Full Report

Grid and Polysilicon

TBEA Co., Ltd. (600089.SS) is a Changji, Xinjiang industrial group that has reported almost exactly ¥97 billion of revenue for four straight years — yet its profit fell from ¥15.9 billion in 2022 to ¥4.1 billion in 2024 before recovering to ¥6.0 billion in 2025. That gap is the company. A durable, share-gaining grid-equipment franchise sits alongside a high-purity polysilicon business whose gross margin collapsed from 58% to near zero. This chapter maps what TBEA is; the report works through whether the steady half is worth owning through the cyclical half.

What the company does

TBEA listed on the Shanghai Stock Exchange in June 1997 and today runs four businesses [1]. The first is power transmission and transformation (输变电): transformers, wire and cable, switchgear, capacitors, and turnkey international EPC projects — the equipment that steps grid voltage up and down, including China's ultra-high-voltage backbone. The second is new energy (新能源): high-purity polysilicon, inverters, and the design, build and operation of solar and wind plants. The third is energy (能源): coal mining plus power and heat generation. The fourth is new materials (新材料): high-purity aluminum and electronic foils [2].

The polysilicon operation runs through Xinte Energy (新特能源, 1799.HK), a Hong Kong–listed subsidiary TBEA controls — its stake was raised to 66.61% in 2025 [3] [4]. Coal sits in another controlled subsidiary, Tianchi Energy, which mines the Zhundong field in Xinjiang [5]. So TBEA is a holding structure: a wholly-owned grid-equipment group, plus controlling stakes in a listed polysilicon maker and a coal miner.

Sources: revenue, net income and total assets from FY2025 Annual Report, Key Accounting Data [6]; market cap and P/E derived from 5.05bn shares at ¥22.01 (share price as reported, 3 July 2026) and reported earnings.

Flat revenue, a swinging bottom line

Revenue has been remarkably steady — between ¥96.5bn and ¥98.2bn every year since 2022 [7]. Profit has not. Net income attributable to shareholders more than halved from 2022's ¥15.9bn to 2023's ¥10.7bn, then fell again to ¥4.1bn in 2024, before a partial rebound to ¥6.0bn in 2025 [8] [9]. A stable top line hiding a profit line that fell three-quarters and bounced is the signature of a commodity cycle running underneath a steady core.

Source: FY2025 Annual Report, Key Accounting Data [10]; FY2024 comparatives from FY2024 Annual Report [11].

The mechanism shows up in the consolidated gross margin, which fell from 38.6% in 2022 to 18.2% in 2024 and held at 18.9% in 2025 — revenue stayed put while the profit inside each sales dollar was cut roughly in half.

Source: derived from reported revenue and cost of sales, FY2022–FY2025 Annual Reports [12].

The two engines

The 2025 business mix is best read segment by segment. Electrical equipment — the transformer core — is now the largest line and the one whose margin is rising. New energy, once the profit engine, earned essentially nothing on ¥13.6bn of sales. Coal and power generation quietly throw off high-margin cash.

Source: FY2025 Annual Report, Principal Business by Product [13].

The number that carries the story is the 0.6% gross margin on new energy. Three years earlier it was 57.8%, on ¥34.4bn of revenue [14]. High-purity polysilicon — the raw material for solar cells — went from a windfall to break-even as Chinese polysilicon prices crashed under industry-wide overcapacity. The transformer business moved the other way: electrical-equipment margin rose to 19.8% as grid demand and capacity additions lifted volumes [15].

Sources: FY2022 Annual Report [16] and FY2025 Annual Report [17].

How concentrated the cycle is

Because Xinte is a majority-owned subsidiary, the polysilicon swing is visible in one clean line: minority interest. In 2023, minority shareholders — chiefly Xinte's — took ¥3.39bn of profit. In 2024 they absorbed a ¥537m loss, a swing of nearly ¥3.9bn at that line alone [18]. The same year, asset-impairment losses reached ¥3.56bn, driven by inventory and fixed-asset write-downs at Xinte, more than double the ¥1.60bn booked in 2023 [19]. Much of the earnings volatility a shareholder feels originates inside one listed subsidiary.

Management's response has been to throttle output rather than chase share. Xinte produced 198,800 tonnes of polysilicon in 2024 [20]; in 2025 it made just 96,400 tonnes, down 51.5%, running its plants at only 37.1% of capacity [21]. The polysilicon engine has been idled, not repaired — the price signal, not company choice, will decide when it restarts.

The durable core, and what it costs to keep

Against that volatility, two things are steady. The grid-equipment franchise is growing — electrical-equipment revenue rose 19.7% in 2025 — and China's ultra-high-voltage and grid-investment cycle underpins demand for exactly the transformers, cable and turnkey projects TBEA builds [22]. And the coal-plus-power complex is a genuine cash engine: coal earned a 22.4% gross margin on ¥17.0bn of sales, power generation 54.8% on ¥7.2bn [23].

What complicates the "durable core" read is capital intensity. TBEA spends ¥17–22bn a year on capex, and in 2024 and 2025 that outlay outran operating cash flow, turning free cash flow negative — from +¥6.9bn in 2023 to −¥4.3bn in 2024 and −¥12.7bn in 2025 [24]. The company kept investing heavily even as polysilicon economics collapsed. Total assets grew to ¥227bn and long-term debt to ¥38.6bn, up from ¥25bn in 2022 [25]. Whether that spend is building the next franchise or funding a down-cycle at the wrong time is a question later chapters take up.

