At Australian Policy Online, Swinburne Politics Professor Brian Costar writes about how the press got the Gippsland by-election (and the state by-election for Kororoit in Victoria) so wrong:
We are told that the Rudd government is “reeling” from the “backlash” it suffered in Gippsland, where it took a “savage swing of 6.5 per cent.” The reaction to the Kororoit result has gone even further – some accounts would have us believe that the Brumby Labor government is set for certain defeat in 2010.
Let’s get real. On Friday 27 June the National Party held the seat of Gippsland; on Sunday 29 June the National Party still held the seat of Gippsland. On Friday 27 June the Labor Party held the seat of Kororoit; on Sunday 29 June the ALP still held the seat of Kororoit.
The rest of the article can be read here.
Good work, Prof Costar.
Thanks, Mark
cheerio
The media spin doctors at work Brian Costar is spot on, I sometimes feel we are back during the Gough years with a total negative anti labor press, even the ABC at present cannot appear unbiased. I now read the blogs before the newspapers how depressing is that, the one term govn expressed by a lib turn lab returned Lib mate of mine received the appropriate response from me even though we were both in Casuality helping family and friend in my case.
Brian Costar is the authority on the Victorian Nationals, and also knows his stuff otherwise.
Bilko @ 2,
We are back during the Gough years, mate. And believe me, it going to get worse. The meedja and those who back rhem are behaving like babies who’ve had their dinner snatched away. It will take them a long time to get over it.
I agree with Brian Costar that the press reports have been hysterical and ridiculous. It is important to note that both seats were retained by the parties that previously held them. I also agree that the addition of a third competitive candidate in each seats makes the contests quite different from normal.
With regards to Gippsland, I disagree with his assertion that the swings away from Labor and National on the primary vote all went to the Liberal Party. That’s too simplistic.
Let’s look at the votes for the minor parties. In this by-election, the Greens gained a swing of 1.4$ to receive 7% of the vote. The LDP received 4.2% and did not stand in the general election. In the general election, there were two independents, one received 2.1%, the other received 3.2%. Family First polled 4.3% but they didn’t contest this by-election. I have no knowledge about those independents but I can confidently say that most people who voted for Family First would not have voted for the Greens or the LDP in the by-election. So where did the Family First votes go? I’d say they’d have gone mainly to the Liberals or Nationals. That would put Costar’s claim out of whack.
I agree that there would be some people who voted Labor in the general election and then Liberal in this by-election. However, I can’t believe that most of their lost votes went to the Liberal party. It is likely that some of their lost votes went to the Greens.
One clear fact of the Gippsland by-election was that the conservative vote increased. However, it’s worth noting that by-elections usually result in bigger than normal swings against the government. The people in Gippsland had a few reasons (in their minds) to be concerned about the new government so some changed their votes to vote against the government. Does it say much about the Australian public’s general attitude towards the new government? I don’t think so.
On reading my post again, I think I would’ve done better to leave out that third paragraph. I think my argument is flawed.
I agree with most of Costar’s article, especially about Kororoit (which The Age remains determined to misinterpret so it can talk up its theme of trouble for Brumby). But if the Gippsland result was “totally predictable”, why did Costar’s own prediction (which sounded about right to me at the time) prove more than 5 per cent off the mark?
BTW, Costar’s piece ends with speculation on how a Labor-versus-Liberal preference allocation in Kororoit might have gone. The VEC has given us the answer: the swing against Labor was 4.1 per cent, compared with Costar’s estimate of 1.6 per cent.
Re – Costar – as Jack Strocchi would no doubt attest, I’ve never been enamoured of the predictions game with elections. But do you agree that it doesn’t dent the soundness of Costar’s subsquent analysis, William?
I have a lot of time for Costar, and his general point is sound, but I don’t think the Vic Liberals should feel too badly about this. They got a 4% swing, which isn’t great, but isn’t too bad either. On top of that, at the state election Walters preferenced them, whereas this time she went to Labor. Without that the swing would have been a percent or so more.
