Tuesday, 30 January 2018

2020 NZ Election Hypothetical

2008 -- Somewhere North of Auckland

A beautiful stretch of coastline. Lots of sun. Nearby beaches. Absolutely gorgeous. And all about to be zoned for residential development by the new government.


National's just won another election. The coastline's still gorgeous. But, at last, the properties are about to start going up. Four for now. The owner-developers celebrate the election win long into the night at a caravan on site.


John, Paul and George go over to Ringo's. It's election night once more and the results are rolling in. It's a resounding victory. The four neighbours can't even remember why people thought this one could be tight. To be fair, neither can anyone else.

They take the party down to the beach, because, you know, it's gorgeous and right there... a stone's throw (or less) from Ringo's garden. And his back door's even closer! 

None of them voted Conservative. Colin Craig, David Cunliffe and all the other losers are mocked, but none more so than Kim Dotcom. The beach isn't sure what's so funny, but the neighbours are.


The neighbours aren't sure if they should celebrate or not. Bill English has just returned a fantastic result. But, like Labour's Jacinda Ardern, English seems to have cannibalised prospective coalition partners' votes to do it. Maori are gone. Without Dunne, United Future was Dunne-in. (The pub does not amuse.) ACT as pathetic as ever. TOP? Stuck at 2% ish. The Conservatives? Did they contest the election? It's going to come down to Winston. The mood is subdued.


There are surveyors at the beach today. Or people who look like surveyors. Ringo's daughter swears they took away some of the sand and some of the water with them. Surveyors don't do that, honey, says Ringo.

March 2019

These aren't surveyors. But they have a report. And some court orders.

August 2019

A Blue-Green Party is established. They call themselves the Turquoise Party. Ringo makes a joke about Aquamarine. It makes him remember the beach. He cries. A lawyer hands him a tissue.

January 2020 -- Ringo's Sister's House, Remuera
It's over. The house has gone. Not literally. For now. They're pulling it down in March. But Ringo hasn't got the keys. He hasn't got the land. He hasn't even got hope. They lost the last appeal just before Christmas. Ringo could tell it was over. John didn't even join them this time. Ringo just focusses on getting things set for school: kind of his sister to let his daughter live with her this year. The verdict will come down any day now. Best to focus on other things. Everyone says.

February 2020 -- Ringo's Cockroach Infested Flat, Dominion Road

"Consents should never have been granted to build in the area." "Gross negligence." "Fundamental failings." "Severe damage to sensitive environment." "Erosion." "Unreasonable reliance." "No compensation."

John, Paul and George came over after the verdict. Ringo's sister told them to push off. They went to a pub, and on the spur of a moment decided to flat. They "outbid" thirteen different groups of students for an awful flat. A month later, they're still in it... and really glad their families could live with other family members in sizeable Leafy Suburb properties. Although a bit confused about why they're not there too.

Ringo stubs his toe. Blames John Key.

George remembers why they're flatting.

Election Time 2020 -- Remuera

Ringo knows one thing. He's not voting National. Damn their environmental record, he says. His wife's not convinced. His sister isn't convinced. John, Paul and George are. Ringo's sister mentions Labour. Everyone laughs. Bloody socialists, says Ringo's daughter. John slaps Ringo on the back, "Good, on ya mate." George chips in, "And the Greens are even worse!" They toast the continued ill-health of the Green Party's polls.

"Who you're voting for this year Ringo?" It's Ringo's boss. She's a committed Nationalite.

"Oh, I want a blue-green party."

"A Turquoise man, eh? I'm going to vote Turquoise myself this year."

Election Day 2020 -- Ringo's Polling Booth

This is it. The moment where a lifetime's worth of National voting was undone. The moment where Ringo became a Turquoise Party voter. 

Wait... what? The Party List? #2 John Key? That bastard was in the Turquoise Party?! And #5? Nick Smith! The prick!

Ringo party votes for Winston Peters. He's not happy. It makes him electorate vote for the Greens candidate. He doesn't even read the name.

Election Night 2020 -- The Cockroach Flat

It's an election night miracle!

National 15% Labour 35% Turquoise 20% The Conservatives 4.7% NZ First 8% The Greens 8% TOP 4.3% ACT 1% Maori 2% Mana 1% and collected unimportant parties 1%.

But it turns out Turquoise won only the one electorate seat. They've split the party and seat votes of the centre-right! National manage to hang on to exactly 15% of the total seats through the electorate seats only (18). Bill English, their leader, won't be in parliament!! Maori and Mana have come storming back, winning a seat apiece! But neither TOP nor the Conservatives have managed to grab one! Disaster!

