Santos paid a dividend again this year off Cooper Basin and Moomba gas that most of the transition commentary insists is a stranded asset waiting to happen. Origin and AGL are still burning gas peakers at a profit on cold mornings and hot nights when the wind drops out. And AEMO’s own Gas Statement of Opportunities keeps flagging a supply gap opening up in the southern states before the end of the decade, not because gas demand is collapsing but because the fields that feed Victoria and South Australia are running down faster than anything new is coming on. None of that fits the story we’ve been telling ourselves about gas in the Australian grid. The honest read is that the industry has been written off in the seminar room a fair bit faster than it’s been written off on the balance sheet.
What “written off” actually means in this debate #
Nobody serious argues gas has a long-term future as the backbone of the National Electricity Market. That ship sailed years ago. Batteries do the two-hour job cheaper now, and the trajectory on renewable energy zone build-out, however messy, points one direction. But there’s a difference between gas losing the baseload argument and gas losing every argument, and a lot of the coverage this year has collapsed those two things into one.
The industry hasn’t been written off by the market. It’s been written off in the framing. Every panel I’ve sat through in the last twelve months treats gas peakers as a rounding error, a bridge fuel we’re already halfway across. Meanwhile the Capacity Investment Scheme keeps signing firming contracts that include gas alongside batteries and pumped hydro, and AEMO’s own reliability forecasts keep leaning on existing gas plant to get through the 2027 and 2028 summers. Follow the money and gas hasn’t gone anywhere near the exit.
The peaker problem nobody actually solved #
I wrote a piece last year comparing gas peakers against big batteries on the specific job of firming, and the conclusion held up: batteries win on speed and short-duration response, gas wins on duration and on the genuinely ugly nights when a high-pressure system parks over the east coast for four or five days straight and there’s no wind anywhere from Ceduna to Cairns. That’s not a hypothetical. It happened in winter 2025 and it’ll happen again. Batteries currently being built across the NEM are mostly sized for two to four hours of storage. That’s brilliant for evening peak. It’s no use at all for a multi-day wind drought, and nobody in the industry has quietly solved that problem, they’ve just stopped talking about it in polite company.
Gas peakers fill exactly that gap. Not glamorously, not cheaply, but reliably, and reliability is the only currency that matters when the lights are actually at risk of going out. My colleague’s rundown of gas peakers versus big batteries makes the duration argument in more technical depth than I’ve got room for here, but the short version is the two technologies aren’t really competing for the same job. They’re doing different jobs, and the industry consensus has spent two years pretending gas will simply be absorbed into the battery story once enough gigawatt-hours get built. That’s not what the dispatch data shows.
Follow the money: who’s still building #
Santos is still spending real capital on Cooper Basin gas and on the Narrabri project in New South Wales, alongside its carbon capture ambitions at Moomba, a project I covered in detail when the company laid out its east-coast supply strategy. That’s not the behaviour of a company that reads its own future as terminal. AGL still runs gas alongside its coal fleet as it works through its own exit strategy, and Alinta has been notably unsentimental about keeping gas and coal generation running while everyone else spruiks their renewables pipeline. Origin and AGL both continue to build or contract new gas peaking capacity under long-term firming arrangements, the same mechanism propping up a good chunk of the Capacity Investment Scheme’s contracted portfolio.
If gas were genuinely finished, none of that capital would be moving. Companies don’t sink money into fields, pipelines and turbine contracts because they enjoy the smell of methane. They do it because someone, usually a state government or AEMO’s own reliability standard, is paying them to keep the lights on when renewables and storage can’t. The Capacity Investment Scheme’s own contracting rounds have kept a lane open for gas peaking capacity precisely because the alternative, a reliability gap in southern states in the late 2020s, is politically unaffordable. I’ve gone through how that scheme actually allocates money in an earlier piece, and gas hasn’t been quietly dropped from the eligible list. It’s still there, still being contracted.
The east-coast supply question hasn’t gone away #
This is where the consensus gets genuinely lazy. The argument that gas is finished usually rests on electricity generation numbers, which are real and declining. It rarely engages with the separate and arguably bigger question of east-coast gas supply for industry, for household heating in Victoria, and for the LNG export contracts that keep the Queensland fields economic. That’s a different market with a different set of pressures, and it’s one where AEMO has been consistently blunt about a looming shortfall as southern basin fields deplete faster than new supply, whether from Narrabri, Beetaloo, or elsewhere, gets approved and built. I laid out the mechanics of that squeeze in an earlier look at the east-coast gas supply question, and the honest read is it’s got worse, not better, in the time since.
You can want gas out of the electricity system entirely and still need it for glass manufacturing in Victoria, for fertiliser production, for the industrial processes that don’t have a battery-shaped substitute yet. The write-off narrative tends to treat all gas demand as one undifferentiated blob destined for the same fate as coal power stations. It isn’t. Household and industrial gas demand is a genuinely separate policy problem, and one Canberra and the states have been slow-walking for years, largely because every proposed solution, more supply, price caps, domestic reservation, upsets somebody with a seat at the table.
Where the consensus has it wrong #
I reckon the mistake most transition commentary makes is treating gas and coal as the same story running on the same timeline. Coal is closing because the plants are old, expensive to maintain, and increasingly unreliable, a pattern my colleague covered thoroughly in looking at whether Australia is closing coal faster than it can replace it. Gas isn’t ageing out the same way. The infrastructure is younger, the plant is far more flexible, and unlike coal it can sit idle for weeks and start in minutes when the grid needs it. That’s a fundamentally different asset profile, and lumping the two together under one “fossil fuels are finished” banner misses the actual engineering argument for why gas peaking still has a job for at least another decade, probably longer, even in a grid running mostly on wind, solar and batteries.
It’s a bit like writing off a batter after a rough series without checking the conditions they were batting in. Gas has had some bad headlines. That’s not the same as a bad balance sheet.
None of which means gas gets a free pass. The Safeguard Mechanism is steadily tightening the screws on emissions-intensive gas operations, and I’ve written before about who actually ends up paying for that squeeze on heavy industry. Gas peakers running at low capacity factor as backup for renewables face a genuinely awkward economics problem too, they need to be paid to sit idle most of the year, and that’s exactly the argument behind the capacity contracts the industry has been lobbying hard for. The industry isn’t being unfairly persecuted. It’s being asked to prove its economics work in a grid that increasingly doesn’t need it most of the time. Some of it will. Some of it won’t.
What actually kills gas, eventually #
The thing that ends gas in Australian electricity generation isn’t a policy announcement or a divestment campaign. It’s long-duration storage and firm renewable generation getting cheap and reliable enough to cover the multi-day gaps that gas currently covers, plus a REZ transmission build that stops slipping its own deadlines. I’ve followed the New South Wales rollout closely and the timeline keeps moving right, which buys gas more runway than the consensus wants to admit. When pumped hydro and long-duration batteries can genuinely handle the week-long wind drought, not the four-hour evening peak, that’s the day gas peaking loses its last real argument. On the current build schedule, that’s not close. It might be the 2030s before it’s close.
So has the industry been written off too early? On generation, mostly yes, though not for much longer. On supply, the write-off never should have started, because nobody’s found the substitute yet. Worth asking whoever’s telling you gas is finished exactly which part of the gas industry they mean, and whether they’ve checked AEMO’s own numbers lately, or the AER’s, before they say it with quite so much confidence.
— Marcus Wren, Editor
Photo by Nikola Johnny Mirkovic on Unsplash