For this post I wanted to put together a bunch of my thoughts the day after the WA election. I will be recording a podcast this afternoon and I hope to get it out before the end of the day.
Firstly, let’s zoom out and look at the scale of the result.
Labor won 41 seats out of 59 at the 2017 election, alongside 18 Liberals and Nationals. The 2021 result saw Labor win 53 seats, to just six for the conservative parties.
At this time of writing, Labor has won 40 seats, the Liberal Party has won five, and the Nationals have won 4. One seat (Albany) will either fall to Liberal or Nationals, bringing the conservative parties to a total of ten. There is also one seat where an independent is in with a strong chance of defeating Labor. That is Fremantle, where the independent was leading for much of last night, but is now on 29% behind Labor on 33%.
There are another eight classic seats where Labor is in a close race against the Liberal or Nationals parties. Labor is leading in Collie-Preston, Dawesville, Kalgoorlie, Pilbara and South Perth. The Liberals are leading in Kalamunda and Murray-Wellington and the Nationals are leading in Warren-Blackwood.
So Labor is currently leading in 45 seats, and are not out of the race in four others. Labor could win anything from 40 to 49 seats, with their number likely to fall a bit short of 45. So that is likely to produce a stronger Labor seat count than 2017.
The Liberals are currently one seat ahead of the Nationals, and are likely to be the bigger opposition party but that is not completely guaranteed. The Nationals would be neck-and-neck if they won Albany at 5-5, and are also leading in Warren-Blackwood. That would give them six seats, but the Liberals are the main opponent to Labor in seven other competitive seats. So it’s likely the Liberals will be the main opposition party but I can’t be certain.
Looking at raw voting figures, Labor’s primary vote is sitting on 41.8%, which is slightly less than the 42.2% they polled in 2017, but their two-party-preferred vote is substantially higher at 58.4% compared to 55.5%. This reflects how a large part of the anti-Labor swing went to independents and minor parties, and mostly (with the exception of Fremantle) has remained below the level where they win seats.
The total vote share for minor parties and independents is currently sitting on 24.5%, which would be the second-highest ever vote share for candidates outside of the Labor, Liberal and Nationals parties.
Last night, at the conclusion of the live blog, I published a map showing the 2CP margins in 2017 and 2025, but also the 2CP swings for 2025. This is based on the ABC’s estimates, and covers 54 seats with classic (Labor vs Liberal/Nationals) estimates at the moment. A few of these seats may switch on further counting but we can use the data we have now. I’ve now updated this map with updated data.
The overall trend has remained the same, but the swings to the Liberal Party are a bit bigger. Generally swings get bigger as you move from the inner city to the outer suburbs.
Another way to look at this trend is to look at the average swing from Labor to Liberals and Nationals in these 54 seats based on geographical classificaton:
- Nine inner city electorates – 8.0%
- 21 middle suburban electorates – 11.4%
- 14 outer suburban electorates – 16.0%
- 10 regional electorates – 14.2%
While Labor did poorly in regional areas, in terms of swing the biggest impact was in the outer suburbs of the metropolitan area.
I’ve also produced this next chart which shows the distribution of seats by margin. It excludes non-classic seats – specifically Roe, Baldivis and Moore in 2017, Fremantle in 2021 and Central Wheatbelt, Cottesloe, Fremantle, Mid-West and Roe in 2025.
The biggest differences between the 2017 and 2025 chart is that the ALP has a lot more seats on margins of 8-12%. They have similar numbers of seats on margins of over 12% or under 8%, but they hold a bunch more seats on those sizeable margins. This also makes a difference in terms of the swing needed for Labor to lose their majority. They hold 32 seats on margins of 8% or more right now, compared to just 26 in 2017.
Unsurprisingly, those extra 8-12% Labor seats are compensated by a reduction of Liberal seats on “safe” margins of 4-16%. There are actually more marginal Liberal/National seats on margins under 4%, before considering the various non-classic races like Cottesloe.
It’s important here to touch on the number of seats where the 2CP pairings are incorrect or unclear. As the minor party vote has increased, we are having more cases where the Commission incorrectly guesses the top two before election night, or even where we don’t know who those top two will be.
