One More Nightmare: Group Tickets And The 2021 WA Upper House

Sorry to be so ranty about this, but I think that it has to be said.  The 2021 Western Australian Legislative Council "election" is a bad joke.  Voting systems should give voters real choices about how they vote and should only reward parties with serious voter support.  WA politicians have failed and disrespected the voters by doing nothing at all to deliver this.  Instead voters are still coerced, eight years after the 2013 Senate farce, into propping up another dishonest Group Ticket grift party fest.  If you can do it without making a mistake, vote below the line.

As noted four years ago (sigh) Western Australia's upper house has the worst voting system in the country.  The chamber is grossly malapportioned, with votes in two bush regions being worth nearly four times more than votes in Perth suburbs.  

Voters can vote by putting a one above the line, in which case their vote may be sent goodness knows where as a result of backroom preference deals and deliberate preference harvesting.  They cannot give preferences between parties as they can in the Senate.  The only other way in which it is permitted to vote formally is to vote below the line.  If you do this you must number every box (about 54 on average) and if you leave any box blank or skip or repeat any number in the sequence, your vote is informal. 

 If you want to vote below the line and try to stop preference harvesters from trying to elect unaccountable and perhaps extremist nobodies off tiny shares of the total vote, then I recommend preparing your vote at home beforehand. I also recommend putting parties you have never heard of below those you merely dislike, and ahead of only the worst of the worst.  If I see any links to any tools for doing so, I'll post them here.  Don't forget that if you make a mistake while filling out your vote, you can ask for a new ballot paper.  Also, you can vote above the line as well as below - if you do this and your below the line vote is informal by mistake, your above the line vote still counts.

In 2013, Group Ticket Voting helped cause the WA Senate election to be rerun, after nearly electing an Australian Sports Party candidate off 0.2% of the vote.  (The direct cause of the rerun was the loss of some ballot papers, but it wouldn't have mattered but for the way in which the Group Ticket system can create very close tipping points between irrelevant parties.)  This should have been the end of the system in WA, but it wasn't.  It was used again at the 2017 state election, but the damage was relatively minor.  The Liberal Democrats (3.9%) won an undeserved seat in South Metro at the expense of the Greens (9.3%) and One Nation (7%), with the assistance of voters confusing them for the Liberal Party.  The Shooters (5.7%) leapfrogged One Nation (11.7%) for an undeserved seat in Agricultural, but the Shooters did have some sort of vote there and besides it was only One Nation, so no-one much cared.  

That WA had got off light should have been highlighted by the incredible results of the 2018 Victorian upper house farce in which fully a quarter of seats were won by preference-harvesting candidates who had no genuine claim to be elected.  Although voting for micro-parties is lower in WA, voters are also less likely to use below-the-line voting to try to wreck preference harvesting there, and Victoria should have been another massive wakeup call to the WA Parliament to do something about Group Ticket Voting (and hey, fix malapportionment while you're at it.)

For doing nothing about this the Western Australian Parliament gets my 2020 Grand Gerry Award for the worst contribution to electoral reform in a calendar year.  (The 2019 winner was Barnaby Joyce.)  And no, a pandemic is not an excuse, this could have been fixed quickly and should have been fixed years ago.  If anything, the extremist views the pandemic is bringing out only increases the importance of ensuring that politicians are elected based on real voter choices. The blame seems to be mostly Labor's, but the conservative side are not helping by continuing to signal that they want to keep malapportionment, which may discourage Labor from scrapping group tickets.

The 2021 Micro-Parties

The 2021 contest (I am trying to avoid the word "election") sees a modest ramp-up of the preference harvesting attempts seen in 2017.  As in 2017 there is a favoured party in each region who several of the tickets have been set up to try to elect, but if that party bombs out early in the cutup, it is possible for others to take their place.  The number of parties has slightly increased, and the effort has been bolstered by the running of lower house candidates who are running solely for the purpose of building ballot paper recognition for the upper house ballot.  There have also been some misleading party renamings.

The favoured micro-parties in each region are:

East Metropolitan: Western Australian Party.  Charles Smith (elected in 2017 for One Nation but left) is the candidate for this party.

North Metropilitan: "Liberals for Climate".  This is a rebadge of the abysmal techno-gimmick fad party formerly known as Flux (also Flux The System!).  A review of Flux's awfulness can be found in my 2017 article.  The party has nothing to do with liberalism, Liberals or climate.

South Metropolitan: Liberal Democrats. This is Aaron Stonehouse, the incumbent elected in 2017 but this time he has drawn very badly on the ballot and will struggle to get a substantial primary vote.

Agricultural: "Health Australia Party". This is another misleadingly named party that in Western Australia's case is a rebadge of the former Fluoride Free WA.  See below for more on this party and its candidate here.

