Sieyuan Electric: A Premium-Priced Export Compounder With Zero Margin of Safety

Meta

• Ticker: 002028.SHE
• Company: Sieyuan Electric Co., Ltd. 思源电气股份有限公司
• Price & market cap: CNY 151.73 close as of 2026-07-10; market cap about CNY 118.74 billion as of 2026-07-10, using 782.57 million shares outstanding.
• Currency: CNY
• Report date: 2026-07-11
• Industry: Power equipment
• One-line positioning: Private-sector transmission and distribution equipment maker with a fast-rising overseas business that reached 26.94% of 2025 revenue.

Research summary

Sieyuan Electric is a private-sector power-equipment manufacturer, not a generic “grid capex” stock. It has spent three decades assembling a product set broad enough to sell a substation rather than a single device. The money still comes mainly from hard equipment, not from software dreams or capital-markets storytelling. In 2025, switchgear was the largest revenue line at CNY 8.94 billion, or 41.5% of sales; transformers were second at CNY 4.63 billion, or 21.5%; protection and automation, power electronics, EPC, and storage made up the rest. That composition matters. It means the company’s growth is being driven by a combination of mature, large-ticket equipment categories and a few faster-growing adjacencies, not by one speculative “second curve.” EPC and storage grew fastest in 2025, but the core earnings engine still sat in switchgear and transformer families, with overseas demand increasingly pulling both.

The market is mainly trading three narratives at once. The first is the global grid-investment upcycle: aging overseas grids, renewable integration, electrification, and data-center load growth are all increasing demand for high-voltage equipment. The second is Sieyuan’s own execution record: 2025 revenue grew 39.3% to CNY 21.54 billion and attributable net profit rose 53.7% to CNY 3.15 billion, while 2026 first-quarter revenue and net profit still grew 41.6% and 23.2% respectively. The third is the idea that Sieyuan is becoming the rare Chinese grid-equipment name with meaningful overseas diversification, not merely export exposure in name only. Overseas revenue rose to CNY 5.80 billion in 2025, 26.9% of sales, and 2025H1 overseas revenue had already reached CNY 2.86 billion, up 89.0% year on year and 33.7% of total revenue. Those are not token numbers. They are large enough to change the mix of customers, the pace of orders, and eventually the quality of margins.

The most important fact-check in this report is that the starting ownership hypothesis was wrong in an important way. Sieyuan is certainly not State Grid-controlled, but the latest primary disclosures do not support the idea of a dispersed company with no controlling owner. The 2025 annual-report summary and 2025H1 report show Hong Kong Securities Clearing as the largest registered holder, but that is a nominee account; among identifiable holders, chairman and general manager Dong Zengping held 16.81% at year-end 2025, with co-founder Chen Bangdong at 12.32%. More importantly, the 2025H1 report states that the controlling shareholder and actual controller did not change during the period, and a 2026 company filing explicitly refers to Dong Zengping as the actual controller. The correct framing is not “widely held” but “founder-controlled private champion with a relatively low outright equity stake but effective control.” That changes how one should think about governance, alignment, and succession risk.

The share price rerated hard over the last two years because the business stopped looking like a solid domestic equipment supplier and started looking like a rare high-growth industrial exporter. The company’s own disclosures show overseas revenue up 85.8% in 2025 and management’s July 2026 investor-relations record says North America is already a real market for high-voltage transformers, dead-tank circuit breakers, and disconnectors that can serve renewable interconnection and AI data-center load access. That is the kind of detail that turns a macro theme into a stock-specific narrative. It also helps explain why Sieyuan trades on a much richer multiple than Xuji Electric, Pinggao Electric, and NARI Technology. On 2026-07-10, the stock traded at roughly 37.6x trailing earnings, versus about 19.3x for Xuji, 19.0x for Pinggao, and 21.3x for NARI. The premium is paying for execution and international optionality.

The core bull-bear disagreement is not whether demand is good. Demand is good. Management’s own 2026 targets call for new contract orders of CNY 37.5 billion ex tax, up 30%, and revenue of CNY 27.0 billion, up 25%; 2026H1 preliminary revenue still grew 27.1%. The real disagreement is whether this growth deserves a structurally higher valuation multiple than the rest of Chinese grid equipment. Bulls say yes, because Sieyuan has built a broader product stack, better private-sector execution, and faster overseas penetration than the state-owned peers. Bears say the stock already discounts several more years of near-perfect delivery while underestimating three frictions: working-capital intensity, foreign-exchange noise, and the fact that some of the overseas uplift now comes with higher minority leakage after Sieyuan’s stake in the important Changzhou Sieyuan-Toshiba transformer subsidiary fell from 90% to 70% in July 2026. The 2026H1 quick report explicitly says the reduced stake lowered attributable profit, even as the business still grew.

