Bennelong shock: McKew leads Howard

The worst-kept secret of recent polling history, the Sunday Telegraph’s Galaxy poll from Bennelong, has now been officially unveiled. As foreshadowed on Lateline, it has Maxine McKew leading John Howard 48 per cent to 47 per cent on the primary vote. McKew’s lead on two-party preferred is 52-48. The following, at least, is news:

Overall, 81 per cent of those within Bennelong who voted for Mr Howard at the last election say they will vote for him again. But his support has been eroded by the 17 per cent of Liberal voters at the last election who now say they are leaning more to Labor. In contrast, Maxine McKew’s vote remains solid – 92 per cent of those who voted for Labor in 2004 are sticking with her, along with 73 per cent of the former Green voters … A key change since August – and one that bodes ill for Mr Howard – is that more Bennelong voters now claim to have locked in their vote. Overall, 77 per cent claim to have made up their mind which candidate they will vote for at the election later this month, up from 69 per cent in August.

Author: William Bowe

William Bowe is a Perth-based election analyst and occasional teacher of political science. His blog, The Poll Bludger, has existed in one form or another since 2004, and is one of the most heavily trafficked websites on Australian politics.

94 comments on “Bennelong shock: McKew leads Howard”

Comments Page 1 of 2
1 2
  1. Sorry to post such an immediate change of subject, but I need to ask… when can we expect to see the preference details of the group voting tickets?

    I can’t get an answer out of the AEC site. The only useful piece of information there is that early voting starts next week. So it must be soon.

  2. Well, well, well….

    This is Very Interesting. While I doubt that Howard will do a Stanley Bruce (reference the number of McKew voters who are willing to reconsider their options), he will certainly be forced to campaign in his own seat – and cut back on appearances elsewhere. None of this is bad for the ALP, or good for the LNP.

    All to the good, IMO.

  3. Is this boring news, somehow?

    Not to me, at least.

    I went out, much earlier, Saturday evening, and we all spent a fun evening, among other things, planning the arrangements for our election party.

    This is more exciting, and more rewarding, than one’s 21st!

  4. Oh, I spy. Something is different.

    The format looks a little, mmmmm, undressed, really.


    Because, the posts are no longer numbered.

    And why?



  5. Ok, glad William posted the link to this article. The first place that I went when I got online today was the DT. I has assumed I would find it on the front page. NOT. I checked the front page of the “election coverage” as well. NOT. The Daily Telegraph did a good job of burying this article and if you were just an average voter cruising their website you would have a hard time finding it. Those who have seen it in the print edition, how prominent was it? Front page or buried in the middle some place ……

  6. I made some brilliant predictions, alas now sadly lost, in yesterday’s threads, but I will be reminding youse of them as and when they come to fruition.

    That I made them, youse will have to take my word for it. It’s all William’s fault.

  7. Post 13, maybe.

    Email, as sent to me. I have excised (hopefully) my name and the links. Otherwise, email in entire. I have not yet viewed or responded to the questionnaire.

    Subject. On Line Opinion/The Australian qualitative research into interest rate rises.
    Date. 3/11/2007.

    Dear XXX

    Thanks for completing the first poll in our collaboration with The Australian.
    We’re doing a second poll on interest rates. The plan is to have the results before the Reserve Bank makes its announcement next Tuesday on whether it will raise rates. To complete the questionnaire you can either click on and paste into your browser address bar. The survey will close on Monday at midday, so please don’t delay.

    We’re also continuing our analysis of the first poll. Today’s paper contained an article by Matthew Franklin based on the responses of some of you from Queensland. There was also an article by me which appears to have missed the online version – I’ve copied my original version below. We’ll be following-up with still more analysis of this data as the weeks go on. The message from it is pretty unequivocal – we’re getting a change of government – but the wrinkles are always interesting.

    Thanks for being part of our online research projects. As always, if you want to be removed from our mailings, please email me by return with “Unsubscribe” or “Remove” in the subject line.


    Howard’s howlers might collar under-dog sympathy (The Australian’s headline)
    Newspoll suggests that after two bad weeks, voters are moving back to the Coalition. Is Howard doing something right? Probably not. It may just be because he’s doing most things wrong.

