Morgan: 53-47 to Labor in Bennelong

Today’s Fairfax broadsheets carry a Morgan survey of 472 voters conducted in Bennelong on Friday and Saturday, which shows Maxine McKew leading John Howard 53-47 on two-party preferred and 45.5 per cent to 42.5 per cent on the primary vote. Under the circumstances, the Prime Minister would be greatly relieved – the result is no worse than most of the other Bennelong polls published this year, not a single one of which has had him in front. Indeed, it marks an almost significant narrowing from the 55-45 recorded by ACNielsen’s last such poll in February.

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.

416 comments on “Morgan: 53-47 to Labor in Bennelong”

Comments Page 1 of 9
1 2 9
  1. Let’s face it: This is new territory.

    The tale of Stanley Bruce is a cheerful historical factoid compared to the gravity of the second-longest serving PM possibly losing his seat. And doing so in slow motion, thanks to the 24 hour news cycle.

    The whole campaign is weird. It’s almost like the Libs are already gone and are just hanging around to get their heads lopped off.

  2. I agree – I would have expected worse after the disasters of last week. Do we know when exactly it was taken? I have said before that the relentless media attacks on JH might generate a sympathy vote for him in Bennelong, as they did for McMahon in Lowe in 1972.

  3. Greg, do you think he can afford to take time out from the national campaign to shore up his own seat?

    In American elections it seems to be a given that any incumbent under 50% is in trouble. People know him, and have decided they don’t like him anymore. Whereas the challenger has more upwards room to move. Goes doeble for this particular seat at the moment.

    And Dinsdale, I agree. It’s like they’re all looking directly up at the guillotine waiting for it to fall, hoping against hope that it will snag on something on the way down.

  4. I don’t think the leadership tussle from last week will have any effect on the polls. I’ve predicted John Howard will win Bennelong for a long time now and I’m still predicting that.

    Election prediction: Coalition returned with a majority of 7

  5. Adam, “sympathy” in Bennelong will not necessarily play in the broader Oz electorate. Biddy McMahon got walloped bigtime nationally.

    The senior partner in Team Smirk/Rodent still faces the serious prospect of “doing a Bruce in Bennelong”. People who are normally switched off from politics, now have Ratty’s woes drawn to their attention on telly and in the tabloids. No one wants to back a loser.

    The Rodent scurried about Bennelong all weekend studiously dodging the Pied Political Piper. Bennelong time, is time he can’t spend pressing the flesh and photo-opping elsewhere nationally. Fighting a war on two fronts is much more difficult when the leadership team hate each others guts. Maxine’s candidacy is one of the great “strokes” pulled so far in this campaign.

    Even if Max loses, Team Tin-Tin wins.

  6. Given the reported sample size of 472, there’s an error margin of about 4.5% in this poll. If I were Howard I wouldn’t draw much comfort from this result being 2% better than the Nielsen poll in February.

  7. A 7% swing from a relatively large sample suggests
    -Bennalong is in danger
    A 7% swing in tonights figures would show 46-54 and Howard would be considered to be bouncing back

    One can read a lot of things into this
    -Howard is less unpopular locally than nationally
    -Obviously McKew will be facing someone who will be on the national stage presenting his case and may well lose % as the campaign starts in earnest
    -If Howard were to produce 46-54 then perceptions would change and more Bennalong voters might consider voting Liberal if they believe that Howard is on the rise.

    Personally I think tonight it will be 44-56 but speculation is more fun.

  8. Judging by the Bennelong poll breakdowns, some of the people who want Howard to “stay” do not inted to vote for him. I wonder if these are the same people who tell a pollster that “the country is heading in the right direction”, and mean they’re happy that Labor looks like winning the election.

    And if there’s been a 7% swing in Bennelong, and more than half the voters still want Howard to stay, what does that suggest agaout swings in other Sydney seats? Without the appeal of an incumbent Prime Minister, you would have to think that other seats nearby may swing more. There may be something in the suggestions that North Sydney isn’t entirely safe for Hockey. And while I’ve always felt that Turnbull had a better chance of retaining Wentworth that Howard did of retaining Bennelong, Turnbull must still be very worried.

    The other thing, which I’d be interested in getting people’s views on, is why is NSW swinging so much? The State Government isn’t very popular, and it’s howard’s home state. I can only think that housing affordibility is biting harder there than elsewhere. And perhaps child care – is it harder to get/afford a child care place in NSW than elsewhere? Or is health a bigger issue in NSW than elsewhere (except perhaps Qld)? Certainly the recent Eden Monaro poll was a hint that health is a bigger issue than most commentators think.