Right now there are three seats where we know that the wrong 2CP pairing was chosen. In Fremantle, Kimberley and Pilbara, the correct 2CP pairing is now known, but the Commission chose incorrectly. This is understandable, and hopefully they will complete the counting soon. In Fremantle, it seems likely that the count will be close between Labor and the independent but there is a lot of uncertainty. In Kimberley and Pilbara, the WAEC chose the Nationals rather than the Liberal Party. Kimberley seems to be a Labor win, but Pilbara is close enough to matter, with Labor narrowly ahead.
There are also eight seats where, on my judgement, I can’t be sure who has made the 2CP.
In three of these seats this could actually matter. Labor has lost Albany, but we don’t know if Liberal or Nationals will make the top two and win on the other’s preferences. Warren-Blackwood has a similar dynamic, but Labor is still in the count. Kalgoorlie is another seat where there are three prominent conservative candidates and it’s not clear which of them would emerge against Labor.
In Bibra Lake and Maylands, the Liberal and Greens are in a close contest for second place. Likewise Liberal and an independent are close in Thornlie. While the Greens and independent did well in these seats, this more reflects a very weak Liberal vote with Labor far out in front. Likewise in Central Wheatbelt and Roe, the Nationals have won easily but it’s not clear if Labor or Liberal will be in the 2CP.
Out of these eight cases, there are four where the WAEC is currently not reporting any 2CP figures: Albany, Murray-Wellington, Central Wheatbelt and Maylands. They also haven’t reported any 2CP figures in Perth, where their original pairing was Labor vs Liberal but for a while it looked like the Greens would make the top two. It now appears that the LIberal Party will come second, so I hope they can report whatever was counted there.
Overall it looks like there could be as many as eight non-classic races, and at bare minimum there will be three.
So before I finish, let’s run through the seats still in play.
Albany will either go to Liberal or Nationals. The Liberal is favoured right now but it is close. A 3CP count would be very handy here but I don’t expect to see one. We also expect Fremantle to be close between Labor and an independent but are waiting on a 2CP count.
Labor is leading the Liberal Party in Collie-Preston, Dawesville, Kalgoorlie, Pilbara and South Perth. Kalgoorlie also has the added complication of the Nationals having the potential to overtake the Liberals.
The Liberal Party is in the lead in Kalamunda and Murray-Wellington, while the Nationals lead in Warren-Blackwood. The Liberal Party is not out of the race in Warren-Blackwood.
I should also mention the Legislative Council. With about half the vote counted, there’s a lot of votes yet to be counted. It’s also worth mentioning that this initial count is only looking at above the line votes, so parties that do well on below the line votes will be underestimated. On the current numbers Labor and the Greens are on track for 19-20 seats between them which is enough for a clear majority, with Legalise Cannabis also winning a seat and Animal Justice an outside chance.
History of these sorts of elections from New South Wales suggests the left generally does well on preferences and later counting, which would suggest the total progressive bloc could reach 22 out of 37 seats. This is not surprising considering the scale of Labor’s win in the lower house. I will return to analysing the Legislative Council in coming days.
Great to read your analysis and info on contests in WA on the night – it’s fascinating and good news for federal Labor too – especially if they utilise the lessons which could improve their own performance soon!
I agree Jackie, great news for Federal Labor.
The thing I was most interested in, based on current leads
2021 Labor gains that stayed Labor: Bateman, Darling Range, Dawesville, Hillarys, Kalgoorlie, Riverton, Scarborough, South Perth
2017 Labor gains that Labor lost: Kalamunda, Murray-Wellington
2013 Labor seats that Labor lost: Albany, Fremantle
Blue not John, I can understand the reasons why those 2013 and 2017 Labor seats were lost this time round.
Kalamunda – open seat with loss of personal vote for the sitting Labor MP
Murray Wellington – potential controversies surrounding the sitting MP Robyn Clarke. Also, a naturally conservative leaning seat that was an upset win in 2017.
Albany – Naturally conservative leaning seat that also saw an upset result in 2013 (Peter Watson retained a highly marginal seat against the statewide swing and even saw a slight increase in the 2PP vote, similar to Bundaberg 2024 in Queensland)
Fremantle – strong campaign by teal like independent Kate Hulett (similar to Justine Davis in Johnston 2024 in NT).
South Perth already has 2PP counts complete for booths including pre-polls and postals on the day. What remains are absent votes and late postals which overall would be expected to lean Labor. With Labor already leading that should firm up to a more clear victory in time.
