Digital campaigning in New Zealand

There are few people who would doubt that digital is becoming an increasingly important component of political campaigns. Sadly I can no longer find the source for this graph, but it’s one that I’ve shown a few times recently to convey that both sides of politics are embracing digital as an increasingly large chunk of their campaign expenditure (in the US at least).

Screen Shot 2015-02-27 at 2.37.14 pm

One of the quaint things about election campaigns in New Zealand is the disclosure of electoral expenses, in order to enforce their spending caps. This allows us to very accurately examine what political parties are spending, and what they are spending it on.

Now that the 2014 election expenses have been released I’ve been able to compile what the various parties spent online, and there are some interesting numbers… (data here if you can’t see the full table)

Social Media Websites Display Ads Misc Digital Total Digital Spend Total Spend % Digital
National $38,214.39 $27,791.44 $393,772.70 $459,778.53 $2,558,211.53 18.0%
Labour $22,353.13  $1,236.30  $85,445.91 $109,033.34 $1,269,298.91 8.6%
Greens $33,099.63 $33,804.42 $52,433.96 $119,338.01 $1,291,419.81 9.2%
NZ First $1,518.00 $1,518.00 $268,530.23 0.6%
Maori $6,621.70 $6,621.70 $202,562.12 3.3%
ACT $30,704.66 $30,704.66 $294,406.09 10.4%
Conservative $5,292.00 $5,292.00 $1,914,072.38 0.3%
Internet Mana $230.00 $40,966.58 $19,237.20 $310.50 $60,744.28 $660,574.07 9.2%

While I wasn’t surprised to see that both the Greens and Internet Mana spent a large proportion of their budget online (at 9.2% it’s a higher proportion than the 2012 Obama campaign) – to be honest I was surprised the Internet Mana party didn’t spend more. The only other minor party that made a substantial investment was ACT – presumably getting David Seymour’s video far and wide.

What is surprising is both the huge proportion of their budget that National spent online (18% is much higher than I’ve seen in any other political campaign). As well as investing reasonably heavily in social media and a rather good looking website, they totally saturated online display advertising. It looks like they totally dominated in this space.

By contrast not only did Labour spend a tiny amount of money, but even the proportion of their total spend on digital was small.

This led me to wonder what this looked like compared to the 2011 election. So I dug out those numbers and compared…

Total Digital Spend 2011 Total Spend 2011 % Digital 2011 Change between 2014 and 2011
National $118,078.73 $2,321,216.06 5.1% +12.9%
Labour $98,790.34 $1,789,151.95 5.5% +3.1%
Greens $95,596.93 $779,618.38 12.3% -3.0%
NZ First $558.90 $155,902.86 0.4% +0.2%
Maori $- $72,172.56 0.0% +3.3%
ACT $41,347.10 $617,035.18 6.7% +3.7%
Conservative $5,300.55 $1,878,337.22 0.3% 0%
Internet Mana $500.00 $60,082.91 0.8% +8.4%

Surprisingly the Greens dropped their digital spend as a proportion of their total campaign budget, but I imagine this is probably because their total campaign budget was so much larger this time, and much of their additional resource was allocated to TV. They still spent almost three times as much online in 2014 as they did in 2011.

While National outspent Labour and the Greens on digital in 2011, they dramatically increased their spend in 2014. To be honest, I’m really taken back by it. You don’t think of the Nats as being on the cutting edge of campaigning, but I’ve never seen a political party invest so heavily in digital media before. Let’s hope the new look Labour team decide to do things a bit differently.


This data has been hand coded from the election expense results. There will be data entry errors on my part.

I’ve broken the 2014 data down into three sub-categories of digital spend: social media, websites and digital display advertising. Apologies for not doing this with the 2011 data.

I’ve compared the 2014 Internet Mana party with the 2011 Mana Party – very different beasts.

If you have trouble viewing this data on my blog, you can see the spreadsheet.


