The United States needs Canada if it wants to loosen China’s hold on critical minerals. That is not a slogan from a mining booth. It is a sentence spoken on September 28, 2026, in Colorado Springs, by Edward Fishman, a senior fellow at the Council on Foreign Relations and a former U.S. sanctions official. He told the Mining Forum Americas that Washington has money and political will, and still cannot do this job alone. “We, as in the United States, cannot do it without the Canadians,” he said. He added the warning that makes the week ugly. If Washington keeps antagonizing Canada, getting out of the China dependence is “virtually impossible.”
The same week, the metals that already move across that border were not a fairy tale. Natural Resources Canada’s 2025 trade figures, as cited by a September policy essay, put Canadian critical-minerals exports at about C$49.4 billion. Aluminum, copper, potash, nickel, uranium, and zinc led the list. The United States took just over half of Canada’s mineral and metal exports. Those tons are the help that already exists. A speech about future mines is a different help, and a slower one.
This piece has one theme. The grip is the plant, not the pit. China critical minerals power sits in refining, separation, and the factory steps after the mine. Canada can reduce U.S. dependence where a metal already crosses the border, or where a refinery can be built and kept full. It cannot do it by listing every deposit. Which Canadian minerals could benefit is therefore a short list with a test. Nickel, copper, uranium, and the processing of lithium, graphite, and rare earths can matter. A junior hole with a flag on it does not, until someone builds the plant the flag is supposed to replace.
This is news and context. It is not investment advice. It is not a recommendation to buy Canadian mining stocks, Canadian copper stocks, Canadian lithium stocks, Canadian rare earth stocks, or any other security. A politician’s need is not a bid. A bid is a contract, a plant, and a price that still works when the speech is over.
One theme
Say the theme once, then make every mineral pass it. Can Canada reduce U.S. dependence on China for critical minerals? Only in the steps where Canada already smelts, refines, or can sell a finished material the United States does not make. A new pit that ships concentrate into a Chinese plant does not break the grip. It feeds the grip. The patriotic reading and the commercial reading split at the loading dock.
Which Canadian minerals are most strategic to the U.S.? The ones Washington already carved out of a fight, and the ones it has no domestic plant to replace. In the summer of 2026, a new tariff round on Canadian goods exempted energy, potash, fish, critical minerals, and goods already covered by other duties, according to accounts of the White House measure. The carve-out is a confession. You do not tariff the rock you cannot live without, even while you tariff the country. Strategy is that confession, not a conference slide.
What Fishman actually said
Keep the quote tied to the room. Fishman left government in 2017. He helped design sanctions on Russia and Iran. He now directs a geoeconomics center at the Council on Foreign Relations. On Monday he said the United States needs Canadian and Australian mining expertise, not just its own cash. Antagonizing Canada makes the exit from China “virtually impossible.” He also said eliminating dependence is a different goal from cutting it to a level a country can live with. Governments should set realistic targets. Miners should tell them what the targets do to real operations.
He was not romantic about the alliance. He said a Canada with a bit more independence “may actually be a better ally.” He said competition over supply chains will outlast one White House, because administration after administration has reached for economic pressure. Mines outlast elections. That is the useful part for a reader. A thesis that only works if this year’s tariff mood vanishes is not a thesis about a mine. A mine takes longer than a mood.
He also separated two demands that speeches mash together. Artificial intelligence and electrification support copper demand. Worry about financial security supports central-bank gold buying. Those are not critical-minerals sentences, but they keep the forum honest. Not every metal in a Canadian portfolio is a China-exit metal. Copper can be both a grid metal and a China-refining story. Gold is a different trade. Do not let one microphone turn them into one ticker.
The grip is a percentage at the plant
China rare earth dominance is the cleanest picture, and it is a picture of furnaces more than shovels. The International Energy Agency, as cited by Mining.com, put China’s share of refined rare earths used in magnets at 91 percent in 2024, against about 60 percent of mined output. The mine is concentrated. The refinery is tighter. A later IEA reading, reported by Reuters in September 2026, said investment in refining in the United States and Malaysia had pulled that refined share down from over 90 percent in 2023 to about 85 percent in 2025. Even if every planned rare-earth refinery outside China arrives on time, the agency still sees China’s share around 70 to 73 percent in 2035. A victory that leaves one country with seven-tenths of the plant is a smaller grip. It is not a broken one.