Source: FY2025 Annual Report, Consolidated Cash Flow Statement [26].

Size, valuation and what is priced

At ¥22.01 (3 July 2026), TBEA's 5.05 billion shares are worth roughly ¥111bn — about 19 times trailing earnings. The company still pays a dividend, declaring ¥0.36 per share for 2025, a yield near 1.6% held up through the downturn [27]. Sell-side consensus looks for revenue to reaccelerate roughly 10% in each of 2026 and 2027 — a bet, in effect, that polysilicon has bottomed and the grid business keeps compounding.

That framing sets the question this report exists to answer: TBEA pairs a durable, share-gaining grid-equipment franchise and a coal-and-power cash engine with a high-purity polysilicon business whose margin has fallen from roughly 58% to near zero — and the case turns on whether the steady core is worth owning through the polysilicon down-cycle, and through the ¥20bn-plus of annual capex still flowing into it, at today's ~19x earnings. Everything that follows tests one part of that sentence.

---

The Grid Franchise

TBEA's polysilicon subsidiary earned almost nothing in FY2025. The company still made ¥6.0bn because a second business kept working underneath it: the power-transmission franchise — transformers, cables and turnkey grid projects. That franchise nearly doubled its revenue over four years, widened its margin while polysilicon collapsed, and, with the coal-and-power engine, supplied roughly four-fifths of group gross profit. Its durability is the reason the down-cycle is survivable — and where a buyer's due diligence should start.

Two engines carried the year

In FY2025 the group's ¥18.4bn of gross profit came from almost everywhere except the business that once defined it. The transmission-and-distribution franchise and the coal-and-power engine together produced about ¥15.1bn — 82% of the total — while new energy, mostly polysilicon, contributed ¥0.08bn, or four-tenths of one percent.

Source: FY2025 Annual Report, main-business segment table (revenue × segment gross margin) [1]; group gross profit per reported financials.

The point is not that polysilicon is small — its ¥13.6bn of revenue is a fifth of the top line. The point is that at a 0.59% gross margin it converts almost none of that revenue into profit, so the earnings the group actually keeps are made elsewhere. The Grid and Polysilicon chapter established the collapse; the question that follows is whether the businesses absorbing it are worth owning. For the grid franchise, the four-year record answers most of it.

A core that grew while the cycle broke

The transformer line — reported as "transformer products" through FY2023 and as the slightly broader "electrical equipment" from FY2024 — grew revenue in every one of the last four years, from ¥13.5bn in FY2022 to ¥26.8bn in FY2025 [2]. Over the same window its gross margin recovered from 15.9% to 19.8% [3]. That is the mirror image of the new-energy line, which went the other way to 0.59% over the same three years.

Source: FY2021–FY2025 Annual Reports, main-business product tables [4] [5] [6] [7] [8].

Source: FY2021–FY2025 Annual Reports, main-business product tables (as above) [9].

The demand behind the growth is visible in the order book. Domestic grid contract signings rose from about ¥26.6bn in FY2021 to ¥40.5bn in FY2023 to ¥56.2bn in FY2025 — the last figure up 14.5% year-on-year, which implies roughly ¥49bn of signings in FY2024 [10] [11] [12].

Source: FY2021, FY2023 and FY2025 Annual Reports, management-discussion domestic-market sections [13] [14] [15]. FY2024 (~¥49bn) derived from the disclosed FY2025 growth rate; the FY2024 report's printed figure appears mis-scanned.

This is not a company-specific miracle so much as a company well-placed in a rising tide. China's grid construction investment reached ¥639.5bn in FY2025, up 5.1%, with ultra-high-voltage (UHV) direct-current project investment alone up 25.7% as wind-and-solar mega-bases in the west need long-distance transmission to demand centres in the east [16]. Global transformer prices have risen 40–60% since early 2022 on tight capacity and extended lead times (industry research), which helps explain how TBEA lifted margin and volume at once rather than trading one for the other.

The moat: scale and UHV, not the fattest margins

TBEA's advantage in transmission is real, but it is a scale-and-capability advantage, not a pricing one. Its electrical-equipment line alone booked ¥26.8bn of revenue in FY2025 — more than the entire turnover of China XD Electric (¥23.8bn) or Sieyuan Electric (¥21.5bn), the two nearest listed grid-equipment peers [17]. Add cables (¥15.6bn) and turnkey grid engineering (¥4.9bn) and the franchise is close to half of group revenue. It ranks first in China's electrical-industry Top 100 and runs three overseas manufacturing bases [18]; its annual transformer output, reported by the company at roughly 240 million kVA, is among the largest of any single maker.

Capability shows up in the products it is first to build: TBEA has developed China's ±1100kV UHV DC converter transformer and ±800kV flexible-DC converter valve as domestic first-of-kind equipment, backed by ¥4.79bn of research spending in FY2025 — 4.92% of revenue [19] [20]. In FY2025 it won the sending- and receiving-end converter stations for the State Grid and China Southern Grid Tibet–Guangdong DC line, a marker of standing at the top voltage classes where the field of qualified suppliers is smallest [21].

What the scale does not buy is superior profitability. TBEA's electrical-equipment gross margin of 19.8% sits below all three focused peers — and well below Sieyuan, which earns a 30.8% gross margin and a 14.6% net margin, roughly double TBEA's group-level returns.

