LEU › analyze
LEU · Analyze from before
Date: 2026-08-27 | Price: ~$194.78 (post-mkt $195.48) | 52wk Range: $142.13–$464.25 | Market Cap: ~$3.89B | Sector: Energy / Uranium (Enrichment & Technical Solutions)
This updates: Output/Stocks/Energy/LEU/analyze-2026-06-11.md (2026-06-11, [7.5] Add at $145–170). No health/moat/value/sentiment file was written on LEU between the baseline and this pass, and no second prior verdict-bearing report exists — this is the first differential pass.
A portfolio-specific passage was removed from the public build.
Event list since 2026-06-11
- Q2 2026 earnings (Aug 6/13) — revenue $176.1M (+14% YoY), GAAP net income $16.8M (-42% YoY), adjusted net income $38.7M. Backlog grew to $4.5B.
- DOE HALEU Production Contract task order signed (Jul 1, 2026) — definitive contract executed under the Jan 2026 $900M award.
- New HALEU offtake agreements with Oklo and X-energy (Aug 6, 2026) — signed, with prepayment provisions.
- All financial contingencies removed on ~$3.0B of LEU/HALEU sales contracts (per CEO, Q2 call) — $2.4B of the LEU backlog now under definitive agreements.
- DOE FY2027 budget proposal excludes further funding for HALEU cascade operations — ~$0.8B of the Technical Solutions backlog (the entire TS segment) is exposed.
- Capex ramp: H1 2026 capex $94.8M vs. FY2025 full-year $19.7M; FY2026 guidance reaffirmed at $350–500M total capital deployment.
- Price +23.5% ($157.77 → $194.78) since baseline; four analyst price-target cuts (Stifel, Evercore, BofA, Roth) alongside two new initiations (Truist Buy $215, Barclays Equal-Weight $207).
- Urenco received NRC authorization (Oct 2025, reported this pass) for LEU+ (up to 10% enrichment) with 2026 first deliveries; Orano's Oak Ridge award now sized at ~$5B.
- Saudi—US civil nuclear pact sent to Congress (Jul 2026); LEU named as a possible beneficiary if Saudi Arabia pursues enrichment with a US partner — speculative, no contract exists.
- Uranium spot roughly flat: ~$86/lb (baseline) → ~$87.55/lb (now).
The delta ledger
Lead items first — the biggest financial-narrative move in this pass is the retraction of the baseline's own bear-case interest-expense forecast, and the supersession of the FCF trend by a sharp capex-driven negative swing.
| # | Claim (baseline) | Type | Status | Evidence |
|---|---|---|---|---|
| 10 | "Interest expense surge to $60–80M+ in FY2026" (baseline's central near-term bear case, driving the Phase 3 debate) | State | ❌ RETRACTED | TTM interest expense (sum of last 4 quarterly filings, Q3'25–Q2'26) = $15.7M ($3.4M+$4.1M+$4.0M+$4.2M) — essentially flat vs. FY2025's $14.0M. Company's own H1 2026 disclosure: $8.2M. The baseline annualized the FY2025 exit-rate debt balance at a market rate without accounting for capitalized interest during construction — Net PPE jumped $29.5M→$142.5M over the same two quarters, consistent with construction-phase interest being capitalized into the asset rather than expensed (standard treatment, ASC 835-20). Corroborated across three independent quarterly filings (Yahoo statement data) and the company's own press release; not a single-source finding. |
| 14/15 | "FCF modestly positive ($20–33M range FY22-25), improving, 16.3% 3yr CAGR" | Trend | 🔄 SUPERSEDED | TTM FCF is now sharply negative: OCF (TTM) −$55.0M (exactly matches Yahoo's own trailing field) minus capex (TTM) −$108.8M = ≈ −$163.8M. (Yahoo's own trailing freeCashflow field shows −$127.3M — a ~$36M discrepancy between the vendor's rolled-up figure and the quarter-by-quarter sum; I use the quarter-sum since it ties directly to the filed capex figures. Flagging the vendor inconsistency rather than picking silently.) Driver: capex went from $19.7M (FY2025 full year) to $108.8M (TTM) — this is exactly the capex acceleration the baseline itself predicted ("expect CapEx to accelerate sharply through 2026-2031"). The trend claim as stated (positive, improving, CAGR-computable) no longer holds — replaced by: FCF is deeply negative during the construction ramp, as the baseline's own "toll bridge" framing anticipated, but the magnitude is now a current fact, not a forecast. |
