← All thesesUSARUnattractiveMedium convictionTuesday 17 November 2026עברית

USA Rare Earth

This is the archived full research. Read the story

A Brazilian heavy rare earth mine with a government-backed buyer, priced before the new shares its build still needs.

Our viewUnattractive Medium conviction
The scores
The profile · 1-4
Growth3/4 ↑

Revenue is tiny today, but the mine began selling in November with 325 tonnes guided for the quarter and analysts expect about $611M in 2027.

Business quality2/4 ↑

A scarce heavy rare earth deposit sold at government floors, but no margin yet, a 5.25% royalty first, and MP and Lynas as rivals.

Financial position2/4 ↓

$1.09B of cash against $667.8M of loan and royalty, but operations use over $250M a year and the build needs about $1.55B more equity.

Execution record2/4 →

Bought a running mine and started a magnet line, but Round Top, Stillwater and Less Common Metals targets slipped and the chief executive changed.

Per share capture1/4 ↓

Shares rose 2.5 times in 2026, about 58% more are needed at today's price, and the royalty and buyer take part of the upside first.

Potential3/4 →

Magnet rare earth oxides are a $7.8B market forecast to grow fivefold, and Stage 2 alone would be about 1.6% of world mine output.

Risk profile1/4 →

One buyer, prices set by China's licensing, a Brazilian licence that expires in 2028, environmental notices and a buyer's option on the mine.

Valuation2/4 ↑

13.2 times Stage 1's sales at the floors and 5.8 times at the basket, before about 58% more shares; cheap only if record prices hold.

Evidence quality2/4 →

Cash, debt and the offtake terms are filed; tonnes, the basket and the magnet letters of intent rest on management's words.

The bet · 2-3 years
Speculation2.2×

If it works 2.2× · if it fails you keep about 16%.

The Verdict

We rate USA Rare Earth Unattractive. The business is weak for now: it owns a scarce heavy rare earth mine in Brazil whose first stage a government-backed buyer takes at floor prices, but it lost $141.4M at the operating line in the third quarter on $3.0M of sales, and its targets slipped this year. The price is full: about 13.2 times the first stage's sales at the floors, before government agreements force new shares that at today's price would add about 58% to the count. As a bet it is speculative and fairly priced. Record prices outside China holding would argue against us.

The long caseChina keeps licensing heavy rare earths while Pela Ema reaches 6,400 tonnes a year, so a government-backed buyer pays floors well above Chinese prices for every tonne.

How we weighed it

The price against our value

Against it

No fair value is drawn, since the mine has only just begun to ship. At the guaranteed floors the price is a high multiple of the first stage's sales, before the new shares.

The business

Cuts both ways

Weak

A scarce deposit and a floor-priced buyer, but no margin earned yet, a royalty taken first and the buyer sharing most of any upside.

Growth against what the price assumes

In its favour

The mine's ramp would lift sales many times over, and analysts expect hundreds of millions of dollars next year.

What could go wrong

Against it

Prices set by China's licensing, a Brazilian licence to renew, environmental notices, a buyer's option on the mine and equity milestones that force new shares.

Against competitors

Cuts both ways

MP holds its own government floor and Lynas already separates heavy elements, but heavy rare earths outside China remain scarce.

How its group is behaving

Cuts both ways

Government support for supply outside China is strong, but the listed group was priced on export-control headlines and fell hard this autumn.

The latest news and report

Against it

The last report missed on losses, revenue fell from the quarter before, targets slipped and the chief executive changed.

The evidence under this - 7 claims
C09facthigh confidence
Pela Ema sells 100% of its Phase 1 output to US SIIE, LLC under a take-or-pay offtake expected to run about 15 years, at floors of $2,050 a kilogram for terbium, $575 for dysprosium and $110 for neodymium and praseodymium, plus 70% of any market value above the floors; carbonate is priced at the floor less a fixed separation cost.
against this

The floors pay only for four elements; yttrium and the other heavy rare earths in the carbonate earn nothing until the company separates them itself.

current state · 2026-09-30written 2026-11-17
C25facthigh confidence
China added export controls on samarium, gadolinium, terbium, dysprosium, lutetium, scandium and yttrium in April 2025, and those controls remained in effect, according to the US Geological Survey.
against this

China began issuing general export licences to selected exporters, so the controls can loosen without being lifted.

current state · 2026-02-01written 2026-11-17
C03facthigh confidence
The Q3 2026 operating loss was $141.4M; selling and administration rose to $103.7M, with consulting and legal costs up $68.0M and $8.6M mainly on acquisitions. Over nine months the operating loss was $224.4M.
against this

Much of the overhead rise is deal costs the finance chief calls one-time.

historical · Q3 2026 and nine monthswritten 2026-11-17
C22facthigh confidence
In March 2026 the company targeted a Round Top prefeasibility study by the end of Q3 2026, Stillwater's second line in Q1 2027 and Less Common Metals at 3,000 tonnes by end-2026; by November these had become a technical report in Q3 2027, the second line in Q2 2027 and new furnaces in Q3 2027.
against this

The chief executive says the Round Top delay buys more drilling and engineering before committing capital, which lowers building risk.

historical · March to November 2026written 2026-11-17
C32inferencemedium confidence
Company value is about $2,264M: $2,676M of market value less $1,090.9M of cash, plus the $305.6M loan, the $362.2M royalty obligation and $11.1M of preferred stock. At Stage 1's 4,000 tonnes the oxide sells for about $388M a year at the fourth-quarter market basket, 5.8 times, and $172M at the floors, 13.2 times.
derivedC31, C05 and C17 for the bridge; C13 and C12 for 4,000 tonnes at $97 and $43 a kilogram.
against this

The vendor's company value of about $164M rests on a share count of 132.6M, before the placement and the Serra Verde shares, and is not used.

