22 May 2025
1300 794 893
One of my favourite quotes is by the late Sir David Butler, eminent UK psephologist “Electoral history is littered with unexpected landslides.” What’s the history of election landslides and what do they say for future polls?
At the time of writing, there is one doubtful seat in the House of Representatives: Bradfield. For the whole second-half of the 20th century, Bradfield in Sydney's Upper North Shore was a safe, blue-ribbon Liberal seat.
At the time of writing, the two-candidate preferred vote is 56,229 for the “teal” independent Nicolette Boele and 56,183 for the Liberal candidate Giselle Kapterian, a margin of only 46 votes. While this result may change following a recount, nevertheless for the purposes of this piece I take it that Independent Boele will be the new MP.
And if this is the case, that Boele carries the seat, we'll have 13 crossbench members in the 48th Parliament. One of those will be the Greens member for Ryan (Queensland) Elizabeth Watson-Brown, the sole survivor in the four seats the Greens enjoyed in the 47th Parliament. Melbourne in Victoria and the Queensland seats of Brisbane, Griffith and Ryan were then held by the Greens before polls opened for 2025.
The non-Greens crossbench secured 12 wins in May 2022. Ten of those - in the seats of Clark, Curtin, Fowler, Indi, Kennedy, Kooyong, Mackellar, Mayo, Warringah and Wentworth - won again in May 2025.
The missing two who won in 2022 but didn't stand in 2025 were Kylea Tink in North Sydney (whose seat was abolished in the redistribution) and Zoe Daniel whose blue-ribbon seat of Goldstein in Melbourne returned to Liberal Tim Wilson. Incidentally, the nail biter for Wilson's seat represents the only gain made by the Liberal Party at the 2025 election. Those two outgoing seats were replaced by Boele gaining Bradfield and Andrew Gee winning Calare as an independent in 2025 where he had won as a National in 2022.
Therefore the number of non-Greens crossbench wins was 12 in 2022 and is 12 again in 2025. I find that interesting because so many experts confidently predicted an increase in the size of the crossbench. Instead it is three smaller, reflecting the loss of three Greens and the failure of the non-Greens crossbench to make any net gains.
The other 137 seats have split as follows:
Take a note of those numbers. They are very relevant in the light of the dummy spit by the National Party, which has now broken up what had been the Coalition.
The over-riding characteristic of these numbers is the fact of Labor winning many more seats than anyone predicted. For example, in my Switzer Daily article posted on Thursday 17 April, I gave an emphatic affirmative answer to the question posed: “Will Anthony Albanese be our PM after May 3?”.
In my prophetic piece, I wrote: “it will be a majority Labor government.” However, when I did the arithmetic on seats I could only find 79 that I predicted would be won by Labor.
So where did those extra 15 seats come from? There were 12 gained from the Liberal Party not predicted by me:
Then there were two gained from the Greens not predicted by me, Griffith and Melbourne. Therefore, I did not predict the defeat of Peter Dutton in Dickson, nor did I predict the defeat of Adam Bandt in Melbourne.
The 15th seat was Bennelong. Because its boundaries had been changed to make it much better for the Liberal Party, it was notionally Liberal on my new pendulum. The Liberal candidate, however, lost badly and the sitting Labor member, Jerome Laxale, retained his seat easily, getting a two-party preferred vote swing in his favour of 9.3%.
I was not alone in my arithmetic being excessively cautious. Former Labor minister Barry Jones wrote an article published in The Saturday Paper for 26 April to 2 May titled “The last majority government” in which he wrote:
“The likeliest result on May 3 will be a small majority for Labor, but it will almost certainly be the last victory for a hegemonic party”.
The first part of that sentence reflected my thinking at the time, but certainly not the second part. He went on:
“All future parliaments will have a three-way split in the House of Representatives, as is already the case in the Senate, except where Labor and the Coalition gang up to pursue their own party interests.”
I have always rejected that view. I think there will be occasional minority governments, but each hung parliament will be followed (as in 1943 and 2013) by a decisive win for either Labor or the Coalition. I spell Coalition with a capital C because I have no doubt that the next conservative government will be the standard Coalition that Australia has known since the 1920s.
A characteristic of this election was the repeated display of wishful thinking by the personalities on the program jocularly known as “Sky News After Dark”. The best example was on the night of Thursday 1 May on the Sharri program. The host, Sharri Markson, asked a panel for their predictions, every one of which greatly overstated the number of likely Coalition seats.
Finally, Markson herself concluded by predicting 72 Labor, 66 Coalition and 12 on the cross bench. She understated Labor by 22 seats, overstated the Coalition by 23 and understated the cross bench by one. Only a tone-deaf person could have made such a prediction on 1 May.
Anyway, on polling day, The Weekend Australian carried its final Newspoll that recorded a two-party preferred vote for Labor of 52.5% and 47.5% for the Coalition. With the final percentages being 54.8 and 45.2, the error in Newspoll was to underestimate Labor by 2.3%.
The difference between Newspoll and Markson, therefore, is that Newspoll picked up some of the swing to Labor caused by the campaign – but not all of it. Markson, on the other hand, made a prediction that would have been quite reasonable if made back in March, but not two days before polling day.
On 13 December last year, Switzer Daily published an article under my name titled “What’s my favourite election for this year?”. The election I nominated was that of the Northern Territory on 24 August. That election was my favourite because it was an unexpected landslide, which saw the then Labor government and installing Lia Finocchiaro, leader of the Country Liberal Party, as the new NT Chief Minister. Labor was reduced from 15 seats to four and the CLP numbers increased from seven to 17.
That article actually features one of my favourite quotations. It was from my friend, the late Sir David Butler (1924-2022). He was the eminent psephologist of the United Kingdom in the second half of the 20th Century. The quotation is:
“Electoral history is littered with unexpected landslides.”
Butler would say that in respect of many elections around the world but all of them had this feature in common: the electoral system was always one of single-member electoral districts.
I can apply that truth to several Australian state elections and at the federal level for particular states, for example the landslide to Labor in Queensland in December 1961 and the landslide to the Liberal Party in South Australia in November 1966. However, my search for such a case at a federal general election finds me comparing the May 2025 general election with that occurring in August 1943.
John Curtin was Labor’s leader and Leader of the Opposition from October 1935 to October 1941 during which period he was able to raise Labor’s vote significantly over two general elections. Effectively, he won the September 1940 election but the circumstances of the day meant he had to wait until October 1941 to take office as prime minister.
I was a four-year-old toddler in August 1943, which means I have no memory of that great event. However, I have read enough history to know that there was great uncertainty during the campaign as to what the result might be. My parents were voters at the time and they confirmed to me the memory of great uncertainty among pundits.
History, therefore, now records a similarity between the Australian federal general elections of August 1943 and May 2025. In each case, there was a Labor prime minister, who had been in the office for a couple of years but was not seen to be in a strong position. Each result was an unexpected landslide to Labor. The main difference was that Curtin died in July 1945, so Ben Chifley took Labor to the 1946 election which he won - before losing in 1949 to Bob Menzies.
\n time, we’ll learn what the fate of Anthony Albanese might be.