| 2 | "$3.9B backlog... contractual committed revenue... customers have locked in supply" | Structural/Trend | 📉 DRIFTED | Backlog grew to $4.5B (+15%) and its quality improved — all financial contingencies removed on ~$3.0B of LEU/HALEU contracts, $2.4B now under definitive agreements (up from an undisclosed mix at baseline). But a new, specific risk surfaced: the entire $0.8B Technical Solutions backlog rests on a DOE contract whose proposed FY2027 budget carries no further funding, and DOE "does not currently intend to exercise additional options" per company disclosure. That's 18% of total backlog now flagged, not zero. Net: backlog grew and firmed on the LEU side, but the "customers locked in, no execution ambiguity" framing was too clean — it breaks at the point the existing break-trigger #3 already names (backlog <$2.5B two consecutive quarters is the level to watch if TS unwinds and LEU growth stalls simultaneously). |
| 3 | "DOE... implicit floor on the business... every incentive to support" | Structural | 📉 DRIFTED | Mixed: DOE signed the definitive $900M HALEU Production Contract task order (Jul 1) — the big-ticket item is intact and funded. But the FY2027 budget gap on the Technical Solutions line is the first concrete instance of DOE support being non-uniform across programs. The "DOE always backstops" framing needs the caveat that it backstops the strategic enrichment build-out specifically, not every Centrus government contract line. Appropriations risk is real, datable (FY2027 budget cycle), and now a named trigger. |
| 12 | "Operating margin more than halved, 20.5%→11.2% (FY22-25)" | Trend | 📉 DRIFTED | TTM operating margin (sum last 4 quarters: −$16.6M+$12.8M+$0.8M+$10.4M = $7.4M / $473.9M revenue) = ~1.6% — compression continued faster than the FY2022-25 annual trend implied, driven by a genuinely negative Q3 2025 quarter (SWU cost/volume mix) partially offset by Q2 2026 recovery. Gross margin TTM ≈23.7% vs. FY2025's 26.2%, also still compressing. CFO attributed Q2 cost pressure specifically to "average unit cost of SWU sold rose 13%" while SWU volumes fell 23% — a genuine mix/cost headwind, not just construction noise. |
CARRIED / REFRESHED (compact list — all present, not dropped):
| # | Claim | Type | Status | Note |
|---|---|---|---|---|
| 1 | Sole US-licensed HALEU producer, structural monopoly | Structural | ✅ CARRIED | No competitor reached HALEU-grade commercial production. Urenco's Oct 2025 NRC authorization covers LEU+ (≤10%), still below the HALEU threshold (5–20%). Orano's Oak Ridge award now sized ~$5B (bigger commitment than baseline characterized) but still years from production — the exact updated timeline to Orano first output was not independently re-verified this pass; treat the "5-7yr lead" figure as carried, not refreshed. |
| 4/5 | Wide-moat grade (8.5/10), Evergreen rating (8/10) | Judgment | ✅ CARRIED | Strengthened if anything — Oklo/X-energy offtakes diversify the customer base beyond DOE dependency, a genuine moat-widening development not present at baseline. |
| 6 | Fluor (EPC) + Geiger Brothers (construction) execution team | Structural | ✅ CARRIED | First centrifuge completion still targeted 2026 per Q2 call; Piketon headcount guidance raised (100→175+ net new hires), a sign of ramp, not delay. Self-reported by management on the earnings call — not independently verified against the 10-Q (SEC EDGAR blocked WebFetch this session); single-sourced, flagged as such. |
| 7 | $1.3B FY2025 raise funds expansion; no near-term additional raise needed | State | 🔁 REFRESHED | H1 2026 ATM usage was modest ($53.9M, "opportunistic," CEO's own word) vs. FY2025's $523.7M equity raise. New offtake prepayments (Oklo, X-energy) function as non-dilutive funding, reducing reliance on further equity issuance. Dilution risk lower than baseline implied. |
| 8 | Net cash fortress (+$742M at FY2025) | State | 🔁 REFRESHED | Net cash now ~+$691M (Q2'26: cash $1.869B − debt $1.178B) — down modestly as capex ramps, still a fortress. Cash sufficient for FY2026 guidance even at the $500M high end. |
| 9 | FY2026 revenue guidance $450-500M | State | ✅ CARRIED | Reaffirmed unchanged; TTM revenue $473.9M is tracking inside the range. |
| 13 | Share dilution trend (+34.3% FY22-25) | Trend | 🔁 REFRESHED | Pace has slowed sharply: basic shares outstanding 19.66M (FY25)→19.95M (Q2'26), +1.5% vs. the FY2025 spike. |
| 16 | EV/Backlog 0.60x, below peer 0.8-1.5x range | Price | 🔁 REFRESHED | Now 0.68x (EV $3.05B / backlog $4.5B) — still below the peer range, though the discount narrowed as price rose faster than backlog. Re-derived fresh in the Valuation section below, not carried mechanically. |
