Centrus Energy

Phase A — Understand the business

Lens 1 · Company Overview

Centrus is a nuclear-fuel enrichment company with two faces: a profitable, cash-generative fuel-brokerage business living off a legacy Russian supply contract, and a capital-intensive, government-funded re-industrialization bet to rebuild U.S.-owned uranium enrichment from scratch. The market is paying for the second; the cash comes entirely from the first.

It reports two segments:

• LEU segment (~77% of FY2025 revenue) — sells the enrichment component of nuclear fuel (measured in SWU, "separative work units") plus natural uranium hexafluoride, primarily to ~utilities operating commercial reactors. Centrus does not currently enrich the LEU it sells — it is a middleman that buys SWU under long-term contracts (largest: Russia's TENEX; second: France's Orano) and resells to utilities under medium/long-term fixed-commitment contracts. Margins here have been structurally inflated since a one-time 2018/2019 market-price reset on the TENEX contract locked in low purchase costs while spot SWU prices ran 5x. International sales were ~33% of LEU revenue since 2023, rising to ~44% in Q1 2026.
• Technical Solutions segment (~23%) — the actual enrichment + advanced manufacturing arm. This runs the HALEU Operation Contract with the DOE (cost-plus-incentive-fee) at the Piketon, Ohio demonstration cascade (16 AC100M centrifuges), plus engineering/manufacturing at Oak Ridge, TN. This is where the future is being built.

Contract structure / key terms:

• LEU sales: medium/long-term, fixed-commitment contracts where utilities are obligated to buy a specified SWU quantity; quantities/prices have some escalation and ±range flexibility. Average SWU order ~$10.2M in FY2025 / ~$13.7M in Q1 2026 — highly lumpy (a single order shifting quarters swings results).
• Government: HALEU work is cost-plus-incentive-fee; the DOE owns the HALEU produced from the demo cascade; the DOE bears all D&D (decommissioning) liability on the leased Piketon facility.
• Backlog: $3.8–3.9B total across both segments, extending to 2040. But $2.3B of the LEU backlog is "contingent" — it only converts if Centrus secures the public + private capital to build new enrichment capacity. That contingency is the whole story.

HQ Bethesda, MD; 467 employees (up from 322 a year earlier — a hiring ramp). The corporate lineage is USEC Inc. (privatized from the U.S. government in 1998), reorganized as Centrus in 2014.

Lens 2 · Supply Chain

Map the chain — names, not categories:

Upstream (SWU supply into Centrus):

• TENEX (subsidiary of Rosatom, Russian state nuclear) — largest supplier. Centrus buys SWU-contained-in-LEU; delivers natural uranium hexafluoride back to TENEX. Contract runs through 2028. Through 2027, "well over one-half" of the LEU Centrus delivers to customers is TENEX-sourced. This is the central single-source dependency and the single biggest risk in the entire company.
• Orano (French-government-owned) — second supplier. SWU-in-LEU, deliveries 2023→2030, priced in USD + EUR with floors/ceilings, annual min/max volume flexibility.
• Secondary sources: inventory, spot purchases of SWU/uranium/LEU, SWU borrowings from fabricators/utilities.
• Transport chokepoint: Russian LEU is shipped by a Canadian carrier using Chinese-built vessels. Requires a Canadian permit (extended to March 2027) and is currently exempt from new USTR Section 301 fees on Chinese-built ships — if that exemption is removed, the route becomes "cost prohibitive".

The company itself: demonstration enrichment at Piketon (HALEU); centrifuge manufacturing at Oak Ridge; no commercial-scale enrichment yet.

Downstream (customers): domestic + international utilities operating commercial reactors. In Q1 2026, three LEU customers were each >10% of revenue ($18.5M / $14.2M / $8.8M) and one Technical Solutions customer (the DOE) was $31.5M — real customer concentration. Forward: advanced-reactor developers — the first named commercial HALEU offtake is Oklo (supply for up to five Aurora powerhouses, deliveries from 2029).