Mining and Pastoral: "Daylight Saving Party."  Does this party even exist or have anything to do with daylight savings?  Who knows; its Facebook page has seen no action since 2017 and it has no website.  That it tried to change its ballot abbreviation to "National Liberals" (which was rejected) should tell you all you need to know.

South West: "Sustainable Australia". This well-known micro-party makes itself out to be a kind of alt-Greens party for centrists, but it is not centrist.  Its main schtick is arguing for immigration restrictions on supposedly environmental grounds that are often laced with xenophobic dogwhistles.  In Victoria the party's MP -while expressing opposition to the most extreme conversion practices - still voted against a government motion to ban gay conversion "therapy".

In general these parties are strongly favoured by the preferences of most of the obscure "independent" tickets in their region (usually they are the highest micro-party preferenced), frequently without any quid pro quo and suggesting that many of the "independent" tickets may be also vote-harvesting devices, although some don't follow this pattern and appear genuine.  

Other parties who have been flagged as competitive in my own scenarios include Australian Christians in East Metro and South Metro (religious right parties are tanking at the moment, but perhaps not if the Liberal vote collapses) and according to Antony Green "Liberals for Climate" in South Metro and Mining and Pastoral, and Legalize Cannabis in South West Metro.  Please feel free to post your own calculator discoveries in comments, or if you can't or don't want to sign up here, please feel free to email me (link in profile) with information clearly marked as for publication.  The ABC calculators are here.  They are not perfect, as they assume all votes are below the line, and in 2017 they also had a distribution order issue that caused them to get one winner wrong (I don't know if this has been repaired.)

The "Health Australia" Spiral

The worst of the preference snowballs that could occur in this election involves the Agricultural region and the Health Australia Party, whose candidate is Bass Tadros.  Tadros describes himself as a "Life Coach, Hypnotherapist, Relationship Coach, Trainer & Television Personality".  Tadros' form guide, from his social media, includes:

* On 5G: "Any Living Being exposed to a New Magnetic Field is Poisoned !!!

Some survive it … Some don’t!!! …

but this excessive electrification of the Earth is not compatible with health!"

* On COVID: "Great to see real leaders like Craig Kelly sharing real information that actually makes sense [..]"

* On vaccines: "as ALUMINIUM is more and more present in Our Bodies via Vaccines, “Care” Products, … Food Packaging, …, … they become Great Receivers to absorb more Electromagnetic Fields !!!"

* Again on vaccines: "Is vaccination playing into the hands of those interested in some sort of ‘reset’?

Will it provoke even more ‘variants’, rather like antibiotics provoke bacterial resistance?"

* On lockdowns: "Was the lockdown convenient? Is the community being held to ransom with fear as the main tactic? Are politicians manipulating the disease data to ensure they appear to be “on top of the virus”? … and “acting in the public interest”? Are we being pushed – or running like lemmings – into the ‘great reset’ in the interests of power brokers and corporate big business?"

There's plenty more of this stuff.  Tadros' views clearly align with a general 5G/vaccine/lockdown conspiracy theory mindset normally considered of the fringe right.  And yet, the Greens have preferenced him directly after themselves.  It's not even clear what if anything the Greens get out of it, or if they are just suckers who have assumed the Health Australia Party are some kind of harmless natural remedies party that their supporters would consider likeminded.

In my initial simulations with the ABC calculator, Tadros sometimes wins and sometimes doesn't, but in cases he gets up off as little as 0.2%.  

I found an especially perverse scenario as follows.  Firstly I took the votes basically as they were at the last election, adding a few votes for new micro-parties that seemed like plausible guesses of what they might get.  Tadros didn't win, the result was 2 Labor 2 National 1 Liberal and One Nation.

Then I took 3% off the Liberals and 2% off the Nationals and gave 5% to Labor.  This resulted in Labor now losing their second seat to Tadros.  The reason this happens is that a possible key point in Agricultural involves whether or not Health Australia can get over Shooters, Fishers and Farmers, who are carrying the National excess (if there is any) over their second quota respectively.  If Health Australia survive at this point, votes come to them from One Nation and the Shooters via the One Nation surplus.

In these simulations a massive leg-up Health Australia receive is the Greens votes, unless the Greens can poll well enough to beat the combined total of Health Australia itself, Daylight Savings, WA Party, "Liberals for Climate", Legalize Cannabis, Liberal Democrats, SUS AUS, Animal Justice and several supposed independents.  (Fake independents have been seen in these contests before).  Perhaps the Greens drawing first on the ballot will be enough to do this for them, but that's optimistic.

Health Australia stopped winning when I took the long handle to the One Nation vote, but then when I looked at a scenario where Labor gained 10% off the Liberals, Nationals and One Nation, they won again.  These calculator scenarios are often fragile - sometimes even the tiny rate of below-the-line voting that WA has can be enough to wreck them - and we really have no idea what some of the micros (like NO MANDATORY VACCINATION) might get in the current environment.   But that Tadros has often popped up as a plausible winner should be a major concern.  Why on earth did the Greens preference him?  