But there's still going to be an overhang! (I think) And the wasted vote is ridiculously high at 10%!

What's Going on Here?

If you ask me, this hypothetical illustrates a number of issues that are currently present in New Zealand's electoral system.

Firstly, we've got the problem of wasted vote induced by the stupidly high threshold. By my hypothetical the Conservatives and TOP should have 6 (5.64) and 5 (5.16) seats respectively. Instead, they walk away with none because they didn't win any electorate seats.

Secondly, when things go wrong, NZ's MMP system actually gets terribly complicated. And difficult to understand. Luckily there is an "MMP SEAT ALLOCATION CALCULATOR" so we can use that to see what happens here, but first some further notes on the example:
  • From above, I have 18+4 seats already allocated (1 each for Turquoise, Maori, ACT and Mana, 18 for National).
  • I'm arguing that the split votes allow Labour, Greens and NZ First to win some electorate seats. I say that this leaves 5 Maori Seats for Labour, 2 extra Labour seats, 28 more Labour seats and 7 apiece for Greens and NZ First.
Hence the result:

So, I was wrong about the overhang, although we could easily change that with different assumptions about the electorate seats won by Labour, the Greens and NZ First. Also, apparently, Bill English will be getting in to Parliament. Let's say Labour wins two fewer electorates and National gets them instead...

We'll use this version as the "true" results of the example when I go on to use the example for posts. But let's see what happens if the Conservatives win one of those extra seats and TOP the other.

You can see how radical the departures from the democratic will NZ's 5% threshold can induce immediately, right?

In the first example, we're looking at a likely coalition of Labour (46), Greens (57), Maori (60) and Mana (61) or Labour (46) and Turquoise (73) because the alternative would rely on Maori or Greens working with National.

In this new version, we're looking at:
  • Turquoise (24), National (42), NZ First (52), the Conservatives (56), TOP (63) and ACT (64)... or maybe without ACT
  • Labour (43) and Turquoise (67)
  • Labour (43), Greens (53), TOP (58), Maori (60) and Mana (61)
  • Labour (43), Greens (53) and NZ First (63)
  • A minority government by Labour relying on ad hoc support of any of the other parties/these voting blocs
  • A minority government by Turquoise relying on ad hoc support of any of the other parties/these voting blocs
Very different worlds, no?

Although, one suspects that if there was to be a Turquoise Party of such attraction to National voters, it would not be a party particularly inclined to work with Labour. Similarly, if environmental concerns were to suddenly become this influential, Labour's electoral performance would reflect that. But, remember, this is a hypothetical, a thought experiment, designed to illustrate certain points not a model.

Thirdly, you can see immediately why parties have incentives to behave in ways that are contrary to the health of our democracy.

National's various victories under John Key and its 2017 performance under Bill English all represent what happens when the centre-right vote isn't split. In contrast, people who lean left vote for a bunch of different political parties. In the example, you see what National's afraid of... becoming second fiddle. Now, those concerns aren't really reasonable, but if you can control the entire right bloc of votes, you end up with a lot more power... and can even push yourselves towards single-party government.

Now, you might say that the last election shows why parties also have an incentive to not behave like this. If National had helped the Conservatives out in 2014, it's likely they'd have managed to be around now even with the Colin Craig nonsense... which might have meant that National could have formed another government (some of NZ First's voters would be attracted to the Conservatives). But my example has a rebuttal to this argument.

You see, in New Zealand, we still use first past the post for electorate results. As a consequence, split votes are a big problem. This is a well remarked on problem with FPTP. I'm not sure what we should do to fix it... preference voting like the flag referendum? Borda Counts? But we need to plug this backdoor issue. Keeping new parties down, helps avoid losing electorates where your "kind" of party is dominant. It stops something like the above from happening... where a party vote swing caused electorate seat collapse due to split votes. Here's what happens if National's "lost" 23 seats went 12/7/1/1/1 to National/Turquoise/Conservatives/Mana/Maori:

National has now won way more seats than it should have. It is over-represented... essentially, MMP has failed to achieve proportionality because the size of parliament doesn't change to even out the overhang, it just adds on the extra electorate seats. It's actually possible that this sort of thing would be more likely with a better electorate seats measure, but it's fairly easy to fix.. we'd just need to be able to add on another 70 list politicians in the case of such extreme overhangs.