current state · 2026-11-16written 2026-11-17
C40inferencemedium confidence
Raising the required $1.55B or more by the end of 2027 at the 16 November close would take about 217M new shares, 58% more than are outstanding.
derivedC15 and C31: $375M plus at least the $300M cash paid for Serra Verde, plus $875M, divided by $7.13; against C17's 375.3M shares.
against this

Up to $300M may be convertible notes, the milestones may be renegotiated, and the cash on hand covers the next year's spending.

current state · 2026-11-16written 2026-11-17
C44management claimmedium confidence
Management says dysprosium oxide outside China sells for 14 to 16 times the Chinese price and terbium for 6 to 8 times.
against this

Prices outside China are thinly traded and fall if Beijing grants more export licences.

current state · 2026-11-09written 2026-11-17
01

The company

USA Rare Earth, based in Oklahoma, is building a chain from mine to magnet outside China. Rare earths are metals whose magnets drive electric motors, wind turbines and guided weapons; the heavy ones, dysprosium and terbium, keep those magnets working when hot.

Its Pela Ema mine in Brazil washes rare earths out of clay. Less Common Metals in England turns rare earth oxide into metal and alloy, and a plant in Stillwater, Oklahoma, makes magnets.

The evidence under this - 3 claims
C01facthigh confidence
USA Rare Earth owns the Pela Ema rare earth mine in Brazil (Serra Verde), a separation demonstration plant in Colorado, Less Common Metals' metal and alloy plant in England, a magnet plant in Stillwater, Oklahoma, and the undeveloped Round Top deposit in Texas.
current state · 2026-09-30written 2026-11-17
C18facthigh confidence
Stillwater's first magnet line, 600 tonnes a year, sold its first magnet blocks in September 2026; a second 600-tonne line is due in the second quarter of 2027, and a $1.2B plant at Blacksburg, South Carolina, is planned for 6,400 tonnes, toward about 10,000 tonnes of US capacity.
against this

Magnet revenue was $0.2M in the quarter and the finance chief calls current sales low-volume prototypes.

current state · 2026-11-09written 2026-11-17
C20facthigh confidence
Less Common Metals has made rare earth metals and alloys in England for decades with about 2,500 tonnes a year of capacity; in the third quarter shortages of dysprosium oxide and samarium metal from China's controls cut what it could make and sell.
against this

Management says its 2026 volumes passed all of 2025's by August and that legacy contracts reprice in the first half of 2027.

current state · 2026-11-09written 2026-11-17

How it makes money

In the third quarter of 2026, $2.8M of its $3.0M of sales came from Less Common Metals and $0.2M from magnets. The work cost $22.9M to deliver, a loss before any overheads.

From the fourth quarter the mine adds sales: all its first-stage output goes to one buyer at agreed floor prices, plus most of any market price above them.

The evidence under this - 3 claims
C02facthigh confidence
Revenue was $3.002M in the quarter to 30 September 2026, $2.8M from Less Common Metals and $0.2M from magnets, down from $5.821M in the June quarter; cost of revenue was $22.884M, a gross loss of $19.882M. Nine-month revenue was $14.521M.
against this

Management blames industry-wide shortages of dysprosium oxide and samarium metal, and Serra Verde's first sales fell in November, after the quarter.

historical · Q3 2026written 2026-11-17
C20facthigh confidence
Less Common Metals has made rare earth metals and alloys in England for decades with about 2,500 tonnes a year of capacity; in the third quarter shortages of dysprosium oxide and samarium metal from China's controls cut what it could make and sell.
against this

Management says its 2026 volumes passed all of 2025's by August and that legacy contracts reprice in the first half of 2027.

current state · 2026-11-09written 2026-11-17
C18facthigh confidence
Stillwater's first magnet line, 600 tonnes a year, sold its first magnet blocks in September 2026; a second 600-tonne line is due in the second quarter of 2027, and a $1.2B plant at Blacksburg, South Carolina, is planned for 6,400 tonnes, toward about 10,000 tonnes of US capacity.
against this

Magnet revenue was $0.2M in the quarter and the finance chief calls current sales low-volume prototypes.

current state · 2026-11-09written 2026-11-17
Where the money comes from, stream by stream
StreamShare of sales (Q3 2026)How it earns it
Less Common Metals metals and alloys93%Rare earth metals and alloys made in England for magnet makers.
Magnets7%Prototype magnet blocks from the first Stillwater line.

Who runs it

Thras Moraitis, who ran Serra Verde, became chief executive on 1 October 2026; Rob Steele is finance chief and Michael Blitzer, whose acquisition company took the business public, is executive chair. Officers and directors owned 9.1% in April, and Serra Verde's former owners now hold about a third of the shares.

The former chief executive was paid $11.7M for 2025, almost all in shares, with bonuses at the board's discretion. The chairman bought 100,000 shares at $21.44 in January.

The evidence under this - 4 claims
C36facthigh confidence
Thras Moraitis, Serra Verde's chief executive, has run the company since 1 October 2026, replacing Barbara Humpton; Rob Steele is finance chief and Michael Blitzer, whose acquisition company took it public, is executive chair.
current state · 2026-11-12written 2026-11-17
C37facthigh confidence
Officers and directors owned 9.1% in April 2026, before the Serra Verde shares; Serra Verde's former owners received 126.8M shares, about a third of those now outstanding.
current state · 2026-11-06written 2026-11-17
C38facthigh confidence
Barbara Humpton was paid $11.7M for 2025, $11.5M of it in stock awards; bonuses were set at the compensation committee's discretion rather than by stated goals.
historical · FY2025written 2026-11-17
C35facthigh confidence
The chairman bought 100,000 shares at $21.44 in January 2026; the former chief executive sold 88,391 at $13.75 in October 2026 after retiring, not under a pre-set plan.
against this

A retiring executive selling is often personal, and the chairman's purchase came at three times today's price.

historical · January to October 2026written 2026-11-17
02

Why it wins, and who could beat it

Why customers come back

Heavy rare earths are mined at scale almost only in China and Burma, and since April 2025 China requires a licence to export them. From 2027 the Pentagon may not buy magnets whose supply chain runs through China, from the mine onward.