Sources: TBEA electrical-equipment segment, FY2025 Annual Report [22]; peer figures derived from reported financials (China XD 601179, Pinggao 600312, Sieyuan 002028, FY2025). TBEA's is a segment gross margin; peer figures are company-level, so the comparison is directional. TBEA net margin is not shown at segment level.

Two things pull TBEA's blended franchise margin down. Cables — ¥15.6bn of revenue at an 8.3% gross margin — are a largely commodity product bundled into the franchise total [23]. And Sieyuan's mix skews to higher-value switching, monitoring and reactive-power gear where specification, not tonnage, sets the price. The honest read: TBEA is the volume and UHV-capability leader of Chinese grid equipment, and a beneficiary of a genuine multi-year buildout, but a focused competitor can and does earn more per yuan of sales. Scale is the moat here; margin premium is not.

The export leg is bigger than it looks

The part of the franchise least visible from the headline numbers is international. Overseas revenue doubled from ¥6.3bn in FY2022 to ¥12.7bn in FY2025, at gross margins in the mid-teens comparable to the domestic book [24] [25].

Source: FY2022 and FY2025 Annual Reports, main-business geographic tables [26] [27].

Behind that revenue is a turnkey engineering business — substations, transmission lines and power plants built as complete projects across Asia, Africa and the Middle East. TBEA signed roughly US$2.0bn of international orders in FY2025 and reported an international turnkey backlog — contracts in execution but not yet recognised, plus contracts still to be performed — of over US$5.0bn [28]. The scale of the footprint is easy to underappreciate: the FY2024 accounts carry foreign-currency working-capital balances in dozens of currencies — from Indian rupee and Pakistani rupee to a long tail of African and Central Asian units — evidence of a genuinely global project book rather than opportunistic exports [29]. That book is a real asset; it also carries the receivable and country risk that comes with building in frontier markets, which a later chapter on working capital should size.

What would change this read

The bull case for owning TBEA through the polysilicon trough leans on this franchise being durable, and the evidence supports that: rising revenue, recovering margin, a growing order book, and a demand backdrop — grid electrification and UHV — with years to run. The read would weaken on three fronts, each checkable in future filings: domestic signings turning down (they have compounded, but China grid capex is policy-set, not contracted), the margin gap to focused peers widening rather than closing, or the international backlog converting into impairments and slow-paying receivables instead of cash. None of those is visible yet. The franchise is doing the job the thesis needs it to do.

---

The Polysilicon Engine

TBEA's largest subsidiary by assets is also its only loss-maker. Xinte Energy — 66.61%-owned and Hong Kong–listed — houses the high-purity polysilicon business, whose product-level gross margin fell from +71.10% in FY2022 to −47.84% in FY2025 as output was deliberately cut to 37% of capacity [1] [2]. Two structural features cushion the group: roughly a third of the loss sits with Xinte's minority holders, and the capacity build is finished.

Sources: FY2025 Annual Report, Analysis of Major Subsidiaries [3]; Photovoltaic Product Information [4].

What TBEA consolidates

Xinte Energy Co., Ltd. (1799.HK) is a self-contained polysilicon and solar company that TBEA controls but does not wholly own. On a full-consolidation basis it is the group's heaviest entity: ¥82.1bn of total assets and ¥45.1bn of liabilities, against ¥37.0bn of net assets [5] [6]. TBEA carries all of that on its balance sheet even though it owns two-thirds of the equity. In FY2025 Xinte generated ¥15.3bn of revenue and a ¥1.33bn net loss — the loss narrowing sharply from ¥4.04bn in FY2024, when revenue was ¥21.2bn [7] [8].

Set against its siblings, the picture is stark: the entity that owns the most assets earns the least.

Source: FY2025 Annual Report, Analysis of Major Subsidiaries [9].

Tianchi's coal-and-power operation earned ¥3.39bn and the transformer group ¥1.70bn; together they more than covered Xinte's loss and delivered the group's ¥5.95bn of attributable profit [10]. Xinte is the reason group earnings swing; it is not the reason they exist.

How far the polysilicon economics fell

The segment disclosure understates the collapse. TBEA's "new energy" segment — which bundles inverters, EPC engineering, and solar-station operations with polysilicon — reported a 0.59% gross margin on ¥13.6bn of FY2025 revenue [11]. The polysilicon product on its own tells a harder story: revenue of just ¥2.92bn at a −47.84% gross margin [12]. Three years earlier the same product sold ¥25.3bn at a 71.10% margin.

Sources: FY2022 [13], FY2023 [14], FY2024 [15], FY2025 [16] Annual Reports, Photovoltaic Product Information.

The margin swing translates into roughly ¥19bn of gross profit lost off a single product line: about ¥18bn of gross profit in FY2022 became a ¥1.4bn gross loss in FY2025.

Source: derived from reported polysilicon product revenue and gross margin, FY2022–FY2025 Annual Reports [17].

Two responses run underneath the margin line. First, TBEA cut volume hard: FY2025 polysilicon output was 9.64万吨 (96,400 tonnes) against roughly 260,000 tonnes of nameplate capacity — a 37.08% utilization rate, and output down 51.51% year on year [18]. The company calls this "self-disciplined production control" (自律控产) — running plants for maintenance and technical upgrades rather than chasing loss-making tonnes. Second, this is an industry price problem, not a TBEA-specific one: the FY2024 report states plainly that the market price of polysilicon had "fallen below producers' cost" and stayed there [19]. A −47.84% gross margin at a third of capacity is a company selling well below the cost of what it makes, while trying to make as little of it as pricing allows.

