China Yangtze Power

Phase A — Understand the business

Lens 1 · Company Overview

China Yangtze Power (CYPC) is the largest listed hydropower operator on earth — a single-asset-class utility that owns and runs the six great cascade dams on the Yangtze and its Jinsha tributary. It does one thing, at planetary scale, and does it with almost no marginal cost.

• What it is: operator of six cascade hydropower stations — Three Gorges (22.5 GW), Baihetan (16 GW), Xiluodu (13.86 GW), Wudongde (10.2 GW), Xiangjiaba (6.4 GW), Gezhouba (2.715 GW) — totalling 71,695 MW (~71.7 GW) across 110 generating units. For scale, that single portfolio is ~30–35× the largest US hydro operator's fleet and roughly the entire installed generating base of a mid-size European country.
• How it makes money: it sells electricity to the grid. FY2024 generation from the six stations was 295.9 TWh (+7.1% YoY, +19.6 TWh). Revenue is generation × on-grid tariff; because the dams are already built and largely depreciating, the incremental economics are extraordinary — FY2024 operating margin ~54% (op income CNY 45.31B on revenue CNY 84.49B).
• Customers: the two grid monopolies — State Grid Corporation of China and China Southern Power Grid — plus large cross-provincial industrial offtake. Three Gorges power is dispatched across a wide footprint (Central China, East China incl. Shanghai/Jiangsu/Zhejiang, and Guangdong via long-distance UHV lines). Customer concentration is total but counterparty risk is effectively sovereign — the buyers are state grid monopolies.
• Suppliers: the river. There is no fuel bill. The main "input" is water inflow (hydrology), and the main cost lines are depreciation (huge, on the dams), financial expense (interest on the acquisition/construction debt), and O&M (small). This inverts the usual utility cost structure — no coal, no gas, no commodity beta.
• Contract structure / tariffs: a hybrid of regulated benchmark on-grid tariffs and a rising share of medium-to-long-term (M2L) market contracts. Historically hydro on the Yangtze enjoyed a favorable "cost-plus-reasonable-return" benchmark; the ongoing marketization (see Lens 3/13) is gradually pushing a slice of volume onto traded prices. Payment terms are effectively take-and-pay against sovereign counterparties — cash conversion is high and bad debt negligible.
• Ownership: controlled by China Three Gorges Corporation (CTG), a central SOE under SASAC, holding ~62–64%. Free float is split between domestic institutional/retail (~17%) and foreign institutions (~19%). This is a state-controlled national-champion utility, not a free-float compounder — read every capital-allocation and governance point through that lens.

Plain-terms model: a bond with a turbine attached. CYPC converts falling water into ~CNY 34B of annual net profit and pays most of it out as dividends, with the "coupon" varying year to year with rainfall.

Lens 2 · Supply Chain

Hydro has almost no upstream fuel chain — which is exactly the point — so the "supply chain" here is (a) the physical water/dispatch chain and (b) the equipment/EPC chain that built and maintains the fleet. Named stakeholders end-to-end:

• Upstream input — the watershed. The Jinsha River (upper Yangtze, Tibetan Plateau snowmelt + monsoon) feeds the cascade in sequence: Wudongde → Baihetan → Xiluodu → Xiangjiaba (all on the Jinsha), then the mainstem Three Gorges → Gezhouba. The cascade is hydrologically linked: upstream reservoirs (Baihetan's huge storage especially) regulate flow to optimize downstream generation — a genuine operating synergy, not just a portfolio. The counterparty here is nature; the "supplier risk" is drought (see Lens 13).
• Equipment / turbines (the build-and-maintain chain): giant Francis turbine-generator units supplied historically by Dongfang Electric, Harbin Electric, and international majors (GE/Alstom, Voith, Andritz) on the earlier stations; Baihetan is famous for the world's first fleet of 1,000 MW (1 GW) hydro turbine units, a Chinese-engineered milestone (Dongfang/Harbin). Ongoing spend is maintenance, digitalization, and life-extension — not new greenfield turbines.
• EPC / dam construction: built by parent CTG and state hydro-engineering arms (Sinohydro / PowerChina, Gezhouba Group). Critically, CYPC buys the finished, commissioned asset from CTG — it is the operator/owner, not the builder. This is the core of the parent-child model: CTG constructs and de-risks (and bears the multi-decade construction overrun risk), then injects the running station into the listco (Baihetan/Wudongde injected Jan 2023 for CNY 80.5B — see Lens 9).
• The company (midstream): CYPC operates the dams, manages cascade dispatch, and sells the output.
• Downstream — transmission & offtake: State Grid and China Southern Power Grid wheel the power via UHV DC/AC corridors (e.g. Three Gorges → East China; Baihetan → Jiangsu/Zhejiang and Guangdong on dedicated ±800 kV links) to industrial and residential end-users across ~9 provinces plus Shanghai.

Chokepoints / single-source dependencies: (1) Hydrology is the ultimate single point of failure — one basin, one climate regime; a multi-year drought hits the entire fleet at once (Lens 13). (2) The grid buyers are a duopoly — CYPC has no alternative route to market. (3) The parent — CYPC's growth pipeline of new capacity is single-sourced from CTG's construction arm (Jinsha upper reaches, Tibet mega-projects). Names or it didn't happen: CTG (parent/EPC), Sinohydro/PowerChina, Dongfang Electric, Harbin Electric, State Grid, China Southern Power Grid — that is the chain.

Lens 3 · Competitive Advantages (moats)

CYPC has one of the widest, most durable moats in global utilities — but it is a moat with a state-imposed ceiling on how much of it accrues to minority shareholders.

• Irreplaceable physical assets (the deepest moat). You cannot build a second Three Gorges. The great dam sites on the Yangtze are a finite, one-time geographic endowment, now fully claimed and owned by CYPC. This is the definition of an unrepeatable asset — no amount of capital lets a competitor replicate it. Switching cost for the grid is also near-total: this is baseload/peaking scale that nothing else can substitute regionally.
• Cost moat. Marginal cost of generation is effectively zero (no fuel). Against coal (fuel-cost-exposed) and even wind/solar (intermittent, needs firming), dispatchable near-zero-marginal-cost hydro at 72 GW is the low-cost, high-value producer. ~54% operating margins are the proof.
• Cascade / scale synergy. Owning the entire linked cascade lets CYPC optimize water storage and release across six stations to maximize total generation and shift output toward higher-value peak periods — an operating edge a single-dam owner cannot match.
• Regulatory / sovereign moat. As a strategic central-SOE asset feeding national grids, CYPC enjoys priority dispatch, favorable historical tariff treatment, and de-facto protection. The flip side (Lens 13): the same state that protects it also sets its price and can redirect the surplus (marketization, "reasonable return" caps, dividend/social obligations).
• Bargaining power — asymmetric. Over suppliers: high — it buys commissioned assets from a captive parent and equipment from competing domestic OEMs. Over customers (the grid): structurally low-to-medium — the grids are state monopsonies and the tariff is administratively influenced; CYPC cannot simply raise price. Its leverage is the physical indispensability of 72 GW of clean baseload in a decarbonizing grid.

Verdict on the moat: as close to permanent as exists in the sector on the asset side; the binding constraint is not competition, it is who captures the rent — increasingly split between minority holders, the state (tariff policy), and reinvestment.

Lens 4 · Segments

CYPC is close to a pure-play: hydropower generation is >90% of the business. There is no meaningful multi-segment story to disaggregate the way a diversified utility would have — which is itself the thesis (clean, concentrated, no conglomerate discount rationale, but no diversification cushion either). What breakdown exists:

• By activity (`` structure, exact splits not cleanly sourced): (1) Domestic six-cascade generation — the overwhelming majority of revenue and essentially all of the earnings power; (2) Ancillary / investment income — CYPC holds strategic minority stakes in other Chinese power names (it has historically been a large financial investor in SPIC, Shanghai Electric Power, Guangzhou Development, and other listed utilities), so a portion of net profit is equity-method / dividend investment income, not operating hydro; (3) Overseas — Peru (Luz del Sur distribution, see Lens 9) plus consulting/O&M in Brazil, Pakistan, Portugal — small relative to the domestic dams but the main international leg.
• By geography of generation (``): almost entirely the Yangtze/Jinsha basin (Hubei, Sichuan/Yunnan border, Chongqing). Output is sold cross-province (Central, East, South China + Guangdong), but the assets are one basin. Overseas revenue is concentrated in Peru via Luz del Sur.
• Trend & cause: the step-change was structural, not organic — the Jan 2023 injection of Baihetan + Wudongde lifted domestic capacity from ~45.6 GW to ~71.8 GW (+57%), which is why FY2023 revenue jumped +13.5% and FY2024/2025 then normalized to low-single-digit organic growth (FY2024 +8.1%, FY2025 +2.1%). Read this clearly: the big capacity leg is now in the base. From here, generation growth is a hydrology-driven wiggle around a flat ~300 TWh, not a growth ramp — until/unless the next asset injection (upper-Jinsha / Tibet projects from CTG) arrives.

Hard-requirement note: segments.csv on the shelf is empty (web-only company), so no segment figure here is . Splits above are labeled /`` accordingly — do not treat the activity mix as audited.

---

Phase B — Measure performance

Lens 5 · Earnings Result

Latest full-year print — FY2025 (reported ~early 2026):

• Revenue: CNY 86.24B, +2.07% YoY (a second source cites CNY 85.88B / +1.65% — minor discrepancy, both ~+2% and consistent).
• Net income attributable: CNY 34.50B, +6.17% YoY (corroborated ~CNY 34.17B / +5.14% ).
• Net income grew ~3× faster than revenue — the tell of a de-levering, high-operating-leverage asset: flat-ish generation + falling interest expense (post-acquisition debt paydown) + fixed cost base = margin and EPS expansion without top-line growth. This is the FY2025 story in one line.

Prior-year context (FY2024): revenue CNY 84.49B (+8.1%); net income CNY 32.50B (+19.3%); EPS CNY 1.33; total profit CNY 38.86B (+19.9%); generation 295.9 TWh (+7.1%). FY2024's outsized profit growth was a rebound off the weak, drought-suppressed 2022–2023 plus a full year of the injected Baihetan/Wudongde capacity.

H1 2025 (interim): revenue CNY 36.587B (+5.02%); net income CNY 12.984B (+14.22%); operating profit CNY 15.872B (+13.93%); six-cascade generation 126.66 TWh (+5.01%). Note Q2 softness: Three Gorges reservoir output −8.39%, offset by Baihetan +16.24% — the cascade diversification working as designed.

• Margins: FY2024 operating margin ~54% (CNY 45.31B / 84.49B). Structurally elite for a utility and stable.
• Balance-sheet flags: total assets ~CNY 568.8B; total debt ~CNY 274.4B; D/E ~1.22. The leverage is the legacy acquisition + construction debt (financing the Baihetan/Wudongde purchase and the dams themselves), not distress — it is long-dated, low-rate, sovereign-adjacent debt against 100-year assets and gushing cash flow. The positive read: this debt is being amortized, and falling interest expense is a multi-year EPS tailwind (see Lens 11).
• Market reaction / what's priced: despite record FY2024/FY2025 profits, the stock is down ~9.85% over the trailing year and sits mid-range in its 52-week band (CNY 25.38–31.19, last ~CNY 27.19). That divergence — record earnings, falling stock — is the single most important market-behavior fact in this dossier (see Lens 8): the market treats CYPC as a rate-sensitive bond proxy, and 2025–26 repriced it on yield/rate dynamics, not on operations.
• Anything unusual vs. its own history: nothing operationally alarming. The one thing to watch is that growth has flattened to ~2% now that the acquisition is annualized — the "growth utility" narrative is over; the "high-quality yield + slow compounder" narrative is what remains.

Lens 6 · Earnings Calls (sentiment trend)

CYPC does not run US-style quarterly earnings calls with transcripts; disclosure is via semi-annual/annual reports and Shanghai-exchange filings (no transcripts/ on the shelf). Management "tone" must be read from the reports and Chinese-market coverage rather than a Q&A transcript — label all of this ``.

• Persistent, unchanging management themes: (1) generation optimization — squeezing more TWh from the same water via cascade coordination and "refined dispatch"; (2) shareholder returns / dividend stability — repeatedly emphasized (the payout commitment is a core part of the equity story); (3) debt reduction — deleveraging the acquisition financing; (4) "world's largest clean-energy corridor" framing tying the company to national decarbonization.
• What's newer (last ~18 months): more language around power-market reform readiness (M2L contracting, spot-market participation), selective overseas/pumped-storage opportunities, and digital/smart-hydro operations. This is a management preparing for a more market-priced world while defending the yield.
• Sentiment trajectory: steady-to-constructive. There is no drama here — a state utility's communications are deliberately measured. The tonal shift over time is from "integrating the big acquisition / capacity growth" (2022–2023) to "optimizing a now-complete fleet, defending the dividend, and preparing for marketization" (2024–2026). That shift itself is the signal: management has pivoted from a growth story to a stewardship/yield story.

If a US-style transcript is later needed, the closest primary sources are the CYPC Semi-Annual/Annual Report MD&A sections (LSE RNS PDFs) and CTG investor-relations commentary — none scrape cleanly today.

Lens 7 · Comps

Peer set: global large-cap regulated/quasi-regulated utilities + Chinese power SOEs + hydro/renewable pure-plays. All multiples `` with source/date; where a specific line isn't sourced it is marked n/a (never fabricated).

Reads:

• Vs. Chinese SOE peers, CYPC is the premium name. It trades above CGN Power (15.9x) and well above Longyuan (8.7x) on P/E, and at a lower yield (3.7% vs 4.4%/6.4%). The market pays up for CYPC's asset quality, near-permanence, and lowest-risk cash flows — a deserved quality premium, but it means CYPC is not the cheap or the high-yield way to play Chinese power; it's the blue-chip way.
• Vs. Western regulated utilities, CYPC's ~18x P/E sits right between Duke (~19x) and below NextEra (~22x) — i.e. it is priced like a high-quality developed-market regulated utility, not at an emerging-market / China discount. On EV/EBITDA 14.3x it is cheaper than NextEra (18.7x). On EV/Sales 11.0x it optically screens rich, but that is a feature of ~54% margins (high-margin businesses always look expensive on sales) — EV/EBITDA is the fairer lens and there it is reasonable.
• The tension the table exposes: you are paying a developed-market quality multiple for a single-basin, hydrology-exposed, state-tariff-controlled asset. The bull says the moat justifies it; the bear (Lens 13) says China-utility risk (marketization, SOE cash claims, RMB) should command a discount CYPC doesn't fully offer.

Lens 8 · Stock-Price Catalysts

What actually moves this stock (mostly ``):

• Hydrology / generation prints (the biggest operational driver). The 2022 record drought (Yangtze flow >50% below 5-yr average, Sichuan hydro halved) crushed 2022–2023 output and earnings and was the last time operations truly dented the stock. Conversely, good water years (2024, H1-2025) drive the earnings rebounds. For this name, the weather is the fundamental catalyst.
• The Baihetan/Wudongde asset injection (2021 announce → Jan 2023 close, CNY 80.5B, +57% capacity). The defining corporate event of the last cycle — it re-rated the growth profile and lifted the earnings base. Now fully in the base; the next injection is the next catalyst.
• Interest rates / bond-proxy dynamics — the dominant driver of the last two years. CYPC trades like a duration asset: falling Chinese rates and the 2024 "high-dividend A-share" defensive bid pushed it to record highs; the trailing-12-month −9.85% despite record profits reflects a give-back / rotation as the defensive trade cooled and yields moved. The market reacts to CYPC's yield spread vs. Chinese govvies more than to its TWh.
• Dividend policy / payout signals. Any change to the ~90%-of-profit payout (up or down) is a first-order catalyst — this is a stock owned for the coupon.
• Power-market reform headlines. Marketization/tariff-mechanism news (M2L rules, spot-market rollout to Sichuan/Hubei) is an emerging swing factor — a genuine risk to the "administered tariff" comfort.
• Consensus stance: aggregators show a "Strong Buy" average rating with a price target of CNY 32.79 (+20.6% upside) from ~CNY 27.19 — the sell-side sees the post-selloff level as attractive on yield + quality.

Pattern in one line: operationally it reacts to rain; as a security it reacts to rates and the dividend.

---

Phase C — Judge people & books

Lens 9 · Management

• Archetype: this is a state-appointed SOE management team, not founder-owners. The chairman is Ma Zhenbo (romanized in some feeds as "Zhen Bo Ma"; appointed to the role in 2024) with a board dominated by CTG/SASAC-aligned directors. Individual-executive granularity is thin in English sources (label ``, low confidence on names). What matters is the structure: capital allocation is ultimately set by the state and the parent, not by an independent, incentive-aligned management.
• Track record (institutional, not individual): the franchise's defining achievements — flawless operation of the world's largest hydro fleet, the smooth CNY 80.5B Baihetan/Wudongde integration (2023), and the $3.59B Luz del Sur (Peru) acquisition (2020) — are real, large-scale execution wins. Cascade dispatch optimization that squeezes record TWh from the same water is genuine operating skill.
• Skin in the game: essentially none at the management level (no meaningful founder/insider equity — this is a professionally-run SOE). Alignment comes from CTG's ~62–64% stake, i.e. the controlling shareholder's interest, which is the state's interest — overlapping with but not identical to minority holders' interests (the state may prioritize energy policy, price stability, or SOE reform goals over per-share value).
• Capital-allocation history — the crux: (1) Dividends — strong and consistent; ~90% payout, a multi-year track record of returning cash, 5-yr dividend growth ~+7%. This is genuinely shareholder-friendly and the core of the bull case. (2) M&A — the parent-injection model (buy commissioned dams from CTG) is accretive and low-risk by design (you buy a running asset, not a construction project). Luz del Sur diversified internationally. (3) Financial investments — large minority stakes in other listed utilities (SPIC, Shanghai Electric Power, etc.) are a mixed use of capital — defensible as sector consolidation/income, but it's a partial "cash-box" behavior that dilutes the pure hydro story. (4) Deleveraging — steadily paying down acquisition debt, which is quietly the biggest driver of forward EPS.
• ROE/ROIC trend: FY2025 ROE ~15.6%, ROA ~5.3%, ROCE ~10.9% — solid and stable for a capital-heavy utility, well above cost of capital.
• Red flags (governance): the standard SOE set — (a) related-party dependence on CTG for the entire growth pipeline (asset injections are related-party transactions; pricing fairness to minorities is a perennial question); (b) state can redirect the surplus (tariff caps, dividend expectations, social/policy obligations); (c) thin minority-holder influence given the ~62% control. None are "fraud" flags — they are structural principal-agent flags inherent to owning a Chinese central-SOE.

Lens 10 · Forensic Red Flags

Acting as a forensic analyst on a business this simple: the accounting risk surface is unusually small. No complex revenue recognition (sell power to grid at tariff), no meaningful inventory, minimal receivables risk (sovereign buyers), no aggressive non-GAAP adjustments culture. The real "quality of earnings" questions are structural, not manipulative:

• Revenue recognition: low risk — metered generation × tariff, settled with state grids. Little room for channel-stuffing or timing games.
• Depreciation policy (the one to actually scrutinize): the single largest accounting judgment is useful-life assumptions on the dams and turbines. Dams are depreciated over very long lives; the choice of life materially affects reported profit. Longer assumed lives flatter current earnings; if lives are conservative, earnings are understated (a hidden positive). This is worth a look in the primary financials (not scrapeable today) — flag as ``/unverified: depreciation-life assumption is the key earnings lever.
• Cash flow vs. earnings: for hydro, operating cash flow should substantially exceed net income (huge non-cash depreciation add-back). If it doesn't, that's the red flag. Reported metrics (high FCF-supporting a ~90% payout) are consistent with the expected profile — no obvious divergence, but the exact OCF/NI bridge is n/a from primary here.
• Receivables / inventory outrunning revenue: low structural risk (sovereign offtake, no inventory). Not flagged.
• Debt & interest: ~CNY 274B debt / D/E 1.22 is the genuine risk item — but it's disclosed, long-dated, and amortizing, not hidden. Interest-rate resets on refinancing are the thing to track.
• Financial-investment / equity-method income: a portion of net profit is investment income from minority stakes (SPIC et al.) — lower-quality, less-repeatable than core hydro earnings. Analysts should haircut the multiple applied to that slice. ``, exact contribution not cleanly sourced.
• SBC: immaterial (SOE, no Silicon-Valley-style equity comp). Not a distortion here.

Regulatory findings (required sub-section):

• SEC enforcement (EDGAR LR + AAER): None possible / none found. Per regulatory/regulatory-findings.md (generated 2026-07-06), CYPC has no CIK and does not file with the SEC, so no EDGAR Litigation Releases or AAERs can name it. total_sec_findings: 0.
• Non-SEC enforcement (web): No material FTC/DOJ/FDA/CFPB/consent-decree/fine hits surfaced for "China Yangtze Power" in the searches run — as expected for a domestic Chinese SOE with essentially no US operational footprint (its overseas exposure is Peru/Brazil/Portugal/Pakistan, not the US). Historical, non-enforcement controversy attaches to the Three Gorges Dam project itself (displacement of ~1.3M people, ecological/seismic debate) — reputational/ESG context, not a securities-enforcement finding. ``.
• 10-K Item 3 (Legal Proceedings): n/a — no 10-K exists (foreign A-share, no EDGAR). Material-litigation disclosure lives in the Chinese annual report, not scrapeable here.
• Net: No material securities-regulatory or legal enforcement findings — verified via SEC EDGAR EFTS (LR, AAER; not applicable — no CIK) and web search as of 2026-07-06. The relevant "risk" is not enforcement but state/policy control and ESG (dam-legacy) exposure, covered in Lens 13.

---

Phase D — Project & stress-test

Lens 11 · Forward Projection

Bottom-up from FY2025 actuals (revenue CNY 86.24B, net income CNY 34.50B, EPS ~CNY 1.41 on ~24.47B shares ). Fiscal years FY2026 / FY2027 / FY2028. All outputs `` with arithmetic; inputs labeled.

Driver assumptions:

• Generation: structurally flat at ~295–305 TWh in a normal-hydrology year — the fleet is built out; no new capacity in the base case until the next CTG injection (not assumed here). ±10% swing on weather is the real variable. ]
• Tariff: roughly flat to modestly pressured as marketization proceeds — assume −0.5% to −1%/yr blended realized price drift in base.
• Interest expense: declining as acquisition/construction debt amortizes and (partially) refinances at lower Chinese rates — the key EPS tailwind. Assume this adds ~2–3 pts to net-income growth above revenue growth.
• Share count: roughly flat (~24.47B); no major issuance assumed.

Base-case path (illustrative, +~6% NI/yr on deleveraging + normal water): FY2026E EPS ~CNY 1.49 → FY2027E ~CNY 1.58 → FY2028E ~CNY 1.67. At an unchanged ~18x P/E that supports roughly the consensus PT area (~CNY 27–30 on FY26–27 EPS; the sell-side CNY 32.79 target implies mild multiple expansion or faster deleveraging).

The valuation reality: total return here is dividend (~3.7%) + low-single-digit EPS growth (~5–6%) ± the weather. That's a ~9–10% base-case total return in a normal year — a high-quality bond-plus, not a compounder. The upside case needs either a wet-year earnings surprise, or rate-driven multiple expansion (the bond-proxy bid returning), or a new accretive asset injection from CTG. The downside case is a bad-hydrology year compounded by tariff give-up.

No forecast.ts create in this unattended --watchlist run (per skill: only log a Brier forecast on genuine committed conviction, not in the sweep). Position seed carries the numbers forward for a later /thesis pass.

Lens 12 · Bull vs Bear

Bull case. You are buying the single best physical energy asset on the planet — 72 GW of clean, dispatchable, near-zero-marginal-cost generation that literally cannot be replicated — at ~18x earnings / 14x EBITDA with a ~3.7% yield, a ~90% payout, 15%+ ROE, and a hidden earnings tailwind (deleveraging shrinks interest expense for years, growing EPS even with flat generation). It is the ultimate defensive/decarbonization compounder: bond-like stability with equity-like inflation and rate optionality, backed by the Chinese state, feeding a grid that will only value firm clean power more as coal retires. Contrarian earnings surprise: a genuinely wet year drops straight to the bottom line at ~54% margins. Capital allocation is disciplined and shareholder-friendly (dividends + accretive parent injections). In a lower-rate world, this re-rates as the premier Asian yield-quality name.

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

1. Tariff marketization is a one-way de-rating risk to the moat's economics (not its physics). As China forces more volume onto traded M2L/spot prices (Sichuan/Hubei spot markets live by end-2025; >40%-hydro provinces' contract floors relaxed from 80% to 60% ), the administered "reasonable-return" comfort erodes. Hydro's near-zero marginal cost means in a merit-order spot market it could be pushed toward the marginal (low) clearing price rather than capturing its full value — a structural margin/tariff headwind that compounds quietly.
2. The weather — and it's getting worse. One basin, one climate. The 2022 drought halved Sichuan hydro and dented earnings for ~2 years; climate change raises the frequency/severity of both drought and extreme-flood years. This is an undiversifiable, intensifying tail risk sitting under the entire 72 GW at once. No balance sheet fixes it.
3. State/SOE claim on the surplus + expectations already high. At a developed-market ~18x multiple, minimal China discount is priced, yet the state controls the tariff, can lift dividend/policy/social obligations, and prioritizes energy-security and price-stability goals that may not maximize per-share value. RMB depreciation erodes the USD return. The stock already fell ~10% in a year of record profits — proof the multiple, not the moat, is the swing factor.

Pre-mortem (18 months out, thesis broke — what happened?): A dry 2026–27 (weak generation) collides with an accelerated spot-market rollout that visibly compresses realized hydro tariffs; simultaneously Chinese long rates back up (or the high-dividend defensive trade unwinds further), so the bond-proxy loses its bid. Earnings miss and the multiple de-rates — the classic double-hit on a yield stock. The dividend holds (that's the floor), but the stock still draws down 20–25% as "safe compounder" reprices to "rate-and-rain-sensitive utility."

Are multiples too high? Not egregiously — 14x EBITDA / 18x P/E for an asset this durable is defensible. But it prices CYPC as a developed-market quality utility while leaving it exposed to emerging-market policy/tariff/FX risk. The multiple gives you no margin of safety for the marketization + drought scenario.

Contrarian view (what the market refuses to see): The bulls treat CYPC as a bond substitute and the bears fixate on China risk — both miss that the terminal value of dispatchable, storage-capable hydro is rising as wind/solar flood the grid and firming becomes the scarce commodity. A reservoir cascade is a giant, pre-built "battery." If China's power-market reform ever properly prices flexibility/capacity (not just energy), CYPC's ancillary/peaking value could be re-rated upward — the reform is framed only as a tariff risk, but it could equally become a flexibility-value windfall. That optionality is unpriced.

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

Dismantling the bull case:

• What structurally breaks the money machine? A regime where hydro no longer captures a protected "reasonable return." Marketization is the mechanism: force the near-zero-marginal-cost asset into merit-order spot pricing and its realized tariff can fall even as it dispatches everything — margin compression with no volume offset. The bulls' "fortress margin" is a policy gift, and policy is being rewritten in real time.
• Where is revenue concentrated, and what if it shifts? 100% one river basin, >90% one activity, sold to a two-buyer state monopsony under an administered price. There is no diversification. A shift in any of the three — hydrology, tariff policy, or grid-dispatch priority — hits the whole company. Peru/overseas is too small to matter.
• Why is the moat weaker than bulls think? The asset moat is genuine and permanent; but the moat that matters for shareholders is the ability to convert the asset into growing per-share cash flow, and that moat is thin — capped by state tariff-setting, ~62% parent control that can claim the surplus, and a growth pipeline that only exists at the parent's discretion and pricing. You own a wonderful asset through a straw the state controls.
• Most dangerous thing bulls underestimate: the multiple, not the business. They've re-rated a flat-growth, rate-sensitive, single-basin utility to ~18x on a "quality/defensive" story. When the rate cycle or the dividend-trade sentiment turns (as it did, −10% in a record year), there is a long way to de-rate before the ~3.7% yield puts in a floor.
• Worst capital-allocation risk: related-party asset injections priced by the controlling parent (fairness to minorities is unfalsifiable from outside), plus the "cash-box" minority stakes in other utilities that dilute the clean hydro story and add opacity.
• What must hold for today's price? Normal-to-good hydrology and broadly stable realized tariffs through the marketization transition and a persistent low-rate/defensive bid and an undisturbed ~90% payout. Several of those are outside management's control.
• If growth disappoints 20–30%? Since base-case growth is only ~5–6% (mostly from deleveraging, not operations), a "20–30% growth disappointment" here really means a drought year that turns +6% into −10% NI — plausibly a 15–25% earnings swing and, if it coincides with a multiple de-rate, a 20–30% price drawdown. The dividend cushions but does not prevent it.
• Single scenario that permanently impairs the business: a structural shift to unfavorable merit-order spot pricing for hydro (permanent tariff/margin reset) — more plausible than the physical tail (a multi-decade mega-drought). Probability: the physical catastrophe is low; the policy/tariff erosion is moderate-and-rising and is the short-seller's real thesis.

Lens 14 · Management Questions (ordered by information value)

1. As Sichuan/Hubei spot markets go live and M2L contract floors relax to 60%, what share of your generation will be exposed to market (vs. administered) pricing in 2026–2028, and what is the realized-tariff sensitivity to a full merit-order transition?
2. What is your explicit dividend policy going forward — is the ~90% payout a floor, a target, or discretionary — and under what conditions (drought year, new capex cycle) would you cut it?
3. What is the concrete pipeline and expected timing/pricing of the next asset injection from CTG (upper-Jinsha, Tibet mega-projects), and how will you protect minority shareholders on related-party valuation?
4. Quantify the interest-expense glide path: how much of the ~CNY 274B debt reprices or amortizes over 2026–2028, and how much EPS growth does deleveraging contribute annually?
5. Run us through a repeat-of-2022 drought scenario: what did/would it do to generation, net income, and the dividend, and what physical or contractual hedges now exist that didn't in 2022?
6. What are the useful-life and depreciation assumptions on the dams/turbines, and how much reported profit is sensitive to a change in those lives?
7. How should we value the minority equity stakes (SPIC, Shanghai Electric Power, etc.) — are these strategic-permanent, income-generating, or eventually monetizable, and what's their contribution to net income?
8. Does China's power-market reform create an opportunity to monetize flexibility/capacity/ancillary services from the cascade's storage, and are you positioned to capture that upside rather than just the energy tariff?
9. What is your overseas strategy post-Luz del Sur — is Peru a template for further OECD/LatAm regulated-asset acquisitions, and what return threshold governs it?
10. How exposed is the equity return to RMB depreciation for foreign holders, and do you have (or plan) any USD-linked cash flows or hedges?
11. What is the maintenance/refurbishment capex required to keep 72 GW at peak availability over the next decade, and how does that trend?
12. How do you think about pumped-storage development — is CYPC a builder/owner of pumped hydro, and how material could it become to the earnings mix by 2030?
13. What climate-adaptation investment (reservoir management, forecasting, basin resilience) are you making, and how do you plan generation around a more volatile hydrological regime?
14. What is your long-term capital-allocation priority stack — dividends vs. domestic injections vs. overseas M&A vs. debt reduction vs. new energy — when they compete for the same yuan?
15. What role does CYPC play in CTG's and the state's broader energy-security and SOE-reform agenda, and where might national-policy objectives diverge from per-share value maximization?

---