From 116 to 6: the Copper Discovery Collapse Reshaping Supply

Six significant copper discoveries in five years. In the 1990s, there were 116.

That arithmetic sits at the centre of one of the most consequential supply stories in the resource sector, and it is why any serious copper discovery analysis now begins with a question about geology rather than price. As of September 2026, the International Energy Agency (IEA) projects that supply from existing and announced mines will fall roughly 25% below primary copper requirements by 2035. The world’s appetite for the metal, driven by electric vehicles, power grids, wind, solar, and data centres, is converging on a pipeline that is running dry at the source.

What follows maps the structural forces behind the discovery drought, so you can judge which parts of the copper supply story are cyclical and which are permanent. This is not a piece about where the price is heading; it is about the geological and capital-allocation reality that will shape supply for the next decade.

From 116 finds to 6: the collapse in significant copper discovery

The clearest way to see the drought is to lay the decades side by side. According to S&P Global’s 2026 assessment, 263 significant copper discoveries were made between 1990 and 2025, holding a combined 1.402 billion metric tonnes of copper. S&P Global counts a deposit as significant only when it contains at least 500,000 metric tonnes of copper across resources, reserves, or historical output.

Track the count across the decades and the trajectory becomes hard to ignore. The 1990s delivered 116 finds. The 2000s delivered 104. The 2010s dropped to 37. Then came the most recent window.

The contained metal figures tell the same story with more force than the count alone.

The 2020-2025 window Six significant discoveries. 8.7 million metric tonnes of copper between them. Not a single deposit met the major-discovery threshold in 2025 alone.

What is sitting in the pipeline today

Discovery counts flatter the real supply picture, because most catalogued deposits are nowhere near production. Among the 263 discoveries tracked by S&P Global, production has yet to begin at 165 of them, feasibility studies remain incomplete at 135, and just 17 have advanced as far as construction or preproduction.

Those 17 projects are the full near-term inheritance of 35 years of exploration. Everything else is either still being studied, still being financed, or still sitting as an entry in a database.

The most recent update to that database added roughly 37 million metric tonnes, a gain of about 3%. The majority of that increase came from expansions of already-known deposits rather than independent new finds, which tells you the pipeline is not just thinning at the front. It is nearly empty at the source.

Where the exploration money actually goes, and why it avoids the frontier

The budget split explains the discovery collapse better than any single geological fact. In 2025, mine-site programs captured 43% of global copper exploration budgets, while grassroots exploration, the search for entirely new deposits, took just 25% of a total that reached US$3.30 billion, a 2% rise on the prior year.

This is not a failure of nerve. It is a rational response to how copper exploration is financed, and understanding the mechanism matters because it is the upstream driver of the discovery numbers in the previous section.

Mine-site programs win capital for reasons that have nothing to do with geology and everything to do with efficiency:

• Existing mills, tailings facilities, and power systems are already built and paid for.
• Permitting routes are established, cutting years off the path to output.
• Capital risk is lower, because the deposit’s ore body and metallurgy are already understood.

– Grassroots exploration offers none of that comfort. It maps unknown geology, establishes access to remote ground, and lays the foundation for projects that may take more than a decade to define, if they materialise at all. When capital is scarce and impatient, the frontier is the first thing to be cut.

The frontier is the first thing cut when capital is scarce, and the distinction between grassroots prospecting methods and mine-site extension work is not merely technical: these are structurally different activities competing for the same budget lines, with very different payoff timelines and risk profiles.

That is precisely what the long-run data shows. Grassroots exploration typically absorbed 50-60% of copper budgets through the 1990s and early 2000s, according to S&P Global. Between 2012 and 2025, grassroots copper exploration spending fell 68% to US$0.9 billion.

The structural driver Grassroots copper exploration spending has fallen 68% since 2012, to roughly US$0.9 billion. This is the part of the cycle that generates genuinely new supply options, and it has been quietly withdrawn.

The consequence flows straight through to the numbers. When S&P Global tallied the 2025 rise in total discovered copper volume, deposits tied to existing mining complexes accounted for well over 60% of that gain. What this tells you is that mine-site extensions cannot fill the grassroots gap; they are two different activities. One squeezes more from known ground, the other finds the ground in the first place, and for analysts assessing junior miners or diversified majors, today’s grassroots drought is a leading indicator of a thinner project pipeline in 10-15 years.

Deeper holes, lower grades, and the compounding cost of finding copper today

Even where the money is spent, the ground itself has become harder and more expensive to work. The barriers do not simply add up; each one makes the next more punishing.

1. Depth. Near-surface deposits are largely exhausted, forcing programs to drill through deeper cover into more geologically uncertain rock. Average drilling depth has risen nearly 50% since 2010, reaching approximately 600 metres.
2. Grade decline. Global average copper ore grades have fallen roughly 40% since 1991, to around 0.62%. Codelco’s plant-average grades slipped from about 0.83% in 2009 to roughly 0.63-0.64% by 2023-2024.
3. Capital cost inflation. Lower grades mean processing far more rock for the same copper, which demands larger plants, more energy, more water, and larger tailings volumes. Capital costs per unit of brownfield capacity have climbed roughly 65% since 2020.
4. Regulatory and social barriers. Permitting approvals have fallen to their lowest level in 15 years, and community consent processes can halt projects regardless of technical merit.

– Read those four in sequence and the multiplicative effect becomes clear. Deeper holes cost more per metre; lower grades mean more of those expensive metres must be drilled to define the same tonnage; and the plant built to process that lower-grade ore carries a heavier capital bill than the same plant would have a decade ago.

