Two cables, 1500 megawatts combined, and a price tag that has roughly doubled since the project was first costed in earnest. That’s the state of Marinus Link in October 2026: still not under construction in the water, still central to Tasmania’s entire pitch as the “Battery of the Nation”, and still the subject of pointed questions from Victorian and federal budget officials about who actually pays for it.
I’ve spent a fair bit of this year going back through Marinus Link Pty Ltd’s project updates and the Tasmanian Networks filings, and the pattern is a familiar one for Australian transmission: the engineering case hasn’t really changed, but the cost curve has moved a lot faster than the political cycle around it.
What Marinus Link actually is, in MW and MWh terms #
Let’s be careful with that number, because “Marinus Link” gets used loosely to mean the cable, the Tasmanian pumped hydro behind it, and the whole Battery of the Nation concept as one lump. They’re not the same thing.
Marinus Link itself is a proposed high-voltage direct current interconnector running under Bass Strait, built in two stages of 750 megawatts each, for a combined 1500 MW of transfer capacity between Tasmania and Victoria’s grid at Heybridge and Hazelwood respectively. That’s power capacity – how fast electricity can move across the strait in either direction at any instant. It says nothing on its own about how much energy actually flows in a day, a week or a year, which depends on what’s generating in Tasmania (hydro storage levels, wind output) and what Victoria needs at the time.
The existing Basslink cable already does a version of this job, at roughly 500 MW one-way. Marinus is pitched as additional, firmer capacity, not a replacement, and the business case leans heavily on Tasmania’s existing hydro fleet acting like a giant battery: store energy when mainland prices are soft, discharge when the mainland is tight. That’s the pumped hydro and conventional hydro half of Battery of the Nation; Marinus is the pipe that makes it worth anything beyond the island.
The cost number everyone argues about #
Early public estimates for the full two-stage project sat in the $3.5 billion range. By the most recent project updates from Marinus Link Pty Ltd, the figure being discussed publicly for the combined project has moved up past $5 billion, with industry commentary suggesting it could run higher again depending on how Stage 2 timing and converter station costs land. Marinus Link Pty Ltd itself has been cautious about locking in a single all-up number publicly while contracts are still being finalised, which is itself telling.
None of that is unusual for subsea HVDC infrastructure right now. Global cable and converter supply chains are stretched, specialist cable-lay vessels are booked out years in advance, and every comparable project, from European interconnectors to our own Victoria-NSW VNI West build, has seen similar upward cost revisions. I’d argue the mistake isn’t that Marinus got more expensive; it’s that early public figures were presented with more confidence than the stage of engineering actually supported.
The AEMO Integrated System Plan has consistently kept Marinus Link in its “actionable” and preferred development path projects, which matters because it’s the document that underpins regulatory investment test outcomes. Readers who want the full planning logic behind projects like this one should look at what the AEMO Integrated System Plan actually says – Marinus sits inside that broader national transmission roadmap, not off to the side of it.
Staging: why Stage 1 and Stage 2 aren’t the same bet #
The project split into two 750 MW stages partly for cost management and partly because the Tasmanian generation and storage side, Battery of the Nation’s pumped hydro and wind additions, doesn’t need the full 1500 MW on day one. Stage 1 is the one with a firmer timeline attached; Stage 2’s final investment decision has effectively been pushed out, tied to how fast Tasmania’s own generation pipeline (new wind, the proposed pumped hydro sites) actually gets built and contracted.
That staging decision is the quiet admission in all of this. Building 1500 MW of cable capacity ahead of the generation that would use it is just capital sitting idle, earning nothing, while regulated revenue requirements still need to be recovered from someone. Splitting it lets the first cable prove the economics, and the construction methodology – before the second gets committed. It’s a sensible risk-management call. It also means the full Battery of the Nation vision, the one with Tasmania genuinely acting as a mainland-scale storage asset, is realistically a 2030s-plus story, not a late-2020s one.
| Element | Capacity / detail | Status (October 2026) |
|---|---|---|
| Stage 1 cable | 750 MW HVDC | Contracting and construction planning underway; still pre-construction in water |
| Stage 2 cable | 750 MW HVDC | FID deferred, tied to Tasmanian generation build-out |
| Existing Basslink | ~500 MW | Operating, separate asset |
| Combined project cost estimate | Public estimates now above $5 billion | Not finalised; contracts still being locked in |
Who actually pays, and the RIT-T argument #
Marinus Link runs through the regulatory investment test for transmission, the AER-administered framework meant to establish that a project delivers net market benefit before cost gets passed through to consumers via regulated revenue. That test has been a genuine battleground. Victorian consumer groups and some industry submissions have pushed back on cost allocation, arguing Tasmanian generators and Tasmanian consumers capture a disproportionate share of the benefit relative to what Victorian electricity users would pay.