  9. I think this is the most significant line in the story, and one that passes by unquestioned:

    “On a two-party basis a Costello-led Government would trail Labor 46-54 per cent.”

    In other words, in Howard’s own seat, he is only 1% more popular than Peter Costello.

    Devastating stuff and I think it might kick off another round of leadership discussions.

  10. Coota – that’s a question for Adam. But I note Mackerras’s estimate at the time that Gorton suffered an 8.9% swing in the 1969 election; and on AEC figues Keating lost 9.1% in 1996. Both figures inflated because they came off pro-govt swings in the previous election.

  11. This is my opinion:
    Howard as a long term encumbant MP nearly always has a smaller swing than the national average. its one of the reasons why this election is so important to Labor, one more term and people start learning the name of their MP.
    Maxiene ws placed there, im sure to be a distraction for Howard in the lead up to the election. it’s her personal capacity that is making her a threat. these so called “celebriy candidates” are being placed in alot of these seats. Hockey’s is another example.

    The swing towards Maxine means Howard will continue to be distracted and less devoted to selling nationally. It’s only a pitty that we don’t have Rove McManus running against Costello.

    – as this is my first post here I’ll start clean and admit that I am a Labor member.

  12. Maybe I’m being naive, but a drive through Bennelong shows the sort of older established comfortable middle class suburban area that wouldn’t normally go Labor (I’m ignoring the underlying state seats – Debnam and even Brogden, as policitians, just weren’t serious players compared to Howard). I think that Howard’s margin in Bennelong would normally be higher than the current 4.2% – that number is artificially low due to the “Not Happy John” campaign last time, plus a strong green candidate (Wilkie). Labor might have a stronger candidate this time, and be properly funding the seat, but I just don’t think that Labor can win a seat where a significant proportion of the voters have probably never voted Labor in their lives and aren’t about to start.

    For what it’s worth, I hold the same view about Wentworth. I think the Peter King vote had a similar effect on the Liberal margin as the “Not Happy John” campaign in Bennelong.

    I would think seats like Hughes, Macarthur or even Greenway (IMO far fewer “rusted on” Liberals than Bennelong, despite the current 2PP margins) would have to fall before Bennelong went. That is, if a swing is really on Bennelong obviously could fall – but a lot of other Liberal seats in Sydney would go first.

    For similar reasons, I don’t accept that seats like Banks would be likely to fall to the Liberals in the foreseeable future, despite the apparent marginal status of that seat. I think that this last point is probably my safest prediction for the election this year!

  13. labor is putting even more pressure on howard with call for 3 debates, if he refuses he is cutting and running,not a good look to be seen as too scared to debate
    the bennelong poll shows, as a lot of poll bludgers expected that howard is going to have to spend more time in his electorate this election than in previous ones
    i expect to see a few more bennelong polls before this election is done

  14. Josh (18)

    Local one-off factors like Wilkie and King need to be taken into consideration if we’re trying to translate national polling into local swings. But if a local poll for local people shows a primary vote of 45.5 per cent to 42.5, then that’s what the primary vote is (subject to the usual uncertainties like MOE and the effect of the real campaign).

    Whatever affected those votes last time around is irrelevant now.

  15. If Maxine gets the Asian vote then she’s home and dry. 10% of Bennelong are Chinese and if they like the fact that Rudd speaks Mandarin and has a Chinese son-in-law they may vote for him. The last swing was because of the “doctors’ wives” but it was not sufficient. This group will still not vote for Howard. I know because I belong to this group. George M said that the Asian vote will determine who wins Bennelong. I’m beginning to see more of the Chinese in Bennelong saying that it’s time for Howard to go. Some of them have not forgotten that Howard kept mum when Pauline Hanson was on the rise. I haven’t forgotten either. As an ethnic Chinese I must add that most of my kind tend to vote with their hip pocket.

  16. Interesting about teh Nexus poll Will. Even though we do not have a trend for it, it seems to be pretty consistent with the other polls at the moment.

    Heard Downer on the radio this morning. After hearing him, I really thought that the Libs will run election ads with the Labor Front bench team, and then ‘Union official’ stamped across them, finishing with ‘Do you really want union to run the country?’ They are already planting the seeds for such a campaign with the constant mentioning of union bosses.