Greens voters in Agricultural (map here, includes Esperance and Geraldton) can't get around this by voting for left-wing micro parties, since all of them feed Tadros too.  Voting Labor above the line is a lower-risk option, since Labor are often in the race for the last seat, but Labor votes could also potentially feed into the HAP spiral or maybe elect the Shooters.  The only safe solution is to vote below the line - very carefully!

Majority Prospects?

There has been a lot of interest in whether there will be a Labor majority or at least a Labor-Greens majority.  I find that a Labor majority relies on a massive primary vote landslide (something along the lines of the Newspoll result for the Lower House), and if that doesn't happen then a Labor-Greens majority is far from certain.

What I have found so far is that the Greens, who presently hold four seats, are very vulnerable.  The four main scenarios I tried were: no swing, 5% swing from Liberal/National to Labor, 5% swing from Liberal/National to Labor with One Nation vote halved, and 10% swing from Liberal/National to Labor with One Nation vote halved.  There was no seat that the Greens always won in these scenarios, and in the latter two they finished up with only 1-2 seats.  Their problem is that an awful lot of tickets put ALP/Greens last, and the Greens primary isn't near a quota anywhere.

This means that even with a medium-sized swing to Labor, consistent with a Lower House 2PP of around 61%, there's no guarantee of a Labor-Greens majority.  In my first runs of the 5% swing scenarios, the combined Labor-Green total was only 17 and 18 seats respectively.  In the 10% swing scenario Labor do get to 19 seats and a majority in their own right, but that's assuming that a swing that large in the Lower House would be matched upstairs.  The difficulty of getting a majority for Labor and the Greens is increased by three things: Group Ticket Voting, malapportionment, and the system of having 6 seats in every region rather than an odd number.

In contrast to the Greens, One Nation seem to do rather well for their vote level, and I found chances for them in every seat if the vote shares do not change very much.  Even taking half their votes away, I still found them winning between 2 and 4 seats in my initial attempts, though this may not be reliable.

The 2017 results are quite worth having a look at in terms of how vote share has converted to seat share in the past:

In a fair voting system, the larger parties should get a seat share that is slightly above their vote share, because they are carrying the preferences of all the micro-parties that did not poll enough to be elected.  However in WA 2017, Labor, Liberal and Green combined were only roughly at their vote share, while the Nationals massively exceeded theirs because of the severe malapportionment.

All else being equal, therefore, a majority of their own is a long way off for Labor, because they would expect to need more than 50% of the vote.  But it's possible that they could get there with slightly less.

I easily found scenarios in which Labor and the Greens could have 54% of the vote between them (a 5% swing) and not win a combined majority.  However, there is a lot of luck in this system and perhaps they could get there off a lower combined vote share than that.  

It's impossible to say for sure how many micro parties will win.  In my initial simulation with vote shares held relatively constant, none of them did, but this relied on the One Nation vote staying high.  In all three scenarios with a swing to Labor, however, the number of micro-party seats won was three or four.  It does seem at a first look that the minor party network is somewhat more potent than in 2017 and we will probably be seeing some undeserved winners.  

More Reading

See also here: An affront to anyone who believes in democracy

Night Before Bonus Scenarios! (12 March)

Some of the polls, like the YouGov Dawesville seat poll, have suggested a massive primary vote swing to Labor coming mostly at the expense of the non-Liberal parties.  This shouldn't be regarded as a reliable statewide pattern - YouGov tends to underestimate Others in seat polls, and much of the crash in the Others vote in Dawesville would be down to an independent getting 10% there last time.  

Anyway I ran a scenario where One Nation lost 60% of their vote, all recontesting micros were halved and the Liberals and Nationals collectively dropped 4%.  This involved primary vote swings to Labor of around 10-18% by region.  In this scenario Labor won 21 seats, the Liberals 7, Nationals 4, Greens 2 and One Nation 1.  There was one very close seat in East Metro where One Nation coming off 3% beat the Greens on the calculator, but by so little that in practice the Greens would have won there taking them to 3.  So this scenario seems effective in generally stopping micros if it happens.

I ran a second scenario with the same but One Nation losing 40% of their vote and all recontesting micros losing a quarter.  In this one Labor won 19 seats, Liberals 7, Nationals 4, Greens 1, One Nation 4 (!), Shooters 1.  "Liberals for Climate" nearly beats the Greens in North Metro.  A few wins off 4% or so but nothing off very tiny vote shares.  

A third scenario, as per the final Newspoll with a 15% swing to Labor, 8% away from the Liberals, One Nation down 60%, Nationals halved and micros generally down a little except where ballot paper position implied otherwise.  This also didn't elect any micros but did throw up a few undeserved One Nation wins: Labor 22 Liberals 6 Nationals 2 Greens 3 SFF 1 One Nation 2.  No Mandatory Vaccination did very nearly knock off the Greens in one seat.