Energy Fuels

Phase A — Understand the business

Lens 1 · Company Overview

Energy Fuels is not a pure uranium miner — it is a US critical-minerals platform with a uranium cash engine and a rare-earth growth bet bolted on top, anchored by one irreplaceable physical asset: the White Mesa Mill near Blanding, Utah — "the only licensed and operating uranium mill, and the only uranium mill capable of producing separated REE products, in the U.S.". The Mill is licensed to produce >8 million lbs U3O8/year and is the structural moat — it lets the company process uranium, vanadium, rare earths, and (prospectively) radium under one set of NRC/state licenses that would take a competitor a decade to replicate.

Three reportable segments:

1. Uranium — conventional + ISR mining (Pinyon Plain, La Sal, Pandora live; Nichols Ranch ISR, Whirlwind, Roca Honda, Bullfrog, Sheep Mountain, EZ as pipeline/standby), processing at the Mill, plus vanadium (V2O5) co-product and an Alternate Feed Materials recycling business. This is the only revenue-and-cash engine today.
2. Rare Earth Elements (REE) — the Mill's Phase 1 Circuit cracks/leaches monazite to produce separated NdPr (and pilot-scale Dy/Tb/Sm). Pre-revenue/strategic — generated ~$0 in 2025.
3. Heavy Mineral Sands (HMS) — monazite-feedstock mines (Vara Mada/Madagascar, Bahia/Brazil, Donald JV/Australia) plus the depleting Kwale/Kenya mine (mining ceased Dec 2024, now in reclamation). HMS exists primarily to feed its own REE chain with monazite, with ilmenite/rutile/zircon (Ti/Zr) sold as commercial co-products.

Customers & contract structure (uranium): Six long-term uranium supply contracts with major US utilities, requiring deliveries 2026–2032; base 3.21M lbs remaining, 3.71–5.29M lbs with buyer options/flex. These are fixed/escalating-price term contracts (a partial hedge — in 2025 contract sales realized $71.06/lb vs $76.90/lb spot, i.e. below spot, the classic term-vs-spot trade-off). Total minimum future contract revenue $173.4M. The company supplements/optimizes with discretionary spot sales (350k of 650k lbs sold in 2025 were spot). REE customers are qualification/offtake relationships, not yet revenue: POSCO International (NdPr sampling), Vulcan Elements (MOU for ex-China magnets), a Korean EV motor-core maker (qualified EF's NdPr into commercial REPMs), and Neo Performance Materials (historical MREC buyer).

In plain terms: EF sells ~650k–2M lbs of low-cost domestic uranium a year into a tight, geopolitically-protected market, and is spending the proceeds (plus a $700M convert and a $1.9B magnet acquisition) to build the only non-Chinese rare-earth supply chain that runs from dirt to finished magnet.

Lens 2 · Supply Chain

Upstream inputs → Energy Fuels → end customer, named at each node:

Uranium chain (live):

• Inputs: Conventional ore from EF's own Pinyon Plain (AZ), La Sal & Pandora (UT) mines; Alternate Feed Materials from third-party industrial generators (uranium-bearing residues recycled under the Mill's license); one 2025 ore-purchase agreement with a third-party miner near the Mill; spot-market uranium purchases to backfill contracts.
• EF node: White Mesa Mill — crushing, leaching, SX, drying to finished U3O8 (and V2O5 co-product). Conversion/storage at North American converters (the ~810k lbs finished inventory sits "at the Mill and at conversion facilities in North America" — i.e. Honeywell/ConverDyn Metropolis is the implied UF6 conversion chokepoint, industry-standard).
• Customers: Six major US nuclear utilities (unnamed, NDA-protected) under term contracts; spot buyers including financial holders (Sprott Physical Uranium Trust-type entities the filing alludes to as "financial entities purchasing uranium products to hold").

Rare-earth chain (building — the strategic spine):

• Feedstock: Monazite sands — from open-market offtake (The Chemours Company, Florida/Georgia mineral sands — the source of EF's current NdPr) and, prospectively, EF's own HMS mines: Vara Mada (Madagascar), Bahia (Brazil), Donald JV (Australia, with Astron Corporation). Also third-party MREC/intermediate REE concentrates.
• Separation: White Mesa Mill Phase 1 Circuit (NdPr + heavy concentrate) → planned Phase 2 Circuit (BFS Jan 2026).
• Metals/alloy (pending ASM, ~July 2026 close): ASM's Korean Metals Plant (one of few ex-China REE metal/alloy plants) + planned US American Metals Plant.
• Magnets (pending VAC, ~early-2027 close): Vacuumschmelze (VAC) — European plants + Sumter, SC (2,000 tpa designed, expandable to 12,000).
• End markets: EV/hybrid drive motors, defense systems, robotics, wind turbines, consumer electronics — via OEMs and magnet buyers (Vulcan Elements, the Korean motor-core maker, VAC's 1,000+ existing customers).

