We've got the latest polls all mixed up together and come up with some thoughts on Winston Peters and a bunch of questions for you to discuss. So off you go...

You'll notice on the left hand side of the homepage that, after the latest round of polls, we've updated Pundit's own poll of polls, which track the country's five biggest surveys. And it's noteworthy to see the trend lines have turned a smidgen.

In the first part of the year the slight swing was towards the left, with a sense amongst many that holes were appearing in National after years of impenetrability. That stopped after a reassuring Budget and swung back to National. But now the upward line for the Nats has turned flat, which offers some relief for both Labour and the Greens. After being neck-and-neck with National they had seemed to lose their heads – Labour with internal efforts to destabilise David Shearer and the Greens seeming to over-reach and lose the discipline of recent years.

Both parties toyed with unpopular policy ideas – the "man ban" and quantitative easing. Both dumped them when it was clear the political price was too high.

Instead, they've found surer ground in the housing market – around foreign ownership – and arguing about spies and government intrusion.

The question now is that, with the trend towards National having stilled for a moment, which way the line will turn next.

What's clear from the numbers we see in this data – and what's been clear for some months now – is that Labour needs New Zealand First. National could get by with either a coalition or Peters on the cross-benches. It might yet be able to continue with the arrangement it has now. But if Labour hopes to lead a government in 2014 it will need New Zealand First alongside the Greens.

So when they're angsting about leadership they might want to ask Winston who he prefers, because it could be crucial. What motivates Peters will be a vital question heading into next year's campaign. The New zealand First leader's personal dislike of Key seems genuine and Key's personal rejection andcondescension for two straight elections can't be ignored.

But might Peters swallow that pride for the sake of, well... of what? For the sake of a continued National government?

There are two lines of logic around Peters, what floats his boat and what he may decide. First, if Peters' policies mean anything to him, he has to go with Labour and the Greens. Labour's policy barring foreigners from buying homes here was as much about coalitions as houses and voters. Labour, the Greens and New Zealand First are now united on most things Peters has most wanted, such Reserve Bank reform, a lower dollar, and a local manufacturing boost. He could even get compulsory super.

On that same line, how could he join a government that still contains Peter Dunne and/or John Banks, who he has so eagerly sought to destroy?

And finally, why would he join a fading political force in its third and – judging by today's polls and history – final term? Would he want to climb on board a sinking ship? Or could he even be used as an excuse for an early election, given this is the man who has a reputation for never retaining his ministerial status for an entire government?

The other line of logic, however, is to ask why he'd choose to be third wheel in a coalition when he could be second? Would his ego allow him to play third wheel behind Russel Norman? Or could Labour come up with a post that satisfies 'the look'? A lot might ride on just what job Peters wants and which party can give it to him.

And what about Shearer? Would Peters side with the Labour leader if Shearer doesn't look stronger by election time and hasn't performed well in the campaign and debates which aren't expected to be a Shearer strength? Would he risk a one-term government and a leader who has such competition within his own caucus?

Perhaps most crucially of all, how could that most astute of political minds back a party with 30-something percent support over a party with 40-something percent? Of course as I've written before it would be entirely right and proper, but the public backlash would be immense. And with voters so dubious of such a coalition, would he be willing to carry much of the blame? It could become a defining mark on his party.

The questions around the other minor parties remain unchanged. Can ACT and United Future find a path to survival? John Key was at an ACT meeting recently showing solidarity and Dunne has been careful not to burn off National altogether even as his partner rode roughshod over his privacy. But ultimately both of their future's depend on National, and any decision by National will depend on the polls.

Could the Maori Party yet be a kingmaker with even two seats? That's depend on the left's ability to find those missing few percent.

And finally there remains the Conservatives. Can they crack five percent? Can they take enough from National and New Zealand First in the next year to be ready to launch in the campaign (where their money will count)? Will National rue not lowering the threshold to four percent? Or will a cup of tea in Rodney suffice?

