Friday, July 21, 2017

Fishy Prospects In The Seat Of Lyons

ReachTEL Lyons: Lib 42 Labor 30.4 Green 12.4 Lambie Network 10 SF+F 2.7 Others 2.5
ReachTEL polls in Tasmania have in the past skewed against Labor and to the Greens
Seats that would be won based on this poll: Liberal 3 Labor 2 (status quo)

The Australia Institute has released a large-sample ReachTEL of the state seat of Lyons.  Lyons has long looked like the most crucial seat in determining whether the Hodgman Government can maintain a majority at the next state election, as on a more or less uniform swing to Labor, the third Lyons seat is the third to fall.  Polling has long appeared touch-and-go as to whether the party is likely to hold three seats there or lose one to the Greens or maybe someone else.

The commissioned ReachTEL also covers fish farms, which are seen as a significant environmental issue in the leadup to the next election.  I am satisfied that the poll has not been selectively released and also that ReachTEL have a good record in not letting commissioning sources tweak the primary vote polling design.  So while all commissioned polls are to be treated with some caution, and all seat polls always require special care, I'll have a look at what the data from this poll suggest.

As usual with ReachTEL the data require a lot of unpacking.  ReachTEL use a different format to most other polls, by initially giving voters a set of options that includes "undecided", and then allowing those who are "undecided" to say which party they are leaning to.  However the "undecided" in ReachTEL polls would be included in other polls' headline figures, while the truly undecided voters (those not even leaning to any party) are excluded, as they are by other pollsters.

With the ReachTEL-undecideds ("soft undecideds") reallocated, the results of this poll are Liberal 42, Labor 30.4 Green 12.4 Jacqui Lambie Network 10 "Shooters and Fishers" 2.7 Ind/Other a mere 2.5.  I put quotes around Shooters and Fishers because that party should actually contest as Shooters, Fishers and Farmers following a name change.  On these figures the Greens (0.74 quotas) would seem to have good prospects of holding off the Liberals (2.52 quotas) and the Lambie Network (0.60).  However, an even enough split of votes between the three Liberal MPs Hidding, Shelton and Barnett, or a good flow of preferences from minor parties, could still see them all retain.  None of these MPs have been stellar performers for the government but none have been consistently awful either, so it may well be the party can hold a similarly even spread between the three to last time.

However, ReachTEL polls in Tasmania have had noticeable house effects at the last two federal and the last state elections.  They have in all cases had the Labor Party too low (by an average of 3.8 points) and the Greens too high (by an average of 2.7).  They had the Liberals too high at the 2013 and 2016 federal elections but too low at the 2014 state election, so the pattern there is less clear.

If, following that, I adjust Labor upwards to 34.2 and the Greens down to 9.7, the Greens and Lambie Network are now only marginally clear of the Liberals' excess after their second quota.   If one of the sitting Liberal MPs does really badly while the other two hit quota, this could see the Liberals lose a seat to either the Greens or Lambie Network.  But more likely on such numbers the spread between Barnett, Shelton and Hidding would be reasonably even and all three would be re-elected.  Even if the spread was not even, the party might still hold three seats on preferences.  So my seat reading of this poll is Liberal 3 Labor 2.

Even with my adjustment the Labor vote is still not all that might have been hoped for given the honeymoon effect for the leadership change to Rebecca White, and backs reports that internal polling is not yet showing Labor winning three seats anywhere.  The party suffered a damaging-looking row over conscience votes at its recent state conference, highlighting internal divisions over social issues and factional balance at precisely the wrong time.  It will be interesting to see how other pollsters see the party going.

The Lambie Network support in this poll is similar to the party's Senate vote of 9% in the same seat.  However the party is not even registered to contest the election yet and I would be surprised if this level of support holds up once voters realise it is not Lambie herself on the ballot. If the Lambie challenge doesn't eventuate or doesn't amount to too much, these votes will probably be more use to Labor than anyone else.  Still it is hard to lift Labor to the sort of vote required for a boilover three-seat result.

The Shooters, Fishers and Farmers have been trying hard to market themselves as an alternative source of protest vote over recreational fishing issues but this poll doesn't show it paying off much at this stage.  Their best hope will be if neither One Nation nor Lambie Network end up actually running and they are left as the last fourth party of any magnitude standing.

The poll quotes a two-party preferred figure of 51-49 (respondent-allocated) to Liberal.  This should be largely ignored.  It is just an automatic feature of the ReachTEL polling package used to do the poll.  2PP is not relevant to Hare-Clark firstly because not all minor party preferences are distributed (many are either used to win Greens seats or left with the candidate who comes sixth) and secondly because voters only need to vote for five candidates and so a lot of preferences exhaust.  However the 2PP figure is consistent with the idea that if the Liberals can retain a majority, it will probably be by the skin of their teeth.