Heavy-mineral-sands co-product chain: ilmenite/rutile → TiO2 pigment & titanium metal producers; zircon → ceramics/nuclear-cladding/foundry buyers. (Kwale's last shipment was April 2025; future Ti/Zr volume depends on Donald/Vara Mada FIDs.)

Chokepoints & single-source dependencies:

• The White Mesa Mill is the ultimate single point of failure — one site, one license, in one US state. Any regulatory, environmental, or operational disruption there severs both the uranium and the REE-separation businesses. (See the standing Ute Mountain Ute Tribe / Grand Canyon Trust license challenges in Lens 10.)
• Phase 1 Circuit cannot run conventional-uranium and REE production simultaneously (shared crack-and-leach circuit) — a literal throughput chokepoint that forces the Mill to alternate campaigns until Phase 2 (a dedicated REE crack-and-leach line) is built (~mid-2029).
• Monazite feedstock is the binding constraint on the entire REE thesis — the Mill can separate more than EF can currently feed it. Vara Mada and Donald are both pre-FID and politically/financing-exposed (Madagascar had a 5-yr suspension and a 2025 government overthrow; Donald needs ~AUD$520M and a 2027 start). Until they produce, EF depends on Chemours open-market monazite.
• Conversion/enrichment sits downstream of EF's U3O8 — a US bottleneck (limited domestic conversion capacity) that the filing flags as a tailwind for term-contracting but a dependency nonetheless.

This lens has names, not generalities — the chain is real and unusually vertically ambitious. But it is a chain of mostly un-built links: today only the uranium node and the Phase-1 separation node actually move product.

Lens 3 · Competitive Advantages (Moats)

Durable moats (strong, and genuinely scarce):

1. Regulatory monopoly on conventional US uranium milling + the only US REE-separation-from-monazite license. This is the crown jewel. White Mesa is the only operating conventional uranium mill in America and the only facility that has separated commercial NdPr (and now pilot Dy/Tb) from monazite on US soil. Permitting a new conventional mill in the US is a multi-year, politically fraught process — this is a true scale/process/regulatory moat, not a marketing claim.
2. First-mover on non-Chinese heavy rare earths. First US company to publicly report Dy production at 99.9% purity (Aug 2025) and first Tb (Mar 2026). With China controlling ~92% of refined NdPr and 98–99% of separated heavy REEs (Dy/Tb), being the only qualified Western heavy-REE source is a scarcity moat — for as long as it lasts.
3. Vertical integration (post-VAC/ASM) no Western peer can match. If both deals close, EF becomes "the first western company with geographically diversified commercial capabilities across every critical step of the rare earth value chain" — mine→separate→metal/alloy→magnet. MP Materials has magnets (Independence/Fort Worth) but a single-mine feedstock (Mountain Pass) and no comparable heavy-REE separation; Lynas separates but does not make magnets. EF's spread of nodes is unique.
4. Cost moat in uranium (newly emerged). Pinyon Plain ore at $23–30/lb all-in (mine+mill) is "among the lowest costs for mined uranium production in the world," against an ~$85/lb spot price. The Mill's ability to blend feed sources to hit contract specs is described as "a unique element… that no other producer has in North America."
5. Inventory + balance-sheet optionality. ~810k lbs finished U3O8 + ~1.37M lbs in stockpiled material + ~905k lbs vanadium, plus ~$800M of marketable securities — lets EF time spot sales and self-fund the REE build without forced selling.

Bargaining power: Over uranium customers, rising (utilities are short non-Russian supply and chasing term contracts — the filing notes "an uptick in interest from nuclear utilities"). Over REE customers, currently low/aspirational (EF is sampling/qualifying, not dictating terms). Over suppliers, mixed — strong vs. AFM generators (they pay EF to take waste), weak vs. monazite (open-market dependent until its own mines produce).

Where the moat is thinner than it looks: The heavy-REE scarcity premium is a function of Chinese policy, not EF's technology — if Beijing relaxes export quotas (as it has cyclically), the "only Western source" premium compresses fast. And the magnet/metal moats are acquired, not organic: EF is buying VAC's 100 years and 400 patents for $1.9B, not building them. The moat is real but partly rented and partly geopolitical.

Lens 4 · Segments

FY2025 revenue & operating income by segment:

*FY2025 full-year segment operating income not cleanly split in the annual table; Q1 2026 segment operating results below are the clearest read.