Pilbara seems to mostly be a case of the electoral commission establishing the correct challenger to Labor and beginning the preference count – but estimates from ABC/PB are giving Labor a healthy lead there.
Albany is a clear loss for Labor, with the only doubt being which candidate is the challenger. The Nat currently has a slim lead and should receive a stronger flow of preferences from defeated candidates, so to my eye they are the favourites there.
Fremantle is currently showing a slender lead for the independent using the ABC’s preference count, which uses 65% flow to Labor on Liberal votes and 20% flow to Labor on Greens. I’m not sure the flow from the Greens will be that low and 65% is probably the minimum to expect when considering that the Liberals preferenced Labor on HTV cards. And taking into account that what is mostly left to come are absents and late postals, which independents are known to underperform on, and I think Labor is favoured here.
Murray-Wellington is still waiting on a prepoll of over 7000 votes. Prepoll swings generally were stronger for the Liberals so it’s possible that once that reports, it will more or less confirm victory there.
Kalamunda is waiting on prepolls as well, since there was no prepoll booth in the electorate and they are instead cast as absent votes. My guess would be that the stronger swing will be included within the absent votes and firm up victory there, but that’s just a guess.
The others I’m not quite prepared to make an assessment on at this time.
This is a great win for Labor, no doubt, but I think we need to put it into context. This, without any doubt, is the most right wing Labor government to have ever occupied our shores. They are pro-gas, pro-mining, pro-industrial relations reform, and pro-big business. Actually, they probably embarrass the Victorian Libs because they are to the right of them. The Federal Government, however, is well left of centre. I believe there will be a very different vote federally compared to the State level in WA. The Liberal Party will pick up seats in WA in the federal election I expect, but perhaps only 2.
WA Labor is definitely not to the right of the Victorian Liberals. Aside from the fact that the Victorian Liberals are infamously right-wing and about one-third of the caucus could fit into the “Freedom” parties, I am quite sure they wouldn’t end native forest logging or introduce voluntary euthanasia if they were in government.
Calling a government with such policy the most right-wing Labor government ever is more than a little exaggerated. They appeal to the centre of politics in the electorate – that’s how most successful governments work.
@jackie the only problem is federal labor is unpopular with an unpopular leader, the federal govt is is huge deficiets as it is and is failing on CoL the direct opposite of their state colleagues
Regarding the “huge deficiets”:
In 2013, Australia’s government debt was $257.4 billion in gross debt and $161 billion in net debt.
At the end of 2022, Australia’s gross debt was $889.8 billion, which was 35.2% of the country’s GDP.
I’m no mathematician, but that seems to be an increase in net debt of $632.2 billion under the watch of the coalition government.
Labor by comparison delivered two budget surpluses ($22b for 2022-23 and $9b in 2023-24 respectively), followed by a forecast deficit of $28.3b for this current financial year.
Even if we want to label $28.3b as a “huge” deficit, it compares favourably to the average $70b deficit delivered during the Abbott-Turnbull-Morrison governments.
You are entitled to your own opinions, but not your own facts.
Adda, I do agree that that Labor in WA do absolutely appeal to the centre of politics in WA as do the ALP in Victoria and ALP in NSW. The federal government will find it very difficult to navigate this given the significant differences on what the ‘centre’ is in various states. Accordingly, I do think the ALP in WA will not perform at the same level federally as they did in the State election. Time will tell.
From what I have seen, Labor has done better in metropolitan Perth than expected. Labor ends up having more metro Perth seats than they did in 2017. Labor held many of the blue wall seats that they flipped in 2021 like South Perth, Bateman, Scarborough.
It would’ve been interesting if the Nats and Libs end up with the same number of seats but that’s unlikely.
There is a theory in economics called balanced budget theorem. If a govt were to balance its budget this has a deflationary impact. This would be more so in the case of surpluses.
Not sure of this as the interest payable on
Net govt dept is $53b pa
I checked that areas that have a sizable Muslim Population whether the Gaza Issue affected the votes and it seem like it did with significantly above average swing towards The Greens
-Kewdale Primary School (16% Muslim in Kewdale) had 12.3% swing to The Greens (19.9%) which surpasses the Liberal Vote (17.2%)
– Boyare Primary School (22.3% Muslim in Mirrabooka) had a 10.2% swing to The Greens (15.7%) which again surpasses the Liberal Vote (14.5%)