NSW 2023 Legislative Council Preview And Advice On Effective Voting

The release of today's ballot order is a good time for a quick post about the NSW Legislative Council contest for 2023.

NSW uses a single statewide electorate with half-in half-out (21 seats) at each election.   Voting is overwhelmingly above the line, and preferencing is fully optional.  A below the line option is available but requires filling at least 15 boxes.  Voters for parties that don't have an above the line box, or for ungrouped candidates, need to vote this way.   See my voter advice section for advice on effective voting below.

The quota for election is 1/22nd or about 4.545%.  The major tickets (Liberal/National and Labor) always win several seats each on filled quotas, the more successful minor parties win a few this way, and there is then a race involving the next candidates in line for the parties that have won full quotas and the lead candidates for parties that have polled less than a quota.  This race typically fills about the final four seats, and the preferences of excluded parties have some impact here and can help some parties to jump over others if their contests are very close.  A party that has well over half a quota on primaries in these races typically wins, those that have well under half a quota lose, and the grey zone is around 0.4-0.6 quotas (1.8-2.7%) - however it varies a little from election to election.

The use of a very proportional system means the major parties are not close to a majority and the question is what sort of house they have to work with.  The 2019 slate was 8 Coalition, 7 Labor, 2 Greens, 2 One Nation, 1 Shooters and 1 Animal Justice.  This is a very close to balanced slate and means that even if there is a lopsided result in 2023 there won't be a very lopsided Legislative Council.  

At the ballot draw today it was revealed that, ludicrously, six out of the 21 tickets are not running enough candidates (15) to qualify for an above the line box and are therefore completely wasting their time and electoral resources as they will not be elected.  In 2015 and 2019 there were only two and one such tickets respectively.  The following is the order of the ballot (left to right):

Group A (Lyle Shelton)
pointless group B (No ATL Box) Craig Kelly

Animal Justice Party

Labor

pointless group  E (No ATL Box) Milan Maximovich

Socialist Alliance

pointless group  G (No ATL Box) Silvana Nile / Fred Nile

Elizabeth Farrelly Independents

Liberal/National

Liberal Democrats

pointless group  K (No ATL Box) Oscar Grenfell

Public Education Party

(spuriously so-called) Informed Medical Options Party

Shooters, Fishers and Farmers

Legalise Cannabis

pointless group P (No ATL Box) Danny Lim

One Nation

Greens

Sustainable Australia

pointless group T (No ATL Box) Milton Caine

Group U (Riccardo Bosi)

Thus while there are eight tickets running that are not registered party tickets, only two of these (Shelton and Bosi) will have above the line boxes.  This will produce a messy gappy ballot paper and something should be done about this next time.  The Shelton and Bosi columns will just be headed "Group A" and "Group U".

The following are my rough and moderately snarky assessments of how the 15 horses with ATL boxes might go:

Coalition: In the years 2007-2015 the Coalition typically polled about 3% below its lower house vote in the Legislative Council but in 2019 this jumped to 6.76%, largely as a result of competition from One Nation.  Currently the Coalition is polling around the mid-30s on average for the lower house.  I suspect the difference between its votes in the two houses this time around will be lower, partly because soft voters are more likely to vote for someone else in the lower house and partly because of the disappearance of the Christian Democrats as a competitor.  An upper house vote of around 30% (6.6 quotas) would probably be good for seven seats but if there is a blowout against the Coalition from current polling it might be the Coalition would only get six.  On the other hand, if the higher estimates of the Coalition's primary are correct and the Coalition manages to close the gap on Labor and to reduce losses to other parties, there is still some chance of eight.  

Labor:  In 2007 Labor polled about the same in both houses but in years since it has increasingly tended to poll better in the lower house, the gap reaching 3.6% in 2019.  With the addition of Legalise Cannabis to the ballot I doubt this will go down.  The party is currently mostly polling in the high 30s on which basis I would expect an Upper House vote of about 33.5.  That would be good for seven seats and a realistic chance of an eighth.  I think Labor would have to underperform current polling badly in the lower house to not be good for seven.

