Election preview: Australia

If you’re not already up to speed, you can get the basics of tomorrow’s Australian election from my first story, four weeks ago, and from this week’s update. Labor starts with a notional 78 seats in a House of Representatives of 150, so it can afford net only two losses before losing its absolute majority. But its opponents, the Liberal-National Party coalition, have only 57, and there are 15 on the crossbench (more about them shortly).

The Coalition’s immediate goal, therefore, is to win 11 seats to get ahead of Labor. But the real target is greater than that, for two reasons. More of its contests than Labor’s are against (actual or potential) crossbench members; if it’s making gains some of them will probably come from the crossbench, so they only count for half as much in terms of getting ahead of Labor. And secondly, most of the crossbench leans towards Labor, so to be in contention to form a government the Coalition, as I said last month, “would need to not just get ahead of Labor, but get well ahead.”

Realistically, a Coalition victory would require it to pick up something like 15 seats. There are certainly that many seats out there in which it rates a chance, but it has been going backwards at a rate of knots during the campaign and most of them now look out of reach. The only gain it would now be confident about is Aston, which it lost to Labor in a 2023 by-election; the betting market still also has it a narrow favorite in Labor-held Gilmore and Greens-held Ryan. Another three or four might be serious possibilities, but that’s about it.

On the other hand, a number of Coalition seats are looking vulnerable, either to Labor or to independents. A YouGov analysis this week predicts five losses (net) in each category, which would take the Coalition down to 47. It also tips Labor to pick up Brisbane from the Greens, which would take it to 84 seats, a majority of 18 against all comers.

That’s very much at the high end of expectations, but the fact that numbers like that are on the radar at all is quite remarkable. As I explained way back in 2010, the most well-established rule of federal politics (for reasons unknown it doesn’t apply at state level) is that first-term governments always lose ground: every one has gone backwards since the two-party system was first established in 1910. That rule could be broken tomorrow.

But the balance between Labor and the Coalition is only half of the story. Understandably, it’s the half that most people focus on; more than anything, we want to know who’s going to form government. In 2022, however, the most dramatic aspect of the election was the rise of the crossbench. After decades in which the major parties almost monopolised the House of Representatives, their hold seemed to have been decisively broken.

Last time 31.7% of voters opted for someone other than Labor or the Coalition, a figure that has risen 17 points in 15 years. But many pundits seemed in denial about the situation, or assumed that things would return to “normal” as a matter of course. There’s no sign of it, though: on the contrary, it seems likely that this time the major party vote will drop below the two-thirds mark. William Bowe’s latest polling aggregate has it at just 66.1%.

The practical impact is obvious, as a larger crossbench makes it more and more difficult for anyone to win a majority. But there’s a psychological impact as well. In particular, the size of the crossbench obscures the plight of the Coalition: because Labor only narrowly won a majority last time it registers in people’s minds as a close election. But in the balance between Labor and Coalition it wasn’t close at all. As a percentage of seats won, the Coalition turned in its worst performance since 1943, before the formation of the Liberal Party.

The non-major-party vote is diverse. Of the current 15 crossbench members, five are Greens or Green-adjacent, eight are centrists (Teals or Teal-adjacent), one belongs to the far right and one is unclassifiable (Dai Le, Fowler). The last two would probably support a Coalition government, and a couple of the centrists might be persuaded, but the rest, when pressed, would almost certainly line up with Labor if it needed them.

Yet the underlying shares of the vote do not reflect this; here the electoral system takes a hand. It’s not just that voters outside the major parties are unrepresented in total – although they are, with their 31.7% yielding only (post-redistribution) ten per cent of the seats – but the ideological skew within that group is quite different. Far-right parties in 2022 won slightly over 12% of the vote, but won only a single seat, and that for a relatively moderate regional outfit, Katter’s Australian Party.

There’s been a fair bit of talk about the far-right vote this time, notably with the additional respectability that the Coalition has given to the neo-fascist party, One Nation. But even there its importance is only framed in terms of the effect its preferences will have. One Nation will not win lower house seats, because single-member districts work against minor parties. Preferential voting mitigates the problem, but it only helps those in the centre (like the Teals and to some extent the Greens), not at the extremes.*

When a centrist force finally reaches a critical mass of support, however, it can clean up under single-member districts. The Teals and Teal-like independents may be starting to reach that point, with eight or ten more Coalition seats now in the firing line. (Kevin Bonham has a handy typology of independents.) If enough of them succeed they could force upon the Coalition parties the reckoning that they so badly need.

More on that tomorrow morning, together with some thoughts about the Senate.

.

* The almost total absence of the far left from the electoral scene illustrates the same point: two far-left parties in 2022 collected 0.3% of the vote between them. Voters at that end of the spectrum have to make do with Labor and the Greens.