Tongwei Co., Ltd. (通威股份): The Feed King Who Built the Silicon Sun
I. Prologue: The Sichuan Fish Pond Paradox
In the spring of 1982, an eighteen-year-old technician at the Meishan Hydropower Bureau in Sichuan strung metal mesh across a slow-moving channel of river water and began raising carp inside it. The concept was straightforward: rather than letting fish disperse across an expansive pond, confine them in a small enclosure and let the natural current oxygenate and flush the water. The following year, to fund the venture, he sold his family's pigs to raise roughly sixty dollars' worth of capital.1 His name was 刘汉元 Liu Hanyuan, and the harvest from that single cage was substantial enough to draw provincial agricultural officials to inspect the site.
Forty-four years later, the enterprise Liu founded produced 384,800 metric tons of high-purity polysilicon and shipped 103 gigawatts of solar cells in a single year—an annual volume capable of meeting the electricity demand of an entire mid-sized European country.23 The company that began with river carp in a wire cage now supplies roughly fifteen percent of the world's solar cells and more than thirty percent of China's domestic polysilicon market.3
Yet today, Tongwei is absorbing heavy losses.
This is the central paradox of the business. In FY2022, 通威股份有限公司 Tongwei Co., Ltd. generated ¥142.42 billion in revenue and ¥25.73 billion in net profit, establishing itself as one of the most profitable clean-energy manufacturers in the world by operating margin.4 Three years later, in FY2025, revenue contracted to ¥84.13 billion and the company posted a net loss attributable to shareholders of ¥9.55 billion—its second consecutive annual deficit and a deterioration of nearly 36% from the prior year's ¥7.04 billion loss.25 Full-year gross margin shrank to 2.70%, down 3.69 percentage points, while net margin dropped to negative 12.96%.2 Through the first half of 2026, the downturn persisted: revenue fell 15.19% year-over-year to ¥34.36 billion, resulting in a net loss of ¥5.12 billion.6
The core thesis under examination. Tongwei operates as the most cost-efficient large-scale process manufacturer in China's solar supply chain through an unusual "dual main business" (双主业) model: a sharply cyclical photovoltaic operation anchored to a steady, recession-resilient livestock and aquaculture feed business. The bullish thesis holds that consistent cash flow from feed preserves the balance sheet while Tongwei's structural cost advantage outlasts competitors in an industry-wide war of attrition. The bearish counterargument—supported by three years of financial statements—is that being the lowest-cost producer in an industry burdened by roughly twice as much manufacturing capacity as end-market demand simply means losing money more slowly than peers, which does not constitute a durable economic moat.
The tectonic shift. In August 2022, at the peak of its earnings cycle, Tongwei broke an unspoken neutrality pact with the downstream module manufacturers it had supplied for years—including 隆基绿能 LONGi Green Energy, 天合光能 Trina Solar, and 晶科能源 JinkoSolar—by bidding aggressively into a state-owned utility module tender.7 The downstream retaliation reshaped the entire sector. Whether that forward integration represented an astute commercial land grab or a miscalculation that undermined Tongwei's core customer relationships is the central strategic question of this story, and the industry's subsequent downturn now provides sufficient operating evidence to evaluate the decision.
The road map. What follows traces seven turning points in that history: the creation of China's aquafeed leader between 1982 and 2004; the chemical expansion into Leshan powered by low-cost hydroelectric arbitrage; the counter-cyclical acquisition of a bankrupt rival's flagship polysilicon plant in 2013; the deployment of fishery-solar integration alongside a disciplined refusal to manufacture modules between 2015 and 2020; the polysilicon supercycle and the fateful August 2022 tender bid; the subsequent supply glut, executive succession, and abandoned Runergy acquisition; and finally the strategic frameworks and stress tests that show whether Tongwei represents an enduring industrial franchise or merely an efficient operator navigating a structurally disadvantaged sector.
The trajectory spans fish feed, chlor-alkali chemistry, a bankruptcy auction, an unprecedented boom, and an industry-wide downturn. It began on a riverbank in Meishan.
Subscribe on LinkedInII. Act I: The Feed King of China: Liu Hanyuan's First Empire
There is a defining moment in commodity businesses when an operator recognizes that enduring profits lie not in the end product, but in the critical input it consumes. For Liu Hanyuan, that insight emerged between 1983 and 1986 while managing river cages of fast-growing carp in Sichuan.
Aquaculture faced hard physical limits. Output was constrained by biology, water rights, and the finite river frontage local authorities would permit an operator to enclose. Expanding production meant navigating another administrative negotiation for every additional ton of fish. Fish, however, required constant feeding. An enterprise supplying manufactured feed could sell to every fish farmer across the province—and eventually nationwide—without acquiring a single additional meter of water rights.
In 1986, at age twenty-two, Liu established the Meishan County Fishery Formula Feed Factory outside his family home, marketing extruded feed under the "Keli" brand.1 At the time, the commercial feed industry scarcely existed in China. Compound aquafeed—nutritionally balanced pellets engineered to float rather than sink and pollute pond bottoms—was an unfamiliar industrial product for Chinese smallholders, who had long relied on mixed farm waste. Liu had to develop the market and its distribution network simultaneously.
The unit economics of that initial plant explain why the model compounded rapidly. Formulated feed converted roughly 1.5 to 2 kilograms of pellets into one kilogram of live fish—a dramatic efficiency improvement over traditional farm waste. For smallholders operating on thin margins, feed conversion efficiency determined whether a harvest yielded a profit or a loss. By offering a quantifiable economic payback, Tongwei expanded through demonstrable arithmetic rather than promotional marketing.
Building the distribution moat. Selling feed to millions of fragmented smallholders across Sichuan, Guangdong, Hubei, and Jiangsu was an operational and logistical challenge rather than a branding exercise. It required constructing feed mills close enough to aquaculture clusters to prevent transportation costs from eroding margins, then establishing township-level dealership networks that extended trade credit, managed local deliveries, and trained farmers on feeding protocols. The model rewarded local mill density and scale while penalizing distant competitors, creating a business protected by freight economics and regional market share.
The founder's approach. Liu avoided the public persona of a charismatic visionary, establishing a multi-decade operating record grounded in engineering solutions to physical bottlenecks rather than financial maneuvering. He built cages when pond space was scarce, and constructed feed mills when river frontage capped cage farming. Later, he acquired a chlor-alkali plant to secure chlorine for polysilicon synthesis, and mounted solar panels over fish ponds when land-use regulations prohibited ground-mounted solar arrays on agricultural land.8 The playbook remained consistent: identify the primary operational bottleneck, then control the upstream asset that resolves it. Yet that approach had a distinct vulnerability that the post-2022 downturn would expose: engineering efficiency cannot insulate a producer from an industry-wide supply glut.
