Inside Story

The turn of the electoral cycle could be a long time coming

Labor is riding high across Australia, and the Greens are doing better than most observers acknowledge. Where does that leave the Coalition?

Tim Colebatch 27 January 2023 3498 words

Best of enemies? NSW premier Dominic Perrottet (centre) with prime minister Anthony Albanese and federal MP Susan Templeman at a roads announcement in the Blue Mountains last week. Dan Himbrechts/AAP Image

The Coalition government in New South Wales is on track for a massive defeat, reports the YouGov poll in last weekend’s Sunday Telegraph. Just two months from the election, the poll found 56 per cent of NSW voters would prefer Labor to form the state’s next government.

That would be a stunning swing of 8 per cent against Dominic Perrottet’s government — a government one could argue is the most progressive and competent the Coalition has produced in the past decade, certainly relative to its counterparts federally, or in Victoria or Queensland.

Four days later, a Resolve poll for the Sydney Morning Herald came up with roughly similar findings, though on those figures the swing seems to be more like 6.5 per cent. It would still be curtains. The bookies give the Coalition’s last government on the mainland just a 20 per cent chance of holding onto office when the voters deliver judgement on 25 March.

If that proves right, it would be a dramatic illustration of how far the Liberal Party has fallen. At its peak in 2014, the Coalition ran every government in Australia except for South Australia (where it won the election on votes but lost it on the seats) and the ACT (where it hasn’t won an election since 1998).

The Abbott government’s 2014 budget of broken promises marked the turning point. Since then the Coalition has lost one government after another.

Three of those governments fell after a single term. Six months after Abbott’s budget, Victoria led the way, following four years dominated by the Coalition infighting that brought down premier Ted Baillieu. Queensland followed two months later, with premier Campbell Newman, the hero of its 2012 election win, losing his own seat in a 14 per cent statewide swing.

The polls suggested Abbott’s own government would also be booted out after a single term, but Liberal MPs forestalled that by dumping him and installing Malcolm Turnbull to win them the 2016 election.

Coalition governments continued to fall: Northern Territory voters dumped the Country Liberal Party government in 2016, leaving it with just two seats; the following year Mark McGowan began his remarkable reign by ousting the Coalition government in Western Australia.

In South Australia, by contrast, the Liberals finally broke through after sixteen years in opposition to win power in 2018. But that government too lasted just one term, and Labor emphatically returned to power last March — two months before voters finally sent the federal Coalition on its way.

But the NSW Coalition government should surely be less vulnerable. While the Morrison government showed no appetite for tackling climate change, NSW treasurer and energy minister Matt Kean has been a national leader in pushing the pace of decarbonisation. No government in Australia is more subject to integrity watchdogs than the NSW government is to its own ICAC. And, with some exceptions (water, for instance), it has been more a reformist government than a reactionary one. On gambling reform, it is far more progressive than Labor.

But this election cycle has become merciless for the Coalition parties. They began it with seven of Australia’s nine federal, state and territory governments. They now have just two, and if the NSW government falls on 25 March, they will be left with just one.

It’s worth recalling that latest cycle, and what it means for the Coalition parties. South Australia’s Liberal government fell in March last year, the federal Coalition government was voted out in May, and the Victorian Liberals received another thrashing from voters in November. In 2021, the Western Australian Liberals were left with just two seats in the Legislative Assembly and the Nationals replaced them as the official opposition.

2020 saw Queensland’s Labor government re-elected with an increased majority, leaving the Liberals with just five seats in Brisbane. The Labor–Greens government was given a sixth term in the ACT, while in the Northern Territory Labor was re-elected for a second term.

The one victory for the Coalition in that time was in Tasmania, where former premier Peter Gutwein called an early election in 2021 and narrowly retained his majority. If Dominic Perrottet loses in March, Gutwein’s successor, Jeremy Rockliff, will lead Australia’s only Liberal government. He doesn’t have to face the voters until early 2025.

Being out of government makes everything harder for political parties. They have far fewer staffers to call on, few public servants to provide policy assistance, less fundraising power and a bigger challenge finding things out. Talented people don’t waste their energies on a party destined for opposition, so the opposition parties are left with rusted-on supporters and branch stackers’ recruits.

But cycles change. In the last years of the Howard government, Labor controlled every state and territory government in Australia. Then, with the Rudd government still at the peak of its popularity, Western Australians voted out their Labor government. Two years later Victorians did the same, and by early 2014 Labor had only two governments left.

