The organizing insight of modern critical minerals analysis is that scarcity almost never lives in the ground. The Earth has plenty of neodymium, plenty of gallium, plenty of graphite. What it has very little of is the one plant that turns the ore into something usable, and whoever owns that plant owns the valve. Find the refinery and you have found the chokepoint. That framework holds up across an enormous range of materials and it explains most of the export control drama of the past several years.
Then there is tin, which does not cooperate. Smelting tin is not hard. Cassiterite is tin oxide, you reduce it with carbon in a furnace, and the resulting metal is refined by processes that have been well understood for centuries and are practiced in a dozen countries. There is no solvent extraction cascade, no isotope separation, no single facility whose loss would strand the world. And yet the tin supply chain spent 2025 and 2026 lurching from crisis to crisis, driving the metal above fifty thousand dollars per ton in February 2026 and setting a nominal record on the London Metal Exchange, because of events at three specific pits: one controlled by an armed autonomous region in northern Myanmar, one a sprawling field of alluvial workings on an Indonesian island, and one in a province of eastern Congo where a rebel offensive keeps interrupting the haul road.
That makes tin the counterexample worth keeping in your pocket. The refinery framework is right most of the time, which is exactly why it needs a case that breaks it. Sometimes the constraint really is the rock, and tin has been proving that for about four thousand years.
The oldest supply chain problem in the world
The Bronze Age was named after an alloy that is roughly nine parts copper to one part tin, and the entire civilizational achievement it describes rests on a logistics problem. Copper is common, found across the Mediterranean and the Near East, workable by anyone with a furnace. Tin is not. Tin occurs in economically useful concentrations in a small number of places, and essentially none of them were near the palace economies that needed bronze for weapons, tools, and status objects.
So the Bronze Age ran on long-distance trade in a scarce metal moving from a handful of sources through intermediaries to consumers who had no domestic supply and no substitute. Cornwall, Iberia, Afghanistan, and central Asian sources have all been implicated in the ancient tin trade, and the archaeological argument about which source supplied which region is still live. What is not in dispute is the structure: a critical input, geographically concentrated, moving through routes that could be cut.
When the Late Bronze Age collapsed in the twelfth century BCE, the tin trade collapsing with it is one of the more durable explanations on offer, and the subsequent shift to iron is at least partly a story about a material whose ore is everywhere replacing a material whose ore is not. Iron is worse than bronze in several respects that mattered to a smith. It is also available locally, which turned out to matter more.
That is a four-thousand-year-old demonstration of a principle the tin supply chain still runs on. Concentration of supply plus absence of substitute equals leverage, and it does not require a modern refinery or a state export licensing regime to produce. It only requires geology to be uncooperative, which in tin’s case it reliably is.
The Cornish chapter is worth a beat because it shows how the pattern persists through completely different economic eras. Cornwall supplied tin to Europe for something like two thousand years, through Roman administration, medieval stannary courts with their own legal jurisdiction, and an industrial peak in the nineteenth century, and then lost the market to cheaper Malayan and Bolivian ore and closed down. South Crofty, the last working Cornish tin mine, shut in 1998 and has been the subject of restart efforts ever since, which are now considerably more interesting at fifty thousand dollars per ton than they were at fifteen. A tin supply chain that ran for two millennia went dormant on economics and has spent thirty years waiting for the price to come back, which is a reasonable summary of how mine supply actually responds to demand: slowly, and usually too late.
Where the tin supply chain starts, and why the map is so short
Tin’s geochemistry is unhelpful. It concentrates in association with highly evolved granites, in hydrothermal veins and greisen zones formed by late-stage magmatic fluids, and the mineral that matters is cassiterite, a dense tin oxide that is hard, chemically stubborn, and resistant to weathering. That last property produced the deposits that have supplied most of the world’s tin: when a tin-bearing granite erodes, the cassiterite survives the journey and accumulates in alluvial placers downstream, concentrated by nothing more sophisticated than moving water and its own density.
Alluvial deposits are wonderful and finite. They can be worked with modest equipment, which is why tin mining supports enormous artisanal and small-scale sectors, and why the metal shows up in the conflict minerals conversation. They also deplete, and the world has been working the good ones for a long time. What remains skews toward lower grades, deeper hard rock, and offshore dredging, all of which cost more.