On 20 September 1992, Liu formally established 通威集团 Tongwei Group, consolidated the feed business, and committed ¥10 million of personal capital to construct a facility with 100,000 metric tons of annual capacity—a substantial capital outlay for a private Chinese entrepreneur shortly after the 1989 political retrenchment and Deng Xiaoping's southern tour that reopened the economy to private enterprise.1 The timing underscored Liu's habit of expanding aggressively as soon as a favorable regulatory window opened.
The 2004 listing and the ceiling problem. Tongwei completed an initial public offering on the Shanghai Stock Exchange under ticker 600438 in 2004, securing public equity and institutional borrowing capacity.9 The business went on to become the largest aquafeed manufacturer in the world by volume, a position it still holds.10
The public listing also established a corporate architecture that continues to shape governance scrutiny. The listed entity, 600438, operates under an unlisted family holding company, allowing assets to be incubated privately and transferred between the parent group and public shareholders. Over the following decade, that dual structure facilitated both the company's most lucrative counter-cyclical purchase and some of its most contentious transactions.
Yet aquafeed presented a structural growth ceiling. Although the business generated stable, defensive cash flows protected by freight economics and dealer networks, its addressable market expanded only at the low-single-digit pace of Chinese aquaculture tonnage. Compounding at roughly four percent annual volume growth with returns on invested capital in the high single digits to low teens produced a dependable business, but one with limited reinvestment runway. In his early forties, Liu began searching for a second growth engine with vastly larger scale.
The stability of that initial cash engine proved decisive during the solar crash two decades later. In FY2025, even as Tongwei absorbed steep solar losses, feed sales volume reached a record 6.53 million metric tons, with self-branded feed volume growing 4.22%.2 In the first half of 2026, the agriculture and animal husbandry division generated ¥13.45 billion in revenue, up 7.69% year-over-year, while the photovoltaic division saw revenue contract 26.34% to ¥19.65 billion.11 An agricultural division capable of expanding through a severe solar downturn provided the underlying cash generation that differentiated Tongwei from pure-play solar manufacturers.
How Liu chose to reinvest that agricultural cash flow, beginning in 2006, marked one of the most consequential capital allocation pivots in modern Chinese manufacturing.
III. Act II: The Hydropower & Silicon Bet: Entering Yongxiang (2006–2012)
Southwestern China possesses a distinctive geographical advantage: the Dadu, Jinsha, and Min rivers drop steeply from the Tibetan plateau with immense hydraulic head and water volume. By the mid-2000s, Sichuan had installed far more hydroelectric capacity than provincial transmission lines could carry to the industrial demand centers along the eastern seaboard. During the summer wet season, excess water routinely bypassed turbines because grid operators had nowhere to transmit the electricity. Stranded power was among the cheapest energy in the world, and Liu Hanyuan recognized the industrial arbitrage.
In 2006, Tongwei acquired an equity stake in 四川永祥股份有限公司 Sichuan Yongxiang Co., Ltd., a chlor-alkali chemical manufacturer in Leshan's Wutongqiao District producing polyvinyl chloride and caustic soda.9 To equity analysts covering an agricultural feed business, the transaction appeared to be reckless corporate diversification. To Liu, it represented a chemical pathway into polysilicon.
A fish feed producer with no semiconductor pedigree, no chemical engineering background, and no existing solar customers was buying into a regional PVC manufacturer in Sichuan. Equity markets struggled to identify any strategic coherence because doing so required understanding two unrelated disciplines at once: the chemical synthesis of high-purity polysilicon and the localized economics of stranded hydropower.
The chemistry, in plain language. Solar-grade polysilicon is manufactured using the modified Siemens process (改良西门子法). Metallurgical-grade silicon—essentially smelted, reduced quartz sand that is roughly 99% pure—is reacted with hydrogen chloride to produce trichlorosilane, a volatile liquid. That liquid is repeatedly distilled to extreme purity and then decomposed at roughly 1,100 degrees Celsius onto heated silicon filaments inside a bell-jar reactor, growing into crystalline silicon rods with a purity of 99.9999999% or higher—often referred to as nine to eleven "nines." In practical terms, the process resembles repeatedly distilling a spirit until scarcely one foreign atom per billion remains, then solidifying the refined vapor into high-purity crystal.
The Siemens process requires vast amounts of electricity and depends on three specific elements that a chlor-alkali facility handles as standard operating procedure: chlorine gas, hydrogen gas, and the technical competence to manage hazardous chemical reagents safely. In practical terms, a chlor-alkali operation already provides approximately two-thirds of the chemical handling infrastructure required by a polysilicon plant. Pairing that facility with cheap, stranded hydroelectricity resolved the two largest operating cost components—power and chlorine feedstock—before manufacturing even began.
Siting the facility in Leshan offered a second critical advantage: regulatory zoning. Polysilicon plants are classified as hazardous chemical operations. Trichlorosilane is volatile and reacts violently with atmospheric moisture, while silicon tetrachloride is corrosive and toxic. Developing a synthesis plant requires a designated chemical industrial park equipped with environmental permits, fire-suppression infrastructure, and deep municipal coordination. Wutongqiao District had operated as a chemical manufacturing hub for decades. Tongwei was not merely securing chemical feedstocks; it was securing an operating license that would take a greenfield competitor years to acquire. That permitting barrier remains a fundamental structural shield protecting upstream polysilicon from the uncontrolled influx of capital that periodically overwhelms cell and module manufacturing.
The mania and the crash. Liu's entry coincided with an unprecedented global solar boom. Generous feed-in tariffs in Germany and Spain ignited a demand surge that the four incumbent Western polysilicon producers could not satisfy. Spot prices for solar-grade polysilicon soared from approximately $50 per kilogram to more than $400 per kilogram between 2007 and 2008. Across China, dozens of regional chemical parks rushed to announce polysilicon expansion projects.
The expansion was short-lived. Following the 2008 global financial crisis, European governments slashed solar feed-in tariffs, sending polysilicon prices crashing below $30 per kilogram by 2011 and 2012. The ensuing shakeout devastated the first generation of Chinese solar champions. 无锡尚德 Suntech Power, once the world's leading module manufacturer whose initial public offering made founder 施正荣 Shi Zhengrong China's wealthiest individual, defaulted on its debt and saw its primary operating subsidiary enter bankruptcy in 2013. 江西赛维 LDK Solar, which had borrowed heavily to build massive wafer and polysilicon plants, collapsed under its debt load. 保利协鑫 GCL-Poly survived, but was forced to spend the better part of a decade repairing a damaged balance sheet.
