In short: What is the copper supply gap? The International Energy Agency (IEA) projects that mines currently operating or under construction will cover only about 70% of global copper demand in 2035 — leaving a ~30% shortfall (roughly 7.4 million tonnes, per S&P Global). Drivers: EVs use about 3x more copper than combustion vehicles; power grids and AI data centres add structural demand. The gap cannot be closed quickly — the average lead time from discovery to first production is 17.9 years (S&P Global, 2024). This is a quantity forecast, not a price target. Not investment advice.
Published: July 18, 2026 | Author: Marco Bozem | Sources: IEA, S&P Global, USGS, BHP Insights, Wood Mackenzie
Thirty percent. That is how much of the copper the world needs by 2035 is simply not covered by today's mine pipeline — not built, not permitted, not even discovered. This is not a number from a bullish research note: it is the calculation of the International Energy Agency (IEA), based on every mining project currently in the pipeline.
Money cannot solve the problem. Even if the copper price tripled tomorrow, a new mine would still take close to 18 years to reach first production. Demand and supply are diverging, and this article explains step by step why — along with how I have invested in this thesis through seven copper-adjacent holdings, with full transparency on gains and one honest loss.
I recorded this analysis as a video — with live charts, my portfolio on Parqet and the full M&A timeline (German language, subtitles available):
Copper is the metal that carries electricity. No other affordable material conducts as well. And the world we are moving into is a world built on electricity.
A conventional combustion vehicle contains roughly 23 kg of copper — wiring harnesses, motors, electronics. A battery electric vehicle? Around 80 kg. That is approximately three times as much copper per vehicle (source: Copper Development Association / International Copper Association via Incorrys; S&P data). And the car is just one wedge.
Four demand drivers are pulling in the same direction simultaneously:
The result: major forecasters project roughly 2–2.5% annual demand growth through 2035 (BHP: 2.6% p.a.; Wood Mackenzie: ~2.2% p.a., +24% by 2035). That does not sound dramatic — but hold that number and compare it against supply in a moment.
Since 1990, there have been 239 significant copper discoveries globally (threshold: at least 500,000 tonnes contained copper; S&P Global Market Intelligence, Nov. 2024). Of those 239, only 14 came from the last decade. Over the past five years, the number of truly large new finds can be counted on one hand. The easy, near-surface deposits are already in production.
In the early 1990s the global average ore grade was around 1.5% copper. Today it sits at 0.6–0.7% — a decline of roughly 40% since 1991 (source: BHP Insights; Benchmark Mineral Intelligence via Investing News Network). This is not a recent trend; it is a 35-year structural shift. In practice: you now need to move almost twice as much rock to extract the same amount of copper. More energy, more cost, less metal.
Say you discover a world-class deposit today. How long until the first kilogram of copper is sold? On average: 17.9 years — exploration, permitting, legal disputes, construction (source: S&P Global Market Intelligence; mines with production start 2020–2023). For comparison, the lead time around 2005–2009 was approximately 12.7 years. The timeline is getting longer, not shorter.
Copper comes from very few countries. Chile supplies roughly 23% of global mine production, the DRC has become number two at ~14–15%, Peru ~11%, China ~8%. The top five producers account for approximately 70% of world output (source: USGS Mineral Commodity Summaries 2025). A strike in Chile, a mine shutdown in the DRC — and a meaningful share of global supply wobbles.
One crucial distinction: China mines only ~8% of global copper, but refines roughly 45% of the world's refined copper production. In 2024, China processed 13.6 of a total 28 Mt of refined copper (source: Shanghai Metal Market; Wood Mackenzie). Mining geography and refining geography are two very different maps — both concentrated.
Demand growing at ~2% per year, supply growing at ~1% per year. Two percent against one percent sounds like small numbers. But when that gap compounds for ten to fifteen years, it becomes a massive physical shortfall.
This is exactly how the IEA reaches its figure: all mines operating or under construction today cover only around 70% of 2035 demand. The remaining ~30% is unmet (source: IEA 2025, confirmed via S&P Global Commodity Insights). The tonnage gap is estimated at approximately 7.4 million tonnes by 2035 (S&P Global) — more than Chile produces in an entire year.
The mining majors see the same gap. And because building from scratch takes almost two decades, they are doing the obvious: buying copper reserves through acquisitions. This explains a consolidation wave not seen for decades.
Here is the honest part. Seven copper-adjacent positions from my portfolio at Trade Republic and Scalable Capital, tracked via Parqet. Not a recommendation — a transparent view of how I have put this thesis into practice, with winners and with one clear loser.
| Name | Type | Shares | Avg. Cost | Cost Value | Current Value | P&L % | Dividends (net) |
|---|---|---|---|---|---|---|---|
| Jiangxi Copper | Pure-play copper (China) | 40 | €1.74 | €69.41 | €144.00 | +104.6% | €3.88 |
| BHP Group | Diversified, copper (Escondida) | 18.19* | €22.91 | €416.74 | €656.18 | +56.5% | €39.46 |
| Rio Tinto | Diversified, copper (Oyu Tolgoi) | 11 | €61.57 | €677.25 | €997.30 | +46.5% | €59.87 |
| Anglo American | Diversified → Anglo-Teck merger | 7.52 | €29.42 | €221.31 | €313.33 | +41.1% | €1.96 (div. cut) |
| Glencore | Copper + Trading | 93 | €5.25 | €488.42 | €564.51 | +10.3% | €53.10 |
| Vale | Iron ore-dominated, growing copper division | 107 | €9.17 | €981.68 | €1,330.01 | +34.2% | €101.81 |
| Central Asia Metals | Copper/Zinc, Kazakhstan | 499 | €1.81 | €905.28 | €813.37 | −10.6% | €36.68 |
| Cluster total | €3,760.09 | €4,818.70 | +26.7% | €296.76 dividends | |||
* BHP topped up ~15 July 2026 from 15 to 18.19 shares. Source: Parqet (Trade Republic + Scalable Capital). As of 16 July 2026. Not investment advice.
For fundamental analysis, fair value estimates and dividend safety checks I use InvestingPro. My affiliate link gives a permanent 15% discount — plus a limited summer sale until 31 July 2026.
Go to InvestingPro (−15% + Summer Sale) → *Affiliate*I receive a commission if you purchase via my link. Your price is lower as a result.
If the copper thesis interests you, there are two fundamentally different approaches:
Pure-play copper cyclicals (Southern Copper, Freeport-McMoRan, Antofagasta, Jiangxi Copper, Teck): maximum leverage to the copper price, volatile, thin dividends. You are buying the cycle — not the income. When copper rises, these names outperform diversified peers significantly.
Diversified majors (BHP, Rio Tinto, Glencore): steadier, broader, decent dividends. Copper is one slice of the pie, not the whole thing. The downside: if copper runs, you capture less upside — but you sleep better at night because iron ore cash flow at BHP buffers copper weakness.
I believe the long-term structural supply shortage is real. But the path there is bumpy, not linear. Investors in copper equities must accept volatility.
Copper demand is rising structurally — EVs, power grids, data centres. Supply cannot keep up — almost no new major discoveries, declining ore grades, 18 years to build a mine. The IEA calculates a ~30% supply gap by 2035. And the mining majors are buying each other rather than building. For me, copper is one of the most compelling hard-asset cases of this decade.
I am invested in seven copper-adjacent holdings with a total cost of approximately €3,760, a current value of ~€4,820, up roughly 27% — plus nearly €300 in dividends received. With clear winners and one honest loss.