    We Australians are eccentric. Our greatest military celebration is a defeat, and our greatest horserace is a handicap. We’ve taken the biblical injunction that “the first shall be last” to heart, and the Victa mower to the field of tall poppies.

    Howard might have seemed not to be listening, but now we’ve got his attention, we can’t help feeling so sorry for him some of us are changing our minds.

    It’s possible to milk this dynamic through what spin-doctors call “expectations management”, and our research suggests that amongst the groups most susceptible to it are those who voted for Howard last election who are now intending to vote for Rudd.

    These are Australians who have strong moral, although not necessarily Christian, values. They are what I call post-materialist conservative Australians and by “conservative” I don’t mean that they would describe themselves as such. They are voters who are more worried than most about moral issues, and who value their society more highly than the economy, often seeing the two concepts as opposed. They are conservative in that they are resistant to change and vigilant of community, the way conservatives used to be before Reagan and Thatcher.

    The most telling thing about them is their hesitations in voting for Howard, which are topped by “lies”, followed by Howard himself. This illustrates the moral dimension of their concern. At a much lower order of importance they also nominate Costello as a hesitation, suggesting that a clear succession is no cure for their dislike of Howard. Secondary concerns with Howard are his lack of “vision”, his “policies” and his handling of various “issues”.

    In contrast, they have few hesitations about voting for Rudd. What hesitations there are centre on his lack of policies and his lack of experience. They are less likely than most to be concerned about his union links, or his party.

    They diverge from the bulk of the poll in the issues that concern them. While the economy is cited more than any other issue, it is mentioned by fewer than 10% – less than half those that mention it in the total sample. “[Climate] change” comes next, but again with less frequency than in the total sample. “Health” is in third place, rather than education, and then come the social concepts like “Australia”, “country”, “government”, and “people”.

    “Work” and “education” are their bottom concerns, with “work” indicating in about half the cases concern about Work Choices, and in the other half relating to some aspect of employment.

    Why would these voters slip back to Howard? I suspect most won’t because their antipathy to his morality is so strong, but some will, for three reasons. Their moral mind-set makes them suspicious of government on principle, and forgiving of human fallibility.

    Rudd’s ME02 Howard-lite approach ensures Rudd’s ultimately just another politician. And they expect Rudd to win easily. They can therefore safely vote for Howard to send Rudd a message that he shouldn’t get above himself either.

    Plus, now John’s down and out, but still trying, he has some appeal to their moral side.

    The best thing that could happen to Rudd now would be a bad few weeks, just to convince swingers that he’s not “Captain Perfect” and might fail. So far this week, he’s on track to achieve this.

    PS by CW. Their table, if it could be called that, is as follows. I could not be bothered attempting to reconfigure it for this post. It makes no sense, in any event.

    The message is more intriguing.

    Concerns of swingers versus concerns of total sample

    Swingers Total Concept Ranking Concept Ranking
    Economy 9.6% Economy 19.8%
    climate 8% Education
    8% change
    7.2% health
    7.2% environment
    7.2% water
    6.4% future
    5.6% people
    5.6% country

  8. To repeat something I posted yesterday:

    If its still clear on th eve of 23/11 that the government is going down, then I suggest that whether JWH is ahead, even, or behind in the polling will be irrelevant because the hassle of voting will cause many, probably most of the non rusted on Liberal voters to go with Maxine to avoid having to repeat the exercise a few months later.

    They may feel sorry for Howard, but I doubt the depth of their sorry will be so deep that they’ll award him what they know will be only a short term personal victory. Why care for a bloke who won’t care enough to stick round for the whole term.

    If I was handing out Lab HTV cards I’d be making that very point to every punter – Maxine is in it for the long haul, not for a 24 hour victory lap!