Greens: The Greens have historically more or less matched their lower house vote in the upper house, though they did exceed their lower house vote by almost 1% in 2011.  They too could be affected by increased competition from Legalise Cannabis in the upper house, but on the other hand their lower house polling is likely to be being slightly suppressed by teal independents.  The Greens are still polling well for the lower house (on average around 11%) which suggests they are a lock for two seats and a reasonable chance for a third, but their lower house polling should be watched for any softening that would suggest a third is difficult.

One Nation: One Nation narrowly won two seats in 2019 off 6.9% of the vote.  This is generally expected to increase or at worst stay level, on the back of dissatisfaction with the government and problems in the Shooters, Fishers and Farmers.  However there is not a great amount of polling evidence - one Morgan has had them on 8% for the lower house (which will be wrong as they are only contesting 17 seats).  There was a breathless media report about the party getting "an average of a 3-4 point increase in the primary vote in basically every single seat west of Parramatta." in Liberal internal polling but even a swing of that size across the whole state would probably not be good for a third seat, which would require a state vote of at least 11%.  I expect One Nation to again win two seats.

Shooters, Fishers and Farmers: In happier days the Shooters had a great election in 2019 and won their seat with quota off 5.5% of the vote, boosted by the western backlash against the Nationals that saw them win three seats downstairs.  However the three MLAs have all quit the party (one over policy disagreements then two over the leader using violent language about the one who had left, and the party subsequently not removing him).  There's a general view that said leader, Robert Borsak, will struggle to retain his seat, but I will point out that the party probably needs to lose more than half its 2019 vote before it does so.  We will see.

Animal Justice Party: The AJP won seats in 2015 and 2019 off rather low primary vote shares, just below 2% in both cases, benefiting from left-wing voters being better at filling out preferences than righties.  Its first ever MLC Mark Pearson is retiring (possibly an enforced casualty of the great 2017 snapper scandal) and there is some feeling it will struggle because of competition from Legalise Cannabis.  On the other hand, the party enjoyed a 1.1% swing to it at the Senate election and may well win if it can match the vote it recorded there.  

Legalise Cannabis:  Legalise Cannabis are controversially recycling former Greens MP Jeremy Buckingham as their lead candidate.  This may cost them some votes on the left but the end of their support base that tosses up between them and One Nation won't be bothered at all.  The party polled 2.6% at the Senate election in the state and would probably win if it could replicate that, since that is 0.57 quotas.

Liberal Democrats: The Liberal Democrats profit from voter confusion when they draw to the left of the Liberal ticket but have a lousy ballot draw here in primary vote terms drawing directly on the Liberal ticket's right.  In 2019 they polled 2.18% with former Senator David Leyonhjelm as lead candidate but that wasn't enough (despite Leyonhjelm prematurely claiming victory and the ABC website prematurely calling the seat for him.)   They are hence capable of polling thereabouts but have a history of crawling on preferences, and few obvious preference sources in this lot.  One scenario where they might get lucky is if they manage to roughly hold their vote and the Liberal/National ticket has about 0.3 of a quota spare.  If the L/NP ticket is eliminated before the LDP then being adjacent to them on the ballot paper could prove very handy indeed.  

Sustainable Australia: This controversial party (anti-immigration on purportedly environmental grounds but often and not entirely without reason accused of xenophobia) was not too far off the pace in 2019 when it polled 1.46% (0.32 quotas).  It would probably have to lift substantially to get in the frame and it may not help that there will be competition for the "anti-overdevelopment" (or if you prefer, blatant NIMBY) vote from Elizabeth Farrelly.  The party has run candidates almost everywhere in the lower house, this being a ramp-up of a strategy it also tried in 2019 which is supposed to increase its upper house vote.

Shelton: Former spuriously so-called "Australian Christian Lobby" director Lyle Shelton is running an independent ticket linked to the rebirth of the Family First brandname.  Family First polled 2% in the Victorian election but Shelton will not have the asset of having the Family First name, a magnet for very clueless voters, on the ballot paper.  What he will have is pole position on the grid, which gives him the donkey vote.  The donkey vote does not seem to be worth that much in these contests though, eg Nell Brown's unrelated Group A polled only 0.68% including it plus her genuine votes in 2007.   A further difficulty is that having a group name on an ATL box makes it hard to attract preference flows compared to having a ticket name.  Shelton may poll substantially off the back of the demise of an ATL Christian Democrats or Conservatives option but I suspect a lot of those voters will just vote Liberal.  One to keep an eye on in early counting.

