40 thoughts on “NZ election

  1. Well, then, it would seem that a Malta-style guaranteed majority of seats for a party with over 50% of votes should be on the agenda for the review.

    It is curious to reflect that these quirks can occur because of the otherwise-vestigial influence of SMDs in the MMP system. If NZ used pure nationwide list PR (with a 5% threshold), neither the scenario above nor the bizarre tactical-voting manouevres in Epsom would be on the table.

  2. The 50% situation should be very rare (for a single party). The current weakness of Labour (IMO largely due to lingering arrogance after a long period in power) isn’t helping.
    Am I right in thinking that raising the required number of electorate seats to be awarded list seats (as many are suggesting) increases the likelihood of disproportionate results (unless other changes are made)?

  3. MMP wins convincingly! After 5 MMP elections (this year’s is the sixth), New Zealanders like it.

    However, an editorial before the vote pointed to the need for review:

    “Fairly obvious changes could be made to ease the widespread annoyance over list MPs entering Parliament without an endorsement from an electorate. A party’s list seats could, for example, go to its candidates who came closest to winning an electorate.

    MMP would be much improved by such modifications. . . Such changes should have been put in train much earlier.

    Parliament’s failure to do so has allowed cynicism about political conduct to grow and fester. Tellingly, however, this has come nowhere near the level of disquiet over the unbridled power afforded by the first-past-the-post system, which triggered the introduction of MMP.”

    In New Zealand, the conservatives come one seat short of a majority and promptly schedule talks with three smaller parties.

    In Ontario, the Liberals come one seat short of a majority and refuse to negotiate.

    What is wrong with this picture?

  4. I would love to see the congratulatory telegram David Cameron will be sending to John Key. In fact, I would like to see the congratulatory telegrams David Cameron sent to Enda Kenny, Alec Salmond and Barry O’Farrell. Nothing will dent the “PR and/or preferential voting = weak, unstable coalitions but FPTP = strong majority governments” bubble at the Spectator and the Daily Mail, but Cameron has a first in politics at Oxford so presumably he must at some point have to at least acknowledge counter-evidence from planet Earth.

  5. The Wikipedia page on the New Zealand election is unusually comprehensive for pages of this type (though it doesn’t cover the referendum), giving the breakdown between the list and the electorate seats, a map of the electorate seats, where the parties were running, which seats changed hands, and so on.

    National got nearly a 5% swing in its favor and came out 21% over Labour. This was mainly due to a disastrous Labour result, though the National vote increased by 3%. I was expecting a huge haul of electorate seats for National with that type of spread, but was surprised to find that they took in 41 seats out of 70, with very few electorate seats changing hands (and in one case Labour took an electorate seat from National, their only victory in a rural electorate). This counts the 5 Maori seats, where National was not competitive, but with this anomaly National took about 56% of the electorate seats with 48% of the vote, a pretty low spread between the seats won percentage and the popular vote percentage for a first place party in a FPTP vote in a multiparty election.

    Labour’s percentage of electorates won, just over 30%, also was close to its vote share, which is more normal for a second place party. They seem to have won their core territory in Auckland, though being beaten in Auckland overall, and done well in the smaller cities, while being wiped out in the rural areas. The Greens, at 10%, and New Zealand First, at 4%, increased their vote without winning an electorate. I suspect in a pure single member district election the Greens would have changed their strategy to try harder to win an electorate, which they have done in the past in New Zealand, though before MMP third parties in New Zealand have won fairly high percentages of the popular vote and failed to win an electorate.

    The New Zealand and German system does allow some comparison between the seats won under proportional and single member systems. I am starting to wonder if the results converge over time as parties and voters change their strategies as winning a share of seats disproportionate to their actual vote becomes more or less important.

    • The Auckland Central race is close enough that special votes could change the result. Both candidates are high enough on their party list to remain in parliament in any case.

      (Special votes are these cast outside a voter’s home electorate.)

