In January 2026, copper broke through $13,000 per metric ton on the London Metal Exchange for the first time in history. By May, the red metal was trading above $13,500. The move was not a spike. It was a signal.

The global copper market is entering its most severe structural deficit in 22 years. Morgan Stanley projects a 590,000-ton shortfall in 2026 — the worst since 2004 — widening to 1.1 million tons by 2029. For chief procurement officers in construction, electronics, automotive, and industrial manufacturing, this is not a commodity analyst's problem. It is a supply chain crisis that will land on their desks in 2027 with allocation letters, price escalators, and force majeure notices.

Twenty-two years of underinvestment in new mine capacity, falling ore grades at existing operations, and a demand surge from electrification, AI data centers, and grid modernization have created a structural gap that cannot be closed quickly. Mines take 15 to 18 years from discovery to production. The copper the world needs in 2027 should have been discovered in 2010. It was not.

The Deficit Is Real and It Is Structural

The International Copper Study Group revised its outlook sharply in October 2025. It now projects a refined copper deficit of approximately 150,000 metric tons in 2026, a reversal from the 209,000-ton surplus it had forecast earlier. The ICSG cites mine production growth slowing to 1.4% in 2025 and refined output growth falling to just 0.9% in 2026 as concentrate supplies tighten. These are not cyclical fluctuations. They are structural constraints.

590,000
Metric ton copper deficit projected for 2026 — worst in 22 years (Morgan Stanley)

Wood Mackenzie estimates the world needs more than 8 million metric tons per year of new mine capacity plus an additional 3.5 million tons per year of scrap supply by 2035 just to balance demand. Without this investment, the consultancy warns of "sustained shortages and price volatility."

The deficit is driven by a simple math problem. Global copper mine production has been effectively flat since 2016. Codelco, the world's largest copper producer, saw its output fall to 1.32 million tons in 2024, its lowest level in 25 years, as ore grades declined and water scarcity intensified at its Chilean operations. Anglo American's Los Bronces mine in Chile saw production drop 32% year-over-year in 2023 due to water shortages. Chile has been in drought since 2010. Water availability has fallen 37% in 30 years.

"The deficit will not resolve itself through price signals alone. The lead time on new supply is measured in decades, not quarters."

Price at $13,400: Where It Goes From Here

LME cash copper opened 2026 at $13,010 per metric ton on January 2. By January 6 it had hit $13,300. Spot prices reached $14,528 on January 29, an all-time nominal high. Prices have consolidated in the $13,400 to $13,600 range through the second quarter of 2026.

Goldman Sachs raised its near-term copper price target to $13,735 per metric ton in June 2026. The bank's longer-term view is more aggressive: LME copper reaching approximately $15,000 per metric ton by 2035, driven by structural electrification demand and AI data center buildout. CRU Group and S&P Global have published similar trajectories, with CRU forecasting persistent deficits through at least 2030.

$13,400+
LME copper per metric ton (Q2 2026)
$15,000
Goldman Sachs 2035 price target
1.1M
Metric ton deficit projected by 2029
18 yrs
Average lead time for new copper mine

AI Data Centers: The New Demand Vector No One Modeled

The fastest-growing source of copper demand was barely on procurement's radar three years ago. AI data centers consume massive amounts of copper in power distribution, busbars, grounding, transformers, and cooling systems. A single hyperscale data center can require 5,000 to 10,000 tons of copper. Goldman Sachs estimates data center copper demand will reach 500,000 tons annually by 2030, up from approximately 200,000 tons today.

S&P Global's 2025 special report "Copper in the Age of AI" identifies data centers as "the most significant new copper demand driver since China's urbanization boom." The report estimates that AI-related copper demand will add 1.2 million tons of annual consumption by 2035, equivalent to roughly 5% of current global production.

This demand is concentrated in time. The AI infrastructure buildout is happening now, in a 3- to 5-year window. It overlaps directly with the period when mine supply is tightest. The result is competition for physical copper between traditional industrial buyers and hyperscalers who can pay a premium and lock in multi-year contracts.

EVs Consume 3 to 4 Times More Copper Than ICE Vehicles

A conventional internal combustion engine vehicle contains approximately 23 kilograms of copper. A battery electric vehicle contains 60 to 90 kilograms. The difference is not marginal — it is a tripling of copper intensity per vehicle.

Wood Mackenzie notes that EVs can use "up to three and a half times as much copper" as ICE vehicles. When charging infrastructure is included — the cabling, transformers, and substations required to support a national EV fleet — the per-vehicle copper demand jumps further.

Global EV penetration is projected to reach 30% of new vehicle sales by 2030, up from 18% in 2024. Each percentage point of EV market share adds roughly 40,000 tons of copper demand. The IEA reports that renewable energy infrastructure, including solar and wind, needs 2.5 to 7 times more copper per megawatt than fossil fuel-based generation.

Supply Side: Mines Are Closing, Not Opening

The supply picture is deteriorating in real time. First Quantum Minerals' Cobre Panama mine — a 350,000-ton-per-year operation — was ordered closed by the Panamanian government in November 2023. The mine remains shut with no clear path to reopening. The loss represents roughly 1.5% of global mine supply eliminated overnight.