Pela Ema is one of few such mines outside China, and a buyer backed by the government takes its first stage for about 15 years at floors of $575 a kilogram for dysprosium and $2,050 for terbium.

The evidence under this - 4 claims
C25facthigh confidence
China added export controls on samarium, gadolinium, terbium, dysprosium, lutetium, scandium and yttrium in April 2025, and those controls remained in effect, according to the US Geological Survey.
against this

China began issuing general export licences to selected exporters, so the controls can loosen without being lifted.

current state · 2026-02-01written 2026-11-17
C26facthigh confidence
From 1 January 2027 the Pentagon may not buy neodymium magnets whose supply chain, from mining through finished magnet, runs through a covered country such as China.
against this

The rule allows exceptions where no compliant supply exists at a reasonable price.

current state · 2026-09-01written 2026-11-17
C09facthigh confidence
Pela Ema sells 100% of its Phase 1 output to US SIIE, LLC under a take-or-pay offtake expected to run about 15 years, at floors of $2,050 a kilogram for terbium, $575 for dysprosium and $110 for neodymium and praseodymium, plus 70% of any market value above the floors; carbonate is priced at the floor less a fixed separation cost.
against this

The floors pay only for four elements; yttrium and the other heavy rare earths in the carbonate earn nothing until the company separates them itself.

current state · 2026-09-30written 2026-11-17
C24facthigh confidence
The US Geological Survey puts 2025 world rare earth mine output at about 390,000 tonnes of oxide, 270,000 of it Chinese and 51,000 American; Brazil produced about 2,000 tonnes, up from 560 in 2024.
current state · 2026-02-01written 2026-11-17

The threat

MP Materials, the US rival, sells under its own government price floor and is building a second magnet plant. Lynas separates dysprosium and terbium in Malaysia. Private magnet makers Noveon and VACUUMSCHMELZE run US plants, and Energy Fuels is buying a metal maker. If China eased its licences, cheaper Chinese supply would compete at every step.

The evidence under this - 4 claims
C27facthigh confidence
MP Materials, the US rival, sells under a Department of War price floor of $110 a kilogram on NdPr and is building a second magnet plant, the 10X Facility, to start commissioning in 2028.
current state · 2026-02-26written 2026-11-17
C28facthigh confidence
Lynas produced its first separated dysprosium oxide in Malaysia in May 2025 and terbium oxide in June 2025, from its Australian mine.
against this

Lynas's ore is mostly light rare earths, so its heavy output is limited, and it has not published tonnages.

historical · May to June 2025written 2026-11-17
C30facthigh confidence
USA Rare Earth's 10-K names MP Materials, Noveon Magnetics, VACUUMSCHMELZE, KSM Metals (being bought by Energy Fuels) and Lynas as competitors, and China as the dominant one.
current state · 2026-03-30written 2026-11-17
C29facthigh confidence
Energy Fuels' 10-K, citing Adamas Intelligence, sizes the magnet rare earth oxide market at $7.8B in 2024, forecast to reach $44.1B by 2040, with demand for the four magnet elements growing 8.7% a year.
against this

The figure is a forecaster's estimate a rival chose to publish, not a measured market.

current state · 2026-02-26written 2026-11-17

The rivals: MP, MP, UUUU, and 3 more that are private, in-house or listed abroad.

Who else does this, rival by rival
TickerRivalWhere it competesHow far alongPriced against it
MPMP Materials, the US mine-to-magnet rivalPriced against it: Same productThe mine-to-magnet build: Serra Verde, separation, Stillwater and Round TopOn saleSame product
MP Materials' metal-makingLess Common Metals: rare earth metals and alloysOn sale
Not listedLynas, the Australian producer separating dysprosium and terbium in MalaysiaPriced against it: Not listedThe mine-to-magnet build: Serra Verde, separation, Stillwater and Round TopOn saleNot listed
PrivateNoveon Magnetics and VACUUMSCHMELZE, private magnet makers with US plantsPriced against it: PrivateThe mine-to-magnet build: Serra Verde, separation, Stillwater and Round TopOn salePrivate
UUUUEnergy Fuels with KSM MetalsPriced against it: Small part of itLess Common Metals: rare earth metals and alloysOn saleSmall part of it
Not listedChinese producers, if export licences loosenPriced against it: Not listedThe mine-to-magnet build: Serra Verde, separation, Stillwater and Round TopOn saleNot listed
03

How big it could get

How big this could get

The world mined about 390,000 tonnes of rare earths in 2025, 270,000 of it in China; Brazil mined about 2,000. A forecaster's figure in a rival's annual report puts the magnet rare earths at $7.8B in 2024. Nobody publishes the heavy rare earth market outside China, so its size cannot be counted; at the second stage's 6,400 tonnes, Pela Ema would sell about $621M a year at today's prices.

The evidence under this - 3 claims
C24facthigh confidence
The US Geological Survey puts 2025 world rare earth mine output at about 390,000 tonnes of oxide, 270,000 of it Chinese and 51,000 American; Brazil produced about 2,000 tonnes, up from 560 in 2024.
current state · 2026-02-01written 2026-11-17
C29facthigh confidence
Energy Fuels' 10-K, citing Adamas Intelligence, sizes the magnet rare earth oxide market at $7.8B in 2024, forecast to reach $44.1B by 2040, with demand for the four magnet elements growing 8.7% a year.
against this

The figure is a forecaster's estimate a rival chose to publish, not a measured market.

current state · 2026-02-26written 2026-11-17
C41inferencemedium confidence
At 6,400 tonnes a year Pela Ema's oxide would sell for about $621M at the fourth-quarter market basket and $275M at the floors.
derivedC13's 6,400 tonnes times C12's $97 and $43 a kilogram.
against this

The basket moves with dysprosium and terbium prices outside China, and Stage 2 output may not all fall under the Phase 1 offtake.

current state · 2026-11-09written 2026-11-17
04

The results

The numbers

Is it profitable?Losing

Operating margin is −16×.