The collapse exposed a structural lesson that still defines the industry. Suntech and LDK were broken primarily by financing structures rather than factory-floor inefficiencies. Both had funded massive expansions with short-dated bank borrowing while signing rigid, long-term take-or-pay polysilicon supply contracts at peak market prices. When spot prices collapsed, both firms were legally obligated to buy expensive raw silicon to manufacture panels they could sell only at catastrophic discounts. The fatal vulnerability was not their manufacturing cost, but the toxic combination of debt leverage and inflexible procurement contracts. Tongwei, by contrast, synthesized its own silicon rather than locking into fixed off-take agreements, and its agricultural feed division provided steady, uncorrelated cash flow to service its obligations.
What Tongwei actually did, and what it cost. Yongxiang did not emerge unscathed. Between 2009 and 2012, its polysilicon division posted heavy operating losses and consumed substantial corporate capital. The aquafeed business effectively subsidized the solar unit throughout the downturn. This period provided the first real stress test of the "dual main business" strategy: without the steady earnings and operating cash generated by animal feed, Yongxiang likely would have faced debt restructuring or insolvency alongside Suntech and LDK. The dual model served as an authentic cross-cycle balance-sheet buffer.
Tongwei used the breathing room afforded by that agricultural cash flow to refine its production processes. The fundamental engineering challenges in the Siemens process involve closed-loop recycling: silicon tetrachloride, a hazardous byproduct, must be efficiently converted back into trichlorosilane rather than neutralized as waste; exhaust heat from deposition reactors must be captured; and overall power consumption per kilogram of output must fall. Western incumbents—including Hemlock Semiconductor, Wacker Chemie, and REC Silicon—had long doubted that Chinese entrants could achieve electronic-grade purity while operating at competitive energy conversion rates. Disclosures from FY2025 demonstrated how far the manufacturing metrics had advanced: Tongwei reported unit silicon consumption of 1.03 kilograms per kilogram of output, a cold hydrogenation conversion rate of 32.5%, and reduction-step power consumption at or below 40 kilowatt-hours per kilogram, compared to historical industry baselines that routinely exceeded 70 kilowatt-hours.3
The lesson Liu drew from the initial solar winter shaped Tongwei's operating philosophy for the decade that followed: in cyclical commodity manufacturing, the lowest-cost survivor of a downturn inherits market share during the subsequent recovery. It was a principle he would put to the test the following year during a distressed-asset auction.
IV. Act III: The Masterstroke of the Solar Winter: The Hefei LDK Acquisition (2013–2015)
By late 2012, LDK Solar was carrying more than $3 billion of debt and could no longer service its obligations. Among the assets imperiled by the collapse was 合肥赛维 LDK Hefei, a modern manufacturing base in Anhui province with roughly 1.6 gigawatts of solar cell capacity and 500 megawatts of module capacity, built at a capital cost that ran into billions of renminbi. The Hefei High-Tech Zone administration had intervened to keep the facility operational, primarily to preserve thousands of manufacturing jobs in a municipal economy that had anchored its industrial policy to advanced manufacturing.
From an engineering perspective, the plant was modern and structurally intact. Financially, it had become virtually unmarketable. By 2013, demand for solar cell capacity had evaporated. Module prices had fallen below cash cost across the sector, the European Union was conducting an anti-dumping investigation into Chinese photovoltaic imports, and every potential buyer with relevant operational expertise was fighting for solvency.
In September 2013, 通威集团 Tongwei Group—the unlisted parent company, rather than the publicly traded vehicle—acquired 100% of Hefei LDK for ¥870 million.12
Why the price mattered. At roughly ¥0.54 per watt of installed cell capacity, Tongwei acquired the facility at a steep discount to greenfield construction costs. Crucially, it purchased a turnkey plant that was already permitted, staffed, and integrated into regional electrical grids and transport logistics. Greenfield expansion typically required eighteen months and a speculative bet that market prices would recover before production lines ramped. A distressed acquisition converted a market-timing gamble into an operational execution problem—a far more manageable risk profile for a disciplined industrial operator.
Yet the transaction carried two important caveats. First, Tongwei executed the purchase through the unlisted parent holding company, excluding minority shareholders of the publicly traded entity (600438) from the initial distressed valuation. Second, the company had virtually no prior cell manufacturing experience. Tongwei was an agricultural feed producer that had mastered chlor-alkali chemistry; it had never operated a cleanroom production line where margins depend entirely on microscopic conversion yields.
The turnaround. The operational turnaround that followed relied on rigorous continuous-process engineering rather than proprietary technological breakthroughs. Plant managers overhauled equipment lines, curtailed consumable and auxiliary material expenses, renegotiated silicon wafer procurement terms, and systematically lifted cell conversion efficiency fraction by fraction. Cell manufacturing ultimately rewards the same core competencies as commercial feed milling: high operational throughput, tight yield control, line uptime, and fractional unit-cost reductions. Within roughly eighteen months, Hefei Tongwei generated positive operating cash flow.
This operational turnaround illustrates a subtle structural advantage. While commercial feed milling and photovoltaic cell fabrication share no technological overlap, their operational management systems are close cousins. Both are high-volume, continuous-process businesses where a fractional variance in yield, an hour of unplanned equipment downtime, or a few wasted grams of consumable input per unit compounds across millions of units into the difference between commercial viability and steep losses. Tongwei's transferable institutional capability was high-volume process discipline. That manufacturing rigor explains why the Hefei turnaround succeeded without prior solar credentials, and why Tongwei's core production costs remained among the lowest in the sector even during subsequent industry downturns.
The 2016 asset injection and the governance question. Between 2015 and 2016, the listed company acquired these incubated solar assets from the parent group in two tranches: 永祥股份 Yongxiang and related clean-energy operations in the first, and 100% of Hefei Tongwei in the second. Hefei Tongwei was valued at approximately ¥5 billion, paid through the issuance of roughly 920 million new public shares.13
The transaction highlighted the inherent governance tension within Tongwei's corporate structure. The unlisted parent had purchased the Hefei facility for ¥870 million in 2013 and transferred it to the public entity three years later at an appraisal of roughly ¥5 billion—representing a 5.7-fold increase in asset valuation on a related-party transaction where the counterparty was the controlling shareholder.
The commercial rationale had genuine operational merit: the asset injected in 2016 was fundamentally different from the distressed liquidation acquired in 2013. Annual production capacity had expanded to multi-gigawatt scale, operating lines were running at high utilization rates, and the business was generating consistent profits. The transaction satisfied independent valuation benchmarks, received regulatory clearance from the China Securities Regulatory Commission, and secured approval from public minority shareholders. Nevertheless, while the underlying asset had demonstrably improved, the controlling shareholder captured virtually the entire distressed-purchase arbitrage. Public shareholders acquired a productive, profitable asset at full market valuation, but the initial windfall accrued directly to the family holding company. This dynamic established a precedent that remains central to assessing Tongwei's corporate governance, particularly regarding how future asset consolidations are structured.
