New Zealand 2020 preview [and discussion of results]

The following originally appeared here on 27 July. I am re-upping it because the election is 17 October–right now. I think most of what I wrote back in July still applies–other than the election date itself (and NZF appearing even deader now than it did at the time). The main question of the election remains what it was then: Will Labour win a majority on its own, or will it need a coalition or other agreement with the Greens? More recent polling suggests the answer might be the latter, but it looks like a close call.


New Zealand’s general election will be 19 September (grrr, they are holding it on Rosh HaShanah). Given the generally good record of the government of Prime Minister Jacinda Ardern in handling COVID-19, at this point the question seems to be, how big will her win be? And how much will her two coalition partners suffer as Ardern’s party gets most of the credit?

Through today, there have been seven recent polls that have put Ardern’s Labour Party over 50% in votes. If this were to happen on election day, it would make for the first time since the mixed-member proportional system was first used (1996) that a single party won a majority of seats or votes.

The last poll showing the main opposition National Party ahead was conducted in February. The party has changed leaders not once, but twice, since then.

Obviously, with a majority, Labour would not need to take on coalition or support partners. However, it likely would want to retain its relationship with the Green Party and thus retain the latter in some decidedly junior position. That is, assuming the party clears the threshold. Even the polls that show Labour as far up as 60% (!) still show the Greens over 5%. So, it may be a close call, but they should at least remain in parliament. There seems to be only one poll in several months that has them below the party-vote threshold (mid-May, at 4.7%).

The other partner in the incumbent government, New Zealand First, is languishing far below the threshold in all polls. Its only faint hope for survival would be if it can take back the electorate (single-seat district) of Northland. The party’s leader, Winston Peters, won this seat in a by-election five years ago. The National Party won the seat back in the general election of 2017. Its candidate for the seat in this election, Shane Jones, is making his pitch for the seat.

For far too long we have tolerated substandard National Party representation for our Northland area. No power or influence. No bite and, in fact, not even a decent bark.

The just-linked NZ Herald article remarks that the by-election showed that “the government [then led by the National Party] knew so little about the North that it thought bridges were what everyone wanted.” Jones has said that most of the promised bridges were never built. (So, evidently they do want bridges, even if that’s not all they want.)

The episode is a nice reminder of how distributive politics can come into play in the MMP system. Despite effectively nationwide proportional representation, the district races are an opportunity for local factors to enter into the campaign. Nonetheless, it would seem a very tall order for New Zealand First to repeat in a general election what it pulled off in the by-election. With National crashing so badly nationwide, however, perhaps it is not out of the question that local National supporters could vote for Jones. The latter has emphasized that he could be in cabinet, while the National member would be an “obscure backbench MP,” continuing the alleged neglect of the region by both Labour and National. (Never mind that NZF is unlikely to be back in cabinet no matter what, if their votes are not needed, as they were after the 2017 election.)

Meanwhile, Peters has launched his party’s campaign with the rather odd slogan, “Back your future.” He is really pushing the idea that he is the only thing standing between New Zealanders and a radical government pulled further left by the Greens. It is about the best case he can make for a vote for his party. Given the overall competence Ardern has exhibited and the fact that the Greens would have hardly any leverage if Labour wins a majority on its own (or even if it is merely close to a majority), it is not a claim with much reality behind it. Still, the always colorful Peters has said that being in government has allowed him and his allies to block “woke pixie dust.”

The Greens are also looking for ways to differentiate themselves from both of their partners. Co-leader James Shaw remarked,

We’ve known for a long time, that the closer we got to election, the more likely it was that NZ First would start misbehaving.

…If you look at some of the difficulties that the Government has had over the last three years, a lot of them have come down to NZ First ankle-tapping them and blaming them for saying they can’t get anything done.

I know they like to say they are a force for moderation; it’s more like an agent of chaos.

Green MP Jan Logie has also called attention to New Zealand First’s opposition to the Sexual Violence Legislation Bill, which she is championing.

The Greens are also at work differentiating themselves from their senior partner, albeit with less divisive words, and a policy focus. For instance, they oppose Labour’s policy on charges to New Zealanders returning from abroad and needing to quarantine. (National supports Labour’s plan, so the latter can pass it without the Greens’ support in the current parliament.)

Peters and David Seymour, the leader of another small party, Act, have done their part to keep the campaign especially classy.