| 18 | Graham's Number doesn't apply to this name | Judgment | ✅ CARRIED | Recomputed: √(22.5×$1.90×$42.37) = $42.56 vs. price $194.78 = 4.6x Graham (worse than baseline's 2.7x, reflecting both price appreciation and a lower TTM EPS). Same methodological conclusion: Graham prices stable industrials, not a pre-earnings-scale strategic monopoly funding a construction ramp. Applying it here would flag "massively overvalued" on a model that doesn't fit the company — noted and set aside, as at baseline. |
| 19 | Analyst consensus stable, overwhelmingly bullish | State | 🔁 REFRESHED | Now 2 Strong Buy / 10 Buy / 6 Hold / 0 Sell (18 covering, was 2/9/5/0=16). Still 0 Sell, still net-bullish, but Hold count grew and four desks cut price targets (Stifel $246→$216, Evercore $390→$337, BofA $240→$205, Roth $230→$195) even as two new firms initiated (Truist Buy $215, Barclays EW $207). Mean target $253, range $180–340 — direction held, conviction among existing holders cooled slightly. |
| 23 | Oklo JV — "early-stage optionality" | Structural | 🔁 REFRESHED (upgraded) | No longer exploratory — signed HALEU supply agreements with both Oklo and X-energy, including prepayment terms. A real upgrade from the baseline's characterization. |
| 24 | Uranium spot pullback to ~$86/lb | State | 🔁 REFRESHED | ~$87.55/lb now — essentially flat, not a new driver either direction. |
| 25 | Piketon execution risk "manageable" | Judgment | ✅ CARRIED | No negative execution news found (no cost-overrun disclosure, no safety incident, no auditor/CFO turnover — CFO Tinelli in Q2 2026 filing is the same CFO as baseline's Q1 2026, contra a churn risk). First centrifuge on track for 2026. |
🆕 NEW — no baseline counterpart:
A portfolio-specific passage was removed from the public build.
How the close calls were decided
Backlog growth vs. backlog risk (claim #2/#3). The evidence pulls two ways in the same disclosure: backlog grew 15% and firmed (contingencies removed on $3.0B), while a new, specific, dollar-sized risk appeared on the remaining 18%. These aren't in tension — they're two different segments of the same number moving in opposite directions. I did not net them into a single verdict; I marked the claim DRIFTED rather than either CARRIED or SUPERSEDED, because "contractual committed revenue, no ambiguity" was the baseline's specific wording and that wording is no longer accurate for the whole backlog, even though the LEU-segment majority is stronger than before. The existing break trigger ("backlog falls below $2.5B for two consecutive quarters") already covers the downside case correctly and needs no rewrite — it would fire if the TS risk materializes and LEU growth doesn't offset it.
Interest expense retraction vs. FCF supersession — are these contradictory? No. They're the same mechanism (capitalized construction costs) producing opposite-looking effects on two different statements: interest gets capitalized (helps the income statement, retracts the bear case) while the same capex shows up in full on the cash-flow statement (hurts free cash flow, supersedes the trend). A reader could mistake "interest expense didn't blow up" for "the cash story is fine" — it is not; the capex-driven FCF collapse is real and current, not deferred. Both forces were weighed together rather than letting the positive one (retraction) implicitly explain away the negative one (supersession).
DOE budget gap vs. DOE's $900M signed contract. One is a signed, definitive contract through Jan 2031 (Piketon HALEU Production Contract); the other is a proposed FY2027 budget line for a different program (Technical Solutions/HALEU cascade operations) that has not yet been finalized by Congress. I weighted the signed contract as the stronger, primary-source force (SEC-disclosed, definitive) and the budget gap as a real but not-yet-resolved risk — hence DRIFTED, not SUPERSEDED or RETRACTED on the DOE-anchor claim as a whole.
Thesis persistence and conviction delta
Structural + Trend claims: 14 total (claims 1, 2, 3, 4/5, 6, 7, 8, 11, 12, 13, 14/15, 19, 23, 24, 25 — grouping the paired rows). Of these: 8 CARRIED/REFRESHED cleanly, 2 DRIFTED (backlog composition, op-margin), 1 DRIFTED (DOE floor), 1 SUPERSEDED (FCF). Thesis persistence ≈ 79% (11 of 14 CARRIED or REFRESHED; the 3 DRIFTED/SUPERSEDED rows are real but each has a named, datable resolution point rather than an open-ended break).