Moreover, some people who might have otherwise voted Liberal would have backed Twentyman, and then followed his preference card to Labor.
I’d guess that without the effect of preferences from the two independents the swing to the Libs would have been around 8% – not enough to terrify Brumby given what by-elections usually do, but not a slap for Baillieu either.
Mark, I agree with all the points Costar makes about mitigating circumstances in Gippsland – big Coalition ad spend, Nats and Lib candidates maximising the Coalition vote, good Nats candidate – and would further speculate that Peter McGauran might not have had much of a personal vote. Bearing all this in mind, I was expecting a status quo result or a slight Nationals swing, as was Costar. What nobody saw coming was the size of the swing in the Latrobe Valley – high enough that it can’t be shrugged off, and different enough from the swing in East Gippsland (10 per cent versus about 4 per cent) that it can’t be put down to a uniform factor like prices. The most likely explanation is that it shows how blue-collar workers in low-to-middle income areas will react to even the suggestion they should have to bear sacrifices on behalf of an emissions trading system.
Well, maybe blue collar workers in the middle of dirty coal producing disticts, William. But if we had any evidence that something similar could be observed in some of the seats in Qld and NSW Labor won last time, I’d be more convinced. If we’re to believe the Quarterly Newspoll, Labor’s well up on primaries in Qld compared to election time.
Wasn’t it also the case that a lot of state factors may have been in play in Gippsland, and particularly in the Latrobe Valley? Didn’t state Labor lose two seats there?
Well I know nothing of Costar, but I totally agree with the notion that the usual suspects are cobbling together a new narrative re comparisons with Whitlam (although the real Dennis Shanahan may well have been abducted by aliens and replaced with a rectal probe, if this is anything to go by):
Labor retains huge Newspoll lead over Coalition
Labor has kept its crushing lead over the Coalition in every state and among every voting group in the past three months, while Kevin Rudd has outpolled Nelson everywhere by at least six to one as preferred prime minister
But yes, the Cowardian pearl-clutching, swooning and smelling-salt antics of Bolt, Akerman, Albrecthson, and quite a few who should know better – for a government this early in the term and facing signficant policy challenges – is a bit disturbing.
The script is written and, as Hunter S Thompson said, “the Fix is in.”
Mark, I too would like more evidence, and I’ll happily admit to filling the blanks with conjecture in its absence (remembering that these Newspoll breakdowns flatter Labor as they are based on three months’ worth of data). With regard to state factors, surely this points to another long-term danger for Rudd – coast-to-coast state Labor governments ranging in age from 6 to 13, each rapidly wearing down its reserves of political capital.
This analysis is entirely correct but does not tell the whole story.
In Kororoit, Les Twentyman came within a few hundred primary votes of winning the seat on preferences. He ended up a two-hundred votes behind the Liberals on primaries and this meant the flow of preferences went to Labor.
While this doesnt mean the seat is suddently marginal or at risk, it should be a wake-up call to the ALP.
Several of the governments have been borrowing that capital from the coalition opposition for years.
It’s a measure of how jaundiced the political coverage has been over the last few months (years?), that when I read something like this (which is really nothing more than a cool, calm look at the numbers) it seems like such a breath of fresh air.
There are defintely significant elements of the media that appear to be (whether consciously or not) death-riding the government. Yet it’s also a sign of how irrelvant the press pack is becoming that these predictions of Rudd’s demise are not being shown in opinion polling – last time I checked 55% was still landslide territiry, and a swing TO the government since the last election. That suggests to me that your average Joe is happy to let the new government “get on with it”.
Related to this is the press (and Lib) narrative that this is a “do-nothing” government, entirely obsessed with spin. This is a ludicrous suggestion, IMHO – by the end of June, parliament had passed over 50 bills, relating to anything from tax relief to dental benefits (and a lot more in between), and there’s clearly more to come. Then again, this is the same press that had written Wayne Swan off in February.