Act is enjoying a bit of a surge lately, from polling oblivion a few months ago to 5% in one recent poll and over 4% in several. It just might clear and win multiple seats. Seymour currently holds an electorate seat. Because New Zealand’s MMP has an alternate threshold–five percent of the party-list vote or a single district win–there is a chance the party could elect more than one member for the first time since 2008 even if it remains below 5%. I’d think their odds are reasonably good, as some more ideological right-wing voters may see National as hopeless to form a government and instead vote Act.

The current government was a somewhat strange one when formed. Labour’s 2017 result was ten seats behind the then-governing National. Even with Green support the left-leaning post-electoral combine was two seats seats short of National (54-56). They needed New Zealand First (9 seats) to choose them over National, which of course it did. Now the government looks to be one more case for the common political-science finding that governing as a junior partner is perilous for a party’s electoral support. New Zealand First may be shut out of parliament altogether, and the Greens may be down a seat or two from their 8 in 2017 and facing a partner that possibly has a majority on its own.

44 thoughts on “New Zealand 2020 preview [and discussion of results]

  1. Thanks for the planting.
    I thought that giving charging for returnees as Labour policy on the basis of a Cabinet Paper was a bit keen. Looking like a version in line with Green stated policy will get in place before the election. Labour have only ever said that they were looking into it.
    There was much dropping out by sitting National MPs as leadership changes and general ruckus occurred.


    • That does tend to confirm my long-repeated hobby horse that, if voters want a majority government, they can get one – even under nationwide PR -by, you know, giving one party a – what’s that term again? – “majority of votes”.
      Absolutely nothing stopping them.


  2. It will be interesting to see whether there are a significant amount of split votes at this election, if these polls are indeed borne out. A 60-25 split on the SMD vote would seemingly make it hard for National to win many seats at all. Obviously, however, National district votes have no impact on the seat distribution (except if National were to win an overhang), and as such National MPs could theoretically campaign for district votes while still allowing voters to give Labour a landslide.

    Whether this does actually happen is a separate question. While at the 2002 election, which was the largest swing against a major party in the MMP era, the National list vote was 10pp higher than the SMD vote, this seems to have been partially because NZ First ran SMD candidates in less than half of all electorates. In 2008, there was no appreciable difference between list and SMD votes for Labour, but there was in 2011.


    • Thanks. That is indeed good.

      On Labour, I like this: “The party is still yet to release any substantial policy for the election… Without her own policies to promote, Ardern will be forced into spending the whole election just ruling out Green Party policies.”

      Classic “valence” campaign! Trust us because we are competent; and we are not extreme.

      Please remind me what the “waka jumping” legislation was.


      • I like this point, too: “Labour itself has plenty of MPs happy to let NZ First take the blame for killing progressive policies they themselves are uncomfortable with.”

        Yes, coalitions can be handy that way.


      • A waka is a canoe, and jumping references leaving one mid-stream. Per Wikipedia “members of Parliament who choose to leave their party, or who are expelled from their party, are automatically expelled from Parliament, with the seat becoming vacant.”
        A Member’s Bill to repeal the latest version was drawn in July 2020, and was supported into Select Committee by the Greens, to be considered post-election.


  3. Extra stressful times as NZ Electoral Commission, as COVID community transmission has been detected in Auckland. Auckland (so a third of voters) in COVID Level-3, rest of country in COVID Level-2. This until midnight Friday, at which point I assume they hope to have found the source and can therefore drop Auckland to Level-2.
    They have planned to hold the election in Level-2 conditions, with small areas at Level-3. There is provision to delay the election if required.
    Level-3 allows food delivery but not restaurant/bars, essential services (our definition of these is reasonable) only, schools in-person for children of essential workers only. Level-2 allows schools etc and most businesses to open with distancing/masks.


    • Dissolution of Parliament was scheduled for this morning, now on hold and will be reviewed on Monday. A vote by Parliament would be required to delay the election by more than days, so this leaves options open.


  4. Pingback: Possible New Zealand election delay | Fruits and Votes

  5. Pingback: NZ2020: Maori Party list candidate attributes and “burning bridges” | Fruits and Votes

      • Very badly-designed, that ACT website. I can’t make the drop-down menu go away, so the largest party I can see is Jeremy Smythe-Fauntlington’s Canberra Liberals on 33%.


      • At this stage Greens could potentially win six seats (Labor’s twelfth seat, wherever it was, has disappeared) which would give them 24% of the seats for 14% of the vote!


