The Dim-Post

November 14, 2013

I just CANNOT let this go by

Filed under: Politics — danylmc @ 10:08 am

I’m supposed to be working on my book today, but Massey Uni Political Scientist Claire Robinson has written something in the Herald about politics:

If recent history is anything to go by, the 2014 general election result has already been decided.

Despite the current centre-left Labour/Greens bloc looking competitive, history tells us National should have the 2014 election in the bag, again.

How is this possible when there is a lot of water to go under the bridge between now and the next election? The Labour Party has only just got a new leader and not a single cent of money has been spent on campaign material and advertisements by any political party. Surely voters will be waiting to see what tricks David Cunliffe can pull out of Labour’s hat before coming to a decision?

Well, it’s counter-intuitive, but election campaigns in New Zealand don’t actually make much difference to the outcome of elections for major parties (although they do for minor parties).

Data gathered from the New Zealand Election Study since 1999 shows that on average almost 54 per cent of voters will make their decision about which party to vote for before the election campaign.

Well that’s true. Lots of voters make their minds up before the election. When these voters are asked if they identify strongly with a particular party they’ll generally say they do. They’re mostly core voters. People who are farmers or members of Trade Unions, or who just have strong loyalty towards a particular party. But over 30% of people who voted during the 2011 election made their mind up during the campaign. 30%! 30 freaking %! The whole reason political parties spend huge amounts of money and energy campaigning during election campaigns is because a THIRD of the electorate makes up their mind during those campaigns. Campaigns are crucial! Campaigns make a HUGE difference! Campaigns!

And moving on . . .

David Cunliffe will need to convince National voters that his recent rekindling of Labour’s relationship with the union movement is also in their interests. It may have worked to shore up Cunliffe’s leadership ambitions, but persuading more conservative centre-right voters to swing to the left will not be such an easy ask . . . Conversely for Labour to grow they need to take votes off the Greens, which means that they can’t become too chummy either.
Here’s how the total voter turnout broke down in 2011:

2011votesAnd of that gigantic white block of non-voters, 30% voted for the Labour Party in 2008, and 45% have a favorable or very favorable view of the Labour Party. So if you’re a Labour strategist are you looking at core National voters who have already made their mind up to vote for the National Party? Claire Robinson thinks you are. And, hell, twelve months ago she was right, let’s give her that. But that’s not the only option. Nor is capturing Green voters. Enough people used to vote for Labour and didn’t vote last time to swing the election for them.

I don’t know who is going to win the election next year. It’s going to be decided at the margin and the margins are filled with uncertainty. How will Cunliffe perform during the actual campaign? Goff was dreadful and you can see that impact in the polls: Labour lost about 4% during the campaign period. And who predicted that the ‘teapot tapes’ thing would happen and that National would handle it so terribly, bleeding voters to New Zealand First and getting Winston Peters back into Parliament?

But I do know the election isn’t in the bag because the people who always vote for the same party have already made up their mind to vote for the same party.

About these ads

48 Comments »

  1. Some simple maths:

    Convincing a non-voter to vote Labour/Green: +1 for the centre-left.
    Convincing a National voter to vote Labour: +2 for the centre-left.
    Convincing a Green voter to vote Labour: +0 for the centre-left.

    The question is: Is it easier to convince two people to vote or one person to switch?

    Comment by pete — November 14, 2013 @ 10:14 am

  2. pete: Your simple maths fail because you cannot assume convincing someone to change their vote to support you will not also lead to someone who was already going vote you to stop voting for you e.g. if Labour suddenly said “we’ll sell all state assets”, they might gain some National/Act voters, but their current voter base will desert them.

    Comment by wtl — November 14, 2013 @ 10:24 am

  3. Convincing a non-voter to vote for your party does actually decrease the proportion of votes that all the other parties get. It’s a smart thing to do!

    Comment by danylmc — November 14, 2013 @ 10:28 am

  4. Hmmm … if history is anything to go by, then Labour have no chance: no two-term Labour opposition has ever won government. In fact, no two-term Labour opposition has ever even increase their vote from the previous election :-)

    oh wait http://xkcd.com/1122/

    Comment by Graeme Edgeler — November 14, 2013 @ 10:42 am

  5. Convincing a non-voter to vote for your party does actually decrease the proportion of votes that all the other parties get.

    It also decreases the value of all the other votes you have received…

    Comment by Graeme Edgeler — November 14, 2013 @ 10:44 am

  6. It also decreases the value of all the other votes you have received…

    Huh? Let’s imagine a voting group of 100 people. Of those 100, Party A has 30 votes (30%), Party B has 40 votes (40%). Then one more person gets added to the group, and supports Party A. What happens?