The chokepoint that defines the thesis: Centrus sits downstream of a Russian state enricher that is simultaneously (a) being legally banned as an import source (U.S. Import Ban Act, effective Aug 2024, waivers through 2027) and (b) being restricted as an export source (Russian Decree, per-shipment licenses through Dec 2027). Both ends of Centrus's largest supply line are on a regulatory clock that runs out end-2027/2028, and Centrus's own replacement capacity (Piketon enrichment) will not exist at scale until ~2029+. This timing gap is the bear case's home.

Lens 3 · Competitive Advantages (moats)

The moat is regulatory and strategic, not yet operational — and that distinction is everything.

Real, durable advantages:

1. The only deployment-ready U.S.-origin enrichment technology. Longstanding U.S. policy + nonproliferation agreements prohibit foreign-origin enrichment tech for U.S. national-security missions. Urenco and Orano operate on U.S. soil but with foreign-origin technology. The AC100M centrifuge is, per the filing, "the only deployment-ready U.S. uranium enrichment technology that can meet these national security requirements." This is a genuine, legally-fenced moat for the national-security/defense-adjacent slice of demand.
2. The only NRC license to actively enrich up to 20% U-235 (HALEU). Centrus is the only operator producing HALEU at any scale in the U.S. With 9 of 10 DOE-selected advanced-reactor designs requiring HALEU, and HALEU named the #1 obstacle ("keeps you up at night") in U.S. Nuclear Industry Council surveys, this is a real first-mover position. HALEU market projected ~$0.26B (2025) → ~$6.2B (2035).
3. Section 382-protected NOLs — $355.4M valuation allowance against federal/state deferred tax assets; a Rights Plan (poison pill) extended to June 2029 protects the NOL shield. Worth real money if the company becomes durably profitable.

Where the moat is thin:

• Global LEU market share <5% of a ~50M SWU/yr market. Centrus is a price-taker in commercial LEU, not a price-setter.
• Bargaining power is weak on the supply side — it needs TENEX more than TENEX needs it, and TENEX/Russia have repeatedly demonstrated willingness to weaponize the relationship (the Russian Decree). On the customer side, the contingent backlog gives some forward visibility but real pricing power only emerges if and when it owns enrichment capacity.
• The HALEU moat is a moat over a market that barely exists yet — current production is ~900 kg/yr from a 16-machine demo cascade, "far below the multi-ton demand from planned reactors". The moat is real but the castle is a model home.

Critical competitive correction (bull-case dent): Centrus is NOT the sole DOE enrichment champion. In January 2026 the DOE awarded a separate $900M LEU task order to Orano Federal Services for "Project IKE," a ~$5B gas-centrifuge LEU plant on former Manhattan-Project land in Oak Ridge, TN. Urenco USA (the NEF plant in New Mexico, ~4.3–4.9M SWU/yr) is privately expanding +700k SWU by 2027. So in commercial LEU, Centrus faces two better-capitalized, already-operating rivals on U.S. soil. Its uncontested turf is specifically U.S.-origin-technology HALEU + national-security enrichment — a narrower moat than the "America's enrichment champion" headline implies.

Lens 4 · Segments

FY2025 vs FY2024, $M:

What's happening underneath the flat top line:

• SWU revenue +21% on +23% volume, −1% price. The growth is volume, not price — and unit cost of SWU fell 13%, which is why LEU gross profit jumped +19% to $111.5M. The TENEX 2018-reset cost advantage is still the margin engine.
• Uranium revenue −54% — lumpy, low-margin uranium component sales swing year to year; their absence is why total revenue looks flat despite the SWU surge.
• Technical Solutions gross profit collapsed −66% to $6.0M. Cause: the HALEU Phase 2 delay (DOE failed to supply enough "5B Cylinders" to store output) pushed performance past Nov-2024, and costs incurred after Nov-2024 carry no fee yet ("undefinitized, subject to negotiation"). So the segment is currently running near break-even — Centrus is funding the U.S. enrichment build-out and the government has not yet agreed what it will pay. This is the segment that is supposed to be the future, and it is the one bleeding margin.

