EMRS: Is Labor Finally Making Some Progress? / Hobart Poll Controversy

EMRS Lib 35 (-1) ALP 31 (+4) Greens 14 (=) JLN 6 (-2) IND 11 (-1) others 3 (=)
IND likely overstated, others likely understated
Liberals would be the largest party but Labor would make seat gains 
Possible seat estimate in election "held now" off this poll LIB 14 ALP 12 Green 5 IND 4

The final EMRS Tasmanian voting intention poll of the year is out and it provides some evidence that the Labor Opposition might be taking baby steps on the road back to government at last.  Labor is up four points, albeit from a poor base.  The Liberal government is at its lowest vote since it got down to 34% in December 2017 (a reading that I doubt was accurate given their rapid recovery months later) and the major party gap is also as low as it has been since then.  This said, Labor still hasn't been above 32% since the last "pre-COVID" poll back in March 2020 and if the ALP is going to make a serious push for government, at some point in the term it will need to break out of the very low 30s.  This is a movement in the right direction for once; let's see if it continues.  The poll comes following a quarter dominated by the Spirits of Tasmania fiasco that led to the forced resignation of Deputy Premier Michael Ferguson from Cabinet.  I suppose that yet again, as with bad polls following the 2023 defections, the Government might say that in it could have been worse.  

There is a bigger change on Preferred Premier where Jeremy Rockliff now leads Dean Winter 43-37, down from 45-30 in the previous poll.  Rockliff's six-point lead is the same as his average lead over Rebecca White, so this at least suggests Winter is now established enough to no longer be disadvantaged on this indicator.  What the indicator actually means is another question - on the Australian mainland preferred leader polls skew strongly to incumbents, but in Tasmania the history of two-leader polls of this sort isn't long enough to say how much they do so.  

The new poll features an extremely welcome inclusion expected to be a regular feature - leader favourability ratings!  I have been calling out for individual leaders to be regularly rated in Tasmanian polling for decades and it's fantastic to see it.  EMRS have not gone with the leader satisfaction/dissatisfaction system used by Newspoll et al but instead are using a score-out-of-10 system that is the same as that used by Essential.  A score below 4 is taken as negative, 4-6 neutral, 7 or above positive.  (I actually think counting 4 as neutral in this system is lenient, but the important thing is we'll have comparable figures for both leaders and over time).  What these ratings show is that neither leader has yet got an army of haters - Jeremy Rockliff scores net +15 (37-22) and Winter net +14 (25-11 with a lot of neutrals and notably still 18% "Never heard of").  The question format doesn't state the person's role and asks the voter to rate them as a "person in public life".  

Election Projection

This is the first poll this term where I've felt it worthwhile to break out the Tasmanian poll crunching spreadsheet and try to work out what could happen on these numbers if an election were "held now".  

Some things to bear in mind here are:

* that polls that use a standalone independent option tend to overestimate "independent" and underestimate "other"

* that the Labor leadership has switched from Lyons to Franklin which all else being equal means a loss of the leader bonus in the former and a gain of it in the latter

* that Jacqui Lambie Network is now on the readout everywhere, meaning its 6% is actually a bigger swing from the election than it looks because some of that 6% would have come from Clark, while none of the 6.7% at the election did

* that the JLN MPs in Bass and Braddon quit the party and are now independents.  Lambie's latest intention was to re-endorse Andrew Jenner in Lyons but not run a big ticket. However it may be a long way to the next election and things may change.

This is a possible conversion of this poll by electorate:

On this model (which uses uniform swing except for adjustments for the movement of Labor's leader bonus, the issue of JLN being on the readout in Clark, and bringing Others up to election level at the expense of IND), the Government holds all of its current seats except for a risk to the third seat in Franklin; it could also potentially gain seats.  Labor is in the mix for possible gains in Bass, Franklin and Lyons.  There are some chances for the Greens to gain a seat.  None of the JLN seats are safe, but that's mainly because of the doubt about whether the party would seriously recontest in Bass and Braddon.