    Party A’s share of the vote (31 of 101) goes up to 30.7%.
    Party B’s share of the vote (40 of 101) goes down to 39.6%.

    Comment by Andrew Geddis — November 14, 2013 @ 10:58 am

  7. Claire Robinson is an ex-National party staffer and is also probably angling for a nice fat job on the “reformed” corporatised university councils that Steven Joyce wants, of course she is going to do her bit for the right by trying on some voter suppression via voter apathy (“No need to vote, the result is already known! Peasants, just accept your overlords!).

    But there is also seems to be a striking failure of imagination in most of our “mainstream’ (Robinson is regular on the telly) punditry at the moment, one is which they seem intellectually incapable of imagining a political environment that isn’t one where the two main parties offer powder pink and pastel blue variations on neo-liberal despotism tempered with populist reaction.

    Comment by Sanctuary — November 14, 2013 @ 11:20 am

  8. Huh? Let’s imagine a voting group of 100 people…

    I think that Graeme’s point is that each individual vote was worth 1% and then when you persuaded a new voter into the pool each individual vote is now only worth 0.99%.

    But it decreases the value of all votes in the pool (yours and opposing), by the same proportion, so it is always be a net win to get someone new to vote for you. Democracy!

    Comment by RJL — November 14, 2013 @ 11:38 am

  9. From memory Robinson was a Ministerial staffer, which is a bit different from being a filthy compromised political party staffer, like my wife. And I suspect this has more to do with innumeracy: Wow! 60% of voters have made their minds up! That’s a majority! The election is over!

    Comment by danylmc — November 14, 2013 @ 11:39 am

  10. In a Stuff bio of Cunliffe, Robinson was quite candid about her dislike of him, dating back to a time when they both worked for MFaT. For years I have regarded her as a biased political commentator, and not a particularly smart one, either,

    Comment by Laura — November 14, 2013 @ 11:49 am

  11. Wow! 60% of voters have made their minds up!

    Not even. According to her own figures, 40.4% of National voter made their minds up a year out from the election (which is where we are now).

    Comment by Andrew Geddis — November 14, 2013 @ 11:53 am

  12. It’s a little sad, because 8%(?) of people didn’t vote last time because they felt their vote wouldn’t make a difference. Stories where people say “the next election is already in the bag” isn’t really going to energise them back to voting.

    Comment by Auto_Immune — November 14, 2013 @ 12:05 pm

  13. I thought National’s strategy was to encourage voters to change their minds, so Colin Craig could prop them up.

    Of course Robinson is fundamentally wrong. National could poll around 45% and win a seat landslide, or around 45% and be well beaten. The minor parties and their thresholds aren’t a postscript, they determine the outcome. Clark could have won a fourth term in 2008, Key could have governed alone in 20011. Just shift a few NZ First votes around.

    Comment by sammy 2.0 — November 14, 2013 @ 12:06 pm

  14. I don’t get it – does she not understand numbers *at all*?

    I find that article just plain weird. Even if you’re a biased political commentator, can you really be that blind to how close things are at the moment?

    Comment by Andrew — November 14, 2013 @ 12:06 pm

  15. “in 20011″ … on Jupiter?

    Comment by sammy 2.0 — November 14, 2013 @ 12:07 pm

  16. Claire – I knew her in a former job quite well. She’s the most unimaginative plodder you could ever meet. Thinking outside the box is impossible for her because she so stuck in it that she reminds me of the joke about the fish asked what it thinks of the water and replies “What water?”

    Comment by Rhinocrates — November 14, 2013 @ 12:18 pm

  17. @wtl – agree, but I think it’s even worse than that: if Labour banks right to gain that swing vote, there’s a good chance they’ll alienate someone to their left and still not manage to convince the swing voter.

    Comment by pete — November 14, 2013 @ 12:41 pm

  18. “Campaigns are crucial! Campaigns make a HUGE difference! Campaigns!”

    Oh, so policy is irrelevant. Labour doesn’t need to release any policy, just run a good campaign. :)

    Comment by Ross — November 14, 2013 @ 12:48 pm

  19. Robinson’s piece is just like the other wierd bit of wisdom doing the rounds – that an alliance with Colin Craig would be good for National, as if all those liberal Nats/ex-ACT voters who currently vote for Key because they’ve been able to convince themselves that National and ACT are the still a viable liberal/neoliberal option won’t balk at the idea of worshipping alongside Colin Craig’s merry band of North Shore Taliban.

    Sometimes I get the impression these talking heads are like besieged generals in a bunker, out of touch with reality but still busily manoeuvering little political flags around an unchanging map of a fictional country.

