EMRS: What Doesn't Kill Rockliff Just Makes Him Stronger

EMRS: Lib 38 ALP 24 Green 13 IND 19 others 6
As Tasmanian polling overstates Independents, poll suggests no change from election
Lowest ALP primary since Feb 2014

Jeremy Rockliff has been through a lot of drama as Premier in the last two and a half years.  In May 2023 two Liberals quit the party and moved to the crossbench, putting his government into minority.  In September 2023 the government went further into minority following Elise Archer's forced resignation from Cabinet and Rockliff threatened to call an election to ward off the risk of Archer sitting as an independent without providing confidence and supply.  In February 2024 Rockliff called an early election after the relationship with the two ex-Liberals deteriorated further.  There was a large swing against the Liberals but they managed to form a minority government with confidence and supply agreements from four crossbenchers.  In August 2024 the Lambie Network collapsed and in the fallout Rockliff no longer had reliable confidence and supply guarantees.  In October 2024 Deputy Premier Michael Ferguson resigned over the long-running Spirit of Tasmania saga to ward off a no-confidence motion.  In November 2024 a crossbench no-confidence motion in Rockliff failed after Labor voted against it when their attempt to remove the crossbench's preferred reasons for it failed.  In June 2025 Labor moved their own no-confidence motion, which passed, and in theory Labor could have taken over government mid-term but they did not seek to do so, and an election was held, with a looming deficit crisis now more evidence for critics of the government to run on.  The Government somehow got a 3.2% swing in its favour.  The newly elected parliament (with very similar numbers overall) still included 17 seats worth of previous no-confidence voters plus two new MPs who were highly critical of the government, and could in theory easily have backed Labor.  

And yet, having just easily survived another Labor no-confidence motion, Rockliff is still there, and what's more he's more popular than for some time.  The just-released EMRS poll taken 25-28 August shows Rockliff with an excellent net favourability rating of +18 (43-25), his highest since EMRS started this question in November 2024.  When one considers how old and battle-scarred the Tasmanian Liberal government is, there would be few if any precedents for such numbers in modern polling.  I did say during the campaign that likeability and satisfaction are not necessarily the same thing, but the one poll in which Rockliff polled a poor satisfaction rating (YouGov) turned out to be way off course.   My view is that in the course of surviving so many political near-death experiences, Rockliff has become a tougher politician than when he took over the Premiership, and also one who is frustratingly hard for opponents to pin down or make criticism stick to.  He did cop some personal damage early in the campaign with his net rating at one stage falling to -1 (not that that is bad as such) before recovering.  

EMRS polls were previously conducted by telephone surveying.  The new poll is the first public poll to use a 50-50 online panel/phone mix, a method that was also used in the state election at which EMRS was the most successful pollster.  This poll has the Liberals and Labor down 1.9 from the state result, the Greens down 1.5, Others up 1.5 and Independents up 3.7.  However compared to the final pre-election EMRS poll commissioned by the Liberal Party (37-26-14-5-19) Independents are not up at all, and the Independent vote was substantially overestimated by all pollsters at the election, as it also was in 2021 and 2024.  Overall then, there is no evidence that voting intention has changed since the July 19 election.  Labor would be possibly relieved with that, given the rather intense online reaction against their first-day-back no-confidence motion and the surrounding negotiations.  The Liberals might however also be relieved given the criticism they received in some quarters for policy changes on greyhound racing and salmon expansion.  A previous EMRS panel poll suggests they may have simply read the public mood very well in terms of Tasmanian voter perceptions that it is time for greyhound racing to start winding up.  

It is the case that Labor's primary vote in this poll (24%) is down a statistically inconclusive two points from the last pre-election EMRS poll, and is also the worst primary vote for Labor in EMRS since they received only 23% in February 2014 (just prior to the March 2014 election at which they actually polled 27.3%).  During the 2010-4 term Labor got down to 22% in August 2011 and again in November 2013.  (These comparisons are based on the same format of results that is now used for the headline numbers - the actual headline numbers at the time were lower).  Given that Labor actually led 31-29 in this series in May, the tale told by EMRS tracking over time is that Labor either caused an election they didn't really want or expect, or else greatly overestimated their ability to convince the voters to throw out the Liberals.  It would be interesting to know what internal polling numbers Labor were looking at at state level in May, if any, and whether any internal polling they had at that time might have simply been contaminated by the federal election.  

The poll has also provided some numbers on a new series of party trust to manage issues.  Here they are using a similar format to that used by Essential (with the Greens included as an option).  This format tends to produce worse results for Labor compared to the Liberals than formats with the Greens left out, but arguably for Hare-Clark it is actually better to leave them in.  It is likely that if, say, health and housing were offered as options between the Labor and Liberal parties only, the split would be very close to even, but health especially is a Labor strength issue where a Labor Opposition that was travelling well would typically be well ahead.  

Even less stock than usual should be placed in the Preferred Premier ratings, which have Jeremy Rockliff leading Josh Willie 50-24.  For what it is worth, excluding the Peter Gutwein COVID boost that lasted for much of 2020 through to the 2021 election, this is otherwise the Liberals' best lead since March 2017 when Will Hodgman led 51-22 just prior to the resignation of Bryan Green.  (Gutwein in August 2020 led Rebecca White 70-23).  However, not only do preferred leader scores skew to incumbents, but they also skew against new leaders who have had no time to establish themselves, and Josh Willie is certainly in this category, having been leader for only a week at the time of this poll.  It is also far too early to make much of Josh Willie's net likeability rating of +4, with two-thirds of respondents giving him a neutral, unsure or never heard of response.