Thirty-two projects. That is roughly how many contracts the federal government had underwritten through the Capacity Investment Scheme by the time the scheme’s first major tranches closed. The dollar exposure runs well past $10 billion in revenue guarantees when you count both the capacity and variable renewable energy components. That is not a subsidy in the traditional sense — it is a government balance sheet standing behind private investment decisions. And that distinction matters, because it changes who actually carries the risk when a project underperforms.
The CIS was designed — you can read the full architecture in our Capacity Investment Scheme explainer — as a technology-neutral underwriting mechanism. The pitch was elegant: instead of picking specific projects through a grants programme, the government would run competitive tender rounds and let the market decide which technologies won. Lower-cost, lower-risk projects would offer tighter revenue floors and ceilings. The government backstops the market, not the megawatt-hour.
Honest pitch. The question I keep coming back to is whether the execution is matching it.
What the tender results actually show #
The pattern across the CIS tenders held so far is not random. Utility-scale solar and wind have dominated the awarded capacity, which few people would find surprising given where the levelised cost of energy numbers sit. Battery storage, particularly grid-scale systems, has also done well. What has not fared as well is dispatchable firming capacity that sits outside the big battery format — pumped hydro at the smaller end, long-duration storage technologies, demand response aggregators.
Pumped hydro projects, for all the political attention they attract, face a genuine structural problem in a short-duration tender: their capital payback is long, their revenue profile is lumpy, and a two-hour battery can undercut them on price over a standard contract term. If you design a tender around uniform product specifications, you will get uniform technology outcomes. That is not neutral. That is a format effect masquerading as market competition.
The honest read is that the CIS, as tendered, is tilted toward technologies with short build times and liquid supply chains. Which, in 2026, mostly means solar-plus-storage at scale, and onshore wind in consented corridors. There is nothing wrong with those technologies. But calling the process technology-neutral whilst the tender design implicitly advantages them is a stretch.
Follow the money: who is actually winning contracts #
Follow the money and a handful of names appear repeatedly. The larger integrated energy players — companies with existing grid connections, balance sheets capable of posting bond requirements, and procurement teams that can price a 15-year revenue contract with the requisite sophistication — are structurally better placed to win CIS rounds than newer entrants or smaller developers. This is not a conspiracy. It is procurement design.
Andrew Forrest’s Squadron Energy has been one of the more visible participants in the renewables tender space; we covered their positioning in some detail in our Squadron Energy assessment. But Squadron is hardly alone. The tier-one developers — those with existing portfolios of consented sites, established engineering-procurement-construction relationships, and ready access to capital markets — are the ones writing the tightest bids. And tight bids win CIS contracts.
Smaller developers, community energy projects, and genuinely novel technologies are not excluded on paper. But a process that requires a bidder to absorb up to two years of pre-bid development cost before knowing if they won, in a market where grid connection queues at AEMO can extend past three years, is a process that selects for incumbents. The barriers are not legislative. They are financial and logistical, which makes them harder to argue with and harder to fix.
The firming gap nobody wants to talk about #
Here is where I think the consensus is getting it wrong. The policy debate around the CIS tends to focus on renewable energy capacity — gigawatts of wind and solar added to the NEM — as the primary measure of success. That framing misses the actual problem the scheme is supposed to solve.
The NEM does not have a renewable energy shortage. Rooftop solar alone — and the numbers behind Australia’s rooftop solar build are genuinely extraordinary — has already transformed the daytime supply picture. What the grid needs is firming: dispatchable capacity that can run when the sun is not shining and the wind is not blowing, for hours or days at a stretch. That is a different product from a lunchtime solar farm, and the CIS tender design, in my view, has not fully grappled with the distinction.
Short-duration batteries — two to four hours — are useful for evening peak management and frequency response. They are not useful for a three-day wind drought in South Australia in July. The difference between those two products is enormous, and a tender that treats them equivalently is not being straight with itself about what the grid actually needs. AEMO’s own Electricity Statement of Opportunities has flagged the dispatchable capacity gap repeatedly; the CIS should be the answer to that specific problem, not a vehicle for adding more variable generation to a system that already has plenty of it at the wrong times.
If you want a deeper look at how the firming question plays out technically, the pumped hydro versus big batteries comparison we ran earlier this year is worth revisiting. The short version: both have roles, but neither is a complete answer, and the CIS is not currently doing enough to pull longer-duration solutions through.
The state-by-state wrinkle #
The CIS operates as a federal scheme layered on top of state-level policy environments that vary significantly. Victoria has its own renewable energy targets and its own offtake contract programmes. New South Wales has the Electricity Infrastructure Roadmap, which runs its own tender processes. Queensland has had its government-owned generators as implicit policy instruments for years.
The result, honestly, is a mess of overlapping incentive structures. A developer in a state with strong complementary policy support can effectively double-dip — securing both state and federal revenue support for the same project, which improves their economics and, critically, allows them to submit a lower CIS bid. That is rational developer behaviour. It is also an outcome that a genuinely neutral national tender should probably account for, and currently does not.
The AEMC and the AER have both pointed toward the need for better coordination between jurisdictional and federal policy mechanisms. Progress on that coordination has been, to be polite about it, incremental. Anyone watching the Default Market Offer debates over the past few years will recognise the pattern: the policy architecture is fragmented, the coordination is slow, and meanwhile the tenders keep running.
Is this actually a problem, or just how markets work? #
Fair question. Incumbents win competitive tenders in most industries. Large balance sheets beat small ones. Experienced bidders beat first-timers. You could argue the CIS is doing exactly what it was designed to do — underwriting the least-cost clean capacity at lowest fiscal risk to government — and that the pattern of winners reflects genuine market efficiency rather than structural bias.
I’d accept that argument more readily if the scheme were actually delivering the firming capacity the NEM needs, rather than primarily adding to the variable renewable stack. It is not that the solar and wind capacity is unwelcome. It is that the grid’s constraint is no longer primarily about renewable generation volume. The constraint is the ability to dispatch power reliably when variable generation falls away. On that specific question, the CIS scorecard is thinner than the government’s announcements suggest.
The Coalition’s own alternative — whatever you make of the nuclear argument, and the cost and timeline questions there are substantial — at least starts from the premise that the firming gap is the central problem. Whether their proposed solution is credible is a separate debate. But they are at least asking the right question. The CIS, as currently structured, is not.
What a genuinely neutral scheme would look like #
A few things would help. Product differentiation in tenders: separate auctions for dispatchable capacity with minimum duration requirements, rather than blending everything into a single capacity payment. De-risking mechanisms specifically for longer-duration technologies — not grants, but first-loss guarantees or subordinated debt facilities through the CEFC that reduce the financing cost penalty on capital-intensive projects with longer payback profiles. And genuine transparency on who is winning, what technology they are building, and how the contracted capacity maps against AEMO’s published reliability projections.
ARENA has done useful work on technology cost transparency; their published assessments of emerging clean energy technologies are worth more than they get credit for. The same rigour applied to CIS tender outcomes — published, reconciled against need, auditable — would at least let the market, and the public, assess whether the scheme is actually solving the problem it was designed for.
Right now, the CIS is running fast and signing contracts. Whether those contracts are the right contracts, for the right technologies, in the right proportions — that audit has not really happened. And the longer the scheme runs without it, the more locked-in the current pattern becomes. By the time anyone officially questions whether the process was genuinely technology-neutral, the portfolio will already be built.
That is a slow pitch down leg side that the policy debate has not yet taken a swing at. Someone should.
— Marcus Wren, Editor
Photo by American Public Power Association on Unsplash