  6. Proviso on the Referendum ‘result’, only the Advance votes have been counted. Given the gap, I wouldn’t expect a result change.

    Note: these tables show advance ordinary votes (excluding special declaration votes) as reported on election night. The Electoral Commission aims to publish the official results for the referendum, including the results for ordinary votes cast on election day and special declaration votes, by 2.00pm on Saturday 10 December 2011, once the official count is complete.

  7. There is a noticable amount of split voting, you should look at the percentage of candidate votes, as well as of party votes. An extreme example is Epsom

    Another interesting result is the dead heat in Christchurch Central (which will be broken by Special Votes).

    Also, Ed, NZ First got 6.8% of the vote, up from 4% last election.

    • In the so-far tied result in Christchurch Central, while the National candidate has a safe list position and thus will win anyway, the Labour candidate is too far down the list for a seat. He’s the incumbent in the district and must retain it, or he’s out.

  8. Few electorate seats changed hands because, unlike the party vote, the electorate vote changed very little since 2008. Specifically, National’s share of the electorate vote went up by only 1.2%, while Labour’s dropped just 0.2%.

    Interestingly enough, in this regard last Saturday’s election is a mirror image of sorts of the 2002 general election, when National’s share of the party list vote nosedived, while its electorate vote fell back just slightly, and the overall distribution of electorate seats didn’t change much; my website’s New Zealand page has nationwide party and electorate vote totals for every general election held there under MMP from 1996 to 2011.

    As for the referendum on the voting system, it is true that only advance votes have been tallied so far, but the distribution of advance votes in the general election itself (as shown on the official election results website) was largely in line with the overall preliminary totals; therefore, I would think the advance vote result in the referendum is likely to be representative of the final outcome as well.

  9. > “unlike the party vote, the electorate vote changed very little since 2008”

    Makes one wonder how elections in high-incumbent-advantage elections (mainly the US) would differ if the electoral districts were only 40-50% of the total and the rest were party lists. All else being equal, sitting legislators for a district have an advantage if voters are not quite annoyed enough to turf out the majority party. Australians like Peter Brent and Antony Green talk about the “sophomore surge” and the conventional wisdom is that personal votes for sitting MHRs in marginal seats saved John Howard in 1998 (when Labor won more preferred votes). I wonder if the NZ district results show this tendency – akin to US voters [re]electing Reagan as president but Tip O’Neill as House Speaker.

  10. Further, it’s indeed curious that National won “only” 41 districts out of 69 with a 48-27-11 margin of votes. For a second party “clobbered” by such a plurality lead, Labour was very lucky to win 30% of districts with 27% of votes. If I’d known only the numbers of popular votes, I would expect National’s score out of 69 district seats to be in the 50s, at least, if not a Canadian/ Jamaican-style clean sweep.

    This result does mean that the number of NZ elections at which the party with a plurality of votes has won an absolute majority of single-seat districts is now 8 out of the last 12 – ie, FPTP’s success rate at ensuring what it’s marketed as ensuring is now 66.67% since 1975.

  11. Tom, it’s not so curious because the electorate vote distribution was 47.8-35.0-6.8. Now, what’s really curious about this is that it’s almost identical to the 47.8-35.1-6.8 outcome of the (FPTP-only) 1990 general election. That said, back then National did better in terms of electorate seats – it secured a massive 67-29 lead over Labour – than this time around.

  12. Aha. Thanks, Manuel, for that reminder. Of course the personal and the list votes can and do vary.

  13. In the advance polls the National Party got 49.81%, but in the total it was 47.99%, 1.82% less. In Canada the advance poll voters are normally a little more conservative, and NZ seems the same. Since MMP got 55.8% of the valid advance vote, it will likely get about 57.6% of the total vote. This is a higher margin than the 1993 referendum which decided to adopt MMP by a vote of 53.9%.

  14. The next election will be interesting, the Conservatives in NZ got over 2% of the party vote. That will be a potential ally of the next National Party government if elected in 2014.

