Version: Final (Round 3) | Date: 2026-04-06 | Validation: Self-verified via StockAnalysis.com (Apr 6) + GPT-4 cross-validation (Apr 2) + macro refresh
Thesis: Japan's nuclear restart is entering the execution phase — 15 reactors operating, government targeting 20% nuclear by FY2040, $40B SMR deal signed March 19. This portfolio targets the hidden supply chain that most analysts mis-categorize — stocks covered as "steel," "instrumentation," or "water treatment" companies whose nuclear exposure is buried in segment notes.
ALL figures from StockAnalysis.com (Apr 6, 2026). NOT yfinance. FCF verified via annual cash flow statements.
| Stock | Ticker | Wt | PE | Fwd PE | P/B | ROE | OpM | D/E | DY | FCF | 52W% |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Japan Steel Works | 5631.T | 17% | 34.1 | 33.2 | 3.42 | 10.5% | 8.4% | 0.34 | 1.0% | -¥19.7B | +81% |
| Yokogawa Electric | 6841.T | 18% | 21.7 | 18.1 | 2.44 | 12.2% | 14.6% | 0.05 | 1.3% | +¥88B | +69% |
| Fuji Electric | 6504.T | 15% | 18.6 | 16.2 | 2.00 | 11.8% | 10.3% | 0.13 | 1.7% | +¥51B | +75% |
| Organo Corp | 6368.T | 10% | 23.5 | 21.8 | 5.07 | 23.5% | 21.4% | 0.24 | 1.3% | +¥19B | +125% |
| Kanadevia | 7004.T | 10% | 17.4 | 12.5 | 0.97 | 5.8% | 2.0% | 0.88 | 2.3% | -¥12.8B | +19% |
| Sumitomo Heavy | 6302.T | 10% | 19.5 | 18.1 | 0.88 | 4.7% | 4.8% | 0.37 | 2.9% | +¥11B | +64% |
| Kurita Water | 6370.T | 8% | 36.2 | 20.5 | 2.38 | 6.8% | 13.7% | 0.32 | 1.5% | +¥25B | +67% |
| Toyo Tanso | 5310.T | 7% | 20.8 | 21.8 | 1.17 | 5.7% | 14.6% | 0.06 | 2.7% | -¥5.8B | +36% |
| Nippon Carbon | 5302.T | 5% | 10.6 | 19.6 | 0.81 | 8.7% | 12.8% | 0.16 | 4.3% | +¥0.8B | +11% |
WHY this stock: Signed strategic agreement with Rolls-Royce SMR (Feb 2, 2026) to deliver data processing and control systems (DPCS) for the entire Rolls-Royce SMR fleet. This is a specific, named, disclosed contract — not speculation. 65+ years of nuclear monitoring pedigree. Also supplied the world's largest radiation monitoring panel (300 channels) to JNFL Rokkasho. Coverage focuses on oil/gas instrumentation — the SMR contract is underreported.
WHY 18%: Best quality-to-valuation ratio in the portfolio. ROE 12.2%, op margin 14.6%, D/E only 0.05 (essentially zero debt). FCF strongly positive at ¥88B — the highest in the portfolio. PE 21.7x / fwd PE 18.1x is reasonable for this quality. Both Claude and GPT rated this "agree; keep." Largest FCF-positive name → deserves top weight.
What could go wrong: Rolls-Royce SMR program delays or cancellation; oil/gas capex downturn affecting core business.
Monitoring trigger: If Rolls-Royce SMR project milestones slip, reassess nuclear optionality premium. Watch May earnings for SMR revenue recognition timing.
WHY this stock: 80% global monopoly in nuclear reactor pressure vessel forgings. The only company capable of making the largest single-piece nuclear forgings. Backlog exceeds ¥135B. Secured forging agreements for 4 SMR designs. METI officially named JSW as a BWRX-300 supplier (Mar 2026). ¥28B CAPEX committed for SMR production upgrades. 18-month manufacturing cycle provides multi-year revenue visibility.
WHY 17% (reduced from 25% original, 20% cross-validated): The monopoly is irreplaceable — no one else can make these forgings. But our fresh validation confirmed negative operating CF (-¥4.6B) and FCF (-¥19.7B) in FY2025, driven by a ¥32.3B inventory build. This is the ¥28B SMR capex program in action — it's growth investment, not operating weakness. However, the stock is up +81% 52W and trades at PE 34x with negative cash flow. We maintain a meaningful position because the monopoly and order backlog are real, but reduce weight to reflect the cash burn phase.
What could go wrong: SMR order pipeline doesn't convert to revenue; cash burn continues longer than expected; competitor emerges for smaller forgings.
Monitoring trigger: If operating CF remains negative in FY2026 results (May 2026), reduce to 12%. If inventory converts to revenue (OCF turns positive), increase back to 20%. This is the single most important metric to watch at earnings.
WHY this stock: Top-5 global in nuclear radiation monitoring (65+ years, alongside Thermo Fisher, Mirion) AND top-5 global in IGBT/SiC power semiconductors for EVs. The only stock where both Claude and GPT independently said "agree; keep" in both nuclear AND EV portfolios. PE 18.6x, fwd PE 16.2x — best valuation in the portfolio for this quality. Raised FY operating profit guidance to ¥128.5B.
WHY 15%: Dual-theme exposure at reasonable valuation. ROE 11.8%, D/E 0.13, FCF +¥51B — clean fundamental profile. Appears in both portfolios but capped at 15% per concentration rules.
What could go wrong: Industrial capex downturn; SiC competitors erode IGBT share faster than expected.
Monitoring trigger: Watch May earnings for nuclear instrumentation order flow and power semi backlog. If either segment weakens, reassess dual-theme premium.
WHY this stock: Nuclear water purification specialist — provides condensate filters, demineralizers, reactor water cleanup, fuel pool cleanup for nuclear power plants. Every reactor restart requires Organo's services. Best quality metrics in the entire portfolio: ROE 23.5%, operating margin 21.4%. FCF turned positive in FY2025 (+¥19B) after negative FY2024.
WHY 10% (increased from 8% in cross-validated version): Increased because this is the portfolio's best quality name and provides "operations" exposure — revenue from ongoing reactor maintenance, not just construction/build-out. Balances the "SMR build bet" (JSW, Yokogawa) with "reactor operations" revenue. 52W +125% means significant re-rating has occurred, but quality justifies premium (P/B 5.07).
What could go wrong: Reactor restart pace slows; semiconductor ultrapure water business (non-nuclear) faces competition.
Monitoring trigger: If P/B expands >6x without earnings support, trim to 7%. Watch for new reactor restart announcements — each one is an Organo revenue event.
WHY this stock: Fukushima decommissioning cask manufacturer with specific TEPCO JV ("Toso Mirai Manufacturing"). First domestic maker of used fuel transport casks since 1978. The Fukushima decommissioning program is a ¥22 trillion, 30-40 year revenue stream — fuel debris extraction is just beginning. Name changed from Hitachi Zosen Oct 2024 — databases still show old name, creating information asymmetry.
WHY 10% (reduced from 15% original, 12% cross-validated): Corrected PE is 17.4x (not our original 10.2x). FCF is negative (-¥12.8B). D/E 0.88 is the highest in the portfolio. Operating margin only 2.0%. This is NOT a quality compounder — it's a specific-catalyst value play on the decommissioning contract. Reduced to 10% to reflect weaker fundamentals than originally framed.
What could go wrong: Fukushima decommissioning timeline extends (it always does); TEPCO JV production delays; balance sheet stress from D/E 0.88 + negative FCF.
Monitoring trigger: If D/E exceeds 1.0 or FCF remains negative for 2+ more periods, cut to 5%. If TEPCO JV begins shipping casks (revenue recognition), consider increasing.
WHY this stock: Sole Japanese manufacturer of PET cyclotrons (100+ deliveries). World's first approved BNCT (Boron Neutron Capture Therapy) cancer treatment system (NeuCure). Japan nuclear medicine market growing $550M → $2.54B by 2031 (21.2% CAGR). P/B 0.88 (below book) with 2.9% dividend yield.
WHY 10%: Unique exposure to nuclear technology outside of power generation. The BNCT system is genuinely first-in-world. P/B below book provides downside cushion. FCF positive (¥11B) though modest. ROE 4.7% and op margin 4.8% are low — this is not a quality name but a thematic bet on nuclear medicine growth.
What could go wrong: BNCT adoption slower than expected; cyclotron demand matures; nuclear medicine not a large enough segment to move the stock.
Monitoring trigger: Track BNCT installation announcements and nuclear medicine market size updates. If no BNCT progress in 12 months, reduce to 5%.
WHY this stock: Nuclear cooling water chemistry specialist + semiconductor ultrapure water leader. Dual nuclear + semiconductor theme. FCF strongly positive (¥25B). Op margin 13.7%. D/E 0.32 is moderate. As reactors restart and semiconductor fabs are built (TSMC, Rapidus, Micron), Kurita's services are needed at each facility.
WHY 8%: Added per GPT recommendation to provide "operations/maintenance" ballast. Trailing PE 36.2x is high, but fwd PE 20.5x (strong earnings growth expected) brings it to reasonable territory. Semiconductor ultrapure water business provides diversification beyond nuclear.
What could go wrong: Semiconductor fab buildout slows (TSMC Fab 2 delayed); nuclear restart pace disappoints; premium valuation compresses.
Monitoring trigger: If fwd PE expands >25x without earnings revision, trim to 5%. Watch semiconductor fab capex announcements.
WHY this stock: Produces isotropic graphite for HTGR core components, fusion reactor wall materials, AND semiconductor graphite. Recently received an order for high-temperature micro-reactor graphite (TSE disclosure). Dual nuclear + semiconductor graphite exposure in a ¥114B micro-cap. D/E only 0.06 — virtually zero debt.
WHY 7% (reduced from 10%): Corrected fwd PE is 21.8x (not 11.7x) — no longer "cheap." FCF is negative (-¥5.8B) from capacity expansion capex. This is a small-cap optionality play on next-gen nuclear (HTGR, fusion, micro-reactors), not a value play. 7% reflects optionality sizing.
What could go wrong: HTGR/fusion timelines extend indefinitely; semiconductor graphite demand weakens; illiquidity (¥114B market cap) makes exits difficult.
Monitoring trigger: If additional HTGR or micro-reactor graphite orders are disclosed, consider increasing to 10%. If no new nuclear orders in 12 months, reduce to 3%.
WHY this stock: Nuclear graphite supplier since 1957 + lithium-ion battery anode materials. Trailing PE 10.6x — cheapest in portfolio. P/B 0.81 (below book). Dividend yield 4.3% — highest in portfolio. FCF marginally positive (+¥0.8B). D/E 0.16. Dual nuclear + EV materials exposure.
WHY 5%: Deep value + yield sleeve. Forward PE 19.6x is much higher than trailing (10.6x) — market expects earnings to decline. This is the key risk. ¥51B market cap = very illiquid. Small position captures the yield and optionality without taking concentration risk.
What could go wrong: Earnings decline materializes (fwd PE 19.6x vs trailing 10.6x implies analysts expect ~46% earnings drop); illiquidity makes exits costly.
Monitoring trigger: If trailing PE expands >15x (earnings actually declining), reassess. Hold for yield (4.3%) as long as dividend is maintained.
| Metric | Value |
|---|---|
| Number of holdings | 9 |
| Top-3 concentration | 50% (Yokogawa 18% + JSW 17% + Fuji Electric 15%) |
| HHI (Herfindahl) | ~0.123 (effective ~8.1 names) |
| FCF-positive names | 6 of 9 (67%). FCF-negative = JSW, Kanadevia, Toyo Tanso |
| FCF-negative weight | 34% (reduced from original 50%) |
| Weighted avg fwd PE | ~20x |
| Avg dividend yield | ~2.1% |
| Layer | Stocks | Weight | Role |
|---|---|---|---|
| SMR / Construction | JSW (forgings), Yokogawa (controls) | 35% | Build-out phase beneficiaries |
| Nuclear Operations | Organo (water), Kurita (water), Fuji Electric (monitoring) | 33% | Ongoing reactor revenue — every restart = revenue |
| Decommissioning | Kanadevia (casks) | 10% | 30-40 year Fukushima revenue stream |
| Nuclear Medicine | Sumitomo Heavy (cyclotrons, BNCT) | 10% | Non-power nuclear technology |
| Next-Gen Materials | Toyo Tanso (graphite), Nippon Carbon (graphite) | 12% | HTGR, fusion, micro-reactor optionality |
| Date | Company | What to Watch |
|---|---|---|
| May 2026 | Japan Steel Works (5631.T) | THE key event. Did operating CF turn positive? Inventory conversion? SMR order updates? |
| May 2026 | Yokogawa (6841.T) | Rolls-Royce SMR revenue recognition. Nuclear order pipeline |
| May 2026 | Fuji Electric (6504.T) | FY results after raised OP guidance (¥128.5B). Power semi + nuclear split |
| May 2026 | Kanadevia (7004.T) | TEPCO JV production update. FCF trajectory |
| May 2026 | Organo (6368.T) | FY results. FCF sustainability after FY2025 turnaround |
| Timeframe | Catalyst | Beneficiaries |
|---|---|---|
| May 2026 | Earnings season — FY results for most holdings | All |
| H2 2026 | BWRX-300 supplier announcements / procurement | JSW, Yokogawa, IHI |
| 2027 | Tomari-3 reactor restart (Hokkaido Electric) | Organo, Kurita, Fuji Electric |
| 2027 | Kanadevia TEPCO JV cask production ramp | Kanadevia |
| 2028 | TSMC Kumamoto Fab 2 (3nm) production | Kurita (ultrapure water), Toyo Tanso (graphite) |
| 2029-2030 | Kashiwazaki-Kariwa Unit 7 restart | All nuclear names |
| 2030s | Japan fusion demonstration target (revised Jun 2025) | Toyo Tanso (wall materials) |
| Action | Trigger |
|---|---|
| ADD JSW to 20% | If May earnings show operating CF turning positive (inventory converting to revenue) |
| ADD Toyo Tanso to 10% | If new HTGR or micro-reactor graphite orders disclosed |
| TRIM JSW to 12% | If operating CF remains negative in FY2026 |
| TRIM Kanadevia to 5% | If D/E exceeds 1.0 or FCF negative for 2+ more periods |
| TRIM Organo to 7% | If P/B expands >6x without earnings support |
| REASSESS all | If any nuclear accident globally — theme reprices overnight |
| Stock | yfinance Value | Verified (StockAnalysis) | Gap | Impact |
|---|---|---|---|---|
| JSW trailing PE | 30.4x | 34.1x | 12% | Minor — directionally correct |
| JSW FCF | Not checked | -¥19.7B | Critical | Major risk flag — missed in original |
| Kanadevia PE | 10.2x | 17.4x | 70% | "Deep value" thesis invalidated |
| Kanadevia FCF | Not checked | -¥12.8B | Critical | Cash flow risk added to thesis |
| Toyo Tanso fwd PE | 11.7x | 21.8x | 86% | "Cheap dual-theme" invalidated |
| Toyo Tanso FCF | Not checked | -¥5.8B | Critical | Capex cycle risk |
| Nippon Carbon fwd PE | 10.3x | 19.6x | 90% | Earnings decline expected |
| Organo (new) | N/A | PE 23.5x, ROE 23.5% | — | Quality confirmed |
| Kurita (new) | N/A | Fwd PE 20.5x, FCF +¥25B | — | Solid addition |
Lesson reinforced: yfinance forward PE is systematically unreliable for Japanese mid/small-caps. Every "cheap forward PE" thesis from our original report was wrong when verified against StockAnalysis.com. The FCF checks (absent from our original analysis) proved to be the most decision-relevant validation step — 3 of our original "value" picks had negative FCF.
Why these stocks are mis-categorized by the market — the core thesis of this portfolio.
| Stock | Analyst Coverage Framing | Hidden Nuclear Angle |
|---|---|---|
| Japan Steel Works | "Steel/machinery company" | 80% global monopoly in nuclear reactor forgings + 4 SMR agreements |
| Yokogawa Electric | "Oil/gas instrumentation" | Rolls-Royce SMR control systems partner (Feb 2026 contract) |
| Organo Corp | "Water treatment company" | Nuclear reactor water purification — every restart = Organo revenue |
| Kurita Water | "Semiconductor water" | Nuclear cooling water chemistry + semiconductor ultrapure water dual play |
| Kanadevia | "Waste-to-energy" (databases still show "Hitachi Zosen") | Fukushima decommissioning cask manufacturer with TEPCO JV |
| Toyo Tanso | "Semiconductor graphite supplier" | HTGR core materials + fusion wall materials + micro-reactor order |
| Fuji Electric | "Diversified industrial" | Global top-5 radiation monitoring + EV power semiconductors |
| Sumitomo Heavy | "Industrial machinery" | Sole Japanese PET cyclotron maker + world-first BNCT system |
| Data | Source | Date |
|---|---|---|
| All valuation + FCF metrics | StockAnalysis.com/quote/tyo/XXXX/statistics/ and /financials/cash-flow-statement/ | Apr 6, 2026 |
| GPT cross-validation | GPT-4 independent review | Apr 2, 2026 |
| $40B BWRX-300 deal | White House Fact Sheet, ANS Nuclear Newswire | Mar 19, 2026 |
| METI supplier confirmation (JSW, IHI) | ANS | Mar 25, 2026 |
| Yokogawa × Rolls-Royce SMR | Yokogawa press release | Feb 2, 2026 |
| NRA anti-terrorism easing | Japan Times | Apr 2, 2026 |
| JSW nuclear forgings | JSW IR FY2025, World Nuclear News | 2025/2026 |
| Kanadevia TEPCO JV | World Nuclear News | Apr 2024 |
| Toyo Tanso micro-reactor order | TSE filing via MarketScreener | 2025 |
| SHI BNCT / nuclear medicine | BioSpace; industry market forecast | 2023-2031 |
| Japan fusion strategy | Cabinet Office via Science Japan | Jun 2025 |
| Reactor operations status | NRA Japan, World Nuclear Association | Jan-Apr 2026 |
| DC power demand | Wood Mackenzie | Aug 2025 |
FINAL VERSION | Round 3: Self-validated (StockAnalysis.com) + GPT-4 cross-validated + macro-refreshed
Generated by JPstock-agent | 2026-04-06 | AI-assisted research, not investment advice