Coalition calculus

So, I’m noticing, again, that people aren’t very up on the combination of maths and political speculation that you’ll need to understand who’s likely to form the next government.

Firstly, let’s be clear, there are only three parties parties in Parliament right now who have either clearly commited to who they’ll support or for whom it’s obvious despite the fact nobody’s bothered to ask them. Of course David Seymour/ACT will support National. The Greens have been very clear they have committed to support Labour to change the Government. And Peter Dunne/United Future, while they have worked with Clark’s Labour Party, will of course support National ahead of them given half a chance.

Naturally, Labour and National will lead their respective blocs, so if you’d like to count that, there are five parties who you can safely vote for knowing what sort of government they’d support.

There are a couple of things that are unclear. Mana will of course be more likely to support a Labour government if voted back in to Parliament, however will likely be on the outside of any arrangement if Labour is actually governing, so the uncertainty there is in whether people are willing to forgive Hone for his willingness to work with Kim Dotcom. Given that the Māori Party have stood aside and essentially endorsed him, however, he’s not discountable. I wouldn’t be surprised at either a Mana or Labour win in Te Tai Tokerau.

I won’t speculate on the odds of Dunne losing Ōhāriu at this stage, other than to say both Labour and Peter Dunne need to be taking that electorate deadly seriously, especially as I wouldn’t be surprised if there’s a significant chunk of it who are fans of neither of the two likely candidates. They will need to reach out to supporters outside of their usual base to win, which is presumably behind Peter Dunne’s hilarious inability to find fonts that support macrons.

This brings us to the two wildcards. Firstly, the Māori Party, who have said they will make a reasonable attempt to work with whoever’s in government, but haven’t talked at all about what happens if they decide the next government. While they’ve most recently been in a non-coalition support arrangement with the National Party, there’s no real certainty that they would support them over Labour. In fact, the List party they best align with on the issues is the Green Party, so if we’re going purely on policy, they might well support Labour. With Tariana Turia’s retirement from the party, the historical disagreements with Labour could be a lot less of an obstacle to doing a deal. In my models, I’ve assumed they will be open to working with governments of both stripes, but I consider them a reverse UF- in that I slightly suspect they will prefer a Labour government if it’s their choice, although I don’t have any hard evidence to back that speculation.

And then there is New Zealand First, who like the Māori Party, haven’t publicly chosen sides yet, and don’t even have the excuse that they still intend to work with both parties, as they’re currently sitting on the opposition benches, collaborating quite nicely with Labour and the Greens. They look like they’re set up to change the government, but they have pulled a switch on their supporters this way before back in the first MMP government. Have they learned their lessons from that? Nobody knows.

Lastly, there’s a little election system maths to consider. The number of overhang seats will be important in this Parliament, as National will need to win a similar percentage of the Party Vote to last election to stay in power, which was already a bit of an ask with John Key at the helm instead of Bill English. They’ll also want a bit of padding, as no government has ever won a by-election under MMP, so if any of their MPs resign, they’ll likely lose a seat.

If all the minor parties win seats, that’s likely to be three overhang seats in Parliament, meaning the “magic number” becomes 62/122-123, not 61/120-121. There’s also a credible chance of an independent upset in Ilam, which would have the weird effect of not changing the number of seats in Parliament at all, but rather robbing whoever would normally have won the 120th list seat of their extra MP. That’s likely to be either National or the Greens, depending on how close the Greens end up getting on their last MP allocation, or on an outside chance it may be Labour. (it’s highly unlikely to be the Māori Party, as small parties generally get allocated their seats way before the end of the process) For those not aware, essentially there’s a divisor formula that goes through and outputs a number for each party with Party Votes, and every one that’s over the threshold or has an electorate gets thrown into the formula. Whoever’s number is the highest gets the first seat, and then their divisor is recalculated on a less friendly basis, and the next-highest number gets a seat, for 120 iterations. Independent MPs are the only disruption to this process- small parties simply get their electorates as overhangs if they don’t qualify for a list seat on the basis of their Party Vote. While Raf Manji is likely to vaguely support National, he does represent a wildcard in terms of who loses a seat to him, and it’s actually as likely to hit Labour or the Greens in the face as it is to cause disruption to a National government.