The geography that results is narrow. The Southeast Asian tin belt running from Myanmar through Thailand and Malaysia into Indonesia has been the dominant global source for over a century. China is the largest producer overall. Peru and Bolivia hold significant Andean deposits. The Democratic Republic of the Congo and Rwanda supply African material. Brazil and Australia round it out. Global mine production sits somewhere near three hundred thousand tons per year against refined demand in the range of three hundred sixty to three hundred eighty thousand tons, with secondary supply covering the difference, which tells you the market is already leaning on recycling to balance.
The project pipeline is the part that gets underweighted. There is no cluster of large undeveloped tin deposits waiting for capital. Tin has not attracted major exploration budgets because it is a small market, the majors are not interested, and the juniors that do work it are financing against a metal most generalist investors could not place on the periodic table. Reserve-to-production ratios for tin are among the least comfortable in the industrial metals complex, and the USGS annual minerals accounting is the place where that arithmetic sits in public view for anyone who cares to run it.
The United States sits at the far end of this and has for a long time. There is no domestic primary tin mine of consequence, domestic supply comes from secondary recovery and imports, and import reliance runs at essentially the whole of consumption. Tin appears on the federal critical minerals list, which is accurate and also somewhat academic, since the standard remedies do not map onto a metal whose problem is that nobody has found a large new deposit in a stable jurisdiction. The tin supply chain is the case where a critical minerals designation identifies a real vulnerability and points at no available action.
Solder, and the half of demand nobody thinks about
Roughly half of global tin consumption goes into solder. Not tinplate, not bronze, not chemicals. Solder. Which means that half of the tin supply chain exists to produce the physical joints that hold electronics together, and every printed circuit board, every semiconductor package, every connector termination, and every module in a data center rack is attached to the world by an alloy that is mostly tin.
This is the least appreciated fact in the metal and it reframes everything else. Tin is not an industrial commodity that happens to have some electronics exposure. Tin is an electronics input that happens to have some industrial legacy applications. When artificial intelligence infrastructure spending accelerates, the tin market feels it, because server boards, power delivery modules, and optical transceivers all consume solder in quantities that scale with unit volume. Photovoltaic ribbon, which connects cells within a solar module, is coated in tin alloy. Electric vehicle battery packs and power electronics are full of soldered and welded interconnects.
The demand is also almost perfectly inelastic in the short run. The tin content of a circuit board is a trivial fraction of the board’s value, so a doubling of the tin price does not cause anyone to redesign anything. It causes them to pay. That combination, half of demand locked into an application where the material is essential and the cost is immaterial to the buyer, is exactly the demand profile that turns a supply disruption into a price spike rather than into demand destruction.
The comparison worth holding is with semiconductors themselves, where the value concentrates in lithography and fab capacity, and nobody thinks about the metal holding the packages down. The chip is the miracle. The solder joint is the thing that fails, and the thing whose supply is genuinely constrained.
Advanced packaging has quietly raised the stakes further. As performance gains have shifted from transistor scaling toward how chips are assembled together, the interconnect count per package has climbed steeply, and technologies built on microbumps and copper pillars with tin-based caps put enormous numbers of tin joints inside a single high-end processor or memory stack. The per-unit tin mass is still small. The reliability exposure is not, because a package with tens of thousands of solder interconnects fails if any one of them does, and the qualification burden that creates flows straight back into how conservative the industry is about changing solder chemistry.
What lead-free regulation did to the numbers
For most of the twentieth century, electronic solder was a eutectic tin-lead alloy, roughly sixty-three percent tin and thirty-seven percent lead, which melts at a convenient one hundred and eighty-three degrees Celsius, wets well, and forms reliable joints. It is one of the great pieces of applied metallurgy and it worked so well that it went essentially unchanged for decades.
Then the European Union’s Restriction of Hazardous Substances directive came into force in 2006 and lead had to go. The replacement chemistries that won are the tin-silver-copper family, commonly around ninety-six or ninety-seven percent tin with small silver and copper additions. Look at what that does to the material accounting. The industry went from an alloy that was roughly two thirds tin to an alloy that is nearly pure tin, which raised tin consumption per joint by something close to fifty percent across the entire global electronics industry, on a regulatory timeline, for reasons having nothing to do with tin.
The knock-on effects went beyond tonnage. Tin-silver-copper alloys melt near two hundred and seventeen degrees Celsius, meaning reflow ovens had to run hotter, which stressed components and boards that had been designed for a lower thermal budget, which forced requalification of laminates, packages, and assembly processes across the industry. Wetting behavior changed. Joint appearance changed enough that visual inspection criteria had to be rewritten.