What the injection did accomplish was assembling the two essential pillars of Tongwei's photovoltaic strategy inside a single publicly traded entity: low-cost upstream polysilicon and high-volume downstream cell manufacturing, supported by the steady cash flow of the agricultural feed business. The strategic question that followed was how Liu would link that manufacturing apparatus back to his original agricultural empire.
V. Act IV: The "Fishery & PV Integration" Dual-Drive Engine (2015–2020)
China's renewable energy expansion faced an underlying geographic mismatch. Electricity demand was concentrated in the industrialized eastern and central provinces, while cheap, open land lay thousands of kilometers away in the arid west and north. Compounding the constraint, Beijing strictly enforced the 基本农田红线—the basic farmland red line—a statutory floor on arable land designed to safeguard national food security by prohibiting industrial conversion. For solar developers seeking to build near coastal load centers, suitable ground was either prohibitively expensive or legally off-limits.
Liu Hanyuan addressed the bottleneck through his experience in commercial aquaculture. Eastern and central China contained millions of acres of commercial fish ponds. These ponds were classified separately from protected agricultural acreage, and their water surfaces were largely unutilized.
渔光一体 (Fishery and PV Integration). Tongwei mounted solar panels on concrete pilings three to four meters above the water. The elevated arrays generated grid-connected electricity sold into regional grids, while the panels partially shaded the water below, moderating summer water temperatures and suppressing algae blooms. Tongwei bundled the installations with its agricultural business, selling aquafeed and aquatic care products directly to the fish farmers operating beneath the panels.
The model provided an inventive commercial synergy by monetizing a single parcel through electricity generation, aquaculture feed sales, and pond leasing. As of FY2025 disclosures, cumulative grid-connected fishery-PV and related power plant capacity reached 5.11 GW.3 That asset base provided predictable, high-margin utility cash flow insulated from factory-gate hardware pricing.
Yet despite being featured prominently in corporate presentations for over a decade as a transformational platform, fishery-PV remained a peripheral earnings contributor rather than a primary growth engine. Capacity expanded gradually rather than exponentially, and China's gradual transition from guaranteed feed-in tariffs toward market-based power trading compressed project-level returns. The model proved to be an effective marketing vehicle and a modest balance-sheet stabilizer, but its earnings contribution remained secondary to Tongwei's heavy manufacturing operations.
The "Switzerland" strategy. The far more consequential strategic choice of this period was what Tongwei chose not to do. Between 2016 and 2020, as peers aggressively integrated downstream into finished module manufacturing to sell branded panels directly to project developers, Tongwei explicitly pledged that it would not enter module assembly. Instead, it positioned itself as an independent, neutral merchant cell supplier to the entire industry.
The operational logic rested on value-chain dynamics. Solar cell fabrication was the most technically exacting, capital-intensive processing phase of the crystalline silicon supply chain—the stage where conversion efficiency and micro-yields determined operating profitability. Module manufacturing, by contrast, was largely an assembly process involving glass, aluminum frames, encapsulant films, and junction boxes, paired with substantial working capital requirements, twenty-five-year warranty liabilities, and heavy global sales overhead. By establishing itself as the low-cost manufacturing partner to LONGi, JinkoSolar, JA Solar, and Trina Solar simultaneously, Tongwei captured upstream manufacturing value without incurring customer-facing overhead or receivables risk. Because Tongwei maintained strict commercial neutrality and did not compete downstream, module makers had little incentive to sink scarce capital into duplicative cell lines.
The PERC transition. The technology driving this expansion was passivated emitter and rear cell, or PERC. In a standard solar cell, a portion of incoming light passes straight through the silicon wafer without generating electricity, while free electrons frequently recombine at the unpassivated rear surface. PERC introduced a dielectric passivation layer on the cell's rear side that reflected unabsorbed light back through the silicon for a second chance at absorption while suppressing surface recombination. In commercial production, PERC delivered roughly one full percentage point of absolute conversion efficiency for a modest increase in capital equipment. In an industry where costs and margins are calculated to the tenth of a cent per watt, that efficiency gain spread across gigawatts of volume was decisive. PERC established itself as the photovoltaic industry's dominant cell architecture for nearly a decade before TOPCon technology superseded it.
Tongwei executed the transition with exceptional speed. It constructed multi-gigawatt manufacturing clusters in Shuangliu, Jintang, and Meishan, standardized its production lines on PERC, and drove down unit production costs through high throughput and process discipline. By 2018, Tongwei had become the world's largest merchant cell manufacturer—a position it held for nine consecutive years, maintaining approximately 15% of global cell market share.3
In business strategist Hamilton Helmer's terminology, this posture represented classic counter-positioning. Tongwei adopted a customer-neutral, merchant-focused operating model that integrated competitors could not replicate without undermining their own branded downstream operations. For five years, the merchant strategy functioned smoothly, generating robust cash flow while insulating Tongwei from module-market distribution battles.
Then polysilicon prices soared, and Tongwei decided to abandon its neutrality.
VI. Act V: The Polysilicon Supercycle & The 2022 Module Shock
In September 2020, Chinese President Xi Jinping announced at the United Nations General Assembly that China would peak carbon emissions before 2030 and achieve carbon neutrality by 2060. The 双碳 (dual carbon) goals transformed Chinese solar manufacturing from an export-reliant sector into a primary domestic policy imperative overnight, propelling annual global installations from roughly 140 gigawatts in 2020 to approximately 260 gigawatts by 2022.
The supply chain could not absorb the surge, driven by a structural mismatch in construction timelines that underlies every cyclical boom and bust across the industry.
The bottleneck, explained. Wafer, cell, and module facilities are fundamentally equipment-assembly operations. An operator can order automated production tools from domestic equipment vendors, install them in standard industrial facilities, and begin commercial production within six to nine months. A polysilicon plant, by contrast, is a heavy chemical manufacturing facility. It requires environmental and hazardous-materials permits, complex pressure-vessel engineering, a two-year construction and commissioning cycle, and an extended ramp period to stabilize yields. When end-market demand doubles, three of the four primary stages of the value chain can respond within a single year; polysilicon cannot.
Polysilicon consequently became the defining bottleneck of the global energy transition. Spot prices surged fivefold, climbing from roughly ¥60 per kilogram in early 2020 to more than ¥300 per kilogram by mid-2022.
That construction asymmetry also generated a severe second-order consequence. Because polysilicon facilities require two years to build and the initial demand surge lasted roughly two years, the wave of capacity commissioned at the peak of the shortage came online precisely as supply caught up with demand. The dynamic mirrored the classic agricultural hog cycle, structurally embedded in chemical engineering schedules. Any cyclical recovery in the sector tends to trigger an expansion that overshoots on that same two-year lag: the resulting glut is not a regulatory failure, but a mechanical byproduct of the construction timeline.