  9. What I would like to see to tip the scales in Bennelong in Labors favor some more is this.

    On Thursday or Friday just before the election, Rudd have an open interview type forum with the Chinese in Bennelong in Mandarin. It would go down very well I suspect and if there are any of the Chinese in Bennelong who are still wavering it might tip them over 🙂 ……. One thing Howard can absolutely not do and with so little time before the actual ballot, it will pull a HUGE impact. Unless I am wrong (someone correct me please) I don’t think Rudd has campaigned in Bennelong so he is “due” for this at some point, he has certainly hit other key marginals already.

  10. Nice to see it this close for Maxine with three weeks to go. I agree with MayoFeral that a wodge of not so rusted on Liberal voters will defect as the campaign ends when it will look more and more likely that Howard, if re-elected, would only be a back bencher in the opposition for as long as you can say ” by election.”

    The other thing against Howard, (and we all must genuflect to the Possum), is that great graph of interest rates and Labor’s primary vote. Need I remind everyone that it’s lock step in a positive correlation?

    So, come next Wednesday, when the Reserve wallops Mr Who’dya Trust with another rate rise, then we can whack on another slab of voters for Maxine.

    Sportingbet still has the best odds for Labor, $2.85, so I’ve got my bet on for what will be an historical night.

  11. Just watched the Galaxy Polls bloke on ‘Greet the Mess’ on 10.. He reckons the vote has polarised at on those 47/46 figures so the seat will have to look at ALL preferences (not just Greens) to get the final result

  12. This is not to say Maxine will win, but taking into consideration all disclosed form to date, IMHO the $2.85 at sportingbet is the biggest overs in a two horse race i have ever seen.

  13. The Galaxy Polls bloke (David Briggs) on ‘Greet the Mess’ on 10 (thanks Alex) said there will be a national Galaxy Poll out tomorrow.

  14. Libs are getting desperate, bussing in uncomprehending “supporters”. I ran a leafleting session down at West Ryde yesterday and while the Libs outnumbered us, they couldn’t connect with the Locals. I was really buoyed by the positive reception I got (and always get) when saying “Maxine McKew” about 500 times in 2.5 hours.

    BTW, if Labor will have to rely on Green preferences, it also means Howard will have to rely on religious extremist preferences.

    But Maxine will just scoop it. She’s got a personal following in excess of the national swing. Howard should have read the writing in the wall and moved seat a few elections ago.

  15. i presume howard realizes how bad it would look to the australian electorate ( a nation of punters) if the betting on bennelong has him being defeated
    i wonder what it would be in % terms of votes nationally for howard if he is seen to be going down in his own electorate a week out from the election

  16. Well I for one,will not be crying any crocodile tears for the Rodent if he goes for the “long walk”.The hubris he’s caused in this country during his tenure is more than enough reason to see the back of him and to consign the Howard years to the scrap heap,where they and he belong:)

  17. yes, it could be a very interesting dynamic. It’s never been a factor in a previous election that the Prime Minister (or Opposition Leader?) has been under threat of losing their seat in an election. What impact might this have on the thinking of potential coalition voters if they think they might end up with the less popular Costello as the party leader? Would they actually prefer Rudd to Costello?

  18. Time to send in the kirribilli removalists?,23739,22693601-953,00.html

    SPORTS bookmakers say Labor is such a certainty to win the federal election that the removal vans might as well start heading to Kirribilli House.

    Halfway through the campaign, punters have plunged on Labor, even identifying the 16 seats the Opposition is likely to take from the Coalition to throw out the Howard government.

    An analysis of the four major betting agencies has Labor favourite to win 15 Coalition seats, including the Queensland electorates of Blair, Bonner and Moreton. Labor is also equal favourite to win the Liberal-held Queensland seat of Bowman with one agency and pushing for favouritism with the others.

    There are also another five seats where Labor is shortening in the betting – including the Townsville-based seat of Herbert.

  19. The dead tree version of the Brisbane Sunday Mail has a story of John and Hyacinth busy renovating their house ready for the move from Kirribilli too.

  20. With ‘all due respect’ to Messrs Leigh and others who have been trumpeting the invisible predictive hand of the ‘markets’ in election betting, everything we’ve seen this year suggests that the markets have been reactive to the polls.