These are all pivotal questions worth remembering over the next 18 months. But as is so often the case, the biggest and most vital questions still centre around Winston Peters and exactly what he wants.


Comments (28)

by Tim Watkin on August 06, 2013
Tim Watkin

On the Key-Peters relationship, consider this one example from a speech Peters gave this week at Vic:

"The predicament Mr Key now finds himself in is that no one actually believes him.

Many would like to trust Mr Key on this but they’re having difficulty in understanding how a lengthy list of who’s who in our Government’s agencies knew what was happening yet Mr Key didn’t.

Those experienced in government are simply bewildered when they’re asked to believe that the only person that didn’t know was the one person with the greatest ministerial responsibility."

And that is the real reason for the Government’s overhaul of the spy laws. It isn’t about clarifying the role of the GCSB. It isn’t about terrorism.

John Key’s aim here is not the security and defence of the realm… It is self-preservation.

This is a classic cover up. It has all the hallmarks:

i)             Denials

ii)            Stone-walling

iii)           Misinformation

iv)           Obfuscation

v)            Scapegoating

John Key and his advisors are following, almost word for word, the Watergate script.

Just as Richard Nixon did, John Key is trying to hide behind a wall of “plausible deniability”.

Our message to him is simply this: Mr Key, you can run but you can’t hide. Sooner rather than later the truth is going to catch up with you.


He's comparing Key to Nixon, politics' most notorious liar. And he's essentially calling Key a liar as well. How can he then form a government with him?

by Andrew Geddis on August 06, 2013
Andrew Geddis

But, of course, Peters' comments have to be viewed in context. In particular, his past form in denigrating his opponents, only to join them in power when the opportunity arises. See the below list of pre-1996 quotes:

Jim Anderton: Is the member going into a coalition with National?

Winston Peters: Oh no we are not.” – Parliamentary Hansards, P14147, 20 August 1996.

There is only one party that can beat National in this election that that is New Zealand First.” – Winston Peters, 69 & 85 minutes into First Holmes Leaders Debate, TVNZ, 10 September 1996.

Of course I am not keen on National. Who is? … This is a government bereft of economic and social performance  [so] that they are now arguing for stability.” – Winston Peters, Evening Post, 25 June 1996.

The prospects are that National will not win this election, that they will not form part of any post-election coalition.” – Winston Peters, The Dominion, 5 October 1996.

It is clear that this National government will use every means at its disposal to secure power… Come October 12…  Two months ago I warned that the National Party would use every trick and device at their command to to retain their Treasury seats.” – Winston Peters speech to Invercargill Grey Power, 26 August 1996.

The Prime Minister [Jim Bolger] is not fit for the job and come 12 October he will be out. He should not get on his phone and call me like he did last time, because we are not interested in political, quisling  behaviour. We are not into State treachery.” – Winston Peters, Parliamentary Hansards, P14146, 20 August 1996.

We believe the kind of politician depicted by Bolger, Birch, and Shipley is not to be promoted into Cabinet. As a consequence we will not have any truck with these three people.” – Winston Peters, NZ Herald, 22 July 1996.

We are a party that says what we mean and mean what we say, regardless of the political consequences.” – Winston Peters, Speech to public meeting, 9 October 1996.

by Stephen on August 06, 2013

He has been known to swallow large dead rats before to gain office. Jim Bolger comes to mind.

by Matthew Percival on August 06, 2013
Matthew Percival

The 2014-2017 term is likely to be Winston's last presuming he runs again in 2014. He will be 72 by the end of 2017 and although you wouldn't totally count out Winston running again at age 72 I'd consider it unlikely.

We also don't know what policy platform NZ First will take into the 2014 election and what issues Winston would make a bottom line in co-alition talks.

I wouldn't be making any predictions just yet, with Winston anything can happen.

by stuart munro on August 06, 2013
stuart munro

The position of small parties can force them into somewhat invidious compromises, and Peters may be reluctant to commit to a partnership that will alienate his 'anyone but.. voters' on one side. Nevertheless, both Peter's own and the public's distaste for espiocracy may lead him to commit somewhat earlier.

