WITH AN EYE to how any movement against Campbell Newman’s government might translate into seats, we look at the election pendulum based on results from the state election in 2012; even with early polls showing a slight bounce in support for the LNP, it still faces a swing to Labor in the order of 10%: enough to cost it office depending on where the votes fall, and certainly enough to terminate the Premier’s tenure in his seat of Ashgrove.
As we discussed shortly after Premier Campbell Newman called a state election for 31 January earlier this week, there is a school of though that the snap announcement — made amid a lot of noise about an overseas trade mission (that he hasn’t left for) and timed to coincide with the silly season and school holidays — was scheduled to get to the polls as quickly as possible, lest the LNP vote continued to erode the longer an election date was delayed.
In the days since, of course, we have seen the first batch of opinion polls: all — Newspoll, Galaxy, and ReachTel — showing a bounce in LNP support; all had previously found the LNP behind Labor after preferences, and these new findings ranged from a 50-50 result (ReachTel) to a 53-47 advantage for the LNP (Newspoll).
Yet even a 53% two-party vote at an election for the LNP effectively represents a 10% swing to Labor, coming as it does from the 62.8% recorded at the election in March 2012: less than the (then) Coalition received in 1995, when it fell two seats short of a majority on election day, and a low enough tally to send it into minority or outright opposition depending on how the distribution of votes falls when they are cast.
I want to talk about a 10% swing to the ALP, because I think it’s a very realistic prospect when considering the outcome of this election, even with a few weeks to go; we will make quite a bit of reference to the Malcolm Mackerras election pendulum that appeared in The Weekend Australian yesterday — you can access it here — but before we do so, I should like to make a couple of broad observations at the outset.
The first is that I would expect a swing of 5% against the LNP irrespective of how it was travelling, what it has done (or not done, depending on your position) or the merits or otherwise of the campaign it now fights. Why? For the simple (and obvious) reason of its 2012 win, which was the biggest election win in Australian political history. There has to be a correction, and whilst it’s an arbitrary call I think 5% is a reasonable (and adequately neutral) figure to put on it.
The second is that whilst it hasn’t exactly set the world on fire, nor presented any alternative to the LNP of substance in any way — how could it, on one level, with less than 10MPs? — Labor in opposition hasn’t been a smoking ruin or an apocalyptic, infighting disaster, either: and whilst the Queensland ALP seems content to merely coast back into office on the back of grotty and intellectually fatuous tactics like its counterparts everywhere else in Australia, it hasn’t done anything over the past three years that would lead to it failing to pick up additional seats, or preclude it from making steep gains in the right circumstances.
(For those who haven’t seen the kind of thing Queensland Labor operatives have deemed fit to flood social media with this week, click here — it gives pause for thought when it comes to any consideration of what the ALP’s agenda for government might be if it were to actually win, but I digress).
The third, of course, is the obvious fact that Newman’s government is not popular; for all the good it has done in either delivering the outcomes it promised or making solid progress toward them, from a purely political perspective it has been a flat-footed government at best, seemingly determined to mishandle what it touches, almost going out of its way to create controversy and self-inflicted crises that are more than enough to weary the patience of the electorate.
And it continues to go without saying that short of a horrible, unforeseen scandal erupting around his Labor opponent, Newman remains virtually certain to lose his seat of Ashgrove: it probably grates on LNP hardheads to hear me say that, bent as they are on tying the government’s fortunes to Newman’s. But whilst I would heartily prefer to see both Newman and the government re-elected, I see no point in deny the bleeding obvious.
Mackerras — whose article on the same subject can be read here — has forecast a two-party split of 54-46 in favour of the LNP, and predicted it will be re-elected with 55 of Queensland’s 89 seats to Labor’s 29 (the remaining five to go to Independents and minor party candidates).
The one observation I want to make from the slew of early opinion polling (and no, we’re not going to pick them apart, even if they are favourable to the LNP) is that whilst the Palmer United Party vote appears to have collapsed — finally — and whilst much of this support seems to have transferred to the LNP, it hasn’t been enough to shift Newman’s poll numbers to a position in which the government’s re-election is assured.
Ominously for the LNP, the Labor vote has held up in the high-30s which, provided an exchange of preferences between the ALP and the
Communist Party Greens is tighter than it was three years ago, is enough to cause the government real problems.
I tend to think (for now) that a 53-47 result is nearer the mark than what Mackerras has predicted; partly because of the 5% corrective swing I outlined earlier, and partly because I think the LNP stands to lose more than half the amount of what’s left after that correction to pull its vote back down to the 50% mark.