The Commonwealth has underwritten a chunk of the equity through its investment in Marinus Link Pty Ltd, a structure designed partly to keep the project moving despite that allocation fight. It’s the same underlying tension that runs through who really pays for the transmission build-out more broadly across the NEM – whoever benefits from new interconnection rarely maps neatly onto whoever’s bill carries the regulated charge.
For a worked comparison: if Stage 1’s 750 MW ran at an average capacity factor of, say, 40 per cent over a year, a reasonable midpoint given it will sit idle during periods of low price differential between the states, that’s roughly 2.6 million MWh of energy transferred annually. Multiply that against a modest average wholesale price spread of $20-$30 per MWh captured in arbitrage, and you get a rough sense of the annual market value the cable needs to generate to justify itself beyond the reliability and emissions-reduction benefits that don’t show up on a trading desk’s ledger. It’s not a precise number – nobody has published a confirmed operating capacity factor yet because the thing isn’t finished, but it’s the order of magnitude the RIT-T modelling has to clear.
Where this sits against Snowy 2.0 and other delayed storage bets #
It’s hard to write about Marinus without the comparison to Snowy 2.0 coming up, and fair enough: both are long-lead, cost-escalating pieces of national storage infrastructure that got pitched on a simpler story than the one that’s actually unfolded. Readers who followed will Snowy 2.0 ever earn its keep will recognise the shape: a headline capacity number, years of tunnelling or cable-laying reality, and a cost curve that keeps climbing ahead of the commissioning date.
The difference with Marinus is that it’s an enabler, not a generator. Snowy 2.0 produces (eventually) its own dispatchable MW. Marinus doesn’t generate anything; it just lets Tasmania’s existing and future hydro and wind assets sell into a bigger market, and lets the mainland lean on Tasmanian storage when the wind drops over Victoria and South Australia. That makes its value harder to isolate in a RIT-T model, because it’s really valuing an option, the ability to move energy, rather than a fixed output stream.
The Battery of the Nation pipeline behind the cable #
None of the proposed Tasmanian pumped hydro projects under the Battery of the Nation banner, Lake Cethana chief among them, make commercial sense without Marinus actually built and carrying real volume. That’s the chicken-and-egg problem that’s dogged this project since it was first floated: generation investors want certainty the transmission will exist before committing capital, and transmission investors (including the Commonwealth, via its equity stake) want some signal that generation will actually use the capacity before locking in the full 1500 MW.
It’s a similar dynamic to what’s playing out with interconnectors on the mainland. The dispute over VNI West and its paddock-level route fight shows how even a project with strong ISP backing can stall on matters that have nothing to do with the underlying engineering case. Marinus has largely avoided a landholder dispute of that scale: the cable runs mostly underwater, but it hasn’t avoided the cost and sequencing version of the same problem.
Where the honest money sits #
My own read, for what it’s worth: Stage 1 gets built. The federal equity commitment and the ISP’s actionable-project status make a full cancellation politically unlikely at this point, and too much has already been spent on easements, environmental approvals and early cable procurement to walk away clean. Stage 2 is the genuine open question, and I think it slips further right than Marinus Link Pty Ltd’s current public materials suggest, simply because the Tasmanian generation pipeline that would use that second 750 MW hasn’t moved fast enough to justify an early commitment.
I’ll admit I was more sceptical about Stage 1 timing a year or two back than the project has since proven out – contracting has moved, slowly but visibly, since then. It’s the kind of infrastructure project that rewards patience over hot takes, which is roughly the same lesson I keep relearning on long training rides: the climb looks worse from the bottom than it turns out to be once you’re actually on it, though it’s never as easy as the strava segment makes it look either.
The AER’s published RIT-T assessments and AEMO’s ISP updates remain the places to watch for the next real signal, rather than press statements timed to a budget cycle. Marinus Link’s own project reporting, lodged with the Clean Energy Regulator and referenced in Commonwealth budget papers, will eventually have to settle on one cost number instead of a range. When it does, that’s the figure worth holding the project to.
– Anjali Rao, Grid & Storage Correspondent
Photo by American Public Power Association on Unsplash