Ore Rich, Refinery Poor: Why the U.S. Copper Deficit of 330,000 Metric Tons Is a Processing Crisis, Not a Mining One

Introduction

Here is a paradox worth sitting with for a moment. The United States has enough copper in the ground, and in its scrap yards, to meet all of its own needs and then some. Benchmark Mineral Intelligence calculates that domestic mine output and scrap together can cover 146 percent of U.S. copper demand. By that measure, America is not a copper-poor country.

And yet J.P. Morgan Global Research projects a refined copper deficit of 330,000 metric tons for 2026, one of the largest shortfalls on record. Manufacturers are dependent on foreign processing facilities to turn American ore into usable metal. Nearly half of the copper concentrate mined in the United States is shipped overseas for smelting, primarily to facilities in Mexico, Canada, Japan, and China, before much of it is bought back in refined form.

The diagnosis is uncomfortable but precise: the United States does not have a copper supply problem. It has a copper processing problem. And that distinction matters enormously, because the policy tools needed to fix a mine shortage are completely different from those needed to fix a refinery shortage. Getting the framing wrong means spending money and political capital in the wrong place.

This article explains how the processing gap formed, why it is widening now, and what the realistic options look like for closing it.

The Numbers Behind the Paradox

Start with the basic arithmetic. In 2024, the United States produced approximately 1.2 million metric tons of mined copper, concentrated in Arizona, Utah, and Nevada. It also collected around 870,000 tons of copper scrap. On the supply side of the ledger, the raw material position looks strong.

The problem emerges the moment you look at processing. The United States has just two operational primary copper smelters: Rio Tinto's Kennecott facility in Utah and Freeport-McMoRan's Miami operation in Arizona. Together they produce slightly more than 400,000 tonnes of refined copper per year, which is less than 40 percent of domestic mining output. Total domestic smelting capacity sits at around 585,000 metric tons, roughly half of what the country mines annually.

The arithmetic on the consumption side closes the gap in an uncomfortable way. The U.S. consumed 1.6 million tons of refined copper in 2024 and imported 720,000 tonnes to bridge the shortfall. Sixty-five percent of those imports came from Chile, with Canada, Mexico, and Peru making up most of the rest. The country that mines more copper than it needs is spending heavily to import the processed version of its own resource.

Albert Mackenzie, an analyst at Benchmark Mineral Intelligence, summarized the situation plainly in February 2026: "Both via scrap and domestic mining the US is more self-reliant. In fact, if you remove all US overseas mines, the US still is self-sufficient for raw materials. The problem is that the US doesn't have the processing capacity." He added that expanding scrap processing could be a faster route to closing the gap than focusing first on overseas mine investment, a point that tends to get lost in policy conversations dominated by upstream supply security.

How the Processing Infrastructure Was Lost

The erosion of U.S. copper processing capacity did not happen overnight. It was the cumulative result of decades of economic decisions, regulatory changes, and global market shifts that made it cheaper to process copper elsewhere than at home.

Stricter environmental standards in the United States, particularly around sulfur dioxide emissions from smelting, raised operating costs significantly relative to facilities in countries with less stringent rules. Smelting is inherently a chemistry-intensive, pollution-generating process: copper concentrate contains sulfur, and burning it off produces sulfuric acid fumes that must be captured and treated. Modern smelters in wealthier countries are required to run elaborate acid-plant systems to manage this, adding capital and operating cost that competitors in lower-regulation environments do not always bear equally.

At the same time, global demand was shifting. As China industrialized rapidly from the 1990s onward, it built smelting infrastructure at a scale no other country matched. Since 2005, China has accounted for over 90 percent of growth in global copper smelter output, lifting its share from around 15 percent to approximately 50 percent of global capacity by 2025. Chinese smelters benefit from lower labor and construction costs, greater economies of scale, newer facilities with better energy efficiency, and, critically, state ownership structures that allow them to continue operating during periods when market conditions would force a privately-owned Western smelter to shut down. The IEA has noted that some Chinese facilities operate against physical output targets rather than financial ones, making them structurally more resilient to market downturns.