I wouldn't put too much trust in the polls at present, phone polls are losing validity as younger people opt out of landlines and the poor can no longer routinely afford them. Polling a thousand people only needs a pretty low level of bias to produce utterly wrong results - twenty or twenty five would be enough.

Personal relations between Shearer/Peters/Norman are probably more defining, but I venture, quite promising. Peters likes cleverness, and is apalled at Key's blind and inept use of GCSB powers. (Had he had those powers himself he would have been either very scrupulous, or very discreet.)That makes it easy for him to respect Norman, who is also sharper than the average MP. Shearer is not a towering ego, as long as he can keep up and doesn't earn Winston's contempt, they can work together. Importantly, Labour seemed to have finally learned not to trash the Greens.

ACT will not poll over the margin of error, but strategic Gnat voters may continue to support Banks in the court jester role, if no fresh scandal arises. Dunne will also struggle, particularly after the details of his emails makes their way, by osmosis, onto the interwebs. There is a possibility that his wife, or Andrea Vance might stab or Bobbit him. The Maori party will continue to assert their alliance with Key was valuable, and probably pick up 1-2 seats, but there will likely be a migration away from the Maori roll.

NZ remains wide open for a lib-dem type party to do what was hoped for by the initial supporters of NZ First and United Future. As such a party fails to materialise, support will accrete around the best alternatives - the Greens, Labour, etc. Mana's big chance is to capture that block - it could get them10-12 seats - but to date they aren't getting the airtime that would make that likely. Minto's mayoral bid was astute, they need more things like that.

by Tim Watkin on August 06, 2013
Tim Watkin

Matthew, I agree that predicting Peters is daft. You'll note I carefully asked questions rather than fall into that trap! But disagree about policies - we know what Peters stands for, he's not going to change.

But the age point is an important one. What will Peters do with his last shot? What does his own legacy and his party's survival mean to him?

by Stephen on August 06, 2013

Who would blink first over the asset sales policy? Or are they both craven enough to come to an "arrangement"?


by BeShakey on August 06, 2013

"the public backlash would be immense"

Is there much to support this? Key seemed keen to suggest this would be the case, but I wondered whether that was more about trying to make it so. The Electoral Commission's polling suggests NZers have a pretty good understanding of how MMP works, and I really haven't seen much strong evidence either way (which makes me think the reaction wouldn't be immense if people aren't wound up about the very real possibility now).

by Tim Watkin on August 06, 2013
Tim Watkin

Shakey, to quote from 2008 polling: "...nearly 80% of people think that the party with the most votes should be the one that gets to lead the government, with just 15% disagreeing."

Other polls have been similar. For me – and I've talked to a fair few political scientists and politicians about this and I can't remember anyone disagreeing – this is a blind spot for NZers. We're still very FFP on this point – it's just never happened before and I think it would rattle a lot of cages. Can't be sure until it happens of course, but would be (pleasantly) stunned if otherwise.


by Tim Watkin on August 06, 2013
Tim Watkin

Andrew, great list of quotes! Wonder if Key will work as hard to win him over.

by Andrew Geddis on August 06, 2013
Andrew Geddis

Andrew, great list of quotes!

A belated hat-tip to Frank Macskasy for it ... sorry for not linking earlier.

by Frank Macskasy on August 07, 2013
Frank Macskasy

@ Andrew


Let's hope Winston doesn't make the same mistake he made on 11 December 1996.

Which ended in tears, and this.

As for Peter Dunne, had 1,775 voters who gave their ELECTORATE votes to Gareth Hughes (a pointless exercise) voted for Charles Chauvel, Dunne would've lost his seat by 129 votes.

And history, as they say would've been different; no asset sales; no scandal after scandal; no Charter Schools; no GCSB/TICS Bills, etc, etc, etc...

