Essential Research: 52-48 to Coalition

Essential Research offers more mediocre post-election poll ratings for the new government, together with findings on climate policy, boat arrivals, industrial relations and manufacturing.

Essential Research has the Coalition’s lead up slightly on a weak showing last week, from 51-49 to 52-48, with primary votes of 43% for the Coalition (steady), 36% for Labor (down one) and 9% for the Greens (steady). Findings of further questions:

• “Direct action” is favoured over carbon pricing 35-31, reversing a 39-29 lead for carbon pricing in May. Support for carbon pricing is down from 43% to 39% with opposition up to 43% to 47%.

• Support for the government’s decision to cease issuing statements when asylum boats arrive is at 39% – surprisingly high, to my mind – with opposition at 48%.

• The re-establishment of the Australian Building and Construction Commission is supported by 29% and opposed by 22%, with the rest down for either no view or don’t know.

• There are also questions on manufacturing which suggest respondents to be broadly supportive of protectionism.

Meanwhile, buttons have been pressed today for Senate contests in Victoria, South Australia and the Australian Capital Territory, which you can read about in the Senate counting thread a few posts below.

Morgan and Essential polls

Neither Morgan nor Essential finds the government enjoying much of a honeymoon, while Morgan has Anthony Albanese well ahead of Bill Shorten as preferred Labor leader.

Morgan has published its first multi-mode poll since the election, and Essential its second online poll (the latter will henceforth publish on Tuesdays rather than Mondays). Even if you doubt the value of voting intention polling at this point of the cycle, the results are of interest with respect to the Labor leadership. If you don’t doubt the value of voting intention polling, the results are of interest in pointing to a weak Coalition honeymoon.

Starting with voting intention:

• Essential Research has the Coalition lead at 51-49 on the current two-week rolling average, combining results from 1042 respondents in this week’s survey from Thursday to Sunday and 844 from the week before. This leaves the Coalition two points down on a less than spectacular showing last time. On the primary vote, the Coalition is down a point to 43%, Labor up one to 37% and the Greens steady on 9%.

• The Morgan SMS, online and face-to-face poll of 2999 respondents, conducted from Saturday to Monday, has the Coalition on 43.5%, Labor on 34%, the Greens on 10.5% and the Palmer United Party on 4%. This compares with election results on current counting of 45.6%, 33.4%, 8.6% and 5.5%. This translates into a headline two-party figure of 50.5-49.5 on respondent-allocated preferences, but it’s a more comfortable 52.5-47.5 on preferences from the September 7 election (though I’m not sure exactly how minor party preference splits were determined given all the votes aren’t in). It is of course enormously unlikely that minor party preference allocations would have changed so dramatically over a fortnight, a further pointer to the dubiousness of respondent-allocation.

• Morgan has good news for Anthony Albanese, who is favoured over Bill Shorten 41% to 23% among all voters, 46% to 32% among Labor voters, 38% to 18% among Coalition voters and 48% to 12% among Greens voters. The gap is widest and narrowest and Albanese and Shorten’s respective home states of New South Wales and Victoria. The qualitative findings here are unusually interesting: “Electors who preferred Anthony Albanese often mentioned Shorten’s role in the demise of former Prime Ministers Kevin Rudd and Julia Gillard, Shorten’s strong links to the unions, and also his links to the Governor-General as well as Albanese’s better policy expertise, experience and personality.”

• Essential finds Tony Abbott with similarly modest leads as preferred prime minister over both Albanese (37-31) and Shorten (37-32).

• Essential at least has had a bounce on personal approval, in net terms at least – his approval is up only one point since his last poll as Opposition Leader on September 2 to 41%, but his disapproval is down 13% to 36% (making for a big increase in “don’t know”.

• Essential finds 45% concerned about the lack of women in cabinet against 50% not concerned, with splits of 39-57 among men, 51-42 among women, 67-29 among Labor voters and 17-80 among Coalition voters.

• Also featured in Essential are questions on trust in use of personal information by various professions and organisations, and the value or otherwise of foreign investment in farm land.