The result in the United States has been what the Columbia University Center on Global Energy Policy describes as a fragmented governance landscape, with no single agency coordinating copper policy across the Departments of Interior, Energy, and Commerce, the EPA, and the State Department. As Richard Holtum, CEO of commodity trader Trafigura, put it: "Mining is not critical, refining and smelting is critical. If you don't have it in your country, then you are at the mercy of someone that does and their ability to turn on or off that smelting capacity."

A Market Already Under Pressure: Supply Disruptions and the TC/RC Signal

The structural processing deficit would be serious enough in a stable market. In 2026, it is unfolding against a backdrop of significant global supply disruption that is squeezing the system from multiple directions.

In late 2025, a massive mudslide at the Grasberg mine in Indonesia, the world's second-largest copper producer, triggered a force majeure. As of early 2026, the mine remains in only a phased restart. A separate seismic event in May 2025 caused flooding at the Kamoa-Kakula complex in the Democratic Republic of Congo, a flagship Ivanhoe Mines project. Further disruptions from strikes and environmental closures in Chile and Peru have compounded the picture. Carlos Piñeiro Cruz, principal copper analyst at Benchmark Mineral Intelligence, noted that pre-disruption growth was originally forecast at around 2 million metric tons for 2026, but has since been downgraded by approximately 700,000 metric tons, with most of the reduction coming from the Escondida mine in Chile.

Price forecasts diverge depending on how analysts weigh these disruptions against longer-term demand signals. J.P. Morgan expects prices to average around $12,075 per metric ton for the full year, with a Q2 peak near $12,500. Goldman Sachs sits more conservatively at around $11,500 per tonne. Citigroup sees potential for prices to approach $15,000 per ton if low inventories persist. LME cash copper briefly hit $13,300 per metric ton on January 6, 2026, a 50 percent year-on-year increase.

Perhaps the most telling market signal is what has happened to treatment and refining charges, known in the industry as TC/RCs. These are the fees that smelters charge miners to process copper concentrate. When they fall, it means smelters are competing aggressively for limited ore supply, a signal of structural tightness. In January 2026, the annual benchmark between Chilean miner Antofagasta and Chinese smelters settled at zero dollars per tonne, the lowest level ever recorded in annual negotiations. Spot TC/RCs had already been negative since 2024. Even China's major smelters have agreed to production cuts of over 10 percent in 2026, and the Chinese government has halted approximately 2 million tonnes of planned new smelting capacity. That even the world's most cost-competitive smelter network is retrenching underlines that this is not a cyclical dip but a structural signal.

The New Pressure: AI Data Centers and the Competition for Electricity

Superimposed on these supply-side pressures is a demand-side shift that few analysts fully anticipated even three years ago: the emergence of artificial intelligence infrastructure as a major copper consumer and, more subtly, as a competitor for the electricity that copper smelters need to operate.

A single large AI data center facility can require up to 50,000 metric tons of copper, three to four times more than a conventional data center. J.P. Morgan estimates that copper demand from data centers alone could approach 475,000 metric tons annually. S&P Global projects total copper demand will reach 42 million metric tons by 2040, a 50 percent increase from current levels, with AI data centers as a primary growth driver alongside electric vehicles and grid infrastructure. The scale of technology sector capital expenditure reinforces the demand trajectory: Google, Microsoft, Meta, and Amazon are collectively projected to spend $725 billion on infrastructure in 2026, up 77 percent from the prior year.

But AI's impact on copper refining runs deeper than simple demand addition. Smelting and refining are among the most electricity-intensive industrial processes in existence. As data centers claim a growing share of the U.S. power grid, the cost and availability of electricity for industrial users becomes more constrained. S&P Global estimates that data centers' share of total U.S. electricity demand will rise from 5 percent today to as much as 14 percent by 2030, with rising electricity costs rippling through to consumer power bills. Industry analysts note that if copper smelters must compete with hyperscale data centers for grid access, the economics of domestic refining, already marginal, become even harder to justify without explicit policy support.