Geography: all long-lived assets are U.S.-based. Revenue split: Q1 2026 SWU/uranium revenue was $17.3M U.S. / $27.3M foreign — i.e., foreign flipped to a majority of LEU sales in the quarter, a swing worth watching given the Russian-transit dependency on foreign re-export.

Trend read: the commercial engine (LEU/SWU) is accelerating on volume and benign cost; the strategic engine (Technical Solutions) is decelerating in margin precisely as spend ramps. The company is mid-transition from "milk the Russian contract" to "build American capacity," and FY2025 is the seam.

---

Phase B — Measure performance

Lens 5 · Earnings Result

Latest print — Q1 2026 (quarter ended 2026-03-31):

• Total revenue $76.7M vs $73.1M (+5% YoY) — missed consensus ~$81.4M.
• Net income $10.0M vs $27.2M PY. The drop is quality-driven: PY had an $11.8M debt-extinguishment gain (8.25% Notes redemption); this year advanced technology costs jumped to $18.9M from $3.0M — the Piketon/Oak Ridge ramp is now hitting the P&L hard.
• Diluted EPS $0.45 vs $1.60 PY; basic $0.51 vs $1.60. Missed consensus ~$0.58 (−22%). (Note a conflicting headline claimed an "88% EPS beat" — this contradicts the Investing.com miss and the YoY decline in the filing; I treat the print as a miss vs. the lowered consensus, consistent with the 10-Q's own numbers.)
• Gross profit $31.5M vs $32.9M; segment GP: LEU $30.1M (incl. intersegment) / Technical Solutions $3.6M.
• Diluted share count 22.4M vs 17.0M PY — a +32% YoY increase in the diluted denominator, driven by ATM equity issuance + convertible-note dilution now counting (2.46M shares from the 2.25% converts). Dilution is real and ongoing.

Guidance: FY2026 revenue guidance raised to $450–500M (from $425–475M). Backlog $3.9B as of 2026-03-31.

FY2025 full-year (context):

• Revenue $448.7M (+2%); operating income $50.2M (+5%); net income $77.8M (+6%).
• Net income flattered by: investment income $44.7M (+247%, from $2.0B cash earning money-market yield) and the $11.8M debt-extinguishment gain; dragged by interest expense $14.0M (+419%, the new converts) and a swing to a $6.8M nonoperating pension loss.
• Balance sheet is the standout: $1,957.2M cash, working capital $1,940.8M (vs $668.4M PY). Inventories $130.2M; receivables $30.7M (down from $80.0M). Total debt $1,207.5M — $402.5M 2.25% converts (due 2030) + $805.0M 0% converts (due 2032), no principal due before maturity. Net cash ~$750M.
• Cash flow: operating $51.0M, investing −$19.7M (capex only $19.7M — the spend is still ahead of them), financing +$1,224.9M (converts + ATM equity).

Unusual vs. own history: investment income is now a material earnings driver — roughly half of pre-tax income is the cash pile earning yield, not the core business. That is high-quality optionality (dry powder) but low-quality earnings — strip it out and the operating business earned ~$50M pre-tax. The market is not paying 50x for $50M of enrichment-brokerage EBIT; it is paying for the build-out option.

Lens 6 · Earnings Calls (sentiment trend)

No transcripts on the research shelf (transcripts/ empty) — this lens is ``.

Management tone over the last several quarters has been consistently and increasingly bullish on the strategic transformation, with a deliberate pivot in language from "demonstration" to "commercial scale". CEO Vexler and CFO Tinelli framed the Q4/FY2025 call as recapping "a remarkably successful 2025" and previewing scale-up. Recurring themes: the $900M DOE HALEU task order (Jan 2026), the Oak Ridge $560M+ manufacturing investment, the Piketon LEU+HALEU expansion, and the $2.3B contingent backlog as the demand proof point.