UPDATE: Morgan has kindly provided me with its qualitative responses from the Labor leadership question, and I’ve run the responses through a word cloud generator. Note that in doing so I’ve merged together a couple of words like “don’t like”, “don’t trust” and “prime minister”. You can get a considerably bigger image by clicking on the images below.

First up, the 443 responses from Anthony Albanese supporters, for whom the primary reason for backing Albanese appears to have been Bill Shorten:

And now the 229 responses from Bill Shorten supporters:

Week two flotsam and jetsam

Another review of the late counting situation, plus the Labor leadership vote, jockeying to succeed Bob Carr in the Senate, and prospects for electronic voting.

Yet another review of late counting, together with a few other things:

• With McEwen continuing to slip from the Liberals’ grasp, the only remaining lower house seat in doubt is Fairfax, where Clive Palmer received a very handy fillip yesterday when provisional votes pushed his lead out from three to 98. Follow the action here.

• Then there are the Senate races in Western Australia and Tasmania, which are unlikely to become clear until the below-the-line data entry is completed and the button pushed to calculate the outcome (there’s a dedicated thread for Senate counting here, although it’s not doing much business). In the former case, there are probably two seats which hinge on absurdly trivial combinations of micro-party votes and whether they work to the advantage of Australian Sports Party candidate Wayne Dropulich – the fates of Labor and Greens incumbents Louise Pratt and Scott Ludlam as much involved as those of Dropulich and the other potential micro-party winner, Zhenya Wang of the Palmer United Party. The early test for Dropulich is whether he stays ahead of the Rise Up Australia party (0.29%) after his own votes (0.22%) are supplemented by preferences from Australian Voice (0.09%), which has been touch-and-go but has improved for Dropulich on today’s counting. As TruthSeeker observes, Dropulich then needs for the current 183-vote lead of Australian Christians over Shooters & Fishers at Count 21 to hold, which it may not do when below-the-line votes are taken into account. Failing that, Dropulich could be saved if, at Count 19, Help End Marijuana Prohibition failed to hold its present 117-vote lead over the Animal Justice Party, for reasons which would do your head in. On any scenario in which Dropulich wins, the other seat looks set to go to Scott Ludlam of the Greens. If he fails, Zhenya Wang will be joined by Louise Pratt rather than Ludlam, as the Palmer United Party’s direction of preferences to the Greens ahead of Labor would no longer be a factor.

• For Tasmania, Kevin Bonham has the various scenarios neatly laid out in a flow chart, two of which (the final seat going to third Liberal Sally Chandler or Jacqui Lambie of the Palmer United Party) are rated more likely than the others (the win for Robbie Swan of the Sex Party currently projected by Antony Green’s calculator and, with a particularly small chance, a win for Family First). So far as the projection of Antony Green’s calculator is concerned, the trend of counting is towards Robbie Swan of the Sex Party in his fight to stay ahead of the third Labor candidate at Count 21. He took the lead on Tuesday, and it has since gradually opened to 382. However, Bonham’s rough calculation is that it will need to be more like 800 to save him from below-the-line vote leakage. Of the many absurdities in this state of affairs, I have two favourites. One is that the Liberals need the Labor vote to be as high as possible to help ensure Swan’s exclusion, which presumably means Liberal scrutineers are fighting with Labor ones to ensure potential Labor votes are included in the count. The second, noted by Kevin Bonham, is that voters confusing the Liberal Democrats with the Liberals is actually to the Liberals’ advantage, as they have various paths to victory which involve the Liberal Democrats staying ahead of the Palmer United Party or Family First, while their own vote total is essentially academic at this stage.