FY2025 uranium detail:

• Uranium concentrates $48.2M (+$10.3M YoY); AFM/processing $1.9M.
• Volume 650,000 lbs U3O8 sold (vs 450k; +44%): 350k lbs spot @ $76.90/lb, 300k lbs contract @ $71.06/lb; blended realized $74.21/lb (−12% YoY).
• Cost applicable $50.89/lb (+38% YoY) → uranium gross margin ~31% in 2025, but rising fast as Pinyon Plain ore flushes through.
• Vanadium: held in inventory (~905k lbs), not materially sold in 2025.

HMS detail: $15.8M of FY2025 revenue was previously-produced Kwale inventory shipped out (final shipment April 2025) at no gross profit (low-grade end-of-mine ore cost more to process than it fetched). HMS revenue effectively goes to ~$0 until Donald/Vara Mada come online (2027+).

Q1 2026 segment operating results — the inflection:

Geography: Operations span US (uranium + Mill), Kenya (Kwale reclamation), Madagascar (Vara Mada), Brazil (Bahia), Australia (Donald JV). Revenue today is ~100% US-origin uranium; the international footprint is all future feedstock/HMS.

The segment story in one line: uranium is accelerating into profitability (Q1'26 segment op income turned positive on Pinyon Plain economics), HMS is rolling off to zero (Kwale done), and REE is a widening cash sink ($5.9M Q1 loss) that is the entire growth thesis.

---

Phase B — Measure performance

Lens 5 · Earnings Result (latest print: Q1 2026, ended 2026-03-31)

• Revenue $35.84M (vs $16.90M Q1'25, +112%). All uranium: concentrates $35.7M (vs $0 — there were no uranium sales in Q1'25), AFM $0.1M. HMS $0 (vs $15.5M; Kwale gone).
• Net loss attributable $(10.84M), EPS $(0.04) — vs $(26.30M)/$(0.13) Q1'25. Loss cut ~59%.
• Uranium COGS $21.5M on $35.8M revenue → ~40% segment gross margin, up sharply from negative in Q1'25 — the Pinyon Plain low-cost ore is now in cost-of-goods. Management guided COGS falling to $30–40/lb in Q1 2026 from the FY2025 ~$43/lb inventory average, en route to the $23–30/lb run-rate as fresh Pinyon Plain pounds dominate the FIFO stack.
• Operating loss $(16.93M) — driven entirely by the REE build ($5.9M) + HMS reclamation drag ($9.1M) + $2.4M VAC/ASM transaction costs; uranium itself was operating-profit-positive.
• Other income +$5.79M (unrealized MTM gain on marketable securities +$3.7M, realized +$1.1M, interest +$1.7M, FX −$0.8M) — the ~$800M T-bill book is now a real earnings contributor.

Guidance (FY2026): Mined 2.0–2.5M lbs U3O8 · Processed 1.5–2.5M lbs · Sales 1.5–2.0M lbs (a 2.3–3.1× jump over 650k lbs in 2025). Conventional Mill run continues through Q2 2026, then pauses until ~Q1 2027 pending feedstock. Tone: confident — December 2025 8-K was characterized in the press as "crushes guidance and charges into 2026".

Balance-sheet flags:

• Cash $? (FY2025 ended $64.7M) + marketable securities $802.2M (debt $778.1M ≈ US Treasuries/agencies + equity $24.1M) + $22.7M restricted.
• Convertible senior notes $676.7M carrying / $816.9M fair value — the convert trades ~21% above par because the stock is near/above the $20.34 conversion strike; effective rate just 1.38%.
• Inventory $69.0M; trade receivables fell to $5.5M (from $16.0M) as 2025 deliveries collected.
• ARO undiscounted $48.7M vs only $22.7M collateral posted — EF is liable for the ~$26M gap.
• Working capital ~$927M (FY2025) — extraordinary liquidity for a sub-$70M-revenue company.

Market reaction: The stock is +22% YTD 2026 but fell on the June 23 VAC announcement (dilution: 65.9M new shares; close dropped from ~$16.12 on Jun 22 toward ~$14.5 by Jun 29). The tape says the market loves the REE narrative but is price-sensitive to dilution — a critical signal for Lens 12/13.

Unusual vs. own history: The Q1 uranium gross margin flip and the sheer scale of the marketable-securities book ($802M vs $81M a year earlier) are both step-changes, not trends. This is a company mid-transformation.

Lens 6 · Earnings Calls (sentiment trend)

No transcripts on the shelf (transcripts/ empty; the compiled layer has 0 quarters). This lens is reconstructed from filings + press, labeled accordingly.