What they say more of: "first-mover," "U.S.-owned," "commercial-scale," "national security," "thousands of centrifuges." What's conspicuously hedged: every expansion statement is wrapped in "subject to the availability of public and private funding" and "no assurance" language. The Q1 2026 call had to address a top-and-bottom-line miss while raising full-year guidance — management leaned hard on backlog and the DOE award to redirect from the soft quarter. Net: tone is genuinely confident on the secular story, but the filing's risk language is the tell that even management treats the build-out as contingent, not committed.

Lens 7 · Comps

Peer multiples are `` with source/date, or n/a. There is no clean operating peer — Centrus is sui generis (a sub-scale enricher with a government-funded option), so the table mixes uranium miners, an enricher-adjacent set, and pre-revenue reactor names for context. Do not over-anchor on these — the read is "LEU is priced far above any uranium/nuclear comp on earnings."

• Peer-average P/E cited at ~11.5x vs LEU's ~50x+. LEU trades at a ~4–5x earnings premium to its nominal peer set.
• Consensus 12-month price target ~$279–285 avg (range $137–410) across ~13 analysts; consensus Buy (≈11 buy / 0 sell / ~5 hold — the buy/hold count is noisy across sources). UBS cut its target to $170 (from $195).

Read: there is no defensible multiple-based valuation for LEU. It is an optionality stock — the price embeds the present value of a build-out that mostly does not exist on the balance sheet yet. On trailing earnings it is expensive by any nuclear/uranium yardstick.

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

Mostly ``; the pattern is the point.

• 2025: +385% full-year rally — the dominant move, driven by (a) the Russia import-ban repricing of Western enrichment, (b) AI-datacenter nuclear-demand narrative, (c) sequential DOE contract wins. ATH $250.88 on 2025-07-21 (+407% off the April low).
• Jan 5, 2026 — $900M DOE HALEU task order: the single biggest fundamental catalyst; "changed the investment story".
• June 2026 — Oklo HALEU supply agreement: stock "climbs" — first marquee commercial offtake.
• −26% drawdown into Apr 2026 and the Q1 2026 top/bottom-line miss — the counter-moves, showing the stock now reacts to execution/quarterly delivery, not just policy headlines.
• Recurring driver historically: anything Russia/sanctions (import-ban news, the Russian Decree, waiver grants) and anything DOE (RFP selections, task orders, funding).

What the tape reveals: for ~3 years LEU traded as a policy/sanctions option — it moved on Washington and Moscow, not on EPS. The Q1 2026 miss is the first sign the market is starting to demand delivery alongside the narrative. The stock is a high-beta expression of "U.S. re-shoring of nuclear fuel" — it will keep moving violently on DOE funding headlines and any TENEX disruption, in both directions. Short interest ~22.7% (most-shorted energy name) adds squeeze risk on good news and air-pocket risk on bad.

---

Phase C — Judge people & books

Lens 9 · Management

CEO — Amir V. Vexler (53; President & CEO since 2024-01-01, joined as Special Advisor Dec 2023).

