 2018 which listed a number of similar contests where the second- and third-placed candidates were much closer. This includes the 2018 Wagga Wagga by-election, the 2009 Frome by-election, the state seat of Rockhampton in 2017 and the federal seat of Denison in 2010.

There was a lot of discussion about this in the final weeks of the 2018 Wentworth by-election, when the (admittedly inconsistent) seat polling suggested independent Kerryn Phelps and Labor’s Tim Murray were neck-and-neck for second place, but Phelps was in a much stronger position to win if she outpolled Murray than the reverse. In the end Phelps performed better relative to Murray than the polls suggested – which could have been due to Labor voters casting a strategic vote for Phelps, or just that the polls weren’t that accurate.

While this scenario can come into play, I think it would be easy to try and apply this argument where it doesn’t apply.

I don’t think it’s worth considering a strategic vote in any of the following scenarios:

  • If Labor or the Greens are actually competitive in a seat, I’d recommend casting a sincere vote for your preferred candidate, rather than helping a less-preferred candidate defeat your preferred candidate. I think this probably applies in a seat like Kooyong. The independent there could be strong enough to win, but so could Labor or the Greens.
  • If the independent is not strong enough to make the final count or to win the two-candidate-preferred vote. There are a lot of Voices-style candidates running all over the country and they won’t all do as well as Zali Steggall. I have particularly low expectations of those who are running in conventional Labor-Liberal marginal seats such as Casey and Boothby. Labor may well run dead to try to avoid this issue in seats where they have no chance, but that won’t happen in seats where Labor has a good chance.
  • The independent needs to have a political positioning that makes them likely to do better on preference flows than Labor or the Greens. A far-left independent isn’t likely to have advantageous preference flows that would justify a strategic vote.
  • A strategic vote doesn’t make sense if the independent isn’t so strong that their position in the top two is in no doubt. Established independents will usually convert second preferences into primary votes. You might as well cast a sincere vote in those races.

I suspect there will only be a small number of seats where this scenario plays out in 2022. But in an electorate with a relatively strong independent, where Labor and the Greens are not strong enough to win but in with a chance of making the top two, voters who don’t support the Coalition may want to consider these factors when deciding how they vote.

Liked it? Take a second to support the Tally Room on Patreon!
Become a patron at Patreon!

61 COMMENTS

  1. The question of tactical voting only arises in single member electorates. A list system such as nz using pr balances this.For some one of the right to vote Labor to try and control the ordering or visa versa is hard to do and risks a backfire.

  2. The NZ electoral system does have tactical voting. Both in individual electorates which affect the party’s total seat count (such as the Maori Party seats, or where ACT was running to get a lifeline into parliament), but also if a party is on the verge of falling below the 5% threshold.

  3. Don’t see the point in finding the ‘correct’ explanation of a deliberately ambiguous promise like Kylea Tink’s. She’ll interpret her ‘mandate’ for kingmaking however she feels like. Barring a formal coalition agreement she’ll vote in parliament however she feels like. If she wanted to commit to determining it in any particular manner, she would have made it clear exactly how she’ll do it.

    Always funny to me how naive some presumed political junkies can be around here.

  4. Well I think there’s some cross purposes here, FL. High Street has made several comments on the theme that the Voices independents will be less reliable on progressive issues – and support of progressive governments – than their campaigning would like progressive voters to believe. You may consider this obvious, but it’s clear to me that it’s a response to an attitude that tactically voting for such an independent is a moral imperative and the idea that these candidates are some progressive ideal that just happens to be able to win safe Liberal seats (never mind how they might do that). When they literally come out and say that they will support a government based on how their electorate voted, I believe it is quite legitimate to spell out exactly what this means (Liberal, no matter how you slice it: even if she were to use the notional two party preferred figure her whole election pitch is that she can win the two candidate preferred even though a Labor candidate could not). Whether she follows through on this or not is another question, but I think you’re misaiming the naiveite criticism by working that question into the mix. High Street is arguing against a naive assumption that these independents are going to be progressive tabulae rasae by simply pointing out a logical conclusion of this statement of Tink’s.

  5. I was actually thinking more of what Glen and Adda said tbh. I largely agree with HS. I don’t agree that what she said exactly means anything- who knows what criteria she’s using to determine which government her voters want? After all they don’t vote for Albo or Morrison, they vote for a candidate, and she never said anything about the 2PP as far as I can tell. In lieu of an actual explanation, she can use that and any number of other excuses to coronate whichever leader she wants. It’s another reason why I said you should just vote for the candidate you want to represent you and not to worry about the ‘moral imperative’, as you call it.

  6. Thanks everyone for engaging once again.

    Particular thanks to Dryhad – I will need to get you to write my posts in future. And thanks for actually realising I am been banging on about this for a while.

    @Glen – As Dryhad said after you – I have made several comments on this theme. You obviously haven’t seen any of them as you have assumed I hold the exact opposite position to the one I do. But that’s ok – do try and inform yourself of past comments though before diving in with the boots. I can assure you, I FULLY know the polling that exists in North Sydney. It’s why I find the calls for strategic or “”switch” voting so counterproductive – and frankly, deceptive. The point is the ALP PV vote won’t be 23% if people switch for tactical reasons – but some will so the ALP ran a poor campaign/had a bad candidate/ got swamped by $$, other will say it was because of “strategic” voting – no-one will know which is true. And finally, I’ve been quoting the 2PP results in non-traditional seat on this blog for months.

    @FL – thanks for the support – largely. If anyone thinks Tink, or any IND, is going to support a party to form government that got 40%, because the lower party won the TPP say 51 – 49, then as Darryl Kerrigan would say 25 years ago – tell ’em their dreaming.

  7. That last comment got garbled at the when it was posted – last para should have said:

    If anyone thinks Tink, or any IND, is going to support a party to form government that got 40%, simply because the lower PV party won the TPP (that only those on this blog would know exists) say, 51 – 49, then as Darryl Kerrigan would say 25 years ago – tell ’em their dreaming.

  8. I fully agree that she may interpret the vote in any way she wishes if she is claiming a mandate. That’s why I said that one would have to trust her intellectual honesty – which, in the end, is not much different to the question of how much one trusts her in general on campaign issues.

  9. Assuming the recent ACT Senate polling is reasonably accurate (unlikely imho), do you think that the second Senate race could be an instance where tactical voting is rational? E.g. should someone who prefers the Greens put Pocock first on the assumption that most Greens preferences would flow to Pocock ahead of Seselja but not the reverse?

Comments are closed.