Management's narrative arc across FY2024 → Q1 2026:

• Recurring themes (rising): "critical minerals hub," "only US REE separation," "non-Chinese supply chain," "lowest-cost uranium," "mine-to-magnet." Each quarter the REE language gets louder and the uranium language more confident-but-secondary.
• New in 2025–26: "heavy rare earths" (Dy/Tb), "metals and alloys" (ASM), and now "magnets" (VAC) — a deliberate, quarter-by-quarter march down the value chain, each step announced as a milestone.
• Things they stopped saying: the pure-play uranium framing of 2022–23 is gone; HMS as a standalone business (Kwale) has been recast purely as "monazite feedstock control."
• Tone shift: from "ramp-up/standby/care-and-maintenance" defensiveness (depressed-price era) to acquisitive, growth-forward, capital-deploying confidence. The $700M convert and $1.9B VAC deal are management acting on that confidence, not just talking.

Sentiment trend: decisively more bullish and more ambitious quarter-over-quarter — which is itself a yellow flag for a serial acquirer (see Lens 13). The substance (Dy/Tb milestones, Q1 uranium profitability) partly earns the optimism; the scale of the bet outruns the proven economics.

Lens 7 · Comps

Peer set: Western/non-Chinese uranium and rare-earth names. Multiples are `` with date or n/a. Energy Fuels' own EV/EBITDA is n/m (negative EBITDA).

Read: The entire complex trades at elevated multiples vs. historical medians (Lynas 83x vs 10-yr median 14x; Cameco 50x vs 33x) — i.e. the market is pricing a structural critical-minerals super-cycle, not current earnings. EF, MP, and (effectively) the developers have no meaningful earnings to multiply — they are valued on optionality and strategic scarcity. The honest comp statement: UUUU cannot be valued on a multiple today; it is an asset/optionality story — ~$1.4B of assets (incl. ~$800M securities + the irreplaceable Mill + mineral properties) supporting a $3.5B equity value, with the gap = the REE/magnet call option. Against MP ($9.8B for a more-built magnet platform) EF looks cheaper on the magnet narrative but less proven on near-term EBITDA.

Lens 8 · Stock-Price Catalysts (>5% moves, ~last 5 yrs)

Mostly ``; pattern-reading.

• 2021 — Sprott Physical Uranium Trust launch + uranium spot squeeze: the entire uranium complex (UUUU included) re-rated violently on financial buyers warehousing physical U3O8. EF's price is highly beta to the uranium spot narrative.
• 2021–2023 — REE pivot announcements: first US monazite processing, Phase 1 Circuit commissioning, first NdPr — each a step-change in the "EF is a rare-earth story now" re-rating.
• Aug 2025 — first 99.9% Dy: "first U.S. company to publicly report Dy production volumes and purities" — meaningful pop.
• Dec 2025 — Dy qualified by major Korean automaker + "crushes guidance" 8-K: strong catalyst into 2026.
• Jan 2026 — uranium spot spike to $100.25 (Jan 28): lifted the whole group; UUUU rode it.
• Jan 20 2026 — ASM acquisition agreement: "mine-to-metal-and-alloy" — re-rate.
• Mar 2026 — first 99.9% Tb: shares +~5% pre-market.
• Jun 23 2026 — VAC $1.9B acquisition: stock fell on dilution (the rare negative reaction to a strategic announcement) — close ~$16.12 (Jun 22) → ~$14.5 (Jun 29).

What the market actually reacts to: (1) uranium spot price (high beta), (2) REE production/qualification milestones (the re-rating engine), and (3) — now — dilution from acquisitions (the new downside trigger). The June VAC drop is the tell: the narrative is fully priced, so incremental bullish news must clear a high bar while dilution is punished. This is a stock that trades on story momentum and is vulnerable when the story requires writing big equity cheques.