The reason I say "actually possible" here is because to investigate we'd need to deal with a realistic example. This one is contrived to make a very specific point very, very obvious. That's an okay thing to do. But the truth is that sometimes examples have to be realistic in order to gain any insight from them. If we wanted to evaluate what an alternative model for choosing the winner in the electorate seats would be, we'd need a realistic example. But what I've done is sufficient to lay bear the incentives parties have to not change the threshold and make new political parties credible votes. You do that and you not only dilute your own "power" but you open yourself up to split votes in safe seats. These are deep problems with our current system.

Fourthly, let's talk about the disconnect between New Zealand's broader constitutional principles (as I see them, anyway) and the electoral system... as seen in coalitions.

There are different ways of understanding constitutions. The one that I think is most widespread (at least among interested lay people such as myself) is the view that the constitution is the rules of government and thus the limits on its power. Sure, it says how everything meshes together but it's the limits on what government can do and how that matters most.

The way NZ's constitution limits/contains government power is fivefold. Namely:
  1. By allowing the GG to dismiss the PM. This would cause a constitutional crisis and it's not clear who would be blamed (or what would happen). This gives the GG incentives to not dismiss the PM and the PM incentives to not push it.
  2. Short terms of three years with transparent government and proportional elections. The idea is that if a government oversteps, it will get punished at an election which always just around the corner. The electorate is given lots of power and assumed to be willing to wield it.
  3. Through coalition government. It's necessary in a coalition for a government to keep several different houses in order, this gets really difficult if dubious stuff is on the cards. You could almost understand this as the belief that a junior government party will find its integrity faster.
  4. Faith. What I mean by this is that while the New Zealand Bill of Rights act cannot prevent legislation contrary to its will from passing, it does require all legislation be compared against it. What I mean by this is that we have citizen's initiated referenda which can't bind the government but they can make it think twice... or give energy to the opposition.
  5. Parliamentary politics. Firstly, we have to acknowledge the Opposition as an idea. A strong Opposition forces the government parties to engage with bad things they're doing instead of just passing them and moving on. Secondly, the role of Opposition party MPs in committees and the like means that the government has to choose which committees it will have a majority in.
Compared to some other countries, there isn't really much that stops a government from doing what it wants. Consider that:
  • Votes are almost always conducted on a party basis. Not in the sense that the whips have to corral parliamentary cats but in the sense that a party says it has X members and thus has X votes which will be voting in whatever way. In the UK, for instance, the party leader basically coerces MPs to vote the party line by holding the ability to punish and reward members in accordance with their behaviour. The whips in the UK are the agents of this coercion.
  • Governments are frequently able to bypass parliamentary measures. The National government we just had was notorious for passing legislation under urgency. It seems to me that this is a fairly easy fix... require the ability to pass legislation under urgency to be decided by a super-majority and if that passes, the legislation still has to get the majority vote. The problem with using urgency frivolously is that slowing the passage of bills down gives more time for opposition to/doubts about them to coalesce/form. As Terry Pratchett once wrote, “People don't like change. But make the change fast enough and you go from one type of normal to another.” Irrelevant filibusters aren't a good means of slowing something down, but it's the same core idea.
  • The PM is essentially God. In theory the prime minister is just the first among equals relative to their Cabinet colleagues... the reverential deference of the Holiness the President of the United States, Serene Ruler of the Americans as seen in the US version of House of Cards or Designated Survivor is completely inappropriate. However, the PM has much more power because in practice what Cabinet says go, and the PM gets to control who's in Cabinet... and one doesn't become a PM without being persuasive (and Cabinet is small enough for groupthink). In the USA, their president has to jump through all sorts of hoops as a result of not being part of the legislature. I don't think separating powers out is necessarily better, but if you don't let the system mesh properly then not separating powers runs several risks.
  • We don't have any Supreme Legislation. Now, I don't see why we couldn't make NZBORA supreme if we wanted to. This would have the net effect of preventing legislation that falls afoul of NZBORA from being able to be passed. This doesn't even have to be in conflict with parliamentary sovereignty... by definition parliament is able to make a supreme version of NZBORA that isn't even double entrenched. In the USA or Australia, the Supreme or High Courts are able to strike down unconstitutional legislation. I don't actually like this system... a lot of the cases the US Supreme Court deals with are actually political questions unsuitable for a court, as was the Barnaby Joyce citizenship saga in Oz. But, the point here, is that this is a kind of "check" on government we lack.
  • There is no upper house. Again, I'm not actually convinced that Upper Houses are needed. The "ideal" kind is filled with experts who are appointed until they wish to retire. But even substantially more imperfect ones still function as a "check" because it means more people have to lack integrity for bad stuff to be enacted. At the moment we're literally one vote from an Act of Parliament saying all Blue Eyed Babies must be incinerated. If we had an upper house we'd be two votes away...
If you're at all familiar with American politics I am sure you've heard of the phrase "checks and balances". Well, I don't believe in checks... but I do think we've got a "balance" problem in NZ. And that's the type of thing that coalitions represent, and the electoral system is creating the associated balance problem. (The difference between checks and balances is that the former stops or changes the way something happens whereas the latter plays with incentives within the same process, i.e. actors balance/tradeoff different concerns/objectives in their decisions.)

