Four gigawatts. That’s roughly the scheduled coal capacity due to leave the NEM between now and 2030 across Eraring, Yallourn and Loy Yang A, against a transmission queue that, on AEMO’s own numbers, is still waiting on several major Renewable Energy Zone links to reach financial close. Put those two facts next to each other and you get the question every state energy minister dodges at press conferences: are we closing coal faster than we can replace it?
My answer isn’t a clean yes or no. It’s worse than that. We’re closing it faster than we can replace it in some places, and slower than the market wants in others, and the gap between those two truths is where the reliability risk actually sits.
The capacity numbers everyone quotes #
Start with what’s leaving. Eraring, the biggest coal plant in the NEM, was meant to close in 2025. Origin Energy pushed that out to 2027 after direct negotiation with the NSW government, a deal that effectively admits the original closure date was premature against the replacement build. Yallourn in Victoria is scheduled to go in 2028. Loy Yang A has flagged an earlier-than-original exit too, though AGL has kept some flexibility in the date.
Against that, AEMO’s Integrated System Plan models roughly 10,000 kilometres of new transmission and tens of gigawatts of firmed renewable capacity needed by the mid-2030s to keep the lights on at the reliability standard regulators expect. The plan itself is sound engineering. The execution against it has not matched the paper pace, and that’s not a hot take, it’s visible in every quarterly AEMO reliability report since 2023.
Where the mismatch is real, not theoretical #
NSW is the clearest case. I’ve written before about how the state’s coal exit is outrunning its REZ rollout, and nothing in the last twelve months has closed that gap. Central-West Orana, the state’s first REZ, is still working through transmission approvals and landholder negotiations well behind the original schedule. New England REZ, which ACEN Australia and others are building toward, has firmed up some solar capacity but the transmission backbone that makes the whole zone work isn’t there yet. Meanwhile Eraring’s four units keep running because nothing else in the state is ready to pick up the load.
Victoria tells a slightly different story. The state has leaned harder into offshore wind ambition through the SEC revival, but offshore wind in Australia has a development timeline measured in a decade, not a budget cycle, and the shrinking of the Illawarra zone after BlueFloat’s reduced footprint is a reminder that paper gigawatts and permitted gigawatts are different animals. Yallourn’s 2028 date assumes Victoria’s REZ and battery build keeps pace. On current contracting rates, that’s an optimistic assumption, not a settled fact.
What the Capacity Investment Scheme was supposed to fix #
The Commonwealth’s Capacity Investment Scheme exists precisely to de-risk this mismatch, underwriting revenue for new renewable and storage capacity so it gets built ahead of, not behind, coal retirements. It’s a sensible mechanism and it has pulled forward some battery and wind contracts that might otherwise have sat on a developer’s spreadsheet for another funding round. But a scheme that underwrites revenue doesn’t underwrite transmission approvals, council planning pathways, or transformer lead times, and those are the actual bottlenecks right now. You can have all the contracted capacity in the world sitting behind an unbuilt line.
This is the bit that gets missed in the political debate. The argument usually runs “renewables versus coal”, as if it’s a straight swap. It isn’t. The constraint isn’t generation capacity, it’s the poles and wires connecting that capacity to load, plus the firming layer, batteries, pumped hydro, gas peakers, that smooths out a wind and solar fleet that doesn’t run on a schedule. Readers wanting the long version of who actually pays for that build-out should look at the funding mechanics in who really pays for the transmission build-out, because the dollars matter as much as the megawatts.
The reliability numbers AEMO actually publishes #
AEMO’s Electricity Statement of Opportunities, published twice a year, is the document to read if you want the honest picture rather than the political one. The most recent editions have flagged reliability gaps in NSW and Victoria within the outlook period under certain coal-retirement and demand scenarios, not catastrophic blackout warnings, but genuine amber flags that say the margin is thinner than planners would like. That’s a regulator’s document, not an activist’s press release, and it deserves more attention than it gets.
What the ESOO doesn’t say, because it isn’t AEMO’s job to say it, is whether the political timetable for coal closures should bend to the construction timetable for replacement capacity, or the other way round. That’s a judgement call, and I’ll make mine plainly: the construction timetable should set the pace, not the other way round. Every time a government negotiates a closure extension after the fact, as happened with Eraring, it’s effectively admitting that. The honest move would be to set closure dates conditional on REZ transmission milestones being met, not on a calendar year picked for political convenience.
The storage and firming layer isn’t keeping up either #
Batteries have had a strong few years in Australia, and projects like the Waratah Super Battery are finally earning their keep as grid shock absorbers rather than sitting idle. But batteries firm hours, not weeks. The deeper firming question, the one that actually determines whether a coal retirement is safe, is whether there’s enough dispatchable capacity for a prolonged low-wind, low-sun stretch in winter. That’s the argument behind Snowy 2.0, behind Tasmania’s Battery of the Nation push through Marinus Link, and behind every gas peaker argument that resurfaces whenever a cold snap hits Victoria. None of those big firming projects are moving at a pace that matches the coal closure calendar. Snowy 2.0 alone has blown past its original budget and timeline by years, and it’s still the centrepiece of the firming plan for the southern states.
Virtual power plants and home batteries are filling some of the gap at the margins, and the reality check on VPPs is worth reading if you want to know how much load they can genuinely shift rather than how much the marketing promises. But aggregated household batteries, however well orchestrated, are not a substitute for gigawatt-scale firming when a REZ transmission line is still in planning.
What a sensible policy response actually looks like #
I don’t think the answer is to abandon coal closures, and I don’t think it’s to pretend the current pace is fine. It’s to make the exit dates genuinely conditional, publicly, on independently verified transmission and firming milestones, rather than political ones set years in advance and then quietly renegotiated when reality intervenes. AEMO’s own Integrated System Plan already models this kind of staged, conditional pathway; what’s missing is a mechanism that forces governments and generator owners to actually follow it rather than treat it as a reference document. The Eraring renegotiation is proof the current approach works only by exception, case by case, under pressure, rather than by design.
There’s also a cricket-adjacent point here, in that nobody declares an innings before the runs are actually on the board. Announcing a closure date is the equivalent of declaring before you’ve checked the required run rate. NSW and Victoria have both effectively had to bat on past the declaration because the chase wasn’t ready.
So, are we closing coal faster than we can replace it? #
In NSW, yes, on the evidence of the REZ timelines and the Eraring renegotiation. In Victoria, it’s closer but the offshore wind ambition carries real execution risk that the 2028 Yallourn date doesn’t fully account for. Nationally, the Capacity Investment Scheme and the batteries coming online are closing the gap faster than critics give them credit for, but “faster than critics claim” and “fast enough” are different statements, and conflating them is where most of this debate goes wrong.
Governments like a firm closure date because it’s a clean political signal. Grids don’t care about signals. They care about megawatts, available, firmed, and connected, on the day the old plant actually switches off. Until the transmission queue clears faster than the coal fleet ages out, that gap is the story, and anyone telling you otherwise probably hasn’t read the ESOO.
– Marcus Wren, Editor
Photo by American Public Power Association on Unsplash