That minority-leakage point is easy to miss because the operating story is so strong. In January 2026, management said the new transformer workshop and equipment were still being installed, commissioned, and rolled into batch production, but that customer deliveries were broadly on track. By November 2025, Changzhou local-government reporting said the second high-voltage plant at Changzhou Sieyuan-Toshiba had entered operation and, after full ramp, would be able to produce more than 400 intelligent high-voltage transformers per year. This capacity expansion is one reason the transformer business is no longer just an add-on to switchgear. It is becoming one of the central pieces of Sieyuan’s export machine. But it is also why the market now expects a long runway of growth rather than one strong year.

Sieyuan, at this point in its life cycle, is neither a cyclical reversal nor a mature cash cow. The best label is high-quality re-rating within a genuine growth phase: a company that already had a real business, then found a much larger market for it, and is now being priced as though the next leg will be even cleaner than the last. That description has two consequences. First, the company quality is real. Founding history, product breadth, customer reach, and multi-year delivery record support that. Second, the stock is vulnerable to being a good company at the wrong price. On my estimates, the current price still embeds a demanding continuation case. This is not a fraud-risk story, nor a balance-sheet accident waiting to happen, but a valuation-discipline story.

Company vertical history

Sieyuan was born in Shanghai in December 1993 as Shanghai Sieyuan Electric Co., Ltd., founded by Dong Zengping, Chen Bangdong, and Li Xia with initial cash contributions and a tiny registered capital of RMB 300,000. The company’s origin story matters because it explains why it still behaves differently from the large state-linked equipment groups. It was a founder-built electrical-equipment company, not a carve-out from a ministry system or an SOE reform artifact, and it expanded by product line, engineering capability, and acquisitions, then later plugged itself into the huge capex flows of China’s power grid and, eventually, the overseas grid cycle. The 2025H1 report still traces the company back to that original 1993 formation, and the 2025 annual-report materials describe Dong as one of the founders who has worked at the company full time since inception.

Its basic listing path was straightforward. The company was restructured into a joint-stock company in December 2000 and listed on the Shenzhen Stock Exchange on 2004-08-05. I did not retrieve the original IPO pricing and proceeds from a primary listing document in this pass, which is one of the report’s declared blind spots, but the listing date and path themselves are clear in the company’s filings. What investors first got was a domestic high-voltage and substation-equipment story. What they own now is a broader power-equipment platform with export, EPC, storage, and auto-electronics options layered on top.

The company’s history divides cleanly into four stages. The first was product validation and early domestic expansion. Older annual reports describe a business centered on high-voltage switchgear, capacitors, instrument transformers, grounding devices, and related T&D components. That early structure established the company’s pattern of building specialist subsidiaries around product lines rather than forcing everything into one plant or one sales channel. It also built long-term relationships with domestic grid customers. By 2018, switchgear was already the largest product bucket at CNY 2.08 billion of revenue, line products were at CNY 0.97 billion, reactive compensation at CNY 0.83 billion, and smart-equipment revenue at CNY 0.46 billion. Even then, overseas revenue had reached CNY 0.84 billion and was growing much faster than domestic sales. The seeds of the later export story were already visible.

The second stage was product-system completion. The most important move here was the 2018 acquisition of 90% of Changzhou Toshiba Shudian Transformer, later renamed Changzhou Sieyuan Toshiba Transformer. That deal did two things at once. It filled a strategic product gap in transformers and reactors, and it gave Sieyuan access to a brand, technology, and customer-acceptance bridge that would later matter in overseas markets. Analyst language sometimes overstates this as if one acquisition changed everything overnight. It did not. The more accurate reading is that the transformer deal completed a one-stop substation and export toolkit. In a 2026 draft prospectus, the company describes itself as able to supply and integrate core equipment in EPC projects and to serve grid, industrial, and renewable customers across more than 100 countries and regions. That claim would have been weaker without the transformer platform.