This is a useful case of a demand shock arriving through regulation rather than through technology or growth, and of a policy targeting one hazardous material substantially increasing consumption of an unrelated constrained one. Nobody weighed the tin supply chain when RoHS was drafted, and there is no reason they should have, which is precisely the point. Material demand gets restructured by decisions made for entirely different reasons, and the people making those decisions are not looking at a mine plan in Myanmar.
Tin whiskers, and the failure mode that came back
Removing lead from solder and plating brought back a problem the industry had solved by accident. Pure tin surfaces spontaneously grow whiskers: single crystals of tin that extrude from the surface over months or years, reaching lengths of millimeters, thin enough to be invisible without magnification and conductive enough to short adjacent circuit traces. Lead in the alloy suppressed the growth. Take the lead out and the whiskers return.
The definitive public record on this lives at NASA’s tin whisker documentation effort, maintained out of Goddard Space Flight Center, which catalogs failures attributed to whiskers across satellites, missile programs, medical devices, nuclear plant instrumentation, and automotive electronics. A commercial satellite lost on orbit in 1998 to whisker-induced shorting from a pure tin plated relay prompted NASA to issue formal parts advisories warning the community against pure tin finishes, and the standing mitigation guidance has been to require a minimum lead content in plating for high-reliability applications.
The mechanism is a stress relief phenomenon. Compressive stress in the tin layer, generated by intermetallic compound growth at the interface with the underlying metal, by thermal cycling, or by the plating process itself, drives atomic diffusion that extrudes material at surface grain boundaries. Mitigations include nickel underlayers to block intermetallic formation, annealing to relieve stress, conformal coatings to contain whiskers mechanically, and matte rather than bright tin finishes. None of them is a guarantee, and there is still no fully accepted accelerated test that reliably predicts whisker-prone products.
The reason this belongs in a supply chain discussion is that it explains why high-reliability sectors, aerospace, defense, medical implants, and nuclear instrumentation, obtained exemptions from lead-free requirements and continue to use tin-lead. Those industries looked at a regulatory mandate, looked at their failure data, and declined. The result is a split tin supply chain on the demand side as well as the supply side: a very large lead-free commercial market and a small, stubborn, high-reliability tin-lead market that will not convert, supplied by a shrinking number of vendors willing to keep a leaded line running for customers who buy in small volumes and cannot accept the alternative. It is one of the cleanest examples available of an engineering community holding a line against a policy consensus on the strength of receipts.
Wa State, and the mine that moves the market
In August 2023, the United Wa State Army, which governs an autonomous region along the Chinese border in northern Myanmar, suspended mining at Man Maw pending what was described as a resource audit. Man Maw was Myanmar’s largest tin asset, Wa State accounts for roughly ninety percent of Myanmar’s tin ore output, and Myanmar was the third largest tin producer in the world behind China and Indonesia. The concentrate had been flowing across the border to Chinese smelters, which had built their feed assumptions around it.
The suspension removed tens of thousands of tons from the market and it has never fully reversed. License applications reopened, permits were issued, restart proceeded slowly and short of expectations, and the recovery was further hampered by rainy season logistics and by infrastructure damage from the major earthquake that struck Myanmar. Through 2025 and into 2026 the material has been trickling rather than flowing, and every incremental announcement about Wa State permits has moved the London price.
Consider what that means structurally. A non-state armed group administering a territory that appears on no map as a sovereign entity holds effective control over a supply source large enough to set the global clearing price for a metal that half the world’s electronics depend on. There is no negotiation channel that runs through a foreign ministry. There is no export licensing regime to petition. There is a decision made in a region whose contested political status is precisely what makes the decision unappealable.
Compare that to the export control episodes involving gallium and germanium, graphite, or antimony. Those were state decisions, announced through a ministry, implemented through a licensing system, subject to diplomatic engagement and eventually to negotiated suspension. Wa State is not that. It is closer to weather.
Indonesia, resource nationalism, and a thousand illegal mines
The second pressure point runs through Bangka and Belitung, the Indonesian islands whose alluvial tin fields have supplied the world for generations and which now host an enormous informal mining sector operating alongside and against the state producer. PT Timah, the state tin company, reported a production drop on the order of a third under competition from unlicensed operators working the same deposits, and the government has estimated that something like twelve thousand tons of tin leaves the country illegally each year.
