Coalition slumps but Morrison gains in Newspoll; electoral changes to curb micro parties

This week’s Newspoll, conducted August 25-28 from a sample of 1,528, gave Labor a 54-46 lead, a one-point gain for Labor since three weeks ago. Primary votes were 40% Labor (up one), 36% Coalition (down three), 10% Greens (down one) and 3% One Nation (steady).

Despite the three-point loss for the Coalition’s primary vote, Scott Morrison’s net approval returned to positive after going negative three weeks ago for the first time since the start of the pandemic. 49% were satisfied (up two) and 47% dissatisfied (down two), for a net approval of +2, up four points.

Read more: First negative Newspoll rating for Morrison since start of pandemic; 47% of unvaccinated would take Pfizer but not AstraZeneca^[1]

Anthony Albanese gained one point for a -7 net approval. Morrison led Albanese by 50-34 as better PM (49-36 previously). Newspoll figures are from The Poll Bludger^[2].

This Newspoll contradicts the Resolve poll^[3] that I wrote about on Thursday. In Resolve, the Coalition led Labor by 40-32 on primary votes, from 38-35 in July. Although Resolve’s issue-based questions have been skewed to favour the Coalition (see below), this would not affect voting intentions.

Newspoll is the most-watched poll in Australia, but that doesn’t mean it is right. At the 2019 federal election^[4], the final Newspoll was 51.5-48.5 to Labor, overstating Labor by 3.0%. Newspoll has made changes to its methods since that debacle, but we need to wait until the next federal election to see if its accuracy improves.

In November 2018, Essential and Ipsos both gave Labor 52-48 leads while Newspoll continued to have Labor ahead by 55-45. Analysts considered Newspoll right, and that convinced other pollsters to “herd” their results to Newspoll for the 2019 election. Labor’s weakness in Resolve compared to Newspoll should not be ignored.

Read more: Newspoll probably wrong since Morrison became PM; polling has been less accurate at recent elections^[5]

Although there was a three-point loss for the Coalition in Newspoll, it went to Others, not Labor, and may be easier to recover. It’s plausible this drop was due to Clive Palmer’s ubiquitous anti-lockdown ads gaining some support from the Coalition’s right.

The improvement in Morrison’s ratings is encouraging for the Coalition. During the pandemic, Morrison’s ratings have been far ahead of what would be expected from voting intentions.

The Guardian’s datablog^[6] shows 27.2% of Australia’s population (not 16+) is fully vaccinated, up from 17.8% three weeks ago. We still rank 35 of 38 OECD countries after we overtook Mexico, but were passed by South Korea. The Age^[7] shows 57.5% of 16+ have received at least one dose and 34.2% are fully vaccinated.

Electoral legislation to curb micro parties passes parliament

Legislation^[8] that will increase the number of members required to register a party from 500 to 1,500 passed parliament with Labor support last week. Existing parties that are not currently represented in federal parliament will have three months to show they have 1,500 members or be deregistered.

A 1,500 member threshold with Australia’s population is not a big ask. If parties cannot meet this threshold, it is unlikely they would be electorally viable.

The legislation also allows older parties to reserve their own names^[9], such as Liberal or Labor. This will force the Liberal Democrats and the proposed New Liberals to change their names as the Liberal party was first. Confusion between the Liberals and Liberal Democrats has assisted the latter at some elections. The same applies for Labor and the Democratic Labour Party.

Analyst Kevin Bonham is tracking parties^[10] that are likely to be affected by these changes. If the government wants these changes implemented, an election is unlikely to be called until next year.

Pre-poll voting will be limited^[11] to the final 12 days before an election; it’s currently the final 19. Sorting of pre-poll booths will start at 4pm on election day, two hours before polls close. This is an attempt to avoid not having results from large pre-poll booths until very late on election night.