The minority buffer

Here the ownership structure matters more than it first appears, and it cuts two ways.

Because Xinte is only 66.61%-owned, roughly a third of its losses accrue to Xinte's own minority shareholders, not to TBEA's. That shows up in the minority-interest line. In FY2024 minorities absorbed a ¥541m loss, which lifted TBEA's attributable profit above the group's total net profit; in FY2025, as Xinte's loss narrowed, the same line flipped to a small +¥47m [20]. Group management attributes the FY2025 recovery in profit chiefly to polysilicon losses narrowing, alongside profit growth in transmission equipment and gold [21]. The listed-subsidiary structure means TBEA's shareholders never ate the full polysilicon loss.

The other side of the same fact is the balance sheet. TBEA consolidates 100% of Xinte's ¥45.1bn of liabilities and 100% of its capital programme, while owning two-thirds of the equity. It is not retreating from that exposure — in April 2025 TBEA raised its Xinte stake from 64.52% to 66.61%, paying ¥417m to buy in shares from employee partnerships [22]. And it continues to fund the capital-heavy businesses partly with outside money: group subsidiaries absorbed ¥4.59bn of fresh minority-shareholder capital in FY2025, and minority equity on the consolidated balance sheet stands at ¥26.5bn [23] [24]. The economics are therefore split: about two-thirds of Xinte's profit and loss reaches TBEA holders, but the whole of its debt and capex sits on the group.

The impairment record confirms the pain is being taken, not deferred. Xinte's fixed-asset and inventory write-downs drove a ¥3.56bn group asset-impairment charge in FY2024 — most of that year's ¥4.14bn attributable profit — easing to ¥976m in FY2025 as the worst of the mark-downs passed [25].

Where the cycle stands

The case set out at the start of this report turns on whether the steady core is worth owning through the polysilicon down-cycle; the state of that cycle is therefore the live variable. Three points frame it.

The price has stabilised but not recovered. N-type polysilicon traded around 31.5–33.7 RMB/kg in late June 2026, a tentative floor after prices spent 2025 and early 2026 below cash cost, with Chinese producers in Xinjiang and elsewhere still cutting output to hold the line. At those levels much of the industry, TBEA included, is still not covering full cost — consistent with a product gross margin that was −47.84% in FY2025.

The capital build is essentially done. TBEA's polysilicon nameplate already stands near 260,000 tonnes a year, with no new lines under construction disclosed [26]. Group capex was still heavy at ¥22.1bn in FY2025, but that spend is now flowing to the coal, power and materials businesses rather than to new polysilicon capacity [27]. The polysilicon plant is built; the question is utilisation and price, not further investment.

And the FY2025 profit recovery did not come from a polysilicon rebound. It came from the loss narrowing — helped by lower impairments, a stronger transmission and gold business, and a mark-to-market gain on a listed equity holding — not from polysilicon returning to profit [28].

The reasonable read is that Xinte is a large, deeply cyclical, loss-making asset whose downside is genuinely shared with outside shareholders and whose capital demands are largely behind it — a drag the group can carry, rather than a hole it must keep filling. The strongest fact against that read is size: at ¥82bn of assets and ¥45bn of consolidated liabilities, Xinte is too big to be treated as a rounding error, and a renewed leg down in polysilicon prices, or a decision to restart idled capacity into a weak market, would deepen the loss and the impairment risk again. What would change the read in the other direction is straightforward to watch: polysilicon prices clearing producers' cash cost on a sustained basis would move Xinte from a roughly ¥1.3bn drag toward breakeven, and the utilisation rate — 37% in FY2025 — is the single line that will show it first.

---

Coal and Power

The coal-and-power business inside Tianchi Energy — TBEA's 85.78%-owned Xinjiang subsidiary — is the group's largest single earner and the cash that funds its polysilicon capex. Tianchi made ¥3.39bn of net profit in FY2025, close to half of group attributable profit [1]. But the engine is cooling: its profit has roughly halved since FY2023 as Xinjiang coal prices fell [2]. A growing, high-margin pithead power business is the partial offset.

The group's largest earner

Tianchi Energy runs open-pit coal mines in the Zhundong (准东) coalfield of Xinjiang — the national reserve base the filing describes as shallow-buried, low-stripping-ratio, and among the most economic in the region — with approved coal capacity of 74 million tonnes a year and 5,040MW of pithead thermal generation (4,040MW excluding captive plants) [3] [4]. In FY2025 the subsidiary held ¥54.34bn of assets and turned ¥23.85bn of revenue into ¥3.39bn of net profit [5].

Sources: FY2025 Annual Report, Principal subsidiaries [6] and product-segment table [7]; share of profit derived from reported financials [8].

The structure is vertically integrated. TBEA mines low-cost coal, burns part of it in its own pithead plants, and moves the electricity out of Xinjiang through the "West-to-East" transmission corridors its own transformers equip — the plants are labelled "疆电外送" (Xinjiang power export) in the filing [9]. This is why the two lines behave differently as coal prices move, and why they belong in one chapter rather than two.

The engine carries the group

Coal and power together produced ¥7.73bn of gross profit in FY2025, about 42% of the group's ¥18.28bn — a slice larger than the entire grid-equipment franchise (The Grid Franchise) [10]. At the net-profit level the concentration is starker. Tianchi's ¥3.39bn, multiplied by TBEA's 85.78% stake, contributes roughly ¥2.91bn to attributable profit — about 49% of the group's ¥5.95bn in FY2025 [11] [12].

