In the copper slump of the mid-2010s, smelters in Chile kept running while several high-cost operations in other countries cut shifts, suspended production or closed entirely. The metal price was the same for everyone. What differed was how much each mine spent to get a tonne of copper out of the ground and to market. That gap — between the price the market pays and the cost a specific operation incurs — is the whole story of who survives a downturn and who does not.
What a cost curve actually is
A cost curve lines up every producing mine in a given commodity from left to right, ordered by their unit cost of production, cheapest on the left and most expensive on the right. The horizontal axis represents cumulative output — tonnes of metal produced per year — so as you move right you are also accumulating global supply. The vertical axis shows unit cost. The resulting shape is roughly a rising curve, sometimes called a supply curve or an industry cost curve, though it is rarely smooth: there are steps and flat sections where many mines cluster at similar costs, then a sharp rise at the right-hand end where a small number of very high-cost operations sit.
The market price, at any moment, sits like a horizontal line cutting across that curve. Every mine to the left of the intersection — those whose costs fall below the price — is, at least in cash terms, making money. Every mine to the right is losing money on each tonne it sells. The mine sitting exactly at the intersection is breaking even. When the price falls, the line drops, and more mines slide into loss. When the price rises, more mines become viable. This is not merely a conceptual device: producers, traders and analysts use actual cost curve data, compiled from public filings and engineering studies, to judge where the price floor might be and which operations are most exposed.
Cash cost, all-in cost, and why the distinction matters
Not all costs are equal in a downturn. Mines distinguish between cash costs — the direct operating expenditure needed to produce and sell a unit of metal, covering labour, energy, consumables, transport and site administration — and a broader measure that adds sustaining capital, royalties and corporate overheads. Industry practice has converged on the all-in sustaining cost, or AISC, as a more honest comparison, because it captures what a mine genuinely needs to spend to keep running at a steady rate without running down its infrastructure.
A mine might have a cash cost comfortably below the spot price but an AISC that sits above it. In the short run it will keep operating, because every tonne sold at least covers the cash going out the door. But it cannot sustain itself: equipment wears, tailings dams need raising, development headings need advancing. If the low price persists, the mine either finds fresh capital or begins a slow decline. Cash costs tell you who shuts down this quarter; all-in costs tell you who cannot survive the next two or three years.
A worked illustrative example
Suppose, for illustration, two hypothetical copper mines. Mine A processes ore at a grade of 1.2% copper with a simple flotation circuit close to a port, and achieves a cash cost of $3,800 per tonne of copper and an AISC of $4,600 per tonne. Mine B processes ore at 0.4% copper in a remote location, requiring an energy-intensive leach circuit and long-haul trucking, producing copper at a cash cost of $5,200 per tonne and an AISC of $6,400 per tonne. These figures are illustrative and do not represent any specific operation.
If the copper price stands at $8,000 per tonne, both mines operate profitably on an all-in basis. Mine A earns $3,400 per tonne above AISC; Mine B earns $1,600. Now suppose the price falls to $5,500 per tonne. Mine A is still generating $900 per tonne above its AISC — it is fine. Mine B is now $900 per tonne below its AISC. Its cash cost is $5,200, so it is still marginally above breakeven on a cash basis and management may choose to keep running to preserve jobs and relationships, gambling on a price recovery. But it is consuming its asset base rather than sustaining it. If the price drops further, to $5,000 per tonne, Mine B falls below even its cash cost and every tonne it sells accelerates the loss. At that point, suspension is the rational decision.
The arithmetic is straightforward, but the management decision is not. Mines carry fixed obligations — debt service, community agreements, union contracts — that do not disappear when production stops. Temporarily idle mines still incur care-and-maintenance costs. Restarting later is expensive and slow. So real-world managers sometimes continue operating at a cash loss for months, particularly if they believe the price dip is brief, if they are in the middle of a debt covenant period, or if a shutdown would trigger penalties. The cost curve predicts the order of exit, not its exact timing.
What shifts a mine's position on the curve
Ore grade is the most fundamental driver. Doubling the grade of ore through the same mill roughly halves the processing cost per unit of recovered metal, all else equal. But grade alone is not destiny. Energy prices matter enormously for operations that rely on diesel generation or intensive hydrometallurgy. Exchange rates shift dollar-denominated costs for mines operating in other currencies — a weakening local currency can improve the dollar cost position of a mine overnight. By-product credits alter the effective cost significantly: a copper mine that also recovers gold, silver or molybdenum can apply those revenues against its copper cost, lowering its reported position on the curve. Strip ratio in open-pit operations — the volume of waste moved per unit of ore — determines how much of the mining cost is genuinely productive. Infrastructure age and location determine sustaining capital requirements.
Over a mine's life, its cost position moves. High-grade ore at the top of a deposit is mined first; grades typically decline as the operation matures, pushing the mine rightward on the cost curve over time. Conversely, investment in new processing technology or a higher-capacity mill can push costs down. The curve, in other words, is not a static ranking but a snapshot of a continuously shifting competitive landscape among producers.
Where this leads
Readers who want to go further should look at how analysts construct marginal cost models to estimate price floors, and at the relationship between cost curves and the incentive price — the all-in cost at which new capacity needs to be built to meet future demand. That territory connects cost curve analysis to capital allocation cycles, project economics and the long-run behaviour of commodity markets.