I’ve been contacted by Labour to point out some of the digital spending in their expense that I missed, I have updated accordingly and removed a paragraph from my post that is now inaccurate with the new figures. If anyone else spots mistakes or omissions I’m more than happy to make corrections.


How much did each vote cost?

The electoral commission has just realeased the party vote expense returns. Well worth a read. The stand out figure is the massive $1.7m that the Conservative Party spent. I would be very interested to know how much of that came from Colin Craig.

I’ve matched the expenditure against the number of party votes each party received to figure out how much each vote cost. The big losers at the Conservatives ($31.78 per vote) and ACT ($25.83). At those rates, it would have cost the Conservative Party $3,579,006 and ACT $2,915,328 to reach the 5% threashold.

At the other end of the spectrum you have the Aotearoa Leagalise Cannabis Party who spent $4,003 and managed to get over 11 thousand votes – achived largely off the back of their party name rather than their advertising efforts.

Party Name Cost Per Party Vote
Conservative Party  $31.71
ACT  $25.83
Democrats for Social Credit  $20.23
Greens  $3.15
Labour  $2.91
Mana  $2.49
Māori Party  $2.26
National  $2.19
United Future  $2.06
Alliance  $1.99
Libertarianz  $1.73
New Zealand First  $1.06
Aotearoa Legalise Cannabis Party  $0.34

Al Gore and the teapot tapes

I’ve begun reading Al Gore’s book The Assault On Reason, and I highly recommend it. This passage jumped out at me:

It is simply no longer possible to ignore the strangeness of our public discourse. I know I am not alone in feeling that something has gone fundamentally wrong. In 2001, I had hoped it was an aberration when polls showed that three-quarters of Americans believed that Saddam Hussein was responsible for attacking us on September 11. More than five years later, however, nearly half of the American public still believes Saddam was connected to the attack.

At first I thought the exhaustive, nonstop coverage of the O.J Simpson trial was just an unfortunate excess – an unwelcome departure from the normal good sense and judgment of our television news media. Now we know that it was merely an early example of a new pattern of serial obsessions that periodically take over the airwaves for weeks at a time.

As I was reading this, one event immediately sprung to mind – the teapot tapes “scandal” that engulfed the final weeks of the 2011 campaign.

People on all sides have acknowledged that it was something of a non-event. It was no Watergate. But still, it took up almost all of the political coverage during the most critical part of the election campaign. Was it simply easier for the media to latch onto a hard and fast “event” like this? Either way, the political discourse in our country suffers when things like this take disproportionate coverage.

It is a long standing bugbear of the beltway to bemoan the lack of quality news. Personally, I think that avenues such as the internet are going to have to play an increasing role. I’m looking forward to reading what Al Gore has to say at it.


Early votes – the numbers

In the second part of my look into early votes I’ve taken a thorough dig through the election results. There was a fair bit of copy and pasting involved, so apologies if I’ve made any mistakes. All figures in this post refer to the party vote. If you’re so keen you’d like to see the dataset I’ve used, feel free to email me and I’ll send you a copy.

Firstly, I’ve taken a look at the number of early votes cast as a percentage of the total number of votes cast.

While there is a little variation at either end, the early vote in most electorates sits comfortably between 10-14% of the total votes cast.

It is noticeable that Christchurch electorates are very well represented (Christchurch East 2, Christchurch Central 5, Waimakariri 8, Port Hills 10 and Wigram at 11) at the top end of the scale. I’m not sure if this is because Cantabrians voted early due to uncertainty about polling booth locations, because of some extra effort on the part of campaigns or the Electoral Commission, or some other factor. I’d welcome suggestions as to what might have happened.

And then I’ve also taken a look at how Labour’s party vote in early votes compared to their share of the rest of the votes (votes cast on the day, and special votes).

Again, while there are some big outliers, particularly Te Tai Tokerau and Tamaki at the bottom end, there is not actually a great deal of variation in the numbers here.

Of the 25 electorates where Labour’s share of the early votes was higher than their share of the other votes (or exactly the same in the case of Ohariu), they won 12, with several others being very close.