      • That sounds like the ACT’s equivalent of Tasmania’s Peg Putt Paradox. (“Too many Green MHAs, with 7-seat electorates. Let’s cut the district magnitude to 5-5 and get rid of these pesky troublemakers.” But then, with a fairly firm hold “quoat-squatting” on one seat per district, the Greens then hold 5 seats out of 25 instead of 5 out of 35.
        But six seats with five districts would mean they’re in line for a second seat in one electorate?
        Interesting that S-Rat has told the media that the Greens would not consider allying with the Liberals as long as it has a leader like Coe who is “too populist” (ie, right-wing). By implication, that leaves open a future Green alliance with the Libs if they ever return to centre-left moderates like Carnell or Humphries. I could certainly see Greens generally preferring a Liberal party under someone like Ted Baillieu (or Chris Puplick,if he were still in politics) over a Labor led by Mark Latham or Graeme Campbell (if either were still in the Labor Party).
        You can always incite some initial interest in Hare-Clark when you tell the average person on the proverbial Bondi tram, “In Canberra they rotate the order of names on the ballot-papers… They also had a Liberal leader named Zed.” (Unfortunately said interest never lasts too long)


      • The ACT Greens look likely to win a second seat in the inner-city seat of Kurrajong, where they got 24% of the primary vote to 26% for the Liberals. I suspect that the Greens’ good result has something to do with the more harmonious relations that exist between Labor and the Greens here relative to those in Tasmania, where Labor has to campaign to workers in the resource sector. Labor campaign literature advised its voters to vote 1-5 down the Labor ticket and then 6-10 down the Greens ticket, which I don’t think is common in Tasmania,


      • Jokes about Zed Seselja and Robson Rotation aside…
        ‘Taimus Werner-Gibbings has a lot going for him. He’s a Sydney University law graduate, holds the plum position of deputy communications director to shadow minister Andrew Leigh, and has a handsome mug which should attract some second glances to corflutes around south Canberra roadsides.
        But on his boss’ own research, the Labor candidate for Brindabella will have to overcome the electoral disadvantage of his hyphenated surname if he is to join the Legislative Assembly in October.
        Labor candidate for the seat of Brindabella, Taimus Werner-Gibbings, with his wife Libby. Research by Andrew Leigh says attractive politicians have an advantage at elections, but those with hyphenated surnames are held back. Photo: Graham Tidy
        In his book The Luck of Politics published last year, Dr Leigh said federal candidates with hyphenated surnames fared an average 2.3 percentage points worse than those without them.
        No hyphenated person has ever become an [ACT] MLA, and there is none in the current House of Representatives either. (Three senators have hyphenated surnames, matching Dr Leigh’s expectation the Senate model of candidates running on party lists [sic] makes the issue less relevant.)…”
        – Matthew Raggatt, “Labor hopeful Taimus Werner-Gibbings hopes to break hyphen drought in ACT election”, Canberra Times (23 July 2016),


    • Around half of votes counted, Labour on 50-51%, with ~7% wasted votes, so able to given alone. Maori close in Te Tai Hauāuru, but probably won’t get it.


  6. My website’s New Zealand page now has nationwide party vote and electorate vote preliminary results of today’s general election in the South Pacific nation. As in a number of past NZ elections under MMP, the electorate vote outcome wasn’t as lopsided as that of the party vote, even though Labour still won the former by a fairly substantial margin.

    As for Winston Peters’ New Zealand First Party, it was wiped out indeed and only managed to finish a distant third in the Northland electorate, with 11.6% of the electorate vote, in what turned out to be the party’s best electorate showing.


  7. The NZ Grins may be foreshadowing (or perhaps more “hoping for”) a more Australian-style coalition between their party and Labour. ie one that remains in place even when the larger party wins an absolute majority of seats on its own (as the Australian Liberals did in 1975 and 1977). Some voluntary sharing of power today as prudent insurance against Rob Oakeshott/ Winston Peters-style coy-playing if a future election doesn’t give a majority on election night.


  8. It is possible for one party to win an absolute majority under a PR system. Amazing that NZ party system fragmented under FPTP, interestingly enough it consolidated under a MMP system perhaps the reason is that voters had an incentive to waste their vote under FPTP, where it wasn’t likely to make that much of a difference where as with the party vote, it makes a difference through the country.

    What would the result had been if NZ First won an electorate seat? Also what would the result had been if their was no threshold?

    Alternative back in 2017, NZ First could have allied with the National Party. National would be in power, and would Labor had defeated them in this election?


    • “Consolidated” is hardly the descriptor I would reach for. While 49% for a party under PR is unusual, the gap between the top two is very large, and that is down to fragmentation of the opposition (including normal National voters having spun off to variously to ACT, Labour, and New Conservative).

      By contrast, under FPTP, even when there was “fragmentation” in the sense of a strong third party (in votes), the top two tended to be relatively close.