Dec 2025: −36×12 mo to Sep 2026: −16×−36×−16×
Does growth pay off?Unclear

Running costs take 1419¢ of each sales dollar, down from 3633¢ in the year to Dec 2025.

Dec 2025: 3633¢12 mo to Sep 2026: 1419¢3633¢1419¢
Does it earn good returns?Poor

Shareholders' money lost 4.9% a year, up from −125% in the year to Dec 2025.

Dec 2025: −125%12 mo to Sep 2026: −4.9%−125%−4.9%
How long can it keep going?Limited

It used $463M of cash over the last twelve months, so the $1.09B it holds lasts about 2.4 years at that pace.

Dec 2024: −$16.1MDec 2025: −$86.3M12 mo to Sep 2026: −$463M−$16.1M−$463M
Is the balance sheet safe?Net cash

It holds $1.09B of cash against $668M of debt, $423M more cash than it owes.

Dec 2024: $14.7MDec 2025: $358M12 mo to Sep 2026: $423M$14.7M$423M
All the numbers

Filed figures from the company's 10-K and 10-Q, read 2026-11-17. Fiscal years end in December.

Dec 2024Dec 202512 mo to Sep 2026
Size and growth
Sales$0$1.64M$16.2M
Profit
Gross margin-12%−131%
Operating margin-−36×−16×
Net margin-−181×−717%
What growth costs, per sales dollar
Research and development-967%301%
General and admin-26×11×
Pay in shares-534%131%
Capital spending-23×13×
Cash
Cash from operations−$13.0M−$49.0M−$256M
Free cash flow−$16.1M−$86.3M−$463M
Balance sheet
Cash and short-term investments$16.8M$360M$1.09B
Debt$2.03M$1.85M$668M
Net cash$14.7M$358M$423M
Returns and shares
Shares (diluted)--286M

What growth costs

Growth has not yet paid. In the third quarter the work cost $22.9M to deliver $3.0M of sales, and over the last four quarters the business lost $250.6M at the operating line on $16.2M of sales. Whether scale pays waits on the mine's first full quarters.

The evidence under this - 3 claims
C02facthigh confidence
Revenue was $3.002M in the quarter to 30 September 2026, $2.8M from Less Common Metals and $0.2M from magnets, down from $5.821M in the June quarter; cost of revenue was $22.884M, a gross loss of $19.882M. Nine-month revenue was $14.521M.
against this

Management blames industry-wide shortages of dysprosium oxide and samarium metal, and Serra Verde's first sales fell in November, after the quarter.

historical · Q3 2026written 2026-11-17
C03facthigh confidence
The Q3 2026 operating loss was $141.4M; selling and administration rose to $103.7M, with consulting and legal costs up $68.0M and $8.6M mainly on acquisitions. Over nine months the operating loss was $224.4M.
against this

Much of the overhead rise is deal costs the finance chief calls one-time.

historical · Q3 2026 and nine monthswritten 2026-11-17
C43facthigh confidence
Over the four quarters to September 2026 revenue was $16.2M, the operating loss $250.6M, operations used $256.1M of cash and capital spending was $206.9M; stock pay was $21.2M.
against this

Serra Verde contributed for under a month and sold nothing in the period, so the trailing year predates the mine.

historical · four quarters to Q3 2026written 2026-11-17

The last report (9 Nov)

Quarter ended September 2026, filed figures.

-48.4%
Growth from the quarter before
against 2.2% the quarter before that
-4712¢
Operating profit
per dollar of sales

In the third quarter of 2026, sales fell to $3.0M from $5.8M the quarter before, as shortages of Chinese inputs held back Less Common Metals. The operating loss was $141.4M, swollen by deal costs; a gain on warrants turned it into a small net profit, and the adjusted loss was $0.44 a share.

The mine was bought in September and began selling in November, with at least 325 tonnes guided for the quarter. Several targets set in March slipped by one to three quarters.

What it did to the thesisIt moved the case from plans to a running mine, but sales fell, losses grew and targets slipped, so the view rests even more on the mine's ramp and on how the new shares are issued.

The quarter in detail

Sales growth from the previous quarter, filed figures. The company has restated its own history, so a quarter either side of the restatement may describe a different business; n/c marks one that does not compare.

Earnings per share $-0.44 against $-0.10 expected; $-0.02 as filed, the gap being what the company adds back.

The evidence under this - 5 claims
C02facthigh confidence
Revenue was $3.002M in the quarter to 30 September 2026, $2.8M from Less Common Metals and $0.2M from magnets, down from $5.821M in the June quarter; cost of revenue was $22.884M, a gross loss of $19.882M. Nine-month revenue was $14.521M.
against this

Management blames industry-wide shortages of dysprosium oxide and samarium metal, and Serra Verde's first sales fell in November, after the quarter.

historical · Q3 2026written 2026-11-17
C03facthigh confidence
The Q3 2026 operating loss was $141.4M; selling and administration rose to $103.7M, with consulting and legal costs up $68.0M and $8.6M mainly on acquisitions. Over nine months the operating loss was $224.4M.
against this

Much of the overhead rise is deal costs the finance chief calls one-time.

historical · Q3 2026 and nine monthswritten 2026-11-17
C04facthigh confidence
Q3 2026 net income of $11.7M came from $148.1M of other income, mostly a fair-value gain on warrant and earnout liabilities; the company's adjusted net loss was $125.6M, or $0.44 a share.
historical · Q3 2026written 2026-11-17
C12management claimmedium confidence
Management guides fourth-quarter Serra Verde sales of at least 325 tonnes of rare earth oxide, worth about $43 a kilogram at the floors and about $97 at market prices after separation costs and the buyer's share.
against this

The $97 rests on record prices for dysprosium and terbium outside China; at the floors the same oxide earns less than half.

forwardwritten 2026-11-17
C22facthigh confidence
In March 2026 the company targeted a Round Top prefeasibility study by the end of Q3 2026, Stillwater's second line in Q1 2027 and Less Common Metals at 3,000 tonnes by end-2026; by November these had become a technical report in Q3 2027, the second line in Q2 2027 and new furnaces in Q3 2027.
against this

The chief executive says the Round Top delay buys more drilling and engineering before committing capital, which lowers building risk.

historical · March to November 2026written 2026-11-17
What management did with the money

The company raised $1.50B by selling new shares in January 2026, then paid $300.0M of cash and 126.8M shares for Serra Verde. It agreed to buy 13.6% of Carester, a French separation plant. Operations used $228.2M of cash over nine months and building spent $182.9M more.