With a total of 287,113 early votes cast (or 12.73% of the total vote), it is clear that this is going to be a very important part of the electorate. If more and more people are going to be casting early votes, it is going to be important that campaigns do not rely on only communicating with them in the last week. Early campaigning is likely to become even more important.

That said, you often hear people talking about the percentage of people who make their mind up in the polling booth. I’d be very interested to see research regarding how this number is moving.

In defence of Peter Dunne

Okay, it really, really pains me to do this.

One of the things that I really dislike in politics are bold faced lies. By claiming that Dunne has no mandate to sell assets, that’s exactly what Russel Norman is engaging in. This sort of politics is never pretty.

The Greens are calling on United Future leader and MP for Ohariu, Peter Dunne, to ask voters in his Wellington electorate whether they want asset sales.

Greens co-leader Russel Norman said National and ACT had campaigned on asset sales but Dunne had not, and he now had the power to stop them.

The thing is, Peter Dunne did campaign on his stance on asset sales. And it just so happens that his policy is compatible with National’s. What a surprise.

I am glad that the Greens are fighting hard against the sale of state assets, and I’m sure that Labour and Mana will also be fighting them at every step of the way. But by claiming that Peter Dunne never campaigned on asset sales, Russel Norman is simply wrong.

Let’s be clear – the only reason Peter Dunne is in this mess in the first place isn’t because of Russel Norman, it’s because his policy platform was as wet as the water resources he doesn’t want to see sold. He campaigned against sales of assets that were never going to be sold. It’s infuriating having someone as limp as Peter Dunne in Parliament – but he should be criticised for his lack of spine, not a lack of truthfulness.

Looking at turnout

I’ve had the pleasure of spending a sunny Sunday afternoon looking at the turnout numbers of the 2011 general election. The total turnout was only 74.21% (down from 79.46% in 2008). In raw numbers, that’s 2,257,336 votes cast, as opposed to 2,356,536 – despite the growing population.

2011 turnout. Click to enlarge

Wellington Central has this time taken out the title of highest turnout, beating the 2008 champion, Otaki (I am expecting a call from Mrs Keall). At a guess I would assume this has a lot to do with both Labour and the Greens running strong get out the vote (GOTV) strategies – perhaps the only place in the country where two parties manage this.

At the other end of the spectrum, we again have all seven of the Maori seats, followed by the three South Auckland Labour strongholds (Manukau East, Manurewa and Mangare). The continuation of this trend is particularly disappointing. Some may find it surprising given that this year there were four parties running strong campaigns in the Maori seats (Labour, Maori, Mana and Greens) – the simple fact of the matter is that Maori are far more disassociated from the political process than perhaps any other group in society.

I’ve also analysed the swing in turnout…

Turnout swing. Click to enlarge.

The dramatic dip in the vote in Botany stands out. Perhaps this is due to them having a by-election so soon, and local body elections (and a council by-election) last year. Are the good people of Botany simply suffering from election fatigue?

Before I pulled these numbers I had a theory that there would be a dramatic drop in turnout in the Christchurch electorates due to the earthquake. We certainly saw that to some extent (particularly Christchurch East, Christchurch Central and Wigram), but not as much as I expected. The other factor to take into account is that a lot of people have simply fallen off the electoral roll as they’ve moved away from Christchurch (the total number of people enrolled in these electorates has dropped significantly). There is more work to be done to see if they have since gone and enrolled elsewhere (and thus will have some pretty big impacts on the next boundary change), or if there are many earthquake refugees who are not even enrolled.

The big question is why did so few people vote?

All sorts of theories are being rolled out: flow on impact from the Rugby World Cup, lack of online voting, simple voter apathy. One thing that I think really did have an impact was having the National Party above 50% in the polls for so long – I’d hazard a guess that a lot of people didn’t cast their vote because it looked so certain that it would be a cakewalk to victory.

The Maori and Green parties are calling for an inquiry into the low turnout (a good idea, but it really does remind me of Saramago’s novel Seeing).

