Dongfang Electric Corporation Limited (1072.HK) - Deep Dive Research Report

Prepared 1 August 2026. Reporting-cadence note: Dongfang Electric is a China-incorporated state-controlled issuer dual-listed in Hong Kong (1072.HK) and Shanghai (600875.SH). It does not hold Western-style earnings calls with published transcripts. It files quarterly, interim and annual results announcements through HKEX and the SSE. The last six reporting periods (used throughout in place of "concalls") are Q1 2026 (approved 29 April 2026), FY2025 annual (late March 2026), Q3 2025 (late October 2025), H1 2025 interim (late August 2025), Q1 2025 (late April 2025) and FY2024 annual (late March 2025). The most recent released period is Q1 2026; the 2026 interim is due around late August 2026 and had not been released as of this writing.

Section 1: What the Company Does

Dongfang Electric makes the large machines that turn heat, water and wind into electricity, and it builds the power plants those machines go into. If you stand inside a Chinese coal, gas, nuclear or hydro power station, a large share of what you are looking at - the boiler, the steam turbine, the generator, the control system - was very likely designed and cast in one of Dongfang's factories in Sichuan or Zhejiang. It is one of a handful of companies on earth that can supply the entire "power island" of a gigawatt-scale plant from a single group, and one of an even smaller handful that can do it for every fuel type: coal, gas, nuclear, hydro, and wind.

The company was formally established in 1984 in Chengdu, Sichuan, with roots in a 1958 machinery plant, and it grew up as an instrument of China's electrification. For decades its core product was the workhorse of the Chinese grid: the coal-fired steam turbine and generator set. When China was adding the equivalent of a large country's entire generating fleet every few years, Dongfang and its two national peers - Shanghai Electric and Harbin Electric - were the three factories the state relied on to build it. It listed its H-shares in Hong Kong in 1994 and its A-shares in Shanghai in 1995, and completed a full group restructuring in 2007 that folded the manufacturing assets into the listed entity.

The core value proposition is heavy, certified, mission-critical hardware that a customer cannot afford to get wrong. A power turbine spins for 30-plus years, and a failure can black out a province or, in the nuclear case, be a safety event. What Dongfang sells is not a commodity casting but a qualified, reference-backed piece of rotating equipment that a utility and a nuclear safety regulator will both sign off on. That qualification - the metallurgy of a turbine blade that survives 1,600 degree combustion gas, the ability to forge and machine a nuclear-grade reactor coolant pump, the reference list of units already running on the grid - is the moat. It takes years and enormous capital to build, and a new entrant cannot shortcut it.

The best single illustration of what the company is trying to become is its gas turbine programme. For most of its history China could not build a heavy-duty gas turbine on its own; the hot-section metallurgy was a guarded technology held by GE, Siemens and Mitsubishi. After roughly 13 years of internal R&D, Dongfang's F-class 50 MW heavy-duty gas turbine (the "G50") reached full-load operation in 2021, making China the fourth country able to independently develop an F-class machine. In 2025 the same turbine won its first overseas orders. That arc - from importing the most valuable component of a power plant to exporting it - is the story of the modern company: a legacy coal-equipment maker deliberately climbing into the higher-technology, higher-margin, and increasingly clean and export-facing corners of the power business.

Section 2: Business Segments

Dongfang reports across five segments. The cleanest way to hold them in your head is: two big manufacturing engines (clean/efficient thermal-and-nuclear, and renewables), one project-delivery arm that wraps equipment into turnkey plants, one after-market-and-finance arm, and one basket of emerging bets. In FY2025 the group's new effective orders split roughly 67% energy-equipment manufacturing, 22% modern manufacturing services, and 11% emerging industries, which is a fair proxy for where the business's weight sits.

High-Efficient Clean Energy Equipment (coal, gas, nuclear)

This is the historical heart of the company and still its largest single revenue and order contributor. It covers coal-fired boilers, steam turbines and generators; heavy-duty gas turbines and combined-cycle equipment; and the conventional-island and increasingly nuclear-island equipment for nuclear plants (reactor coolant pumps, steam generators, turbine-generator sets).