---

Phase C — Judge people & books

Lens 9 · Management

• Track record: Mark Chalmers (CEO since 2018, retiring April 15, 2026) — ISR/conventional uranium operator; ex-Paladin Energy (ran Langer Heinrich/Namibia + Kayelekera/Malawi), ex-General Atomics (Beverley ISR), ex-Cameco (Highland); mining-engineering degree, 10 yrs Chair of the Australian Uranium Council. He executed the uranium-restart-plus-REE-pivot — a genuinely impressive strategic re-positioning of a sleepy uranium miner into a critical-minerals platform.
• Incoming: Ross Bhappu (President → CEO, eff. April 15, 2026) — 35+ years mining + private equity, ~25 yrs at Resource Capital Funds (M&A, project finance, valuation, capital sourcing for mining). Chalmers stays two years as consultant. Archetype read: this is a deliberate hand-off from an operator (Chalmers) to a financier-dealmaker (Bhappu) — exactly the profile you appoint to run an aggressive, capital-markets-funded roll-up. The VAC/ASM strategy is the Bhappu strategy. Bullish if you believe the roll-up; a flag if you think mine-to-magnet execution needs an operator more than a financier.
• Tenure & skin in the game: Long-tenured board/management; the 2022/2023 SAR grants vested in Oct 2025 only after the 90-day VWAP cleared $12/$14/$16 — i.e. comp was tied to share-price hurdles that were actually hit. Insider ownership not quantified on the shelf (insider-transactions.csv absent) — n/a; performance-based options struck at a 10% premium to grant-date VWAP are a reasonable alignment design.
• Capital allocation history: This is the crux and it is mixed-to-aggressive. (1) Brilliant opportunistic financing: raised $272M via ATM into share strength + a $700M 0.75% convert at a 32.5% premium with a capped call lifting the effective strike to $30.70 — then parked ~$800M in T-bills earning more than the 1.38% convert cost. Issuing near-zero-cost paper and earning positive carry on it is genuinely shrewd. (2) Heavy, persistent dilution: shares went 157.7M (2022) → 198.7M (2024) → 240.4M (Dec 2025) → ~241.6M (Q1'26) → pro-forma >370M after VAC's 65.9M + ASM + convert (Ara Partners alone ends up ~19.9%). (3) A $1.9B acquisition of a business doing $29M EBITDA — 65× trailing EBITDA, justified only by the 12,000-tpa Sumter expansion case ($400M run-rate EBITDA someday). ROE/ROIC: deeply negative — accumulated deficit $489.7M, three of the last four years in net loss. The bet is that future integrated EBITDA validates the spend.
• Red flags: One related-party item — Saleem Drera (ex-RadTran owner, now VP Radioisotopes) holds 83% of a 2% royalty on future radium revenue + up to $14M in milestone consideration — small and disclosed. No excessive-comp or strategy-pivot-churn flags beyond the inherent acquisitiveness. The promotional cadence (a milestone press release nearly every month) is borderline but backed by real production data.

Verdict on management: A capable team that has correctly read the critical-minerals macro and is aggressively monetizing its narrative via the capital markets. The skill is real; the risk is that they are writing $2.5B of cheques (VAC + ASM + Phase 2 + mine FIDs) against a business that earns ~$0 today, and the new CEO's instinct is dealmaking, not operating.

Lens 10 · Forensic Red Flags

Accounting risk scan:

• Revenue recognition: Clean and conservative. Uranium/HMS recognized on transfer of control (shipment); term-contract variable consideration handled via the optional exemption; KPMG (auditor since 2017) issued an unqualified opinion on both the financials and ICFR. The single Critical Audit Matter is the ARO estimate ($22.2M) — judgmental but standard for a miner.
• Marketable securities / fair-value option: The biggest non-operating line. EF elected the fair-value option on ~$788M of marketable debt securities (mostly Treasuries/agencies) and runs MTM gains/losses + interest through "Other income." This inflates/deflates reported other-income with rates and is not operating cash — analysts must strip it out. The Donald JV "Advance" is also carried as a Level-3 fair-valued debt security (discount rate 6.18–6.29%) — a soft, model-dependent mark, though small ($10.5M).
• Cash flow vs. earnings divergence: Operating cash flow $(89.5M) in 2025 vs net loss $(86.1M) — i.e. the loss is roughly cash (no flattering accruals); the company is genuinely burning cash to build, funded by financing ($895M raised). This is honest but capital-intensive: EF does not self-fund — it is structurally dependent on equity/debt issuance.
• Receivables/inventory vs revenue: No red flag — receivables fell as deliveries collected; inventory roughly flat. Uranium finished-goods carried at ~$43/lb weighted-average cost (below $74 realized) — conservative.
• SBC: Rising fast — $12.6M in 2025 (from $5.4M), $3.6M in Q1'26; ~5% of the share count is reserved for equity comp. Material dilution vector but disclosed and partly performance-gated.
• Goodwill/intangibles: Modest today (RadTran IP $4.3M); watch VAC closing — a $1.9B purchase of a 100-yr-old business will create substantial goodwill/intangibles and future impairment risk if the magnet thesis disappoints.
• Going concern: Not an issue — ~$927M working capital. The opposite problem (capital deployment risk), not survival risk.