Resolve poll’s skewed question on reopening

I wrote last Thursday^[12] that in last week’s Resolve poll, 62% supported state and territory leaders sticking to the plan to ease restrictions at 70% fully vaccinated, with only 24% saying the leaders should go their own way.

However, there was a paragraph added to the bottom of this article^[13] after I had first looked at it. This paragraph contains the actual question the poll asked, and includes the clause “such as using less severe restrictions once their populations reach 50 per cent vaccination or easing restrictions at 70 and 80 per cent if case numbers are still high”.

So the choice in this question was between sticking to the plan or easing restrictions faster, with no option for easing restrictions slower.

This is not the first time Resolve has skewed questions to favour the Coalition. I wrote about a skewed Resolve question on climate change in its June poll.

Read more: Victorian Labor holds comfortable lead; flawed climate change question in federal Resolve poll^[14]

Despite lockdowns, unemployment rate falls to 4.6% in July

Although Sydney was in lockdown for all of July, the ABS reported^[15] on August 19 that the national unemployment rate for July fell 0.3% from June to 4.6%. The unemployment rate was last this low before^[16] the 2008-09 global financial crisis.

A 0.2% drop in the participation rate to 66.0% partly explains the unemployment rate’s fall, but the employment population ratio – the percentage of eligible Australians who are employed – decreased less than 0.1% to 62.9%, just off its peak that was higher than for the last ten years.

Economic analyst Greg Jericho wrote in The Guardian^[17] that many people during lockdowns were still counted as employed, but worked zero hours. If those people were counted as unemployed, the unemployment rate in July would be 6.8%, up from 6.5% in June and 5.1% in May, when there were no lockdowns. By this measure, the unemployment rate in April 2020 was 14.4%, much higher than the official 6.4%.

Very large poll for Australian Conservation Foundation finds strong support for climate action

Nine newspapers reported on Monday^[18] that a poll of 15,000 respondents conducted by YouGov for the Australian Conservation Foundation found 67% believed the government should be doing more to address climate change.

I am very sceptical of this poll because it was conducted by a lobby group with a vested interest in promoting climate change action. And the actual results at the 2019 election were not favourable to climate change action.

NSW state redistribution finalised

Boundaries for the 93 NSW lower house seats that will be used at the 2023 NSW state election have been finalised. ABC election analyst Antony Green^[19] wrote that Labor gains one notional seat from the Coalition, reducing the Coalition’s seat lead over Labor to 47-37 from 48-36, with three seats each for the Greens, Shooters and independents.

Opposition Leader Chris Minns’ seat of Kogarah has had its Labor margin cut from 1.8% to just 0.1%.

References

1. ^{^} First negative Newspoll rating for Morrison since start of pandemic; 47% of unvaccinated would take Pfizer but not AstraZeneca (theconversation.com)
2. ^{^} The Poll Bludger (www.pollbludger.net)
3. ^{^} Resolve poll (theconversation.com)
4. ^{^} 2019 federal election (theconversation.com)
5. ^{^} Newspoll probably wrong since Morrison became PM; polling has been less accurate at recent elections (theconversation.com)
6. ^{^} Guardian’s datablog (www.theguardian.com)
7. ^{^} The Age (www.theage.com.au)
8. ^{^} Legislation (parlinfo.aph.gov.au)
9. ^{^} reserve their own names (theconversation.com)
10. ^{^} tracking parties (kevinbonham.blogspot.com)
11. ^{^} voting will be limited (antonygreen.com.au)
12. ^{^} wrote last Thursday (theconversation.com)
13. ^{^} this article (www.smh.com.au)
14. ^{^} Victorian Labor holds comfortable lead; flawed climate change question in federal Resolve poll (theconversation.com)
15. ^{^} ABS reported (www.abs.gov.au)
16. ^{^} this low before (tradingeconomics.com)
17. ^{^} The Guardian (www.theguardian.com)
18. ^{^} reported on Monday (www.smh.com.au)
19. ^{^} Antony Green (antonygreen.com.au)