It seems to me that various over-paid journos (and I don’t mean the real ones, who actually work for a living, but the puffed-up columnists) are desperate to write their retrospectives about the Rudd government, even though, after just 7 months, that seems a ridiculous proposition to anyone outside of their self-important little bubble.
“The largest factor influencing the final vote in each by-election was the entry of a competitive third candidate who had not contested the previous general election. ”
Jeez, you wouldnt have known that reading the press reports.l fairly important point.
So much for the end of honeymoon thesis!
(mind you, It wont surprise me if Brumby is in trouble next election.)
Paul @ 4 agree a bit of a worry,I just want Rudd to get a move on although Alan Ramsey’s col last Sat was disturbing. Hugo # 16 is on the ball also
“This analysis is entirely correct but does not tell the whole story.”
Geoff Baxter@14 I totally agree.
Possum dismisses Paul Austins piece because…
“The Age remains determined to misinterpret so it can talk up its theme of trouble for Brumby”.
While there is plenty of room to argue against the conclusions made by Austin on the figures from that by-election result his skills as a journalist, fairly telling the story, are still encouraging.
It was important for the public to be exposed to the dirty tricks campaign, run by Labor, and their manager Newnham and the pathetically hands off comments by Premier Brumby that “I don’t run byelection campaigns”, for instance.
Good to remember the difference in style between the actually elected Bracks and the ‘wild horse at a gate’, Brumby. His abrasive ways may not have become obvious to those outside of Victoria but you would have to give credit to Austin for his little chat about the new “Action Man Premier”.
http://www.theage.com.au/opinion/sledgehammer-politics-won-kororoit-20080702-30o0.html?page=-1
Actually Joe2 – that was Pollbludger whom said that which you quote, just for the record and all.
Well, again, my own view would have been considerably different if someone in the MSM had highlighted the new competitive 3rd candidate factor – rendering all talks of “swings” rather irrelevant vis-a-vis the previous elections.
Whoops!
Extreme apology Possum.
Geoff Baxter @ 14, that isn’t right: as you can see from the preference distribution, Labor’s Marlene Kairouz reached 50 per cent after exclusion of the CEC and Family First-turned-independent Tania Walters, and would comfortably have won the seat regardless of who had been excluded first out of Twentyman and the Liberal candidate. We don’t know for certain, but we can be extremely confident that distribution of Greens preferences would in fact have put Twentyman ahead of the Liberal, but Kairouz would still have beaten him by 5684 votes (9.0 per cent). If the Liberal candidate had stayed ahead – and assuming this gives us by far the best yardstick for comparing the result to the 2006 election – the margin would have been 21.5 per cent. There really is very little evidence here to suggest Labor faces any sort of backlash in the western suburbs.
I do work with Brian and he’s smarter than the average pol scientist. But I don’t follow his logic on Gippsland. If 3 cornered contests could inflate the non-Labor vote by such a percent, there would be irresistible pressure for them to be common at federal elections.
Reality check: if you were a ‘l’iberal who voted Labor in prior polls because you saw the Nationals as unreconstructed agrarian socialists or reactionaries, why would you then forget that ideological position and blindly follow the Liberal how-to-vote card?
The average swing in a by-election is around 4-6% which ever way you look at it the State Government is not under any threat by the K by election result. If there was any loser in this election it had to be twentyman and his sidekicks artist Phil Cleary. (I still blame his lot for the reason Australia is not a republic today).
In many ways it is easier mathematically to win a 60% “Safe” Seat then a marginal 515% seat. Twentyman needed to peg back the ALP to below 50% and most importantly he needed to secure 26% of the primary vote and make sure he was in a good position to collect all other preferences,
Problem for Twentyman is is not the local hero he thought he was and as long as he sides with the likes of Cleary and others he is unlikely to succeed in winning the seat. A win in a full state election would be impossible unless the government is really on the nose which it isn’t.
Another example of an independent winning a safe seat is the former member for Mildura. Kennett was on the nose and he came up the middle. Now that Labor has regained its position such a win is highly unlikely under the circumstances.