Sources: group attributable profit, FY2025 Annual Report key accounting data [13]; Tianchi net profit at 85.78% stake, FY2023–FY2025 principal-subsidiary tables [14] [15] [16].

FY2024 is the year that shows what the engine does. Tianchi's attributable contribution (about ¥4.26bn) actually exceeded the group's entire ¥4.14bn of attributable profit — everything else the group did, netted together, lost money [17] [18]. The "steady core" of the investment case is, at the bottom line, substantially one coal-and-power subsidiary in Xinjiang.

The engine is cooling

The reason the engine cannot be treated as a fixed floor is that its profit is falling with the coal price. Coal-product gross margin has dropped every year since FY2022 — 47.63%, then 46.41%, 32.42%, and 22.39% in FY2025 — a 25-point decline in three years [19] [20] [21] [22]. This is a price effect, not a volume one: management states coal output was held roughly flat and sales volume "stable," against a backdrop of surging Xinjiang capacity and "continuously falling" coal prices [23]. Coal-product gross profit has fallen from about ¥8.46bn in FY2023 to ¥3.80bn in FY2025, a decline of roughly 55%.

Source: coal-product and power-generation lines of the FY2022–FY2025 product-segment tables [24] [25] [26] [27].

The internal hedge

The same falling coal price that squeezes the mining margin is fuel-cost relief for the power plants, and this is where the vertical integration pays. Power-generation revenue grew 28.2% in FY2025 to ¥7.18bn, and its gross margin held at 54.75% — up from a ¥4.51bn base in FY2022 [28] [29]. Power gross profit (¥3.93bn) overtook coal gross profit (¥3.80bn) for the first time in FY2025 — the chart above shows the crossover. The growth came from newly commissioned pithead units burning captive coal that gets cheaper as the market falls; thermal output reached 15.54 TWh at an average 3,846 utilization hours [30].

Source: FY2022–FY2025 product-segment tables [31] [32] [33] [34].

The hedge is real but partial. Power generation added about ¥1.79bn of gross profit between FY2023 and FY2025; coal-product gross profit shed roughly ¥4.66bn over the same span. The offset covered a bit more than a third of the coal decline, which is why total energy-segment gross profit still fell from ¥10.60bn (FY2023) to ¥7.73bn (FY2025). Vertical integration slows the bleed; it does not reverse it while coal prices are falling.

What it funds, and what to watch

The engine's job in the group is to generate the cash that the capital-hungry businesses consume — the polysilicon build (The Polysilicon Engine) and the group's continuing ¥22bn annual capex. That capex is not slowing: alongside the completed polysilicon lines, Tianchi is building a ¥17.03bn Zhundong coal-to-gas plant (about 18% complete) and the group a ¥6.78bn alumina project, so the energy and materials arms are themselves absorbing capital rather than only returning it [35].

The read: the coal-and-power engine remains the most valuable, highest-return part of TBEA, but it is a still-large cash machine that is shrinking, not a fixed floor. The strongest fact against a bearish view of it is the power business — a 55%-margin, growing, captive-fuelled stream that structurally benefits from the very coal-price weakness that hurts the mining line. What would change the read, in either direction, is checkable in the next two annual reports: the coal-product gross margin (22.4% and falling), whether power-generation volume keeps climbing fast enough to offset it, and whether Tianchi's net profit stabilizes above the ¥3bn mark or continues down toward the group's other earners.

---

Cash Conversion

TBEA's reported profit does become operating cash — over FY2022–FY2025 cumulative operating cash flow ran about 1.9 times net income, so this is not an accruals story. The problem sits one line lower: a capital-spending programme now running at 2.4 times operating cash flow has pushed free cash flow to -¥12.75bn, a third straight year of deterioration [1]. The gap is filled by new debt and by fresh cash from minority shareholders, while the dividend keeps flowing.

Sources: FY2025 Annual Report, Consolidated Statement of Cash Flows [2]; net debt and ratios derived from reported financials, FY2022–FY2025.

Earnings convert to operating cash; capex takes it all

The accrual test comes out clean. In each of the last four years operating cash flow exceeded net income — by 1.4x in the FY2022 peak, and by more than 3x in the depressed FY2024 as depreciation and working-capital release cushioned a thin profit. Cumulatively, ¥69.9bn of operating cash against ¥36.7bn of attributable profit is a conversion rate a skeptic cannot fault at the earnings-quality level.

Source: derived from reported financials, FY2022–FY2025 annual reports; FY2025 and FY2024 cash flows per the Consolidated Statement of Cash Flows [3] and FY2023/FY2022 per the FY2023 report [4].

The free-cash line tells the opposite story. Operating cash flow fell from ¥25.81bn in FY2023 to ¥12.91bn in FY2024 and ¥9.33bn in FY2025 [5][6], even as attributable profit rose in FY2025. Capital expenditure moved the other way — to ¥22.08bn, its highest in the window [7]. Free cash flow has gone from +¥6.91bn to -¥4.29bn to -¥12.75bn.

Sources: FY2025 Annual Report, Consolidated Statement of Cash Flows [8]; FY2023 Annual Report [9].