Regulatory findings (required sub-section):

• SEC Litigation Releases / AAERs: None. Verified via SEC EDGAR EFTS (LR + AAER), period 2021-06-30 → 2026-06-30: "total_sec_findings: 0".
• Non-SEC enforcement (web): Targeted search ("Energy Fuels" + lawsuit/investigation/settlement/fine/securities class action, 2025–2026) returned no securities class action or regulatory enforcement against Energy Fuels Inc. — results were all other "Energy"-named issuers (Eos Energy, New Era Energy, Just Energy).
• 10-K/10-Q Item — Legal Proceedings: Only routine/environmental matters, all characterized by the company as without merit and not material: (1) standing Ute Mountain Ute Tribe / Grand Canyon Trust / Uranium Watch challenges to White Mesa Mill air-quality orders, the groundwater-discharge permit renewal, and Alternate Feed Material license amendments (open since 2011–2021, in stipulated suspension/ALJ limbo); (2) a Kenya Ports Authority stevedoring dispute at Kwale (~$4.6M disputed, $3.2M in escrow); (3) a Kwale community/environmental petition (Mivumoni B) before the Kenyan Supreme Court (hearing March 2026, ruling pending). These are operational/ESG litigation, not accounting or securities fraud.
• Conclusion: No material regulatory or legal findings — verified via SEC EDGAR EFTS (LR, AAER), web search, and 10-Q Note 15 (Legal Proceedings) as of 2026-06-30. The books are clean and KPMG-audited; the only forensic cautions are (a) other-income distortion from the T-bill MTM book and (b) the future goodwill/impairment risk that VAC will introduce.

---

Phase D — Project & stress-test

Lens 11 · Forward Projection (FY2026–FY2028 EPS)

All `` with arithmetic shown; built bottom-up from FY2025 actuals + FY2026 guidance. EF is mid-transformation and not consensus-EPS-driven, so these are scenario sketches, not precise calls. No forecast.ts create (watchlist mode).

Key drivers: (1) uranium volume ramp (650k → 1.5–2.0M lbs guided 2026) at falling cost ($50.89/lb → $23–30/lb run-rate); (2) realized uranium price ~$71–85/lb (contract + spot mix vs ~$85 spot / ~$93 term); (3) REE = net cash drain through 2027, only "minimal" profit until 2028–2030; (4) ~$800M T-bill book earning ~$30M+/yr other-income; (5) rising share count (240M → pro-forma >370M); (6) VAC closing ~early 2027 adds ~$29M EBITDA now, ~$400M someday — but also ~$718M debt + 65.9M shares.

FY2026 (base): Uranium revenue ≈ 1.75M lbs × ~$75/lb = ~$131M; uranium gross margin ~45–55% at $30–40/lb COGS → uranium gross profit ~$60–70M. REE/HMS ≈ $0 revenue, ~$25–35M combined segment operating loss (build + reclamation). Corporate SG&A + SBC ~$60–70M. Other income ~$25–35M (T-bill carry, net of convert interest). Net result: roughly breakeven-to-modest-loss at the net line; EPS ~$(0.05) to +$0.05. The swing factor is how many spot lbs they opportunistically sell at $85. FY2026 bull: 2.0M lbs at $80+/lb + big spot sales + $40M+ other income → EPS ~+$0.10–0.15. FY2026 bear: uranium spot fades to ~$65, they hold inventory, REE cash drain widens → EPS ~$(0.15).

FY2027 (base): Mill run pauses ~H1 (guidance: next conventional run ~Q1 2027), so uranium volume may dip vs 2026 pending feedstock; REE still pre-scale. VAC closes early 2027 → consolidates ~$29M EBITDA + integration cost + interest on $718M new debt + full dilution of >370M shares. Net: likely still a net loss, EPS ~$(0.10) to $(0.20) — 2027 is the "paying for the transformation" trough year.

FY2028 (base): The first year REE could matter — expanded Phase 1 (Dy/Tb) online (2027), Donald monazite potentially flowing (H2 2027 start), VAC Sumter ramping toward 4,000 tpa. If REE+magnet EBITDA reaches ~$100–150M and uranium runs ~2M lbs profitably, EPS could turn solidly positive (~$0.15–0.40). This is the year the thesis is supposed to prove itself — and it depends on FIDs, financings, and ramps all landing.

Honest summary: EPS is not the right 2026–27 lens for this name — it is near-zero/negative by design while the company spends. The real forecast question is "does the ~$2.5B of committed/planned spend (VAC $1.9B + ASM + Phase 2 $410M + mine FIDs) convert into >$300–500M of integrated EBITDA by 2028–30 before the balance sheet or the REE-price tailwind runs out?" Base case: partially, and later than the bulls model.

