2020 New Zealand general election & referendums (17 October)

[Lord Halifax]

Curious if there's any reason why the South Island is restricted to exactly 16 seats while the North Island's number can vary. Is the population size and distribution really that much more dynamic?

It's to guarantee it a minimum representation, otherwise it would gradually lose seats.

[Pericles]

It seems there was a big age divide in the cannabis referendum. It isn't really surprising, but the scale of it is interesting nonetheless. https://thespinoff.co.nz/politics/16-11-2020/new-zealands-cannabis-referendum-results-were-defined-by-age/

[CumbrianLefty]

So that one is just a matter of time?

[Gary JG]

Curious if there's any reason why the South Island is restricted to exactly 16 seats while the North Island's number can vary. Is the population size and distribution really that much more dynamic?

South Island had the larger population in much of the 19th century, but its relative share of the New Zealand population has been falling since. Extract from the 1990 edition of the New Zealand Official Yearbook.

"The initial alluvial gold rushes had ended by the 1870s. However, after a short lull high levels of immigration were maintained, bolstered by Vogel's assisted passage scheme. From 1867 the proportion of the total population resident in the North Island began to steadily increase. A strengthening economic base and suppression of Maori opposition to land alienation made the prospect of settling in the North Island more attractive. Farming in the North Island became increasingly feasible and profitable with innovations in agricultural methods and, in particular, the development of refrigeration. Large estates were also being broken up by the 1890s and this contributed to more intensive settlement. The greater opportunities seemingly available ensured that the North Island became the usual destination chosen by both external and internal migrants. At the same time the South Island lost people due to migration and the North Island population experienced a higher rate of natural increase. By 1901 these factors resulted in the population of the North Island again exceeding that of the South Island.

Since that time the North Island's population has continued to expand at a greater rate, and its share of the total population has continued to grow. A little over 60 percent of the population resided in the North Island by 1916. This figure rose to 70 percent by 1966 and to over 73 percent by 1986.

From the 1940s the centre of gravity of the North Island population has itself gradually moved further north."

[Pericles]

Also, this is a good breakdown of the results. I'm surprised Simon Bridges only won by 4%. How the electorates lean based on the party vote is also interesting.

[CumbrianLefty]

Only found out yesterday that the often quoted (almost exact) 50% share for Labour is referring to their list vote, the constituency vote was actually slightly lower at "only" 47%.

I think its pretty unusual for the winner at least to have those figures that way round?

[Pericles]

Only found out yesterday that the often quoted (almost exact) 50% share for Labour is referring to their list vote, the constituency vote was actually slightly lower at "only" 47%.

I think its pretty unusual for the winner at least to have those figures that way round?

Where did you find those numbers? I'd like to see them. It makes some sense, people are more attached to their local MP than the party brand so National has a higher floor in the electorates. Maybe it's also an old lingering FPP mindset, so maybe people are splitting their votes the wrong way.

[Pericles]

Wow, some amazing person put the electorate votes for each party in each MMP election up on Wikipedia. I made these two graphs from the data.



This basically illustrates how opposition parties have lower floors in the party vote than in the electorates. I don't have the 2020 electorate vote numbers though, if CumbrianLeftie has numbers for National and Labour that would complete the comparison.

Also, random thought about the 2020 results but I'm surprised about Takanini. This new electorate was supposed to be National-leaning, instead Labour won it 55%-30% in the party vote (basically the same as nationwide) and 53%-32% in the electorate vote. Here is the electorate profile. It's a younger electorate than average, it's pretty ethnically diverse-actually plurality Asian. Education levels seem pretty similar to NZ overall. It's possible that Asians swung to Labour at a higher rate than expected. Labour's immigration stance appeared to depress its 2017 vote among that demographic and so maybe that not being much of a factor gave Labour an additional boost, or there is some other explanation. I might need to research this further.

[CumbrianLefty]

Sorry, just citing what somebody else (who I have no reason to disbelieve) said re the 47% figure.

[Pericles]

So here are the final National-Labour swings, in each of the general electorates that remained very similar or the same since the last election (this does make the analysis less reliable). The nationwide swing was 16.0% by the way.

Auckland Central-13.2%
Bay of Plenty-19.7%
Botany-21.0%
Christchurch Central-16.1%
Christchurch East-15.8%
Coromandel-17.1%
East Coast-15.3%
East Coast Bays-20.7%
Epsom-16.3%
Hamilton East-16.8%
Hamilton West-18.5%
Hutt South-17.1%
Ilam-18.3%
Invercargill-15.3%
Kaikoura-18.9%
Kelston-13.7%
Mana-15.5%
Mangere-7.1%
Manurewa-15.0%   
Mt Albert-10.3%
Mt Roskill-16.0%
Napier-15.5%
Nelson-15.4%
New Lynn-15.2%
New Plymouth-16.4%
North Shore-19.5%
Northcote-18.1%
Northland-16.3%
Ohariu-19.3%
Otaki-18.5%
Pakuranga-20.3%
Palmerston North-15.8%   
Papakura-11.6%
Rangitata-18.3%
Rangitikei-18.6%
Remutaka-17.4%
Rongotai-11.4%
Rotorua-17.0%
Selwyn-20.5%   
Tamaki-19.3%
Taranaki-King Country-17.7%
Taupo-18.7%
Tauranga-17.7%
Te Atatu-16.3%
Tukituki-16.7%
Upper Harbour-19.3%
Waikato-18.1%
Waimakariri-21.2%
Wairarapa-17.8%
Waitaki-18.2%
Wellington Central-10.6%
West-Coast Tasman-12.4%
Whanganui-16.8%
Whangarei-17.3%
Wigram-16.6%

Most of the correlations were weak.





