The base case is zero. The reality is 20 per cent over. Mining’s most expensive assumption isn’t going away.

Will Kendall has a line that lands with the quiet force of something everyone in the room already knew but hadn’t quite said out loud: "Base case is cost overruns of over 20 per cent," Will says, "That’s your base case."

Will is National Head of Mining and Mining Services, Financial Advisory at Grant Thornton, and the co-author of the firm’s most recent From development to production: financial management across the mining lifecycle report. In an interview for The Rock Wrangler, he walked through the funding landscape reshaping how Australian mining projects get built: private credit, royalty finance, JV-style contractor arrangements, government-backed critical minerals facilities, and the broader shift in appetite for coal and battery metals. All of it is useful context. None of it is as operationally consequential as the single structural pattern he keeps returning to.

The industry has a systemic capex problem. Every feasibility study, every NPV calculation, every debt-service coverage model, every equity raise pitch deck starts from a capex number that history says will be wrong by at least a fifth. This is based on Will’s own experience but he also references a report he read which surveyed 100+ projects showing base-case cost overruns above 20 per cent as the industry-wide norm. And yet the number that goes to the board, to the lenders, and to the market is almost always presented as if the base case is zero.

For mining professionals who have been through the ramp-up phase of a project that didn’t land where the FID model said it would, none of this is news. What surfaces from the conversation with Will is the structural reason it keeps happening.

The copper leach that ate the shareholders

His worked example is a copper project he has chosen not to name, but the pattern will be familiar to anyone who has watched a junior go through a distressed restructure.

"Fourth-quartile, heavily debt financed, and it was a leaching operation, so the cash flow just took ages and ages to come," Will says. "There were three layers of debt. A secured facility from a hedge fund, second-ranking from a contractor, and third-ranking from a commodity trader."

None of this was hidden. All of it was disclosed. On the FID model, the numbers worked.

They didn’t work in practice. Ramp-up took longer than budgeted. Costs came in higher than budgeted. The debt-service obligations didn’t care.

"It all got too hard," Will explains, "and it ultimately resulted in the company being suspended on the ASX. There was about two years before a restructuring was worked out, and the existing shareholders probably ended up with 10 to 20 per cent of the company. A massive dilution."

The dilution the board had structured the funding to avoid arrived anyway, an order of magnitude worse than a properly-sized equity raise at FID would have delivered. This is the pattern Will says he sees repeatedly across the junior developer space, and it is almost always traceable back to the same origin: a capex number that assumed the base case was zero when the industry’s own empirical record says it isn’t.

The question nobody wants to answer

Ask Will which of the three board-level financial questions he would want a developer to answer before committing serious capital to construction, and he will tell you they are all important: capex confidence, funding buffer for delay, and whether the Management team can actually make the transition from explorer to developer to producer.

Ask him which one boards most commonly struggle to answer honestly, and the answer is immediate.

"I’d say the first one," Will says, "because everyone doesn’t want to be the guy or the person to say, do we really believe in those capex numbers? We might have to increase them just a little bit more, because ultimately that’s going to impact your NPV, which is what you’d be projecting to the market and what’s the base of your funding."

The structural tension is that the CAPEX estimate supporting the investment case is generally produced or commissioned by the same project sponsor that needs the economics to work. JORC imposes formal professional accountability for resource and reserve disclosures, but there is no universal equivalent for CAPEX. Banks often bring in an Independent Technical Engineer before providing project finance, but that review comes later and does not cover every funding pathway.

The consequence is a well-documented industry-wide bias toward optimism at the exact point in the project lifecycle where optimism costs the most money. And it flows straight into the funding decision Will identifies as the single most damaging one juniors make.

Debt over dilution, dilution anyway

If the capex number is too low, the funding structure built on top of it will be wrong. Will’s observation is that juniors, particularly those with equity-constrained shareholder registers and no appetite for a larger raise at FID, systematically choose debt over equity in scenarios where the underlying project economics can’t carry it.

"You’ve got fourth-quartile cost operations that are generally owned by fairly equity-constrained juniors who don’t want to raise a whole heap of equity because they’ll dilute their existing shareholders," Will explains. "So they rely on a lot of debt. That’s fine at the time, assuming everything pans out perfectly. But in reality, it doesn’t."