Radio New Zealand had a particularly interesting interview with Thomas Patterson, a professor of government and press at Harvard University’s John F. Kennedy School of Government. He looks at our low turnout and points out that it’s in line with global trends. Take a listen here.

I’d be very interested to hear what you think about our low-level of turnout, and what can be done to slow or stop the trend.

(Almost) Final results released

With the exception of judicial recounts in Waitakere and Christchurch Central, the final results of the 2011 general election have been released. Congratulations to Carmel Sepuloni in particular – knocking off a popular cabinet minister in a year when the tide is out against Labour is no mean feat. Commiseration to those who have lost their seats.

Turnout was only 74.21% (down from 79.46% in 2008). This is very disappointing. The pundits will be discussing this until the cows come home, and from the conversations I’ve had with others already, there are some interesting theories emerging. I’ll be looking at some of these issues in the coming weeks.

I’ll also be updating my Labour candidate vote statistics in the next day or two. I also plan to investigate the variance in party vote and turnout , electorate by electorate. I have some theories already I intend to test.

But for now, I’m off to celebrate with the good folks from the Campaign for MMP. They can be particularly proud of their 57.77% result.

Labour’s electorate vote

One of the most paradoxical things to come out of the 2011 general election was that while Labour’s share of the party vote was decimated (down to 27.1%  – or a 6.9% swing against), a reasonable number of their electorate candidates managed to substantially increase their majorities.

Electorate Candidate Percentage swing to Labour
Mangare Sua William Sio 20.7
Rimutaka Chris Hipkins 11.3
Wellington Central Grant Robertson 7.1
Tamaki Makaurau Shane Jones 6.8
Manukau East Ross Robertson 6.7
Te Tai Tokerau Kelvin Davis 6.5
Napier Stuart Nash 6.4
Ikaroa-Rawhiti Parekura Horomia 6.4
Manurewa Louisa Wall 5.5
Auckland Central Jacinda Ardern 4.7

Firstly, congratulations to all those candidates who managed to hold their seats (if Brendon Burns holds Christchurch Central on specials, then Labour have only lost one electorate – Waimakariri –  despite a massive swing against them).

Also, before we go any further, it is worth noting that there are lots of variables at play. For example, the Te Tai Tokerau swing is against the 2008 general election result, not the very close 2011 by-election. So please take things with a grain of salt.

For me there are some pretty obvious insights that jump out of the raw numbers.

If you look at the three “M” seats, Mangare, Manukau East and Manurewa, Labour managed to significantly increase their share of the vote, despite the turnout being in line with the nation-wide trend. This has to be due to a combination of factors working in Labour’s favour:

  • The absence of Taito Phillip Field. In 2008, despite being booted out of Labour and facing corruption charges, he stood against Sua William Sio, and his Pacific Party stood candidates in the other “M” seats. They never had a chance of beating Labour, but they took a noticeable amount of the votes.
  • It’s possible that elements of Labour’s policy platform – $15 minimum wage and GST off fruit and veges – struck a chord in these electorates which have high levels of social deprivation. That said, if we’re going to make untested, sweeping generalisations we have to admit that the people at the bottom of the heap are also the ones most likely to be disconnected from the political process.

Another point to note is that three of the top 10 seats in terms of swing are Maori electorates. One of the two new seats Labour picked up was Te Tai Tonga. Their candidate, Rino Tirikatene, pushed hard on the line that Maori Party voters were dissatisfied with the deal with National. So hard, that on election night, even Tariana Turia was admitting it

“And it may well be … that they haven’t liked the relationship with National.”

The final point that jumps out is the fact that Labour obviously has some decent campaigners. Everyone on this list has done well (and to be fair, there are also another dozen or so candidates who have pulled in very respectable results). Robertson, Hipkins, Nash and Ardern all ran excellent campaigns. It really shows that one of Labour’s greatest losses on election night was Stuart Nash. I personally hope he sticks with it – victory in Napier is well within his grasp for 2014.

