EMRS Feb 2017: Lib 35 ALP 29 Greens 19 Ind/Other 11 One Nation 6
On these numbers a hung parliament would be inevitable (approx 11 Liberal, 10 Labor, 4 Green, though one more Liberal seat might fall to One Nation or the Greens)
EMRS Nov 2016: Lib 40 ALP 28 Green 18 Ind/Other 13 (One Nation not in readout)
On these numbers most likely result would have been a hung parliament (approx 12 Liberal, 10 Labor, 3 Green)
Current seat aggregate of all polls: Lib 12 ALP 10 Greens 3
Note: EMRS tends to skew to Greens and Others and against ALP. No evidence on skew for or against One Nation is known.
Once again, Tasmanian phone pollster EMRS has released two of its quarterly polls, for November and February, in a single release. See also the trend tracker, which shows that the Liberal vote has been falling for four years now.
For the first time, One Nation has been included in the readout, and immediately the Hodgman Government has lost five points to 35%, its worst position of the term. Even the Roy Morgan series, which was obviously skewed against the Government (and hasn't been seen since October) never had it below 37.
The most obvious reading is that the Hodgman government has lost votes directly to One Nation. It may be more complicated than that - there may be transfer from the government to Labor and from both parties to One Nation. This would be most consistent with what has happened elsewhere.
But in any case this poll would see the government a very long way from a majority. After adjusting for the tendency of EMRS to skew to the Greens and others, and to skew against Labor especially (and when the Others vote is high, I would expect the Liberals too) the following is my model of how this poll might pan out at an election "held now".
The most likely result here is a hung parliament with 11 Liberal, 10 Labor and 4 Green. In this scenario the Liberals lose their fourth seat in Braddon and their third seats in Bass and Franklin to Labor and lose their third seat in Lyons to the Greens. They probably retain their third Braddon seat, in roughly the same way as they won four last time (splitting of their vote between multiple candidates enables them to beat the Greens and One Nation). Another disadvantage for One Nation here is possible competition from the Jacqui Lambie Network, who have yet to decide whether to register and run candidates.
It should be noted that this blank-page One Nation figure might not last when their candidates are exposed to a campaign. Palmer United at times polled as high as 7-8% but collapsed to 5% because they were exposed as a rabble. The current WA election also shows how fourth parties can struggle under media scrutiny. One Nation will need to endorse high-profile candidates and minimise bad endorsements to have good chances of winning seats (see more on their chances here.)
(As there are no One Nation polling breakdowns, I have used the last Senate election results to distribute their 6%. I have also included a weighting for the votes of PHON-like parties that could have taken votes from them.)
The leadership figures stand in remarkable contrast to the above. Will Hodgman continues to lead Bryan Green as preferred Premier, with his lead increasing from 50-25 last August to 50-22 in November and now 52-20. This is Hodgman's biggest lead since the 56-19 lead he held in November 2015, at which time the Liberal primary lead was a whopping 48-25. For some time, poor leadership figures for Green have been excusable by the fact of his party trailing heavily, him being not so long in the job and preferred leader scores skewing to incumbents, but the first two excuses no longer apply. If we had two-party-preferred voting in Tasmania this poll would come out at close to 50:50, so Green should really be only about 15 points behind.
While we have unfortunately not had any leadership approval ratings to say whether it is more that Bryan Green is unpopular or Will Hodgman is popular, a Labor leader shouldn't be lagging nine points behind his party's own primary on this score.
It may be that Labor has internal polling on their leadership that they place more trust in than EMRS. If this is not the case then the leadership question may well assume a greater currency in light of this result. If Labor do want to change leaders then the next few months would be the right time to do it, to give the new leader a proper run into the next election.
Complacency about the apparent certainty of a hung parliament (so who cares who the leader is) isn't really an option for the ALP here - the primary votes in this poll may not be representative, or votes may snowball back to the government of the day to avoid a hung parliament as they did in 2006.
Incidentally Green's commitment to the position has been questioned in light of his involvement in a fishing business.
What's driving this?
Frankly I'm a surprised the government's result in this poll is this bad. As with any outlying sample in Tasmania I suspect some sample noise, and also probably interference from currently bad federal polling and perhaps the current penalty rates issue. (The poll is very fresh, being conducted from 1-4 March.) The Tasmanian economy is in improving shape and the government does not seem to have as many problems as during previous runs of lacklustre EMRS polls. The current result should be treated with some caution until we see more polling, but even the November poll was pretty bad as well.
At least the poll confirms my view that culture war politics from the Liberals does not pay. The attempt by Resources Minister Guy Barnett to reignite a forests fight that nobody but him seems to want to be involved with has hardly set this poll alight.
This is the first fresh polling data we have seen since the mid-November ReachTEL meaning that everything else in the aggregate is pretty old rope now. All the same I have concerns about the reliability of any Tasmanian polling at all, so I've weighted this one at 40% and the November one at 10%. I've taken the One Nation figures as per the model above and applied no further adjustments.
The aggregate is still quite cautious about the current 35% result, especially as it is most likely an underestimate. It has the Liberals dropping a Franklin seat to Labor and a Lyons seat to the Greens, though the latter is saveable on preferences if the primary is a little higher and the spread of votes between the three Liberal Lyons incumbents is fairly even. The fourth Liberal seat in Braddon goes to Labor. In Bass, Labor picks up a seat, but it might be from either the Greens or the Liberals. On these figures the Liberals would probably get the seat on preferences.
In the aggregate model the Liberals are still actually quite close to majority government despite a primary lead of only 7.4 points over Labor. They benefit from the uneven distribution of the Liberal vote around the state.
While we have one poll now showing the Liberals a very long way from holding onto majority government, the lessons of 2006 must be remembered. It is very easy to write off the chances of incumbent governments retaining their majority only to find they get back in. Tasmanian voting intention is volatile and a lot can change in the year before the next election.
Something else I think deserves attention is the question of who will be the primary Greens candidate for Lyons. Tim Morris was defeated in 2014 and has been little heard from since. It would seem that the present issue of fish-farming (a flashpoint around Orford, within Lyons) is an opportunity for the Greens to build up profile for a new candidate if they have one. Instead the Green appearing with fishermen in a recent media photo-opp was Franklin MP Rosalie Woodruff. I am wondering generally whether the Greens are being active enough in managing the importance of candidate profile for the upcoming election, with this possibly also an issue for their Bass incumbent Andrea Dawkins.