Hopefully, voters in Ohariu will be a bit more canny next time...

by James Green on August 07, 2013
James Green

Stuart: I wouldn't put too much trust in the polls at present, phone polls are losing validity as younger people opt out of landlines and the poor can no longer routinely afford them. Polling a thousand people only needs a pretty low level of bias to produce utterly wrong results - twenty or twenty five would be enough.

This point gets trotted out a lot, but I'm not sure that it's the real problem (even as I am considering abandoning my landline later this year).  There are two reasons for this, (1) that the people that don't have landlines are also disproportionately less likely to vote, and (2) the level of people responding with with hangups/f*** offs/call screening with caller ID is ranging 60-90%. In my recent experience (research rather than polling), it was toward the high end of that band, and the people that did talk to us were overwelming female (~70%) and older. I've seen a very few other surveys where they have been a bit transparent about their response rates, and it really is stunningly bad. At least for now, this is still a bigger threat to the validity of polling than a shift away from landlines.

by Frank Macskasy on August 07, 2013
Frank Macskasy

@ James - Roy Morgan calls both landlines and cellphones.

by stuart munro on August 07, 2013
stuart munro

Maybe it's time polling went to something a bit more objective: some of these eye tracking response systems might give quite interesting data on politicians - and being subconsciously determined contain less of some kinds of bias.

by James Green on August 07, 2013
James Green

@Frank - Thanks, that's interesting, but they still don't mention their refusal rate. So perhaps they might capture 50-100 people on mobiles that they would have missed, but they could easily have 1200-3000 people who would not answer their questions.

I note that in Australia Roy Morgan conduct face-to-face polls to help get around this problem (eg). The difference ismost prounounced for the liberal party - with liberal party supporters seemingly 20% more likely to respond in person (leading to 4-10% increases in their apparent performance).

by Tim Watkin on August 08, 2013
Tim Watkin

Re the polls, I think most if not all the pollsters phone mobiles these days. And the proof of the pudding is in the results, not the methods anyway. If you look at the polls around elections, they haven't been significantly out of kilter with the final results.

Also, don't forget that polls show trends. So even if they might be out a little here and there in any one poll, the trend is what you're trying to guage. All the better when you bring the main polls together as we do here and get an overview.

by James Green on August 08, 2013
James Green

Also, don't forget that polls show trends. So even if they might be out a little here and there in any one poll, the trend is what you're trying to guage.

I entirely agree with this broader point, but the 'smidgen' that the polls have turned is no bigger than the average noise twitch that occurs 5-10 times a year. For now, I think, the trends are still well intact.

by Tim Watkin on August 08, 2013
Tim Watkin

I'm not sure which trend? This year we started with a swing to the left that had Labour/Greens neck-and-neck with National. Since the Budget we've swung back and National could just about govern alone. See the left-right gap line on our polls page.

Are you saying the swing back to National is still going?

by Andrew Robertson on August 08, 2013
Andrew Robertson

Darn! I haven't checked this blog in a wee while and I completely missed this discussion.

So at my blog I've put together a grid of the different methodologies about the public polls. It's here: http://grumpollie.wordpress.com/nz-public-poll-methods-grid/

There is a bit of guess work - it's a work in progress.

I'm also planning a blog post about why polling companies don't report response rates with their polls. There are other post there about landline vs cell phone polling, whether a response rates is an indicator or sample quality, etc. 

by Andrew Robertson on August 08, 2013
Andrew Robertson

To add to the discussion though - I do think a telephone survey with very good fieldwork practices can achieve response rates in excess of 30%. I agree though - I think non-reponse is one of the biggest issues with telephone polls.

Something to keep in mind is though it is that response rates aren't even very relevant to some methodolgies. Quota surveys, for example, aren't designed to maximise responde rates. They're designed to fill quotas.


by Alan Johnstone on August 08, 2013
Alan Johnstone

In the end Winston will do whatever he wants. Who knows what his price may be, but it'll be paid by someone.