1. Track record: Genuinely the right résumé for this moment — prior CEO of Orano USA (the exact competitor now building Project IKE) and ~20 years at GE including CEO/Chairman/COO of Global Nuclear Fuel (the GE–Hitachi JV). He has run nuclear-fuel manufacturing, commercial ops, and federal engineering services. For a company whose entire thesis is "industrialize enrichment," a centrifuge/fuel operator is the correct archetype.
2. Tenure & skin in the game: ~2.5 years in seat — short, and over a period of huge stock appreciation, so his ownership is mostly equity comp, not founder stock. Reported aggregate "insider" ownership figures are unreliable/conflicting across sources (54% "insider" in one tally almost certainly conflates the Class B holders / large institutions and an old register) — treat the headline insider % as n/a — not reliably sourced.
3. Capital allocation on watch: the defining moves are (a) raising $805M in 0% converts (Aug 2025) + $402.5M 2.25% converts — cheap, non-dilutive-until-deep-ITM financing, a genuinely shrewd use of a hot stock and low effective cost; (b) $524.7M raised via ATM equity in 2025 at high prices — opportunistic but dilutive (+32% diluted share count YoY); (c) redeeming the 8.25% Notes early for an $11.8M gain. Building a ~$2.0B war chest before the capex hits is defensible treasury management. ROE/ROIC is not yet a meaningful gauge — earnings are dominated by investment income and the operating base is small.
4. Red flags: insiders are net sellers — 3 selling, 0 buying in the last 12 months, ~$1.06M sold (CFO Tinelli, CMO Donelson, others at $96–204). Not alarming in size, but zero insider buying into a 385% run while issuing equity is a mild tell. A $50M buyback was referenced in one source — this conflicts with the 10-K's explicit statement that there were no share repurchases in 2025; I trust the filing — treat the buyback claim as unverified/likely an authorization, not execution.
5. Archetype: professional turnaround/scale operator, not founder. Correct for the industrialization stage; the risk is that professional managers building government-funded megaprojects have every incentive to keep raising and spending regardless of returns.

CFO — Todd Tinelli (joined Aug 2025, ex-Sprague Resources/Hartree, energy trading/treasury background) — a trading/liquidity CFO, fitting for a company managing a $2B cash book and commodity-linked contracts. SVP Field Ops Patrick Brown (nuclear engineer, ex-Urenco New Mexico commissioning, U.S. Navy nuclear) — directly relevant operational hire.

Lens 10 · Forensic Red Flags

Acting as a forensic analyst across the three statements:

• Revenue recognition (Technical Solutions): cost-to-cost / cost-plus percentage-of-completion with significant estimation of total costs at completion. The undefinitized HALEU Phase 2 fee (costs incurred post-Nov-2024 carry no agreed fee) means revenue/margin on that work rests on management's estimate of what the DOE will eventually pay — a soft spot, and the visible cause of the −66% TS gross-profit drop. Watch for catch-up adjustments (positive or negative) when the fee is definitized.
• Cash vs. earnings divergence: FY2025 net income $77.8M but operating cash flow only $51.0M. The gap is benign and explained — net income includes the non-cash $11.8M debt-extinguishment gain and the timing of deferred-revenue drawdown ($90.2M deferred-revenue liability, down from $152.5M, i.e. the company delivered against prepayments, consuming a cash-flow tailwind from prior years). Not a manipulation flag, but it means reported EPS is higher quality on paper than in cash this year.
• Receivables/inventory vs. revenue: receivables fell to $30.7M from $80.0M while revenue was flat — good (no channel-stuffing signal; collections strong, ~$450M cash collected). Inventory fell to $130.2M. Clean.
• SBC / non-GAAP flattering: equity comp only $5.8M (incl. a one-off $3.6M reclass of board RSUs from equity to liability) — small, not a non-GAAP-flattering vehicle. Refreshingly un-egregious vs. typical high-multiple names.
• Intangibles/goodwill: $21.2M identifiable intangibles (fresh-start backlog + customer relationships from the 2014 reorg), amortizing, ~3¾ yrs remaining on customer relationships. Immaterial impairment risk.
• Tax/valuation allowance: $355.4M valuation allowance remains against deferred tax assets; partial releases ($10.2M in 2025) are judgment-driven off LEU backlog visibility. A future large release would flatter GAAP EPS without cash — watch for it as a non-recurring earnings boost.
• Pension: mark-to-market accounting recognizes actuarial gains/losses immediately in Q4 → can swing results materially. Obligations largely de-risked via 2023–2024 annuitizations (~$420M transferred to insurers). Lower go-forward volatility, but the immediate-recognition policy means a bad-return year hits the P&L.
• Off-balance-sheet: only the SWU purchase commitments (TENEX/Orano) + the DOE technology-license royalty (1–2% of DOE-tech SWU revenue, $100k min, $100M lifetime cap). Disclosed, modest.
• Going concern: none — $2.0B cash, stated adequate liquidity for ≥12 months. Altman Z ~2.75 ("some stress") reflects the small operating base relative to market cap, not distress.