The pattern repeats in metals Canada talks about more often. Published accounts of the IEA’s 2026 outlook put China as the leading refiner of 19 of 20 strategic minerals, with an average share around 70 percent. For gallium, graphite, manganese, and rare earths, the top refiner’s share is above 90 percent. Lithium refining is in a range that published notes put from the mid-60s to the mid-70s. Cobalt refining is about three-quarters, though the ore is mostly Congolese. Battery-grade graphite stays Chinese even if more graphite is mined elsewhere. One Reuters summary of the same agency said China in 2025 processed 70 to 95 percent of lithium, cobalt, phosphate, manganese, and graphite, made about 98 percent of lithium-iron-phosphate cathode material, and about 80 percent of battery cells. Those are not mining statistics. They are factory statistics.
Copper is the mineral people think is diversified because the pits are. Chile, Congo, and Peru dig a lot of it. China’s refined share still sits in the mid-40s in IEA-based accounts, and the agency has said China accounted for more than three-quarters of the growth in global refined supply of these materials from 2023 to 2025. North America has only three primary copper refineries left, two in the United States and one in Canada, according to the same family of reporting. A continent that cannot refine its own copper is not independent because a Canadian pit is busy. Copper mining Canada can be real and still end at someone else’s tankhouse.
Export controls showed what the percentages mean in a factory. China tightened magnet rare earths in April 2025. The IEA said the controls disrupted magnet supply and forced some automakers to cut or briefly stop production. In the fall of 2025 Beijing widened controls on graphite and on high-performance LFP cathode material, and floated a regime that would have touched goods with only a trace of Chinese rare earths. A later trade report said that wider regime was paused for a year after an October 2025 meeting between President Trump and President Xi. A pause is not a repeal. A pause is a clock. The critical minerals supply chain still runs through a gate one capital can close.
What Canada already is, and what it is not
Canada critical minerals policy is not new in 2026, and it is not a wish list invented after Fishman’s speech. The federal strategy dates to 2022. A government progress note says that by 2024, domestic output had risen more than 10 percent for nine minerals: aluminum, graphite, lithium, magnesium, molybdenum, niobium, platinum-group metals, scandium, and uranium. Fourteen projects had moved through federal impact assessment. Sixty-eight technical reports had been filed, including 20 for lithium, 10 for copper, and 7 for nickel. Ottawa has also talked about offtake deals, including scandium with Rio Tinto and graphite with Nouveau Monde, and a First and Last Mile Fund of C$1.5 billion from 2026 to 2030 on top of earlier infrastructure money. In October 2025 it announced a first wave of partnerships it said would help unlock about C$6.4 billion of projects. Those are processes and cheques. They are not tons in a U.S. magnet plant.
The tons that exist are less fashionable and more important. Critical-mineral export value is dominated by aluminum, copper, potash, nickel, uranium, and zinc. The United States is the buyer for just over half of Canada’s mineral and metal exports. That is integration. It is also concentration of another kind. Canada’s risk is not only China’s refinery. It is a customer that takes half the cargo and is willing to fight the shipper. Fishman’s point is that the fight and the cargo cannot both be maximized.
Aluminum shows the integration in a way rare earths do not. Canadian primary output was about 3.3 million tonnes in 2024, from nine Quebec smelters and one at Kitimat, in the policy essay’s account of government figures. U.S. smelters fell from about 30 in 1980 to a handful still running. The essay says Canada supplies more than half of U.S. aluminum use, and that Canadian aluminum sales into the United States fell by about half between July 2025 and July 2026. Aluminum is not the mineral in Fishman’s rare-earth paragraph. It is the mineral that proves a cross-border metal system can be damaged by policy faster than a new mine can be built. Remember that speed when someone says a lithium project will “break the grip” by 2028.
Potash is the other confession. More than 80 percent of U.S. potash imports, and by some counts closer to 90, come from Saskatchewan, in reporting that draws on U.S. Geological Survey data. Potash is not a China-refining story. It is a Canada-supply story. It was also on the tariff exemption list. Washington already knows which Canadian tons it cannot punish. The strategic list starts there, with the carve-out, not with the explorer who has not poured concrete.