To go through the seats one by one:

Bass: Here Labor would be fighting for the seat currently held by Rebekah Pentland (IND ex-JLN).  If JLN ran and were not damaged by Pentland recontesting as an indie, JLN would probably still just hold the seat based on the 2024 preference flows.  However other possibilities are Pentland consolidating the high Bass indie vote plus taking some of the intended JLN vote and winning, or Pentland's departure splitting the JLN vote (or some of it going back to Labor) and Labor winning the seat.

Braddon: On the numbers above JLN still wins a seat leaving Craig Garland and Miriam Beswick fighting for either one seat or no seats, but given Jacqui Lambie's statement I'm not sure this would actually occur.  I think the most likely outcome would be the Liberals getting a fourth seat and either Beswick or Garland but not both retaining; another possibility is that there is enough of a downturn in the overstated Independent vote for the Greens to get one.  (Garland has run into a spot of bother but I'm not sure if it will damage his future chances.)  

Clark: 2-2-2-1 as per election

Franklin: The model suggests a close three-way fight for the last seat between the third Liberal, third Labor and second Green.  However based on the election, the Liberals by spreading their vote across three MPs will make the final two against either Labor or the Greens.  Labor are probably able to do this too if they can find a strong enough third candidate now they have two incumbents, at least to the extent that they might be able to stay ahead of the Greens and win on Green preferences (they nearly kept their third candidate ahead of the Greens this year off a much lower vote). It's quite plausible on the current poll numbers that Labor could just win the seat for a 2-5 right-left split in Franklin.  

Lyons: Here the fight is between Labor's third candidate and JLN's Andrew Jenner.  Paradoxically despite me slugging Labor 2% for the loss of their former leader their prospects are actually better because some oi what they've lost is votes that would be lost as leakage anyway, and now they can spread out their vote between their top three candidates, the second of whom will be a first-term recount-elected MP (these tend not to poll massive votes).  If they can keep two candidates ahead of Jenner then they win.

Overall my read of this poll in combination with Jacqui Lambie's latest stated intentions is that there is a fair chance the JLN element would largely disappear from parliament with Labor the main beneficiary.  

I should note that the numbers above are based on existing boundaries.  During the next federal term Tasmania will be redistributed and the balance of the seats will change.  Clark must grow and will presumably take territory from Franklin in the Kingston area and perhaps Lyons around Molesworth, unless the redistribution committee go for more radical surgery.  I'll be taking more interest in this one than I usually do in these things.  

Hobart Council Poll Controversy

The current round of EMRS polling included some commissioned questions that have raised more eyebrows than usual.  Recently EMRS was purchased by FontPR, an omnipresent Tasmanian public and government relations and campaigning firm which in addition to its lobbying and campaigning roles owns multiple local newspapers and runs the state's most (if not only) significant political podcast.  Recent Fontcast episodes have frequently teed off at Hobart City Council's decision to install bicycle lanes on Collins Street, a project which the government has declined to fund.  

FontPR is also often regarded as a Liberal-aligned outfit on account of Brad Stansfield being former Chief of Staff to Will Hodgman and working prominently on multiple Liberal campaigns.  However, when not doing so his commentary can be quite critical of the government, and managing director Becher Townshend is a former Labor staffer.  The Fontcast also favourably interviewed new Labor leader Dean Winter, though at times they have criticised some of his decisions too.  In short, FontPR isn't straightforwardly partisan, but I don't think we'll see it doing the Greens' campaigns any time soon.  And it has fingers in a lot of pies in a state where potential conflicts of interest are everywhere.  

Furore erupted on the shrinking remains of Twitter (and doubtless other social media) when it was claimed that EMRS had been asking a poll with questions that someone had said were as follows, albeit not claimed to be verbatim.  (The author of this question list is unknown to me and I also don't know yet if it was a social media post or a text message):

"1. Do you think councils should stick to core services of roads, rates and rubbish.[..]

2. Do you support converting Collins St into a bike lane.[..]

3. Do you think the State Government should appoint administrators to the Hobart City Council."

As summarised this looks like the classic hook, line and sinker of the Yes Prime Minister Leading Questions variety.  Prime the respondent with a red-meat right-wing culture war trope that misrepresents what councils are actually tasked with the project, throw in a false statement about what the bike lane project was then invite them to conclude that the impossibly woke Hobart City Council must be canned.  But I understand this is not what actually happened at all.  Among other things the bike lanes question described the project accurately and it was placed after the administration question not before.