The third stage was the move from domestic equipment vendor to international challenger. The company’s 2025 draft prospectus says sales had expanded to more than 100 countries and regions by September 30, 2025, with 20-plus overseas subsidiaries serving local customers. It also says the company’s overseas revenue grew 44.7% in 2024 and 72.7% in the first nine months of 2025, and that overseas growth had become one of the core engines of financial performance. That is the real turning point in the equity story. Investors stopped seeing Sieyuan as just another domestic electrical-equipment supplier exposed to Chinese grid tender timing. They started seeing a multi-continent exporter with real entry into developed markets. The July 2026 company IR record adds important color: North America is no longer a vague aspiration. Sieyuan says it is already selling high-voltage transformers, dead-tank circuit breakers, and disconnectors there, and it is explicitly tracking AI data-center demand as a pull factor for power-equipment deployment.

The fourth stage is the current one: capacity-led scaling under a much higher capital-markets expectation set. In 2025, revenue rose 39.3% and attributable net income 53.7%. The main product lines all moved, but at different speeds. Switchgear grew 29.3%, transformers 38.4%, power electronics 49.5%, EPC 130.9%, and storage systems and components 120.5%. The trap is to treat all of those lines as equally important to normalized future profit. They are not. EPC and storage are clearly growing from smaller bases; they matter, but they are not yet the same quality of profit pool as switchgear and transformers. Management itself sounds careful on newer areas. In July 2026, the company said solid-state transformer products were still only in early-stage research and feasibility evaluation, not yet in formal R&D project approval. That restraint is encouraging. It suggests management is not yet trying to sell every frontier-power buzzword as current earnings power.

A few key nodes still shape the present. One is the 2018 transformer acquisition, already discussed. Another is the accumulated overseas-market build-out, which management describes as a long-term strategy rather than a short-term order hit: the July 2026 IR record says the company continues to invest in overseas products and markets, while the prospectus says customers now include overseas grid operators, generators, and industrial users across more than 100 countries. A third node is the supercapacitor push, which has gone from a goodwill headache into a plausible but still unproven growth option. In May 2026, management said CNY 100 million of goodwill impairment had been taken in each of 2023 and 2024 for the controlled supercapacitor asset, but no impairment was recognized in 2025 because product progress had improved in new power systems and data centers. That is not proof of a durable profit engine so much as proof that the asset stopped deteriorating. The difference matters.

The latest node is more subtle: Toshiba’s contractual repurchase right over part of the Changzhou transformer JV was exercised, and by July 2026 Sieyuan had completed registration of the ownership change, reducing its stake from 90% to 70%. That does not change the industrial logic of the asset, but it does change the economics at the listed-company level. The 2026H1 quick report states directly that attributable net profit is now calculated using a 70% ownership share. Investors who only look at revenue and order growth will miss that some operating success is no longer fully flowing through to the parent.

Financial vertical review

The long financial arc is one of acceleration, not mere scale. Revenue was CNY 4.81 billion in 2018, CNY 15.46 billion in 2024, and CNY 21.54 billion in 2025. The recent step-up is especially important: the draft prospectus shows revenue rising 24.1% in 2024, then the 2025 annual report shows another 39.3% jump. This is a business moving onto a different demand curve, not normal late-cycle industrial drift, helped by exports, transformers, and more system-level selling. The 2026 first quarter then added 41.6% year-on-year revenue growth, and the 2026H1 quick report still showed 27.1% growth against a tougher base.

Margins improved less dramatically than revenue, which is healthy. It suggests this was not a pure accounting or one-off subsidy surge. In 2025 the overall product gross margin was 30.77%, down 0.48 percentage points from the prior year despite rapid growth, while attributable net profit still rose 53.7%. That combination points to scale benefits, mix, and expense leverage, not just a single gross-margin windfall. The company also continued to invest: sales expense rose 33.8%, management expense 29.3%, and R&D spending 17.2% to CNY 1.30 billion in 2025. This looks like a business still pressing its advantage rather than harvesting near-term earnings at the cost of future competitiveness.

Cash conversion is the part of the financial story that needs the most discipline. It is not broken, but it is bumpier than headline EPS suggests. In 2025 operating cash flow was CNY 2.23 billion, down 9.3% year on year despite much stronger profit. The company said rapid business growth increased purchases, supplier payments, payroll, and taxes, while inventories rose 17.3% to CNY 4.08 billion. In 2025Q1, operating cash flow was negative CNY 559 million. In 2025H1 it was negative CNY 713 million, with the company explicitly linking the outflow to inventory growth and supplier payments. The 2026H1 quick report showed a rebound to positive operating cash flow of CNY 225 million, up CNY 938 million year on year, but the broader lesson remains: Sieyuan is a real manufacturer doing real project deliveries, not a software company. Growth consumes working capital first and releases it later.