    I think Labor had better be prepared for it. One way to go would be to go history and show that without unions in the past we would have no limits on working hours, safety etc, as well as pointing out the other experiences of their team.

  17. Everyone already knows the ALP is made up of lots of union officials anyhow. It makes 0 difference. They’re all elected at state level.

  18. If you have three debates you might have:

    1. Howard
    2. Costello
    3. A mystery leader to be chosen by phone in.

    No matter, Kevin “The Iron Chef” Rudd will slice and dice the lot. His specialty is chicken and there are plenty of chooks running around with their heads cut off to satisfy his culinary talents.

  19. Hurray another conservative leader scraps across the line!

    The Centre-Right New Democracy Party won 153 of 300 seats in the 2007 Legislative Elections in Greece!

    I’m predicting a margin of about that in favour of the Coalition i cant see them having a majority of more than 5 if they manage to pull off a miracle.

  20. “Given the reported sample size of 472, there’s an error margin of about 4.5% in this poll.”
    Well it’s a long time since I’ve been a statistician, but I would have thought that sample size & POPULATION SIZE would both have an effect on MOE.

    To take an extreme example, a sample of 472 from an electorate of 500 would not have an MOE of 4.5%

    Another factor to bear in mind is that the Bennelong population is more homogeneous than the Australian population . EG no farmers in Bennelong.
    But this is nitpicking. I’d rather be in Maxine’s shoes (anytime) than John Winston’s .

    Call the election please Says:
    September 17th, 2007 at 7:12 am
    “I don’t think the leadership tussle from last week will have any effect on the polls. I’ve predicted John Howard will win Bennelong for a long time now and I’m still predicting that.

    Election prediction: Coalition returned with a majority of 7”

    You may be right, but your predictions, specially about the coalition majority are certainly at the lower end of the perceived probability scale. Possibly even outside the Moe?

  21. In 2004, some 33 seats were polled individually in the 3 months before the election, some of them several times. Bennelong was not one of them. On average, the polled result in those 33 seats OVERestimated the ultimate ALP TPP in those same seats by about 2%. The various national polls over that same three month period also overestimated the national ALP TPP by 2%. But, applying the national ALP TPP polled swings to Bennelong in 2004 would have UNDER-estimated the ALP TPP by about 2.7%.

    Take your pick.

  22. Whatever one can say about your comments Glen, one has to admire your never say die spirit.

    You are certainly my choice to be leading the chorus of “Always look on the bright side of life” for the Liberals on election night.

  23. Gippslander the stats brigade will be in shortly to jump on you, but since I’m here already… I believe the size of the target population actually has minimal effect on the MOE for a sample size. It has SOME effect, but it’s small. The example you gave was extreme. Considering a target population of 50,000 against a national electorate of 13 million, the effect is only pont one of a percent or thereabouts.

    Call the Election: bad mood today? So much gloom…

  24. re 40: Crispy

    I have a cold so that could be it! However, I’m just going to be pessimistic from now on and then be pleasantly surprised if the ALP win. I’ve been considering putting a bet on the Coalition so I’ve got something to be happy about.

  25. Geoff Lambert #37
    The 3 month average straddles the Latham Implosion. polls taken say one month out would have shown a smaller discrepancy, but Possum might be able to shed light on this.
    The “swing” in Bennelong would have been affected both by national trends, demographic changes, prime ministerial incumbancy, and a raft of other issues we don’t know about. Picking over the entrails can be fascinating, but the smaller the sample the less meaningful any results. As for prediction, all I can say is that I’d rather be ahead in the current polls than behind .

  26. As an ethnic Chinese I must add that most of my kind tend to vote with their hip pocket.

    So does most of the rest of Australia, which is why we’ve had this Government for so, so long.

    When is the Newspoll out?

  27. Just to remind you Glen, this is the biggest one in the world. And apart from the Australian election the one I want Hillary to win more than anything.

    While on the subject. I’ll point out that since George Bush has been too busy with Iraq, South America has swung sharp left. (No more right wing death squads). The only two countries not left wing are Paraguay and Columbia. Nicaragua has even gone back to a left wing government. George Bush has restore democracy but not the way he intended.

  28. I wonder whether the chinese component is as well polled as the white middle class, I can imagine there are issues here that may underrepresent their opinion.

    Also, it may come down to whether the chinese vote as typical
    – imigrants (Labor)
    – small business owners (Liberal)

Comments are closed.

Comments Page 1 of 9
1 2 9