Reserve a share of production and you exempt the plant that produces least
Australia's draft gas reservation scheme hangs the obligation on how much an exporter produces. The export plant accused of draining the east coast market is the one that produces least of its own gas.
Australia exports large volumes of gas as liquefied natural gas from plants in Queensland, while households and factories on the east coast buy from the same gas supply. The government has proposed a "reservation" rule: exporters would have to keep some gas back for Australian buyers. The draft says up to 20 per cent of what each exporter produces. Draft legislation is expected within weeks.
Tying the set-aside to how much gas a company produces lets off the export plant that produces least and buys the rest here — the very behaviour the scheme exists to curb. Measure net withdrawals instead.
The word doing the damage
Write a rule reserving a fifth of what a company produces, and the company that produces least owes least. That is not a loophole in the usual sense. Nobody has to hide anything, lobby for a carve-out or find a clever lawyer. It is simply what happens when you hang an obligation on the wrong variable.
The federal government’s draft scheme, released last month, would require liquefied natural gas exporters to set aside up to 20 per cent of their production for Australian buyers. The idea is to force more gas back into the east coast market and push prices down. Legislation is expected within weeks. The Australian Workers’ Union’s national secretary, Paul Farrow, says the draft now contains “Santos-sized loopholes”. The phrase is a union official’s, not a finding. The arithmetic underneath it is not.
Three export projects sit on Curtis Island in Queensland. According to the ABC’s reporting, two of them — Shell’s, and the Origin and ConocoPhillips venture — have overwhelmingly been net suppliers to the domestic market. The third, the Gladstone LNG plant operated by Santos, was built with 7.8 million tonnes a year of liquefaction capacity without anything like enough gas of its own to fill it, and an ABC investigation found it has exported the equivalent of 20 per cent of eastern Australia’s domestic demand over a decade.
None of that is misconduct. Building a plant larger than your own reserves and buying the balance on the market is a legitimate commercial structure, and processing other people’s gas is how shared infrastructure is meant to work. The point is narrower and more awkward: a percentage-of-production reservation is close to costless for a facility whose defining feature is that it does not produce enough gas.
Why a buyer with spare trains sets the price
This matters more than a fairness complaint between three ventures, because the buying is the price mechanism. An export plant with unfilled capacity and no gas of its own has to bid for molecules. It bids against manufacturers, gas retailers and power generators, and what it can afford to pay is set by the international price less the cost of chilling and shipping. That is the channel through which world prices become Australian prices. It is not a side effect of the export industry; it is the industry’s main point of contact with domestic customers.
A reservation scheme that measures production rather than net withdrawal aims at the wrong end of that transaction. It taxes the act of owning gas and ignores the act of competing for it. The producers who already sell into the east coast will carry administrative and contractual cost to deliver volumes they were largely delivering anyway. The buyer whose bidding sets the marginal price keeps bidding.
“Up to” is not a number
The draft also says up to 20 per cent. A ceiling is not a floor. It tells exporters the maximum of their exposure and tells domestic buyers nothing at all about the minimum of their supply. Every discretionary ceiling in a resources policy becomes a negotiation, conducted project by project, with reasons that may never be published. If the government intends the reserved share to vary with market conditions, it should say what triggers the variation and who decides. If it does not intend that, it should write a number.
The case for leaving it alone
The strongest defence of the production test is that it is administrable. Equity production is metered, audited and already reported. Net domestic position is harder: it requires netting purchases against sales across contracts of different tenor, some of them confidential, and it invites gaming through intermediaries and swaps. A capacity-based test would be worse still, penalising exactly the tolling and third-party processing arrangements that let small producers reach a market at all. And forcing a plant to reserve gas it does not own means forcing it to buy at market and resell below it — a levy on a midstream service, dressed as a supply measure.
That case is real, and it explains why officials reach for production. It does not survive contact with the objective. The scheme is not trying to allocate ownership fairly; it is trying to stop the domestic market being drained. If the chosen metric cannot see the draining, it has failed before the first ship loads. The netting problem is solvable — gas sales are already reported into regulatory systems, and a net position can be struck annually with a true-up, as tax law does with far messier concepts than molecules.
What would settle it
One number would end the argument, and nobody in government has published it: for each Curtis Island facility, the gas bought from the east coast market in a year, the gas sold into it, and the difference. If the Santos-led venture’s net figure is close to zero, the critics are wrong and the production test is adequate. If it is deeply negative while the other two are positive, then a reservation scheme built on production has been designed to miss its target, and the ministers signing it will have been told so before the Bill was introduced.
The government should define the reserved quantity in the Bill as a net domestic position — gas withdrawn from the east coast market less gas sold into it — rather than a share of equity production, and should replace "up to 20 per cent" with a stated minimum plus published criteria for any variation.
And before the vote, Treasury and the resources portfolio should publish, for each east coast LNG facility, its annual domestic purchases, domestic sales and net position for the past ten years. Parliament should not legislate a reservation scheme without knowing which facilities it will actually bind.
What would change this view
Publish the Bill with the obligation measured against something other than equity production — shipped LNG volumes, nameplate liquefaction capacity, or a net domestic withdrawal test — and my criticism of the instrument is wrong on its face. Equally, if the government publishes per-facility figures showing the Santos-led GLNG venture has been a net supplier to the east coast over the past decade, the premise of the complaint collapses.
Either outcome is observable within weeks: the exposure legislation is expected shortly, and the definition of the reserved quantity will be in the first dozen clauses. If it says "production", the design flaw is in the statute, not in the commentary about it.
Worked from
- Critics warn of 'Santos-sized loopholes' in gas reservation plan
ABC News · 2026-10-11