Which Canadian minerals are most strategic to the U.S.
Which Canadian minerals are most strategic to the U.S.? Rank them by how hard they are to replace, not by how often they appear in a deck.
Nickel is first among the metals that match this week’s headline. Nickel mining Canada is not a concept. Sudbury and Voisey’s Bay already feed cathodes and the specialty steels used in defense. The United States closed its last primary nickel refinery in 1985. Ore from the remaining large U.S. nickel mine, in Michigan, is sent to Ontario to be smelted. That is dependence in the honest direction. The United States depends on a Canadian plant for a metal it still digs. A fight that chills that plant does not create an American nickel industry. It creates a gap China and Indonesia are happy to price. If any Canadian mineral “benefits” from a serious U.S. effort, nickel benefits as a flow that must not be broken, and as extra refining capacity if someone pays to expand it.
Copper is next, and it is bigger than a battery story. Grids, munitions, electronics, and data centers all eat it. Copper mining Canada produces concentrate and some refined metal. The bottleneck is the refinery count. Three primary copper refineries in all of North America is not a system with slack. Canadian copper stocks, as a category, are a claim on that scarcity only if the company sells copper into a North American circuit and can say what it costs. A copper explorer with no road and no smelter slot is a claim on geology. Geology is not the mid-40s percent of refining that sits in China. Do not pay the geology price for the refining problem.
Uranium is strategic for a reason that has nothing to do with electric cars. Canadian supply is a large slice of the fuel U.S. reactors use. Published accounts put the share near a third of U.S. reactor fuel and a large fraction of Canada’s uranium exports. The number moves with the year and the definition. The direction does not. Nuclear fuel is a critical mineral in the security sense even when it is bored by the battery conference. A benefit here is a contract with a utility, not a spike in a junior’s volume.
Lithium is the crowd’s answer, and it is only half right. Lithium mining Canada is no longer zero. Output was among the nine minerals that grew more than 10 percent by 2024. Twenty feasibility-style reports in a few years say the rocks are being measured. Canadian lithium still has to be converted. China refines most of the world’s lithium chemicals and makes nearly all of the LFP cathode. A Canadian spodumene mine that exports concentrate to a Chinese converter has not loosened the grip. It has rented it. Canadian lithium stocks benefit if, and only if, conversion, cathode, or a binding offtake sits on this continent. The rock alone is a press release.
Rare earths are the mineral in the headline and the weakest Canadian ton. Canadian rare earths exist as deposits, including work in Quebec and the Northwest Territories that promoters cite. They do not exist as 91 percent of magnet supply. China rare earth dominance is separation, metal, alloy, and magnet. The IEA’s own arithmetic says non-Chinese refining can rise and still leave China with about seven-tenths of the plant in 2035. A Canadian rare earth mine that stops at concentrate is a customer of the grip. Canadian rare earth stocks are a bet on separation and magnets, or they are a bet on a story. Those are different bets. The story is cheaper to tell and more expensive to hold.
Graphite belongs beside the rare earths because the offtake is realer than the speech. Ottawa has pointed to a graphite offtake alongside the scandium one. Battery-grade graphite is one of the minerals where China’s share stays near nine-tenths even under generous mine forecasts. A North American anode plant with a Canadian feed is a chip out of that share. A graphite occurrence without an anode customer is a map. Scandium is tiny and real in the same way. Small metals with a named buyer beat large metals with a dream.
Can Canada reduce U.S. dependence on China for critical minerals?
Can Canada reduce U.S. dependence on China for critical minerals? Yes, at the margin, in nickel, copper refining, uranium, aluminum, potash, and a handful of processing projects. No, if the sentence means “replace China’s plants before the next election.” The IEA’s 2035 rare-earth case is the scale check. Years of announced projects still leave China in the lead. Canada is a serious mining country. It is not a second China waiting behind a ribbon.
The margin is still worth money, and it is worth policy. Dependence falls when a magnet, a cathode, a tankhouse, or a nickel smelter on this continent runs at capacity and has a second customer so one cancelled contract does not kill it. Fishman asked governments to pick a level of dependence they can live with, not a fantasy of zero. Zero is how projects get approved on slides and stalled in the real cost of power, permits, and skills. A livable share is how a plant gets built.
