Poll Roundup: "Deeply Unpopular By December"

2PP Aggregate: 53.1 to Labor (unchanged in two weeks)
Labor would easily win federal election "held now"

This week's federal poll roundup title is brought to you by an AFR article entitled Abbott on the nose in regions. The article was published in mid-April, just before the Budget, at a time when the government had recorded a 48% 2PP in a Nielsen poll (remember those?) and was just starting to head south from a borderline winnable 49:51 position it had been locked in for five months.  The final paragraph of the article was:

"Those who are aware of the severity of some of the budget decisions that have been made, are warning that the [polling] situation is likely to get worse before it gets better. The Coalition is bracing to be "deeply unpopular by December" said one source."

It's December now!  So how did that prediction go?



My suspicion is that the AFR's source expected budget anger to be some kind of slowly increasing dissatisfaction rather than an immediate slap.  Nonetheless things did indeed get worse, then they briefly got back to about where they were, and then they got worse again.  There has not, yet, been any "better".

In the long term, the average 2PP polling position for incumbent governments is exactly 50:50.   About 15% of the time governments are running at 46:54 or worse, and this is my definition of "deeply unpopular".  The Coalition spent five weeks on the fringes of that zone following the Budget, and is now in a position that I wouldn't call "deeply unpopular", but that isn't too far off it.

This week's polls

This week's polling started with a Morgan multi-mode that had Labor up 53:47 (with exactly that expected from the primaries).  This was actually a relatively good result for the Coalition given this series' house effect to Labor of 1.5 points.  I should note that Morgan's purely SMS polls in the Victorian state election did not go well at all, usually showing massive skew to the Greens and producing very volatile 2PP figures, but so far their national multi-modes have not displayed these foibles.

Newspoll produced a 54:46 lead for Labor off primaries of 37 each for the major parties and 13 each for the Greens and Others.  Yet again the combined figure for Greens and Others is high compared to that for other pollsters and without much doubt reality.  Newspoll also seems to be on a bit of a pro-Labor run at the moment, with my modelled 2PPs for the last four running 1.2 points above the aggregate, and the published 2PPs 1.45 points above.

Essential meanwhile has released a 53:47 (up one) to Labor off primaries of 40-40 with 9 for the Greens and 3 for PUP.

The strong Newspoll and weak Morgan for Labor cancel out and my aggregate is the same as two weeks ago at 53.1 to ALP.  Here's the tracking graph:

Leaderships

Newspoll delivered another nasty result for Prime Minister Abbott, who now has a net satisfaction rating of -24 (33-57).  It can be argued whether he is "deeply unpopular" (18% of PM netsats have been that bad or worse, but only 10% for PMs who won re-election at the end of that term).  However, it is his worst apart from five consecutive readings in the range of -26 to -31 from mid-May through mid-July.  His career average netsat is now -15, not much higher than the -19 for Gillard, for example.

Abbott's dissatisfaction rating has now been 50 or worse in every Newspoll for seven months.  This is by no means fatal as the same had been true of Paul Keating for eleven months when he was re-elected.  (However, the same disclaimer always applies with Keating's ratings: the win wasn't about him.) Malcolm Fraser also had a similar run lasting for most of a year in 1979-80 before being voted back in, but John Howard's longest such run in eleven years in office was four months.

Bill Shorten's netsat is a nondescript -4 (39-43) and his "better Prime Minister" lead of seven points is the highest it has been since the initial Budget backlash (when it was ten points three times.)

An Essential question about whether the the PM had "fundamentally kept faith with the Australian people" as concerns election promises found 31% agreeing that he had and 56% disagreeing.  Those disagreeing included 21% of Coalition supporters, although only 3% of Coalition supporters strongly disagreed.  I am a little wary of this question because of its preamble "In response to claims that he has broken election promises," and would have preferred one that referred to the general question of keeping election promises, but not to specific claims.  All the same the jaded response from some Coalition supporters is significant.

Essential also had Abbott-vs-Shorten attribute polling.  Compared to the last poll taken in May the polling for Abbott showed small to modest improvements in eight attributes and small worsenings in four, while that for Shorten showed five improvements and seven declines with no major difference on anything.  Shorten continues to outperform Abbott on almost every attribute, many of them by large margins.

Other Essential polling found that respondents continue to oppose remaining budget bills, mostly strongly (though 39% continue to support a six-month dole wait for the under-30s, in my view the worst of the lot).  Essential also had some polling on the defence force pay increase issue that is rather hard to make sense of because it is not clear whether all respondents who found the below-inflation 1.5% increase unfair were objecting to it being too low as opposed to too high.

Next week unless something startlingly impossible happens, the government will have trailed in polling for one year, showing that while its unpopularity isn't that deep by historical standards, it is proving stubbornly resilient.  I will have more about that unwanted (except for its opponents) anniversary next time.