There was one that was very strong, but this is unsurprising.


Now for some trivia from these. The most educated electorate of these is Epsom, which had an average swing but interestingly was the one electorate to party vote National while still not electing a National MP lol. The least well educated electorate is Mangere, which was a huge Labour landslide as it always is but also had the weakest swing of all of them. Mangere also has the lowest share of its population being 65 years or older. The oldest of these was Otaki, which was a Labour gain. The electorate with the highest Asian share seems to be Mt Roskill, which is a Labour seat with an average swing. I'll let you have a look and see if you find anything interesting and worth questioning perhaps.

[𝕭𝖆𝖕𝖙𝖎𝖘𝖙𝖆 𝕸𝖎𝖓𝖔𝖑𝖆]

1. Mangere has a lot of working class Pacific Islander immigrants I assume?

2. What is the other electorate just below Otaki in the age graph and why are the two of them so significantly older than all the others? Big retirement communities à la The Villages, FL?

[Pericles]

1. Mangere has a lot of working class Pacific Islander immigrants I assume?

2. What is the other electorate just below Otaki in the age graph and why are the two of them so significantly older than all the others? Big retirement communities à la The Villages, FL?

1. Yes, it is 59.7% Pacific Islander.

2. It's Coromandel. Actually I got it wrong, Otaki is the second oldest and Coromandel is the oldest with 28.4% being 65 and older. As you can see, it had a slightly above average swing, but is still held by National. I don't think there's anything quite on the scale of The Villages though.

[Pericles]

I just saw the Wikipedia article has been updated. Labour received 48.07% of the electorate vote to 34.13% for National. There was a 10.06% swing from National to Labour in that vote, compared to a 15.99% National-Labour party vote swing. National's electorate vote was 8.55% higher than its party vote. It's the third largest gap between the electorate and party vote for one of the main parties, with Labour 2014 being second and National 2002 being first still. The difference between Labour's margin in the party vote and its margin in the electorate vote does seem to be the largest ever though.


This is what I expected, given how National only won the party vote in one electorate that they don't even hold.

[Estrella]

I've already mentioned this account in the France thread, but someone on DeviantArt made maps and write-ups of every New Zealand election from 1890 to 1990. Definitely worth a read.

[Pericles]

I did some more calculations, this time on the difference in between National's electorate vote and its party vote in all the seats it won in 2017.
Auckland Central-6.1% (Open)
Bay of Plenty-11.9%
Botany-13.8% (Open)
Coromandel-12.3%
East Coast-8.9% (Open)
East Coast Bays-15.5%
Hamilton East-12.2%
Hamilton West-11.1%
Hutt South-19.2%
Ilam-8.2%
Invercargill-15.0%(Open)
Kaikoura-12.0%
Kaipara ki Mahurangi-11.0%
Maungakiekie-16.1%
Nelson-17.3%
New Plymouth-11.7%
North Shore-11.5% (Open)
Northcote-13.9%
Northland-9.9%
Otaki-14.9% (Open)
Pakuranga-19.6%
Papakura-12.1%
Port Waikato-2.5%
Rangitata-8.0% (Open)
Rangitikei-14.6%
Rotorua-14.0%
Selwyn-15.0% (Open)
Southland-14.8% (Open)
Tamaki-14.8%
Taranaki-King Country-9.8%
Taupo-15.4%
Tauranga-10.3%
Tukituki-12.1%
Upper Harbour-6.5% (Open)
Waikato-13.3%
Waimakariri-17.7%
Wairarapa-5.3% (Open)
Waitaki-13.3%
Whanganui-8.2%
Whangaparoa-15.8%
Whangarei-13.3%

So National did better in all of its 2017 electorates with the electorate vote than the party vote, and the average difference was 12.4%. There seems to have been a marginally worse performance in seats where an incumbent was not standing. Seats like Upper Harbour and Rangitata would have been won if they weren't open (well if Andrew Falloon hadn't disgraced himself), while it's possible National could have won in East Coast and Wairarapa. The result there does show that Kieran McAnulty was a strong candidate for Labour, since National didn't seem to make any special errors there. Port Waiikato stands out as an underperformance, especially since Andrew Bayly is now number 3 on their frontbench as treasurer, but it seems a right-wing splinter party randomly got 21% of the vote there. Gerry Brownlee's loss may also have been especially avoidable as his overperformance was below average, perhaps he was a weak incumbent for his electorate (and his vague Covid conspiracy theories were unhelpful). The standout performances are Chris Bishop in Hutt South and Simeon Brown in Pakuranga. Bishop lost anyway because his electorate is Labour-leaning. Simeon Brown is interesting because he is a social conservative, and from what I've seen he has a reputation for making gaffes (he also looks a bit like an even more baby-faced Pete Buttigieg imo, take a look). Despite that (or maybe Pakuranga is into social conservatism), he actually has the largest majority of any National MP now.

[Pericles]

There has been some funny news related to the 2020 election. Somehow, Judith Collins is still National leader (probably because nobody else wants the job quite yet), and the election review she promised has finally been completed. National is not actually giving its MPs a copy lol, and is claiming that the review is property of the party board and MPs get a sanitised version without the 'gory details'. They don't plan on making either version public, which is disappointing since I wanted to read it, but I'm optimistic that a version will get leaked. Clearly they still aren't a strong team.
https://www.newshub.co.nz/home/politics/2021/03/national-creates-two-versions-of-election-review-one-with-gory-details-taken-out-shared-with-mps.html