Lens 12 · Bull vs Bear

Bull case. Energy Fuels is the only Western company assembling a complete, geographically-diversified rare-earth chain from mine to magnet — White Mesa is an un-replicable regulatory monopoly, and the VAC + ASM acquisitions buy the metal/alloy and magnet links that took rivals decades. Heavy rare earths (Dy/Tb) are ~99% China-controlled and structurally short; EF is the first qualified non-Chinese source, into a market (defense, EVs, robotics) where Western OEMs and governments will pay up for security of supply. Underneath the REE call option sits a newly profitable, lowest-quartile-cost uranium business (Pinyon Plain at $23–30/lb vs ~$85 spot) with six contracted utilities and 810k lbs of inventory — a real cash engine funding the build. The balance sheet is a fortress ($800M T-bills, a 0.75% convert earning positive carry), and management has shown it can finance opportunistically. At ~$3.5B market cap vs MP's ~$9.8B for a less-diversified platform, the integrated-chain re-rating is far from priced. Earnings surprise: any of (Vara Mada/Donald FID, a major magnet offtake, a US-government critical-minerals support package, a Dy/Tb price spike on a fresh Chinese export curb) could re-rate the stock hard.

Bear case (the 2–3 things that could permanently impair).

1. The REE economics never clear the cost of the build. EF is spending ~$2.5B (VAC $1.9B + ASM + Phase 2 $410M + mine FIDs) to chase a business that earned $0 in 2025 and whose own filing says profits are "minimal until… 2028–2030." If monazite feedstock (Vara Mada is in coup-prone Madagascar; Donald is pre-FID) doesn't arrive, or if China floods the Dy/Tb market and collapses the scarcity premium (it has repeatedly used export policy as a weapon in both directions), the chain becomes a stranded, high-cost asset. The moat is rented from Beijing's quota policy.
2. Dilution + leverage permanently caps per-share value. Shares go from 157.7M (2022) to pro-forma >370M — a ~2.3× increase in four years — plus $700M of converts (dilutive at $20.34) and $718M of new VAC debt. Even if the enterprise succeeds, the per-share outcome can disappoint; the market already punished the June VAC dilution.
3. Single-asset concentration. Both the uranium and REE businesses funnel through one mill in one Utah county under licenses that tribes and NGOs have litigated for 15 years. One serious regulatory/operational hit at White Mesa impairs the whole company.

Pre-mortem (it's Dec 2027, the thesis broke — what happened?): Uranium spot faded to the $60s as the speculative bid unwound; the Mill paused its conventional run on schedule and uranium revenue dipped; Vara Mada's FID slipped again on Madagascar politics and Donald's financing stretched; China announced a Dy/Tb export-quota increase, halving heavy-REE prices and erasing the "only Western source" premium; VAC integration cost more and ramped slower than $400M-EBITDA dreams; and the stock sits at ~$8 with >370M shares, the REE option looking like an expensive, debt-laden science project rather than a cash machine.

Are multiples too high? There is no earnings multiple to be too high — the question is whether ~$3.5B equity (vs ~$1.4B of assets, ~$800M of which is just Treasuries) fairly prices the REE/magnet option. Relative to MP/Lynas the optionality looks reasonably priced; on a sum-of-parts it is rich unless REE EBITDA actually materializes by 2028.

Contrarian view (what the market refuses to see): The market treats EF as a rare-earth story and prices the uranium business as a free option. The contrarian read is the inverse — the uranium business is the only part that is real, profitable, and de-risked today, and the rare-earth empire is a capital-destroying narrative that consumes the uranium cash flows and dilutes shareholders to chase a Chinese-policy-dependent premium. If you're bullish UUUU, be bullish because Pinyon Plain is one of the cheapest uranium mines on earth into a tight market — not because magnets will save the day on the bulls' timeline.

Lens 13 · Devil's Advocate (short-seller)

Dismantling the bull case.