Depreciation and amortization of about ¥6.51bn (derived from reported financials) sets a rough floor for maintenance spending, so most of the ¥22.08bn of capital expenditure [10] is discretionary growth capital — the polysilicon nameplate now complete (The Polysilicon Engine) and the ¥17.03bn coal-to-gas and ¥6.78bn alumina builds still running (Coal and Power). That is the two-sided point: the free-cash deficit is a choice, not a distress signal, and it reverses the year management stops building. Until then, the businesses funding the spend are earning less than they were — the same coal and polysilicon margins that fell in the prior chapters.

The working-capital build under the cash line

Part of the operating-cash decline is a balance sheet that has absorbed cash as the order book grew. Trade receivables rose to ¥19.51bn from ¥12.93bn in FY2022 — a 51% increase against essentially flat revenue [11]. Inventory climbed the same 51%, to ¥21.39bn [12]. On top of that sit ¥6.57bn of contract assets — engineering revenue recognized on a percentage-of-completion basis but not yet billed, including frontier-market projects in Tanzania and Uganda whose settlement conditions are not yet met [13].

Source: FY2025 Annual Report, Consolidated Balance Sheet [14]; FY2022–FY2023 balances from reported financials, as reported.

Counting notes receivable (¥1.65bn), receivable-financing balances (¥3.64bn), trade receivables and contract assets together, roughly ¥31bn of the group's cash is tied up in what customers and projects owe it [15]. Some of that build is genuine growth in the grid and export franchise (The Grid Franchise); part is the slower collection that comes with a larger share of frontier-market turnkey work.

How operating cash is held up

The mechanism behind operating cash deserves a direct look, because it is doing quiet work. TBEA monetizes its receivable book rather than waiting on it. In FY2025 it endorsed or discounted ¥3.71bn of bank-acceptance notes with full derecognition — moved off the balance sheet, pulling their cash forward into operations [16]. It ran a further ¥1.67bn of bill-discount borrowing [17], and one of its FY2025 bond issues earmarked ¥700m of proceeds specifically to fund a supply-chain factoring book for its trading partners [18].

None of this is improper — bank-acceptance-note discounting is routine for a Chinese industrial, and the amount has actually shrunk from about ¥5.8bn at the FY2022 peak. But it is a lever with a ceiling: operating cash is being supported by turning receivables into cash faster, not by the underlying businesses generating more of it. A reader should treat the reported ¥9.33bn of operating cash as flattered, modestly, by financing-adjacent activity rather than understated.

Who funds the gap

With free cash flow at -¥12.75bn and the dividend still being paid, the shortfall is covered from three places. The group took on net new debt of about ¥9.0bn — ¥27.08bn of borrowings plus ¥1.5bn of bonds against ¥19.55bn repaid [19]. It drew ¥4.59bn of fresh cash from minority shareholders of its subsidiaries — outside investors putting money directly into Xinte and the energy arms rather than into TBEA itself [20]. The balance came from running down cash, which fell ¥2.49bn to ¥25.69bn [21].

Source: FY2025 Annual Report, Consolidated Statement of Cash Flows [22].

The minority-injection line is the same structure the polysilicon chapter flagged from the other side: TBEA consolidates 100% of Xinte's debt and capex but owns two-thirds of its equity (The Polysilicon Engine). On the funding side, that asymmetry works in TBEA holders' favour — outside shareholders financed ¥4.59bn of the group's capital programme in FY2025, against ¥0.66bn a year earlier. It is real capital, but it is not TBEA's, and it comes with a claim on a third of the subsidiaries' future cash.

The dividend, the equity, and the leverage

Two clarifications matter for judging balance-sheet resilience. First, the share count. Weighted shares jumped from about 4.48bn to 5.05bn between FY2023 and FY2024, which reads like an equity raise into the capex. It was not: share capital rose from ¥3,885m at end-FY2022 to ¥5,053m at end-FY2023, matched by a reduction in capital reserve — a capitalization of reserves that raised no cash and diluted no one proportionally [23][24]. The ¥22bn capex was funded by debt and internal and minority cash, not by selling new shares to the public.

Second, the dividend is held to a set share of profit rather than defended at all costs. The FY2025 distribution is ¥0.36 per share, ¥1.81bn in total, pegged at 30.35% of attributable profit [25] — the same 30% ratio as FY2024, and well below the 64% three-year cumulative payout the group ran through the higher-earning years [26]. Management is conserving cash, not stretching to sustain a headline yield. Against that, ¥4.0bn of perpetual bonds sit inside reported equity of ¥74.4bn — securities that carry coupons and behave like debt, so a stricter lens would nudge leverage up by that amount [27].

Net debt roughly doubled in a year, from ¥8.31bn to ¥16.56bn, and total borrowings — long-term loans, bonds and leases — reached about ¥42bn from ¥27bn in FY2022 (derived from reported financials). At 22% of equity, net leverage is not yet a concern; against ¥25.69bn of cash [28] and a debt-to-equity ratio near 0.57, the group can carry the current spend. The condition that would change that read is duration: if capex stays above ¥20bn for another two or three years while coal and polysilicon margins hold at today's depressed levels, the funding mix leans harder on debt and minority capital, and the balance-sheet cushion that looks comfortable now thins out.

Consensus expects revenue to grow about 10% in FY2026, to roughly ¥107bn. Faster top-line growth does not, on its own, close a cash gap driven by capital intensity and margin — it is the capex decision and the coal and polysilicon price path, not the revenue line, that determine when free cash flow turns positive again.