Elizabeth Farrelly Independents: The nearest thing to a teal-style campaign, with a high-profile lead candidate but also a figure of fun for online pundits, especially after failing to declare she was a registered Labor candidate for council elections that she didn't contest.  Farrelly polled 9.8% in the Strathfield by-election but that was in a field of six not a field of 15 groups.  Will probably poll negligibly west of the Red Rooster line and anywhere much outside Sydney so would have to do it all in the inner city seats.  At least has a party name above the line after taking over the Keep Sydney Open registration.  Another one I'll keep an eye on as early in the night as I can, though the pundit consensus is nope.

Public Education Party: This is a rebadge of Reason in a registration sense but more a case of Reason giving away its registration to what was the Fairer Education Party, whatever that was.  A single issue party that seems not to be very high profile and that I don't think should much trouble the scorers.

Bosi: There has been some online interest in the prospects of Riccardo Bosi, leader of the unregistered AustraliaOne extreme right nationalist conspiracist party.  As with Shelton Bosi has a box but not a party name, so even if he manages to poll something non-negligible he'll struggle for preferences, not that anyone much in this lot will send preferences his way anyway.  Will probably generate a lot more hot air than votes.  

IMOP: I should bloody well hope not.

Socialist Alliance: No.

Overall balance

The 2019 result can be taken as 11-10 right-left if counting the Shooters as right.  If current polling is correct and Labor is on track to win the 2PP vote then there's a good prospect of the left winning at least 11, since in that case Labor should get at least seven and the Greens at least two, with Labor, the Greens, Legalise Cannabis and Animal Justice likely to have enough between them for two more.

A 11-10 left-right split would leave an incoming Labor government with a finely balanced upper house and having to work with the Shooters (for instance) to get bills passed.  If Labor does really well a 12-9 upper house result, giving the left a majority, is possible as it's conceivable three of the left's four main chances would then get up.  

If the Coalition wins the election overall it could well be doing so off a weaker voting position than in 2019 and that might well come with an adverse split upstairs leading to a Legislative Council where it had to work with Animal Justice or Legalise Cannabis, for example.  The current government has been quite good at working with crossbenchers with different political views downstairs and would surely take such a situation if offered it right now.  

Effective Voting Advice

If voting above the line, you can just vote 1 for your preferred party and stop, but you shouldn't.  If your party says to vote 1 and stop, you should question its intelligence.  If you do vote 1 and stop, once your party has had all its candidates excluded or elected, your vote plays no further role and cannot help other parties you might like or consider OK beat parties you dislike.

Ranking the 15 parties with above the line boxes from 1-15 is not that hard and if you have the time I recommend doing it.  If you don't, at least make sure you rank (in order of preference) all the parties you like, followed by all the parties you think are overall OK.  If you don't think you know enough about a party, I suggest demote it below parties you do feel you know enough about, or leave it out.  

For the vast majority of voters in NSW, voting below the line in the upper house is pointless and far more bother than it's worth.  I suggest only voting below the line if you, for instance:

* wish to vote for or preference an ungrouped candidate or a candidate without an above the line box (bear in mind they are simply not going to win)

* wish to re-order your preferred party's ticket (bear in mind this is not even remotely likely to affect the results)

* wish to jump around between candidates from different parties, for instance based on their social issue positions

* have a burning desire to put someone last (this too, will make you feel good if you have the time but your vote's very unlikely to get anywhere near there)

* have way too much time on your hands and are up to no good tactically (warning: that section is for advanced players only and the NSW system is slightly different) but don't want to even tactically vote 1 for, say, Socialist Alliance. 

The important thing about the NSW system is that, like the Senate system and unlike the Victorian system, your vote only goes where you send it.  There are no group ticket preferences and even if your party has put out a how to vote card suggesting preferences to someone else, it's your decision who your preference flows to.  (The counting process is complicated in NSW by archaic random elements, but in terms of the impact your vote on average has it may as well be the same as the Senate.)