    What would this election result been like if the Conservatives were able to win one electorate seat?

    Where is NZ party system moving toward? It seems to me like it is a 2 and a half party system similar to Ireland and Australia.

  15. Errol, I think Wilf is arguing that “vote National” and “vote MMP” trend oppositely. So if National does worse in the final count compared to the advanced vote count, then MMP should do better in the final count.

  16. > ” I am starting to wonder if the results converge over time as parties and voters change their strategies”

    Eg, I wonder if MMP actually encourages tactical voting at district level as voters can still (most of the time) express their true first preference with their list vote. Usually such tactical voting would mean “squeezing” the smaller parties (quirks like Epsom aside). Enid Lakeman noted that in Germany voters tended to do this.

    It is one advantage of MMP over simple nationwide List-PR. With the latter, someone who votes for an under-5% party has their vote wholly thrown awa. But with MMP, someone who “wastes” their vote on the Christian Family Heritage Party or the Socialist Worker ticket at list level can still help elect an MP if they vote National or Labour in their district. Thus the divergence between list and district votes respectively starts to look something like the divergence between primary votes and two-party-preferred under optional- (not compulsory-) preferential systems of AV.

  17. Tom (#22): completely agree. It must be the case that many Green voters selected a Labour district candidate. Note further that several Labour MPs won in their district who would not have won given their list rank. So such tactical votes may affect the overall outcome–on the intra-party dimension, that is. Once I have the chance, I want to look at the electorate-level returs and see if there indeed were districts where this sort of tactical voting could have been decisive.

    Suaprozzodi (#19), the Conservative Party is really just one rich guy who attempted to use a Christian conservative brand–and the funds to put up A LOT of billboards– to get into parliament. It’s a segment of opinion not well represented by any current parliamentary party, but whether it could gain seats next time, and whether it would be an acceptable partner to National if it did, are very much open questions.

  18. These are lots of interesting stats that should be available in time, including split votes by party. 2008 results

  19. New Zealand would never do this. What would a one vote MMP system be like?

    One would not have the strategic voting anomaly problem.

    If we can calculate from the electorate vote as a test. This was the original German system until they it was switched to the 2 vote system. I think currently only one German state has a 1 vote MMP system.

    I thought United Future was a Christian Democratic party.

    Assuming election 2014 it is doubtful that United Future and ACT will exist, The Conservative Party even if it did get representation would not support a Labour lead government unless National wants to form a government with NZ First which they said they would not. They would rather find another party to support them as well, and they can play the parties against each other.

    National could form government with the Greens, but such a coalition would be unnatural like the one with the Maori party. They disagree with each other on economics. I wonder what common ground can the National Party and the Green can find.

    It seems funny that the NZ First Party came out of nowhere to win 8 seats. How could this have happen? Did the pollsters under estimated their strength or are voters reluctant to tell pollsters on the phone that they are voting for NZ First? Is the NZ First Vote Right-Wing Labour voters looking for a way to prevent a National Party majority? It looks like National does not need NZ First for support as it has ACT, and United Future to form government albeit with a very slender majority..

    Why would anyone want to bring back FPTP? Doesn’t anyone remember what happen in 1978, and 1981, reverse plurality election? Who wants that, twice in a row? If that happen in the U.S electoral college twice in a row, it would have been abolished.

    It seems like most NZ governments are bare majority/minority governments. The is a complete change from the old days of FPTP when a party won a thumping majority. It seems to me that Western Countries these days are going through a period where elections are very close indeed.

    • United Future is not a Christan Democratic party, though at one time it did have Christian social conservatives on its list (e.g. In 2002). Now it’s just the Peter Dunne party. Really nothing more, despite Dunne’s effort to cultivate the “outdoor enthusiast” vote with a position against the 1080 pesticide (see the billboard I posted a photo of some days ago, which was from the West Coast Tasman district on the South Island).