The core capability here is heavy metallurgy and rotating-machinery engineering at scale, plus the nuclear qualifications that almost nobody else in China holds. Dongfang has consistently described itself as holding the leading domestic market share in both nuclear power and gas power equipment - a claim it repeated in its H1 2025 interim, where it also flagged a breakthrough overseas order for its 50 MW heavy-duty gas turbine. Nuclear is the crown jewel: China is building reactors faster than any other country, each new unit needs a conventional and often a nuclear island's worth of Dongfang steel, and the supplier list is short and regulator-gated. Gas turbines are the technology bet - the G50 programme took China from zero indigenous F-class capability to exporting units.

Strategically management treats this as both the cash engine (coal service and new coal build in China remains real) and a technology-upgrade story (gas and nuclear are the higher-value, less-cyclical growth inside the segment). Its competitors here are Shanghai Electric and Harbin Electric domestically, and GE Vernova, Siemens Energy and Mitsubishi Heavy Industries in gas turbines globally.

Renewable Energy Equipment (wind, hydro, high-end equipment)

This segment makes wind turbines (onshore and offshore), hydropower turbines and generators, and a grab-bag of high-end equipment including power electronics, control systems and chemical/pressure vessels. It is the segment most exposed to China's energy-transition capex and the one generating the company's most eye-catching engineering headlines.

The signature capability is very large rotating machinery at the frontier of scale. Dongfang is one of China's leading hydropower equipment makers, supplying turbine-generator units for the giant cascade projects on the Yangtze and its tributaries; hydro is a technically demanding, high-barrier niche where the reference list matters enormously. In wind, Dongfang installed roughly 8.4 GW globally in 2024 and in September 2025 rolled out and installed a 26 MW offshore wind turbine - a 310-metre rotor, 153-metre blades, standing nearly 200 metres tall - which it describes as the world's largest single-capacity offshore unit, and for which it began taking early orders within weeks.

Why it exists separately: the customer base (wind developers, hydro authorities), the technology (aerodynamics, composites, power electronics) and the economics (more competitive, more cyclical, subsidy- and auction-driven) are all different from the thermal island. Wind in particular is a brutally price-competitive Chinese market; Dongfang is a credible top-tier but not dominant player, sitting behind Goldwind, Envision and Ming Yang on volume. Management frames renewables as the growth engine and the face of the company's clean-energy identity.

Engineering and Supply Chain (EPC and trade)

This segment takes the group's own equipment and wraps it into turnkey power projects (general contracting / EPC) and runs associated trading. Its reason to exist is commercial pull-through: bundling Dongfang boilers, turbines and generators into a single engineered plant, especially for overseas customers who want one accountable contractor rather than a parts list. It is lower-margin and working-capital-heavy (large negative operating cash swings around project milestones are normal, as seen in the Q1 2026 operating cash outflow), but it is a lever to export the manufacturing segments' hardware and to win Belt-and-Road-style international plant contracts. Competitors are the international EPC arms of power-equipment majors and Chinese engineering houses.

Modern Manufacturing Services (power-station service and finance)

This is the after-market and financial-services arm: servicing, upgrading, spare parts and long-term maintenance for the installed base of power stations, plus group financial services. It exists because Dongfang's decades of shipped equipment create a large, sticky installed base that needs parts and overhauls for 30 years, generating recurring, higher-margin, less-cyclical revenue than new-build. It represented roughly 22% of FY2025 new orders. Strategically this is the annuity - the cash cow that smooths the new-build cycle - and management has been explicitly growing its share of the order mix.

Emerging Growth Industries (hydrogen, energy-saving/environmental, power electronics, intelligent manufacturing)

The venture basket: green-hydrogen electrolysers and systems, energy-saving and environmental-protection equipment, power electronics and control, and intelligent manufacturing. It exists to give the company optionality on the next energy wave - most prominently hydrogen, where Dongfang has built electrolyser and integrated green-hydrogen capability. It was roughly 11% of FY2025 new orders and is best understood as a strategic option, not a current profit driver.

Section 3: Products and Business Detail

Thermal island. Coal-fired boilers, ultra-supercritical steam turbines and large synchronous generators remain the volume backbone. These are hundreds-of-tonnes castings and forgings machined to fine tolerances, certified for decades of continuous operation. The process knowledge is in large-ingot forging, blade metallurgy, rotor balancing and the reference base of units already running on the Chinese grid.