Insiders in 2026Q3: 4 purchases and 0 sales, from their Form 4 filings.

The evidence under this - 4 claims
C07facthigh confidence
In January 2026 the company sold 69.8M shares in a $1.50B private placement, about $21.50 a share.
historical · January 2026written 2026-11-17
C08facthigh confidence
Serra Verde was bought on 3 September 2026 for about $2.57B: $300.0M of cash and 126.8M shares valued at $17.85 each.
historical · September 2026written 2026-11-17
C23facthigh confidence
The company agreed to buy 13.6% of Carester, whose French heavy rare earth separation plant is due to start in early 2027, for about €40.0M, with rights to buy part of its oxide.
current state · 2026-09-30written 2026-11-17
C06facthigh confidence
Operations used $228.2M of cash over nine months to September 2026; capital spending and equipment deposits were $182.9M and Serra Verde cost $190.0M net of its cash.
against this

The finance chief says the purchase price and deal costs do not recur.

historical · nine months to September 2026written 2026-11-17
05

The price, and where we disagree

In short. Full. At the floors the buyer guarantees, the price is about 13 times the first stage's sales, before the new shares the build still needs.

Share price
$7.13
close,
Company value for each $1 of sales
10.1
its 6-year normal 4.0

The shares closed at $7.13 on 16 November, 78.3% below their high, and about 15.5% of them are sold short.

The next test is the fourth-quarter report, around 31 March 2027: whether the mine sold at least 325 tonnes, and how the company met or changed the equity raise its government agreements require by that date.

The evidence under this - 4 claims
C31facthigh confidence
The shares closed at $7.13 on 16 November 2026, 78.3% below their 52-week high of $32.80 and down 63.0% over 60 sessions; at 375,319,930 shares the market value is about $2,676M.
current state · 2026-11-16written 2026-11-17
C34facthigh confidence
58,085,746 shares were sold short at 30 October 2026, about 15.5% of shares outstanding, up from 28.4M in mid-August.
against this

The vendor's float is stale and reads 30%; against shares outstanding it is about half that.

current state · 2026-10-30written 2026-11-17
C12management claimmedium confidence
Management guides fourth-quarter Serra Verde sales of at least 325 tonnes of rare earth oxide, worth about $43 a kilogram at the floors and about $97 at market prices after separation costs and the buyer's share.
against this

The $97 rests on record prices for dysprosium and terbium outside China; at the floors the same oxide earns less than half.

forwardwritten 2026-11-17
C15facthigh confidence
The Commerce agreements require the company to raise $375.0M of equity plus Serra Verde's cash acquisition costs by 31 March 2027, another $875.0M by 31 December 2027, and to set up a revolving credit line of up to $250.0M by 30 June 2027.
against this

The 10-Q says the company may instead seek to change the milestones with the Department.

forwardwritten 2026-11-17
The betSpeculation
If it works
2.2×
from today's value, at the stated multiple
If it fails
keep 16%
of today's market value: the rest of the company, less its debt
How we value it

No fair value: USA Rare Earth lost $250.6M at the operating line over the last four quarters on $16.2M of revenue, and the mine that would earn a profit began selling only in November; the bet prices it instead.

How we read the multiple

With no profit to price, we compare the whole company's value with what the mine's first stage would sell.

  1. The whole company, netting its cash against its loan and royalty, is worth about $2.26B.
  2. The first stage, 4,000 tonnes a year, sells about $172M at the floors and $388M at today's prices.
  3. That is 13.2 times the floor sales and 5.8 times at today's prices.
  4. The government agreements require about $1.55B of new equity by the end of 2027.
  5. At today's price that is about 217M new shares, 58% more.

Against its own history. It listed in 2025 and had almost no sales before the mine, so there is no history of what buyers paid for each dollar of sales.

The evidence under this - 4 claims
C32inferencemedium confidence
Company value is about $2,264M: $2,676M of market value less $1,090.9M of cash, plus the $305.6M loan, the $362.2M royalty obligation and $11.1M of preferred stock. At Stage 1's 4,000 tonnes the oxide sells for about $388M a year at the fourth-quarter market basket, 5.8 times, and $172M at the floors, 13.2 times.
derivedC31, C05 and C17 for the bridge; C13 and C12 for 4,000 tonnes at $97 and $43 a kilogram.
against this

The vendor's company value of about $164M rests on a share count of 132.6M, before the placement and the Serra Verde shares, and is not used.

current state · 2026-11-16written 2026-11-17
C40inferencemedium confidence
Raising the required $1.55B or more by the end of 2027 at the 16 November close would take about 217M new shares, 58% more than are outstanding.
derivedC15 and C31: $375M plus at least the $300M cash paid for Serra Verde, plus $875M, divided by $7.13; against C17's 375.3M shares.
against this

Up to $300M may be convertible notes, the milestones may be renegotiated, and the cash on hand covers the next year's spending.

current state · 2026-11-16written 2026-11-17
C12management claimmedium confidence
Management guides fourth-quarter Serra Verde sales of at least 325 tonnes of rare earth oxide, worth about $43 a kilogram at the floors and about $97 at market prices after separation costs and the buyer's share.
against this

The $97 rests on record prices for dysprosium and terbium outside China; at the floors the same oxide earns less than half.

forwardwritten 2026-11-17
C13management claimmedium confidence
Management targets Stage 1 at Pela Ema running at 4,000 tonnes a year by the second quarter of 2027 and Stage 2 at 6,400 tonnes by the end of 2027, and says Stage 2 is 40% built with over 75% of its capital committed.
against this

Brazil's whole rare earth mine output was about 2,000 tonnes in 2025, so the target triples it within two years.

forwardwritten 2026-11-17
Priced against its rivals

Only MP Materials sells the same thing to the same buyers as a large part of what it does. Energy Fuels does too, but as a small part of a much bigger company, so its price says little about this one. A price comparison needs at least two, so the price is judged against its own history instead.