• Australian Workers Union national secretary Paul Howes was thought by many to have jumped the gun yesterday when he refuted media speculation he might replace Bob Carr in the Senate, given Carr is yet to announce any intention on that front. However, the universal expectation that it will be so is indicated by jockeying to fill the spot. Troy Bramston of The Australian reports that Carr wishes to be succeeded by Graeme Wedderburn, who has been his chief-of-staff both as Premier and Foreign Minister. However, it is today reported that state secretary Jamie Clements has called for the position to go to Deborah O’Neil, who lost her seat of Robertson at the September 7 election, pleading affirmative action. Graeme Wedderburn held senior positions with Westpac and Origin Energy following Carr’s retirement as Premier in 2005, before being lured back to the job by Nathan Rees in 2009 in part by the promise of a Senate seat down the track. However, he was denied a vacancy at the 2010 election due to an arrangement in which Matt Thistlethwaite, who is now entering the lower house as Peter Garrett’s successor in Kingsford-Smith, was given a Senate seat to ease him out of the state secretary position.

• At the beginning of what promises to be a bumper season of electoral reform debate, the Electoral Council of Australia and New Zealand offers a paper on Internet voting in Australian electoral systems. A trailblazer on this score has been Estonia, which has provided for voting over the internet at national elections since 2007, and allowed for voting over mobile phones at the 2011 election, at which the overall take-up rate was nearly a quarter of all votes cast. However, simplifying matters somewhat in Estonia’s case is its national identity card. The paper observes that survey research by the Western Australian Electoral Commission found satisfaction that internet voting would be secured had increased from a third of all respondents in 2005 to a half in 2013. Electronic voting more broadly, including “kiosk” voting conducted within polling stations, is spruiked as offering lower costs, improved formality, more accurate capture of preferences (trials with overseas personnel in 2007 found a higher take-up rate for below-the-line voting), and opportunities for assisting vision-impaired or non-English speaking voters.

• I’ve had too little to say on the Labor leadership election process, of which I’m all in favour, but there’s a useful review of the New Zealand and British precendents from David Donaldson in Crikey.

• Six months out from the state election, there was an EMRS opinion poll from Tasmania out yesterday, which you can read all about in the post below.

• Another new post directly below deals with the state by-election for Miranda in New South Wales, to be held on October 19.

Essential Research: 53-47 to Coalition

The new government has its first poll, sort of. Also featured: an overview of in-doubt seats from the real election, of which I count four (two House, two Senate).

The new government’s first opinion poll is testament either to a striking weakness in its honeymoon effect, the fact that it’s only partly a post-election poll, or the observed tendency towards constancy in results from the pollster in question. That pollster is Essential Research, and its poll is its routine fortnightly average of federal voting intention conducted online from samples of about 1000 respondents each week. The latest result was thus half-conducted over the period of the election itself, such that one might dispute its provenance as a post-election poll (which you can pile on top of general doubts about the value of any polling conducted immediately after a change of government). For what it’s worth, the poll has the Coalition on 44% of the primary vote (45.7% at the election on current figures), Labor on 36% (33.5%) and the Greens on 9% (8.4%). The published 53-47 two-party preferred (the current election result being 53.4-46.6) is weaker for Labor than the primary vote shifts suggest it should be, which may be because they are still using preference allocations from the 2010 election.

Further questions, which unlike voting intention were derived from this week’s sample only, have 38% rating the election of micro-parties to the Senate as “good for democracy” against 25% for bad, although I’d like to see more specific questions in relation to this topic. Forty-four per cent believe the lack of a Coalition Senate majority will make for better government against 30% for worse. Respondents were asked about various aspects they might expect to get better or worse under the new government, including the surprising finding that cost of living and interest rates are expected to be worse. A finding on the state of the economy is an instructive insight into the influence of partisan considerations on such polling. Overall, 40% describe the state of the economy as good and 25% as poor, compared with 36% and 30% when the question was last asked in mid-July. Tellingly, the good rating among Coalition voters is up 14 points to 32% while poor is down ten points to 35%, while Labor voters are down nine points on good to 50% and up four points on poor to 18%.