Heavy-duty gas turbines. The G50 (F-class, ~50 MW, firing temperature around 1,600 degrees, thermal efficiency above 36%) is the flagship indigenous machine. The hard part is the hot section: single-crystal blade metallurgy and thermal-barrier coatings that survive combustion gas hotter than the melting point of the metal itself. Mastering this put China in a club previously limited to the US, Germany and Japan. The turbine reached full-load operation in 2021 and, in 2025, entered export markets, including a reported North American order for multiple units.

Nuclear equipment. Conventional-island turbine-generator sets and increasingly nuclear-island components (reactor coolant pumps, steam generators). This is the highest-barrier product line: each component is safety-graded and regulator-qualified, and the qualification is effectively non-transferable to a new entrant. Dongfang repeatedly claims leading domestic nuclear market share.

Hydropower. Large Francis, Kaplan and pumped-storage turbine-generator units for China's giant river-cascade projects. Pumped storage is a growing sub-line because it is the cheapest grid-scale storage complement to intermittent wind and solar.

Wind. A full onshore and offshore turbine range topped by the 26 MW offshore machine (310 m rotor, 153 m blades) unveiled and installed for testing in September 2025 - marketed as the world's largest single-unit offshore turbine. Offshore is where Dongfang's heavy-engineering DNA gives it an edge over lighter onshore-focused rivals.

Hydrogen and emerging. Alkaline and PEM electrolysers, integrated green-hydrogen systems, environmental-protection equipment, power electronics and control systems.

Manufacturing and geography. The manufacturing base is concentrated in Sichuan (Deyang / Chengdu, Dongfang Turbine, Dongfang Electric Machinery) and Zhejiang (Dongfang Electric Wind Power, Hangzhou), with additional specialist plants across China. Exports run through the Engineering & Supply Chain arm into Belt-and-Road markets and, increasingly, into higher-end markets as the gas-turbine North American order and the 50 MW overseas breakthrough show. Milestone markers that reshaped the business: HK listing 1994; 2007 group consolidation; 2021 indigenous F-class gas turbine full-load run; 2025 26 MW offshore turbine and first heavy-duty-gas-turbine exports.

Section 4: Customers

The customers are, overwhelmingly, large power producers and the state. The buyers are China's "big five" power generation groups (Huaneng, Datang, Huadian, State Power Investment, China Energy) and their peers, the national and provincial nuclear operators (CNNC, CGN, SPIC), the large hydro and grid authorities, wind and offshore-wind developers, and - through the EPC arm - overseas utilities and governments. Many of these are, like Dongfang, central state-owned enterprises, so the customer relationship is partly commercial and partly a matter of national industrial policy.

The buying decision sits with engineering procurement teams at the utility, but for nuclear and for frontier machines it is effectively co-decided by safety regulators and by central planning. The criteria are: proven references on the grid, safety and regulatory qualification, ability to deliver a full island on schedule, price, and increasingly indigenous-content credentials (buying a Chinese-made gas turbine instead of an imported GE/Siemens unit is itself a policy goal). Sales cycles are long - months to years from tender to contract, then years of manufacturing and commissioning - which is why the reported "new effective orders" figure and backlog matter more than any single quarter's revenue.

Switching costs are high and structural. A turbine is a 30-year installed asset; once a plant runs Dongfang equipment, the spares, overhauls and upgrades flow back to Dongfang for decades (that is the entire logic of the Modern Manufacturing Services segment). For nuclear, requalifying a new supplier is a multi-year regulatory exercise almost no operator will undertake casually. Concentration is real but is better read as a reflection of an oligopolistic national market than a company-specific fragility: the same handful of state generators buy from the same three domestic equipment makers, and the state has an interest in keeping all three viable.

Contract structure is milestone-based project revenue: large orders booked into backlog, recognised as manufacturing and commissioning milestones are hit, which produces lumpy revenue and the large negative working-capital swings visible around project cycles (the Q1 2026 operating cash outflow of roughly RMB 2.7 billion is normal seasonality for this model, not distress). The service and finance segment provides the steadier, recurring counterweight.

Section 5: Competitive Landscape

The structure is an oligopoly at home and a challenger-versus-incumbents dynamic abroad, and it differs sharply by segment.

