When it comes to explanations for the election result, we’ve heard it all. Like, really. Everything.
But is there any data that can help us unpick what happened?
Well, our secret ballot voting system means we can’t figure out how individuals or specific groups of people voted.
But we do know a fair bit about the voters who make up different electorates in Australia, and we also know how those electorates voted.
We’ve brought together those two sources of information to see what patterns emerge. There’s no one big ‘gotcha’ that explains every result, but then again, we didn’t expect to see that.
What does emerge is that some of the key voting trends in 2022 were already brewing in 2019.
Let’s start with climate change.
The dots on this chart represent each of the 151 electorates in Australia. The colours show the winning party back in 2019.
They’re a bit hard to tell apart, we realise. But bear with us, that’ll change.
The seats are ordered by the percentage of voters who said climate change was the most important issue to them.
So the seats on the left of the chart are the least concerned about climate change, and those on the right are the most concerned.
We’ve highlighted a few seats that are worth keeping an eye on.
Think of them as markers to keep your bearings.
Now, let’s add another dimension.
We’ve added in the Coalition’s two-candidate preferred vote (2CP) in 2019 on the vertical axis.
The electorates above the 50 per cent line were won by the Liberal and National parties last time round.
On its own, this doesn’t tell us that much.
But let’s see if anything changes in 2022.
Can you see a pattern in the changes? (Try scrolling up and down).
There’s a bit going on so don’t worry if you can’t.
The three seats we’re keeping an eye on — all at different levels of concern for the climate — moved in different directions.
Wentworth and Ryan moved below the 50 per cent mark, meaning the Coalition lost them at this election. Lyons swung towards the Coalition, but not quite enough to change hands.
Let’s rewind to the 2019 results for a moment.
Now, keep your eye on that cluster of blue Coalition seats towards the right of the chart, near Wentworth. Remember, those were won by the Coalition in 2019 but have high levels of concern about climate change.
Watch them as we transition to 2022 results again. Here we go.
You might notice they all drop below that 50 per cent line, and they all turn ‘grey’.
That’s because they were won by independent candidates who argued in favour of greater action on climate change — these are all ‘teal’ seats now.
But looking across all the seats, it’s hard to tell whether there’s a pattern here, right? There are movements in both directions as we change years.
Let’s try something different.
Instead of the Coalition’s vote, we’re now looking at the change in Coalition vote between the 2019 and 2022 elections, which is often called the swing.
There’s a bit going on here, so let’s step through what this chart is telling us.
Seats that swung against the Coalition in 2022 now appear below the 0 per cent line. The lowest seats saw the biggest change in vote towards Labor, minor parties or independents.
And seats that swung towards the Coalition appear above the line.
It’s a little easier to see the trends without the soup of colour. We’ll add a trendline as well.
So what we’re looking at now is the relationship between change in the Coalition vote and concern about climate change.
As a general rule, the tighter the grouping of dots around the trendline, the more confident we can be that there’s a stronger trend.
Electorates concerned about climate change did swing slightly more strongly against the Coalition than electorates with lower concern, but the relationship is fairly weak.
What about in 2019?
We’ve now changed the chart to show us the change in Coalition vote between 2016 and 2019.
See how the dots are much more closely grouped around the trendline?
That means that the correlation between concern about climate change and swing away from the Coalition in 2019 was far more pronounced.
However, as we know, the Coalition held on to win government in 2019. The swings against it were not large enough for it to lose seats.
But take a look at Wentworth and Warringah.
They’re outliers here. Knowing what we do now, they were swinging ahead of the independent pack.
In an early stirring of the teal storm that arrived in 2022, independent Zali Steggall defeated Tony Abbott at the 2019 election in Warringah.
Dave Sharma narrowly won Malcolm Turnbull’s former seat of Wentworth back for the Liberals, after they had lost it to independent Dr Kerryn Phelps in a by-election.
Keep in mind that swings in successive elections build on each other.
So in 2019, when seats concerned about climate change swung against the Coalition, the stage was being set for the ‘teals’ in 2022.
