Nuclear, Uranium & Photonics Daily
SEQH Capital Research · Trading & Research Desk
Edition 005 · Thursday, October 1, 2026 · Post-close
Prices as of 4:00 PM ET close unless stamped otherwise.
The Tape in 60 Seconds
$120 billion for American reactors, and the nuclear basket still fell 1.2%.
The biggest financing headline of this nuclear cycle landed Wednesday evening. The U.S. and South Korea announced a framework for up to $120B of Korean investment to finance as many as eight large reactors in the U.S. using Westinghouse technology. That speaks directly to the bottleneck we flagged yesterday, which is financing, not permits.
The equities didn’t follow. CCJ, which owns 49% of Westinghouse, fell 1.13% on 4.09M shares, the heaviest volume of the week. Our five-name nuclear basket lost 1.20%, while the S&P 500 rose 0.19% to 7,666.45. Week-to-date, the basket is down 3.38% vs. −0.99% for the S&P. Optics went the other way: Coherent rose ~10% on its PhotonLink platform and a Bernstein Outperform initiation, and the whole group bid up behind it.
Three takeaways:
Big framework, cautious stock reaction. The deal includes a potential 5–10% Korean equity stake in Westinghouse, and Cameco’s own filing flags the possible effect on its ownership interest. A framework is not a construction contract.
The developers keep bleeding. OKLO (−2.38%) and SMR (−1.39%) both closed within cents of their session lows. Large-reactor financing doesn’t flow to SMR developers.
SPUT’s discount has narrowed for three straight sessions, from −13.63% to −12.99%, a 64 bp improvement and the first sustained narrowing in our tracking.
Key numbers
U.S.–Korea framework: up to $120B · up to 8 reactors (6 AP1000 + 2 APR1400)
Contingency reserve built into that framework: $20B (17% of the total)
Nuclear basket (5 names): −1.20% vs. S&P +0.19% · week-to-date −3.38% vs. −0.99%
SPUT discount: −12.99% (9/30) · narrowed 3 sessions in a row
U3O8 spot (SPUT-implied): $89.59/lb · flat
COHR: ~+10% intraday on PhotonLink + Bernstein Outperform
01 · Macro Tape
S&P 500: 7,666.45 · ▲ 0.19%
Dow Jones: 50,926.56 · ▲ 20.51 (+0.04%)
Nasdaq Composite: 26,871.60 · ▲ 0.04%
Russell 2000: outperformed after a rough September and Q3
Treasuries: the 10-year and 30-year both touched their highest levels since 2002 intraday, then pulled back into the close
ISM manufacturing (September): 54.5 vs. 54.9 consensus, the ninth straight month of expansion. The prices index jumped 6.8 points.
Read: stocks dropped in the morning as yields hit 20-plus-year highs, then recovered as yields eased and chip stocks led. Manufacturing is expanding with input prices rising fast, and that keeps the Fed on hold-or-hike. Friday’s jobs report is the last major data point before October hike pricing settles. For our long-duration names, the morning’s yield spike mattered more than the afternoon recovery.
02 · Reactors, SMRs & Power
CCJ / BEP / Westinghouse · U.S.–Korea framework for up to $120B of reactor financing
The deal: Korea would invest up to $120B, drawn from its $200B U.S. investment fund, to finance up to eight large U.S. reactors. That covers six Westinghouse AP1000s (three two-unit plants at federal sites) and two Korean APR1400s, which are built on Westinghouse technology.
What Westinghouse gets: an upfront payment, guaranteed work scope and a fuel-fabrication contract, plus a potential 5–10% cornerstone equity investment from Korea.
Risk budget: $20B of the $120B is a contingency reserve for cost overruns, per Korea’s trade ministry.
Legal and policy fit: it requires a waiver under Westinghouse’s 2025 intellectual-property settlement with KEPCO and KHNP. It builds on the October 2025 U.S. Commerce Department partnership to deploy at least $80B of Westinghouse reactors, and is separate from the DOE’s American Nuclear Supply Chain Loans commitment.
Desk math:
$/unit: $12.5B per reactor before contingency, $15.0B including it.
$/watt: roughly 9.4 GW across the eight units, or ~$12.7/W all-in. The capacity is our estimate from standard unit ratings.