Private or foreign rivals that cannot be priced: Noveon Magnetics Inc., VACUUMSCHMELZE GmbH & Co. KG, Lynas Rare Earths Ltd, KSM Metals Co., Ltd., China's state rare earth groups.

The bet: what it pays if it works

This bet is on the mine-to-magnet build. If the second stage ran at full rate with today's prices outside China, and buyers paid a generous multiple of sales, the company would be worth a little over twice today's value; after the new shares the build needs, each share gains much less.

If prices fall to the floors and the second stage stalls, the first stage alone earns too little to cover the losses, and new shares keep coming.

The mine-to-magnet build: Serra Verde, separation, Stillwater and Round Top revenue in two to three yearsMillion dollars of oxide and magnet sales a year neededAt the stated multiple
$126.00Mabout 126 at 100% kept0.5×
$621.00M (the big win)about 621 at 100% kept2.2×
$1.48Babout 1.5K at 100% kept5.2×

To pay 2× (a Fair bet) it would need about $564.01M of The mine-to-magnet build: Serra Verde, separation, Stillwater and Round Top revenue (about 564 million dollars of oxide and magnet sales a year); to pay 5× (a Cheap bet), about $1.41B of The mine-to-magnet build: Serra Verde, separation, Stillwater and Round Top revenue (about 1.4K million dollars of oxide and magnet sales a year), at the same multiple.

Each row is The mine-to-magnet build: Serra Verde, separation, Stillwater and Round Top alone, plus the rest of the company (Less Common Metals: rare earth metals and alloys) counted at about $16.00M.

The steps it has to take

  • Done Serra Verde bought and Pela Ema's first stage built
  • Done First magnet blocks sold from Stillwater
  • In progress At least 325 tonnes sold in the fourth quarter due 2026-12
  • Not started Equity milestone of $375M plus acquisition cash due 2027-03
  • In progress Stage 1 at 4,000 tonnes a year due 2027-06
  • In progress Stage 2 at 6,400 tonnes a year due 2027-12
The detail behind the label
Rating
Fair
the big win's payoff against the 2× and 5× lines
Cheap bet at about
≈ $3.75
47% below the price (close, )
The option
2.2×
what The mine-to-magnet build: Serra Verde, separation, Stillwater and Round Top is worth if it works, against what the price pays for it
The evidence under this - 4 claims
C41inferencemedium confidence
At 6,400 tonnes a year Pela Ema's oxide would sell for about $621M at the fourth-quarter market basket and $275M at the floors.
derivedC13's 6,400 tonnes times C12's $97 and $43 a kilogram.
against this

The basket moves with dysprosium and terbium prices outside China, and Stage 2 output may not all fall under the Phase 1 offtake.

current state · 2026-11-09written 2026-11-17
C40inferencemedium confidence
Raising the required $1.55B or more by the end of 2027 at the 16 November close would take about 217M new shares, 58% more than are outstanding.
derivedC15 and C31: $375M plus at least the $300M cash paid for Serra Verde, plus $875M, divided by $7.13; against C17's 375.3M shares.
against this

Up to $300M may be convertible notes, the milestones may be renegotiated, and the cash on hand covers the next year's spending.

current state · 2026-11-16written 2026-11-17
C32inferencemedium confidence
Company value is about $2,264M: $2,676M of market value less $1,090.9M of cash, plus the $305.6M loan, the $362.2M royalty obligation and $11.1M of preferred stock. At Stage 1's 4,000 tonnes the oxide sells for about $388M a year at the fourth-quarter market basket, 5.8 times, and $172M at the floors, 13.2 times.
derivedC31, C05 and C17 for the bridge; C13 and C12 for 4,000 tonnes at $97 and $43 a kilogram.
against this

The vendor's company value of about $164M rests on a share count of 132.6M, before the placement and the Serra Verde shares, and is not used.

current state · 2026-11-16written 2026-11-17
C43facthigh confidence
Over the four quarters to September 2026 revenue was $16.2M, the operating loss $250.6M, operations used $256.1M of cash and capital spending was $206.9M; stock pay was $21.2M.
against this

Serra Verde contributed for under a month and sold nothing in the period, so the trailing year predates the mine.

historical · four quarters to Q3 2026written 2026-11-17

Where we disagree with the market

The market believes

The only US-listed heavy rare earth producer, with a government buyer, 78% below its high.

We believe

A real mine priced at record prices, before the new shares its build still needs.

The market sees the only US-listed heavy rare earth producer, with a government buyer, after a 78% fall, and analysts expect sales of about $611M next year. We see a price that pays for the first stage at today's record prices, before more than half again as many shares.

The evidence against us: the second stage on time with prices outside China holding, or the equity raise met well above today's price, would show the price was cheap.