But before we draw too many conclusions from one trend, let’s look at things from another angle.
This chart orders the electorates by how far they are from their nearest capital city.
See how the markers for 10km, 100km and 1,000km are evenly spaced across the bottom?
That’s to account for the uneven distances between electorates in Australia.
Now, here’s that distance measure plotted against the change in Coalition vote between 2016 and 2019.
There are the same two outliers again — this time representing major swings away from the Coalition in seats close to the city.
Outliers aside, the general trend is that regional seats swung towards the Coalition, while closer to the city there were small swings away.
So, what happened three years later?
Similarly to the story on climate, the results were mixed. But the trend was the same — the Coalition vote was weaker in the city than in rural areas in 2022.
And this time around, the swings against the Coalition in the city were much stronger.
OK, so you probably could have told us that the Coalition wasn’t as popular in the inner city as in rural Australia.
If that one didn’t surprise you though, this next one might.
While some trends — like the two we’ve just looked at — continue in successive elections, others do the opposite.
Here are the electorates again, this time ordered by the percentage of voters who said the economy was their “most important issue” in the 2022 election.
Unsurprisingly, in line with the general view that the Coalition benefits when the economy is the focus, seats where the economy ranks highly are a sea of blue.
When we add in the swing in 2019, we can see the Coalition’s success in campaigning on their credentials as the better economic managers, and attacking then-Labor leader Bill Shorten as “the Bill Australia can’t afford”.
The electorates most concerned about the economy swung more heavily towards the Coalition.
Once more, our two outliers are down the bottom, bucking the trend.
But here’s where it gets interesting.
There was a complete reversal in 2022. The electorates most concerned about the economy swung the most against the Coalition in this year’s election.
There are many different ways to interpret this.
One reading is that the economy wasn’t as important to Australians in 2022, and they voted on other issues, or based on their views of the party leaders.
Another possibility is that while the Coalition campaigned on the economy again, its message didn’t cut through — perhaps voters instead preferred Labor’s plan, including its vow to lift wages.
The combinations of these three trends — climate, the regional divide, and a seeming lack of Coalition cut-through on the economy — helped hollow out the Liberal Party’s heartland: affluent inner-city seats. What started with signals from two seats in 2019 became a teal wave in 2022.
La Trobe University political scientist Andrea Carson says the seats the Liberals lost are “highly educated electorates with politically engaged constituents and strong independent challengers. Usually, it would be in their economic interest to vote for the Coalition, but we’re not seeing that this election.”
Dr Carson also pointed to climate change as a factor: “I think that shows that the warning wasn’t heeded by the two major parties. They didn’t spend much time talking about climate change and those that cared about it have. It’s been a great benefit to the Greens and independents.”
Indeed. This election has seen unprecedented lows in the primary vote for the major parties in Australia, and corresponding highs for independents and minors.
Labor’s primary vote of 33 per cent at this year’s election is the party’s lowest since the 1930s. And they won the thing.
The biggest swings to the Greens and independents were seen in very high and very low socio-economic areas, in what could be a sign that perhaps only ‘middle Australia’ feels represented by the major parties.
Dr Carson puts this trend down to post-materialism in richer areas.
“When you’re pretty comfortable and well-off, you can afford to care about things beyond material concerns. And I think that’s why you’re seeing the independents and Greens pick up there,” she said.
As for poorer areas, “the Coalition hasn’t promised any policies that benefit the lower socio-economic, so it’s really not in their economic interests to be supporting the Coalition”.
Electorates in lower socio-economic areas such as the west of Sydney and northern South Australia saw swings of more than 20 per cent to independent candidates. While Rob Priestly fell short in Nicholls, Dai Le was successful in Fowler. She distanced herself from the teals, saying, “My electorate here has a different need to the electorates that the teal independents are representing.”
Looking at the chart of socio-economic trends above, Fowler sits towards the far left, compared to a teal seat like Goldstein towards the right — this suggests there are indeed very different groups of voters in the two seats.
