A long-term gas contract is a capital decision

Most manufacturing boards I know approve a new kiln or dryer through a capital paper that runs to dozens of pages. The gas or electricity contract that will feed it for the next five or ten years is usually signed by a procurement team under delegated authority and reported to the board, if at all, as a line in the cost forecast. Marrangal worked that way until 2022. I now think it is backwards. For an energy-intensive plant, the energy contract is often a larger financial commitment than the asset itself, and it decides whether the asset has a future.

The contract outlives the budget

Clay bricks are fired in kilns at temperatures above 1,000 degrees Celsius over many hours, and plasterboard is dried in long gas-fired dryers. Before 2022, gas made up about a fifth of the cash cost of our oldest kiln, at the Bendarra works north of Melbourne. The contract supplying our Victorian kilns expired in June 2023, after the price shock of 2022 and the emergency measures that followed it. The offers we received for a new three-year contract were close to double the old price. On those terms gas would have been nearly a third of that kiln's cash cost, and the kiln could no longer cover its costs at the volumes we expected in a downturn.

Nothing physical about the kiln had changed. Its refractory lining was due for renewal at a cost of A$31 million, and its burners were older than our most junior engineers. Read alone, the capital paper for relining looked reasonable. Read beside the gas contract that would have fed the relined kiln, it did not. We closed the kiln in September 2023 and moved its output to two newer kilns that use about 11 per cent less gas per thousand bricks.

That episode taught me that energy contracts set the remaining life of an asset more decisively than its engineering does. A ten-year gas contract signed for an ageing kiln is, in practice, a decision to keep that kiln running for ten years. A short contract leaves room to close, convert or replace it. Neither choice is wrong in itself, but both are strategic, and a delegated procurement limit is not the place to make them. The budget that approved the contract will have been forgotten long before the contract ends.

Price the connection queue

Electrification adds a second constraint that many boards have not yet absorbed. Replacing gas heat with electric heat in a dryer or a kiln requires a large grid connection, and in much of Australia a new industrial connection of tens of megawatts takes years to study, approve and build. The binding constraint is often not the technology. It is the queue.

We met this directly on the second line at our Brisbane plasterboard plant. A federal programme that co-funds industrial decarbonisation offered up to A$62 million towards a fully electric dryer, which would have needed a new connection of about 32 megawatts. Our network operator indicated four to five years. A hybrid design, with a heat-pump first stage that recovers heat from the dryer's exhaust and a gas-fired final stage, needed about 9 megawatts, which an upgrade of the existing connection could supply within eighteen months. We chose the hybrid and A$28 million of support. The line was commissioned in March 2026 and uses 38 per cent less gas per square metre than its older neighbour.

Our mistake was not the choice of design. It was failing to lodge an application for the larger connection at the same time. A connection application is an option. It costs a modest sum in studies and fees and preserves the ability to electrify fully later, when the technology has matured and the price of electricity is clearer. By not lodging it in 2024, we lost two years in the queue that no amount of money can now buy back. I would ask every energy-intensive manufacturer to keep a register of the connection each plant would need to electrify, with the queue time attached, and to acquire places in that queue as deliberately as it acquires land.

What the board should sign

I have three proposals for boards of energy-intensive manufacturers.

First, approve long-term energy contracts above a threshold in the same meeting and the same paper as the assets they feed, and state the remaining-life decision each contract implies. At Marrangal the board now approves every gas contract longer than three years or larger than A$20 million a year. Contracts covering about 70 per cent of our forecast kiln and dryer demand to 2030 have gone through that process, staggered so that no more than a third reprices in any one year.

Second, require every capital proposal for a thermal asset to show its electrification path, even when the honest answer is that none is economic yet. The paper should state the technology, the connection size, the expected time in the queue and the gas and electricity prices at which conversion would pay. That turns decarbonisation from a statement of intent into a list of decisions with dates, and it lets directors see which of today's commitments would close tomorrow's options.

Third, treat government co-investment as part of the energy decision rather than as a windfall to be maximised. Programmes of this kind typically attach conditions on operation, reporting and the sharing of results, and the larger the support, the more those conditions shape the design. Support for a technology a site cannot connect is not support at all. It is a commitment the company may be unable to honour.

One objection deserves an answer: boards are not energy traders, and pulling procurement into the boardroom can slow decisions that markets reward for speed. I accept that risk, and at Marrangal short-term and spot purchasing stays with management. But a contract that decides whether a kiln runs until 2035 is not a procurement matter. It may be the most consequential capital decision the company takes that year, and directors should read it as one.