---

Sum-of-the-Parts

At ¥22.01 on 3 July 2026, TBEA is a ¥112.9bn company — 19x reported and 25x recurring FY2025 earnings, 1.5x the equity attributable to its shareholders. Split that price into its pieces and the surprise is what carries it: the one part with a live market quote, the listed polysilicon stake, is worth only about ¥6bn — roughly 5% of the whole. The other 95% is the grid and coal-and-power engines, and coal, not the grid, is the larger earner.

What ¥22 buys

TBEA earned ¥5.95bn attributable to its shareholders in FY2025, ¥1.161 per share, for a reported P/E near 19x [1]. But ¥1.40bn of that profit was non-recurring, most of it the mark-to-market gain on two listed equity stakes — Huadian New Energy (600930.SH) and CSG Energy (003035.SZ) — revalued at year-end prices [2]. Strip it out and recurring profit was ¥4.55bn, ¥0.8821 per share, lifting the operating multiple to 25x [3]. Return on equity was 8.75% reported, 6.65% on recurring profit [4].

Sources: share price and count as of 3 July 2026 (market data); earnings, book value and recurring profit from FY2025 Annual Report [5].

Those are blended multiples on trough-ish earnings: FY2025 profit of ¥5.95bn sits against ¥10.7bn in FY2023 and ¥15.9bn in FY2022 [6]. Whether 19x is cheap or full depends entirely on which parts of the group are doing the earning, and what each is worth on its own.

The parts, by what they earn

TBEA discloses net profit for its four principal subsidiaries — the cleanest way to see where the money is made [7]. The grid franchise — the electrical-equipment group that the investment case leans on (The Grid Franchise) — earned ¥1.69bn on ¥37.1bn of revenue. The coal-and-power engine, Tianchi Energy (Coal and Power), earned twice as much: ¥3.39bn. Polysilicon, Xinte Energy, lost ¥1.33bn (The Polysilicon Engine), and the new-materials arm (XJZH) added ¥0.66bn.

Source: FY2025 Annual Report, major-subsidiary analysis (电装集团, 天池能源, 新特能源, XJZH) [8].

Adjusting for minority interests — TBEA owns all of Tianchi (85.78% directly, the balance through subsidiaries) but only 66.61% of Xinte and 36.81% of the new-materials arm [9] — the coal engine contributes its full ¥3.39bn to TBEA holders, the grid ¥1.7bn, materials only about ¥0.24bn, and polysilicon a roughly ¥0.9bn drag. The "durable core" the report has emphasized is real, but on FY2025 earnings it is the second-largest of the three engines. The largest is a coal-and-power business whose profit has already halved from ¥6.93bn in FY2023 and whose coal-product margin fell from 47.6% to 22.4% (Coal and Power).

The polysilicon option is priced near zero

The polysilicon business is the one piece with an arm's-length mark. Xinte Energy is separately listed in Hong Kong (1799.HK) [10], and the market there values the whole company at roughly HK$10bn — about ¥9bn — trading at close to a quarter of its book value. TBEA's 66.61% share is therefore worth about ¥6bn, or some 5% of TBEA's ¥112.9bn market capitalisation.

Set that against what sits inside TBEA's own accounts: 66.61% of Xinte's ¥37.05bn of net assets is about ¥24.7bn of book value [11]. The market prices the stake at roughly a quarter of that carrying amount. In plain terms, TBEA's shareholders are being asked to pay almost nothing for the polysilicon arm — its ¥82.1bn of consolidated assets, its idle capacity, and any recovery it might deliver.

Sources: subsidiary net profit and stakes, FY2025 Annual Report [12]; Xinte stake at 66.61% [13]; Xinte listing on 1799.HK [14]; Hong Kong market capitalisation per market data, 3 July 2026.

What the price implies

Take the ¥6bn Xinte stake out of the ¥112.9bn and roughly ¥107bn remains for the grid, coal-and-power, new-materials and the listed-stake portfolio, net of ¥16.6bn of net debt (Cash Conversion). Against combined recurring earnings from those operating engines of about ¥5.3bn, that is close to 20x — a full multiple for a mix in which the biggest single contributor is a coal business already past its own peak, and the grid franchise, though growing, earns a mid-single-digit net margin.

There is a real bull answer to that. FY2025 earnings are depressed: at the FY2022 level, polysilicon alone generated close to ¥19.9bn of gross profit (The Polysilicon Engine), and even a partial normalisation would swamp today's multiple. The Huadian and CSG stakes are worth more than their contribution to reported profit suggests — the ¥1.52bn mark-to-market gain in FY2025 is a hint at scale [15], and Huadian only listed during the year [16]. That is why the four analysts covering the stock carry an average target of ¥32.96, about 50% above the current price.

The bear answer is the arithmetic above: a 25x recurring multiple, a 6.65% recurring return on equity, the largest earner cyclically supported and structurally shrinking, and free cash flow of -¥12.75bn while the polysilicon "option" still consumes capital (Cash Conversion). What would move the read is specific and watchable: a polysilicon price back above Xinte's cash cost, which turns the option live; a coal price that stabilises the ¥3.4bn Tianchi contribution rather than eroding it; and capex rolling off so the core stops leaking cash. Until then, the price pays for the core the company has, not the recovery the label implies.

---

Related Parties

A recurring worry sits under the valuation and the receivable book: with a controlling parent (新疆特变电工集团) atop a web of affiliates, how much of TBEA's reported revenue and its ¥19.5bn of trade receivables is genuinely earned from third parties rather than recycled through the group? The audited FY2025 disclosures answer it. Related-party sales were ¥1.15bn — 1.2% of revenue — and zero among the top-five customers; related-party receivables were roughly ¥0.2bn, about 1% of the book. The dependence runs the other way: TBEA buys ¥4.57bn (4.6% of procurement) from parent-group suppliers.