In thermal, nuclear and hydro islands, the domestic market is a three-way division of the state's power-equipment needs among Dongfang, Shanghai Electric and Harbin Electric. This has been stable for decades; the state has no interest in one player collapsing and effectively rations big national programmes among them. Dongfang wins where it has the strongest technology reference - it claims domestic leadership in nuclear and gas equipment - and shares the rest. Barriers to entry here are close to prohibitive: the metallurgy, the qualification, the reference base and the capital intensity mean no fourth domestic entrant has emerged in a generation.

In heavy-duty gas turbines, the incumbents are the Western/Japanese majors - GE Vernova, Siemens Energy, Mitsubishi Heavy Industries - who have dominated the hot-section technology for decades. Dongfang is the domestic-substitution challenger: it wins on price, on Chinese industrial-policy preference, and now on a credible indigenous F-class product, but it is still climbing the reliability-reference curve that the incumbents built over decades of fleet hours.

In wind, the picture inverts: Dongfang is a strong but not dominant player in a hyper-competitive Chinese market led on volume by Goldwind, Envision and Ming Yang. Margins here are thin and pricing is punishing; Dongfang differentiates on offshore and giant-turbine engineering (the 26 MW unit) rather than on being the cheapest onshore supplier.

(Market caps are approximate peer-size references only, move daily, and are stated as of roughly mid-2026.)

Where Dongfang is strong: nuclear and gas-turbine domestic leadership, offshore-wind and giant-turbine engineering, and a full-island capability few can match. Where it is exposed: wind margins are commoditised and competitive, thermal new-build is structurally mature in China, and in export gas turbines it is still the challenger to entrenched, fleet-proven incumbents.

Section 6: Industry

Demand is driven by one thing above all: the build-out and decarbonisation of China's electricity system, plus a growing slice of overseas power investment. China adds more generating capacity per year than any other country - well over 100 GW of new wind alone in recent years - and is simultaneously the world's most active nuclear builder and a large hydro and pumped-storage investor. Every one of those plants needs the heavy equipment Dongfang makes.

The industry is large and, unusually for capital goods, growing, because the energy transition adds new demand (wind, hydro, storage, hydrogen, nuclear) on top of a still-real base of thermal replacement and gas peaking. The wind-turbine market globally runs into the hundreds of billions of dollars over the coming decade, and China is the dominant manufacturing base within it. Nuclear is a multi-decade tailwind given China's reactor-build pipeline, and gas is growing as a flexible complement to intermittent renewables.

In the global supply chain Dongfang sits as a vertically integrated original-equipment manufacturer and EPC contractor - it forges, casts, machines, assembles and installs, and it increasingly exports both components and whole plants. The import-substitution dynamic is central to the investment thesis: the highest-value component, the heavy-duty gas turbine hot section, was historically imported from GE, Siemens and Mitsubishi, and China's national goal is to localise it. Dongfang's G50 is that localisation, and its move into exports flips the historic import flow.

Regulation is pervasive and mostly a barrier that protects incumbents: nuclear safety qualification, grid interconnection standards, and industrial-policy preference for indigenous content all favour the established domestic three. Cyclicality is meaningful but muted by policy - power capex is tied to five-year plans and state priorities rather than pure market cycles, and the growing service/annuity and multi-fuel mix dampen the swings of any single technology. Tailwinds: nuclear pipeline, offshore wind, gas peaking, pumped storage, export policy, hydrogen optionality. Headwinds: thermal maturity in China, brutal wind pricing, and the working-capital intensity of the whole model.

Section 7: Growth Triggers

(Sourced from the last six results announcements. Dongfang publishes results releases and results-evaluation/action-plan documents rather than earnings-call transcripts, so triggers are drawn from those disclosures.)

• 50 MW heavy-duty gas turbine (G50) overseas expansion. The H1 2025 interim flagged a "breakthrough in overseas orders for 50 MW heavy-duty gas turbines," and subsequent 2025 disclosures pointed to entry into the high-end North American market. (H1 2025 interim, ~Aug 2025; repeated in FY2025 annual, ~Mar 2026.)

  Management highlighted the market share of nuclear power and gas power remaining the industry leader, "with a breakthrough in overseas orders for 50 MW heavy-duty gas turbines." (H1 2025 interim.)