President Prabowo Subianto responded with an enforcement campaign, ordering the closure of a thousand illegal mines in Sumatra, seizing hundreds of tons of metal, and arresting operators. Layered on top of that is an annual work plan permitting requirement that has repeatedly delayed legal exports while approvals grind through. Indonesian refined tin exports fell by roughly a third to around forty-six thousand tons in 2024 under the combination of tighter quotas, permit delays, and regulatory scrutiny, with official quotas subsequently raised but actual shipments dependent on execution.
Indonesia has done versions of this before. The country has periodically tightened export rules, imposed purity and traceability requirements, mandated exchange trading, and used tin policy as an instrument of industrial development, and each round has produced a supply interruption that the market absorbed and then forgot. This round is larger because the enforcement is aimed at a sector that had grown to represent a substantial fraction of actual output.
The uncomfortable part is that the crackdown is defensible on every ground except supply. Unlicensed alluvial mining in Bangka-Belitung has produced real environmental destruction, real fatalities in collapsing pits, tax evasion at scale, and a smuggling economy. Enforcing the law is the right call and it removes metal from a market that was already short. Those two statements do not conflict, and the tin supply chain has to live with both.
Bisie, M23, and the original conflict mineral
The third pressure point is Bisie in North Kivu, in the eastern Democratic Republic of the Congo, where the Mpama North operation accounts for something on the order of six percent of global tin supply. It sits in a province where the M23 rebellion has been running an offensive that forced a temporary suspension of operations in March 2025, with production subsequently resuming into a security situation that remains unresolved.
Tin is also where the modern conflict minerals framework started. The 3TG designation, covering tin, tantalum, tungsten, and gold, emerged from exactly this geography, and the reporting obligations imposed on American issuers under the Dodd-Frank framework were written because armed groups in eastern Congo were financing themselves through mineral production. The compliance apparatus that grew up around it, smelter certification, chain-of-custody documentation, and audit programs, applies to the same metal and the same region twenty years later.
Whether that apparatus accomplished what it intended is a genuinely contested question. It produced traceability systems that did not exist before and it also produced a period of de facto embargo that damaged legitimate artisanal livelihoods without dislodging the armed groups. The same debate attaches to tantalum and to the cobalt trade, and the honest position is that certification changed the paperwork more reliably than it changed the ground.
What is not contested is the supply effect. A rebel offensive in North Kivu removes a measurable percentage of world tin supply, an outbreak of disease in the surrounding region raises the risk premium, and neither event is responsive to anything a purchasing manager in Taiwan can do. Add Wa State and Bangka-Belitung and roughly a quarter of global supply sits behind political conditions that no commercial counterparty controls.
Why the tin supply chain has no refinery chokepoint
Now the part that makes tin analytically interesting. In almost every other constrained material, the analysis pivots at this point to the midstream: who refines it, where the separation capacity sits, which single plant everyone depends on. That pivot does not happen with tin, and the reason is metallurgical rather than political.
Cassiterite is tin dioxide. Reduce it with carbon at temperature and you get metal, with the main complications being the removal of iron, arsenic, lead, bismuth, and other impurities through processes that are well established and not proprietary. Electrorefining produces high-purity metal where it is needed. The capital intensity is real but not extraordinary, the process chemistry is a century old, and smelters operate in China, Indonesia, Malaysia, Peru, Bolivia, Thailand, Brazil, and Belgium among others. Nobody holds a valve.
The absence of a midstream chokepoint has a specific consequence: tin’s price responds almost purely to mine supply and to inventory, without the buffering that a concentrated processing sector sometimes provides. It also means the standard policy toolkit, which is built around reshoring processing capacity, does nothing here. Building a tin smelter in Ohio would not help, because there would be no concentrate to feed it.
That distinction deserves to be made explicitly, because industrial policy tends to reach for the same instrument regardless of the disease. Processing bottlenecks respond to capital, permits, and offtake guarantees, all of which a government can supply. Deposit scarcity in politically inaccessible jurisdictions responds to exploration over a decade, mine development over another decade, and diplomatic conditions nobody controls. Treating the tin supply chain as a processing problem produces a plant with nothing to process, which is an expensive way to discover you diagnosed the wrong constraint.
There is a further wrinkle in secondary supply. Tin recycling is meaningful, running from solder dross, tinplate scrap, and process residues, but the tin dispersed into finished electronics is in the same category as indium in a display film: milligrams per device, laminated into an assembly, uneconomic to chase. Manufacturing scrap comes back. Consumer product tin mostly does not, and the collection economics are the binding constraint rather than the metallurgy.
