Renata opens her Yarra Valley Water bill in July 2026 the way most Melburnians now do: bracing before she reads it. Beneath the usual creeping rise in water and sewerage charges sits a newer line, the "desalination cost recovery" — her share of a decision the state government made three months earlier to order 150 billion litres of water from the Wonthaggi desalination plant, its full contracted capacity, for the first time since it opened. She has no idea whether a drop of desalinated water has ever reached her tap. She has, however, helped pay for the plant standing ready every year since 2012, whether Melbourne needed it or not. That is the plant's whole design, and it is why a facility built to insure against drought has become one of the most argued-over lines on a Victorian water bill.
What the government did
The Victorian Desalination Plant at Wonthaggi, on the Bass Coast, was announced by the Bracks Labor government in June 2007, in the depths of the Millennium Drought, when Melbourne's reservoirs had fallen toward levels not seen since records began and the city faced its tightest water restrictions in generations. His successor, the Brumby Labor government, awarded the design-build-operate contract to the AquaSure consortium on 30 June 2009 (Department of Treasury and Finance, Partnerships Victoria project records). Construction ran from October 2009 to December 2012, finishing under the Baillieu Coalition government elected in November 2010 — one of the few points in this story where responsibility genuinely crosses party lines.
The contract both sides of politics have now lived with for close to two decades is a public-private partnership running approximately 27 years from commissioning, to around 2039. AquaSure built and owns the plant; Melbourne Water buys its output and recovers the cost from the four Melbourne retail water businesses, which pass it to household and business bills through prices the Essential Services Commission (ESC) sets each year. Payments to AquaSure have two parts: a large fixed "water security" charge, payable in full whether the government orders a single litre or not, reported at approximately $608 million a year — about $1.8 million a day — when first widely disclosed, and indexed to inflation since; and a smaller variable charge scaled to the volume actually ordered. The plant can deliver up to 150 gigalitres (GL) a year, roughly one-third of Melbourne's typical annual water use, and every year the Minister for Water decides, by 1 April, how much of that capacity — if any — to order for the financial year ahead.
The scale of the commitment was flagged early. The Victorian Auditor-General's Office, reporting on the state's 2009-10 annual accounts (tabled 6 October 2010), estimated the plant would cost the state more than $18 billion over the following 30 years — around $2 billion more than the government's own published estimate at the time. Later reporting on the full contract, including its escalation clauses, has put the ceiling on total payments over the contract's life as high as approximately $23.9 billion, or roughly $860 million a year in years when the plant is ordered at capacity (approx.; figures widely reported from AquaSure contract disclosures and Auditor-General material) — numbers that make the annual order decision, however modest it looks beside the fixed charge, one of the more consequential calls a Victorian water minister makes each year.
The strongest case
A drought-proof water source is not a luxury in a drying climate
Take the government's justification for keeping, and now leaning on, this plant seriously, because the underlying risk is real and getting worse, not better. Every other major source of Melbourne's water depends on rain falling in the right catchments at the right time of year. Desalinated water does not — it is the one part of the water grid immune to drought, and the case for paying an insurance premium against that risk was built on hard experience. The Millennium Drought pushed Melbourne's storages down to levels that genuinely threatened supply, and the current decade has delivered an eerily similar warning: inflows into Melbourne's storages in the year to June 2025 were around 36% below the 30-year average — the lowest January-to-June inflow on record — and the 24 months to mid-2026 produced some of the driest conditions since the worst years of the Millennium Drought itself (Victorian Government, water.vic.gov.au, seasonal water outlook commentary, 2025-2026). By 8 July 2026, Melbourne's total storage stood at 65%, seven percentage points below the same date a year earlier, continuing an unbroken slide from above 91% in mid-2024. The Bureau of Meteorology's own outlook gave only a 20-40% chance of exceeding median rainfall across most of Victoria for the following three months.
Set against that trend, ordering the plant's full 150 GL capacity for 2026-27 is not obviously reckless. A government that waited until storages were critically low to place an order would be gambling with a supply source that, by contract, needs lead time to ramp up. Paying a standing charge every year, including years when it turns out not to have been strictly necessary, is close to the textbook definition of insurance: the premium is the same whether or not the insured event occurs that year, and building desalination capacity only after a drought has already bitten would be far more expensive, and far riskier, for a city of five million people.