The next political cycle began with the Andrews government taking power in Victoria in November 2014. If the Perrottet government is voted out, that cycle will culminate on 25 March.

But would that also imply that Labor is now at the top of its cycle? That it will be all downhill for Labor from here, and the Coalition will soon start taking back the governments it has lost?

Not necessarily. A looming Coalition revival would be a reasonable expectation if you assumed that its losses of government reflected cyclical factors alone: bad luck, changing political fashions, one-off factors, and in the NSW case, the wear and tear of twelve years in government.

But if that were so, we should be seeing signs of it in the state that was first to switch its allegiance to Labor. And those signs should have been apparent in November’s election there. As we shall see, though, there was little if any sign of a Liberal comeback.

It’s early days yet for the Albanese government. We’ve seen many governments enjoy an extended honeymoon before crashing — the Rudd government, for example — but eight months after the election, all the polls suggest it’s the Coalition that has lost a lot of ground.

Labor won the 21 May election with a tad over 52 per cent of the two-party vote. The latest Resolve poll in Tuesday’s Age/Sydney Morning Herald implied that the figure has jumped to 60 per cent since the election. Peter Dutton’s poll ratings remain far behind Anthony Albanese’s.

In this Labor cycle, the Coalition lost its first state government just eight months after reaching its peak of running seven of the nine governments. But the Coalition’s next cycle might not begin anytime soon.

After New South Wales votes it will be almost a year and a half before the next election anywhere in Australia, and then there will be a bunch of them: the Northern Territory, the ACT and Queensland in the second half of 2024, and then Western Australia, Tasmania and federally in the first half of 2025. While you’d think the Coalition will surely win one of them, it could face a long wait before its own cycle gets going.


First, since 2000, the Coalition has mostly been in government in Canberra, but Labor has dominated government at state and territory level. So far this century, in the states and territories, Labor has been in power for roughly three-quarters of the time, and the Coalition for just one-quarter. Outside New South Wales and Tasmania, Labor has become the natural party of government, and the Liberals/Nationals/LNP/CLP the natural parties of opposition.

In Victoria, the Coalition has been in power for just four of the past twenty-three years. In Queensland, three. In South Australia, six. In the Northern Territory, five and a half, and in the ACT, the Liberals have had less than two years in office and twenty-one in opposition.

Maybe that’s logical. The federal government’s main jobs are economic management, social security and foreign relations — and on two of those, voters have a conservative bias. The states, by contrast, are mainly responsible for running the services we rely on: hospitals, schools, roads and public transport. On those issues, voters have a bias towards improving services. Keeping the budget in the black seems to have gone out of fashion.

But former NSW premier and federal Liberal president Nick Greiner put it well when he said the Liberals’ problem is that its activist base is well to the right of its voter base. (Just as the Greens’ activist base is well to the left of its voter base.) And the longer a party is out of power, the more deep-seated that becomes. Failure breeds more failure.

Take the ACT Liberals. In its first twelve years of self-government the ACT mostly had Liberal governments. Red Hill pharmacist Kate Carnell was parachuted in to become party leader. She won two elections, presenting a moderate, modern face that Canberra voters could relate to. But in 2001, the Libs lost office, went into opposition, and stayed there.

By 2023, it’s come to feel as though the Labor–Greens coalition is the permanent government of the ACT, and the Liberals are the permanent opposition.

Could the same be happening in Victoria? There, the Coalition parties have won a majority of Victoria’s seats in just one of the last sixteen elections, federal or state. For a Liberal Party that once governed the state alone for twenty-seven years straight, defeat has come to be seen as normal.

The Coalition’s one victory was at the 2010 state election under another moderate, modern Liberal, Ted Baillieu. But in truth they were not ready to govern, were profoundly disunited and had no coherent agenda. When voters turfed them out in 2014, there was little to show for their four years in power.

Love him or loathe him, Daniel Andrews has been full of ambition to do things, above all investing in transport infrastructure and tackling cutting-edge social issues. In 2018 Labor was returned with a massive majority — so large that we all assumed 2022 would see a swing back to the Coalition.

That assumption seemed certain after the government’s mishandling of Covid. It imposed harsh, unpopular lockdowns that effectively closed down Melbourne for eight months, then changed course 180 degrees and dropped all controls as the death toll from Covid soared. More than one in 1000 Victorians have now died of the disease — 76 per cent more than in the rest of Australia.