The balance sheet is sound enough for growth, though not frictionless. At 2025H1, cash was CNY 3.09 billion and trading financial assets were CNY 2.52 billion; short-term borrowings were only CNY 215 million and long-term borrowings CNY 27.5 million. Total assets were CNY 25.29 billion. The company was carrying goodwill of CNY 541 million, manageable against equity, and fixed assets plus construction in progress of nearly CNY 2.87 billion, which is consistent with ongoing capacity build-out. Contract liabilities of CNY 2.56 billion show the business still has customer prepayments supporting the model. The financial risk is therefore less about leverage and more about execution stress showing up via receivables, inventories, and FX translation.

Price and valuation history

I did not retrieve a clean full-price series from listing to the present in this pass, so the valuation history here is built from recent market data and the business inflection itself. The important phases are clear enough. For years, Sieyuan traded as a capable but not obviously scarce domestic power-equipment name. That changed when exports, transformers, and broader integrated solutions began to lift the growth rate above what the local peer group could match. By 2025-2026, the market was no longer applying a peer-like multiple. On 2026-07-10, Sieyuan traded at about 37.6x trailing earnings. Xuji was near 19.3x, Pinggao near 19.0x, and NARI near 21.3x. The market is therefore pricing Sieyuan as a structurally better grower, not merely as another beneficiary of domestic UHV and substation capex.

That multiple premium can be explained. It is anchored in faster earnings growth, a private-sector execution culture, and a materially larger overseas growth runway. But it is still a premium, and large premiums require constant proof. If overseas growth moderates, or if parent-level earnings growth slows because of FX losses and higher minority interests, the multiple can compress even if the business remains healthy. That is the stock’s central tension today.

Business model, industry and peers

Sieyuan’s business model is easiest to understand as a layered equipment platform rather than a collection of unrelated factories. The company sells primary equipment such as GIS, GIL, circuit breakers, disconnectors, transformers, reactors, bushings, capacitors, and instrument transformers; secondary equipment such as protection and automation; power-electronics and power-quality products; EPC services; and a smaller but fast-growing storage and supercapacitor portfolio. The company itself says it is one of the few domestic names able to cover both primary and secondary equipment and to provide engineering services on top. The prospectus adds that direct sales still accounted for over 99% of sales during the track-record period, which tells you customers are buying solutions and vendor accountability, not low-touch channel goods.

The profit logic resides in three places. First, switchgear remains the largest pool. It benefits from technical barriers, qualification requirements, installed-base credibility, and grid-customer stickiness. The prospectus says Sieyuan’s 750kV GIS has been delivered to major grid operators with internationally leading performance parameters, and its 1000kV GIS has passed major type tests. Second, transformers are now a strategic export lever. The company supplies oil-immersed, gas-insulated, and dry-type transformers up to 750kV, and Changzhou Sieyuan-Toshiba’s second plant gives room to scale. Third, system-level selling matters more than the raw margin of any one box. Once a vendor can bundle primary equipment, protection, automation, and EPC, it can take a larger share of project value and become harder to displace.

Cost structure and operating leverage look like those of a high-quality industrial, not a capital-light business. Materials and manufacturing are meaningful, with copper and logistics explicitly cited by management as cost variables. Working capital rises with orders. Capacity must be added ahead of demand. Yet there is also real operating leverage. In 2025, revenue grew faster than sales and management expenses, and net profit grew far faster than gross margin. R&D stayed heavy, which means scale is not coming from starving future products. The key operating-leverage test is whether the company can keep lifting revenue with a stable-to-slightly-rising gross margin while cash conversion normalizes after a growth spurt. 2026H1’s sharp cash-flow rebound is modest evidence in that direction.