In terms of polling, it's all very static and has been since 2008. The right wing holds a very small lead over the left wing bloc. I sense poeople are very fixed in their views, this country contains very few floating voters and elections are won and lost on turnout.

The national party governs today because labour hass been unable to get it's vote out in south and west auckland like it did in 2005. Everything else is fluff. If labour mobilises well it wins, if it doesn't it goes down to defeat.

If all labours efforts aren't focused here I'd be surprised.


by Tim Watkin on August 08, 2013
Tim Watkin

Alan, I think that broadly you have a point, but you over simplify. The polls have not moved dramatically since 2008, but they've moved more than you allow – and the movement within the left and right blocs matters under MMP and shouldn't be dismissed. New Zealand First, for example, was nowhere in 2008 and the Greens half what they are now.

Yes Labour's turn out is vital, but if you think simply motivating south and west Auckland is enough when they've been wiped out in the provinces, have leadership woes and a lack of fresh talent, I'd disagree. Those issues are more than fluff. What's more "the National Party governs today" less because of Labour's turnout machine and more because of Key's political skills.

by Tim Watkin on August 08, 2013
Tim Watkin

Andrew thanks for popping in with that. James says he wouldn't put much trust in the polls these days. You?

And if you want to submit that post you're planning as a Your Punt (as well as on your site, of course), I'd be interested in sharing it with Pundit folk and happy to link back to yours.

by James Green on August 09, 2013
James Green

Are you saying the swing back to National is still going?

Yes. Or at least I don't think that the apparent 'signal' currently exceeds the typical range of 'noise'. Though picking trend breaks is always easier in hindsight. But for example, what is now quite obviously a swing back to National on the left-right gap, has probably only truly become apparent in the last 2-3 months. And if you look at the gentle slope back to the left* (which I'd argue has been apparent ever since the election), there are plenty of twitches in that line that are easily the same size or greater than the current one. Having said that, as a researcher, I should probably have a more conservative interpretation than you.

*The particular approach I've drawn in here is more like the type of trend interpretation used by a certain class of mathematically oriented share traders (so called 'technical analysis'), but I think my conclusions would be robust if I was to apply a more social science oriented trend analysis.

by James Green on August 09, 2013
James Green

James says he wouldn't put much trust in the polls these days.

I wouldn't go that far. I was just trying to say that non-reponse is a bigger threat to validity than the non-coverage of those without landlines. But as you noted above, at least up until the last election, the opinion poll results are still mapping well onto the electrion results, which is the best (only?) test of their validity that we have.

If I was a polling company (which I'm not), I think I would be starting to experiment with collecting some additional data but not including it in my publicly released polls. This would mean that if the election result did differ from my current polling strategy, I'd have a headstart on updating it, by checking the additional data against the election.

Andrew - Thanks for that grid, it's an excellent resource.

by Andrew Robertson on August 09, 2013
Andrew Robertson

Tim - Thanks, I'll let you know when it's ready to post. Not sure I should answer your question about trusting polls. I'm hardly neutral on the topic.

James - thank you. I complete agree with you. I think non-response is a bigger problem that non-coverage.

by Alan Johnstone on August 09, 2013
Alan Johnstone

Tim, of course I over simplfy, it's hard not to in 3 sentences.

However in 2005 the turnout was 80.92%, in 2011 it was 74.21%. I'm convinced that the missing voters aren't people that voted National.

Let's look at the 3 safe south Auckland Labour "M" seats.

In 2011 despite having broadly similar or high vote percentages than in 2005 labour only got 50,545 party votes. In 2005 it got 57,254. If I had time I could do the same stats for other urban labour voting areas and get the same picture, i'm sure I could find 50,000+ "missing" labour voters that stayed home.This is the difference between power and opposition.

Movements between Labour and the Greens don't really matter.




Post new comment

You must be logged in to post a comment.