In Chile, the world's largest copper-producing nation, water scarcity, labor unrest, and declining ore grades are compounding output declines. Miners at two major Chilean operations downed tools in August 2024, disrupting production. Codelco's output has fallen 30% from its 2017 peak of 1.84 million tons. Anglo American's copper production from its Chile and Peru operations was 6% lower in Q2 2024 versus Q2 2023.

350,000
Annual metric tons lost from Cobre Panama mine closure alone

Scrap supply, often cited as a buffer, cannot close the gap. The IEA projects scrap volumes growing from 16 million tons today to 19 million tons by 2030 — a 19% increase that falls far short of covering projected demand growth. Copper recycling can provide up to 30% of global supply by 2030, but that still leaves a structural deficit in primary supply of 3 to 5 million tons by mid-decade.

Why 2027 Is the Inflection Point for Procurement

The 2026 deficit is being absorbed by inventory draws. LME and COMEX warehouse stocks have fallen to critically low levels. Shanghai Futures Exchange inventories are near multi-year lows. Once visible stocks are exhausted, allocation becomes the market mechanism.

This is what makes 2027 different. In a deficit market with low inventories, buyers cannot simply pay more to get supply. They are rationed. Suppliers allocate volume based on contract history, relationship, and willingness to accept price escalator clauses. CPOs who have not locked in volume agreements by early 2027 will face allocation letters — formal notices from suppliers that they will receive a fraction of their contracted volume.

Five realities CPOs must accept about the copper market in 2027:

  1. Price will not solve the problem. At $15,000 or $20,000 per ton, there is no new mine coming online in 2027 that is not already in construction. The supply curve is vertical in the short term.
  2. Contract leverage shifts to sellers. Multi-year fixed-price agreements will be replaced by quarterly price resets, surcharges, and volume-dependent pricing tiers. Standard payment terms will shorten from 60 to 30 days or less.
  3. Chinese buyers will set the floor. China consumes 54% of global copper. Its state-owned enterprises and fabricators have the balance sheets to pre-pay for volume and lock supply for years. Western CPOs compete against Beijing's procurement machine.
  4. Substitution is limited. Aluminum can replace copper in some power cables, but not in motors, transformers, or electronics. The substitution ceiling is estimated at 10 to 15% of current copper applications.
  5. Scrap quality degrades with each cycle. Recycled copper loses conductivity with repeated processing. High-purity copper for electronics and EV motors requires primary metal. Scrap can buffer construction-grade applications but not the highest-performance uses.

What CPOs Must Do Before Q1 2027

The window to act is closing. Here is the procurement playbook for the copper crunch.

Audit direct copper exposure now. Most CPOs underestimate their actual copper spend because it is embedded in components: wiring harnesses, motors, transformers, connectors, and busbars. Conduct a full bill-of-materials audit to identify every line item with copper content. The exposure is typically 2 to 3 times what raw copper purchasing records show.

Lock multi-year volume agreements. In a deficit market, volume guarantees matter more than price. Supply agreements with fabricators and distributors should include minimum volume commitments with take-or-pay clauses. Suppliers will prioritize customers who commit to volume over those who shop spot markets.

Develop scrap partnerships. Secondary copper supply is not a substitute for primary metal, but it can buffer price volatility and reduce sourcing risk for non-critical applications. Partner with scrap processors and refiners to secure dedicated scrap feedstocks. Consider tolling arrangements where you supply scrap and receive refined copper.

Redesign for copper efficiency. Engineering teams should review every application where copper can be reduced, substituted, or eliminated. Every kilogram of copper avoided is a kilogram of supply chain risk removed. The payback period on copper-reduction design changes is measured in months at current prices.

Build inventory buffers. The cost of carrying 60 to 90 days of copper inventory is lower than the cost of a production line shutdown. Finance teams need to approve working capital increases for strategic copper stockpiles. The ROI is a function of avoided downtime, not price speculation.

FAQ: The Copper Crunch

How big is the copper deficit in 2026?

Morgan Stanley projects a 590,000-ton deficit. The ICSG forecasts approximately 150,000 tons of refined copper deficit. The discrepancy reflects differing assumptions about scrap supply and demand growth rates. Both agree the market is in structural deficit.

Will copper prices hit $15,000 per ton?

Goldman Sachs targets $15,000 per metric ton by 2035. Spot prices briefly touched $14,528 in January 2026. At current trajectory, sustained trading above $15,000 is likely within 12 to 24 months unless demand growth slows sharply.

Can recycling close the supply gap?

No. The IEA projects scrap volumes reaching 19 million tons by 2030, up from 16 million today. This covers only a fraction of the 8+ million tons per year of new mine capacity Wood Mackenzie says is needed by 2035.

Which industries are most exposed to copper allocation risk?

Automotive (EV wiring harnesses, motors, batteries), construction (power cables, plumbing, HVAC), electronics (connectors, PCBs), and data center infrastructure are the most exposed. Any industry with copper content in its bill of materials faces allocation risk by 2027.

How long does it take to bring a new copper mine online?

15 to 18 years on average from discovery to production. Permitting alone takes 5 to 10 years in most jurisdictions. No new mine that has not already broken ground will produce copper before 2032.

What caused the 22-year underinvestment in copper mining?

Low prices from 2014 to 2020 discouraged investment. The average copper price in 2015-2016 was below $5,000 per ton, well below the incentive price needed to justify new mines. Capital discipline, shareholder pressure for returns, and rising geopolitical risk further suppressed investment.

Sources