The real moat is a combination of engineering breadth, qualification and delivery credibility, and the time cost of becoming an accepted vendor in mission-critical networks. Brand alone is not enough in this industry. But long product-design lives, 30- to 50-year equipment life cycles, and demanding customer standards do create switching costs. In July 2026 management said overseas customers care far more about brand reputation, product quality, and delivery assurance over lifecycle cost than about low headline price, and said the company follows a value-pricing strategy rather than using low pricing as a core tactic. That matters because it suggests the overseas mix is not being bought at unsustainably low margins. The 2025H1 report supports that view: overseas revenue had a 35.7% gross margin, above the 31.9% margin for the broader T&D segment.

Channel is not the moat here; acceptance is. Grid operators, large generators, industrial operators, and infrastructure developers want proven equipment and predictable delivery, because equipment failure is expensive and political. The prospectus says the domestic customer base spans major national power-grid groups, top-five generation groups and their subsidiaries, regional power companies, transit, petroleum, and industrial and mining users. Overseas, it says the company has stable partnerships with grid operators, power-generation companies, and industrial customers across more than 100 countries and regions. This breadth reduces customer-concentration risk relative to a pure State Grid tender specialist, even though domestic grid customers remain important.

Management and governance are stronger than the initial prompt implied, but with a different risk profile. Dong Zengping is a founder, chairman, general manager, largest identifiable shareholder, and actual controller according to company materials, not just a hired CEO. That aligns management economically, which is usually positive. It also means succession and power concentration sit closer to the center of the governance map than they do at the large central-SOE peers. The good news is that there is no evidence in the recent extracted filings of balance-sheet abuse or material governance breakdown. The 2025 annual report said the company and its controlling shareholder and actual controller had maintained good integrity status, and there is no sign in recent filings of serious regulatory penalties. The less positive point is that founder-control can deserve a discount if the organization becomes too dependent on one decision-maker. At present, the market is not applying such a discount.

Industry structure is favorable, but not frictionless. The short version is that grid investment is expanding both because more power must move and because power quality matters more than it used to. Management’s July 2026 IR comments cite two structural changes driving overseas demand: rising renewable penetration, which creates grid-stability and power-quality needs, and faster user-side electrification from loads such as EVs and data centers. The company also points to overseas grid replacement cycles and industrialization. On the domestic side, management referenced State Grid’s “15th Five-Year” fixed-asset investment plan as historically high. None of that means every supplier wins equally. Profit pools usually concentrate in technically certified high-voltage products, delivery reliability, and integrated project capability, not in the easiest-to-copy line items.

From a cycle perspective, Sieyuan is exposed to a capex cycle, a policy cycle, and a global industrial cycle. It is less exposed to the kind of pure inventory boom-bust that dominates semiconductors, and less defensive than a regulated utility or software maintenance annuity. In upcycles, orders, shipment mix, and overseas utilization are the main accelerants. In downcycles, working capital, fixed-cost absorption, and project delays are the main pressure points. That is why this is a better business than its recent cash-conversion dips might suggest, but also why it is not appropriate to value it like an asset-light compounder with perfect cash passthrough.

Horizontally, the most useful domestic comparison set is Xuji Electric, Pinggao Electric, NARI Technology, and China XD Electric. Xuji is heavily tied to State Grid and is stronger in grid control, protection, and power-electronics-related systems, but its growth and valuation are more sober. Pinggao is a classic high-voltage switchgear and GIS name with deep grid ties and lower valuation, though its international breadth is much smaller. NARI is the highest-quality domestic listed grid-technology platform, with software, automation, dispatch, and digital-grid strengths; it deserves a premium to most peers, yet still trades well below Sieyuan on earnings. China XD is broad in high-voltage primary equipment and benefits from UHV and domestic transmission investment, but it is still a more state-anchored, lower-margin perception case. Against those companies, Sieyuan’s niche is clear: founder-led, more international, more product-balanced between high-voltage primary equipment and selected secondary/solution layers, and currently the market’s preferred “China industrial export” version of the grid-equipment theme.

Key data table

Table sources: company annual reports and quote pages; market-cap values are author calculations from closing prices and reported share counts.

The table shows why the stock is tricky. Sieyuan is not outrageously large relative to its peers. In fact, NARI is materially larger in both revenue and market cap. What Sieyuan has is a much steeper growth profile and a more export-heavy narrative. That is why its multiple sits well above Xuji and Pinggao and even above NARI. For that premium to hold, Sieyuan must keep proving that its overseas order book is not just early-cycle pull-forward and that core lines like switchgear and transformers can scale without a cash-flow or margin accident.