Adding this result (as adjusted for house effects) into my aggregate of state polling (I've given it a weighting of 20%), the current reading for Lyons is Liberal 41.1% Labor 35.4 Green 12.3 others 11.3. In quota terms, Liberal 2.47 Labor 2.12 Green 0.74 Others 0.68.  At this stage I will leave the aggregate in the sidebar as it is but any further signs of weakness in the Green vote are likely to result in moving the seat to the Liberals, in view of the Hare-Clark issues mentioned above.

Fish Farm Polling

Fish farms are being seen as a possible saviour for the Green vote at this election.  There is a case that the Greens could be in lots of trouble otherwise - the federal party is in crisis because of infighting and eligibility issues.  The state MPs (aside from leader Cassy O'Connor) are probably not high-profile enough and Labor has elected a new leader who is young, female and relatively left-wing.

The salmon farming issue functions partly as an environmental issue and partly as a local NIMBY issue.  The environmental issue is primarily in Macquarie Harbour, where stocking levels have produced local "dead zones" that are seen as threatening the endangered Maugean skate. The NIMBY issue arises anywhere where fish farms are proposed if they are seen as having potential local impacts on recreational fishing or scenery.  In the case of Lyons, a proposed fish farm near Okehampton Bay has been controversial, but this is complicated by the fact that the towns near the proposed farm are far from Green strongholds (and are indeed job-hungry following the downturn in the timber industry.)  Shack-owners concerned about fish farms often don't live in the electorate.

Anyway question 2 of the poll finds that 35% of the electorate, without any prodding, agrees that fish farming companies have "too much influence" on the government.  This sentiment is strongest among Greens voters (63.8% - hardly surprising) but it is interesting that 20% of Liberal voters agree with it.

Question 3, however, is "Do you agree or disagree that the fish farm industry is risking jobs by not investing in future-proofing the industry?"  In my view this question produces a skewed response by inviting the respondent to accept as fact that the industry is not investing in "future-proofing" and then inviting them to draw a conclusion from that supposed fact.  A more appropriate design would be to ask respondents whether they believe the industry is investing in future-proofing, and then asking those who say no whether this is putting jobs at risk.  In such a format I would expect that the don't know rate on the first question would be massive since most voters probably have little idea what "future-proofing" is, let alone whether the industry is engaging in it.  What is interesting about this question is that Liberal voters took the bait almost as readily as Labor voters.

Question 4 asks respondents to consider the impacts of fish farming on other fisheries.  This question may be tainted by the attack on the industry implied in the wording of question 3, but even so it is interesting that 16.9% of respondents consider fish farms to have a positive impact.  I think this points to some respondents thinking about the question in an economic sense and believing that the growth of any industry is good for other related industries.  It might however also point to a kneejerk tendency among some voters to reject the views of the Greens.  This is another question where Greens voters are very much more prone than others to see fish farming as bad.

Other state polling is coming this weekend according to the Mercury!

Saturday, July 15, 2017

Scott Ludlam Mess Scores Four Bob Days Out Of Five

Well here we go again.  After the departures of Senators-who-sort-of-never-were Rod Culleton and Bob Day we've lost another one.  After nine years in the Senate, one of the sharper minds in the place, Greens Senator Scott Ludlam, has suddenly realised he has been a dual New Zealand citizen all along and was never validly elected in the first place.  That sound you heard all afternoon was at least 200,000 Greens supporters banging their heads on the nearest available tree in disbelief.  As for me, I was so distracted by this situation that I needlessly got off a bus in the middle of Hobart city, forgetting it continued past a common stopping point to much closer to home.  No problem though, since I then managed to beat the bus to its next stop on foot and catch the same bus again.  Ludlam's path to getting his seat back, should he want to, would be rather less straightforward.

For the most part this one is a familiar situation.  Although Ludlam has resigned, the fact that he has raised eligibility issues as his reason for doing so should prompt an immediate referral to the Court of Disputed Returns (the High Court in theory though it may well get kicked downstairs to the Federal Court if there are no new legal issues) to determine whether Ludlam was validly elected in the first place (to which the answer is evidently no) and to supervise the filling of the vacancy.  The vacancy will be filled by a recount (called a "special count") as with the vacancies for Day and Culleton.  The Greens won two seats in the original election and in the Culleton recount, beating the WA Nationals' Kado Muir by 25175 votes in both cases.  The recount could shave a few thousand off this (about 2800 personal votes for Ludlam leak out of the Greens ticket based on the original counts) but there's no doubt the Greens would keep two seats.  One of these will be their other existing Senator, Rachel Siewert, and the other will be the third candidate on the original ticket, Jordon Steele-John.