What analysts say, and the evidence
Analysts positive
100%
of 9, 1 Nov
Four months earlier
100%
of 8, 1 Jul
The evidence under this - 5 claims
C33facthigh confidence
Analysts rate the shares five buys and no holds; four forecast sales of about $611M in 2027 and $1.48B in 2028, with a loss of about $0.71 a share in 2027.
against this

The adjusted loss of $0.44 a share in Q3 2026 was over four times the $0.10 analysts expected.

forwardwritten 2026-11-17
C42facthigh confidence
MP Materials' company value was about 29 times its trailing sales at the end of September 2026.
against this

MP earns most of its sales from separated oxide and a government floor, so its multiple prices a running integrated producer.

current state · 2026-09-30written 2026-11-17
C32inferencemedium confidence
Company value is about $2,264M: $2,676M of market value less $1,090.9M of cash, plus the $305.6M loan, the $362.2M royalty obligation and $11.1M of preferred stock. At Stage 1's 4,000 tonnes the oxide sells for about $388M a year at the fourth-quarter market basket, 5.8 times, and $172M at the floors, 13.2 times.
derivedC31, C05 and C17 for the bridge; C13 and C12 for 4,000 tonnes at $97 and $43 a kilogram.
against this

The vendor's company value of about $164M rests on a share count of 132.6M, before the placement and the Serra Verde shares, and is not used.

current state · 2026-11-16written 2026-11-17
C40inferencemedium confidence
Raising the required $1.55B or more by the end of 2027 at the 16 November close would take about 217M new shares, 58% more than are outstanding.
derivedC15 and C31: $375M plus at least the $300M cash paid for Serra Verde, plus $875M, divided by $7.13; against C17's 375.3M shares.
against this

Up to $300M may be convertible notes, the milestones may be renegotiated, and the cash on hand covers the next year's spending.

current state · 2026-11-16written 2026-11-17
C44management claimmedium confidence
Management says dysprosium oxide outside China sells for 14 to 16 times the Chinese price and terbium for 6 to 8 times.
against this

Prices outside China are thinly traded and fall if Beijing grants more export licences.

current state · 2026-11-09written 2026-11-17
06

What could go wrong, and how we would know

The risks

  • New shares. About $1.55B of equity is required by the end of 2027.
  • China's licences. Easier export licences would pull prices toward the floors.
  • The buyer's option. If the mine stops for 60 days, the buyer may take it at a fair value.
  • Licences in Brazil. The mine's licence expires in December 2028 and it carries environmental notices.
  • Slipping targets. Three targets slipped one to three quarters in 2026.
The evidence under this - 5 claims
C40inferencemedium confidence
Raising the required $1.55B or more by the end of 2027 at the 16 November close would take about 217M new shares, 58% more than are outstanding.
derivedC15 and C31: $375M plus at least the $300M cash paid for Serra Verde, plus $875M, divided by $7.13; against C17's 375.3M shares.
against this

Up to $300M may be convertible notes, the milestones may be renegotiated, and the cash on hand covers the next year's spending.

current state · 2026-11-16written 2026-11-17
C25facthigh confidence
China added export controls on samarium, gadolinium, terbium, dysprosium, lutetium, scandium and yttrium in April 2025, and those controls remained in effect, according to the US Geological Survey.
against this

China began issuing general export licences to selected exporters, so the controls can loosen without being lifted.

current state · 2026-02-01written 2026-11-17
C11facthigh confidence
If the mine stops for 60 days, the offtake is breached or a lender forecloses, the buyer may purchase the whole Brazilian subsidiary at a fair value set by independent experts.
current state · 2026-09-30written 2026-11-17
C39facthigh confidence
Pela Ema runs on a corrective operating licence that expires in December 2028 unless renewed, and the company took on Serra Verde's notices of environmental violations, including operating facilities without required licences.
against this

Stage 1 was completed and shipping began, so the notices have not stopped production.

current state · 2026-09-30written 2026-11-17
C22facthigh confidence
In March 2026 the company targeted a Round Top prefeasibility study by the end of Q3 2026, Stillwater's second line in Q1 2027 and Less Common Metals at 3,000 tonnes by end-2026; by November these had become a technical report in Q3 2027, the second line in Q2 2027 and new furnaces in Q3 2027.
against this

The chief executive says the Round Top delay buys more drilling and engineering before committing capital, which lowers building risk.

historical · March to November 2026written 2026-11-17
What we still do not know
  • What heavy rare earths actually sell for outside China, element by element.
  • How much of the second stage's output falls under the floor-priced offtake.
  • What Round Top would cost to build and produce, which no study has shown.
The evidence under this - 3 claims
C12management claimmedium confidence
Management guides fourth-quarter Serra Verde sales of at least 325 tonnes of rare earth oxide, worth about $43 a kilogram at the floors and about $97 at market prices after separation costs and the buyer's share.
against this

The $97 rests on record prices for dysprosium and terbium outside China; at the floors the same oxide earns less than half.

forwardwritten 2026-11-17
C09facthigh confidence
Pela Ema sells 100% of its Phase 1 output to US SIIE, LLC under a take-or-pay offtake expected to run about 15 years, at floors of $2,050 a kilogram for terbium, $575 for dysprosium and $110 for neodymium and praseodymium, plus 70% of any market value above the floors; carbonate is priced at the floor less a fixed separation cost.
against this

The floors pay only for four elements; yttrium and the other heavy rare earths in the carbonate earn nothing until the company separates them itself.

current state · 2026-09-30written 2026-11-17
C21facthigh confidence
Round Top's rare earth value is about 72% heavy elements, planned for a heap-leach process; the company owns all of it since buying its partner in August 2026 for about 3.8M shares, and targets a technical report in the third quarter of 2027 and a build decision in the second half.
against this

Round Top has no reserve under SEC rules and no feasibility study, so its cost and output are unknown.

current state · 2026-11-09written 2026-11-17

What would change our view

  1. Unattractive→NeutralA price that pays for Stage 1 at the floors

    It needs:

    • ○The whole company's value falls to 8 times the first stage's sales at the floors.about ≈$4.32 today, 39% below the last close
  2. Unattractive→NeutralStage 2 delivered with prices holding

    It needs all of:

    • ○Pela Ema sells at a 6,400-tonne yearly rate in a quarter of 2027 or 2028.watched as it happens
    • ○Prices in that quarter stay above the floors, so the buyer pays more than the guaranteed price.watched as it happens
We would also be wrong if
  • The equity raise due by 31 March 2027 is waived, deferred or met well above today's price.
  • Stage 1 runs at 4,000 tonnes a year by mid-2027.
  • China tightens its licences further and prices outside China rise again.