There are of course massive issues with the party vote, and turnout generally. I’ll save those for another day!

Labour perspectives on the election result

My main motivation for starting this blog was to get some thoughts that have been bouncing around my head, particularly about the election result, Labour’s leadership contest, and party reform, down on paper. I figured a useful way to start this off would be a quick re-cap about what others on the blogosphere are saying.

I’m not sure if I’m going to find enough material to do a full spectrum of material, but there has already been plenty written by some of my fellow Labour supporters. We have a lot to think about…

Jordan Carter – Just Left

In the wake of Labour’s most serious election defeat since the 1920s, a comprehensive and critical re-examination of almost all of what Labour’s politics is about is an absolute necessity for our party.

On the table must be our policy, our campaigning, our organisation from branch to national level, our candidate selection, our structure, our communications, our tone, the way the parliamentary party works, what the staff do in the party and in parliament, and on it goes.

There was a bit of reflexive back patting after election day.  I’m not too worried about that, but the time for that is now past.

Carter calls for a long hard rethink of what we’re doing. Amen. He acknowledges that we have some serious organisational short-falls, and failed at the basics. When push comes to shove – the voters simply didn’t connect with Labour. He calls the election as he saw it –

Jordan acknowledges that leadership was an important part of the election result – but takes a strong stance that a new leader will not be a silver bullet.

We have to say it clearly: NO leader can do what needs to be done on their own.

It requires every single one of us in the Labour Party to stand up, to do things differently.

He’s dead right – we all have to lift our game. We have identify and work towards fixing our shortcomings, and understand why voters felt so disconnected from Labour.

Phil Quin – The New Tasman

Phil for the most part observed the 2011 general election from overseas. Which makes his views particularly interesting. He is somewhat less pessimistic:

Okay, it was a shitful result for Labour; no two ways about it. 27 percent is a meagre return for 95 years’ worth of mostly honest toil, especially when a charming huckster like Winston Peters can score a lazy seven for a fortnight’s work. (Notwithstanding, that is, the contribution of the hitherto well-hidden youth wing of NZ First Party who Peters credited for their stellar effort on “the social pages”).

But, in the scheme of things, last night’s result was well short of the worst imaginable scenario for Labour.

“WTF?” quoth the doomsayers. Allow myself to explain myself.

  • Well, 27 is not twenty, which is what National scored in an analogous scenario in 2002. From which the Nats recovered sufficiently in a single term to come within a bee’s floppy of beating Helen Clark in 2005. With Don Flipping Brash as leader.

I have to say I agree. Despite the many obvious problems Labour faced going into, and during, the 2011 campaign, we managed to win back two electorate seats, increase our margins in many more including holding two that the National Party expected to win on the night (Rimutaka and Palmerston North). The massive drop in the party vote is of course the biggest problem we face. I’m looking to explore this in depth in future posts.

Rob Carr – Political Dumpground

Disclaimer: Rob was a member of my campaign team in Wellington Central, and did a very impressive job managing our GOTV effort.

Rob has done a reasonably in-depth analysis of the election result, and there is lots to comment on. His entire post is well worth a read.

One very salient point he makes is that while it was certainly a bad result for Labour, it was also a very unsatisfactory result for National…

Less than a quarter of the New Zealand population voted for the Government when you combine those not able to vote, those who voted for the left and those who did not which is a bit alarming. It is a result neither side is happy with because National does not have a stable Government with the centre-right share of the seats decreasing by 4-5 and Labour obviously doesn’t get to be Government. The centre-left probably couldn’t have had a much better result other than that 2 extra seats to form Government with the Maori Party being possible but it was a bad result for Labour with their vote being eaten up by the Greens and New Zealand First.

All three agree that it was a very bad result for Labour, but all with slightly different views.

From my own perspective – I’m very pleased that Labour bloggers are already looking towards the future, and thinking about the sort of changes we need to make as a party to ensure that we never see a repeat of the 2011 result.