However this recount does raise some new ground. Firstly it's the first time a state will have had to be recounted for two disqualifications from the same election, meaning that the new count will be without both Culleton and Ludlam. Secondly and more interestingly, it creates previously unseen complications with the original allocation of three and six year terms. Scott Ludlam was elected third in 2016 with Rachel Siewert elected 12th.  In the special count to replace Ludlam, Siewert will be elected third and Steele-John will be elected 12th.  So if Steele-John replaces Ludlam and serves out Ludlam's term, then this will create a bizarre situation of the candidate second on the Greens ticket being a Senator for three years while the third candidate on the ticket is a Senator for the balance of six, clearly not the preference of the party's voters.

The original motion passed by the Senate to allocate the terms was:

That, pursuant to section 13 of the Constitution, the senators chosen for each state be divided into two classes, as follows:
(1) Senators listed at positions 7 to 12 on the certificate of election of senators for each state shall be allocated to the first class and receive 3 year terms.
(2) Senators listed at positions 1 to 6 on the certificate of election of senators for each state shall be allocated to the second class and receive 6 year terms.

Section 13 of the Constitution refers to the Senate meeting after each election to allocate seats but does not contain any explicit instrument for revising the allocation should it be needed to revise it. In some hypothetical cases it could be necessary to do so - multiple Senators serving different term lengths might be disqualified at the same time (or one might be disqualified and another unelected as a result, as in the Melbourne City Council Michael Caiafa case) and it might not be possible to clearly identify who had taken whose seat from the original count.  Mostly the opinions I am seeing are that the Senate can (and/or will have to) recommit the matter and could make a fresh decision on the term lengths, but I have not yet seen anything from anyone known to me as a constitutional law expert on it.  [Update: See comments - there is significant doubt about whether a decision to reallocate Siewert's term would be constitutional.]

An important aspect of this situation is the pickle it puts the potential replacement in.  Jordon Steele-John is a 22-year old student who works in disability advocacy (for what hours or pay is not stated) and now appears to have a lucrative job opportunity should things pan out as it seems they will.  But if Ludlam wanted his seat back and the Greens wanted Ludlam to have the seat, or if the Greens wanted it to go to someone else for some reason, Steele-John would be in a very difficult situation (as was Irina Dunn, who took the seat vacated by Robert Wood's ineligibility and was kicked out of her party for her troubles).

So far, Steele-John has posted that he would want the seat to go back to the party for a decision (which would have a similar effect to a casual vacancy, assuming the WA Parliament did the right thing and backed the party's proposed replacement).  As the party might well then decide that he should keep the seat anyway, this doesn't necessarily mean he doesn't want the seat or that it will necessarily create a casual vacancy.  (If the party decided they wanted him to keep it then there would be no point in him resigning.)  At least it appears that he is eligible as he renounced his UK citizenship at age eighteen.  The Alanis factor in that one has got to hurt.  [Update: Despite earlier reports that he was not going to take up the seat, Steele-John has now said he is taking it and has the party's support to do so.]

In ten years, Scott Ludlam was elected to the Senate four times - 2007, 2013, 2014, 2016.  Second candidates who could have replaced him on recounts had the issue been discovered earlier were Alison Xamon in his first term and Dr Christine Cunningham in his second.  Kate Davis was the Greens #2 candidate in the original 2013 election but that election was voided.  Had it not been voided because of a loss of ballot papers combined with problems with the old Group Ticket system, there is even a scenario in which Ludlam's ineligibility might have caused the Greens to lose his seat altogether and unelected another party (albeit only the Sports Party) in the process.

The loss of Ludlam, if more than temporary, is a disaster for a federal Greens party already struggling with internal process issues that to this observer have been farcical (if they're really unlucky I will write about that soon).  Even if it is only temporary, it is highly embarrassing.  The right, to whom Ludlam has long been a particularly sarcastic bane, has been gleeful in response and has had a lot of fun pointing out that Ludlam had rubbed it in when Bob Day was rubbed out.

It raises the question: in the absence of any change to the Constitution to change these rules, or any electoral law change to require candidates to declare status (see Michael Maley's submission here), when will political parties learn?  This has been a known issue for decades.  How hard is it to do due diligence on all candidates before preselection by asking them simple things like "Were you born in another country?" and "If yes, what precise steps did you take to renounce your citizenship of that country?"  This shouldn't have been picked up by a member of the general public searching New Zealand records after ten years.  The party, as well as the candidate, should have been on to it.  Why weren't they?

See also Antony Green.

Update 19 July: Another One's Gone!