• Revenue concentration today is ~100% uranium, ~100% US, ~6 utility contracts. The "diversified critical-minerals platform" is aspirational — strip the narrative and you have a small uranium miner selling 650k–2M lbs/yr, with everything else a cash-consuming promise. If uranium spot mean-reverts (it pulled back from $100 to $85 in H1 2026 already), the only real business compresses.
• The moat is rented from China. EF's entire heavy-REE premium exists because Beijing restricts Dy/Tb exports. China has cycled export policy loose→tight→loose for 15 years. A single quota loosening (entirely within Beijing's gift, and a plausible trade-war bargaining chip) structurally breaks the "only Western source" pricing — and EF will have spent $2.5B building capacity into a market that no longer pays a security premium.
• The most dangerous competitor bulls underestimate: MP Materials. MP has a producing US mine (Mountain Pass), operating US magnet capacity (Independence/Fort Worth), DoD backing, and a head start on scale — at ~$9.8B it's already further down the integrated path. EF is buying (VAC/ASM) what MP is building, at 65× EBITDA, with borrowed money and printed shares.
• Capital allocation is the short thesis. Paying $1.9B for $29M of EBITDA (65×) is defensible only on a heroic 12,000-tpa Sumter ramp years out. Funding it with $718M debt + 65.9M shares (Ara ends up owning ~19.9% — a PE seller cashing out into EF stock) while the core business loses money is exactly the serial-acquirer-funded-by-dilution pattern that ends badly when the narrative stalls. The promotional monthly milestone cadence supports a stock that needs high prices to keep printing acquisition currency.
• Assumptions that must hold for ~$3.5B: uranium stays >$75, Dy/Tb stay China-short, Vara Mada and Donald reach FID and produce on time, Phase 2 ($410M) is financed and built by 2029, and VAC integrates to ~$400M EBITDA. That's five independent things, several outside management's control.
• −20–30% growth-disappointment scenario: if REE ramps slip and uranium softens, FY2027–28 EBITDA stays near zero against >370M shares and ~$700M+ of debt — the equity is then a leveraged option on a Chinese policy decision, worth a fraction of $3.5B.
• Single scenario that permanently impairs: a White Mesa license revocation/major operational halt (the tribes/NGOs have litigated for 15 years) — low probability, catastrophic, and it takes both segments down at once.

Short-seller's bottom line: Not a fraud (clean books, real assets, real Dy/Tb production). The short case is valuation + capital-destruction + China-policy dependence: a profitable small uranium miner whose stock prices a $300–500M integrated-REE-EBITDA future that is years away, financing-dependent, and contingent on Beijing's export quotas staying tight.

Lens 14 · Management Questions (ordered by information value)

1. The VAC purchase price implies ~65× 2025 EBITDA. What integrated EBITDA, by what year, underwrites that price — and what happens to the deal economics if Sumter only reaches 4,000 tpa, not 12,000?
2. Heavy-REE (Dy/Tb) prices depend on Chinese export quotas. What does the entire REE/magnet investment case look like if China loosens Dy/Tb exports and the price halves — and how much of your $2.5B planned spend is reversible at that point?
3. Vara Mada and Donald are both pre-FID and politically/financing-exposed. What is your honest probability and timing for each reaching production — and what is the REE plan if neither monazite source arrives on schedule?
4. You're committing ~$2.5B (VAC + ASM + Phase 2 + FIDs) while the company earns ~$0. Walk me through the funding stack and the maximum additional dilution shareholders should expect through 2028.
5. Pro-forma shares go from ~240M to >370M. What is your framework for protecting per-share value, not just enterprise scale, given the market's negative reaction to the VAC dilution?
6. Why is the right structure to own magnet manufacturing (VAC) rather than contract it via offtake — what specifically does owning 100-year-old European magnet plants add that long-term supply agreements wouldn't?
7. The Mill can't run conventional uranium and Phase-1 REE simultaneously, and the next conventional run is paused to ~Q1 2027. How do you manage the uranium-cash-flow gap while prioritizing the REE build?
8. Roughly $800M sits in Treasuries earning carry over a 0.75% convert. At what point — and into what — do you deploy that, and what's the hurdle rate before you'd rather buy back stock than make another acquisition?
9. Ross Bhappu comes from mining private equity, not operations. As the strategy shifts from acquiring the chain to operating it (separation, metallization, magnet-making), where does the operating bench come from?
10. Your uranium contracts delivered at $71/lb in 2025 vs $85+ spot. What is the contract-vs-spot strategy from here, and how much 2026–27 volume is uncommitted and exposed to spot?
11. White Mesa faces 15 years of standing tribal/NGO license challenges. Quantify the worst realistic regulatory outcome and its cost/timeline — and what is the contingency if the Mill's license or AFM authority is curtailed?
12. ARO is undiscounted ~$48.7M against ~$22.7M posted collateral. How does the reclamation liability evolve as you add international HMS mines, and is the collateral gap fully funded?
13. What is the realistic commercial timeline and addressable revenue for the TAT/radium (RadTran) initiative — or is it an option you're carrying cheaply rather than a near-term business?
14. If uranium spot returned to the $50s, which growth initiatives get cut first, and what is the minimum uranium price at which the integrated strategy still makes sense?
15. Five years out, is Energy Fuels primarily a uranium producer, a rare-earth/magnet company, or a diversified critical-minerals major — and which single metric should shareholders judge that transition by?

---