      Dunne has represented a district in suburban Wellington since the 1980s, which he narrowly held in 2008. He won with a larger margin this time. Counting him out for 2014 might be a bit premature.

  20. NZ First didn’t ‘come out of nowhere’. Most polls detected an increase in the final lead-up to the election, which ‘feels right’ to me (I won’t go into the perceived reasons), and they were 4.1% in 2008 (although lower since then). My guess is that a factor was the pollster’s models not taking into account a higher voting ratio among NZF voters in this low-turnout election, as well as a high proportion of votes from undecideds/stated Nat voters

    And a fine point, John Key (Nat leader) has said he would not form a Government with Winston Peters (NZF leader). The issue isn’t with NZF policies, just with the oath-breaking thief who has a track record of breaking coalitions.

  21. Let’s remember that it’s possible to make order of election depend on “best losers” at district level without getting rid of the second list vote.

    Lakeman pointed out, quite correctly IMHO, that the Hansard Society’s 1970s proposal for Britain (allocating seats to parties based on district-level votes for their candidates only, with “best losers”) would have encouraged small parties to stand more candidates, and made district MPs (or more accurately, “the” plurality MP for each district) even less representative than under FPP by encouraging sincere voting for minor parties.

    I believe Germany used a one-vote version of MMP for the first Bundestag election or two, then switched to a separate list vote in the 1950s.

    Other variations are possible: for example, non-plurality district candidates who personally outpoll their party in their district could be moved up the list; below that, list order prevails.

  22. I came up with an idea for a MMP system that addressed the main flaw (in my view) of MMP, in that the individual members of the legislature holding the compensatory or “top up” seats are not in any real sense elected (and the voters can’t throw them out without throwing the entire party out of the legislature). In straight open lists systems, all members of the legislature are directly elected, both in the sense that each extra x% of the party’s vote always makes the difference between the marginal candidate getting in or not, and in the sense that voters can opt to move individual candidates up or down the list. Both these features are absent with current iterations of MMP.

    My version was inspired by the Jenkins Commission proposals and runs something like this:

    1. A number of members of the legislature, probably most, are elected from single member districts. In electoral systems that contain a proportionate element I don’t think it makes much difference whether the single member districts are filled by plurality or majority elections.

    2. A large minority of members of the legislature are elected via an open party list system.

    3. Each candidate running for a district seat must pair up with a party list in the same electoral region. The party list can have a single candidate, so its legal to effectively run a district as an independent and have a party list in the region consisting of your brother or your spouse. Also vice versa, a party running a regional list must endorse a district candidate if the list is to be on the ballot in that district.

    4. Voters get one vote only under this system. The ballot shows who is running in that district and the list they are paired with.

    5. If a candidate comes in first in a district, a number of ballots equal to the number of ballots received by his closest opponent are randomly selected and not counted for the regional list vote. Essentially, the ballot of an individual voter will either be counted in the regional list vote OR will be used to select a district MP, but not vote.

    This will probably result in a bonus of seats for the largest party, comparable to what the largest party already has in some purer proportional representation systems depending on the count used. In a situation where you have one big party and lots of small parties, the big party might wind up winning a lot of districts where their candidate gets 45% and the closest opposition candidate gets 20% and not get many votes deducted. But in this case there would be an incentive for the opposition to coalesce into a big party of their own.

    This does sacrifice the two ballot feature of most MMP systems, but should be workable, with some additional paperwork, for independents and regional parties, assuming the party list electoral district isn’t a single nationwide district.

    I think MMP has worked so well in Germany because Germany has a had a relatively honest political culture. But I’ve noticed when reviewing the results of Bundestag elections that a complete list of who the district winners are, where they ran, and how many votes they got are listed right away in the additional returns, but it was hard to get information for non-district members. I remember finding Merkel’s district and her margin of victory quite easily but still have no idea how her predecessor as chancellor wound up in the Budestag. This lack of transparency would become a real problem in a political culture dominated more by party machines and chicanery.