Current fundamentals and bull-bear divergence

The last four reported periods show both momentum and stress. 2025Q1 revenue grew 21.4% and attributable net profit 22.9%, but operating cash flow was negative. By 2025H1, revenue growth had accelerated to 37.8% and profit growth remained strong, while overseas revenue surged to CNY 2.86 billion and 33.7% of sales. Full-year 2025 then delivered 39.3% revenue growth and 53.7% net-profit growth. 2026Q1 stayed strong at 41.6% revenue growth and 23.2% net-profit growth. The newly released 2026H1 quick report shows revenue of CNY 10.80 billion, up 27.1%, and attributable net profit of CNY 1.487 billion, up 15.0%, with operating cash flow improving by roughly CNY 0.94 billion year on year to a small positive inflow. That sequence says growth is real, but margins and attribution are now meeting FX and minority-interest headwinds.

The market is trading three fundamental drivers and one narrative overlay. The fundamentals are overseas orders, transformer capacity ramp, and stronger domestic grid demand. The overlay is AI-related power demand, especially in North America. That overlay is not fiction. Management itself says its North American products can serve renewable interconnection and AI data-center load access, and that it is following the opportunity closely. But the current stock price probably gives too much weight to the AI angle and too little to the more mundane reality that this remains a project-heavy equipment business where receivables, logistics, and certification still matter.

The strongest bull argument is that overseas growth has crossed from marketing phrase to measurable profit driver. In 2025, overseas revenue was CNY 5.80 billion, 26.94% of total revenue, up 85.84% year on year. In 2025H1, the overseas share had already risen to 33.68%, with a gross margin of 35.69%. Management also says global peers are expanding cautiously because of return discipline, supply bottlenecks, and uncertain payback periods, leaving a window for Chinese firms with capacity elasticity and delivery advantage. If that remains true for several more years, Sieyuan has the right product set to keep taking share.

The second bull argument is that the domestic story has not rolled over just because the international story is stronger. Management says 2026 State Grid batch procurement and special bids are proceeding on a normal, constructive rhythm, and that domestic grid and non-grid revenue mix has not changed much. In 2025, the company also highlighted major domestic breakthroughs, including the South Grid ±800kV flexible DC converter-transformer valve-side bushing bid, first 220kV protection win in South Grid centralized procurement, and a State Grid grid-forming SVG project in Hunan. Those are not the signs of a company losing its home base while chasing exports.

The third bull argument is that capacity additions are still in front of demand, not behind it. Changzhou Sieyuan-Toshiba’s second plant is now operating, and management said in January 2026 that the transformer workshop and equipment were entering batch production while deliveries remained broadly on schedule. In June 2026, the company also announced that a wholly owned transformer subsidiary would invest CNY 480 million, excluding land, to expand transformer-product capacity; earlier in the month a controlled switchgear subsidiary announced a CNY 600 million capacity expansion to meet growing demand. That is the profile of a company still widening the runway.

The bear case is more valuation- and quality-of-growth-focused. First, the stock already trades at nearly double the P/E of Xuji and Pinggao and at a clear premium to NARI. That premium can persist, but it leaves little room for even modest disappointment. Second, working-capital demands are rising. Inventory growth and weaker 2025 cash conversion show that expansion is not frictionless. Third, part of the transformer growth engine now leaks more profit to minorities after the Toshiba-related ownership change, and the 2026H1 quick report confirmed this hit attribution. Fourth, some of the newer optionality areas remain small or early. The company itself says solid-state transformers are still pre-R&D-project. Supercapacitors improved, but turning that into a durable, large-margin business is still a future question rather than a present answer.

Valuation analysis

Historically adjusted valuation work is constrained by the data extracted in this pass, but the relative picture is clear. At CNY 151.73 on 2026-07-10, Sieyuan trades on about 37.6x trailing 2025 earnings of CNY 3.15 billion. That is far above Xuji’s 19.3x and Pinggao’s 19.0x, and well above NARI’s 21.3x. The market is therefore valuing Sieyuan not as a normal A-share grid-equipment peer, but as a private-sector export compounder. The question is whether the business quality and runway justify that premium at this exact price, not whether the company is good.