The result is that the capital required to find and define a tonne of copper in the ground has structurally increased. Marginal discoveries that would have attracted development capital in the 1990s are simply uneconomic now, which is why rising exploration budgets do not translate into rising discovery counts. More money is producing fewer finds because the easy ground is gone.

The regulatory and social barriers compounding the technical picture

The non-technical barriers are structurally distinct from geology, but they compound the same problem. Regulatory approvals at a 15-year low reflect stricter environmental review and shifting tax regimes, and they add uncertainty to projects that are already slow and capital-intensive.

Duplicative permitting processes across federal, state, and local regulatory layers compound the timeline problem: a project that clears one approval regime may still face years of parallel review under overlapping frameworks, each with independent rights of legal challenge.

Community consent is the other pressure point. Flawed implementation of Free, Prior and Informed Consent (FPIC), the principle that affected communities must be properly consulted and must agree before a project proceeds, can derail otherwise sound projects. When consultation delivers overly technical information, compresses deliberation timelines, or is treated as non-binding, the practical result is mistrust, litigation, and suspended development.

Ecuador makes the point concretely. Legal challenges to Decree 754 created a permitting vacuum in 2026, stalling approval processes and leaving developers uncertain about how consent will actually be operationalised. Technical merit offered no protection.

The lag you are inheriting Sources place the timeline from initial discovery to production at roughly 16 to 17.5 years. That is the gap between today’s discovery drought and the moment its consequences arrive in the supply figures.

What the supply math looks like from here

Set the demand and supply lines against each other and the gap emerges on its own. The IEA’s 2026 outlook projects refined copper demand, excluding direct-use scrap, climbing 26% from approximately 27.8 million metric tonnes in 2025 to around 35.0 million metric tonnes by 2040.

Supply is moving the other way. The IEA projects output from existing and announced mines falling roughly 25% below primary copper requirements by 2035, an improvement on the roughly 30% gap in the prior year’s outlook, but a structurally significant shortfall nonetheless.

The clearest signal of what closing that gap demands is the capital figure.

The capital requirement The IEA’s 2026 outlook estimates copper mining and refining will need around US$310 billion in capital investment through 2040.

Other institutions frame the deficit differently. S&P Global has projected a 23.8% deficit by 2040, McKinsey an 18% gap by 2031, and BloombergNEF a 19 million tonne shortfall by 2050. These are useful for range, though they are unverified and should not be treated as anchors.

The copper structural deficit is not a single number but a range of institution-specific projections, each using different assumptions about demand growth rates, scrap recovery curves, and project commissioning timelines; the spread across those projections matters as much as any individual figure.

Secondary supply and substitution soften the picture, but only so far:

• Secondary refined copper reached 4.5 million tonnes in 2023, about 20% of total output.
• Material substitution, such as aluminium replacing copper in some applications, can trim demand at the margin.
• Higher capacity utilisation at existing operations adds incremental tonnes, not new deposits.

– Here is what the arithmetic actually tells you. A US$310 billion capital requirement, set against a discovery pipeline that produced six significant finds in five years, means the scale of investment being called for has no plausible supply of new projects to absorb it on current trajectories. For investors, that gap defines the opportunity: projects with defined resources already advancing toward development carry disproportionate strategic value precisely because new alternatives are not being found.

What the discovery drought changes for investors assessing copper supply

Step back from the data and one distinction organises everything: what is cyclical, and what is not.

Some of the drought will respond to a rising copper price. Budget allocation, financing conditions, and junior access to capital all loosen when the metal rallies, which means the grassroots share of spending could recover in a stronger price environment.

• Cyclical factors: exploration budget allocation, financing conditions, and junior miners’ access to capital.
• Structural factors: grade decline, rising drilling depth, and the regulatory and social consent environment.

– The structural factors will not self-correct with a commodity cycle. Ore grades will not rise because copper is expensive. Near-surface deposits will not reappear. Permitting timelines and consent frameworks will not become predictable because a spot price crossed a threshold.

That is why the 17 projects at construction or preproduction represent a genuinely scarce class of asset. The geological and regulatory conditions that would allow new alternatives to be found and developed in time to address the 2035 gap simply do not exist right now.

Investors wanting the full market-dynamics picture alongside the discovery data will find our dedicated guide to the 2026 copper supply deficit, which covers spot price behaviour, inventory drawdown signals, and the near-term trading implications of the structural supply gap this article maps.

The 16-17.5 year development clock makes the point unavoidable. A project entering feasibility today will not reach production before the IEA’s 2035 supply gap arrives, and secondary copper at around 20% of supply is a ceiling on recycling’s contribution, not a rescue.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions. Past performance does not guarantee future results, and these projections are speculative and subject to change based on market developments.

The window for new supply is narrower than the capital figures suggest

The core finding is not that copper faces a capital problem. It faces a time and geology problem that capital alone cannot solve at the pace the energy transition demands. Money can be raised faster than ore bodies can be found, defined, permitted, and built.

Three variables will determine whether the 2035 gap widens or narrows: the rate at which grassroots exploration recovers and begins delivering feasibility-stage projects, the pace at which regulatory and social consent frameworks become more predictable in key jurisdictions, and the degree to which secondary copper and substitution can structurally expand as primary mine output falls short.

Read those as a checklist, not a forecast. Given the development clock and the current state of the grassroots pipeline, anyone assessing copper supply for the 2030s is already working with most of the material that will exist. The open questions are regulatory timelines and secondary supply growth, not discovery volume.

—