What is coming

  • 1 January 2027 Pentagon's whole-chain magnet sourcing rule takes effect: Whether defence buyers turn to suppliers outside China.
Every dated event ahead
  • 1 January 2027 Pentagon's whole-chain magnet sourcing rule takes effect: Whether defence buyers turn to suppliers outside China.
  • 31 March 2027 Equity milestone under the Commerce agreements: How much new stock is issued, and at what price, or whether the milestone is changed.
  • 31 March 2027 Q4 2026 results: Whether Serra Verde sold at least 325 tonnes and at what basket.
  • 30 June 2027 Stage 1 at 4,000 tonnes a year and Stillwater's second line: Whether the mine and magnet plant ramp on the new schedule.
Since we wrote this
Written
17 Nov
Price the thesis read
$7.13
16 Nov
  • 17 NovThesis writtenUnattractive, medium convictionWritten a week after the Q3 2026 results, with the shares 78% below their 52-week high.

Sources 15

Where this came from

Sources - 15

Every claim above cites one of these, or states the reasoning that derived it. The quote is frozen at the date it was read: a link can rot, and an agent fetch returns a reading of a page rather than the page. Backs counts the whole claim ledger, so it can name a claim this page does not surface.

  1. S01primarycompany filing2026-03-30
    the business, Round Top's heavy rare earth basket and flowsheet, Less Common Metals' capacity, Stillwater's lines, the named competitors, 2025 results and headcount
    backsC01C18C20C21C30C43C45
  2. S02primarycompany filing2026-11-12
    revenue by business and region, customer concentration, the Serra Verde purchase, the offtake terms and floor prices, the call option, the development-bank loan and royalty, the Commerce agreements and equity milestones, cash flows, share count, the Brazilian licence and the liquidity statement
    backsC01C02C03C05C06C07C08C09C10C11C14C15C17C20C21C23C31C34C36C37C39C43C46
  3. S03primarycompany release2026-11-09
    the quarter's results, the fourth-quarter Serra Verde outlook and basket values, the 2027 operating targets, Blacksburg, Carester, the TMRC purchase and the chief executive change
    backsC01C03C04C10C12C13C18C22C23C44
  4. S04primarytranscript2026-11-09
    magnet letters of intent, Stage 2 progress, the Round Top decision timing, raw-material shortages, the cash plan against the equity milestones and the basket explained
    We have signed letters of intent and MOUs covering approximately 2500 tons of magnets, and secured purchase orders across the industrial mobility, aerospace, and defense sectors.
    backsC04C12C13C16C18C19C20C21C23C36
  5. S05primarycompany filing2026-04-23
    the board, the chairman's background, beneficial ownership and 2025 executive pay
    backsC36C37C38
  6. S06primarygovernment data2026-02-01
    China's export controls on the heavy rare earths and their status
    In April 2025, China tightened its export controls on rare-earth elements, adding specific controls on alloys, compounds, metals, and oxides of samarium, gadolinium, terbium, dysprosium, lutetium, scandium, and yttrium.
    backsC25
  7. S07primarygovernment data2026-02-01
    2025 mine production: world about 390,000 tonnes of oxide, China 270,000, the United States 51,000 and Brazil 2,000, up from 560 in 2024; US net import reliance on compounds and metals of 67%
    World Mine Production and Reserves: Significant revisions were made to the 2024 production for Australia, Brazil, Burma, Madagascar, Nigeria, and Vietnam based on Government and industry reports.
    backsC24
  8. S08primaryregulator2026-09-01
    the Pentagon's rule barring magnets whose supply chain runs through a covered country, from mining onward, from 2027
    Effective January 1, 2027, for neodymium-iron-boron magnets, this restriction includes the entire supply chain from mining of neodymium, iron, and boron through production of finished magnets
    backsC26
  9. S09primarycompany filing2026-02-26
    MP's Department of War price floor on NdPr and its second magnet plant, the 10X Facility, to begin commissioning in 2028
    the Company entered into the Price Protection Agreement, which provides a price floor of $110 per kilogram
    backsC27
  10. S10primarycompany filing2026-02-26
    Adamas Intelligence's size of the magnet rare earth oxide market as a rival reports it, and Energy Fuels' own separation capacity and planned purchase of Australian Strategic Materials
    The total market for magnet rare earth oxides is forecasted to increase five-fold, from $7.8 billion in 2024 to $44.1 billion by 2040
    backsC29
  11. S11secondarynews2025-06-18
    Lynas's first separated dysprosium oxide in May 2025 and terbium oxide in June 2025 in Malaysia, quoting its chief executive
    Lynas can now supply customers with the two heavy rare earths required for rare earth permanent magnets used in electric motors.
    backsC28
  12. S12secondaryvendor data2026-11-17
    Research pack (FMP market tier and FINRA short interest), frozen 2026-11-17
    content/USAR/research-pack.json
    the close, the 252-session high and low, returns, the analyst split and estimates, earnings history and short interest
    backsC02C31C33C34
  13. S13primarycompany filing2026-10-02
    each open-market purchase and sale, its price and whether it was made under a pre-set plan
    backsC35
  14. S14secondaryvendor data2026-09-30
    MP Materials research pack, frozen with its thesis (FMP market tier)
    content/MP/research-pack.json
    MP's company value against its trailing sales
    backsC42
  15. S15primarycompany release2026-03-30
    the 2025 operating loss and capital spending, and the targets then set for Stillwater, Less Common Metals and Round Top
    backsC22C45
Rendered from content/USAR/thesis.json · thesis reviewed 2026-11-17. This page explains a company; it does not say what to pay, when to buy or how much.