There is a second way Canada reduces dependence, and it is dull. Do not break the flows that already exist. Michigan nickel ore going to Ontario is a working chain. Canadian aluminum going into U.S. cars and fighter parts is a working chain. Potash going to U.S. farms is a working chain. A trade war that exempts the mineral and still poisons the permit, the rail, and the mood can shrink those flows without a single Chinese decree. The Bank of Canada figures cited on aluminum, a drop of about half in a year, are the warning. You can lose a secure supply to your own tariff faster than a rival can embargo it.
Fishman’s other line belongs here. Canada selling more to Europe, Japan, and Korea does not have to weaken the U.S. alliance. A seller with more than one buyer is harder to bully and harder to abandon. That is good for Canadian critical minerals and awkward for a Washington that wants exclusive friendship and the freedom to antagonize. He thinks a slightly more independent Canada may be a better ally. The investment version is plainer. A mine with only one political customer is a political mine. Political mines live and die by elections.
The fight and the carve-out
The trade facts of 2026 are the part a minerals bull will try to skip. On July 1, 2026, the scheduled six-year review of the North American trade pact produced no renewal. The agreement slid into annual reviews. Weeks later, accounts of a White House measure described a new 50 percent tariff on a wide set of Canadian goods, with effect in late August, and with exemptions for energy, potash, fish, critical minerals, and items already covered by other tariff regimes. Read the exemption twice. Critical minerals were named as too sensitive to tax. The relationship around them was not.
That split is the whole investment problem. Canadian mining companies can be inside the carve-out and still live in a country whose other exports are being hit. Workers, power contracts, Indigenous agreements, and provincial politics do not sit inside a customs code. A smelter needs a calm border even if the concentrate is exempt. Fishman said antagonizing Canada makes the China exit nearly impossible. The tariff list says Washington heard the mineral point and not the country point. Hearing half is how alliances fray while white papers stay friendly.
For a reader of stocks, the carve-out is not a buy signal. It is a map of what officials think they cannot replace this year. Potash, uranium fuel, nickel units, copper units, and aluminum metal are on that map because the tons exist. Lithium chemicals and rare-earth magnets are mostly not on it, because the tons do not exist here in size. The minerals that could benefit first are the ones already exempted by necessity. The minerals that could benefit later are the ones that earn an offtake and a plant. The minerals that benefit only in a headline are the ones that match a keyword and a drill hole.
What a plant has to clear
A refinery fails for ordinary reasons. Power is expensive. The permit is slow. The skill to separate rare earths is held by few teams, many of them in China. The customer wants a discount to the Chinese price, and the Chinese price can be cut. A plant that only works when the rival is polite will close in the first price war. Ottawa’s funds and offtakes are an answer to that, not a proof it is solved. A cheque that covers the first miles of road does not cover a decade of below-cost competition.
Nickel shows the skill problem in reverse. Canada already has the smelter the United States gave up in 1985. Expanding a thing that runs is easier than inventing a thing that does not. It is still not easy. Nickel’s other giant is Indonesia, which now mines more than half of world supply in IEA-based tables, with China a major refiner of the rest. A Canadian nickel unit sold into a battery plant has to beat an Indonesian unit on reliability, carbon rules, and politics, not on a speech in Colorado. Reliability is a contract. Politics is this year’s tariff.
Lithium shows the opposite problem. The growth in Canadian reports is real. The chemical plant is the prize, and China already holds it. Hard-rock mines in several provinces can be good mines and still be captive feed. Brines have their own clock, often longer. Canadian lithium becomes strategic when a cathode maker on this continent is short of hydroxide and will sign for years. Until that signature, the mineral is promising. Promising is not strategic. Strategic is the signature.
Rare earths show the hardest problem. Mined share and refined share are different numbers, 60 and 91 in the 2024 magnet figures, and the gap is the business. Separation is a chemical plant with nasty waste and a customer who needs a specific oxide, then a metal, then an alloy, then a magnet. Stopping at the first step and calling it independence is how the last cycle disappointed. If Canadian rare earths benefit, they benefit at the step the IEA says is still scarce outside China in 2035. Count magnets, not drill holes.