Regulatory findings (required sub-section):

• SEC Litigation Releases / AAERs: None. "No LR found" and "No AAER found" for Centrus Energy 2021–2026 via SEC EDGAR EFTS.
• Non-SEC enforcement (web search per the file's guidance): no material FTC/DOJ/FDA/CFPB consent decrees, settlements, fines, or penalties surfaced for Centrus Energy. The Russia exposure is managed through the legal waiver/license regime (DOE import waivers, Russian export licenses), not enforcement actions against Centrus.
• 10-K Item 3 / Note 17 Legal Matters: a cluster of toxic-tort cases tied to the legacy Portsmouth GDP / Piketon site — class action (McGlone, off-site contamination, 7-mile radius + Zahn's Corner Middle School) and individual wrongful-death/cancer suits (Lykins, Rose, Shaw, Dunham, several involving leukemia/childhood cancer). Centrus + Enrichment Corp. + 5–6 other DOE contractors are co-defendants. Centrus asserts Price-Anderson Act indemnification by the DOE and has filed the required notifications; courts have dismissed some claims and let others proceed (discovery ongoing). Separately, the DOE asserts a ~$9.6M Power-MOU claim (Joppa Power Plant D&D) — no liability accrued, merits not yet assessed. Company position: not expected to be individually or in aggregate material.
• Net: No accounting or securities red flags. The legal exposure is real but (a) legacy/government-era, (b) Price-Anderson-indemnified, (c) shared across many contractors. The forensic verdict is clean books, genuine legacy-environmental tail risk.

---

Phase D — Project & stress-test

Lens 11 · Forward Projection

Bottom-up from FY2025 actuals + FY2026 guidance. Every input labeled; output ``. Per the --watchlist rule, no forecast.ts create is logged in this loop.

Anchors: FY2025 revenue $448.7M / net income $77.8M / diluted EPS basis. FY2026 revenue guide $450–500M. Diluted shares ~22.4M and rising (ATM + converts). Investment income ~$45M/yr on the $2.0B cash. Critical modeling fact: advanced-technology + Technical Solutions ramp costs are rising sharply (ATC $18.9M in Q1 2026 alone vs $16.9M for all of FY2025) — the build-out suppresses near-term EPS even as revenue grows.

The honest projection range (fiscal years 2026 / 2027 / 2028; EPS = diluted, GAAP-ish, $/sh):

The decisive point: on a base case, LEU's EPS is flat-to-declining for 2–3 years because the company is spending to build capacity that won't generate revenue until ~2029+. At ~$251 and ~50–60x trailing earnings, the stock is not valued on FY2026–2028 EPS at all — it is valued on the terminal enrichment-capacity option (the bull path). The base-case Brier-style call I would log (not logged here per loop rules): "LEU FY2027 GAAP diluted EPS < FY2025's level (i.e. EPS does NOT grow through the build-out), p≈0.65" — i.e. I expect the earnings to disappoint even if the story progresses, because that is what funding a multi-billion-dollar plant does to a $450M-revenue company.

Lens 12 · Bull vs Bear

Bull case. The West has a structural, government-mandated need to rebuild non-Russian — and specifically U.S.-owned — enrichment, and Centrus is the only domestic player with deployment-ready U.S.-origin centrifuge tech and the sole active NRC HALEU license. The Import Ban Act forcibly removes ~Russian supply by 2028 from a market that is undersupplied without it, SWU spot has repriced ~5x to ~$200/SWU and stays structurally high, and the AI-datacenter power build-out is conjuring a wave of advanced reactors that only run on HALEU (9 of 10 DOE designs). The $900M DOE task order + Oklo offtake + $2.3B contingent backlog are the first hard proof the demand converts. With ~$2.0B cash, cheap 0%/2.25% convert financing, and §382-protected NOLs, Centrus can fund the first cascades and inflect into a high-margin, sovereign-critical enrichment monopoly-of-one for the national-security/HALEU niche. If even the base build-out lands, today's market cap is small relative to a domesticated multi-billion-dollar SWU/HALEU franchise.