Sources: related-party sales and purchases FY2025 notes [1]; related-party receivables [2]; non-operating fund occupation [3].

Revenue is arm's-length

The customer disclosure is the cleanest read on revenue quality. In FY2025 the five largest customers bought ¥30.9bn (31.77% of sales), of which the related-party portion was ¥0 — zero percent [4]. That is not a one-year result: related-party sales were also zero within the top five in FY2024 (¥9.0bn, 9.23% of revenue) and in FY2022 (¥19.8bn, 20.67%), the peak-polysilicon year [5][6].

Source: major-customer disclosures, FY2022/FY2024/FY2025 annual reports [4][5][6].

The rising top-five percentage overstates concentration: from FY2025 the exchange requires customers under common control to be aggregated, so a group like State Grid's regional companies now counts as one buyer rather than several. Total related-party sales across every affiliate came to ¥1.15bn, up from ¥0.59bn a year earlier but still 1.2% of the ¥97.3bn revenue base [1].

The receivable book follows from the customers. The ¥19.5bn of trade receivables that grew 51% on flat revenue (Cash Conversion) is not an intra-group construction: related-party receivables and contract assets totalled roughly ¥0.2bn gross — the largest single line, a Gurbantunggut desert new-energy developer, was ¥66m — against a ¥19.5bn book [2]. Whatever collection risk sits in that book, it is third-party risk on frontier-market EPC and grid contracts, not receivables owed by the parent.

The dependence runs the other way

Where the group does lean on itself is procurement. Related-party purchases totalled ¥4.57bn in FY2025 — 4.6% of the roughly ¥100bn TBEA spends on goods and services, and up modestly from ¥4.38bn in FY2024 [1]. The audited supplier disclosure frames the same fact: ¥4.15bn (4.14% of procurement) of the top-five suppliers' ¥22.2bn was related-party, against ¥0 of related-party sales in the top-five customers [4].

Source: related-party transaction notes, FY2025 annual report [1].

The concentration is in the parent itself, and it touches the transformer core. TBEA bought ¥2.50bn of oil tanks, copper parts, cores and tower brackets from 特变集团 — 42.53% of its total spend in that category — plus ¥1.22bn of construction and engineering services and ¥0.22bn of transport [7]. In other words, close to half of the fabricated tank-and-copper content that goes into TBEA's transformers is supplied by the parent group rather than made in-house — a real qualifier to the self-sufficient scale that underpins the Grid Franchise. Sales back to the parent were trivial by comparison, ¥0.09bn (0.21% of that transaction class) [7].

Source: major related-party transactions with the first-largest shareholder, FY2025 annual report [7].

The prices are set by disclosed formula rather than by negotiation each time — the tank price, for instance, is steel cost (referenced to named benchmark grades) plus a processing fee — which is what makes the 42.53% dependence defensible even though it is high [7]. The residual concern is not fraud but pricing power: a parent that supplies 42% of a key input has room to move the margin split between listco and parent within the formula, and minority holders rely on the audit and the independent directors to police it.

The items that are clean

Three checks a skeptic would run come back empty. First, fund occupation: at year-end the controlling shareholder, the actual controller and other related parties had ¥0 of non-operating funds tied up in the company, and the year saw ¥0 of new occupation [3]. This matters more than usual because TBEA owns a licensed group finance company (特变电工集团财务有限公司), a common conduit for lending listco cash upstream — but it is a wholly consolidated subsidiary, and the report marks financial business between it and related parties "not applicable" [8].

Second, guarantees run downhill, not up: the parent 特变集团 provides ¥1.07bn of pro-rata guarantees to a TBEA subsidiary's affiliates, and there were no related-party loans (资金拆借) at all [9]. Third, key-management remuneration was ¥33.3m, down from ¥40.2m — modest for a group this size and moving the right way [9].

The texture that remains

The one place the parent uses the listco actively is asset consolidation. During FY2025 特变集团 transferred three businesses into TBEA subsidiaries — a 67%-then-larger stake in a Fangchenggang new-materials company, a 100% engineering-repair unit for ¥46m, and a recycling company for ¥10m — each accounted for as a business combination under common control at appraised or mutually-agreed prices [10]. The sums are small and the direction is inward, but combination-under-common-control prices are negotiated between affiliates rather than tested against an outside bidder, so they are not fully arm's-length in the way a third-party acquisition would be.

Governance sits on top of this as a live, annual process rather than a one-time sign-off. TBEA's independent directors reviewed the 2025 and 2026 daily related-party dealings with 特变集团 in dedicated sessions; the December 2025 opinion states the transactions follow market pricing, do not harm minority shareholders, and "will not cause the company to become materially dependent on related parties" [11]. The numbers support that read on the revenue and balance-sheet side, and qualify it on the supply side.

The evidence points one way: TBEA's reported revenue and receivables are real and third-party — the steady core is earning from customers, not from the group — and the meaningful related-party fact is a ~4.6% procurement reliance concentrated in the parent's transformer components. The read would change if the annual related-party caps started to be filled rather than left largely unused, if related-party receivables climbed off their ~1% base, or if the pricing formulae on parent-supplied components were quietly loosened in the listco's disfavour — each of them a line item to watch in the next report.