Scale: $120B is 1.5x the original ≥$80B commitment.
Cameco’s stake: if Korea’s 5–10% came as newly issued shares, Cameco’s 49% would fall to roughly 44.1–46.6%. That’s illustrative only; terms aren’t public.
The stake’s value: at the reported $50B IPO valuation, Cameco’s 49% (~$24.5B) still equals ~65% of CCJ’s $37.85B market cap.
Why CCJ fell anyway: $85.69 · ▼ 1.13%. It traded as low as $83.80 (−3.3%) on 4.09M shares, about 1.3x its 3.21M average, before recovering. After hours it bounced to $86.46 (+0.89%). Consensus target slipped to $126.85. Our read: the market is pricing three things. Possible dilution of Cameco’s stake, a framework that isn’t yet a binding contract, and first power that is years away. In our view, this is the first concrete financing package on the scale large nuclear actually needs, and the after-hours bounce suggests the market agrees.
BWXT: $136.79 · ▼ 0.15%
Range 134.51–137.87 on 1.33M shares. Market cap $12.53B.
Won a ~$189M contract to manufacture naval reactor fuel through its Nuclear Fuel Services subsidiary.
Confirmed Q3 results for Monday, Nov. 2, after the close.
Consensus target drifted to $217.69 from $222.07 after this week’s Investor Day target cuts.
Developers
OKLO: $36.14 · ▼ 2.38%. Range 36.03–37.53, closing 11¢ off the low, on 8.45M shares. 5.1% above its 52-week low of $34.38. Market cap $6.72B. Down 5.0% week-to-date.
SMR: $7.79 · ▼ 1.39%. Range 7.78–8.08, closing 1¢ off the low, on 30.4M shares. That makes three of the last four sessions closing at or near the low. Down 7.5% week-to-date.
Read: this week has delivered a lot of policy and financing support for nuclear: the Clinch River permit, BWXT’s 2030 targets and now $120B from Korea. None of it reached the SMR developers. The money is flowing to licensed large-reactor designs (AP1000) and established suppliers (BWXT). Until OKLO and SMR sign binding customer contracts, every sector-wide positive headline seems to widen the gap between the developers and the rest of the sector.
03 · Uranium & Fuel Cycle
Physical market: Sprott Physical Uranium Trust (U.U / SRUUF)
NAV per unit (9/30): $21.52 (+0.01%)
Market price (9/30 close): $18.72, a −12.99% discount
Discount over the last four readings: −13.63% (9/28) → −13.21% (9/29) → −12.99% (9/30), a 64 bp narrowing in two sessions
U3O8 held: 81,697,348 lb, unchanged
Uranium value at market: $7.319B
Implied $/lb at market price: ≈ $78.86, excluding cash
NAV YTD: +8.69%
Q3 final, per Sprott’s updated table: 0 premium days and 63 discount days
The curve. SPUT-implied spot has stayed in a $0.25/lb band for five readings ($89.57–$89.82). The discount narrowing is coming from the trust’s market price rising, not from NAV falling, which is the right direction for a sentiment turn.
Equities (Thursday close)
UEC: $9.36 · ▼ 0.95%. Range 9.11–9.52 on 9.64M shares. RBC initiated at Sector Perform with a $10 target, only 6.8% above the price and 40.6% below the $16.83 consensus. Jefferies has a Hold at $11.50. The two newest initiations are both neutral, which suggests the post-earnings re-rating is being capped by valuation.
LEU: $139.11 · ▼ 0.27%. It printed a new 52-week low of $135.75 intraday, then closed 2.5% off the low on 543K shares. It’s 30.3% below its $199.64 September offering price. Consensus target $247.40 (+77.9%).
04 · Photonics & Optical Interconnect
COHR / LITE / CIEN / ANET · Bernstein initiates; Coherent’s PhotonLink platform draws buyers
Bernstein initiations: Outperform on Coherent, Lumentum, Ciena and Arista, calling networking and optical suppliers potential structural winners in AI infrastructure. Market Perform on Cisco and Corning.
Coherent PhotonLink: an integrated-optics platform for AI data centers, introduced in September, with more than 10 customer engagements each in co-packaged and near-packaged optics.