Bear case (permanent-impairment risks).

1. The 2028 supply cliff with no bridge. TENEX supplies "well over half" of delivered LEU through 2027 and the contract dies in 2028; Centrus's own replacement capacity is ~2029+ and contingent on funding it doesn't have. A gap year (or a sudden Russian/U.S. action that severs TENEX early — Russia has already weaponized this once via the Decree) would gut the only segment that actually makes money. This is a genuine, identifiable path to a step-down in earnings power, not just a wobble.
2. It is not the chosen LEU champion. Orano (Project IKE, $900M DOE LEU award, ~$5B Oak Ridge plant) and Urenco USA (operating + expanding) are better-capitalized, already-enriching rivals on U.S. soil. Centrus's uncontested moat is the narrow U.S.-origin-tech/HALEU slice — much smaller than the "America's enricher" valuation implies.
3. Valuation is a terminal-option bet on a still-tiny company. ~50–60x trailing EPS, ~22% short interest, EPS likely flat-to-down through 2028, DCF base-case intrinsic values cluster far below price ($47–48 in one bear DCF ). The entire premium is the build-out option, and that option is gated on DOE appropriations (annual, discretionary, politically exposed — Executive Order 14154/14272 risk to IRA credits) and on private capital Centrus has not yet raised.

Pre-mortem (18 months out, thesis broke): It's late 2027. The $900M task order stalled in definitization; DOE appropriations slipped under budget pressure; the §48C credit got tangled in EO 14154; a TENEX shipment disruption clipped 2027 LEU deliveries; FY2026–2027 EPS came in down year-over-year as ATC costs ran past $80M while commercial cascades remained years away. The multiple compressed from 55x toward the peer ~12x as the market re-rated it from "enrichment monopoly option" to "sub-scale broker funding a megaproject," and the 22% short base pressed it. The stock is 50–70% lower despite the narrative being intact — because the earnings and funding slipped while the price had discounted perfection.

Are multiples too high? On any earnings basis, yes — indefensibly so vs. comps. They are only justifiable on a probability-weighted terminal build-out, and the market is currently weighting that closer to the bull path than the base path.

Contrarian view (what the market refuses to see): The market is treating Centrus as the sovereign U.S. enrichment champion and pricing near-certainty of the build-out. What it is underweighting is that (a) the DOE explicitly chose Orano for the big LEU task order, (b) Centrus's profitable core is a Russian-supply brokerage on a 2028 fuse, and (c) the company must still raise large private capital and definitize a government contract before a single commercial cascade exists. The bear's edge is not "nuclear won't happen" — it's "Centrus's earnings get worse before they get better, and the price has no room for that."

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

Dismantling the bull case as a skeptical short:

• Where revenue is concentrated / what breaks it: ~77% of revenue is LEU, "well over half" of which is resold Russian TENEX SWU on a contract expiring 2028, under an active U.S. import ban (waivers only to 2027) and a Russian export-license regime (to 2027). Three customers were each >10% of Q1 revenue. The profitable business is a concentrated, sanction-fused, single-supplier brokerage — the most fragile possible foundation under a $3.6B market cap.
• Why the moat is weaker than bulls think: "Only NRC HALEU license / only U.S.-origin tech" is real but addresses a market worth ~$0.26B today. In the large commercial-LEU market Centrus is <5% share and now faces Orano's DOE-funded Project IKE and an expanding Urenco USA. The moat is over the pond, not the lake.
• Most dangerous competitor bulls underestimate: Orano — Centrus's CEO's former employer, now wielding a $900M DOE LEU award and a $5B Oak Ridge plant. Bulls frame Orano as a supplier; it is also the better-funded competitor for the exact commercial-LEU prize.
• Worst capital-allocation / incentive issues: +32% YoY diluted share count from ATM + converts; insiders selling (zero buying) into a 385% run; a "$50M buyback" claim that the 10-K contradicts; reliance on issuing stock at peak to pre-fund a project whose returns are unproven.
• Assumptions that must hold for today's price: (1) DOE appropriations keep flowing and the $900M task order definitizes + funds; (2) Centrus raises large private capital on top; (3) commercial cascades hit ~2029 on budget; (4) SWU prices stay elevated; (5) no early TENEX rupture; (6) IRA §48C credit survives EO 14154. All six must hold. Each is individually plausible; jointly they are a narrow needle.
• If growth disappoints 20–30%: since the valuation is terminal-option, not near-earnings, a 20–30% haircut to the build-out's expected scale/timing doesn't trim EPS — it re-rates the multiple, and from 55x the downside to a 12–20x "sub-scale enricher" frame is 60–75%.
• Single scenario that permanently impairs the business: an abrupt, permanent TENEX cut-off (U.S. sanctions Rosatom entirely, or Russia stops export licenses) before domestic capacity exists — collapsing the only profitable segment for years, forcing reliance on a thinner Orano contract + spot, right as the company is burning on capex. Plausibility: not the base case (both sides have so far kept the contract limping), but materially non-zero and rising into 2027–2028 — and the price embeds ~none of it.

Lens 14 · Management Questions (ordered by information value)

1. The $900M DOE HALEU task order is "subject to negotiation of a definitive agreement" — what is the realistic timeline to definitize and fund it, and what scope/price terms would make you walk away?
2. Of the ~$2.3B contingent LEU backlog, how much converts to firm, and what private capital quantum must you raise (beyond DOE) to unlock it — and on what timeline and instrument?
3. Walk us through the 2028 TENEX cliff precisely: what volume gap opens in 2028–2029 between TENEX expiry and your own commercial cascade output, and how is it bridged (Orano max volumes, spot, inventory)?
4. What is the all-in capex for commercial-scale LEU+HALEU at Piketon, the funding stack (DOE / converts / equity / partners), and at what cumulative spend do you reach first commercial SWU?
5. Given Orano won the $900M DOE LEU task order for Project IKE in Oak Ridge, how do you see the U.S. commercial-LEU market splitting between Centrus, Orano, and Urenco USA — where do you actually win?
6. The HALEU Phase 2 fee for post-Nov-2024 work is still undefinitized — what is the expected fee outcome, and what is the downside if the DOE definitizes it below your accrued estimate?
7. How exposed are you to Executive Orders 14154/14272 on the §48C credit and the Section 232 critical-minerals tariff track — what's the base case for monetizing the $62.4M credit?
8. With ~$2.0B cash earning ~$45M/yr of investment income, how much is committed to the build-out vs. available, and at what point does the cash pile start drawing down materially?
9. You issued $524.7M of ATM equity in 2025 — what is your go-forward dilution philosophy, and is there a stock price at which you stop issuing?
10. Insiders were net sellers with zero buying through 2025 — and a "$50M buyback" was referenced that the 10-K says didn't happen in 2025. Can you reconcile that, and is a buyback authorized?
11. What is the realistic annual SWU/HALEU output trajectory of the first commercial cascade, year by year, from first production to nameplate?
12. If a U.S. sanction permanently severed Rosatom/TENEX tomorrow, what is the quantified earnings impact in 2026, 2027, and 2028, and your mitigation?
13. Which advanced-reactor offtakes beyond Oklo are closest to firm HALEU contracts, and what volumes/start dates?
14. What ROIC do you underwrite the commercial enrichment build-out to, at what assumed long-run SWU price, and what SWU price breaks the economics?
15. What is your contingency if DOE appropriations for the HALEU/LEU programs are cut or continuing-resolution-frozen in the next federal budget cycle?

---