The political dimension adds complexity. The current administration is simultaneously pushing to accelerate AI data-center buildouts and reversing clean-air regulations that had imposed stricter emissions limits on the two existing U.S. copper smelters. Critics at the World Resources Institute have argued that loosening environmental standards on existing smelters, while offering no comparable incentive framework for new low-emission capacity, signals reactive rather than strategic industrial policy. A truly modern copper smelter requires an acid-plant system to capture sulfur emissions regardless of regulatory minimums, and that cost does not disappear by rolling back a rule. It simply shifts from a compliance obligation to an economic question that investors must answer without regulatory certainty.

Policy Levers and Prospective New Capacity

The policy response to date has centered primarily on tariffs. In July 2025, the Trump administration announced a 50 percent tariff on copper imports, though subsequent clarifications limited the most aggressive application: articles made entirely or almost entirely of copper face a flat 50 percent on their full value, while derivative articles substantially made of copper pay 25 percent. Commerce Secretary Howard Lutnick framed the goal plainly: "The idea is to bring copper home, bring copper production home."

The tariff logic is straightforward in theory. By raising the cost of imported refined copper, the administration hopes to improve the return on investment for domestic smelting and refining. In practice, analysts point out that a modern copper smelter costs between $1.8 billion and $2.5 billion to build, requires five to ten years from planning to operation, and demands sulphur-capture and acid-plant systems that add substantially to both capital and operating cost. Even with tariff protection, the TC/RC environment makes new smelter economics extremely difficult: smelters earn money by charging miners for processing, and with TC/RCs at or below zero, there is no processing fee income to anchor a business case.

Several prospective projects are nonetheless advancing. Grupo México is considering modernizing and restarting its Hayden smelter in Arizona, which has been shut since 2019. Falcon Copper is evaluating a $2 billion greenfield smelter and refinery in the western United States, with potential Japanese participation as suppliers or off-takers highlighted in a recent U.S.-Japan joint factsheet. And Aurubis has already opened its Richmond facility in Augusta, Georgia, the newest copper smelter in the country, designed specifically to process scrap rather than concentrate. Once fully operational, Richmond will produce around 70,000 metric tons of refined copper annually. That is a meaningful addition, though it represents less than 10 percent of current annual copper imports.

The designation of copper as a critical mineral in November 2025 opens the door to financing tools, stockpiling authority, and interagency coordination mechanisms that were not previously available. Whether those tools are deployed through the kind of integrated industrial strategy that experts have called for, or remain available on paper without institutional follow-through, is the question that will shape the next several years of market development. Building on my analysis of the U.S.-India Critical Minerals Cooperation Framework in June 2026, the recurring pattern is that the gap between diplomatic and policy ambition and the institutional architecture needed to deliver it is where critical minerals strategies most often fail.

What to Watch

The 330,000 metric ton refined copper deficit is not a forecast about geology. It is a forecast about infrastructure, policy coherence, and investment timelines. The ore is there. The scrap is there. What is missing is the industrial capacity to complete the value chain on American soil, and that gap cannot be closed quickly even under the best conditions.

In the near term, watch whether TC/RC signals stabilize or continue deteriorating, as this will determine whether any of the prospective smelter projects can assemble a credible financing case. Watch the electricity price trajectory, particularly in the Southwest where both copper mining operations and data-center construction are concentrated. And watch whether the critical mineral designation for copper translates into the kind of coordinated interagency investment framework that experts consistently say is missing, or whether it remains a label without a delivery mechanism.

For manufacturers and supply chain managers, the practical implication of the 2026 deficit is already visible in import prices, lead times, and procurement strategies. The structural story, however, plays out over a longer arc: S&P Global projects global copper production peaking in 2030 at 33 million metric tons even as demand accelerates toward 42 million metric tons by 2040. Benchmark Mineral Intelligence estimates that 100 new copper mines will be needed by 2035 to meet growing demand, requiring over $210 billion in capital investment. None of that investment will deliver usable metal unless the processing infrastructure to handle it also exists. That is the lesson of 2026, written in a 330,000-ton shortfall in a country that never ran short of the ore.