Opponents were quick to suggest that FontPR was up to no good and was using EMRS to further its tricks.   EMRS was also accused of push polling.  However my understanding here (from sources from both sides of the argument) is that these are in fact client-commissioned questions, not questions conducted by either the pollster or the PR firm off their own bat.  It is normal for EMRS polls to include a string of questions from various external clients (those about councils in this case being followed by others for a presumably different client about salmon farms).  I have seen a claim that the client for the council administration question has a particular connection to FontPR but have not verified that the claimed client is in fact the client, and in any case Hobart is Hobart so it would be bemusingly surprising if they didn't.  The questions asked would, I think, have been accepted by the pollster in either its standalone or C|T incarnations.  It's possible that the fact that the firm owning EMRS has broadcast views against the bike lanes has had some influence on the client's decision to use EMRS for this poll, but this isn't Font using a poll it owns to canvass support for its own messages.  

"Push polling" as strictly understood is the dark art of conducting a fake poll to smear a political opponent.  The sample size is huge, the data are thrown away, the point is simply to get the (bogus) message out there.  It is vanishingly rare in Australia.  The name comes from the fact that the supposed poll is really pushing a message.

However, in Australia the term is often used loosely for polls that use misleading wording or suggestive question orders to push respondents to respond in a certain way.  This is usually done for the purpose of generating an exaggerated support figure in the poll that can then be used to exert pressure on decision-makers, or simply the poll can be used to get media attention for the message the client wants out there.   It is also sometimes done simply through incompetence - organisations commissioning polls don't realise that the questions they are asking are faulty, and here there is a role for a pollster to advise the client accordingly.   There can also be an element of pushing in terms of even a relatively small sample poll "seeding" discussion of a proposal in the community - and it is the latter that opponents are concerned about here with the asking of a question about putting a council into administration.  It may also be that the intention of the poll is a sort of Streisand effect bait, to generate outrage from the council's supporters that will give more publicity to the core idea.

At best, asking a question about putting a council into administration is not useful research if that question comes after any question that primes the respondent with a question about the function of councils, so if the results of this question are ever released I would treat them with enormous caution.  Another problem with polling on the issue at all is that it is a relatively novel proposal that has only attracted limited public debate, though there was recently some local reporting of a call to put the Council into administration that even included a threat to cause an elector poll on the bike lanes issue.  (An elector poll is a very expensive voluntary non-binding postal vote of the entire Hobart electorate).  Respondents may say yes or no to the EMRS administration poll according to what they think of Hobart City Council without being at all familiar with the issue.  They may not have thought about it before, and thus the question may not be sampling actual views that are out there.  

Tasmanian councils are typically put into administration because they are disfunctional beyond repair (infighting between councillors and the Mayor, or between councillors and staff) or because they have severe integrity issues.  As tedious as this term's constant bickering, noise and culture warring involving some councillors has been, and although the council has faced other problems including industrial disputation and an anti-discrimination loss, at this stage I'm not seeing much here beyond a political disagreement. Part of the business lobby affected by the bike lane issue is displeased that a left-wing council elected by a very left electorate is doing left-wing things.  If the government was inclined to agree with the push I'd be expecting it to show better reasons.

I may add updates on this poll if more information comes to hand, especially any actual results!

Update 26 Nov: Bike Lane Results

Results of the bike lane question are circulating.  The question was asked in Clark with a sample size of 190 including 92 in Hobart LGA.  The question was:

"The Hobart City Council has announced plans to install bike lanes on both sides of Collins St in the city between Molle Street and Murray Street.  To what extent do you support or oppose this plan?"

The broader Clark sample was opposed to the plan 38%-58%.  The Hobart-specific subsample was opposed 41%-56%.  I think only the Hobart sample would have been asked about putting HCC into administration, which could have contaminated the question.  

The Hobart LGA sample is too small to be useful.  Even ignoring the issues of contamination from prior questions, it is just not possible to representatively poll a city with a sample of 92 and to be sure such a sample is representative.  The margin of error in effect is much higher than the c. 10% for a purely random sample that is cited.  So I would not take any notice of this result, beyond that it isn't close to 100% one way or the other, which might have meant something.