      As for the hypothetical of NZF striking a deal with National in 2017, I suspect that would have killed them off at least as thoroughly as what happened to them here. There was a very large swing to Labour in that election, even though National remained the largest. I doubt Peters seriously wanted to prop up National, as opposed to extracting what he could in protracted bargaining to produce a change of government. Of course, in the end his party was killed off anyway. Coalition junior partners often suffer, all the more so when the senior partner has about the best valence one can imagine.

      When I think of elections won on valence–such as leadership competence–rather than ideology, this will be the one I will think of for a long time to come.


  9. Is anyone else as surprised as I am that New Zealand First got as high a party vote percentage as it did? 2.7% is well above what I was expecting. I guess that’s their hardcore, as there was little question at any time over the last few months of their clearing 5%. And they did not come close in Northland or any other t electorate. So unless a voter was just not paying attention, no one could have voted for NZF thinking it would be an effective vote.


  10. It is quite notable that the Greens won an electorate seat (Auckland Central, with 34.5% of the vote) in a close contest with the Labour candidate. This is their first electorate win since 1999. I wonder if a block of Labour voters in the electorate “lent” their nominal vote to the Greens as threshold insurance. Surely it also helped that the Green candidate herself is well known.

    Also, the Maori Party made a small comeback, winning one seat (Waiariki, with 45% and a narrow margin over the Labour incumbent).


  11. ‘… “When you look at the numbers, New Zealanders have essentially been satisfied with their government since 1999,” said Stephen Mills, the head of UMR, Labour’s polling firm….’

    Charlotte Graham-McLay, “Why New Zealand rejected populist ideas other nations have embraced: Labour’s historic win delivered Ardern a second term while voters punished politicians who embraced populism”, The Guardian (19 October 2020),

    Hmmm. 1999. Interesting. Perhaps something happened shortly before that time that contributed to this sense of satisfaction.


  12. Pingback: MMP in NZ: An example of “best of both worlds” in action | Fruits and Votes

  13. Pingback: NZ 2020: Strategic coalition voting? | Fruits and Votes

  14. Labour-Green talks expected to finish today, with Green ratification (or not) by Sunday.

    Preliminary referendum results will be out shortly
    Around 15% of votes will not be in the preliminary results (late enrolments, out-of-district votes, overseas votes, those incarcerated in the Covid camps etc)

    Final results will be next Friday the 6th (Thursday evening US time).


  15. Labour-Green agreement
    There is little policy detail in the proposed cooperation agreement, other than a general agreement to work together on child poverty, climate change, and the environment.

    The co-leaders will be ministers outside of Cabinet, bound only by collective responsibility in relation to their portfolios.

    This will allow the Greens to vocally oppose the Government on other matters. There is also an agree-to-disagree provision on the matters the Green Party do have portfolios in, so it can note a difference of view there.
    But one area that Labour and the Greens have agreed to look at is the reform of our electoral laws.

    The agreement says the reforms will fall under three areas: the recommendations of the 2012 report into MMP, electoral finance law, and the length of the parliamentary term.


    • How low will the threshold go? What is the pros and cons of a 3 year terms vs 4 year term? Australia, Mexico, and El Salvador have 3 year terms. Do voters prefer shorter or longer terms for legislative bodies? It seems to me that Australians and NZ have rejected 4 year terms in the past.


  16. Labour and Māori gained a seat each off National, the Māori one being a ‘coat tail’ as their Party vote share increased.
    The results of the recreational cannabis and euthanasia referendums remain unchanged. A majority voted against the legalisation of recreational cannabis, 50.7 per cent voting “no” and 48.4 per cent voting “yes”.
    There was a fractional change for the End of Life Choice referendum, with 65.1 per cent voting “yes” and 33.7 per cent voting “no” – a 0.1 per cent shift towards “no”.
    Green MP Chloe Swarbrick claimed an unexpected win in Auckland Central, but with a margin of 492 votes Labour candidate Helen White was unwilling to concede. The final vote tally increased Swarbrick’s lead, to a 1068 margin.


    • The Māori Party are raising the once-per-census limit on changing between the Māori and General Rolls, and the 5% threshold via the mechanism of applying for judicial recounts in two seats.
      Note that the original Royal Commission for electoral systems in the 1990s recommended that the threshold not apply for Māori parties, but didn’t offer a mechanism for identifying them!

      “He said the party was “bombarded with complaints” about voting during the two-week advance voting period. The recount, sought for both candidate and party votes, would allow the party to analyse informal votes (votes considered incorrect that don’t count towards the result), he said.”


Leave a Reply

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

You are commenting using your 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 )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.