    The only advantage the markets have is the ability to factor in the temperamental (as opposed to ideological) conservatism of the Australian electorate come polling day. That is, and with apologies to Possum, waverers do NOT always split with the long term trend (although one can always define ‘long term’ to get the desired trend). In most elections, unless there is a clear case for change + reassurance that change is not a risk, the waverers tend to plump for the status quo.

    This is reinforced by compulsory voting. It’s the greatest of all incumbency benefits in a stable, well-off society like ours, especially since the parties converged into relatively undifferentiated machines. It helps explain Lab States + Tory Commonwealth, which is a long-standing pattern in Oz politics, better than the other explanations. (People don’t reason ‘Can’t have two parties in govt at the same level’. Rather they think ‘why risk change at either level’, and this choice becomes self-reinforcing, as the losing party gets weaker and hence looks riskier next time, and the winner rolls itself in the other incumbency benefits).

    And it’s the one question of macro interest left today. Will Labor, in seat terms, win handsomely, or very modestly? Answer depends on this: Will apolitical waverers decide there isn’t a clear case for change and/or feel it a bit risky?

    Understanding this explains the Coalition’s otherwise flaccid and contradictory (to our ears) messages that ‘Labor = me-too’ + ‘Labor = risk’. And conversely Labor’s modest and barely less contradictory messages of ‘add up your grievances on health + education + workchoices’ and you have a reason to change, but ‘we aren’t promising big changes’.

  21. Bluebottle and another poster (sorry, I have forgotten which one) were collection predictions of seat outcomes. Is that list still around or has it vanished into the digital ether?

    For the record: Mine is 80 to Labor.

  22. It will be hard for John and Janette if he loses Bennelong.

    No butler, chef, housekeeper and gardener. No unlimited wine cellar for the bottle or two of SA red that he likes to imbide. No chaffuer or free petrol. No harbour views.

    Will have to buy their own groceries and pay for their own petrol, maybe even fill up the car himself, then he can truly understand the pain Aussies feel at the petrol pump. Will have to watch the driving bit though with his known love of the SA red, doesn’t want to be remembed like a former premier who went to scotch (not the college).

    It truly will be a massive culture shock for the Howards if he does lose Bennelong, back (partly at least) in the real world after 11 years living in la la land.

    I think they should be offered counselling to help them adjust.

  23. Graeme, you are quite correct about the money following the polls. How else does one make sense of $2.85 for McKew in Bennelong? Now that the Galaxy poll has been published, and the media has picked it up, we can expect to see Labor shorten accordingly. If the punters were ‘ahead of the curve’ so to speak, those odds would be much closer together for Lab/Lib than they currently are, and hence look like the poll results.

    As for your analysis of the voter’s desire for change, I think you have it about right, and note that the under 50’s have pretty much decided Howard is passed his use by date, and obviously aren’t all that attracted to his replacement (Costello).

    Try as he might, Howard cannot reverse time, and his has well and truly run out in a significant proportion of the electorate’s mind.

  24. Rx,

    Official nominations have closed, but unofficially you can always keep your own tally. They put a closing date on the list just so that they didn’t have to be trolling PB daily for that sort of thing.

  25. The odds currently available on McKew are crazy.

    There have now been 4 polls in Bennelong since May, and all of them show McKew winning with at least 52-48 of TPP. All 3 of the galaxy polls have her primary vote on 47, unmoved since May.

    In addition, we’ve heard a number of times about internal party polling showing Howard in trouble.

    On this basis, you’d have to rate McKew as *at least* an even money bet wouldn’t you? And yet you can still get fantastic odds of 2.5+ on her. I suspect they will shorten during the next couple of weeks (perhaps between now and wednesday) as people begin to wake up and smell the roses.

    Anyone know whether another Bennelong poll is going to be done before the election? A poll with a week to go could cause chaos with another 52-48 result.

  26. Ashley,

    One possible reason for the apparent discrepancy – relative to the betting in other marginal electorates – in the odds on offer for a McKew victory, is that the Bennelong markets have been fiddled with by Liberal Party operatives. Simon Jackmam recently mentioned that it was indeed possible that some individual seat betting markets have been tampered with by individuals from both sides of politics.

