Here is a puzzle that ought to bother anyone who owns, or is thinking about owning, a rare earth stock. The material these companies dig up and process has gotten more expensive. The companies themselves have gotten cheaper. Both things are true at the same time, and according to a recent note from Canaccord’s critical minerals team, they have been true for most of 2026.
The numbers in that note are stark. Outside China, prices for neodymium and praseodymium oxide, the workhorse input for the strongest permanent magnets, are up about 23% so far this year. Canaccord says its coverage of rare earth producers is down 28% on average over the same stretch, and its developers are down about 17%. If rare earths are the story of the decade, someone forgot to tell the stock market.
A Price Floor Changes What a Price Means
To understand the gap, start with the floor. In July 2025 the Pentagon struck a deal with MP Materials that included a 10-year price floor of $110 per kilogram for its neodymium-praseodymium output. The Center for Strategic and International Studies, in an April brief by Gracelin Baskaran and Meredith Schwartz, describes the package as $400 million in equity that made the Defense Department MP’s largest shareholder, a $150 million loan, and a 10-year offtake commitment for 100% of the output of a planned second magnet plant. Canaccord notes that the same $110 level now applies to Lynas and USA Rare Earth as well.
Now look at where the market sits. Canaccord reports that Chinese and Rotterdam prices for neodymium and praseodymium have converged, and the Western price is above $110 per kilogram. Chinese prices, by its account, swung from roughly $175 down to about $50 and back, and have settled near $110 over the past ten months. So the market price has drifted up to the level the government guaranteed. That is good news for the fundamentals. It is also a quiet reminder that a floor only matters when prices fall below it.
Is the floor now just a number on a chart? Maybe. A guarantee is most valuable when the market is cheap, and when the market catches up, investors start asking what else the company has going for it. That is a fair question, and it is one the equity market seems to be asking loudly.
Where the Real Squeeze Is
The most dramatic prices are not in the headline magnet metals. They are in the heavy rare earths, the obscure ones that most retail investors have never had a reason to learn. Canaccord’s figures put dysprosium in Rotterdam at about $1,074 per kilogram against $227 in China. Terbium is about $4,594 against $983. Yttrium, used in aerospace coatings and other demanding applications, is the extreme case. A Financial Times report this month, citing Argus assessments, put yttrium oxide at roughly $9 per kilogram inside China and about $1,150 outside it, which is about 179 times the price outside China two years ago.
Those are not typos, and they are not the result of a sudden jump in demand. They reflect access. The CSIS brief traces it to China’s April 2025 export restrictions on heavy rare earths and magnets. Yttrium exports from China to the United States totaled 17 tons in the eight months between April and December 2025, compared with 333 tons in the eight months before the restrictions began. Aerospace manufacturers, CSIS says, have reported shortages and rationing. When a market is cut off from its main supplier, prices stop reflecting cost and start reflecting scarcity.
Here is the catch for investors. A shortage of yttrium in a Western spot market does not automatically translate into earnings for a company that does not yet produce yttrium at scale. Canaccord itself forecasts that its yttrium price will fall from about $740 per kilogram this year to about $100 by 2030, as supply outside China catches up. The scarcity premium, in other words, is something the analysts expect to fade.
So Why Are the Stocks Falling?
Canaccord’s own explanation leans on “global macroeconomic pressures” rather than on rare earths: a 10-year Treasury yield that has hovered around 5%, a wobbling technology sector, and a fading appetite for speculative assets. That matters because many rare earth names are, bluntly, bets on future capacity. When the cost of money rises, a dollar of earnings promised in 2029 is worth less today than it was a year ago.
The price action shows it. Canaccord’s chart shows the REMX rare earth ETF peaking in May above $100 and slipping to the high $50s since. MP Materials traded at $40.17 when the note came out, about half of Canaccord’s $82 target. Canaccord says producers are trading at roughly 0.57 times net asset value and developers at about 0.47 times. Analysts publish targets constantly and they are often wrong, so treat those as one firm’s opinion, not a forecast of anything. The valuation discount itself, though, is what the market is actually saying about risk.
Politics has not helped. On June 22, Bloomberg reported, and Mining Weekly republished, that China’s Commerce Ministry added MP Materials and USA Rare Earth to its export control list, along with eight other American companies. Chinese exporters are barred from shipping dual use items to the listed firms. Both companies have been expanding capacity since the 2025 restrictions. Bloomberg described the curbs as mostly symbolic, since both companies say they have largely cut off supplies of equipment and materials from China.
The Test Is Execution, Not Spending
Washington has certainly done its part on paper. CSIS counts more than $7.3 billion committed across five federal departments and agencies, plus another $2 billion for the National Defense Stockpile, along with equity stakes, loans, and offtake agreements around the world. Yet the same brief points out how far there is to go. U.S. production at Mountain Pass reached 8,900 tons in 2025, covering only about a third of American consumption. The other 18,100 tons were imported, and 71% of that came from China.
That is why the authors argue that execution, measured in actual production and private investment, is now the main test. It is hard to disagree. A subsidy can build a plant. It cannot, by itself, deliver a skilled workforce, a reliable heavy rare earth separation circuit, or the metallization capacity downstream. Canaccord estimates China controls more than 90% of global metallization capacity this year, and that strip caster lead times can run up to two years. Money is fast. Machinery is slow.
There are also dates that investors can circle. The current two month extension of the truce between Washington and Beijing expires on January 10, 2027, according to ZeroHedge’s write-up of the note. Separately, CSIS notes that, starting January 1, 2027, the Pentagon will bar rare earth materials and magnets sourced from China, Russia, Iran, and North Korea, and questions whether Western capacity can be ready in time. Those two deadlines, nine days apart, will tell the market a great deal about whether the policy scaffolding is holding.
What a Commodity Rally Does and Does Not Tell You
The easy lesson from this year is that a rising commodity price and a rising stock price are different things, and that confusing them is expensive. A miner earns money from what it sells, at the price it gets, minus what it costs to produce, financed at whatever interest rates prevail. Change any of those and the picture changes. Rare earths in 2026 are a case where the first input looks strong and the last one looks hostile.
The longer view, to be fair, is not gloomy. Canaccord projects magnet demand growing at about 7.3% a year to roughly 365,000 tons by 2035, driven by electric vehicles and, more speculatively, humanoid robots. It sees neodymium and praseodymium in deficit through 2028. It also cautions that Western premiums may not last and that some magnet makers are already moving toward cheaper ferrite magnets. That caveat deserves as much weight as the bullish demand forecast, because substitution is how high prices end.
Notice what is missing from this picture: certainty. The market is weighing a real supply shortage, a government that has promised to back its domestic producers, a rival that can tighten or loosen the tap at will, and a bond market that has made patience costly. Those forces do not resolve neatly. They get repriced, one headline at a time. Anyone studying this sector would do well to read the primary documents, check how each company is actually financed, and keep track of the two January dates. The gap between the price of the metal and the price of the stock is real, and it is the part of this story that rewards close reading.
This article is written for educational and informational purposes only and does not constitute financial or legal advice. The views and analytical frameworks presented draw on publicly available information and reported commentary from industry participants. Readers are encouraged to consult primary sources and form their own informed views on these complex topics.
