Exit poll – it was a Seinfeld election

This election was a Seinfeld election – it was more about personalities than plot, and not a lot happened. Before the election cost of living had been a major issue, but it disappears out of our qualitative research, apart from that, the issues remain much the same apart from perceptions of Anthony Albanese and Scott Morrison, and respondents optimism about the direction of the country.

This is the take out message from our polling taken on the weekend of the election which you can download from here.

On our quantitative analysis Albanese went into the election with a -5% net favourability rating and came out with +4%. Coalition voters didn’t change their view of him, but his own Labor voters did by 20 percentage points, Greens 15, Nationalists 12, Others 16 and Independents 38.

In our first poll most respondents didn’t express a very detailed impression of Albanese, but by the second poll a number of descriptive words were present including “decent”, “genuine”, “caring” and “honest”. By contrast Morrison’s approval fell from a net -27% to net -36%. This was driven by a significant drop of net support from Liberals of 19 percentage points, and Nationalists 22 percentage points.

There was a general positive move in sentiment on the question of whether the country is heading in the right direction between our two polls, but it went from -36% to -24%, so it was still quite negative. Sentiment became more positive amongst all groups apart from Other and Liberal where it went in the other direction. The election of a left of centre government can explain the improved sentiment amongst the left, but the change in the Nationalist vote is either a statistical flute, or it can be inferred they wanted the Liberals to lose.

Nevertheless, preferred prime minister ratings remained quite stable, indicating that voters didn’t determine their vote on the basis of their approval of the two leaders.

What appears to have happened is that the Coalition ran a protest vote campaign. The essence of such a campaign is that you tacitly admit that the other side is likely to win. “It won’t be easy under Albanese” fills that role. You also don’t run hard on your own policies, but on the other sides. At the same time Albanese ran a small target strategy, trying to avoid Bill Shorten’s fate which was largely due to his wide and radical policy promises. So we had two parties not promising very much, and trying to leverage their vote off the shortcomings of their opponents.

This resulted in the extremely low first preference votes for the major parties, and a swing to minor parties. Electors protested, but they didn’t protest by going to the coalition, they went to minor parties who offered something in the way of positive policies. Most of those parties were ones who supported radical change in climate policy, but our qualitative results suggests that climate as an issue was no more important this election than previous ones, and was a concern mostly of voters on the extreme left. The corollary of this is that the Teals and the Greens won a lot of votes for reasons unconnected to the substance of their policies, and that these voters can easily defect to another party in the future.

Election issues were pretty similar to the first poll with Coalition and Nationalist voters clustered around economic and national security issues, while Labor, Greens and Independents were worried about climate change, integrity and corruption, and health (including Medicare). Morrison was also an issue for the left, while Albanese was not particularly an issue for the right (indicating a failure of the protest vote campaign and suggesting Labor’s negative campaign against Morrison was more successful than the Coalition campaign against Albanese).