    I have no doubt that the Bennelong markets have been fiddled.

    It is a bad look for Howard to be behind in Bennelong opinion polls; it would be catastrophic for him to be behind – or even on level terms – with McKew. Think perceptions/win expectations.

    I have no doubt that the bookies are holding a large amount of money for the Bennelong market.

    The bookies are basically giving free money away to backers of McKew in Bennelong.

    Stanley Melbourne Bruce will soon have company.

  27. I am amazed at the consistency in the Bennelong polls. Knowing the seat well, it strikes me that it would be one of the hardest in the country to get a representative sample in. It covers all the way from harbourside mansions in Hunters Hill to ex Housing Commission fibro in Ermington, with big lumps of low grade 60’s and 70’s era units in Ryde and Meadowbank. Lane Cove is full of ‘leafy’ suburbia, with lots of ’empty nesters’. Eastwood and Epping similar but with lots of Asians mixed in. On top of that there is the very big proportion of Koreans and Chinese immigrants, including many recent arrivals, many of whom would not answer a ‘phone poll. How all of these factors are sampled for, and then weighted correctly is beyond me. My congratulations to Galaxy if they have got it right.


    Alan H

  28. fagin, i agree with you about the fiddling in bennelong,
    it would be interseting to know how much money and where it went in bennelong maybe after the election we might find out
    it’s perception, howard cannot be seen to be losing his seat
    why vote for howard when he won’t be in parliment

  29. Well, I’ve taken one for the team, Just put $100 on Maxine, I don’t know how often they change their odds, but it had no effect, maybe the pool is much larger!
    Which then leads to questions about who would be betting on Howard?

  30. # Rxon 04 Nov 2007 at 9:16 am
    Bluebottle and another poster (sorry, I have forgotten which one) were collection predictions of seat outcomes. Is that list still around or has it vanished into the digital ether?

    Is this what you are after, Rx?

    Predictions @ Poll Bludger
    as at 28th October 2007

  31. It looks like Howard has been tipped off by the Reserve Bank Board that they are going to rise rates on Tuesday.

    {“But I do know this: interest rates now are lower than they were at any time under a former Labor government, and at 8.3 per cent, they are less than half than the notorious peak of 17 per cent reached under the former Labor government.”

    He said interest rates were under pressure from inflation and the voting public would have to ask itself whether the Coalition would be better able to manage those pressures than Labor.

    He said there are also other challenges facing the economy, predicting further repercussions from the sub-prime meltdown in the United States.

    “I would say to the Australian people: there are inflationary pressures in the economy which put pressure on interest rates, is this a time to hand over economic management to a party that has an inflationary industrial relations policy,” he said. },22049,22700300-5001028,00.html

  32. At Sportingbet you can back Howard to win Bennelong at odds of 1.38.

    Why would anyone (other than the liberal party) do this, when you can back Labor nationally with the same bookie and get 1.32.


  33. Scorpio #46,

    People forget two facts when talking interest rates:

    1) In 1989, the whole world was in recession – that sent interest rates everywhere through the ceiling. Labor, Liberal, Conservative, Democrat, Republican – the government couldn’t stop it anywhere. Bush lost the 1992 election because of it, despite the huge popularity boost from the first Gulf War. Thatcher was killed by it in the UK. It simply didn’t matter what your economic policies were, or what party you belonged to – it was a killer everywhere.

    2) Today’s interest rates are at least 1% above the OECD average. To the Lib naysayers, I post this link:

    You will note that, among other nations, the UK, USA, Canada, Korea and – get this, that lefty nanny state – Sweden, are each at least 1% lower, alongwith a great many other nations’ rates.

  34. Can I suggest a project. SportingBet offer a $100 free bet as soon as you sign up with a $30.00 deposit. Bet your $30 however you like, but put your free $100 on Maxine. I have just done this and it felt good! Let’s see if we can get Maxine below $2.00! Tell your friends, pass it on, letterbox, let’s have a Maxine tsunami.


    Alan H

Comments are closed.

Comments Page 1 of 2
1 2