• 26 MW offshore wind turbine commercialisation. The world's-largest single-capacity offshore unit was rolled out and installed for testing (Sept 2025) and began attracting early orders; a ramp from prototype to serial orders is the forward driver. (Q3 2025, ~Oct 2025; FY2025 annual, ~Mar 2026.)
• Nuclear and gas market-share leadership feeding order intake. Management repeatedly states leading domestic share in nuclear and gas equipment, framing China's reactor and gas-peaking pipeline as a continuing order source. (H1 2025 interim; FY2025 annual; repeated theme.)
• Record new-order intake building backlog. FY2025 new effective orders rose to roughly RMB 117bn (mid-teens growth), following ~RMB 101bn in FY2024, and Q1 2026 booked ~RMB 37bn with ~70% from energy-equipment manufacturing - management frames the rising backlog as forward revenue coverage. (FY2025 annual, ~Mar 2026; Q1 2026, ~29 Apr 2026.)
• Higher-quality hydropower orders. The H1 2025 interim noted the quality (margin/scale) of hydropower orders improved significantly, pointing to the giant cascade and pumped-storage pipeline. (H1 2025 interim, ~Aug 2025.)
• Shift toward services and emerging industries in the order mix. FY2025 orders were ~22% manufacturing services and ~11% emerging industries, and management's 2026 action plan emphasises growing the recurring-service and new-energy (hydrogen) share. (FY2025 annual and 2025 evaluation / 2026 action plan, ~Mar-Apr 2026.)
• Margin and returns focus. Q1 2026 showed gross margin expansion (+0.45 pts y/y) and a sharp profit rise, which management ties to a "quality, efficiency, returns" programme flagged in the 2026 action plan. (Q1 2026, ~29 Apr 2026; 2026 action plan, ~Apr 2026.)

Section 8: Key Risks

Thermal maturity and order-mix rollover. The historical cash engine - coal-fired equipment in China - is structurally mature. If new coal build and even coal-service demand fade faster than gas, nuclear, hydro and wind can fill the gap, the volume base erodes. The mechanism is slow and visible in the order mix rather than sudden, making it a high-probability moderate drag rather than a cliff. Management's own pivot toward services, gas, nuclear and emerging industries is a direct response.

Wind margin compression. Chinese wind is a price war. Dongfang competes against Goldwind, Envision and Ming Yang in a market where per-MW pricing has fallen for years. Even record MW volumes can produce thin or negative segment economics. This is a high-probability, moderate risk; the company's defence is to fight on offshore and giant-turbine engineering rather than onshore commodity pricing.

Working-capital and cash-conversion intensity. The EPC-and-milestone model ties up large working capital; Q1 2026 posted a roughly RMB 2.7 billion operating cash outflow. In a normal year this reverses as milestones are hit, but a large project delay, a customer payment problem, or a downturn in one of the state generators could turn a seasonal swing into a genuine liquidity strain. Moderate probability, moderate severity, and worth monitoring across quarters rather than judging on any single one.

Gas-turbine reliability and reference risk. The G50 export story is the highest-upside trigger and also a concentrated technology risk. Heavy-duty gas turbines are judged on decades of fleet hours; an indigenous machine still building its reliability record is one high-profile field failure away from a reputational setback in exactly the high-end markets (North America) it is trying to crack. Low probability, high consequence for the growth narrative.

Policy and state-ownership dependence. As a central SOE selling mostly to other SOEs under five-year-plan capex, Dongfang's demand, and to a degree its margins, are policy-set. A shift in national energy priorities, a slowdown in nuclear approvals, or the state directing a big programme to a peer can move the order book independent of Dongfang's own execution. This cuts both ways - policy is also the tailwind - but it means the company does not fully control its own demand.

Geopolitics on exports. The push into North American and other developed gas-turbine and equipment markets runs straight into the trade and technology-security frictions facing Chinese industrial exporters. Tariffs, procurement bans or security reviews could close the very high-end export doors the growth story depends on. Rising probability, and largely outside management's control.

Section 9: Walk the Talk

The six periods assessed: FY2024 annual (~Mar 2025), Q1 2025 (~Apr 2025), H1 2025 interim (~Aug 2025), Q3 2025 (~Oct 2025), FY2025 annual (~Mar 2026), and Q1 2026 (~29 Apr 2026). The most recent, Q1 2026, was approved on 29 April 2026, roughly 94 days before this writing, which is the freshest released period given the quarterly cadence (the 2026 interim is due late August). Because Dongfang publishes results announcements rather than guidance-heavy earnings calls, "walking the talk" is judged against its stated operational priorities - growing orders, shifting mix toward higher-value nuclear/gas/service, indigenising and exporting the gas turbine, and improving returns - rather than against numeric revenue guidance it does not give.