Subtracting 5% from a 62.8% two-party at the outcome leaves a “margin,” if you like, of 8%. Mackerras is claiming the LNP will lose half of it. I say it will lose a bit more than that. To this end, if there is movement either way from these predictions on polling day at all I expect it to be further towards Labor: thus, an eventual outcome of 52-48 or even 51-49 cannot be discounted.
We will, however, confine ourselves to a 10% swing for today’s purposes — and this takes us across to the Mackerras pendulum.
A uniform swing of 10%, based on that resource, would see the ALP win 26 LNP-held seats in addition to the nine it already holds, for a nominal total of 35 seats. This includes Labor’s by-election gains (Redcliffe and Stafford) but some comment needs to be made on a couple of the seats within the swing range and on Labor’s side of the ledger.
In Gladstone, where popular conservative incumbent Liz Cunningham is retiring, I expect Labor to win this seat; by contrast, the seat of Mt Isa — shown on the LNP side of the pendulum and presently held by Robbie Katter under the banner of his father’s party — is one I think will remain in non-Labor hands (but cannot decide which of Katter or the LNP will win it).
Of the two seats won by Labor at by-elections, both can be expected to swing to Labor based on the 2012 results and I am a little surprised Mackerras has shown them here as Labor seats: his usual practice (with which I agree), on the basis by-elections are often aberrant, is to show the pendulum as it stood at the previous election.
Nonetheless, I expect Labor to be re-elected in Stafford; provided its candidate selection has been astutely made in Redcliffe, however, I rate the LNP an excellent chance to reclaim the seat lost after the grub Scott Driscoll (who should never have been endorsed in the first place) was forced to vacate it.
I expect the LNP to easily retake Gaven from the snivelling, disloyal, self-important lemming who was lured from the party’s fold by the promise of God knows what by Clive Palmer, only to walk away from Palmer’s party (and its “leadership”) on a stand of “principle:” If Alex Douglas polls any more than 2-3% in this seat I’ll be stunned.
Similarly, I expect the ALP to easily win in Yeerongpilly, and for Condamine to be won back by the LNP. Dalrymple — held by the Katter crowd — should be an LNP gain, but in the febrile political climate in Queensland at present I wouldn’t stake my house on it.
I don’t expect the retirement of Tim Mulherin to make any material difference either in his seat of Mackay or to the election outcome; this is — after all — normally a very, very safe Labor seat despite its present margin of 0.6%; by the same token, the Labor Party’s candidate issues in Lytton — a seat that actually held for the ALP in the near-2012 massacre of 1974 — will probably be no bar to it reclaiming that seat in Brisbane’s Bayside, although many of us said similar things about a comparably traditional Liberal electorate (Indooroopilly) that took three terms, a change of party and a sexual harassment scandal to dislodge Ronan Lee after his narrow win in 2001.
Mackerras has published these pendulums for decades, but with an unwavering disclaimer: that he acknowledges swings are rarely (if ever) uniform, but that the pendulum is nevertheless a reliable predictor of the number of seats that change hands at any given election based on a given swing: using the pendulum I have linked to here, let’s say the swing to Labor is 1%: the pendulum shows just one seat (Bulimba) falling to the ALP on such a movement.
But in practice, for the purposes of making the point, the ALP might indeed win a single seat at an election where it achieves a 1% swing; that seat could be further up the pendulum. Or Labor could win three and lose two. Either way, movements outside the swing range cancel out the movements within it, according to Mackerras, to offer a reasonably reliable indication of how many seats might fall even if the swing is not uniform.
On this occasion I disagree with him. The boundaries in Queensland (despite a small weighting in five rural electorates) contain a bias toward the ALP of an indiscernible amount between 2% and 4%; part of this derives from the swag of ultra-safe, mostly former National Party seats in rural Queensland, although after 30 years of exclusively presiding over redistributions in office it is undeniable that Labor routinely won up to 80% of the Brisbane seats, and sometimes with as little as 55% of the two-party vote in the capital.
Taking all of these factors into account — and looking at Mackerras’ pendulum — I can’t seriously dispute that any of the seats below the 10% cut-off would be lost by the LNP if the swing was really on, and in a movement of 10% against the government it should be noted that none of them were held by the LNP before its thumping win in 2012.
Indeed, the only two seats in that portion of the pendulum I might quibble over are Barron River and Keppel*; both regional seats in far north Queensland, these are the electorates that at various times over the past few decades have retained a history of support for the conservative parties even when the statewide trend was against them.