Surely the 2022 Victorian election should have been the start of a new cycle of Liberal renewal. But no. At face value, the Libs made no progress at all. Labor had fifty-five seats in the old Assembly, the Coalition twenty-seven, and the Greens and independents three each. In the new one, Labor has fifty-six, the Coalition twenty-eight (assuming it retains Narracan at Saturday’s delayed election), and the Greens four. Labor won easily — and while the Nationals gained three seats, the Liberals went backwards.

But there are nuances to that narrative — and in politics, nuances matter. One reason why the Albanese government is riding so high is that it has the freedom of having a clear majority in the House of Representatives. But that majority came so close to being a minority.

If fewer than 500 votes in the relevant booths had changed hands, the Liberals’ Andrew Constance would have won the federal seat of Gilmore from Labor, the Greens’ Steph Hodgins-May would now be the member for Macnamara — and Labor would be a minority government, negotiating every bill with the crossbenches or the Coalition. Small changes, big difference.

For Victoria, nuance one: The pre-election redistribution of boundaries, on the Victorian Electoral Commission’s estimates, gave Labor a net gain of three seats (because Melbourne had grown rapidly, and Labor held most of the seats with swollen enrolments). That was the real starting point: so Labor lost two seats, and the Coalition gained two. Not much difference, sure, but there was some movement the Coalition’s way, however inadequate.

Nuance two: As in the federal election, Labor was lucky. A shift of just 350 votes in the relevant booths would have seen the Greens take Northcote and the Liberals win Pakenham and Bass. Instead of the seats going from 58–26–4 to 56–28–4, they would have ended up as 53–30–5. Not much difference there either, but it would have shown clear movement from Labor to the Coalition.

Nuance three: On the votes, there certainly was movement. For the eighty-six seats contested by both sides at the last two elections, Labor’s share of the two-party-preferred vote fell from 57.5 per cent in 2018 to 54.8 per cent this time. That’s a swing of 2.7 per cent to the Coalition.

But in the twenty-five marginal seats — those where the election is decided — it was a very different story. The average swing from Labor to the Coalition there was just 0.3 per cent. The Coalition won four seats that were notionally Labor’s on the new boundaries (Caulfield, Hawthorn, Nepean and Morwell), while Labor won three seats that were notionally the Coalition’s (Bayswater, Bass and Glen Waverley).

(Labor also lost Richmond to the Greens and the Coalition won Shepparton from independent Suzanna Sheed. That’s why, in net terms, Labor lost two seats, and the Coalition gained two).

Mostly, the big swings to the Coalition were in safe Labor seats in Melbourne’s western and northern suburbs, where it had no chance of winning: seats like Greenvale (15.1 per cent swing to the Liberals), Mill Park (13.5), St Albans (12.4), Thomastown (11.4) and Kororoit (11.1).

Across the nineteen middle and outer suburbs northwest of the Yarra, the average swing to the Coalition was 7.1 per cent. But it won none of them, and even after those swings, only two of them were even mildly marginal.

To win a majority in the Assembly, the Coalition needed to make a net gain of nineteen seats. On the traditional pendulum, admittedly a rough measure, that required a swing of 10.4 per cent.

At the 2026 election, to win a majority, it will need to make a net gain of seventeen seats. On the pendulum, that would require a swing of 8.1 per cent. No opposition has gained a swing of that size since 1955, when Labor split in two.

The start of a cycle that will see Australia swing back to Coalition governments? I think not.

At this election, twelve Labor seats were marginal against the Coalition. At the 2026 election, only eight will be. Most of the seats it needs to gain to win office will need to be won by swings of 6 to 10 per cent.

These are normally classified as “fairly safe” seats. If Labor is still riding high with voters in 2026, they won’t be threatened. But if it loses support in the next four years, as governments often do in their third terms, some could come under threat. Labor would have far more territory to defend.

Nuance four: A significant development was lost in the focus on Labor’s triumph and the Liberals’ woes. This election saw the Greens almost double their territory. Richmond was the only seat they picked up, but they came close to winning five other seats: Northcote (lost by 0.2 per cent), Pascoe Vale (2.0), Preston (2.1), Footscray (4.2) and Albert Park (4.5).

One in ten seats in the Assembly, all in the inner suburbs, are now either held by the Greens or within their reach. That’s a position they’ve never been in before, anywhere in Australia.

In the Legislative Council, they now hold the balance of power, winning four seats, up from one in 2018. Labor will have to negotiate with them on any legislation the Coalition opposes.