The evidence
The record shows a plant whose use has tracked, roughly, an arc from caution to reliance. The first desalinated water did not reach customers until 2017, almost five years after construction finished; cumulative deliveries since then total approximately 505 GL, as at the 2025-26 order (Melbourne Water, Desalination data). Orders were modest through the late 2010s, climbed to around 100-125 GL in some drier years between 2019 and 2022 (approx.), then fell to around 15 GL for 2022-23 as three wetter-than-average years refilled the system. No order at all was placed for 2023-24 or 2024-25, and ministers said at the time that healthy storages meant households would not face an additional desal charge. That changed abruptly: a 50 GL order was announced on 4 April 2025 for 2025-26, as Melbourne's storages had fallen to 78.1% (from 91.5% a year earlier) and Geelong's had hit a six-year low of 42.2%. A year later, on 7 April 2026, the government ordered the plant's full 150 GL capacity for 2026-27 — the largest single order in the plant's history.
The bill impact is not hypothetical. The ESC's approved prices for 2026-27, effective 1 July 2026, lift the average Melbourne owner-occupier's annual water bill by around $73, or about 7%, with the regulator citing recovery of the latest desalination order's cost as part of that increase; regional Victorian bills, largely insulated from the desal charge, rose by a smaller approximate 5% on average. Melbourne Water was separately cleared to spend approximately $10 billion on network and supply infrastructure over the following five years — spending in which the desalination contract's fixed cost is a permanent, non-discretionary line, not a variable the utility can trim in a lean year.
The critique
None of the above requires calling the plant a mistake to ask a sharper question: does the machinery that decides how much water to order each year track genuine supply risk, or does it also track something closer to convenience? The 50 GL order for 2025-26 and the full-capacity order for 2026-27 both arrived as storages were visibly falling, and both can be defended on the evidence above. But the plant's own history includes years when orders were placed against a healthier backdrop, and the criteria the Minister for Water actually applies each 1 April — how far storages must fall, over what period, before an order is triggered — are not published as a clear, numerical rule the public or an independent body can check in advance. The government publishes the outcome of its deliberations, not the trigger behind them. That gap between a real risk and a discretionary, largely unreviewable annual decision is where public trust in the whole arrangement tends to erode fastest.
Melburnians have paid an insurance premium on this plant every year since 2012. The premium does not fall when it rains.
The fixed-charge structure compounds the problem rather than disciplining it. Because roughly $608 million a year (approx.) is owed to AquaSure regardless of the order, the marginal cost that actually shows up when the government adds gigalitres to an order is only ever a fraction of the plant's true annual cost to the state. That insulates the order decision from full fiscal scrutiny in a way ordinary spending is not: a minister weighing whether to order 50 GL or 150 GL is choosing a variable add-on to a sunk cost paid either way, not choosing whether to spend $608 million or save it. The Auditor-General flagged the scale of that fixed commitment as far back as October 2010, warning the contract would cost billions more than the government's own published figure. Successive governments — Labor at the outset, the Coalition through construction, and Labor continuously in office from November 2014 to today — have had fifteen years to renegotiate, restructure or better explain that contract to the public. None has done so in any way that shows up in a materially different payment structure.
The Essential Services Commission's role compounds the opacity rather than resolving it. The ESC sets the maximum prices retail water businesses can charge, and its determinations pass the desalination order's cost through to bills as a matter of regulatory mechanics; it does not, and is not designed to, independently second-guess whether the Minister for Water's order decision was right on the merits. That leaves no independent body routinely testing the trigger criteria against the outcome — a gap that would matter less if storages stayed comfortably high, and matters a great deal more now that both the frequency and the size of orders are rising together with genuine climate risk. When Victorian water bills already carry a widening gap between promised cost-of-living relief and what households actually pay, a charge that arrives without a transparent, checkable rule attached to it is a harder one for any government to defend, even when the underlying decision turns out to be sound.
The risk to the future
A drying climate meets a contract with no off switch
The trend lines in this topic point the same direction: storages down from above 91% to 65% in two years, inflows at Millennium Drought-era lows, and a desalination order that has gone from zero to full capacity in twelve months. If that trajectory continues — and Victoria's own water-outlook data give no confidence it will reverse quickly — Melburnians should expect more years like 2026-27, not fewer: full or near-full orders stacked on top of a fixed charge that was already, on the Auditor-General's 2010 numbers, one of the most expensive infrastructure commitments in the state's history. The insurance case for the plant gets stronger as the climate dries; the household bill case gets harder at the same time, because a rising desal charge lands on top of Melbourne Water's approved $10 billion, five-year investment programme and every other cost-of-living pressure already stacked on Victorian household budgets.
The deeper structural risk is that the contract runs to around 2039 on payment terms fixed in 2009, with no mechanism forcing a transparent, published trigger for the order decision that determines how much of that fixed cost gets a variable top-up each year. A government that keeps making defensible individual order decisions but never publishes the rule behind them is spending down a form of public trust it cannot easily rebuild once a single order looks political rather than hydrological — and trust is exactly what will be needed to secure support for whatever comes after 2039, when Victoria has to decide how it replaces or renegotiates the state's single largest piece of drought insurance.