Bad news for government in the pre-polls

Time for another update on Australia’s ballooning pre-poll voting. Yesterday almost another 415,000 were taken, the biggest daily total yet, bringing the total to just over three million: 3,003,473. (You can download all the figures here.)

With three days of pre-polling left, that’s already very close to the 2016 total.* Nonetheless, the rate of increase is definitely slowing. I produced a table last Friday; here’s how it continues for the last few days:

A week ago, it seemed possible that the total would pass five million; now that looks unlikely, although it should still reach four and a half.

That suggests that this is not just a matter of a general trend towards increased pre-poll voting (although that’s certainly a large part of it); there’s also been a shift involving people voting earlier within the pre-poll period, presumably because they’ve made up their minds and want to get it over with.

And that’s a bad sign for the Coalition government, since on all accounts it was starting from behind and depending on being able to swing around some support during the campaign.

Although the total pre-poll is no longer anything like “almost double the number that voted at the same time” in 2016 (as the ABC inexplicably reported this morning), it’s still an awful lot. Some pre-poll centres have taken huge numbers of votes: I count eight with more than 20,000 each. (Five of those are in Victoria, which is probably another bad sign for the government.)

Many centres are now running at more than 1,500 votes a day, and new centres are still being opened. The Ponds, in Sydney’s western suburbs, has taken nearly six thousand votes in the three days that it’s been open. Cooranbong, in the Hunter Valley, has taken nearly three thousand in just two days.

Last week I talked about concerns that many people have with the rise in pre-poll voting, as to whether some important element of democracy is being lost. But there’s also a big practical problem.

Come Saturday night, some centres will be in the position of having to count 30 or 40 thousand votes; a large number will be counting ten thousand or more. That’s a very big task – compare with a normal polling place, which rarely has more than about four thousand.

The electoral commission has promised it will all be done on the night. But even if it is, it will be very late on the night, and because the numbers are both so big and so different from last time, it will be very difficult to get clear results without them.

Antony Green, who knows more about the subject than pretty much anyone, was today warning of delays but nonetheless expressing confidence overall: “we should know the result on the night unless it is very close.”

“On the night”, however, could easily be past midnight, so it’s very possible that a lot of Australians will go to bed with a greater or lesser degree of uncertainty about the result – even if, when all is done, it isn’t particularly close. As William Bowe warned last week:

The implications of this can go beyond the simple matter of the theatre of election night. Whether one leader or another is decisively able to claim victory before the networks close shop can have a material impact on the public’s perception of the strength of their mandate and authority in office.

That fear may be overblown, or it may be that this election will be sufficiently decisive that the question will be academic. But it’s another implication of greater pre-poll voting that at some point will have to be looked at.

* Which raises the surprisingly difficult question of just what the 2016 total pre-poll was. On the spreadsheet available at the electoral commission the final figure is 2,980,498, but that’s clearly incomplete; many centres have no return for the final day, and some are missing the previous day as well. Adding up the totals from the published results (see here, under “Two candidate preferred by candidate by polling place”) I get 3,082,505, but that doesn’t include informals, of which there should be rather more than 100,000. So a round figure of 3.2 million is probably pretty close: that will easily be passed today. (After I had written that I noticed that the electoral commissioner cites the same number, although he doesn’t sound very certain about it.)