And looking above the 10% mark, there are other seats that might be expected to fall even as others on smaller margins are retained; Sunnybank, Albert, Kallangur and Everton are all seats that could hardly be described as natural Liberal territory over the past 30 years. All four — perhaps not coincidentally — are in Queensland’s south-east, and three of the four in Brisbane.
Just as there was a “wild card” seat last time that I thought was exposed to the LNP despite sitting on a big Labor margin — Ipswich — so too is there a seat this time, with the seat of Redlands, on the southern outskirts of Brisbane, posing a big question mark.
It’s a seat that has changed hands many times over the past 30 years or so. Will the “penis plonker” scandal, combined with the LNP’s very late preselection of a replacement candidate, gift an unexpected extra seat to Labor?
Or will the 21% buffer established in 2012 save the LNP in this seat? Time will tell.
One thing I am certain of, however, is that the LNP’s Brisbane seats will be hit harder in any large movement away from the party than its real estate in the bush will be: and it is here some surprising results might come to pass.
Faced with the long list of Brisbane seats it presently holds, it’s easy to forget that Liberals could muster no more than four seats in Brisbane at seven of the eight elections between 1989 and the smashing LNP win in 2012: and with variously 28 to 38 seats technically sitting in the greater Brisbane area — the total has grown over the past three decades — that’s an awful lot of Brisbane with a longstanding tradition of voting Labor.
The biggest wildcard of all in this election is whether — and how many of — these seats revert to form en masse in light of the obvious unpopularity the Newman government has experienced for most of the time it has held office.
If everything I have referenced in this article were to come to pass — with the Brisbane electorates staying reasonably firm for the LNP — I’d see a 53% result delivering 50 seats to a re-elected, but heavily depleted, state government: with the difference between such an election result and the 1995 debacle to be found in the half-dozen extra seats this scenario would yield for the conservatives in and around Brisbane.
And there is a further disagreement with Mackerras: as best as I can estimate, such a 50-seat result would elect a government composed of 28 ex-Liberals and 22 ex-Nationals; some of my contacts in Queensland keep trying to wear me down over the old Liberal/National divide (saying it no longer exists) but it does exist in one very important respect.
Health Minister Lawrence Springborg is also a three-time election loser as Coalition and LNP leader; I remain to be convinced otherwise, but Queensland’s conservatives are unlikely to amount to much if they can’t win a fair proportion of the seats in Brisbane — and under Springborg, they made virtually no headway whatsoever in the capital over three elections.
The era of Joh Bjelke-Petersen and his brand of folksy populism is gone; so too is the gerrymander that helped translate it into seats — including in Brisbane, even if they were held by the Liberals he so detested.
My point is that I don’t think people in Brisbane (or the south-east more broadly) are interested any more in leaders from the bush; and rightly or wrongly, Springborg (and his vastly unpopular rural colleague, Jeff Seeney) shouldn’t feature in any calculations over the LNP leadership.
Yet calculations there will be, for whichever way you cut these numbers, Campbell Newman will not win Ashgrove; the LNP will need a new leader; and he (or she) will need to represent an electorate in the urban corridor around Brisbane.
The likely pro-Liberal balance within the LNP after the election reinforces this view.
I think the post-election Premier is most likely to be the Treasurer, Tim Nicholls, or it should be; neither a rural MP or a multiple election loser as Springborg is, it seems Nicholls’ greatest crime in the eyes of some sections of the LNP is his friendship with former Liberal powerbroker and disgraced Senator Santo Santoro.
My response to that — in short — is “get over it.”
Poor judgement and badly prosecuted political strategies are the reasons the LNP is even contemplating whether or not it will survive this election at all; making a rural MP — no matter how impressive — its replacement for Newman would go a long way toward writing off the subsequent election three years hence before the LNP’s second term had even started.
And an election result featuring a 10% swing against it and involving the loss of some 28 seats from its 2012 result — including the Premier in Ashgrove — satisfies the prediction I made during the week of “a belting” in every sense except for outright electoral defeat.
A 10% swing against Campbell Newman’s government? I hope readers have enjoyed the discussion; feel free to post comments and counter points.
Of course, I could be wrong; the polls could shift again; Labor could shoot itself in the foot; some scandal could befall the government — who knows.
But right now, this is how I would see such a scenario playing out.
Tomorrow we will be back to something more issues-based than this exercise in crystal ball gazing.
*Yes, Keppel was created in 1992 — and won by the Nationals’ Vince Lester from opposition. He also held it in opposition after the fall of the Borbidge government in 1998.