The impact on Tongwei's income statement. Tongwei had spent the preceding decade assembling precisely the asset class that had become scarce, at a unit cost that maximized scarcity rents. Its cash production costs sat in the range of ¥35 to ¥40 per kilogram, supported by standardized 100,000-metric-ton production lines across Leshan in Sichuan, Baotou in Inner Mongolia, and Baoshan in Yunnan, paired with preferential industrial electricity tariffs. With spot prices surpassing ¥300 per kilogram, gross margin on polysilicon climbed past 85%.
For FY2022, revenue expanded 119.7% to ¥142.42 billion, while net profit grew 217.2% to ¥25.73 billion.4 Tongwei produced roughly 1.6 times as much solar polysilicon that year as the world's second-largest manufacturer.14 Return on equity exceeded 50%.
Yet an 85% gross margin on an undifferentiated industrial commodity reflects temporary shortage rents rather than an impenetrable economic moat. The distinction proved decisive. Provincial governments, private equity funds, and industrial conglomerates across China studied those public earnings and directed capital into polysilicon synthesis. Across 2021 and 2022, more than ¥100 billion in capital poured into domestic polysilicon and next-generation cell lines. Tongwei expanded aggressively into the same upswing—a counterpoint to its historical reputation for counter-cyclical discipline: the management team that had acquired Hefei LDK at the bottom of the cycle also expanded aggressively at its peak.
In historical context, Tongwei's FY2022 net profit exceeded the cumulative earnings of its first eighteen years as a publicly listed enterprise by a wide margin. That windfall illustrated the financial leverage of controlling an upstream choke point during a demand shock. It also explained the scale of the subsequent capital response: an 85% gross margin published in an audited annual report served as an unmistakable commercial signal that sector returns far exceeded perceived operational risks. In commodity processing, extraordinary reported profitability is rarely an enduring equilibrium; it is an open invitation to new capital.
17 August 2022: the day the industry changed. State-owned utility 华润电力 China Resources Power published the bidding results for the fifth batch of its centralized photovoltaic module procurement. Tongwei—which historically had avoided commercial module manufacturing—placed first with a bid of ¥1.942 per watt for a 3-gigawatt package of 545-watt modules valued at approximately ¥5.83 billion, scheduled for delivery between 10 September and 10 December 2022. In winning the contract, Tongwei undercut bids from both 亿晶光电 EGing PV and LONGi.715
Public markets grasped the structural shift immediately. Shares of Chinese module manufacturers dropped sharply on the announcement: their dominant merchant cell supplier had transformed into a direct rival in premier utility tenders.16
The retaliatory cascade. The downstream response arrived swiftly. LONGi, Trina Solar, and JinkoSolar accelerated investments in captive cell manufacturing lines to diminish their operational dependence on Tongwei. Tongwei responded by unveiling ambitious module capacity expansions, completing its transition into full vertical integration from polysilicon synthesis to finished panel assembly.
From a strategic perspective, the move traded a defensible, customer-neutral position for an integrated posture burdened by weaker industry economics. The counter-positioning advantage Tongwei had maintained over the prior five years did not erode naturally; the company voluntarily abandoned it. Crucially, Tongwei made that transition just as industry-wide capacity expansions were preparing to outstrip demand. The company committed capital to downstream module lines during an emerging supply glut while simultaneously giving its largest merchant cell buyers an urgent strategic incentive to internalize production.
Management consistently defended the pivot by arguing that vertical integration across the value chain had become inevitable—that downstream module manufacturers were already building captive cell capacity, threatening to leave Tongwei stranded as a merchant supplier without off-takers. That rationale carried practical weight. Yet it amounted to an admission that the merchant model was becoming untenable, rather than proof that capital allocation into module assembly would generate attractive economic returns. Four years into that strategic pivot, operating results provide substantial evidence to assess the outcome.
The transition also introduced a notable shift in corporate messaging. Throughout 2021 and early 2022, Tongwei's investor communications had explicitly highlighted its identity as a specialized supplier that chose not to compete with its downstream clients, framing that neutrality as a primary competitive moat. Management abandoned that stance without reconciling the pivot against its prior commitments. While corporate strategies must adapt to changing market conditions, unaddressed reversals in fundamental strategic positioning introduce governance friction and complicate investor assessment of long-term capital allocation plans.
The polysilicon supercycle reached its zenith in the second half of 2022. The downturn that followed was not a routine cyclical correction, but an industry-wide war of attrition.
VII. Act VI: The Great Solar Glut, Leadership Succession, & The Runergy Stress Test
On 19 March 2023, 谢毅 Xie Yi, the chairman and chief executive who had led Tongwei through the supercycle, resigned for personal reasons.17 Two days later, the board elected his successor: 刘舒琪 Liu Shuqi, born in 1989, educated at Queen Mary University of London, previously a company director, assistant president, and general manager of the photovoltaic business department—and Liu Hanyuan's daughter.18
She was thirty-four. She had managed raw-material procurement and the photovoltaic business unit during years when soaring polysilicon prices generated record corporate earnings—an operating environment in which almost any producer appeared exceptionally profitable. She assumed leadership precisely as the cycle turned.
The succession drew immediate market skepticism, centered less on technical competence than on governance authority: whether a founder's daughter, installed in her mid-thirties, possessed the institutional standing to idle production lines, cancel capital projects, and push back against family-driven expansion targets. Three years of operating history provide an empirical record. While losses widened in each of her first two full fiscal years, she also demonstrated the discipline to abandon an active cross-border acquisition. She subsequently guided the company toward industry-wide capacity rationalization structured on terms distinct from the holding company's historical distressed transactions.
The leadership transition also carried a distinct structural implication. Xie had been the executive most closely associated with Tongwei's rapid capacity expansion and its downstream entry into module assembly. His departure occurred five months after the China Resources Power tender and roughly one quarter after polysilicon prices crested. Tongwei characterized the resignation as personal, and regulatory disclosures contain no contradictory filings.17 The practical consequence was that the executive who executed the supercycle's defining strategic pivot departed before absorbing its aftermath. While an incoming executive often uses succession to reassess or reverse prior commitments, Liu Shuqi chose not to alter the course. Tongwei continued expanding module capacity and pursuing full vertical integration rather than retreating to its former merchant model.