    Comment by Sanctuary — November 14, 2013 @ 12:57 pm

  20. @pete – But there is at least a chance that the alienated left person may vote Green instead, which is still helpful to the left. I think there’s a bunch center-left voters who feel ignored right now, but that could be my own bias.

    Comment by Andrew — November 14, 2013 @ 1:00 pm

  21. I think if they were going to end up voting Green then they would have already (and it looks like a lot of them did that last time around). I’m sure the NZES would have something to say on this (I swear I’ll read it soon).

    Comment by pete — November 14, 2013 @ 1:04 pm

  22. Yeah you’re probably right. I’m just down in the dumps because there’s no party I want to vote for at the moment. In my imagination, there are others like me. It helps.

    Comment by Andrew — November 14, 2013 @ 1:14 pm

  23. You make the statement in this post that “of that bloc of non-voters, 30% voted for Labour in 2008″.
    I have seen this sort of statement made a number of times but I have never seen the evidence given for such a claim.
    How can you possibly know who did not vote, assuming you don’t have access to all of the voting records, and even if you did, how can you possibly know how they voted in the previous election?

    Comment by Alwyn — November 14, 2013 @ 1:29 pm

  24. I am always confused by the assumption that the vast majority of non voters, if forced to, would vote on the left.

    Is this because people who don’t vote are considered to be morons and thus more likely to find a home on the left?

    Comment by King Kong — November 14, 2013 @ 2:05 pm

  25. How can you possibly know who did not vote, assuming you don’t have access to all of the voting records, and even if you did, how can you possibly know how they voted in the previous election?

    Because that’s what the NZES data tells us – they interviewed a whole heap of people, some of whom voted (and so they asked them who for and why); some of whom didn’t (and so they asked them why not, and if they voted last time, who for).

    Not that this is the same data that Claire Robinson relies on in her article. So if it is rubbish … then so is her whole analysis.

    Comment by Andrew Geddis — November 14, 2013 @ 2:27 pm

  26. I think you’re missing that her major argument for the inevitability of National’s victory is that the party leading a year and a bit ahead of the election has gone on to win (sample size = 5), which is your run-of-the-mill bad pundit argument rather than an outrageously stupid one. The 54 or 40 percent of voters make up their minds early bit is just an explanation for why this occurs (unconvincing and badly expressed, but it’s not like she’s a communications expert or anything), not the reason for her prediction. The real problem is the one in the xkcd comic: she’s making predictions based on arbitrary and limited data, while ignoring other highly pertinent information — for instance, that it’s currently very close between LAB + GRE and NAT + whoever, so it’s much more likely the pattern will be broken this time than it was in the Bad for Phil Goff era.

    Comment by bradluen — November 14, 2013 @ 2:30 pm

  27. bradleun: My favourite example of that type of effectively useless statistic is the sports commentators favourite ‘So far this season, team XXX has won every time they lead at half time’. It’s a combination of stating the obvious (well duh, the team which has more points at any given time is always going to be more likely to go on to win) and ignoring the fact that it has limited predictive value (like the xkcd example).

    Comment by wtl — November 14, 2013 @ 2:42 pm

  28. @Auto_Immune,

    It’s a little sad, because 8%(?) of people didn’t vote last time because they felt their vote wouldn’t make a difference.

    And they’re right to think that. As individual voters, our choice doesn’t matter AT ALL, unless we live in a small handful of close-run electorates held by minor parties.

    Comment by Phil — November 14, 2013 @ 2:47 pm

  29. Phil – that’s a bit of an odd statement.
    By extension you could probably argue that 100% of individual voters ballots don’t matter at all unless they happen to live in a tightly contested electorate.

    Comment by Gregor W — November 14, 2013 @ 3:36 pm

  30. Ah, misread. I see that’s what you actually are saying. How do you figure that, particularly wrt party as opposed to electorate voting patterns?

    Comment by Gregor W — November 14, 2013 @ 3:39 pm

  31. I agree that clearly, Labour’s best strategy is to motivate those non-voters to vote. But it seems like the big media narrative at the moment is “Labour needs to appeal more to people who vote National”. Given that this is the context that all of Labour’s policies are being announced, it might be in their interests to announce more ‘centrist’ policies (whatever that means).

    Comment by Matthew — November 14, 2013 @ 3:53 pm

  32. I just had a look at the New Zealand Election Study data. I’m not that familiar with it. From what I can tell, of those who did not vote in 2011, 38% liked the National Party the most on Election Day, and 25% liked the Labour Party the most. I applied the weight to the analysis.

    I’ll keep looking for other interesting variables. The sample size, without weighting was 386 for those who did not vote. Make of that what you will.

    Comment by Andrew — November 14, 2013 @ 5:15 pm

  33. Here’s another interesting one – of those who did not vote in 2011 46% wanted National to win and 32% wanted Labour to win. Respondents could choose Labour, National, Neither, or Don’t know.

