Where a building-products strategy is actually decided

Our volumes follow detached and medium-density commencements in five states, with a lag that differs by product: bricks and frames are bought early in a build, plasterboard and windows months later. I built Marrangal's commencement model in 2009 and still own it. It forecasts by state, dwelling type and product from approvals, housing-finance commitments, builders' order books and land-title releases, and it is reconciled each quarter against our own despatches. Since 2022 every division's plan has gone to the board at three levels: a central case, a stretch case and a trough case that assumes a 25 per cent fall in detached commencements held for two years. The trough case is binding for capital decisions. It stopped a second Queensland brick kiln in 2022, set our walk-away price for a roof-tile acquisition in 2021 and shaped the timing of our 2024 frame-and-truss purchase. I also publish to the executive team how far each forecast missed, because a model that is never scored becomes a belief.

Our kilns and plasterboard dryers turn gas into product, and gas is between a fifth and a third of their cash cost, depending on the contract. Energy strategy sits in my team, with three specialists who work alongside procurement and the CFO. We plan contracts asset by asset rather than state by state, so that each kiln's remaining life and each contract's term are decided together. Since 2023 the board has approved every gas contract longer than three years or larger than A$20 million a year. Contracts now cover about 70 per cent of forecast kiln and dryer demand to 2030, staggered so that no more than a third reprices in any one year. For electricity we hold two wind-linked power purchase agreements covering 45 per cent of our load. I keep a register of the grid connection each site would need to electrify, with an estimate of its time in the queue, because that constraint now governs our decarbonisation path more than the technology does.

I test every product line against three properties: weight relative to value, how quickly customers need it and how far it is made to measure. Bricks and plasterboard are protected by freight. Standard aluminium profiles are not. Made-to-measure windows and prefabricated frames are protected by time and fit. As general manager of windows I applied that test when I closed our extrusion press, withdrew from high-rise apartment windows where imported systems from Chinese fabricators were 25 to 30 per cent cheaper, and invested A$19 million in insulated glass made to order. I also insist on an honest landed cost for imports, including breakage, delay, the safety stock a long supply line requires and the cost to a builder of an idle crew. The test cuts both ways. We import some hardware, fixings and specialty glass ourselves, because on those three properties it would be vanity to make them here.

Since 2012 I have led or advised on eleven Marrangal transactions, completed or abandoned. They include the sale of our loss-making concrete-pipe business for A$64 million in 2015, the purchase of a Perth window fabricator for A$38 million in 2014, the sale of 40 per cent of our plasterboard business to Hayatsu Gypsum for A$165 million in 2016 and the purchase of Kurrawong Frame & Truss for A$96 million in 2024. I set a walk-away price before any negotiation, based on trough earnings rather than the vendor's best year, and write it into the board paper so that it cannot drift in the excitement of a process. I also sit on the board of the plasterboard joint venture as Marrangal's nominee, which obliges me to act in the joint venture's interest when it differs from Marrangal's. Working with Hayatsu's nominees has taught me to put a disagreement in writing well before a meeting, so that it is settled by evidence rather than by seniority in the room.

Programmes that co-fund industrial decarbonisation and onshore manufacturing have become a regular part of capital planning in Australia. I treat each offer as a contract rather than a windfall, and I price its conditions, which typically cover operating periods, performance reporting and the sharing of results, as obligations alongside the cash. In 2024 the plasterboard joint venture was offered two routes for its second Brisbane line: A$62 million towards a fully electric dryer or A$28 million towards a hybrid. I recommended the smaller sum, because the larger one depended on a 32-megawatt connection that our network could not deliver for four to five years. I also work with state governments on planning and energy questions for our brickworks, and I keep those conversations apart from any grant application so that neither is mistaken for a condition of the other. My test is simple: would we make this investment, in this form, without the money? If not, the grant is designing the plant.