The FY2024 annual was a genuinely mixed report and management did not hide it: new orders grew strongly (up ~17% to roughly RMB 101 billion) even as net profit fell ~18%. The honest read is that they delivered on the thing they most control - order intake and market position - while profitability was squeezed. What they promised into 2025 was continued order growth, a mix shift toward higher-value equipment, and progress on the gas turbine and returns.

Across 2025 they largely delivered on the order and mix promises. H1 2025 showed revenue and profit both growing again (profit up ~13%), and crucially the specific claims were concrete and checkable: leading nuclear and gas share, improved-quality hydro orders, and a genuine overseas breakthrough for the 50 MW gas turbine. That gas-turbine claim was not vapour - by late 2025 the G50 had entered export markets including North America, exactly as flagged. The 26 MW offshore turbine, teased as a technology ambition, was physically rolled out and installed in September 2025 and began drawing orders within weeks. When Dongfang says a frontier machine is coming, it has tended to actually ship metal.

"The market share of nuclear power and gas power remains the industry leader, with a breakthrough in overseas orders for 50 MW heavy-duty gas turbines." (H1 2025 interim.)

The FY2025 annual is where the "returns" promise landed. After the disappointing FY2024 profit, FY2025 net profit rose ~31% on ~13% revenue growth, orders hit a record ~RMB 117 billion, and the board raised the dividend. The 2025-evaluation / 2026-action-plan document explicitly foregrounded "quality, efficiency, returns and investor commitment," and Q1 2026 then opened with gross-margin expansion and a ~37% profit jump. So the arc reads as: acknowledged a weak 2024, told shareholders they would prioritise margin and returns, and then produced a margin-up, profit-up, dividend-up year and a strong opening quarter.

The one honest caveat is that a meaningful part of Q1 2026's profit jump came from fair-value gains on equity holdings (~RMB 325 million) rather than pure operations, so the returns improvement is partly market-aided. And FY2024 is a reminder that this is a lumpy, cyclical business where profit can fall even as the franchise strengthens. Net assessment: management is specific, ships what it says it will build, does not paper over bad years, and delivered on the orders-and-mix-and-returns story it laid out. This reads as a credible, execution-focused management that under-promises on numbers (it barely gives forward numbers at all) and lets delivered orders and shipped hardware do the talking, rather than a serial over-promiser.

Section 10: Shareholder Friendliness Index

Dividends. Dongfang pays an annual final dividend and the three-year record is a dip-then-recovery, tracking earnings. The final dividend was RMB 4.75 per 10 shares for FY2023, cut to RMB 4.03 per 10 shares for FY2024, then raised to RMB 5.30 per 10 shares for FY2025 (paid to H-shareholders at roughly HK$6.07 per 10 shares). The FY2024 cut was not a policy retreat but a mechanical follow-through of that year's ~18% profit decline; the FY2025 increase to a new three-year high tracked the ~31% profit rebound. The company frames dividends within a stated "returns and investor commitment" plan, and the payout has stayed at a moderate, earnings-linked level rather than being stretched.

Buybacks and dilution. There is no meaningful open-market buyback-and-cancel programme. The only repurchases in the last three years were tiny technical cancellations tied to the Restricted A-Share Incentive Scheme - about 17,334 restricted A-shares bought back and cancelled (roughly RMB 98,000) after a participant was disqualified. Far more consequential in the other direction: in 2025 the company completed a private placement of about 272.88 million A-shares and 68 million H-shares, raising roughly RMB 4.12 billion net. So the share count grew, not shrank - this is a company that raised equity capital to fund its expansion, not one retiring stock. MoatMap's trailing-~90-day buyback feed recorded zero repurchases (window since 3 May 2026), consistent with the no-buyback picture, and the multi-year check confirms none beyond the trivial incentive-scheme cancellation.

Verdict: Neutral - it pays a moderate, earnings-linked and now-rising dividend, but it issues equity to grow rather than buying back stock, so net capital return is modest and share count is rising.