The behavioural logic is straightforward. The dilution from an equity raise is immediate and visible. The dilution from a distressed restructure sits in the future, is not yet real, and can be argued against. Boards choose the risk they can’t yet see over the certainty they can. And so the fourth-quartile cost project ends up carrying a debt load that assumes flawless execution, on a cash flow curve that assumes flawless ramp-up, at a commodity price that assumes flawless market conditions.

Will’s point isn’t that debt is wrong. It’s that debt on a fourth-quartile asset owned by a single-project junior is a very specific kind of leverage, and it needs a very specific kind of buffer. When the buffer isn’t there, the emergency equity raise that comes 18 months into ramp-up dilutes the existing register by 80 to 90 per cent. The dilution the board avoided arrives anyway at ten times the cost.

More funding, same arithmetic

There is a legitimate argument that the range of options available to Australian mining developers is wider now than it has been in a decade. Will runs through the shifts: private credit specialist debt funds stepping into the space traditional banks retreated from on coal (Whitehaven is the reference case, Boab Metals with Merit Capital another); royalty finance emerging from its North American origin into Australian gold and silver deals (KGL Resources with Wheaton Precious Metals at USD 300 million, Minerals 260 with Franco-Nevada at USD 170 million); and mining contractor JV and profit-share models being deployed on satellite pits and short-life gold assets by MMS, Mega Resources and others.

Will points to the strong gold price as the catalyst for the contractor-funded model in particular.

"It made a lot of these satellite pits that might only have one to two years, maybe three years at best of mine life, very profitable," he says. "So you have a situation where mining contractors came in and said, look, if you the mine owner can get access to a plant, we’ll go and mine that satellite pit for you. You don’t have to put in any money in terms of the pre-strip or to get it into production. If you give us a share of the profits, we’ll fund all of that until the cash from the gold sales come through."

Government-backed vehicles have added another layer. NAIF, Export Finance Australia and the Critical Minerals Facility are all making real capital available, with real drawdowns and, in some cases, real defaults. NAIF has carried exposure to mineral sands projects that ended up in administration. At the same time, Iluka’s Eneabba rare earths refinery is drawing on the Critical Minerals Facility, while Arafura and Alpha HPA have secured support as part of the Government’s push to build domestic critical minerals supply chains. The optionality is genuinely useful. But none of it fixes the underlying capex confidence problem, because every funding structure, no matter how sophisticated, is priced off the same feasibility inputs. Royalty finance at a 20 per cent capex underrun is expensive dilution just as debt at a 20 per cent capex underrun is expensive default risk. The funding source doesn’t change the arithmetic. The capex number does.

The fix nobody’s building

Will doesn’t say it directly, but the implication of his analysis is that the industry needs an equivalent to Competent Person discipline on the capex side. Not necessarily a regulatory imposition, but a cultural one: mandated stress-test disclosure showing what happens to NPV, debt-service coverage and equity dilution risk at 10, 20 and 30 per cent capex overrun scenarios. If those numbers were routinely disclosed at FID, boards would find it harder to defend the base-case-is-zero assumption. Lenders would price differently. Shareholders would have a clearer view of the risk they were carrying.

None of this is technically difficult. The models already exist inside every developer’s finance team. The disclosure hurdle is cultural, not analytical.

Until that shifts, the industry will keep producing the same pattern. Fourth-quartile assets funded with debt structures that assume flawless execution. Ramp-up delays that shouldn’t be surprising but always are. Emergency equity raises that dilute the register worse than a properly-sized raise at FID would have. And a slow, predictable procession of ASX suspensions and restructures that everyone in the sector will describe, correctly, as regrettable and, incorrectly, as unforeseeable.

What actually matters

The next 12 months are going to bring more junior developers into production than any period since the last cycle turn. Gold and copper are strong. Lithium is restarting. Coal sentiment is easing. The funding market is more diverse than it has been in years.

The projects that survive their own ramp-up won’t be the ones with the cleverest capital structures. They’ll be the ones whose boards asked the uncomfortable question at FID, believed the answer they didn’t want to hear, and built a funding stack with the buffer to absorb it.

The base case is 20 per cent over. Everyone in the industry knows it. The projects that price for it will still be here in three years. The ones that don’t will be case studies in someone else’s report.