Their first-preference vote rose just slightly, from 10.7 per cent to 11.5. But there’s a simple reason for that which media commentators ignored: at this election there were three times as many minor-party candidates as last time. Animal Justice and Family First ran in every seat, and the Freedom Party in most of them. Those candidates took votes from all other parties. As the table shows, the minor parties’ vote more than doubled, from 5.2 per cent last time to 11.7 per cent.

Note: while the independents’ vote fell by 0.6 per cent, the Liberals by 0.8 per cent and Labor’s by 5.8 per cent, the Greens’ vote rose by 0.8 per cent. They clearly did the best of the major players.

We can compare apples with apples by looking at the vote for the three main parties after minor-party preferences. The comparison is limited by the fact that only in twenty-seven seats were Labor, Liberals and Greens in the final three in both 2018 and 2022 and the commission distributed everyone else’s preferences between them. (The VEC has promised that this time it will eventually distribute all preferences in all seats. Thank you, at last.)

When just Labor, Liberals and Greens were left in those twenty-seven seats, the Liberal vote was unchanged from 2018. But Labor’s vote was down 3.2 per cent, while the Greens’ vote rose by 3.2 per cent. In the new marginal seats of Footscray, Pascoe Vale and Preston, the swing from Labor to Greens was three times that size.

But isn’t that just because the Liberals changed their preferences from Labor to Greens? No. These are three-party figures: the Liberal preferences hadn’t been distributed at this stage. And note: when they were, they didn’t change the outcome in a single seat, not even Richmond. As the VEC has repeatedly shown, most inner-suburban Liberals don’t follow their party’s how-to-vote card.

Where Liberal preferences were distributed, they lifted the Greens’ two-party vote by an average of 4.5 per cent. The Greens were slowly expanding their territory even when the Liberals preferenced Labor, but in 2026 that 4.5 per cent would make it far easier for them to keep doing so. And conversely, Labor would breathe a huge sigh of relief if the Liberals preferenced them instead.

In war and politics, they say, the enemy of my enemy is my friend. Helping the Greens win Labor seats brings the Coalition no closer to government, as the Greens would always support Labor. But making Labor fight on two fronts dilutes its campaign resources: and of Labor’s thirteen most marginal seats after this election, in five its opponent is the Greens. If the Liberals redirect preferences to Labor in 2026, the only one Labor need worry about would be Northcote.

The Greens often declare war on themselves, and do or say silly things that discourage you from taking them seriously. But they have stood up for principles that matter to many Australians — tackling climate change seriously, treating refugees humanely, ending persecution of whistleblowers — which is why they have become the first party since Federation to establish itself as a lasting independent alternative to the two major parties.

The Country Party, now the Nationals, began as an independent group but abandoned that to form what has been a permanent coalition with the Liberals. In the last term of the NSW Coalition government, their one public stand on principle seemed to be to defend the interests of land developers against koalas.

The DLP and the Australian Democrats each lasted a generation and won significant support as independent alternatives to the two majors. But neither established a territorial base that could win them lower house seats, and eventually they withered away.

The Greens have carved out their own territory by making themselves the party of young voters in the inner suburbs. In those nine seats in inner Melbourne, they won 38.7 per cent of the three-party vote this time, ahead of Labor with 38.4 per cent and the Liberals with 22.9 per cent. At the federal election, they also won all three seats in inner Brisbane.

And the young, and the inner suburbs, just keep growing. It’s a good constituency to have.

New South Wales has become the Greens’ weakest state. At the federal election they polled just 10 per cent there, compared with 13.3 per cent in the rest of Australia. While they are now competitive in nine state seats in Melbourne, in 2019 they were competitive in just two seats in Sydney (Balmain and Newtown, both safe Green seats) and two in the hippie north of the state (winning Ballina, just losing Lismore). This state election will test them too.

For Victoria, the bottom line is that the Liberals have lost not only the 2022 election but probably the 2026 one as well. What happens then will be determined by the events of the next four years, and how voters respond to them. But things would have to go very badly for Labor for the Coalition to emerge with a majority.

For Australia, the bottom line is that if the Coalition loses what many see as its best government of recent years then it could be a long road back to power. And if the Greens’ MPs, senators and activists manage to avoid alienating their potential voters, it’s possible that the ACT will not be the only parliament in which the successor to a Labor government is a Labor–Greens one. •