Conviction: 7.5 → 7.8. Driven by named rows, not a general vibe: - + pressure: claim #10 RETRACTED (the baseline's own central near-term bear case — interest expense eating operating income — did not happen and won't on the current debt/capitalization structure); claim #23 REFRESHED-upgraded (Oklo/X-energy offtakes are signed, not exploratory, and diversify away from pure DOE dependency); claim #2 backlog grew 15% and firmed via contingency removal. - − pressure: claim #14/15 SUPERSEDED (FCF is sharply negative right now, a fact the market has to keep absorbing through 2026-27); claim #2/#3 new TS budget risk (18% of backlog, real appropriations exposure); claim #12 op-margin compression ran hotter than the FY22-25 trend implied; price already ran +23.5% since baseline, so less of the improvement is still available to be "discovered" — the EV/Backlog discount narrowed from 0.60x to 0.68x even before any of this quarter's good news is fully priced.
Net: the retraction of the single biggest bear-case number in the baseline's own debate section is a genuine, well-corroborated positive that most of the market has not obviously repriced (four analyst desks cut targets over the same window, suggesting the Street is if anything more cautious, not less). That asymmetry — thesis strengthening on the specific point the baseline was most worried about, while sentiment cooled — is worth a modest conviction increase, not a large one, because the price has already captured a meaningful share of the good news and a new, real risk (TS funding) has appeared to replace the old one.
Updated Valuation
Standard models, re-derived from scratch (Price rows never carry forward):
- Graham's Number: $42.56 (EPS $1.90 TTM, BVPS $42.37) vs. price $194.78 → 4.6x Graham. Does not apply — see Judgment row #18 above. Not weighted.
- DDM/DYT/Bogle: N/A — no dividend, and per the CLAUDE.md guardrail, the vendor forward P/E is not reliable here.
forwardEps($3.82) andepsCurrentYear($3.49) mix normalized non-GAAP add-backs (growth costs, stock comp) with a TTM GAAP EPS ($1.90) that itself carries a recurring, unexplained "gain on sale of security" line (see New Finding above). A forward-P/E-based trim would be built on a numerator this report cannot vouch for. Set aside, as at baseline — the earnings are genuinely too lumpy/reclassified for a P/E anchor to mean anything yet.
EV/Backlog (primary method, as at baseline, re-run):
| Metric | Value |
|---|---|
| Market Cap | $3.89B |
| Enterprise Value | $3.05B |
| Total Backlog | $4.5B ($3.7B LEU + $0.8B Technical Solutions, $2.4B of LEU under definitive agreements) |
| EV/Backlog | 0.68x (was 0.60x at baseline) |
Still below the 0.8–1.5x range cited for comparable long-cycle infrastructure/defense contractors. At 1.0x backlog: EV $4.5B + net cash $0.69B = $5.19B market cap → ~$260/share (≈20M shares). At 0.8x: ~$215/share. At 1.3x (bull, reflecting the Oklo/X-energy diversification and contingency removal): ~$328/share — close to the analyst high target ($340) and the level this report anchors the trim to.
Fair value range:
| Scenario | Driver | Fair Value |
|---|---|---|
| Bear | TS budget line stays unfunded, capex overruns, backlog effectively haircut to ~$4.0B, EV/Backlog stays at 0.5-0.6x | $140–165/share |
| Base | LEU-side execution continues on schedule, TS risk resolves neutrally or partially, EV/Backlog normalizes to 0.8-1.0x | $215–260/share |
| Bull | Saudi optionality or further SMR offtakes lift backlog past $5B, multiple re-rates to 1.2-1.5x | $300–390/share |
Fair value range for the frontmatter: $180–280 (wide by design, weighted toward base case, acknowledging the TS-budget uncertainty pulls the low end down from a pure base-case midpoint and the analyst mean target of $253 sits comfortably inside it).
Entry: $170–195. Current price sits at the top of this band — this is a "hold, add on weakness" zone, not a fresh-money green light at today's exact print. Strong buy re-confirmed near structural support: <$155 (near the 52wk low $142.13).
Trim: 1.3x EV/Backlog (~$325–330/share at current backlog + cash), explicitly not a forward-P/E multiple. Basis stated per the CLAUDE.md rule for pre-earnings-scale names: forward EPS here mixes non-GAAP add-backs with an unexplained recurring gain line, so a P/E trim would be anchored to a number this report doesn't trust. EV/Backlog is the metric the entire thesis is actually built on (backlog exceeds market cap, the moat is regulatory/contractual, not earnings-multiple-driven) — the trim moves automatically as backlog and net cash update at each future pass, the same self-updating property CLAUDE.md wants from a forward-P/E trim on a normal compounder.