Stocks are not the mineral
Canadian mining stocks are a loud version of a slow industry. They move when a speech moves, and they need a plant to justify the move. Canadian mining companies that already produce nickel, copper, uranium, or aluminum are exposed to the flows Fishman says Washington cannot replace. They are also exposed to costs, to the dollar, and to a customer that is picking a fight. A producer can be strategic and still be a bad purchase at the wrong price. This article will not name a purchase.
Critical mineral companies farther down the chain deserve a colder eye. Canadian copper stocks with a producing mine and a smelter slot are not the same object as a copper story on the venture exchange. Canadian lithium stocks with a feasibility study are not the same object as a lithium mine with a chemical plant. Canadian rare earth stocks with a magnet ambition are not the same object as a rare-earth intersection in a news release. The words on the ticker can match. The cash flows do not.
The test is short, and it is public. Does the company sell a mineral the United States already imports from Canada in size? Does it refine, or does it only dig? Is there an offtake with a name, a volume, and a term? What happens to the project if the Chinese price is cut for two years? What happens if the border fight lasts longer than the exemption? A company that cannot answer in a filing is not a way to break a grip. It is a way to rent a headline. Headlines expire. Plants, if they are real, do not.
Three ways this can age
These are paths, not predictions.
One. The relationship is repaired enough to build. The tariff fight cools. Nickel, copper, and uranium contracts lengthen. One or two lithium-chemical and graphite-anode plants in Canada or the United States actually run, fed in part by Canadian rock. Rare-earth separation grows in North America and in allies, and China’s refined share slips toward the IEA’s 70 percent case rather than back to 90. In this path Canada reduced dependence at the margin. Fishman was right that it could not be done without Canada. He was also right that zero was the wrong goal. Investors who needed a clean break with China by 2027 are disappointed. The plants are not.
Two. The fight continues, the carve-out stays, and nothing new gets built. Existing nickel, potash, uranium, aluminum, and copper flows survive because they are hard to replace. New refineries stall because nobody will finance a plant into a moody border. China’s share in magnets and graphite barely moves. In this path the strategic minerals are exactly the ones already shipping. The minerals that “could benefit” on a slide do not. Canadian mining stocks tied to existing tons muddle through. Stocks tied to future plants give back the speech.
Three. The antagonism wins. Exemptions narrow. Aluminum’s drop becomes a template for other metals. Ontario’s nickel plants and the lone Canadian copper refinery run below plan. The United States discovers that money and will do not smelt nickel by themselves, which is what Fishman said in the first hour. China does not have to seize a mine. It has to watch an ally quarrel. In this path the grip tightens by neglect. The minerals that benefit are not Canadian. They are whoever still ships.
How to read the next speech
This is a filter, not a plan.
Ask whether the sentence is about a plant or a pit. If a minister names a mineral and cannot name a refinery, a smelter, or an offtake, the sentence is a pit sentence. Pit sentences do not move China’s 85 percent. Plant sentences might, over a decade, and only if the plant stays open when prices fall.
Ask whether the mineral is already in the carve-out. Nickel, copper, uranium, potash, and aluminum are strategic because the tons are awkward to lose. Lithium and rare earths become strategic when the tons exist on this continent in chemical or magnet form. Until then they are candidates. Candidates are not the same word as strategic, whatever the headline needs.
Ask who the customer is. A mine with a U.S. buyer, a Japanese buyer, and a European buyer matches Fishman’s hope that a more independent Canada is a sturdier ally. A mine with one political sponsor matches the risk that the sponsor changes his mind in July. The critical minerals supply chain is a set of contracts. Count them.
Ask what China would still control if the project works. If the answer is the chemical, the magnet, or the cathode, the project has not broken the grip. It has moved one step. One step is allowed. Calling it independence is not.
What this does not mean
It does not mean the United States will treat Canada as a partner next quarter. Fishman said the opposite risk in plain words. Hostility makes the exit from China nearly impossible. The tariff record of 2026 is hostility with a mineral footnote. Footnotes do not run smelters. People do.
It does not mean every Canadian mineral benefits. Potash benefits from being unavoidable, and it is not a China story. Nickel and copper benefit if the plants stay busy. Uranium benefits if utilities sign. Lithium and rare earths benefit only at the processing step. A lot of other names benefit only until the next risk-off week. “Could benefit” is a filter. It is not a basket.