    Comment by Andrew — November 14, 2013 @ 5:20 pm

  34. Haven’t read the article yet. Put it off until I was somewhere I could swear.

    She was the lady that said this on Q &A:

    http://www.scoop.co.nz/stories/PO1309/S00260/qa-discussion-in-response-cunliffe-parker-interview.htm

    CLAIRE”Absolutely. Annette King, Trevor Mallard, Phil Goff – you know, they’re the soul of the Labour Party, and so it’ll be really interesting to see how they’re—”

    and also this seems to be the same Massey University which has Dr Damien Rogers, as has been discussed Fran Mold’s partner, ex-NZ spy agency staff, who seems to also have plenty of opinions about Labour that he likes having published in the Herald, something that seems to have started particularly after Shearer lost the leadership.

    And I now know from reading above that she was an ex-Nat staffer and I had forgotten her take on Cunliffe at MFAT.

    So yes, not exactly an independent impartial commentator.

    Comment by sheesh — November 14, 2013 @ 6:07 pm

  35. Also seems to be a pundit, but not a pol scientist, but something to do with creative arts?

    Comment by sheesh — November 14, 2013 @ 6:10 pm

  36. “Respondents could choose Labour, National, Neither, or Don’t know.”

    What’s the point in them asking a FPP question about an MMP election?

    Comment by Sacha — November 14, 2013 @ 7:40 pm

  37. The study has been carried out for many years. Who knows? Comparability over time maybe. They ask for preference toward any party too. I’ve only really just started looking at it.

    It’s nice to have data to analyse/report/discuss that’s not confidential to clients.

    Comment by Andrew — November 14, 2013 @ 7:55 pm

  38. Thank goodness for raw data. Sadly, the 2011 analysis all seems to be dead links to Vowles’s own website in the UK.

    Comment by Sacha — November 14, 2013 @ 8:08 pm

  39. Exit polls are illegal in NZ (sadly), so the best source of data in other democracies is not available here.

    The NZES data is problematic, at best. Asking people who didn’t vote, long after the fact (that they didn’t), who they might have voted for if they had, does seem remarkably useless data. Obvious bias towards the subsequent winner, and obvious problem with value of response, given that they didn’t vote even when surrounded by media coverage, and are being asked to recall something that they didn’t actually do.

    Comment by sammy 2.0 — November 14, 2013 @ 8:14 pm

  40. No disagreement here. All survey data are problematic (that’s why I drew attention to the sample size for non-voters). That doesn’t means you can’t get useful information from it.

    There are a lot of assumptions about what a higher voter turnout would mean for an election result. They are big assumptions.

    Comment by Andrew — November 14, 2013 @ 8:22 pm

  41. Sheesh @ 35 – when I knew her she was a graphic designer.

    Comment by Rhinocrates — November 15, 2013 @ 12:57 pm

  42. Sheesh, i thought she was a Departmental rep staffer in a Nat Minister’s office. That’s a different category altogether.

    Comment by Tinakori — November 15, 2013 @ 5:54 pm

  43. The truth of NZ elections seems to be that National wins if too many voters stay home. When more people vote, National does less well.

    Comment by Steve (@nza1) — November 16, 2013 @ 12:29 pm

  44. The truth of NZ elections seems to be that National wins if too many voters stay home. When more people vote, National does less well.

    The turnout at the 2002 election was 77.0%. National did not do well.

    Comment by Andrew Geddis — November 17, 2013 @ 8:27 am

  45. Perhaps it is simply that fewer voters turn out if they think the election result won’t be a close-run thing.

    Comment by Adze — November 17, 2013 @ 9:58 am

  46. Perhaps it rains.

    Comment by Andrew — November 17, 2013 @ 12:41 pm

  47. Claire Robinson is a funny sort of political scientist. By my count from a bit of Google Scholar’ing she seems to have published a grand total of *one* peer-reviewed journal article (excluding book reviews). And yet she is a Professor (the highest possible academic rank at Massey), and is a Pro Vice Chancellor in the bargain. Did she have a name change or something that I’m missing?

    Comment by AcademicPeon — November 18, 2013 @ 4:36 pm

  48. The Professor also tweets …

    Googling “Simon Cunliffe” would only have overloaded her with information, so she didn’t (but she’s definitely not a Nat-hack, no sir, just a regular commentator, here to help us understand …).

    Comment by sammy 2.0 — November 18, 2013 @ 7:05 pm


RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

The Rubric Theme. Create a free website or blog at WordPress.com.

Follow

Get every new post delivered to your Inbox.

Join 388 other followers

%d bloggers like this: