China Three Gorges Renewables: The Dam-Builder's Bet on Wind and Sun

I. Introduction & Episode Roadmap

On the evening of April 29, 2026, a filing landed on the Shanghai Stock Exchange server that would have been unthinkable to the people who bought this stock five years earlier. China Three Gorges Renewables — 三峡能源 in the ticker tape, the listed renewable-power arm of the state enterprise that built the Three Gorges Dam — reported that its 2025 net profit had fallen 39.2%, to ¥3.71 billion, on revenue of ¥28.4 billion, itself down 4.4%.¹² Strip out one-off gains from selling stakes in subsidiaries and the picture was worse: adjusted profit fell 48.03%.¹³

The fourth quarter had actually been a loss — ¥598 million of red ink in the final three months of the year, against a ¥498 million profit in the third quarter.¹ For a company whose entire business is selling electricity from assets that had already been paid for, into a grid that is legally obliged to buy the output, a quarterly loss is not a rounding error. It is a signal that something structural has changed.

Here is the part that makes this a story rather than a bad quarter. Over the same period in which profits nearly halved, the company did not shrink. It grew — enormously. Installed grid-connected capacity reached 52,374 megawatts at the end of 2025, up from 47,961 MW a year earlier and roughly 22,896 MW at the end of 2021.¹⁴ Power generation rose 5.99% to 76.26 billion kilowatt-hours.¹ By any operational measure, 2025 was a year of expansion. The company built more, generated more, and connected more than ever before.

And it earned barely half of what it had earned in 2023.

The One Number That Explains Everything

The reason sits in one number. In 2023, the average price CTGR received for every kilowatt-hour it fed into China's grid was ¥0.4883. In 2024 it was ¥0.4195. In 2025 it was ¥0.3767.¹⁵ That is a decline of nearly a quarter in two years, in a business where roughly the entire cost base — depreciation on turbines and panels, interest on the debt that bought them — is fixed the day the asset is switched on. When the price falls and the costs do not, the arithmetic is brutal and immediate.

A Policy With a Serial Number

The proximate cause has a document number. On February 9, 2025, China's National Development and Reform Commission (国家发展和改革委员会) and National Energy Administration (国家能源局) jointly issued 发改价格〔2025〕136号 — Document 136 — mandating that wind and solar generation move from administratively set tariffs to market-formed prices, with a "mechanism price" transition determined province by province and, for new projects, set by competitive auction.⁶⁷ The state that had spent fifteen years guaranteeing the economics of Chinese renewables announced that it would stop.

This episode is about what happens to a company built entirely for one regime when that regime is rewritten. Along the way we will look at how a hydropower giant decided to hedge its own core business; how ¥22.7 billion of IPO money and a captive relationship with China's grid produced one of the fastest asset buildouts in global power history; how that buildout was financed by debt that has never once been repaid out of the company's own free cash flow; how nearly ¥50 billion of unpaid state subsidy sits on the balance sheet as an asset that management does not control the timing of; and how a brand-new management team, installed almost exactly as the policy landed, is now being asked to learn a skill — selling electricity in a competitive market — that this company never needed for the first fifteen years of its life.

We will also test the bull case honestly. There is a real one. Scale in offshore wind is genuine, the parent's balance sheet is genuine, and China is still adding renewable capacity faster than any country ever has. But we will test the bear case with the same seriousness, because as of August 2026 the bear case is no longer hypothetical. It has already shown up in the numbers — and it kept showing up in the first quarter of 2026, when profit fell another 57.49%.⁸

Let us start with the dam.

---

↑ Back to Top

II. Origins: A Dam-Builder Hedges Its Bet on Water

Running Out of Rivers

There is a specific kind of anxiety that afflicts a company that has already built the largest thing of its kind in the world. 中国长江三峡集团有限公司 China Three Gorges Corporation — CTG — completed the Three Gorges Dam, a 22,500 MW hydroelectric station on the Yangtze that remains the single largest power plant on Earth by installed capacity. It then built and operated a cascade of further mega-dams upstream. By the 2010s, CTG had a problem that most industrial companies would envy and no growth-minded management team enjoys: it had run out of rivers.

China's best hydropower sites are finite, and the good ones were taken — largely by CTG itself. A state enterprise whose mandate is clean energy, whose engineering culture is built around enormous civil-works projects, and whose balance sheet is among the strongest in the Chinese power sector, needed a second act. Wind and solar were the obvious answer. They shared the essential economics of hydro — heavy upfront capital expenditure, near-zero fuel cost, long asset lives, revenue from selling electrons to a grid — while being infinitely repeatable in a way that a river is not. You cannot build a second Yangtze. You can build a thousand wind farms.

The vehicle CTG assembled for this began life as China Three Gorges New Energy, folded together from earlier state assets — the group absorbed China Water Investment Group in December 2008, which brought a portfolio of small hydro and early wind projects with it.¹ For a decade it operated as a side business: useful, growing, but not the point. That changed in June 2019, when the company was renamed China Three Gorges Renewables (Group) Co., Ltd.¹ A renaming is a cheap signal, but in a Chinese central state-owned enterprise it is rarely a casual one. Dropping "New Energy" for "Renewables" and attaching the parent's most valuable brand asset — the Three Gorges name — to the vehicle was an internal declaration that this was no longer a hedge. It was the growth engine.

The Best-Timed Listing in Chinese Power

The timing was, in retrospect, close to perfect. In September 2020 China committed to peaking carbon emissions before 2030 and reaching carbon neutrality before 2060 — the 双碳 dual carbon goals that would reorganise a decade of Chinese industrial policy. Every province, every grid company, and every central SOE now had a decarbonisation quota to fill, and the companies that could build renewable capacity fastest were about to become instruments of national policy rather than merely participants in a market.

On June 10, 2021, CTGR listed on the Shanghai Stock Exchange's main board. It sold 8.571 billion new shares at ¥3.00 apiece, raising ¥22.713 billion gross — at the time the largest IPO in the history of China's power industry.⁴ The stock closed its first day up more than 44%, took the company's market value through ¥100 billion on day one, and then ran up eight consecutive daily limit-ups over the following fortnight.⁴

What is worth pausing on is what kind of IPO this was. This was not a spinout under duress, or a private developer selling into a hot market to fund a cash-burning balance sheet. CTG had methodically executed what the company itself described as a three-step reform roadmap — corporatisation, bringing in strategic investors, then listing.⁴ Strategic investors including 浙能资本 Zheneng Capital, 都城伟业 Ducheng Weiye and Sichuan Chuantou Energy had already taken stakes before the listing, and they are still on the register today.¹ The float was engineered, not extracted.

That structure conferred two advantages that a private developer could not replicate. The first was cost of capital. A central-SOE subsidiary with an implicit and often explicit parent guarantee borrows at rates that private Chinese renewable developers cannot access; CTGR's finance costs actually fell 3.85% in 2025 even as its debt pile grew, because, in management's own words, it caught a favourable funding window and cut its blended project financing cost.¹ The second was resource access. Offshore wind in particular is not a market you enter by being clever. It requires sea-area use rights, provincial development quotas, grid connection commitments, and the political capital to secure all three simultaneously in Guangdong, Fujian and Jiangsu. CTG had that capital. Nobody outside the national team did.

The result was that CTGR entered the 2020s with the two things that mattered most in a capacity race: money that was cheap, and permission that was scarce. For roughly four years, that combination worked exactly as designed. What it did not do — and this is the thread that runs through everything that follows — was create any advantage in the thing that would eventually matter most: the price at which the electricity is sold.

---

↑ Back to Top

III. How This Utility Actually Makes Money

A Business With No Customers to Win

Strip away the megawatts and the policy documents and CTGR's business model, for most of its existence, could be written on the back of a boarding pass. Raise capital. Build a wind farm or a solar plant. Connect it to the grid. Collect a price that the government set, from a buyer that the government owned, for twenty years. Repeat.

That sentence contains something genuinely unusual, and it is worth dwelling on because it inverts almost everything a Western investor assumes about competitive business. In a normal company, growth requires winning customers, defending price, and out-executing rivals on product. In CTGR's world, none of those variables existed. The customer was fixed: in 2025, the top five customers accounted for 98.10% of revenue, with 国家电网 State Grid alone at 71.74%, 中国南方电网 China Southern Power Grid at 15.58%, and Inner Mongolia Power at 10.20%.¹ The price was fixed by the price bureau. The product — electrons — is the most perfectly commoditised good in existence and requires no marketing whatsoever.

So what was the constraint? Capital and construction. That is the entire game. If you could raise the money and physically build the asset, revenue followed with near-mechanical certainty. This is why CTGR's income statement for its first decade looks less like a business and more like a compounding machine: revenue rose from ¥8.96 billion in 2019 to ¥11.31 billion in 2020, ¥15.48 billion in 2021, ¥23.81 billion in 2022, ¥26.53 billion in 2023, and ¥29.72 billion in 2024.⁴¹⁹

The Machine That Never Paid For Itself

But a business whose only constraint is capital has a specific and dangerous financial signature, and CTGR's balance sheet shows it clearly.

Consider 2025. The company generated ¥20.94 billion of cash from operations — a genuinely strong number, up 10.81% on the prior year, and evidence that the physical assets do throw off cash.¹ It then spent ¥32.11 billion buying and building fixed and intangible assets.¹ The gap — more than ¥11 billion — was filled by borrowing ¥48.31 billion of new debt against ¥30.06 billion of repayments, plus ¥3.65 billion of fresh minority equity injected into project subsidiaries.¹ The prior year followed the identical pattern: ¥18.89 billion of operating cash against ¥30.89 billion of capex.¹

In plain English: this company has never funded its own growth. Not once. Every megawatt added since the IPO has been financed by external capital, and the moment that capital becomes expensive or unavailable, the growth stops. Total debt reached ¥187.54 billion at the end of 2024, a 15.24% increase in a single year, against EBITDA of ¥24.90 billion — a ratio of about 7.5 times, up from roughly 7.0 times a year earlier.⁹ The consolidated asset-liability ratio has climbed steadily from 64.73% at the end of 2021 to 71.91% at the end of 2025.⁴¹ The current ratio fell from 101.48% at end-2023 to 86.65% at end-2024 — meaning short-term obligations now exceed short-term assets.⁹

Lianhe Ratings, which maintains an AAA domestic rating on the company with a stable outlook, describes this candidly. Its June 2025 tracking report notes that operating cash inflow "still cannot meet investment expenditure requirements," forcing external financing that has increased the debt burden.⁹ The rating agency's stated downgrade trigger is instructive: an adjustment in clean-energy policy direction that significantly weakens operating cash flow.⁹ That trigger was pulled less than four months before the report was written. The rating held anyway — which tells you something about how much of the AAA is CTGR's own credit and how much is CTG's.

Lending ¥48 Billion to the State, Interest-Free

Now the second structural feature, and the one that deserves to be understood early rather than treated as a footnote: the receivables.

At the end of 2025, CTGR carried ¥48.25 billion of net accounts receivable on a balance sheet with ¥28.4 billion of annual revenue.¹ The gross figure was ¥51.00 billion against a bad-debt provision of ¥2.75 billion.¹ Receivables are worth more than a year and a half of sales. For a utility selling to a state-owned grid, that is extraordinary.

The explanation is the 可再生能源电价附加 — the renewable energy tariff surcharge. Under the old regime, renewable generators received a base grid price plus a national subsidy funded by a levy on electricity consumers. The base price was paid reasonably promptly. The subsidy was not. The subsidy fund ran a structural deficit for years, and disbursement depends on fiscal budget allocation progress rather than on any commercial obligation. In the underwriter's verification opinion for CTGR's 2025 corporate bond issue, the breakdown was disclosed: of ¥44.76 billion of net receivables at end-2024, ¥42.43 billion — 94.80% — was national subsidy.¹⁰ That proportion has been rising, from 92.03% at end-2022.¹⁰

Trace the trajectory and the picture sharpens. Net receivables were ¥26.73 billion at end-2022, ¥36.69 billion at end-2023, ¥44.76 billion at end-2024, and ¥47.69 billion by the end of March 2025.¹⁰ Revenue over the same period grew roughly 25%. Receivables grew roughly 80%. A Sina Finance screening alert in April 2025 flagged precisely this: receivables grew 22.04% in 2024 against 12.13% revenue growth, pushing receivables to 150.61% of revenue from 138.47% a year earlier.¹¹

When an investor asked management directly in 2024 whether ¥44.36 billion of receivables — roughly six years of the company's then-annual profit — might not be collected, the company's response did not dispute the size. It restated that the state disburses the surcharge according to fiscal budget progress, and pivoted to what it could control: accelerating project deployment, improving market trading capability, and growing green-certificate revenue.¹²

That answer is honest, and it is also the whole problem. CTGR has been extending unsecured, uncollateralised, interest-free credit to the Chinese state, at a scale that now exceeds half of its shareholders' equity, on a repayment schedule it does not set and cannot influence. The company books the subsidy as revenue when the electricity flows, because under Chinese accounting standards the state's credit makes eventual collection probable.¹⁰ The auditor, Shinewing, identified accounts-receivable bad-debt provisioning as a key audit matter for 2025 precisely because the judgment is so material.¹ The provision ratio has been raised steadily — 3.21% at end-2022, 3.93% at end-2023, 4.84% at end-2024 — and CTGR has fully provisioned one legacy receivable against State Grid Jiangsu relating to benchmark tariffs from a 2012–2014 test turbine project.¹⁰

Nothing here suggests fraud or aggressive accounting. It does suggest that reported net income and distributable cash are two very different things at this company — a distinction that becomes critical when we get to the dividend.

---

↑ Back to Top

IV. The Segments: Where the Megawatts (and Margin) Actually Sit

Offshore: The Crown Asset and Its Price Tag

Walk the coastline of Guangdong province and, roughly forty kilometres offshore near Yangjiang, you will find the physical expression of CTGR's strategic identity: a cluster of turbines called Qingzhou, phases five, six and seven, each a full gigawatt, each costing north of ¥13 billion.¹ These are not wind farms in the sense that a Texas ranch has a wind farm. They are offshore civil-engineering campaigns, with subsea cabling, offshore converter stations running at ±500kV, and installation windows dictated by typhoon season.

Offshore wind is where CTGR makes its claim to being something other than a generic Chinese renewable developer. At the end of 2025 the company had 7,546.8 MW of offshore capacity connected, which it calculated as 16.06% of China's national total — a genuine leadership position in the highest-value, highest-barrier segment of the industry.¹ It added 497 MW of offshore capacity during the year, 7.54% of the national additions.¹ The strategy has an internal slogan, 海上风电引领者 — "offshore wind leader" — and the FY2025 annual report commits to building a full-coastline offshore wind corridor spanning Liaoning, Shandong, the Yangtze delta, Guangdong, Hainan and Guangxi, and to pushing into deep-water, far-shore sites.¹

Why does offshore matter disproportionately? Two reasons. First, capacity factors: offshore wind blows harder and more consistently, so a megawatt offshore produces meaningfully more kilowatt-hours per year than a megawatt onshore, and far more than a megawatt of solar. Second, location: offshore farms sit next to the coastal provinces that consume the most power, which means the electricity is absorbed locally rather than needing to be wheeled two thousand kilometres across a congested grid.

The cost, however, is spectacular. Using CTGR's own disclosed project budgets, Qingzhou 5 was planned at ¥14.05 billion for 1,000 MW; Qingzhou 7 at ¥13.36 billion; Qingzhou 6 at ¥13.76 billion; the 400 MW Putian Pinghaiwan DE project in Fujian at ¥4.70 billion.¹ That is roughly ¥12–14 million per megawatt. Compare the same report's onshore projects: a 500 MW onshore wind project in Qinghai at ¥2.13 billion, and the 312.5 MW Xinxian onshore wind project in Shandong at ¥1.72 billion.¹ Onshore costs roughly ¥4–5.5 million per megawatt. Offshore, in other words, costs about three times as much per megawatt to build.

Under a guaranteed-tariff regime with a premium offshore price, that trade worked. Under market pricing, it becomes a much sharper bet: three times the capital, exposed to the same market clearing price as everyone else, with harsher operating and maintenance economics on top.

Onshore Wind and the Solar Problem

The onshore wind fleet is the quiet workhorse. Total wind capacity was 24,432.6 MW at end-2025, so onshore accounts for roughly 16,900 MW — more than double the offshore fleet.¹ It is cheaper, more mature, and generally more reliably dispatched than solar. Wind as a whole delivered ¥18.84 billion of revenue in 2025 at a 46.47% gross margin, down 6.55 percentage points year on year.¹

Solar is the volume story and, increasingly, the problem. CTGR's solar capacity reached 26,780.5 MW at end-2025, having added 2,514.8 MW during the year — more than wind's 2,000.5 MW of additions.¹ Solar is now the largest single block of megawatts the company owns. It is also the segment where the economics have deteriorated fastest: solar revenue was ¥8.82 billion at a 33.98% gross margin, down a striking 19.10 percentage points year on year.¹ The realised solar tariff fell to ¥0.3265/kWh in 2025 from ¥0.3673 in 2024 and ¥0.4938 in 2023 — a 34% decline in two years.¹⁵

This is the single most important structural fact about CTGR's asset mix, and it separates the company from its closest peer. 龙源电力 Longyuan Power, the wind arm of China Energy Investment and the original Chinese wind pioneer, ended 2025 with 32,147 MW of wind against 13,841 MW of solar — a fleet that is 70% wind.¹³ CTGR's is 51% solar. Solar has lower capacity factors, lower realised prices, and worse curtailment exposure. When market pricing arrived, the company with the heavier solar weighting was always going to be hit harder. It was.

The Optionality — and the Gas Plant

Two smaller lines deserve a sentence each. The small and medium hydro portfolio, which the company had carried since the 2008 absorption of China Water Investment, was deconsolidated from January 2025 — CTGR no longer controls hydro capacity, and hydro generation is absent from 2025 results entirely.¹ Separately, "other" capacity of 1,161 MW consists of standalone energy storage projects.¹

And then there is the optionality. CTGR discusses pumped storage, new-type storage, hydrogen, and concentrated solar power as future growth vectors, and it has a 2,400 MW pumped-storage station under development at Golmud Nanshankou in Qinghai with a planned investment of ¥17.88 billion.¹ These are real, strategically coherent, and — as of today — immaterial to profit. Treat them as options, not as earnings.

One item in the "emerging" bucket is not optionality and deserves flagging plainly: in June 2025 the board approved investing in a gas-fired combined-cycle peaking plant in Yingkou, Liaoning, and CTGR bought 65% of the parent's Yingkou energy investment vehicle for up to ¥123.82 million as a related-party transaction.¹ At year-end 2025 the company had 7,960 MW of thermal capacity under construction, alongside 3,600 MW of pumped storage.¹ A "pure-play renewable" is building fossil generation. The strategic logic — flexible capacity supports the absorption of intermittent renewables and earns capacity payments — is defensible. But investors who bought CTGR as a clean, undiluted read on Chinese wind and solar should note that the definition of the business is being widened, and that the widening was initiated by the parent.

Everything else — EPC, equipment, consulting — is a rounding error. Power generation was ¥27.66 billion of the ¥28.40 billion total.¹ This is a wind-and-solar generator, and the analysis should be sized accordingly.

---

↑ Back to Top

V. The Hypergrowth Decade: From IPO to 50+ GW

Building a Second Three Gorges

In its 2021 annual report, CTGR permitted itself a moment of genuine institutional pride. By the end of that year, its combined wind and solar capacity reached 22,681 MW — surpassing, for the first time, the 22,500 MW nameplate of the Three Gorges Dam itself.⁴ Management gave it a name: 风光三峡, a "wind-and-solar Three Gorges." A subsidiary had, in capacity terms, out-built the monument its parent was named after.

That was the mood of the era. Over the four years that followed, CTGR roughly doubled again, ending 2025 at 52,374 MW.¹ To put that in perspective: the company added, in four years, more generating capacity than the entire installed base of most European countries. Longyuan Power, the incumbent leader that had been building Chinese wind farms since the 1990s, ended 2025 at 45,994 MW — smaller than a company that had been a listed entity for less than five years.¹³

The revenue curve tracked it faithfully. From ¥15.48 billion in 2021 to ¥29.72 billion in 2024, revenue nearly doubled while capacity slightly more than doubled.⁴⁹ That relationship — revenue growing roughly one-for-one with megawatts — is exactly what you would expect in a business where price is administratively fixed and offtake is guaranteed. It is the financial fingerprint of a regulated build-out.

When Megawatts Stopped Meaning Money

Profit told a subtly different story, and this is where the careful reader should have started paying attention well before 2025. Net profit attributable to shareholders was ¥5.64 billion in 2021, ¥7.16 billion in 2022, and ¥7.17 billion in 2023.⁴¹⁴¹ Then it went the other way: ¥6.11 billion in 2024, and ¥3.71 billion in 2025.¹

Read those two series together. Between the end of 2021 and the end of 2025, installed capacity grew 129%. Over the same period, net profit fell 34%. Against the 2023 peak, profit is down 48% while capacity is up 97%.

That divergence did not begin with Document 136. It began the moment the share of CTGR's output sold into competitive markets started rising. In 2021, market-traded electricity was 29.23% of on-grid volume.⁴ By 2024 it was 58.9%.⁵ By 2025 it was 65.80% — 48.74 billion kWh of the company's 74.08 billion kWh sent to the grid was priced by a market rather than by a bureau.¹ The FY2024 annual report, published in April 2025, explicitly named this as the first cause of that year's 14.81% profit decline: a change in the mix of on-grid electricity, with the market-traded proportion rising and blended average tariff falling.⁵

So the honest framing is this: the era was easy, but it had already stopped being easy before the policy landed. Market exposure was creeping up year by year, and each increment shaved the realised price. Document 136 did not start the erosion. It removed the possibility that the erosion would stop.

The Target That Quietly Disappeared

What about the forward growth targets? Here the analysis has to be careful, because the answer is itself informative. CTGR's FY2025 annual report contains no published numeric capacity target for 2027, 2030, or the 15th Five-Year Plan period.¹ The strategy section is entirely qualitative: consolidate offshore leadership, push desert-and-Gobi bases in Inner Mongolia's Kubuqi and southern Xinjiang, develop system-regulating resources, explore new business models, pursue M&A, and cautiously expand internationally.¹ Where numbers do appear in board records, they appear as conditions rather than commitments — in October 2025 the strategy committee approved two fishery-solar hybrid projects in Jiangsu explicitly subject to completing the mechanism-price auction and confirming that returns satisfy group requirements before construction begins.¹

That is a meaningful behavioural change, and it cuts both ways. A company that used to chase megawatts is now gating construction on realised project returns — which is precisely what a shareholder should want. It is also a company that has quietly stopped publishing the growth numbers that once anchored its equity story. Investors valuing CTGR on a growth multiple should notice that management has removed the growth target from the document where such targets belong.

---

↑ Back to Top

VI. Capital Allocation, the Parent Relationship & M&A Discipline

Who Got the Better End of the Trade?

Every controlled subsidiary of a large state enterprise lives with the same unanswerable question: when the parent sells you something, who got the better end of the trade?

CTGR's relationship with CTG runs in every direction at once. CTG is the controlling shareholder, holding 8.25 billion shares directly — 28.86% — with a further 20.99% held through the wholly-owned 长江三峡投资管理有限公司 Yangtze Three Gorges Investment Management and 3.49% through 三峡资本 CTG Capital, giving the group comfortably over half the register.¹ CTG is also CTGR's third-largest supplier, at ¥3.56 billion of procurement in 2025, or 10.99% of the total.¹ CTG affiliates provide financing through 三峡财务 CTG Finance and leasing through 三峡租赁. CTG Capital is a co-investor in CTGR's planned REIT. And CTG has been an open-market buyer of CTGR's shares.

Start with the asset-injection question, because the outline for this story rightly asks it. The honest answer, based on 2025 disclosure, is that the pattern is smaller and less aggressive than the framing implies. The one material same-control acquisition in 2025 was the 65% stake in the Yingkou energy investment vehicle, approved by the board in June 2025 at up to ¥112.21 million and revised in December to up to ¥123.82 million.¹ That is a rounding error against a ¥387 billion balance sheet. Because it was a business combination under common control, Chinese accounting standards required CTGR to restate the prior year as though the entity had always been owned — which is why 2024 net profit appears in the FY2025 report as ¥6.110 billion "restated" against ¥6.111 billion as originally reported.¹ The restatement moved the number by less than ¥2 million.

There is a real analytical point buried in that small transaction, though, and it is not about price. It is about direction. The asset injected was a gas-fired peaking plant, and the reason the board gave was to secure supportive provincial policy in Liaoning.¹ Capital allocation at CTGR is not purely a return-maximising exercise; it is partly an instrument of the parent's and the state's broader system objectives. That does not make it wrong. It does mean that a discounted-cash-flow purist evaluating each project on standalone IRR is using the wrong model for how decisions actually get made here.

Now the harder benchmarking question — is CTGR building offshore wind efficiently? Its own disclosed offshore project budgets of roughly ¥12–14 million per megawatt sit within the range that Chinese offshore developers have reported through this cycle, and the trend within CTGR's own portfolio is downward: Putian Pinghaiwan DE at ¥11.75 million per megawatt is meaningfully cheaper than the Qingzhou projects.¹ But a genuinely rigorous per-megawatt comparison against 中国广核新能源 CGN New Energy, 国家电力投资集团 SPIC, and 中国华能 Huaneng is not possible from public filings, because those peers do not disclose project-level capital budgets on a comparable basis. Anyone claiming a precise cross-company offshore capex ranking from public data is asserting more than the disclosure supports, and this analysis will not do that. What can be said with confidence is that CTGR's offshore leadership is measured in installed megawatts — 16.06% national share — not in demonstrated cost advantage.¹

A 31% Payout, and Why It Is Low

Then there is the dividend, and here the picture is more interesting than the headline.

For 2025, CTGR proposed ¥0.041 per share, a total of ¥1.172 billion — a payout ratio of 31.56% of net profit.¹ Cumulative cash dividends across the three most recent financial years were ¥5.32 billion against average annual net profit of ¥5.67 billion, which the company presents as a 93.89% three-year payout ratio — though that arithmetic compares three years of dividends to one year of average profit, so read it as roughly 31% on a like-for-like basis.¹

Thirty-one percent is a low payout for a utility. European and North American renewable IPPs commonly distribute 60–80% of earnings. The standard defence is that CTGR is a growth company reinvesting for scale, and that defence has merit. But there is a second explanation that the balance sheet makes hard to ignore: much of the "profit" that would fund a higher dividend has never arrived as cash. It sits in that ¥48.25 billion receivable. A company cannot distribute a subsidy it has not been paid. On this reading, the low payout is less a strategic choice than an accounting consequence — and any investor modelling dividend growth at CTGR should be modelling subsidy collection, not earnings.

Worth noting for calibration: Longyuan Power proposed ¥0.1625 per share on 2025 EPS of ¥0.5414 — approximately 30%.¹³ The two largest listed Chinese renewable IPPs are running near-identical payout discipline. This is a sector convention, not a CTGR idiosyncrasy.

The Buyback That Bought the Floor

The most-discussed capital-allocation signal of the past eighteen months was the parent's buyback. On April 9, 2025 — two months after Document 136, with the share price under pressure — CTG announced it would purchase CTGR shares in the open market within twelve months, committing to no less than ¥1.5 billion and no more than ¥3.0 billion.¹⁵ The programme completed on April 8, 2026: CTG had bought 350,766,795 shares for ¥1,500,897,227.54, excluding transaction fees.¹

Read that number carefully. CTG spent ¥1.5009 billion against a stated range whose ceiling was double that. It bought the floor. A controlling shareholder that genuinely believed its subsidiary's shares were dramatically mispriced, and that had CTG's balance sheet, had both the authorisation and the means to spend twice as much. It chose not to. That is not a bearish signal — a ¥1.5 billion purchase is real money and a real vote — but it is a materially weaker signal than the announcement range implied, and it deserves to be read as support rather than conviction.

Two Governance Items Worth Watching

Finally, two governance items that a skeptical investor should have on the radar.

The first is the equity incentive plan, which was effectively unwound during 2025. CTGR had a 2021 restricted share plan. Over the course of the year the company repurchased and cancelled restricted shares from 255 grantees, totalling 32,089,054 shares, in tranches announced in February and July 2025.¹ The income statement carries the consequence: a ¥93.64 million one-off share-based payment charge classified as non-recurring, arising from the cancellation and modification of the equity incentive plan.¹ Executive shareholdings collapsed accordingly — General Manager Liu Zi's holding fell from 370,000 shares to 123,300; Vice President Lü Pengyuan's identically; the pattern repeats across the senior team.¹

Whatever the mechanical cause, the outcome is that the only meaningful ownership link between CTGR's operating managers and CTGR's shareholders was substantially dismantled in the same year the business model changed. That matters for how one assesses everything management says next.

The second is a genuinely new and underappreciated development: capital recycling. On August 6, 2025 the board approved issuing an infrastructure public REIT using the Dalian Zhuanghe III offshore wind project as the underlying asset, with CTGR subscribing 34% of the fund units and CTG Capital 10%.¹ The application was formally accepted by the China Securities Regulatory Commission and the Shanghai Stock Exchange on December 26, 2025.¹

If it completes, this would be the first time CTGR monetises a completed offshore asset and rotates the proceeds into new construction — a genuine alternative to the debt-only funding model that has defined the company since listing. For a business whose central financial weakness is that it cannot self-fund growth, an equity-like recycling channel is arguably more strategically important than any single wind farm. It is also, notably, a related-party transaction with the parent's capital arm on the other side. Investors should watch the pricing when it prices.

---

↑ Back to Top

VII. The Reckoning: Document 136 and the End of Guaranteed Returns (2025)

Eleven Pages That Changed the Model

The document that rewrote CTGR's business model is eleven pages long and reads, as Chinese policy documents do, with almost no drama at all.

On February 9, 2025, the NDRC and the National Energy Administration jointly published 发改价格〔2025〕136号 — the Notice on Deepening Market-Oriented Reform of Renewable Energy On-Grid Pricing to Promote High-Quality Development of Renewable Energy.⁷ Its operative instruction was simple: in principle, all electricity generated by new-energy projects enters the electricity market, and on-grid prices are formed by market transactions.⁶

To soften the transition, the document created a settlement mechanism the regulators call the sustainable development pricing mechanism, and which practitioners shorthand as the mechanism price. It works as a two-way contract for difference: where the market price falls below the established mechanism price, the project is compensated; where it exceeds it, the excess is clawed back.⁶ Projects were divided at a hard cutoff of June 1, 2025. Existing projects that reached grid connection before that date maintain policy continuity through difference settlement against a mechanism price broadly aligned with their previous pricing. New projects connecting on or after that date have their mechanism price determined by regional competitive bidding, with the volumes eligible for the mechanism calibrated against each province's renewable development targets.⁶ Provinces were required to complete implementation by the end of 2025.⁶

That last clause is the one that matters most for the future. For legacy assets, Document 136 is a grandfathering exercise with a wrapper. For everything built from June 2025 onward, the mechanism price is not granted — it is won at auction, against every other developer in the province, in a country that added 120 GW of wind and 317 GW of solar in 2025 alone.¹⁶ Competitive auctions among capital-rich state enterprises bidding for a scarce policy resource have one predictable outcome: the price falls to the level at which the most aggressive bidder is willing to accept a marginal return.

What It Did to the P&L

The 2025 results are the first read on what this does to a real P&L, and they are unambiguous. Revenue fell 4.43% despite generation rising 5.99%.¹ Operating cost rose 17.36% to ¥16.53 billion, because the depreciation on all those newly commissioned assets landed regardless of what price they earned.¹ Gross margin on the power business fell 10.55 percentage points.¹ Operating profit fell 40.42%.¹ Management attributed the decline to three causes: generation below expectations in certain regions due to changing absorption conditions; a year-on-year decline in average on-grid tariff driven by market conditions; and increased impairment provisions taken on a prudential basis against certain equity investments, fixed assets and goodwill.¹

Those impairments are worth isolating. Asset impairment losses were ¥1.42 billion in 2025, more than double the ¥661 million of 2024, on top of ¥541 million of credit impairment losses.¹ Goodwill fell from ¥1.60 billion to ¥1.50 billion.¹ In 2024 the company had already taken ¥1.45 billion of impairment provisions, itself ¥892 million more than 2023.⁵ Two consecutive years of rising write-downs, on assets largely built during the hypergrowth years, is a question about acquisition and development discipline during that period — not proof of poor decisions, but a fact pattern that deserves an explanation management has not yet given in detail.

Did Management Warn Anyone?

Now the management-credibility test, which is where this section earns its keep. Did CTGR warn investors, or did it discover the problem alongside them?

The evidence is mixed, and leans toward partial credit. The FY2024 annual report, published in April 2025, listed 政策调整风险 policy adjustment risk first among its risk factors and named the market-pricing reform explicitly as an accelerant of full marketisation that could depress project returns.⁵ It also listed 电费回收风险 — power tariff collection risk — as its second-ranked risk, tying slow collection directly to balance-sheet pressure.⁵ And the FY2024 profit-decline explanation named rising market-traded volume as the leading cause.⁵ So management was not silent. The direction of travel was disclosed.

What management did not do was quantify it. Nowhere in the FY2024 disclosure is there an estimate of how far realised tariffs might fall, what proportion of the fleet would be exposed by when, or what that would do to earnings. Risk was acknowledged in category; it was never sized.

The FY2025 risk section then does something more troubling. Compare the two lists side by side. The 2024 list ran: policy adjustment, tariff collection, resource acquisition, construction and safety, post-investment management.⁵ The 2025 list runs: policy and market change, marketing risk, absorption risk, resource acquisition, engineering quality and safety.¹ Tariff collection risk — the second-ranked risk in 2024, tied to a receivable that grew from ¥44.76 billion to ¥48.25 billion over the year in question — is no longer a named risk factor at all.¹⁵

There may be a defensible internal rationale. The company has consistently argued that state-credit receivables carry manageable collection risk. But dropping a disclosed risk in the year the underlying exposure grows by ¥3.5 billion is a disclosure choice that an investor is entitled to question, particularly when the same report opens by asserting that no major risk affecting the company's going concern exists.¹

The new risks that replaced it are, at least, the right ones. 市场营销风险 marketing risk — described as intensifying because market trading proportions keep rising and competition keeps sharpening — is a category that would have been meaningless at this company five years ago.¹ Its stated mitigations are concrete: build out the generation-and-retail system, strengthen medium-and-long-term contracts as ballast, coordinate competitive bidding for incremental projects, and invest in marketing information systems.¹ The 2026 operating plan goes further, committing to a marketing benchmarking system, station-by-station bidding and post-mortem review mechanisms, dynamic spot and medium-to-long-term trading strategy optimisation, and — the phrase that captures the whole transition — a shift "from selling electrical energy to green services."¹

That is the language of a company that has understood the problem. Whether it can execute is an entirely separate question, and the answer will not be visible for years.

---

↑ Back to Top

VIII. Current Management: New Leadership Meets the Hard Part

A Handover in the Middle of the Storm

Consider the calendar. Document 136 published on February 9, 2025. On April 7, 2025, Zhang Long (张龙) resigned as director and general manager of CTGR.¹ On April 8, Liu Zi (刘姿) was appointed general manager, joining the board on April 24.¹

Whatever the internal reasoning — and Chinese SOEs rotate senior cadres routinely, so this need not signal anything adverse — the operational reality is that CTGR changed its chief executive within eight weeks of the most consequential policy change in the sector's history. Chairman Zhu Chengjun (朱承军) had himself only taken the role in July 2024.¹ The two people responsible for navigating the transition both arrived immediately before or during it.

Zhu Chengjun's background is CTG's institutional core. He holds a doctorate and the rank of senior engineer, and his career ran through CTG's General Office as deputy director, then head of enterprise management, then party secretary and deputy director of the group's basin hub operations management bureau — the organisation that runs the Three Gorges Dam itself — before serving as chairman and party secretary of 湖北能源 Hubei Energy, CTG's other listed subsidiary.¹ He is, in other words, a systems-and-governance operator from the hydro side of the house, with prior experience running a listed company inside the group.

The Planner Who Got the Job

Liu Zi is a more unusual profile, and it is worth telling properly. She holds a doctorate and the senior rank of professor-level engineer. Before joining CTG, she worked in the United States — as an associate researcher at a Texas engineering research centre's energy systems company, and then as a senior vice president at a US consulting firm.¹ She returned to join CTGR's solar business unit as assistant to the director, rose to deputy director, then moved into planning and development, becoming deputy general manager and then head of that department. She subsequently served as CTGR's chief economist while running the investment and M&A department, then as deputy general manager, chief economist and party committee member — before taking the top operating job.¹

That trajectory is worth dwelling on because of what it is not. She is not a construction engineer promoted for delivering projects on time, which is the archetypal path in a build-out era. She is a planner, an economist, and a dealmaker, with international experience in energy systems consulting. If a board were deliberately selecting for the skill set required to navigate a shift from administered tariffs to market pricing — analytical, commercial, comfortable with valuation and capital allocation rather than pouring concrete — this is close to the profile you would draw up.

That is the most charitable reading, and it is a plausible one. The less charitable reading is that no evidence yet exists either way. Liu Zi has been in the seat since April 2025. Her first full reporting year produced a 39.2% profit decline, and the first quarter of 2026 produced another 57.49% decline.¹⁸ Both are attributable to a policy shock that predates her appointment. Neither tells you anything about her.

The rest of the senior team is also new. Chief Accountant Yang Qinghua joined in December 2024, having come through Yangtze Power's finance department, a secondment to SASAC's revenue management bureau, and CTG International's European operations.¹ Vice President Wang Zhongliang was appointed January 2025 and Zhang Liyi April 2025; board secretary and chief compliance officer Yang Liying took her role in June 2025.¹ Deputy General Manager Lü Pengyuan, who ran CTGR's offshore wind affairs department before becoming a vice president, is one of the few points of continuity on the technical side.¹ Total headcount is 7,427.¹

Pay, Ownership, and the Missing Alignment

The compensation structure is where the incentive analysis gets pointed. Zhu Chengjun's 2025 pre-tax remuneration was ¥1.1238 million. Liu Zi's was ¥1.1650 million. The combined pre-tax remuneration of all directors and senior executives was ¥10.1188 million.¹ Independent directors received ¥180,000 each.¹

By the standards of a company with ¥387 billion of assets and 52 GW of generating capacity, these are modest, salary-driven packages typical of central SOE governance — where senior appointments carry political rank as well as commercial responsibility, and where pay is capped by SASAC policy rather than set by a market. There is no performance-linked equity of consequence: as covered above, the 2021 restricted share plan was repurchased and cancelled during 2025, taking executives' already-small holdings down by roughly two-thirds.¹ Liu Zi's remaining 123,300 shares are, at any plausible price, a trivial fraction of her personal net worth.

The implication is not that these managers will do a bad job. It is that the mechanism by which Western investors normally assume alignment — executives who get rich if the stock works and poor if it does not — is effectively absent. Alignment here runs through a different channel: performance against SASAC and CTG group targets, career progression within the state system, and the parent's own capital discipline. Investors underwriting CTGR are underwriting CTG's institutional judgment, not the personal financial incentives of the executives running the subsidiary. That is neither good nor bad in itself, but it is different, and it should change how you weight management commentary.

One further point on credibility that is testable rather than speculative. Management has been consistent about what it will not do. Since 2024 the disclosed operating plans have emphasised cost control, receivables management, and financing-cost reduction, and the 2025 results delivered on the last of these — financial expenses fell 3.85% to ¥4.14 billion despite a growing debt load.¹ The 2026 plan explicitly invokes 过紧日子 — the state-enterprise idiom for tightening the belt — and commits to comprehensive cost control across all categories.¹ That is a small but real instance of promises matched to outcomes. It is also, so far, the only one available for a team this new.

---

↑ Back to Top

IX. Competitive Landscape & Industry Structure

The National Team

Imagine a war game where every combatant has the same commander. That is a reasonable first approximation of Chinese renewable power generation.

The field is not populated by scrappy independents fighting incumbents. It is populated by the subsidiaries of central state-owned enterprises — what the industry calls the 国家队, the national team. 龙源电力 Longyuan Power, the wind arm of China Energy Investment, was the original Chinese wind pioneer and held the capacity crown for years. 中国大唐 China Datang, 中国华能 China Huaneng, 中国广核新能源 CGN New Energy, and 国家电力投资集团 State Power Investment Corporation — the country's largest utility-scale solar developer — round out the core. CTGR is the newest of them, and in installed capacity terms it is now the largest listed pure-play.

These companies compete, genuinely, for provincial development quotas, for construction contracts, and increasingly for price. But they share ultimate ownership, they buy from the same handful of contractors, they sell to the same two grid companies, and they are all measured against the same state decarbonisation targets. Rivalry here is real but bounded — closer to competing divisions of a conglomerate than to Coca-Cola versus Pepsi.

Five Forces, Badly Balanced

Run Porter's five forces across this structure and the picture is unusually lopsided.

Buyer power is close to absolute. When 98.10% of your revenue comes from five customers and 71.74% from one, you are not negotiating; you are receiving.¹ Under the old regime this did not matter, because the buyer paid an administered price. Under Document 136, the buyer's structural position becomes economically live for the first time. A monopsony that is also the market operator, also the transmission owner, and also the entity deciding which generators get dispatched, is not a counterparty a supplier can push back against.

Supplier power runs the other way, and this is the one force clearly favouring CTGR. Its five largest suppliers accounted for 68.18% of procurement in 2025, led by 中国电力建设集团 PowerChina at 25.35% and 中国能源建设集团 China Energy Engineering at 23.27%.¹ But these are construction contractors, and China's turbine and panel manufacturing base has spent the past three years in brutal, margin-destroying overcapacity. Turbine prices and module prices have both collapsed. A developer buying into that market gets steadily better terms — which is precisely why solar capex per megawatt has fallen and why CTGR could add 2,514.8 MW of solar in 2025 at all.¹ Falling equipment costs are the one genuine tailwind offsetting falling tariffs.

Threat of new entry is low, and the barrier is not technology. Anyone can buy turbines. The barrier is permission and capital: provincial development quotas, sea-area use rights, grid interconnection agreements, and the ability to carry ¥48 billion of unpaid state receivables without going bankrupt. That last item is a moat that nobody discusses as a moat. A private developer with CTGR's receivables profile would have failed years ago. The ability to be the state's involuntary creditor is, perversely, a competitive advantage available only to those the state itself backs.

Substitute threat is negligible in the medium term. The product is decarbonised electricity, and China's 15th Five-Year Plan for the new-type energy system targets wind and solar exceeding 50% of total installed capacity by 2030, non-fossil sources supplying half of all electricity generated, and new energy alone reaching 30% of output.¹⁷ Demand for the product is mandated. What is not mandated is the price.

Which brings us to rivalry — the force that has genuinely changed. Before 2025, competition among the national team was competition for quotas and sites. Price was not a variable, because price was set. Document 136 made price a competitive variable for the first time in the industry's history, and it did so by requiring new projects to bid for their mechanism price against each other.⁷ Every one of these companies has cheap state capital, aggressive volume targets, and management incentives tied to scale rather than to return on capital. That combination, pointed at a competitive auction, has a well-documented outcome across industries: bids get aggressive, and marginal returns compress toward zero.

Seven Powers, Two Survivors

Now the harder question — what durable power does CTGR actually possess? Run Hamilton Helmer's seven powers honestly and most of them fail immediately.

Brand: none. The grid does not pay a premium for Three Gorges electrons. Switching costs: none. Kilowatt-hours are perfectly fungible; the offtaker has zero attachment. Network economies: none — the value of a wind farm does not rise as other people build wind farms. Counter-positioning: none, and structurally impossible, since CTGR is the incumbent, not the insurgent. Process power: unproven; the annual report's disclosed operating model of remote centralised monitoring, unmanned stations and regional maintenance centres is sensible and probably better than a small developer's, but every national-team peer runs a comparable system.¹

Two powers have a real claim.

The first is scale economies, and here the evidence is mixed rather than conclusive. CTGR's ¥48.31 billion of new borrowing in 2025 at a falling blended cost is a real advantage over a subscale developer.¹ Its centralised procurement across a 52 GW fleet plausibly extracts better turbine pricing. But scale economies are only a power if they translate into a cost per unit that rivals cannot match, and here the peer comparison is uncomfortable. Longyuan Power, with 45,994 MW of capacity, generated 76.47 billion kilowatt-hours in 2025 — almost exactly the same output as CTGR's 76.26 billion — and earned ¥4.53 billion of net profit against CTGR's ¥3.71 billion, with a return on equity of 6.08% against CTGR's 4.23%.¹³¹ A smaller fleet produced the same electricity and more profit.

The explanation is mix, not incompetence: Longyuan's fleet is roughly 70% wind, CTGR's is majority solar, and wind simply earns more per installed megawatt in the current pricing environment.¹³¹ But that is precisely the point. If a peer with less scale earns a higher return because it made different asset-mix choices, then scale is not doing the work the bull case says it is doing.

The second claimed power is a cornered resource: China's best offshore wind sites, permitted and built. This one holds up better. A 16.06% national share of offshore wind capacity is a genuine leadership position, and offshore development rights are genuinely scarce — the permitting, marine spatial planning, and provincial coordination required cannot be replicated by capital alone.¹ CTGR added 497 MW of offshore capacity in 2025 against national additions of 6,590 MW, a 7.54% share of new build, which suggests peers are catching up faster than CTGR is extending its lead.¹ China connected 7.192 GW of offshore capacity in 2025 — 78% of the global total — and officials expect more than 15 GW a year through 2030, on a path to roughly 100 GW cumulative.¹⁸ If national offshore capacity roughly doubles by 2030 and CTGR's share of new additions runs at half its share of the installed base, its leadership erodes arithmetically.

Precise per-company offshore rankings among CGN New Energy, Huaneng and SPIC are not consistently disclosed on a comparable basis, so this analysis will not assert one. What the disclosure does support is narrower and more useful: CTGR leads on installed offshore megawatts, that lead is real, and its rate of extension is slowing.

The uncomfortable synthesis is that CTGR's competitive position is strong on access and weak on economics. It can get sites and capital that others cannot. It cannot, on current evidence, convert those into a superior return on capital. In a guaranteed-tariff world, access was sufficient. In an auction world, it may not be.

---

↑ Back to Top

X. Bull vs. Bear: Why It Wins From Here, Why It Might Not

The Bull Case, Stated Fairly

Every investment case eventually reduces to a single question: what would have to be true?

Start with the bull case, stated at its strongest, because it deserves that.

The core claim is that CTGR owns 52 GW of long-lived, zero-fuel-cost generating assets in the world's largest and fastest-growing electricity market, financed at a cost of capital almost no competitor can match, with a controlling shareholder that has both the balance sheet and the demonstrated willingness to support the equity. The physical assets are real, they are producing, and their output grew 5.99% in 2025 while the world was worrying about the price.¹ Operating cash flow grew 10.81% in the same year — a fact easily lost in the profit headline, and one that suggests the assets themselves are functioning exactly as designed.¹ Depreciation, which is the largest single reason profit fell, is a non-cash charge. The bull would argue that a business generating ¥21 billion of operating cash against a ¥28 billion revenue line is not in distress; it is in a period where accounting profit has decoupled from cash generation.

The bull case also rests on three specific mechanisms. First, equipment deflation: every megawatt CTGR builds in 2026 costs materially less than the same megawatt in 2022, which partially offsets the lower tariff. Second, offshore scarcity: deep-water and far-shore sites are a genuinely constrained resource, and CTGR's permitting relationships in Guangdong, Fujian and Jiangsu are not easily replicated. Third, the capital recycling channel — if the Dalian Zhuanghe III REIT completes, CTGR gains an equity-like funding source that breaks the debt-only dependency that has defined it since listing.¹ And a policy floor exists: the mechanism price is explicitly designed as a two-way contract for difference that compensates generators when market prices fall below it, which is not the same as leaving developers fully naked to spot volatility.⁶

The Bear Case, Which Is Already a Description

Now the bear case, which as of today is not a thesis but a description.

The leverage is the starting point. Total debt of ¥187.54 billion against EBITDA of ¥24.90 billion, a current ratio below one, and an asset-liability ratio that has risen every year since listing to 71.91%.⁹¹ Free cash flow has been negative in every recent year — ¥32.11 billion of capex against ¥20.94 billion of operating cash in 2025, and ¥30.89 billion against ¥18.89 billion in 2024.¹ This is a company whose growth depends entirely on continued access to debt and equity markets. Chinese onshore bond markets have been accommodating; if they tighten, or if CTG's own credit profile shifts, the funding cost rises across a very large balance sheet at once.

The receivables are the second pillar of the bear case, and the activist question here writes itself: what is the actual economic value of an asset representing more than a year and a half of revenue, comprising 94.80% unpaid state subsidy, carried at a 4.84% provision, with a collection schedule the company neither sets nor forecasts?¹⁰¹ An activist would push harder still. Why did tariff collection risk disappear from the named risk factors in the very year the balance grew?¹⁴ Why does the report open by stating that no major risk affecting going concern exists, while the auditor simultaneously designates receivable provisioning a key audit matter?¹ These are not accusations. They are questions that a company with a genuinely comfortable position would find easy to answer, and CTGR has not answered them in its filings.

The third pillar is the most important and the least appreciated: the economics of new capacity are structurally worse than the economics of existing capacity. Projects connected before June 1, 2025 were grandfathered into mechanism prices broadly aligned with their prior tariffs. Everything built after that date must win its mechanism price at a provincial auction against every other state developer.⁶ CTGR ended 2025 with 22,634 MW of capacity under construction — nearly half its existing fleet — and the overwhelming majority of that will connect under the new regime.¹ The company is, in other words, adding megawatts whose per-unit economics are almost certainly inferior to the megawatts already online. Growth that dilutes returns is not the same thing as growth.

The impairments compound this. Two consecutive years of rising write-downs — ¥1.45 billion in 2024 and ¥1.42 billion of asset impairment plus ¥541 million of credit impairment in 2025 — invite a genuine question about whether some assets acquired or built during the volume-chasing years were underwritten on tariff assumptions that no longer hold.⁴¹ Management's explanation, that provisions were taken on a prudential basis against certain equity investments, fixed assets and goodwill, is accurate but not illuminating.¹

Fourth, governance and float. CTG group entities control comfortably more than half the shares, the largest single free-float holder outside the group is Hong Kong Securities Clearing at 1.76%, and the parent has been an open-market buyer.¹ Price discovery in a stock where the controlling shareholder is also a buyer is limited, and minority investors have essentially no mechanism to influence capital allocation. The unwinding of the equity incentive plan removed the one structural link between management wealth and share price.¹

Fifth — and this is the item that closes the argument — the damage is not in the past. First-quarter 2026 revenue fell 8.92% to ¥6.95 billion and net profit fell 57.49% to ¥1.04 billion, with management again citing weaker-than-expected generation in certain regions, lower average on-grid tariffs, and reduced disposal gains.⁸ Whatever stabilisation the bull case requires had not appeared by the most recent reporting period.

The Honest Synthesis

Where does that leave the honest synthesis? CTGR's answer to "why do we win" is well-evidenced on two dimensions — access to scarce sites and access to cheap capital — and unproven on the dimension that now matters most, which is converting either into a superior return on capital. Its answer to "why might we not" no longer requires imagination. The tariff reform has landed, the P&L has absorbed the first two years of it, and the capacity being added today will earn less than the capacity added yesterday. An investor's view on this company is, in the end, a view on a single variable: where provincial mechanism-price auctions settle over the next three years. Everything else is second-order.

---

↑ Back to Top

XI. Risk Radar

Not all risks are equal, and the discipline of a risk section is refusing to pretend otherwise. Four risks materially govern CTGR's economics. One commonly cited category does not, and saying so is more useful than padding the list.

Regulatory and pricing risk — highest priority, and already realised. This is not a tail risk; it is the base case. Document 136 required provinces to complete implementation by the end of 2025, and each province designs its own mechanism-price auction, sets its own eligible volumes, and calibrates against its own renewable targets.⁶ That means CTGR faces not one pricing regime but roughly thirty, each capable of moving independently. The company already participates in market trading across twenty-seven provinces and regions.¹ Provincial divergence cuts both ways — a favourable auction outcome in Guangdong does not protect a project in Gansu — and it makes forecasting realised revenue materially harder than it was when a single national tariff schedule applied. The mechanism is still being calibrated, which means the second-order effects have not yet been observed.

Counterparty and receivables risk. The mechanism deserves restating precisely because it is so often described loosely. CTGR sells electricity, books the full price including the subsidy component as revenue, and receives the base tariff from the grid company reasonably promptly. The subsidy portion arrives when the central fiscal budget disburses it, on a timetable the company does not control and does not forecast. Because that portion is booked as revenue when the power flows, reported earnings run ahead of cash earnings by whatever the subsidy accrual is in any given period. The consequence is that headline leverage understates true financial exposure: a company with ¥48.25 billion locked in a non-earning, non-controllable asset has less financial flexibility than its debt ratios alone suggest.¹ The one mitigating dynamic worth noting is that as legacy subsidised projects age out and new projects earn market prices without a national subsidy component, the incremental receivable stops growing. The stock of existing receivables, however, remains.

Curtailment risk (弃风弃光). This is the risk most easily missed by investors reading capacity figures alone, and it has a simple mechanism: a wind farm that is built, connected, and ready to generate can still be told by the grid operator to stop, because the transmission system cannot move the power or local demand cannot absorb it. Those unproduced kilowatt-hours are simply lost — the capital was spent, the depreciation accrues, and no revenue arrives.

National utilisation rates held up reasonably well in 2025: 94.3% for wind and 94.8% for solar.¹⁹ But the national average conceals what matters. Tibet's wind utilisation was 68.6%, Xinjiang's 91%, western Inner Mongolia and Jilin 91.8%, while Shanghai, Fujian and Chongqing achieved 100%.¹⁹ The same geography constrains solar. CTGR's own generation data shows exactly this geography at work: Inner Mongolia is its single largest generating region, and Qinghai, Gansu, Xinjiang and Ningxia together represent a substantial block of its solar fleet.¹ Management named absorption capacity as a distinct risk in the FY2025 report for the first time, and attributed part of the year's revenue shortfall to changing absorption conditions in certain regions.¹ The strategic response — the desert-and-Gobi bases in Kubuqi and southern Xinjiang — deliberately places new capacity in precisely the regions where curtailment is worst, on the assumption that new ultra-high-voltage transmission corridors will arrive to move the power. That assumption is a genuine execution dependency on a third party.

Refinancing and cost-of-capital risk. Given permanently negative free cash flow, CTGR must roll and grow its debt continuously. Its AAA domestic rating carries a stable outlook, and Lianhe's stated downgrade trigger is a clean-energy policy adjustment that significantly weakens operating cash flow.⁹ Operating cash flow has not weakened — it grew in both 2024 and 2025 — which is why the rating held.¹ But the rating explicitly rests on CTG's support as well as CTGR's standalone profile, so the relevant exposure is not only to Chinese credit conditions generally but to any change in CTG's own capacity or willingness to backstop. Neither is currently in question. Both are worth monitoring.

Execution risk on the double transition. The company must simultaneously build roughly 22.6 GW of projects under construction and develop, from a standing start, a commercial capability it has never needed: power marketing, spot and medium-to-long-term trading, contract portfolio management, and price forecasting.¹ Its 2026 plan commits to exactly this — dynamic trading strategy optimisation, station-by-station bidding review, a marketing benchmarking system.¹ But building a trading desk inside a construction-engineering culture is a genuine organisational challenge, and one worth noting: R&D expense fell 53.97% in 2025 to ¥19.58 million, with total R&D investment of ¥554.80 million, 82.25% of it capitalised.¹ A company entering a capability-building transition cut its expensed R&D by more than half.

What does not belong here: AI disruption and cybersecurity are not material threats to the economics of selling electrons into a state grid. CTGR's own 2026 plan treats digitalisation and AI as efficiency tools — smart station upgrades, blade fault diagnostics — rather than as existential variables, and that framing is correct.¹ Forcing them into a risk section would be padding.

---

↑ Back to Top

XII. Durable Lessons, KPIs to Watch & What This Story Teaches

The Error at the Heart of the Story

There is a specific intellectual error that CTGR's story exposes, and it is not unique to China.

The error is treating a policy-created return as though it were an economic one. For fifteen years, Chinese renewable developers earned regulated-utility returns without regulated-utility scrutiny, because the state wanted capacity built and was willing to underwrite the economics to get it. Investors capitalised those cash flows as though they were contractual. They were not. They were a policy instrument, and policy instruments are retired when they have done their job.

By 2025, they had done their job spectacularly. China's wind and solar capacity reached 1.84 billion kilowatts, 47.3% of total generating capacity — surpassing coal for the first time in history.¹ The country added 120 GW of wind and 317 GW of solar in a single year.¹⁶ When an industry has been built to that scale, the subsidy that built it becomes an expensive anachronism, and a government focused on electricity-price competitiveness for its industrial base has every reason to withdraw it. The guarantee was always going to end. The only question was when, and whether investors had priced it.

That is the first durable lesson: when a government both creates and underwrites an industry, the underwriting is the most valuable asset on the balance sheet and the one least likely to appear there. Investors in state-directed utilities anywhere — Indian renewables, European capacity markets, American tax-equity structures — should price the day the guarantee is renegotiated into their base case, not their downside case.

The second lesson is about cash conversion, and it is the more transferable one. CTGR's receivables grew from ¥26.73 billion at end-2022 to ¥48.25 billion at end-2025 while revenue grew roughly a quarter.¹⁰¹ For three of those years, nothing went wrong. Profits were reported, dividends were paid, bonds were issued, the rating held. The gap between accounting earnings and collected cash widened silently, year after year, and was visible to anyone who read the balance sheet alongside the income statement. It only became a story when a separate shock — the tariff reform — forced the whole business model into public view.

That is how balance-sheet risk usually works. It does not announce itself. It compounds quietly in a line item that nobody discusses until something else breaks, and then it turns out to have been the more important number all along.

There is a third, more uncomfortable observation. Between the end of 2021 and the end of 2025, CTGR more than doubled its installed base and its net profit fell by a third. If a company can grow its asset base that dramatically while earning less money, then the metric management and investors were both tracking — megawatts — was never the metric that mattered. Return on the capital deployed to build those megawatts was. Chinese renewable developers spent a decade optimising for a number that turned out to be a proxy for value only as long as the price was fixed.

Three Things Worth Tracking

So what should an investor actually track from here? Three things, and deliberately not more.

First, realised average on-grid tariff. This is the single number that determines whether CTGR's earnings stabilise or keep falling. The company discloses it annually by generation type — wind and solar separately — and the trajectory over 2023 to 2025 has been relentlessly downward.¹⁴ The question for 2026 and 2027 is not whether it falls further but whether the rate of decline decelerates as mechanism-price settlements bed in. A stabilising realised tariff would be the first genuine evidence that the reform's impact is finite.

Second, receivables and their collection. Watch both the absolute balance and its relationship to revenue, and watch the bad-debt provision rate, which has climbed from 3.21% to 4.84% over three years.¹⁰ A balance that stops growing would indicate the subsidy backlog is finally being worked down. A provision rate that keeps rising would indicate management's own confidence in collection is falling — and management's provisioning judgment is the most honest signal available on an asset whose value is otherwise unobservable.

Third, free cash flow and the debt trajectory. The gap between operating cash flow and capex is the mechanical driver of leverage, and leverage is the constraint on everything else. If capex moderates as projects gate on mechanism-price auction outcomes, and operating cash flow holds, this company could approach self-funding for the first time in its existence. That would be a genuine regime change in the business, and a far more meaningful milestone than any capacity target.

The specific disclosure to look for in the FY2026 report is the split between capacity earning grandfathered legacy pricing and capacity earning auction-determined mechanism prices. As the second block grows relative to the first, blended realised tariff falls mechanically regardless of what any individual auction produces. That mix ratio, more than any headline, will determine whether margin compression stabilises or continues.

---

↑ Back to Top

XIII. Epilogue

The Three Gorges Dam took seventeen years to build and cost, by most estimates, more than any single infrastructure project in Chinese history. It was justified not on returns but on purpose: flood control, navigation, and power for a country that needed it. Nobody underwrote it on a discounted cash flow.

Its corporate offspring was underwritten on one. CTGR was sold to public investors in 2021 as a growth utility with guaranteed economics — the rare asset that combined the visibility of regulated infrastructure with the expansion rate of a technology company. For four years the model performed exactly as advertised. The company built a wind-and-solar fleet larger than the dam its parent is named for, then doubled it again.

Today it stands as the largest listed pure-play renewable generator in China, mid-transition between two entirely different businesses. The one it was — a capacity-deployment machine that collected an administered price — has ended. The one it is becoming — a merchant power generator that must forecast, bid, hedge and market its output into thirty provincial markets — has barely started. Its leadership team arrived weeks before the transition began. Its balance sheet carries the accumulated cost of the old model in the form of leverage and unpaid subsidy. Its construction pipeline is nearly half the size of its existing fleet, and most of it will earn prices set by auction rather than by decree.

None of that makes the company a failure. The assets are real, the output is growing, and the cash generation from operations is substantial. What has been destroyed is not the business but the certainty — the specific quality that made this stock feel like something other than an ordinary levered utility.

The broader lesson travels well beyond China's power sector. Governments that want an industry to exist will build it, and the fastest way to build it is to guarantee the economics. But a guarantee is a subsidy, a subsidy is a cost, and every government that has ever used this playbook has eventually reached the point where the industry is large enough that the cost outweighs the purpose. At that moment the same state that created the returns withdraws them, in the name of market discipline, and the companies that optimised entirely for the guaranteed world discover what they are worth without it.

CTGR is finding out now, in public, quarter by quarter. Its parent built a wall across the Yangtze that will stand for a century. Whether the company it spun out can generate an adequate return on capital in a market it never had to compete in is a question that will take considerably less than a century to answer — and the first two years of evidence are already on the table.

References

1. China Three Gorges Renewables (Group) Co., Ltd. 2025 Annual Report — Shanghai Stock Exchange filing, 2026-04-30 ↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩

2. 三峡能源：2025年年报净利润为37.14亿元、同比下降39.20% — Sina Finance / Gelonghui, 2026-04-29 ↩

3. 三峡能源2025年年报解读：扣非净利润下降48.03% — Sohu, 2026 ↩

4. China Three Gorges Renewables (Group) Co., Ltd. 2021 Annual Report — company disclosure ↩↩↩↩↩↩↩↩↩↩↩↩↩

5. China Three Gorges Renewables (Group) Co., Ltd. 2024 Annual Report — company disclosure ↩↩↩↩↩↩↩↩↩↩

6. 国家发展改革委、国家能源局有关负责同志就深化新能源上网电价市场化改革答记者问 — National Development and Reform Commission, 2025-02-09 ↩↩↩↩↩↩↩↩

7. 关于深化新能源上网电价市场化改革促进新能源高质量发展的通知（发改价格〔2025〕136号） — National Development and Reform Commission, 2025-02-09 ↩↩↩

8. 三峡能源(600905.SH)：2026年一季报净利润为10.40亿元、同比下降57.49% — Jiemian News, 2026-04-30 ↩↩↩

9. 中国三峡新能源（集团）股份有限公司2025年跟踪评级报告 — Lianhe Ratings, 2025-06-06 ↩↩↩↩↩↩↩↩

10. 中信证券股份有限公司关于中国三峡新能源（集团）股份有限公司2025年面向专业投资者公开发行公司债券之主承销商核查意见 — Shanghai Stock Exchange, 2025-09-13 ↩↩↩↩↩↩↩↩

11. 鹰眼预警：三峡能源应收账款增速高于营业收入增速 — Sina Finance, 2025-04-30 ↩

12. 三峡能源443.61亿元应收账款引投资者关注 — Solarbe, 2024-09-27 ↩

13. 龙源电力：2025年净利润45.26亿元，同比下降28.78% — Economic Observer, 2026-04-01 ↩↩↩↩↩

14. China Three Gorges Renewables (Group) Co., Ltd. 2022 Annual Report — company disclosure ↩

15. 三峡能源控股股东增持计划公告 — Sina Finance, 2025-04-09 ↩

16. 2025年可再生能源并网运行情况 — National Energy Administration, 2026-02-12 ↩↩

17. 《新型能源体系建设"十五五"规划》发布 2030年初步建成清洁低碳安全高效的新型能源体系 — Xinhua, 2026-06-26 ↩

18. China leads global offshore wind with 78% of new capacity in 2025 — CGTN, 2026-06-18 ↩

19. 2025全国新能源并网消纳情况：全年风电利用率94.3% — Sina Finance, 2026-02-06 ↩↩