Section 11: Insider Activities

(Source: MoatMap cross-market disclosure database, market HK, current to 2026-08-01. The HKEX Disclosure-of-Interests portal is API/bot-gated, so this block is the canonical source for recent dealings. Important context: every "insider" below is a substantial-shareholder financial institution - custodian banks, prime brokers and asset managers crossing the 5% / 1% HKEX disclosure thresholds on the H-share line - not a director or officer of Dongfang. There were no director/officer open-market dealings in the window. These are position moves by global banks holding H-shares in custody and prime-brokerage books, so they carry far weaker signal than an executive buying with personal money.)

Buys - reading the signal. Over the trailing 12 months MoatMap records 5 buys, 4 sells and 25 "other" (grants/deemed-interest/custody) across 7 institutions, with a net buying market-trade direction. The largest clean purchase was Morgan Stanley buying ~5.3m shares for ~HK$120m on 3 July 2026, and JPMorgan and Goldman both accumulated in June. But these are institutional trading and custody flows, not insider conviction: prime brokers and custodians cross HKEX disclosure thresholds constantly as they lend, borrow and hold H-shares for clients. Artemis both bought (18 June) and sold (11 June) within a week, which is characteristic of index/position management rather than a directional bet. None of these is a director or officer of the company buying with personal capital, so none warrants a "very bullish signal" flag.

Sells - the why. The material sells are JPMorgan trimming its H-share prime-brokerage book (~2.0m on 6 July, plus smaller lots) and Artemis reducing ~3.75m on 11 June. No filing footnote gives a company-specific reason, and none is inferable beyond ordinary portfolio and financing turnover; treat these as position management, reason not disclosed.

Net assessment. Activity is entirely among substantial-shareholder financial institutions, is broad-based across seven global banks and asset managers, and is directionally net buying by count. Critically, there is no director or officer dealing in the window - the people who run Dongfang neither bought nor sold on record. The institutional flow is a mildly positive-to-neutral tape (global banks net accumulating H-shares) but it carries little insider-conviction weight. Read: neutral, with a faint positive tilt from institutional accumulation, and no genuine management-level buy or sell signal either way.

Section 12: Scenarios

Bull case. The gas-turbine gamble pays off in full. The G50 builds a clean reliability record, the North American and other export orders convert into a repeatable franchise, and China's push to localise heavy-duty gas turbines makes Dongfang the national champion for the most valuable piece of a power plant. At the same time the 26 MW offshore turbine moves from prototype to a serial order book, giving Dongfang a differentiated, high-margin perch above the onshore wind price war, while China's reactor pipeline keeps the nuclear order machine humming and pumped-storage hydro adds a second growth leg. The service and emerging-industry mix keeps climbing, smoothing the cycle and lifting group margins, so the returns-and-quality programme management promised turns into a multi-year margin expansion. In this world Dongfang stops being a cyclical thermal SOE and becomes a diversified, partly-export, higher-technology energy-equipment group whose record order backlog converts into steadily rising, less-lumpy profit.

Base case. Management roughly delivers what it has been delivering: mid-teens order growth, a slow but real mix shift toward nuclear, gas, hydro, offshore wind and services, and gradual margin improvement off the FY2024 trough. The gas turbine wins some overseas orders but remains a promising challenger rather than a dominant exporter; the 26 MW turbine ships in modest volumes; wind stays competitive and thin-margined but is offset by nuclear, hydro and the growing service annuity. Dividends keep tracking earnings, drifting up in good years, and the company continues to fund expansion partly with equity rather than returning large capital. Profit grows but stays lumpy, with the occasional soft year like FY2024, and the stock's fortunes ride China's energy-capex cycle and policy priorities. A solid, state-backed, cyclical compounder doing what it says.

Bear case. The pivot stalls where it hurts most. Chinese thermal build fades faster than gas, nuclear and renewables can replace it, and the order mix rolls over. Wind's price war deepens and drags the renewables segment into losses even at record MW. A high-profile field problem with the young indigenous gas turbine damages its reliability reputation just as it tries to sell into demanding export markets, and geopolitics - tariffs, procurement bans, security reviews - closes the North American and developed-market doors the growth story needed. Working capital, already heavy, turns against the company in a project downturn or a state-customer payment squeeze, and the equity-funded expansion dilutes shareholders without the profit to justify it. Policy, which giveth, taketh away: a national programme is steered to a peer, nuclear approvals slow, and Dongfang is left as a lower-growth, capital-hungry thermal legacy with a wind albatross and an export dream deferred.