  23. As expected, Confidence and Supply agreements with ACT and United Future have been announced, including Minister-outside-Cabinet roles for those parties’ sole MPs.
    If National lose a seat on Special Votes as expected, this will give (59+1+1)/121.


    Expect less formal agreements with Maori and Greens as well.

    • Wouldn’t we expect confidence and supply support again from the Maori Party? With Greens, most likely continuation of the “Memorandum of Understanding” approach.

  24. I was trying to think of a better word than ‘formal’, and failed. A MoU is a formal document as well, after all. MaoriP won’t get Ministerial positons, I expect. Select Committee positions will be finely balanced.

    • The terminology can be troublesome, for sure. But I would think that the political science literature would regard a Memorandum of Understanding as “informal” in the sense that it does not commit the smaller party to supporting confidence and supply, nor guarantee it guarantee any specific policy outcome, let alone offices. Even if it is a written document, signed and publicized, it is “informal” in that it only commits them to working together outside of the “formal” governing arrangements on areas of mutual interest.

      • Errol, don’t you think both National and the Maori Party would want to continue the arrangement whereby the latter held at least the Minister for Maori Affairs? It seems as if it is in mutual interest for that to continue, even if the Maori Party wants a somewhat more arms-length relationship with the Nats this time around.

        If the Nats lose a seat on the final result, they may be somewhat more interested in keeping the Maori Party on board than they would be otherwise. On preliminary results, Nats + Act + UnitedFuture are on 62 of 121 seats. If they lose one, the three parties still have a majority, but can’t afford to lose either small party MP (or even one of their own). So an agreement with Maori provides more stability.

  25. There’s “formal” in the sense of legally enforceable by the courts (this would probably have to be in the Constitution if you want courts to get drawn into issues of hiring and firing cabinets and PMs – see cases mainly from Germany, with a few from India, Pakistan, and Malaysia).

    Then there’s “formal” in the sense of externally enforceable by some arbiter other than courts of law. I can imagine the President in, say, Germany, Italy or Finland – possibly even the UK Monarch if the breach was egregious – intervening to hold politicians to some commitment given in a written document (other than a statute), eg a supply ‘n’ confidence MOU or the Cabinet Handbook or (one alternative mooted in Australia – see George Winterton’s “Monarchy to Republic”, 1986) a non-judiciable declaration of “Westminster” principles.

    Then there’s “informal but enforceable through political means” – eg, if one party rats on its side of the bargain, the aggrieved party (in both senses) avails itself of the remedy of reprisal which, note, is the usual mechanism for enforcing rules of international law in the absence of a powerful judicial writ at that level.

  26. For reasons best known to themselves (I suspect somebody slips something into their baby good) Australian lawyers are utterly convinced that the rules of responsible government cannot be codified into constitutional law. It is a little like English-speaking journalists ‘knowing’ that all forms of PR are complicated. MMP was described as unbearably complex in coverage of this election in Australia.

    Australian lawyers ‘know’ this despite the majority of parliamentary republics and monarchies and the Australian Capital Territory itself having codified rules for changes of government.

    Winterton and others have suggested any number of ways to work around this inescapable fact and the non-justiceable, if codified, declaration of principles is one such example.

  27. Alan, I suspect the lawyers’ fear is less of having these rules written down than of having them interpreted and enforced through the judicial process – which could, in theory, see a premier removed by his/her own party, or via floor-crossing, “sue” in the courts for “unfair dismissal”. SA de Smith cited one case – African, I think, or perhaps the Subcontinent – where a court held that the head of state had acted wrongly in removing the chief minister based on a letter signed by an absolute majority of MPs; instead, the court ruled, it should have been a resolution passed on the floor of the Parliament. So the court ordered the premier reinstated. A month or two after the original removal. Alongside the case for seeking certainty by writing down is the case for seeking finality.

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 )

Google+ photo

You are commenting using your Google+ 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.