The other half of demand
Solder dominates, and the remainder is a spread of applications that keep tin embedded in ordinary industrial life. Tinplate, meaning steel sheet coated with a thin layer of tin, is still how a substantial share of the world’s food is packaged, exploiting tin’s corrosion resistance and non-toxicity in an application that has been running since the nineteenth century. Tin chemicals, particularly organotin compounds, serve as heat stabilizers in polyvinyl chloride, which is why a share of every vinyl window frame and pipe run traces back to a tin mine.
Float glass production uses molten tin as the bath on which glass floats to achieve flatness, an elegant process invented in the 1950s that consumes tin as a working medium rather than as a product input. Bronze and brass alloys continue in bearings, bells, marine fittings, and architectural applications. Tin compounds appear in catalysts, in ceramic glazes, and in specialty coatings.
There is also a research thread worth flagging with appropriate skepticism. Tin has attracted attention as an anode material candidate in next-generation battery chemistries, on the strength of high theoretical capacity, and tin-based perovskite formulations appear in photovoltaic research aimed at lead-free cells. Both are real research programs and neither is a commercial demand line today. Anyone building a tin demand forecast on battery anodes is doing the same thing people did with vanadium flow batteries and lithium metal anodes, which is projecting a laboratory result onto a supply chain.
Tinplate is the application worth watching for the opposite reason, because it is the one large segment that could shrink. Beverage cans went to aluminum decades ago, food packaging faces steady substitution from plastics, laminates, and aseptic cartons, and the tinplate share of consumption has been eroding for a generation. That erosion is the quiet reason the tin supply chain has stayed roughly in balance despite the solder story: a mature application has been slowly handing tonnage back to the market while electronics absorbed it. When that cushion is exhausted, and it largely is, the market loses its shock absorber.
The distribution matters for how the market behaves. Because solder is half of demand and the rest is spread across mature industrial applications with slow growth, tin has a demand base that neither collapses nor surges, sitting under an electronics-driven top half that moves with the semiconductor cycle. That produces a market that is chronically tight rather than dramatically volatile, punctuated by supply events.
What tin teaches about criticality
The value of a counterexample is that it forces you to state your rule precisely. The refinery framework is not wrong; it is a claim about where value and leverage concentrate in most materials, and it is right often enough to be the default. Tin shows what the exception looks like, and the exception has identifiable features worth learning to recognize.
The first is processing simplicity. When the midstream conversion is metallurgically straightforward and geographically distributed, no chokepoint forms there, and the constraint migrates back upstream to the deposits. The second is deposit scarcity that is genuine rather than institutional. Tin is not abundant-but-unrefined in the way the rare earths are; economically minable tin is legitimately uncommon, the good alluvial ground has been worked hard for a century, and the pipeline of replacements is thin. The third is that the deposits that remain sit in jurisdictions where the operative authority may not be a government in any conventional sense.
Put those together and you get a material whose risk profile inverts the standard model. There is no plant to build, no separation technology to master, no export licensing regime to negotiate around. There is a set of holes in the ground in Myanmar, Indonesia, and the Congo, and the question of who controls them is answered by rebel movements, enforcement campaigns, and permit queues rather than by industrial policy. The stockpiling and reshoring reflex that works reasonably well for processing bottlenecks has almost no purchase on this.
The last thing tin offers is perspective on how old all of this is. The Bronze Age palace economies that imported tin from sources they could not control, through intermediaries they could not verify, for an application they could not substitute, were running the same exposure that a contract electronics manufacturer runs today. Four thousand years of technological progress has changed what the metal is for and has not changed the shape of the problem, which is a reasonable argument that the shape of the problem is not really about technology at all.
Every material in the critical minerals conversation has a mechanism like that one, and the mechanism is almost never the headline. The 24-lecture Rare Earth Elements and Critical Minerals course runs the full slate the same way, element by element, from the magnets inside every traction motor and the heavy rare earths that make them survive heat to the samarium-cobalt alternative, the yttrium spike nobody forecast, the hafnium inside advanced logic, the noble gases you cannot drill for, the helium that keeps running short, the uranium fuel cycle, the nickel and battery metals complex, the lithium story everyone thinks they know, the platinum group catalysts, the rhenium holding up a turbine blade, the scandium nobody produces at scale, and the cobalt coming out of the Congo. The instinct it builds is the useful part: find the refinery first, and when there isn’t one, start counting the mines.