The mechanics of the collapse. Global solar installations continued to expand, reaching roughly 450 gigawatts in 2023 and climbing further in 2024. End-market demand was not the bottleneck. Rather, Chinese manufacturing capacity across the supply chain surpassed 1,000 gigawatts. By mid-2026, the sector was carrying an estimated 1,100 gigawatts of annual production capacity against approximately 536 gigawatts of global demand.19
Polysilicon spot prices plunged from more than ¥300 per kilogram to between ¥33 and ¥38 per kilogram by mid-2024, falling below the fully loaded production cost of even the sector's lowest-cost producers. Module prices broke below ¥0.70 per watt. The industry entered a prolonged phase where virtually every participant was selling below total cost.
The persistence of that uneconomic supply illustrates a fundamental paradox of capital-intensive manufacturing. In facilities with heavy fixed capital, operating decisions are dictated by marginal cash cost rather than fully absorbed accounting cost. As long as market prices cover variable cash expenses, running a plant burns less liquidity than idling it, because an idle facility continues to incur depreciation, maintenance, interest charges, and substantial restart expenses. Because competing manufacturers face the identical financial logic, supply fails to clear naturally even when the entire sector absorbs heavy losses. This coordination breakdown made industry-wide capacity retirement proposals attractive across the sector in 2025—and explained why regulators scrutinized the initiative as a potential cartel in 2026.24
FY2024 and FY2025: the numbers behind the narrative. For FY2024, Tongwei's revenue fell 33.87% to ¥91.99 billion, producing a net loss of ¥7.04 billion—its first annual deficit since listing on the Shanghai exchange in 2004.520 Operating volume, however, continued to rise: polysilicon shipments climbed 20.8% to 467,600 metric tons, cell shipments rose 8.7% to 87.68 gigawatts, and module shipments jumped 46.9% to 45.71 gigawatts.5 Operational throughput remained robust, but pricing power had collapsed.
FY2025 was more financially punishing, though different in operational structure. Revenue contracted a further 8.55% to ¥84.13 billion, and the net loss widened 35.73% to ¥9.55 billion, with the net loss excluding non-recurring items reaching ¥9.88 billion.2 Polysilicon output was deliberately curtailed—falling 17.71% to 384,800 metric tons as Tongwei throttled operating rates—yet domestic market share still climbed two percentage points to top 30%, reflecting even deeper production cuts by higher-cost rivals.3 Cell shipments expanded 17.51% to 103.03 gigawatts, while module shipments reached 43.25 gigawatts, ranking fifth globally according to industry tracker InfoLink.321
Two dimensions of the 2025 financial data warrant attention. First, operating cash flow remained positive at ¥1.38 billion despite the ¥9.55 billion net accounting loss, supported by substantial non-cash depreciation on recently completed facilities alongside working-capital releases.2 Second, quarterly losses narrowed sequentially over the course of the year. Following net losses of approximately ¥2.6 billion in the first quarter of 2025 and ¥2.4 billion in the second quarter, the third-quarter deficit compressed to approximately ¥300 million as polysilicon spot prices rose roughly 20% sequentially, and management stated that the polysilicon business achieved operating profitability in the second half.2223 For investors assessing credit risk, positive operating cash flow and sequential loss compression distinguished a severe cyclical downturn from an immediate solvency event.
The Runergy stress test. In August 2024, Tongwei announced an intention to acquire a controlling stake of at least 51% in 江苏润阳新能源 Jiangsu Runergy New Energy, proposing to invest up to ¥5 billion through a capital increase and share transfer.25
The transaction addressed Tongwei's foremost structural vulnerability. Runergy maintained operational cell and module facilities in Thailand and Vietnam—a Southeast Asian production footprint that, on paper, offered an export conduit into the United States. Trade barriers, including the Uyghur Forced Labor Prevention Act, Section 301 tariffs, and anti-circumvention duties, effectively barred mainland Chinese production from the American market, which remained the world's most lucrative solar pricing environment. Constructing comparable overseas manufacturing facilities greenfield would have required years of capital commitment and regulatory approvals.
Due diligence, however, uncovered severe financial and operational distress. Runergy carried heavy short-term debt, faced commercial litigation in the United States, and had idled substantial portions of its Southeast Asian capacity as American trade authorities broadened tariff investigations across the region.
On 14 February 2025, Tongwei terminated the transaction, stating that after financial audits, legal reviews, and multiple rounds of negotiations, the parties were unable to agree on commercial terms.262728 Runergy subsequently restructured through a debt-for-equity swap with its equipment suppliers.
Weighing the outcome. Walking away from an unviable transaction reflects managerial discipline, particularly given the sunk-cost pressure that builds over six months of active due diligence and the common pitfall of pursuing opportunistic empire-building during an industry trough. Yet two constraints qualify that assessment. First, Tongwei was itself heading toward an annual net loss of more than ¥7 billion when it made the decision, meaning balance-sheet constraints, rather than managerial discipline alone, limited its ability to finance the transaction.27 Second, signing a public intention agreement before verifying whether the target's debt maturities and litigation liabilities were manageable created avoidable deal friction—a cost that accompanied publicizing an acquisition framework ahead of comprehensive diligence.
The enduring operational consequence is that Tongwei entered late 2026 with virtually no manufacturing presence outside mainland China, leaving it with minimal direct access to the premium-priced U.S. market. That geographic concentration remains the most significant structural divergence between Tongwei and internationally diversified peers such as JinkoSolar and Canadian Solar, with substantial implications for long-term margins.
VIII. Business Architecture, Segment Economics & Capital Allocation
Strip away the narrative, and Tongwei presents an unusual industrial architecture: four distinct businesses stacked in a single corporate structure.
High-purity polysilicon (高纯晶硅). With annual nameplate capacity of roughly 900,000 metric tons, Tongwei is comfortably the world's largest polysilicon producer.3 Its output concentrates on N-type material—the ultra-pure chemical grade required for the TOPCon cell architecture that now dominates the market. Disclosed operating metrics for FY2025 remain industry-leading: unit silicon consumption of 1.03 kilograms per kilogram of output, reduction-step power consumption at or below 40 kilowatt-hours per kilogram, and a cold hydrogenation conversion rate of 32.5%.3 Yet the unit economics of polysilicon synthesis depend almost entirely on two variables: power tariffs and plant utilization. When electricity costs are low and capacity runs at full tilt, polysilicon represents the most lucrative phase of the solar value chain. When market prices fall below cash production costs, however, heavy fixed depreciation turns the segment into the company's heaviest source of operating losses.
Solar cells (太阳能电池). With more than 150 gigawatts of N-type capacity, Tongwei has maintained its position as the world's largest cell manufacturer for nine consecutive years, and in the first half of 2026 it became the first company in the industry to surpass 400 gigawatts of cumulative cell shipments.11 Disclosures for FY2025 showed unit power consumption declining 13% to 41,900 kilowatt-hours per megawatt and water usage falling 23%, with non-silicon processing costs dropping 14% excluding the impact of silver prices.3 That silver qualification is critical: silver metallization paste represents the second-largest cost input in cell fabrication after the silicon wafer, and a sharp global price rally across 2025 and 2026 was the primary reason gross margin in the first quarter of 2026 fell to negative 4.03%, worsening from the prior-year period.3 It serves as a reminder that holding the lowest conversion cost relative to peers offers no structural immunity to raw-material inflation.
