Hare-Clark is basically unique amongst Australian electoral systems because there is a realistic possibility that a member of parliament will lose their seat to a fellow member of the same party – indeed it happens fairly regularly. Another unusual feature is the use of countbacks. Members of parliament often choose to quit mid-term, and are replaced by a fellow member of their party without the need for a by-election. But those mid-term replacements often struggle to retain their seats against other candidates at the following election.
For this post I am looking at the statistics on how MPs leave parliament, and how mid-term replacements perform at the next election.
For this first chart, let’s look at the cause of departure for MPs elected at one election, but not at the following election. For this one, we are not considering how mid-term replacements perform, just those elected for a full term.
1986 was an extreme example, when 13 full-term sitting MPs were defeated. This was on top of four MPs quitting mid-term and two others retiring at the election. This means that less than half of MPs elected in 1983 made it to the next parliament.
No other election has produced such a high rate of MPs losing their seats – not even in 1998, when ten seats were abolished. On average, 3.7 MPs have lost their seats at each election since 1989.
Something noticeable in the more recent data is the rare cases of MPs retiring at the election. Michelle O’Byrne is the only MP to retire at the 2025, but no MP retired at the 2021 or 2024 election.
Mid-term retirements have been much more common in recent years – five MPs were replaced via countback during the 2021-24 term, but in this term there has only been one – Rebecca White, who quit to run successfully for the federal seat of Lyons. This probably reflects the short length of the term. It’s likely others would have retired before a full-term election in 2028.
Over the last twelve elections, there have been 41 MPs who had filled countbacks. All but one of those MPs then contested the next election. So how did they do?
Just over half of these MPs (24 out of 41) were re-elected.
Quite a few have been defeated – nine have lost to their own party, seven to another party. That latter category includes Lara Alexander, whose seat returned to her old Liberal Party after she became an independent.
The rate of casual vacancy MPs to be re-elected was substantially lower than for full-term MPs – 58.5% compared to 76.9%. And this is despite substantially more full-term MPs choosing to retire. Amongst those who do lose their seats, full-term MPs and those who have filled a casual vacancy are about as likely to lose their seats to a member of their own party or to someone else. Just over half of these defeats came at the hands of their own party, which brings me to the next chart.
In the 7-seat era up to 1996, same-party defeats were more common than other-party defeats. But since 2014 in particular same-party defeats have become much less common. There was just one in 2024 – a particularly strange one where an MP who had filled a casual vacancy (Dean Young) was defeated by the very MP who had vacated his seat previously (Jacquie Petrusma).
This final chart distils this data into three rates: the proportion of seats that were a full-term MP being re-elected, or any MP being re-elected, and the proportion that went to the same party. I’ve excluded the ten newly-created seats in 2024.
There seems to have been an uptick in the proportion of seats being retained by sitting MPs in recent elections, but this includes 2024 when there was understandably a high proportion of MPs being re-elected with the Assembly growing by ten seats.
It will be interesting to see how these trends continue in 2025: the major parties are both polling at low levels, and the seven-seat electorates mean there is more room for change inside and between the parties.
The DemosAU poll published on Monday gives us some hints of potential for MPs to lose their seats to their party colleagues. Specifically:
- In Bass, former federal MP Bridget Archer is way out in front of her Liberal colleagues, potentially putting the seat of Simon Wood or Rob Fairs at risk if the party again wins three seats (as the poll predicts).
- In Braddon, retired federal MP Gavin Pearce is ranked second amongst the Liberal candidates, ahead of incumbents Felix Ellis and Roger Jaensch. While the party has some chance of winning four seats, if they fall short it is likely one of those two would lose.
- In Clark, Liberal candidate Marcus Vermey is ahead of both incumbents Ogilvie and Behrakis. The Liberal Party is a long way from winning a third seat so this could be another case. If there was no polling, you’d also wonder about Labor’s Luke Martin, but the poll suggests he is a long way behind his party’s incumbents.
- In Franklin, the least-prominent incumbent on both sides is being slightly outpolled by a non-incumbent. Josh Garvin is the third-polling Liberal, just ahead of Nic Street, while prominent unionist Jessica Munday is ahead of Labor incumbent Meg Brown. These gaps are not wide, though.
- In Lyons, new Labor MP Casey Farrell, who replaced Rebecca White earlier this year, is trailing White’s federal predecessor Brian Mitchell. This would not be surprising. The DemosAU poll also suggests that ex-MP John Tucker is the more likely candidate to win a seat for the Nationals, rather than incumbent MP Andrew Jenner. So arguably that would also be an intra-party change.
That is seven plausible cases where sitting MPs could lose their seats to their colleagues.
I think part of the story here is that both major parties are a long way away from a majority, which has motivated them both to seek out big names in the hope they can win extra seats. Three MPs who were sitting in the federal parliament just three months ago are all favourites to win seats off their intra-party colleagues. But with the vote for the major parties stagnant, those big names are more likely to take seats off their colleagues.