What are the important takeaways?

1. It’s very likely that whoever governs after the election will need 62 seats, as I don’t consider it a serious likelihood that both Dunne and Harawira will lose their electorate votes.
2. If National wants to govern, they need those 62 seats without the Māori Party or New Zealand First, if at all possible. It’s entirely realistic that the National Party could win 59 seats for themselves but lose the government because they only have two seats worth of solid friends, and both the less predictable parties decide to swing Labour’s direction. Likewise, Labour will want a situation where they can narrow the margin enough that they have a choice between New Zealand First and, say, the Māori and Mana MPs, so that he’s not forced into too much of a pro-nationalist agenda, but such a situation looks rather unlikely at the moment without National seriously dropping the ball, or Little scoring some serious body blows on English that he just hasn’t gotten a chance to do yet.
3. It’s highly likely that New Zealand First will decide who gets to govern based on current polling, if it holds accurate, and if events don’t shift too fast for our sporadic polling to measure accurately. The chance of National governing without them is pretty slim, although it’s now within the realm of reasonable probability. I expect the recent trend back to National to have arrested with the announcement of Jacinda Ardern as deputy, but we’ll see if that eventuates. I also expected that the dip on English becoming PM would become a trend, and that hasn’t eventuated yet, possibly due to poor capitalisation on his weaknesses on behalf of Andrew Little, but again, this is where Jacinda will actually serve very well as deputy: she’s far more persausive with the emotive argument that the Government is failing people than King was, even though Annette King was a highly effective MP and debater, she doesn’t have Ardern’s charisma.
   1. Relevant to Ardern, it’s rather hilarious that the media is now openly speculating on whether it causes a problem for Labour if she overtakes Little in popularity. The answer is no. That’s a great problem for Labour to have, and it’s likely she would want to do further work before even considering the No. 1 spot in her party, so expect for that to happen at some stage and for Jacinda to endorse Little as leader even when she’s more popular than him. Being the most popular isn’t always a qualification to be leader, and that is why people have to put themselves forward as candidates first. If it’s ever a serious issue, Labour will vote on whether she gets to take over, like you would expect, although of course, they’ll tilt the odds in favour of their caucus’ vote for some bizzarre reason.
4. If you’re giving your party vote or electorate vote to the Māori Party, you might want more information about what their plan is if they’re in a position to decide if Bill English stays Prime Minister or not. They’ve already said they can work with Labour, but would they prefer to? Nobody knows at this point, and the largely vacuous excuses for political journalists we have atm haven’t got the nuance yet that they need to ask a very specific question about this and see if the answer will change or not. (There are real political journalists out there who absolutely would understand this nuance, but they’re never given the job of interviewing or moderating party leaders for this sort of question in a format that significant numbers of NZers watch)
5. Pushing for Raf Manji to win Ilam is only particularly relevant if National wins back the government again, and his vote becomes relevant for key legislation. (not that in principle I object to Gerry Brownlee being sent a message, or Christchurch having a bit more power in Parliament, those are good things, it’s just unlikely to move the electoral calculus in favour of changing the Government) It’s actually far more relevant to National whether O’Connor stands a chance of unseating Peter Dunne and whether Hone wins Te Tai Tokerau, as unlike the MP he is unlikely to support the National Party.
6. As before, Preferred Prime Minister is a huge irrelevancy in polling and deserves to be replaced with favourable/unfavourable polling for the PM and the Leader of the Opposition, as the open-ended nature of the question prevents it from being useful as a gauge of whether the leader is an actual drag on either of the biggest two parties, and the lack of data as to whether it has any impact at all on the Party Vote means that experimenting with different metrics is a great idea, in case anyone who actually polls is listening.