Let's take another look at the potential coalitions from the ("true") example election:
  1.  Labour (46), Greens (57), Maori (60) and Mana (61) 
  2.  Labour (46) and Turquoise (73)
  3.  Labour (46), Greens (57), NZ First (68)
  4.  Turquoise (27), National (47), NZ First (58) and Greens (69) / Maori (61)
As we can see some of these aren't really plausible. We could also chuck in some extra partners if we wanted... parties do do that sort of thing, see: Maori in coalition with National.

When I first wrote about coalition options last night I thought (b) was a realistic option. In theory, it has to be. However, unless we want a complete breakdown of the "suspension of disbelief" it really probably needs to be excluded, even though the point of a blue-green party is that it will go either way... unless we see the idea as a "neoliberal" environmentalist party. Thus, we're actually looking at another case of where a realistic example is really required.

The question of why I didn't think about (c) is a difficult one to answer. After all, this is the current government's arrangement. I guess I just got lost in my hypothetical. In fact, this would quite likely win out. While I think the Greens, Maori and Mana would all be relatively happy to work together, there is a bit of bad blood with Labour and they have no slack... 61 is the absolute minimum number required for government. And (d) just makes the bad blood even worse because it's now National.

But here's the thing... coalition negotiations ought to reflect the relative power of the different parties involved... tradeoffs. If we're thinking about how (d) might happen, it would be very difficult for Maori to encourage it over the line. They need the presence of NZ First and only just manage to bring the numbers up. As a consequence, Maori would have to negotiate few policy concessions to make up for its relatively low number of seats. The Greens, in contrast, have a fairly good negotiating position to extract concessions... they're involved in a lot of other potential coalitions and they have a reasonable number of seats.

Now, you might say that the Greens have disproportionate negotiating power. Well, that's not true. The Greens are never able to get more than their 11 seats worth because every concession that they get has to be balanced with what prospective other partners want. And, after all, Labour would be able to say, "Waoh, hold your horses, remember the Turquoisers?"

Once the negotiations are over, political parties have basically only as much power as their seats allow them. That's the first rebuttal of the anti-micro party wing. The second is basically an extension of the above. If we had a parliament of 120 parties the problems with that would be fairly negligible... negotiations would probably take forever but they'd be sped up enormously by the fact that anyone who tried to power play literally can't offer anything more than their next best alternative... so micro-parties ultimately cancel each other out. But I've distracted myself. Remember the first rebuttal?

When you look at the 2017 election, NZ First might have had a bit more power than it should have had. It was able to assume that the Greens would stick with Labour no matter what which allowed it to be fairly ambitious. If there had been actual substantial policy differences with Labour on the big things this could have been an issue. Indeed, NZ First is probably NZ's most modern political party... even though it was far more similar to Labour (despite its leader's ex-National background I might add), NZ First was willing to consider a deal with National. But the point is that 2017 reveals the constitutional threat posed by electing few parties to Parliament. We really do rely on coalitions to stop a party forcing through objectionable policies. And hence we rely on an electoral system that will return many rather than few parties.

Finally, I'll just note that I also talked about some subsidiary issues above. Believe me, I wouldn't spend all day typing (I wake up at 1pm, sue me) if I did not actually think I was talking about real problems. The issue is that the reason why I spend all day typing is my limitations as an author... which is why these issues are being labelled subsidiary issues even though the electorate seat issue is something that I feel quite strongly about. It's also true that there are other problems involved here. After all, I didn't talk about rotten boroughs... that's only vaguely represented in this case.


I will be using this hypothetical for at least one more post... that's why I started off with Ringo.

No comments:

Post a Comment