Astonishingly the Greens have - at least for now - lost a second Senator in the same manner with Queensland Senator Larissa Waters also resigning after finding she was a dual citizen of Canada.  This one raises no electoral complications - assuming a special count is ordered as per usual, former Democrats leader Andrew Bartlett will win the seat.  It also doesn't unsettle any of the other seat results. It is unclear whether Bartlett would take the seat or resign so Waters could be reappointed or so someone else could take it.

What this twin disaster does do is indicate a staggering degree of incompetence and complacency, over a long period, in candidate screening by the Greens. So out comes Bob Brown with excuses about the Greens being mainly reliant on volunteers - which may not be true anyway - but even if it is true ignores the fact that the Greens have a particularly enthusiastic volunteer base including a lot of very capable and educated people.  Where Brown may be getting closer to the problem is in his interview with Emma Alberici where he refers to the lack of central federal control in the party, such that the Tasmanian branch might have been across such issues but have had no way of ensuring other branches were.  (Whether those who understood the problem communicated it with other states properly then is a rock that hasn't yet been turned over here.)  This weak federalism of the Greens is also a large component of the Lee Rhiannon mess.  It's a coincidence that those issues have shown up at about the same time, but it's not a coincidence that they have both happened.

Predictably for Brown, those who quite reasonably use the Greens' inability to get their facts right on eligibility as a proxy for factual inaccuracy on other things are "involved in a venal and nasty and hateful sort of politics - knock everybody down, kick them as best you can, never let a sucker have a go, as the old saying goes." In other words they are just like St Bob on his environmental crusades against a wide range of demonised industries, except that they have a valid argument, which St Bob frequently doesn't.

It was also interesting to see this on the Facebook page of the Byron Greens:

This post has the same structure as the final paragraph of my original article, even including some of the same words like "due diligence" and "simple", but it shifts the blame from the poster's own hapless party to the Australian Electoral Commission.  Whether this is plagiarism of a sort or just coincidence (I don't know which) it is unreasonable to accuse the AEC of sloppiness for not discharging an electoral burden that the law doesn't give it.  And as Antony Green points out, it is a bad idea for the AEC to be charged with candidate screening anyway.  In the form stated by the Byron Greens above, Waters and Ludlam would both have just answered "yes" to question 2 and known no better.

Meanwhile debate about whether Section 44 needs fixing (which can only occur by referendum) is likely to increase.  Examples like Nick Casmirri's (and also Sam Dastyari spending $25,000 on legal fees) suggest that the present situation is confusing and can be very onerous for some candidates.

Thursday, July 6, 2017

How Often Are Federal Newspolls Released?

A humble little subject, but I just thought I'd put up a resource piece about how often federal Newspolls are released and have been released over time.

The timing of federal Newspolls is frequently a subject of discussion, much of it clueless or biased.  One fairly prominent claim links the switch to federal Newspolls in the early 1990s with the frequent turnover of major party leaders, although there is actually no evidence that this is true at all.  On social media, Newspolls are eagerly awaited and considered "due" every second Sunday, mostly by one-eyed Labor supporters.  If Newspoll fails to appear this is claimed to be evidence that it is being "hidden" because the results are bad for the Coalition.  If this were actually the case, Newspoll would skew towards the Coalition compared to other polls (it doesn't), Newspoll would have gone AWOL during obvious Coalition low points like the Hockey budget, the Prince Philip knighthood and even the collapse of Malcolm Turnbull's "utegate" attack on Kevin Rudd (it didn't), and Newspoll conspiracy cranks would be able to post reliable advance predictions of when Newspoll would come out (they don't.)  But those tweeting these nonsense never let the facts get in the way of their inane barracking.

This week I saw a new strain of the viral dumbness that is Newspoll truthism - a claim that Newspoll was becoming less frequent in order to string out the time it would take for the Coalition to lose 30 consecutive Newspolls on Malcolm Turnbull's watch.  (Turnbull has lost 14 in nine months, while Abbott's 30 spanned sixteen months, making Newspolls 17% more widely spaced so far during Turnbull's losing streak.)

Tuesday, July 4, 2017

All Polling On The Plebiscite Has Problems

In the last few weeks we've seen some new polling results concerning the Coalition's proposed plebiscite on legalising same-sex marriage (or as it is more accurately described, marriage equality).  It is extremely well-established in polling that a clear majority of voters support legalising same-sex marriage, but whether voters support deciding the matter by plebiscite or parliamentary vote has been less obvious.  The very inconsistent results from various polls on this subject are causing a fair degree of interest and confusion.

In this article I suggest that the range of results we are seeing on the question of a plebiscite vs a parliamentary vote is largely a result of differences in design between different polls.  Which poll is right and which poll is wrong?  My view is that all of them are suspect.   The question is a difficult one to poll and none of the polls offered thus far have even got close to a design that accurately reflects the choices the parliament, voters and activists face.