It does not mean China is about to lose the refinery. Eighty-five percent of magnet rare earths, a path to about 70 percent in 2035, mid-40s in refined copper, and near-total shares in battery graphite and LFP cathodes are not a crumbling monopoly. They are a monopoly that has been nicked. Nicked is the word. Broken is the fundraising word. Use nicked.
It does not mean you should buy a miner because a CFR fellow said Washington needs Ottawa. He also said targets should be realistic, mines outlast elections, and blocs can raise costs and inflation. Those sentences are the risk. The need is the opportunity only for projects that survive the risk. This desk does not pick them for you.
It does not mean aluminum, potash, and uranium are side issues because the keywords say lithium and rare earths. The side issues are the tons. The keywords are often the hopes. A grown-up reading of Canadian critical minerals starts with the export table and only then opens the drill map.
The close
U.S. needs Canada to break China’s critical-minerals grip. Edward Fishman said the blunt version on September 28, 2026. Without the Canadians, the United States cannot do it. Antagonizing Canada makes the exit virtually impossible. He also said the right goal is a dependence you can live with, not a fantasy of none, and that a slightly more independent Canada may be a better ally. The IEA’s numbers say why the fantasy fails. China refined about 91 percent of magnet rare earths in 2024, still about 85 percent in 2025, and is still seen near 70 percent in 2035 if the rest of the world builds what it has promised. The same maps show Chinese plants dominant in graphite, lithium chemicals, cobalt refining, and battery cathodes, and large in copper refining, while North American copper refineries can be counted on one hand.
Which Canadian minerals could benefit? The ones already in the pipe, and the processing steps that do not yet exist. Nickel, because the United States has not run a primary nickel refinery since 1985 and still sends Michigan ore to Ontario. Copper, because three tankhouses cannot be a continent’s plan. Uranium and potash and aluminum, because the United States already depends on them and, in 2026, carved related minerals out of its own tariff. Lithium and rare earths and graphite, only where a chemical, an anode, or a magnet is contracted on this continent. A pit that feeds a Chinese plant does not break a grip. It extends one.
Can Canada reduce U.S. dependence on China for critical minerals? At the margin, yes, if the border stays calm enough to run the plants that exist and to finish a few that do not. As a substitute for China’s refineries, no. The theme is that margin. The grip is the plant. Canada’s help is the plant it already runs, and the plant it might run, not the list of minerals a headline can name. Fishman asked for a partner. A partner is a smelter that stays open when the speeches stop.
Important information
This article is for information and education only. It is not investment advice and not a recommendation to buy or sell any security, including Canadian mining stocks, Canadian copper stocks, Canadian lithium stocks, Canadian rare earth stocks, or any miner, refiner, or royalty. Critical mineral companies face price, permit, construction, political, and dilution risk. Export rules can change. Tariffs can widen. A strategic mineral can still be a poor investment. Past policy and past prices do not indicate future results.
Edward Fishman’s remarks are from his September 28, 2026, appearance at the Mining Forum Americas in Colorado Springs, as reported by Mining.com, including the quoted lines on Canada and on dependence. The 91 percent refined and about 60 percent mined rare-earth magnet figures for 2024 are the International Energy Agency figures cited in that report. The move from over 90 percent to about 85 percent refined share by 2025, the 70 to 73 percent 2035 case, the 70 to 95 percent processing range, the LFP and battery-cell shares, and the April and October 2025 export-control descriptions follow Reuters and other accounts of IEA work and should be read in the agency’s outlook. Copper-refinery counts, nickel-refinery history, Saskatchewan potash shares, aluminum tonnages, the C$49.4 billion export figure, and the U.S. share of Canadian mineral exports follow a September 2026 policy essay citing Natural Resources Canada, the U.S. Geological Survey, the IEA, and industry groups. Tariff timing and exemptions follow accounts of 2026 White House measures. Canada’s production gains, report counts, offtakes, the C$1.5 billion fund, and the October 2025 project figure are from Government of Canada strategy updates and releases. Shares and dates move. This article does not consider any person’s goals or finances.