While these two groups of electorates are on opposite ends of the scale, the common theme is discontent with what the major parties are offering.
Another new trend that emerged was the rejection of the Coalition in areas with high proportions of Chinese ancestry. The Coalition’s rhetoric on China seems to have made a difference in seats like Bennelong and Reid, two seats in Western Sydney that Labor picked up.
Election analyst Ben Raue did a similar analysis, but he used more granular voting booth data instead of just electorate-level data. He found that “while suburbs with low proportions of people of Chinese ancestry moved in both directions, those with higher proportions almost entirely swung towards Labor, some moving quite far.”
Don’t worry, we haven’t forgotten about that other thing that happened in the past three years.
Yes, the pandemic.
The loss of three previously ‘safe’ Liberal seats in WA — Curtin, Tangney, and Pearce — have been attributed to the Mark McGowan effect. The WA Premier remains extremely popular in the west, having clashed with Scott Morrison over border closures for COVID-19.
The huge swing in the west is evident in this chart showing the relationship between high rates of COVID vaccination and Labor’s vote.
All the seats in the top-right corner of that chart are in Western Australia.
While we can’t attribute the overall result at this year’s election to any one of these trends, the combination of them hints at an explanation for our new-look parliament.
And together, they contributed to the most visible trend of all: the number of traditionally ‘safe’ seats that fell last Saturday.
Let’s have a look at how that played out.
Here are the electorates ordered and coloured based on the Coalition’s vote at the 2019 election.
Electorates around the 50 per cent mark — the marginal seats — tend to change hands at election time.
What’s unusual about 2022 is the number of ‘safe’ ones that have been lost, particularly by the Coalition.
To take a closer look at that, let’s remove the marginal seats from the chart for the moment.
Here are seats that were supposed to be ‘safe’ before Saturday night, at both ends of the political spectrum.
The major parties can usually rely on winning these as they have strong voter bases there.
To get an idea of how these seats usually behave, here’s what happened in 2019.
There was just a single outlier.
It’s our old friend Warringah.
To unpack where the votes went in these ‘safe’ seats in 2022, let’s switch both axes to show the results from this year.
That’s a wildly different outcome.
The blue dots were nicely bunched together in 2019, now they’ve all spread out.
That’s not what ‘safe’ seats are meant to do. They’re meant to sit quietly and not move very far.
But the voters in Liberal heartland didn’t do what they’re ‘meant to’. As we know, a number of these seats swung far enough to change hands.
We can see the teals and the Greens winning Ryan, Goldstein, North Sydney, and Mackellar.
They all ranked highly on concern about the climate, and most are nestled in inner-city areas.
This is despite them all being considered ‘safe’ before the count.
Another casualty was Bennelong in Sydney.
Labor claimed Bennelong for only the second time in more than 70 years.
Bennelong has one of the highest Chinese-Australian populations in Australia.
Three previously ‘safe’ seats for the Liberals fell in WA: Curtin, Tangney, and Pearce.
By looking closely at the past, we start to see trends in what is moving votes. These can help us to understand the results that shape our parliament.
But one thing that’s clear is there’s no guarantee these trends won’t undergo a complete reversal, or be overshadowed by new trends that emerge in the next three years of Labor rule.
What should I read next?
About the data
- ‘Most important issue’ (climate change, economy): ABC Vote Compass 2022
- 2022 and 2019 election results: AEC; ABC Election Computer
- Distance from capital city: AEC commonwealth electoral boundaries
- Socio-economic: ABS – Census of Population and Housing, 2016, TableBuilder. The Index of Relative Socio-economic Advantage and Disadvantage (IRSAD) summarises information about the economic and social conditions of people and households within an area, including both relative advantage and disadvantage measures.
- Chinese ancestry: ABS – Census of Population and Housing, 2016, TableBuilder. Ancestry first choice (Chinese Asian)
- Covid vaccination rate: Federal Department of Health (electorate analysis courtesy of The Guardian)
- Trendlines calculated using linear regression
- Electorates without a candidate from the target party finishing in the top two removed from 2CP charts.