Prices (late-session feed unless noted)
COHR: ~$317.28 · ▲ ~10.2%
CIEN: ~$374.67 · ▲ ~6.5%
GLW: ~$159.82 · ▲ ~3.9%
LYTE (Roundhill Photonics & Optics ETF): ~$25.18 · ▲ ~3.4%
ANET: ▲ ~1.1% · CSCO: ▲ ~0.9%
LITE: +9% to $1,058.42 at 10:56 AM ET. Late-session feeds conflict, and the close is not verified at publication.
Read: Bernstein’s ratings split along the lines that matter. The Outperforms went to companies selling into the AI data-center buildout (optical components, switching and transport), while Cisco and Corning got Market Perform. Coherent gained more than twice the ETF’s ~4%, so traders treated PhotonLink as company-specific news, not just a sector move. This is the third straight day the group has traded on fundamentals: Citi’s switching forecast, Micron’s guidance and now Bernstein. Rotation flows have stopped driving these stocks for now.
05 · Catalyst Calendar
Fri 10/2: September jobs report, the last data point before October hike pricing settles (macro)
TBD: formal signing of the U.S.–Korea framework by both governments, Westinghouse, KEPCO and KHNP (CCJ, BEP)
Oct (TBD): Westinghouse public IPO filing possible, per Bloomberg. Watch for how the Korean equity stake is treated (CCJ, BEP)
Oct FOMC: rate decision
Fri 10/30: Cameco Q3 results, the first chance to hear management on the Korea framework and Westinghouse dilution (CCJ)
Mon 11/2, after close: BWX Technologies Q3 results (confirmed)
Early Nov (estimated): LEU (~11/4), SMR and LITE (~11/5), OKLO (~11/10)
06 · Desk Read: What We’re Watching
i. Financing has arrived for large reactors. Yesterday we said the constraint had moved from the NRC to financing. Within 24 hours, $120B showed up for the AP1000/APR1400 pathway, with an explicit $20B (17%) contingency budget. That is the industry admitting cost overruns are a base case and funding them in advance. That’s a structural change for large nuclear. The stock reaction hinges on Westinghouse’s IPO terms and how Korea’s equity stake is structured, which makes Cameco’s Oct. 30 call the key event for CCJ.
ii. Policy support isn’t reaching the SMR developers. This week brought a permit (Clinch River), a 2030 target set (BWXT) and $120B for large reactors (Korea), yet SMR fell 7.5% and OKLO 5.0% week-to-date. The market has stopped rewarding sector-wide headlines for developers without binding customer contracts. Until OKLO or SMR announces one, we treat them as a bet on interest rates, not on nuclear policy.
iii. SPUT’s discount is the turn signal to watch. Three straight sessions of narrowing (+64 bp) is the first sustained move toward NAV in our tracking, driven by the trust’s price rising, not NAV falling. Moving to about −10% would put SPUT within range of being able to raise money and buy uranium again. That would return the largest financial buyer to a spot market that has held $89.6 without it.
Corrections to Edition 004:
SMR’s Wednesday close was $7.90 (+1.80%); we published a 3:29 PM print of $7.92 (+2.06%).
SPUT’s Q3 final count is 0 premium days and 63 discount days; Sprott updated its table after our publication to include 9/30.
Method
SPUT-implied spot = market value of U3O8 held ÷ pounds held.
$/lb at market = unit price ÷ pounds per unit (0.2374 lb), excluding cash.
Baskets are equal-weighted simple averages of verified closes. The nuclear basket is CCJ, UEC, OKLO, SMR and BWXT. Week-to-date is measured from the 9/25 close.
$/W for the Korea framework uses ~1.12 GW per AP1000 and ~1.35 GW per APR1400; these are approximate standard ratings, not framework disclosures.
Where a close was not verified at publication, the line shows the last verified print with its stamp.
For informational and educational purposes only. Nothing here is investment advice or a recommendation to buy or sell any security. Figures come from third-party sources believed to be reliable but not guaranteed; verify before acting. Past performance does not indicate future results.
© 2026 SEQH Capital Research · Next edition: Friday 10/2, post-close (Jobs Day)