One operational dynamic in the cell division warrants closer scrutiny. Tongwei's cell shipments expanded in every year of the downturn—rising 8.7% in 2024 and 17.51% in 2025 to reach 103.03 gigawatts—even after its merchant relationships with tier-one module makers were ostensibly ruptured.53 That volume resilience is revealing. It indicates either that downstream retaliation was less comprehensive than the 2022 narrative implied, or that Tongwei's cost and scale advantages remain compelling enough that competitors continue buying its cells regardless. Either interpretation reinforces the company's continuous-process manufacturing power and softens, without dismissing, the strategic friction created by its entry into module assembly.
Modules (高效组件). Scaling from zero in 2022, Tongwei's module operation ranked fifth globally by volume in 2025.21 The most significant development in this segment is geographic: overseas module shipments expanded 164% in FY2025 to 9.52 gigawatts, driven by volume growth of 370% in the Middle East and Africa, 279% in the Asia-Pacific region, and 129% in Europe, while overseas cell sales grew 73% to 16.73 gigawatts.293 By the first half of 2026, international shipments represented roughly 40% of module revenue, matching the strategic target management outlined at its May 2026 briefing.1129 This export expansion represents the company's most direct response to domestic price erosion, shifting volume into markets where Chinese state-owned utility tenders do not establish pricing floors.
Agriculture and animal husbandry. Animal and aquaculture feed reached record sales volumes of 6.53 million metric tons in FY2025, generating ¥13.45 billion in revenue during the first half of 2026 alone—a 7.69% year-over-year increase supported by a 27.57% surge in overseas feed sales as the company pushed into Romania, Poland, and Southeast Asia.211 Generating roughly one-quarter to one-third of consolidated revenue, the agricultural division stands out as the company's only business segment currently delivering dependable operating profit.
Capital allocation, assessed against the record. Between 2020 and 2024, Tongwei deployed massive capital expenditure across polysilicon synthesis, cell fabrication, and module assembly. The resulting track record presents a sharp dichotomy.
The company's record in distressed acquisitions is demonstrably strong: the Hefei facility was acquired at a deep cyclical discount and turned around into a profitable asset, even though the initial valuation arbitrage accrued to the private holding company rather than public minority shareholders. The track record for organic capacity expansion is more conflicted. Operational execution was disciplined—Tongwei built facilities faster, cheaper, and with higher conversion yields than competitors, cementing its cost crown. Yet the capital timing was distinctly pro-cyclical. Tongwei expanded aggressively into the 2021–2022 boom alongside the rest of the industry, and its own multi-gigawatt additions directly enlarged the manufacturing glut that subsequently depressed corporate earnings. While management has openly cited the sector's "periodic overcapacity" (阶段性产能过剩), its investor disclosures do not acknowledge the company's own role in creating that supply overhang—a gap in corporate accountability that investors must weigh when evaluating forward guidance.
The balance sheet reflects that capital outlay. By the end of March 2026, total assets stood at ¥186.99 billion against total liabilities of ¥138.81 billion, lifting the debt-to-asset ratio to approximately 74.2%, up from the end of 2025.6 Cash and cash equivalents stood at ¥21.27 billion, down from ¥33.23 billion in cash and trading assets a year earlier.322 Outstanding bond liabilities totaled ¥17.6 billion at the time of the mid-2026 profit warning.30 For a capital-intensive manufacturer navigating the third year of an industry price war, a leverage ratio in the mid-seventies does not represent an imminent solvency crisis, but it leaves little operating buffer and substantially narrows management's strategic room to maneuver. Credit rating agencies have maintained their investment-grade assessments through the downcycle, though the upward trajectory of leverage remains the critical variable to monitor rather than the headline ratio alone.31
The activist stress test. A skeptical institutional investor would target four core vulnerabilities. First, the governance history: a controlling family that captured a large mark on the company's premier operating asset before injecting it into the listed vehicle, and subsequently pursued further asset consolidations partly financed through equity during a depressed share-price environment. Second, disclosure framing: Tongwei's investor presentations highlight volume milestones, patent counts, cell efficiency gains, and market share—metrics that have indeed improved—while financial results have deteriorated across three consecutive fiscal years without a prominent accounting reconciliation in primary corporate presentations. Third, portfolio coherence: maintaining that the agricultural feed business provides an essential balance-sheet anchor while continually directing corporate capital into a loss-making solar expansion prompts the question of whether the two units belong in the same public entity, or whether defensive agricultural cash flows are merely subsidizing an industrial capital sink. Fourth, balance-sheet strain: rising leverage through a prolonged downcycle paired with the lack of an overseas manufacturing base to insulate export margins from mounting Western trade barriers.
Counterarguments exist for each point, and none is fatal on its own. Yet an investor who has not formulated a clear stance on all four is not underwriting an industrial manufacturing business; they are simply taking an unhedged directional view on polysilicon spot prices.
Ownership and incentives. 通威集团有限公司 Tongwei Group holds 43.85% of the listed company; Liu Hanyuan holds an 80% stake in Tongwei Group and serves as the ultimate controller, commanding a combined direct and indirect equity holding of roughly 45%.32 Liu stepped down from formal corporate offices at the listed vehicle in 2019 while retaining voting control.18 In January 2024, Tongwei Group announced an open-market share acquisition plan of ¥1.0 billion to ¥2.0 billion—a clear alignment signal that a controlling shareholder with a 45% position can afford to make.33
The resulting governance structure is simple and highly concentrated: a family-controlled corporation where the founder's daughter manages day-to-day operations and the founder controls the voting equity. That architecture eliminates standard principal-agent friction between executive management and the controlling owner, but it heightens the structural tension between the controlling family and minority public shareholders. The 2016 Hefei asset injection established the historical precedent that illustrates that governance risk; the proposed Lihao transaction serves as the active test of whether capital allocation discipline has genuinely shifted.
IX. Strategy Playbook: Porter's 5 Forces & Hamilton Helmer's 7 Powers
Evaluating Tongwei through Michael Porter's Five Forces framework reveals an enterprise that excels at operational cost discipline while remaining structurally exposed across nearly every external commercial dimension.
