Peter Dunne's cosying up with National is less ideological and more a practical reality of minor party survival, plus more poll analysis

I’ve always suspected Peter Dunne wants to be Prime Minister. It sort of befits someone who was reading Hansard when others were smoking behind the school bike-sheds. The dim prospect of that dream eventuating became apparent early on in his career, when Labour stopped promoting and started shedding its right-wingers. He didn’t need to read Hansard to realise he was never going to lead the Labour Party, let alone the country.

Ambition thwarted, Dunne still wanted to lead. He has since been leader of an array of groupings. He has joined coalitions led by National and Labour. He has enjoyed the ministerial trappings. And he is now one of Parliament’s more venerable members.

Outside his firmly middle-class electorate, he has never been especially popular. There is often a pomposity about his public demeanour – notably his impressive contretemps with TVNZ’s Mark Sainsbury on election night in 2005. In person, however, he is friendly and approachable.

There are three things which are certain about Peter Dunne. First, about two weeks out from polling day, he will craft a photo opportunity with the National Party leader of the day. Then, he will pledge his preference for National’s policies of tax cuts and ‘sensible’ policies for the family. And, third, he will claim that Labour’s closeness to the Greens rules out any possible post-election arrangement with the left.

To be fair to Dunne, he has little choice. For the past few elections, he has held Ohariu-Belmont because its National voters give him their electorate vote. To earn that, he needs to, at the very least, genuflect to the current National icon, and that’s best done just a few days prior to polling day. (And having behaved himself during the last three years as a minister in a Labour-led government, he will have sewn up some Labour votes too.)

The nature of electorate boundary changes mean his Ohariu-Belmont electorate tends to grow slightly more National-centric each election, as it sheds comparatively less well-off suburbs in the north and picks up more affluent areas to the south. As a former Labour MP, he has his own personal links to those on the left. He hired Rob Eaddy, Jim Bolger’s former advisor, to be his chief of staff and nurture relationships on the right.

This election, Dunne is unlikely to bring in any other MPs (we will all miss Judy Turner and Gordon Copeland). His poll ratings now compete with Jim Anderton’s Progressives. That, more than any public dalliance with National and protestations of ideological incompatibility with the centre-left, will count against him being involved in any Labour-led arrangement.

Speaking of polling, the dichotomy between the two polarised camps of published polls is striking. Calling it for the centre-right are TVNZ, Fairfax and Herald polls. Firmly in the red corner are TV3 and Roy Morgan.

It’s interesting to average the last six of the blue-trending polls. This predicts a centre-right landslide with National over 50 percent and Labour marooned on 34.5%. The Greens are on 6.7%, NZ First and the Maori Party on 2.5%, Act on 1.7% and United on 0.7%.

Doing the same for the red-trending polls see National fall to 44.3% and Labour rise to 36.7%. The Greens are on 8.4%, NZ First on 3.7%, and Act on 2.3%. The Maori Party is 2.4% and United 0.6%.

Which group is correct? Remember, TV3 and Roy Morgan called it right at the 2005 election, while TVNZ, Fairfax and the Herald were all miles out.

Of concern to the centre right is that, either way, we are likely to see four parties of consequence post election – Labour and National, the Greens and the Maori Party courtesy of its dominance of the Maori seats. The rest will be stragglers and independents.

A sign of National starting to worry will be scaremongering about the nature of a possible Labour-led coalition. Oh, that’s right, they started that last week.

Comments (6)

by Graeme Edgeler on October 28, 2008
Graeme Edgeler

at the 2005 election ... TVNZ, Fairfax and the Herald were all miles out.

I've truncated, and simplified a little, but in their last prelection poll TVNZ had National at a 95% chance of being getting 38-43% of the vote, and Labour a 95% chance of 35-41%.

That's not "miles out".

by Michael Appleton on October 29, 2008
Michael Appleton

Well, Labour beat National by 2% in 2005. Let us look at how close the polls got to predicting this:

TV3: Labour by 1.8%

Roy Morgan: Labour by 1.5%

Herald: Labour by 7.2%

TVNZ: National by 6%

Fairfax: National by 6%

So, if we're basing our trust in these polls predicting the Labour-National horse race on 2005 performance, we should be paying closest attention to TV3 and Roy Morgan. (TV3's poll was closest to predicting both Labour and National's party vote - Roy Morgan underestimated both.)

What about the minor parties?  TVNZ was closest to predicting NZ First's 5.7% and the Greens' 5.3%. The Herald was closest to predicting Act's 1.5% and United's 2.7%.

I just think it's a shame we don't have more regular polling data during the campaign. Perhaps I'm spoilt by the avalanche of data available daily in the US Presidential race - but wouldn't it be great to have a daily tracking poll here? I know it'd cost a lot - but maybe the major media organisations could pool their resources and share the results? Or is such collectivism pie-in-the-sky in this competiive media world?

by Tim Watkin on October 30, 2008
Tim Watkin


I've been having the same thoughts myself. It would be great to have the kind of polling the US offers, and yes TNS/TV3 did score best last time.

But does that mean they'll do best this time? They and RM are known to score Labour slightly better. In an election where Labour is the most popular party that lean probably captures the trend. But now that National is ahead in all the polls, will the National-leaning polls provide a better snapshot of the trend/mood? Or, are TNS and RM just better polls? I'd be interested in any answers to those questions, as I simply don't know the answers.

by Graeme Edgeler on October 30, 2008
Graeme Edgeler

An apology. In looking at the pre-election TVNZ poll result in my comment below, I relied on Wikipedia's table of 2005 pre-election polls. It appears to be inaccurate in respect of the TVNZ poll.

by Hardcentre on October 30, 2008

An interesting analysis of Dunne's chances. In 2002 it was the taming of the worm that brought him to a wider popularity with the Country (and 7 more MP's). Since then there has been little opportunity for any of the less polarizing Leaders to shine. Perhaps he should persuade Sainsbury to interview him in the Green Parrot - or at least get on Dancing for the Stars.

by Michael Appleton on October 30, 2008
Michael Appleton

Well, for what it's worth, Colmar Brunton pretty much nailed both the 1999 and 2002 election results (they predicted Labour +8 and Labour +23; the results were actually Labour +8 and Labour +23). So, their inexactness seems to be a recent phenomenon.

I suspect (but this is complete speculation) that their 2005 issues related to: 1) How late the movement to Labour was; 2) A possible failure to tweak their metodology for the increasing number of (relatively centre-left) voters who are only reachable by cellphones.

No matter - whatever happens, we're going to have some pretty embarrassed pollsters on Nov 9, unless we see some convergence to a mean over the next week and a half...

Post new comment

You must be logged in to post a comment.