The cash-flow passthrough test tempers the headline multiple debate. In 2025, operating cash flow was CNY 2.234 billion against attributable net profit of CNY 3.150 billion, a ratio of about 0.71. That is not disastrous for a project-heavy manufacturer in a strong growth year, but it is enough to say that owner earnings are lower than accounting earnings. Using 2025 as the anchor year, and treating roughly CNY 500 million of annual capex as maintenance-like rather than pure growth capex, owner earnings were closer to CNY 2.65 billion than CNY 3.15 billion. On that basis, the current market cap implies an owner-earnings multiple around 44.8x, meaning the stock is even more demanding than the headline trailing P/E suggests. I cannot verify a precise five-year OCF/net-income average from primary filings in this pass, so I am anchoring the valuation on the latest full year plus the 2026H1 cash-flow rebound rather than overstating precision.

For absolute valuation, the most suitable methods are owner-earnings-based P/E and peer-adjusted forward P/E, with price-to-sales as a rough cross-check. At the current market cap, Sieyuan trades around 5.5x 2025 sales, which is a rich industrial multiple in China even with strong growth. I assume 2026 revenue of about CNY 27.0 billion, consistent with company targets, and attributable net profit of about CNY 4.1 billion in the base case, somewhat below the most bullish sell-side figures because of FX noise and the lower parent ownership of Changzhou Sieyuan-Toshiba. I then move to 2027 owner earnings to avoid one-year noise.

Valuation scenario table

Scenario basis: owner-earnings-led valuation framework anchored on 2025 reported earnings and cash generation, company 2026 operating targets, and current peer multiples. This is valuation-scenario analysis within a research framework, not investment advice.

The business reason behind these numbers is simple. Sieyuan probably does deserve a premium to Xuji and Pinggao because its overseas growth and private-sector execution are better. It may even deserve a premium to NARI for a period if the export growth is sustained. But the stock already trades so far above the domestic peer center that even a good operating outcome can produce mediocre returns if the premium narrows. That is the hallmark of a good-company-bad-price setup.

Expectation-gap analysis points to three metrics that matter most over the next twelve months. The first is overseas order and revenue mix, especially whether growth in North America and Europe remains broad-based rather than concentrated in a few projects. The second is cash conversion, because rapid revenue growth with permanent working-capital stretch is less valuable than the market is assuming. The third is parent-level profit conversion after the Changzhou Sieyuan-Toshiba stake reduction. The market seems to be pricing market-share gains and revenue growth correctly; it may be underpricing attribution leakage and balance-sheet intensity.

Margin of safety is the decisive issue. The current price of CNY 151.73 is about 60% above my conservative fair value of CNY 95. That means the margin of safety is zero on a conservative basis. The most fragile assumption in the base case is not revenue, but the market’s willingness to keep paying a 30x owner-earnings multiple if growth merely becomes very good instead of exceptional. Cut that multiple assumption to 70% of base, while keeping the business outcome the same, and the base value falls from around CNY 132 to roughly CNY 92. If earnings simply stay flat around 2025 levels for three years, the annualized return from the current price would be poor and likely below the long-term opportunity cost of capital. My sufficiency verdict is therefore: none.

Tracking dashboard

Tracking-date sources and thresholds are based on company filings, IR records, and quote data.

These indicators matter because they separate real execution from narrative heat. If orders and overseas mix remain strong while cash conversion improves, the bull case stays intact even if the stock moves sideways for a while. If orders stay fine but cash conversion, gross margin, or attributable profit weaken, then the market will have to decide whether it is still paying for a compounder or merely for a cyclical winner that briefly outperformed.

Research uncertainties and sources

The largest blind spots in this report are fivefold. I did not retrieve the original 2004 IPO pricing and proceeds from a primary prospectus. A complete primary five-year cash-flow series was also out of reach, so owner-earnings work is anchored on 2025 and 2026H1 rather than a full historical run. Nor did I extract a complete 2025 product-level gross-margin table from the full annual report. The latest disclosed overseas split by country or region is still limited in the pages reviewed; the company discusses breakthroughs in Europe and North America but does not give a fully quantified geographic revenue map in the extracted text. Finally, some market-position claims in the horizontal analysis rely on the company’s 2026 draft prospectus quoting Frost & Sullivan rather than a directly retrieved third-party methodology document.

The most heavily used primary materials were the 2025 annual-report summary, the 2025 annual-report-derived business-analysis extracts, the 2025H1 report, the 2026Q1 report, the 2026H1 quick report, and multiple 2026 investor-relations activity records. I also used the company’s 2026 draft prospectus for customer mix, product architecture, and international footprint, and used exchange or quote pages only for market pricing and scheduling.