As to proper election results, this site continues to follow close counts in dedicated posts as linked to on the sidebar. As far as I’m concerned, there are four seats which are still in serious doubt – two in the House, and two in the Senate. The 1550 votes in Indi are too few to reverse Sophie Mirabella’s 405-vote deficit against Cathy McGowan, while the 849-vote lead of Labor’s Julie Owens in Parramatta is enough to withstand anything the outstanding 3258 votes might conceivably throw at it. That leaves:

Fairfax. Continuing an ongoing trend, Clive Palmer’s lead shrank yesterday from 502 to 362. This resulted from a heavy flow of postals against him (758-465) being greater than an advantage on absents (722-569 in his favour on yesterday’s batch), both of which reflect the earlier trend of postal and absent counting. The number of outstanding absents and postals has diminished to around 1000 each, which leaves the ball in the court of about 2500 outstanding pre-polls, which have so far gone nearly 57-43 against Palmer. If all existing trends continue over the remainder of the count, Palmer will land a few dozen votes short. He will then perhaps take the matter to the Court of Disputed Returns, his current Federal Court injunction to have counting stopped presumably being doomed to failure. Palmer has been invoking an anomaly in the count, much remarked upon on this site, in which the Coolum Beach pre-poll voting centre result had a more-than-plausible number of votes for LNP candidate Ted O’Brien and a mismatch with the number of votes recorded for House and Senate. However, much as Palmer might wish to invoke a ballot box-stuffing operation at once brilliantly efficient in execution and bone-headedly stupid in conception, the AEC’s explanation that the Coolum Beach and Nambour PPVC results had been entered the wrong way around is likely to stand up in court. It is a duly troubling prospect that Palmer’s Senate representatives may emerge as important players in the looming round of electoral reform.

McEwen. After late counting initially flowed heavily against him, Labor member Rob Mitchell has rallied with a strong performance on absents and late-arriving postals. He now leads by 192 votes, which will widen if the tide continues to flow his way. However, it remains to be seen what as many as 5000 pre-polls hold in store. The first batch favoured Mitchell 497-458, but the remainder might come from less favourable areas.

Western Australian Senate. The most excellent Senate modelling of PB regular Truth Seeker illustrates the delicate balance of the count here, and the stars that need to remain aligned if Wayne Dropulich of the Australian Sports Party is indeed to find his way to the Senate off 0.2% of the vote. Key to the outcome is Dropulich remaining ahead of the Rise Up Australia party after distribution of preferences from Australian Voice, after which his snowball builds all the way to a quota. This might yet be undone by a gentle trend towards RUA on late counting, together with the unknown quantity of below-the-line votes. Should Dropulich fall short, not only will his own seat instead go to Zhenya Wong of the Palmer United Party, but the complexion of the race for the final seat between Scott Ludlam of the Greens and Labor’s Louise Pratt will change. This is because the comfortable win presently projected for Ludlam is achieved off Palmer preferences, which won’t be available to him if the votes are used to elect Wong. Truth Seeker’s projection is that Pratt will “almost certainly” defeat Ludlam on a scenario in which Wong is elected.

Tasmanian Senate. The issue here can be neatly observed on the ABC results calculator, the crucible of the outcome being the second last count (Count 24). Here the calculator, which treats all votes as below-the-line, has the Liberal Democrats leading Palmer United Party candidate Jacqui Lambie by 29,705 votes to 28,608. Since Palmer preferences favour the Liberals over the Liberal Democrats, their candidate’s exclusion then delivers victory to the third Liberal, Sally Chandler. However, if that gap of 1097 should close, the Liberal Democrats will be excluded instead, and most of the votes then distributed will flow to Lambie and secure election for another PUP Senator. The size of the gap might make that appear unlikely, but Tasmania has an unusually high rate of below-the-line voting, and one might surmise that it will favour the greatly more visible PUP over the Liberal Democrats. UPDATE: Looks like I wasn’t taking the Sex Party challenge with due seriousness – they win the last seat that might otherwise go to Liberal or the Palmer United Party if they stay ahead of Labor at Count 21, as they get Palmer preferences ahead of the Liberals. The current count has them doing this by the grand total of 14,275 to 14,274.