Updated verdict
Rating: [7.8/10] — ACCUMULATE on weakness, HOLD at current price.
The core thesis — sole US-licensed HALEU producer, DOE-anchored, backlog exceeding market cap — is intact and, on the single point the baseline worried about most (interest expense crushing operating income), demonstrably wrong to worry about at the level baseline modeled. Two new commercial offtakes (Oklo, X-energy) diversify the customer base in a way the baseline only had as speculative optionality. Against that: free cash flow has gone sharply negative as the Piketon/Oak Ridge capex ramp hits in earnest — exactly the "toll bridge" period the baseline named, now arrived — and a real, sized, datable new risk appeared on 18% of the backlog via the FY2027 federal budget cycle. The stock is up 23.5% since baseline and the EV/Backlog discount to peers has narrowed from 0.60x to 0.68x, so less of the improving picture is still available to be bought cheaply.
Position guidance: unchanged in kind from baseline — this remains a satellite/speculative-sleeve position, not a core holding, given the pre-earnings-scale cash burn and the appropriations-cycle exposure now on the table. Do not chase at $195; add in the $170–195 band on weakness, add more aggressively under $155.
Thesis break triggers (carried, one reworded to reflect the new evidence): 1. DOE cancels or materially reduces the HALEU Production Contract (Piketon) — unchanged. 2. Piketon expansion cost overrun exceeds 30% without a commensurate DOE backstop — unchanged, untested this pass (self-reported on-schedule by management only). 3. Backlog falls below $2.5B for two consecutive quarters — unchanged; now the correct instrument to watch if the TS budget gap fails to resolve and LEU growth doesn't offset it. 4. Another equity raise at <$150/share — unchanged; less likely near-term given the modest H1 2026 ATM usage and new non-dilutive prepayment funding, but not retired as a trigger. 5. NEW: FY2027 federal budget finalizes with the Technical Solutions/HALEU cascade line unfunded and no replacement contract is signed by year-end 2026 — names the new risk with a hard date (the FY2027 appropriations cycle).
Upgrade conditions: TS budget line restored or replaced by a new definitive contract; TTM FCF turns a corner (even a narrowing quarterly burn would matter) as the Piketon buildout starts contributing revenue; a third major SMR/utility offtake signed on Oklo/X-energy terms.
What this pass did NOT test
- Orano/Urenco exact production timelines — carried the baseline's "5-7yr lead" framing without independently re-verifying the current dates; only spot-checked that neither has reached HALEU-grade commercial production and that Orano's award is now sized ~$5B. A dedicated competitive-timeline refresh is due next pass.
- The "Gain on Sale of Security" recurring line — flagged as a new open item (three consecutive quarters, consistent magnitude, no explanation surfaced this pass). Needs a direct 10-Q footnote read; SEC EDGAR blocked WebFetch this session (403).
- Piketon execution/cost detail beyond management's own earnings-call statements — "on schedule," "175+ net new hires," "first centrifuge in 2026" are all self-reported by the company on the Q2 call; not cross-checked against the 10-Q construction-in-progress footnote or any independent trade-press site visit. Single-sourced; carried as CARRIED but flagged, not treated as fully corroborated.
- Capital-allocation detail for the Oklo/X-energy prepayments — the dollar size of the prepayments and their exact accounting treatment (deferred revenue timing, refundability) was not obtained this pass.
- Uranium/SWU forward curve — only spot price was refreshed; SWU contract pricing (relevant to the margin story) was last available as a full-year-2025 average ($108.70/SWU, +11% YoY) with no August 2026 update found.
Sources: Yahoo Finance MCP (quarterly income/balance/cashflow statements, stock info, insider transactions, analyst recommendations/upgrades), python .mcp/fin.py LEU --news, Centrus Q2 2026 earnings release and call transcript (Motley Fool, Yahoo Finance, Globe and Mail summaries), Investing.com/StockTitan Q2 2026 summaries, ANS Nuclear Newswire (DOE contract, Urenco/Orano competitive updates), World Nuclear Association country profile, EnkiAI nuclear-sector trackers, PowerMag (Centrus HALEU delivery milestone), UxC/carboncredits.com/ANS (uranium spot price), Yahoo Finance/Stocktwits (Saudi nuclear pact coverage). SEC EDGAR 10-Q direct fetch was blocked (403) this session — flagged as an untested gap above.