Buyer power: extreme. The dominant domestic buyers of solar modules are China's state-owned power generation groups—the "big five and four small" power utilities, including 国家电力投资集团 State Power Investment Corporation (SPIC) and 中国华电集团 China Huadian Corporation. They procure through centralized, sealed-bid reverse auctions. The August 2022 China Resources Power tender is the archetype: a single procurement event with a lowest-price shortlist and no relationship premium carried into subsequent bidding rounds. When procurement is dominated by state-owned enterprises running blind auctions, price discovery mechanically converges on marginal cash cost. This dynamic exerts relentless downward pressure on domestic margins, providing the primary commercial impetus for Tongwei's push into international export markets.
Supplier power: moderate, but with a live problem. Metallurgical-grade silicon feedstock is abundant across Xinjiang, Yunnan, and Sichuan. Electricity, which accounts for roughly 30% to 40% of polysilicon cash costs, is the critical variable. Because power rates are established through negotiated agreements with local grid operators rather than an open spot market, Tongwei's cost advantage depends partly on provincial relationships that remain subject to renegotiation. Silver represents the other pressing input challenge: silver metallization paste cost inflation directly compressed cell gross margins through 2025 and 2026, which explains why Tongwei's heterojunction (HJT) technology roadmap prioritizes reducing silver consumption below 4 milligrams per watt.293
Threat of new entrants: bifurcated across the value chain. Polysilicon synthesis presents substantial barriers to entry: hazardous-chemical permitting, pressure-vessel engineering, a two-year construction timeline, and multi-billion-renminbi capital outlays to achieve minimum efficient scale. Solar cell and module assembly, by contrast, present minimal barriers, because turnkey equipment vendors such as 捷佳伟创 S.C New Energy Technology and 迈为股份 Maxwell Technologies supply fully automated lines to any entity with capital financing. Tongwei's most defensible asset base sits upstream, whereas its largest volume operations are downstream. This structural asymmetry—the direct result of the 2022 vertical integration decision—dilutes overall capital efficiency by shifting resources toward the least protected tiers of the value chain.
Threat of substitutes: low at the industry level, real at the process level. At the macro level, utility-scale solar remains the lowest-cost source of new power generation across most global markets, leaving it with no immediate substitute at scale. Within polysilicon processing, however, 协鑫科技 GCL Technology's fluidized bed reactor (FBR) granular silicon (颗粒硅) consumes substantially less electricity than the modified Siemens process. If granular silicon consistently achieves electronic-grade N-type purity at scale, it would erode the specific thermal and power cost advantages Yongxiang spent eighteen years refining. In cell fabrication, perovskite-silicon tandem architectures threaten crystalline silicon's theoretical efficiency ceiling over a ten-year horizon. Tongwei has deployed a fully automated 5-megawatt tandem pilot line and achieved a small-area tandem cell efficiency of 34.94%—an initiative that functions simultaneously as a technological hedge and an operational acknowledgment of crystalline silicon's physical limits.29
Rivalry: severe. Pricing across polysilicon, cells, and modules has remained below fully absorbed cash costs for three consecutive years, with no individual participant possessing the market power or incentive to exit production unilaterally.
Applying Hamilton Helmer's 7 Powers framework.
Scale economies — dominant and verified. Operating the world's largest polysilicon production clusters with single-site capacities reaching 100,000 metric tons allows Tongwei to amortize fixed overhead, maintenance, and chemical-recovery systems across hundreds of thousands of tons of output. The competitive impact is reflected in market-share data: in FY2025, even as Tongwei curtailed its own polysilicon output by 17.71%, its domestic market share expanded by two percentage points to top 30%, because higher-cost competitors were forced into even steeper operating cuts.3
Process power — strong and institutionalized. Eighteen years of continuous closed-loop refinement at Yongxiang have generated proprietary operating efficiencies that peers struggle to replicate. While the 4,070 authorized photovoltaic patents the company accumulated by mid-2026 reflect substantial research activity, formal intellectual property in commodity solar manufacturing provides limited protection against fast followers.11 The stronger evidence of process power lies in physical conversion metrics: standardized unit silicon consumption of 1.03 kilograms per kilogram of output and cold hydrogenation conversion rates that benchmark industry efficiency.
Counter-positioning — extinguished. Historically the company's most effective strategic posture, merchant neutrality was voluntarily surrendered in August 2022 when Tongwei entered utility module tenders, triggering defensive vertical integration across its primary customer base.
Cornered resource — weak to moderate. Preferential access to low-cost hydroelectricity in Sichuan and Yunnan provides an undeniable cost advantage, but the resource is governed by provincial power contracts rather than proprietary ownership. These arrangements remain subject to policy renegotiation, and competing manufacturers have negotiated comparable regional energy tariffs across Inner Mongolia, Xinjiang, and Qinghai.
Switching costs — virtually nonexistent. Polysilicon is purchased against standardized chemical purity specifications, while modules are procured based on nameplate wattage, warranty terms, and project bankability. Utility-scale buyers can switch between qualified tier-one suppliers from one tender to the next with negligible friction.
Network effects — none. Commodity photovoltaic manufacturing generates no demand-side increasing returns or user network advantages.
Branding — limited in utility procurement. State-owned utility procurement managers award tenders primarily on bid pricing, delivery certainty, and financial bankability. Brand reputation carries modest value in distributed rooftop generation and select international markets, explaining why Tongwei's overseas expansion carries strategic importance beyond immediate volume growth.
The competitive landscape. Evaluating Tongwei against its peer group sharpens the strategic contrast. GCL Technology pursues an alternative chemical process route with structurally lower electricity consumption. 大全新能源 Daqo New Energy operates as a pure-play polysilicon manufacturer with an unleveraged balance sheet and zero downstream exposure, allowing it to capture the upside of any polysilicon price recovery without funding module working capital, though it lacks an earnings cushion during downcycles. JinkoSolar and 阿特斯 Canadian Solar operate diversified manufacturing footprints overseas, securing direct access to premium pricing in the United States that Tongwei cannot tap from mainland China. LONGi maintains the industry's premier commercial module brand. Tongwei's distinguishing characteristic across the sector is the pairing of global upstream scale with an uncorrelated agricultural cash engine. That combination provides an exceptional balance-sheet survival advantage, but it does not generate superior cycle-average equity returns.
The strategic takeaway. Tongwei commands two genuine forms of competitive power under Helmer's taxonomy—scale economies and process power—and both operate exclusively on the cost side of the ledger. Cost powers dictate which manufacturer endures a prolonged war of attrition. They do not enable a producer to generate excess economic returns when an entire industry is selling at or below cash cost. For investors, that distinction represents the core question under examination.
X. The Investment Spine: Historical Falsification, Risks & The Bull/Bear Case
Every investment thesis on Tongwei rests on one of three core claims. Each can be tested directly against the company's operating record.
Claim 1: "Tongwei's cost curve makes it resilient through the cycle."
The audited financial statements provide clear disconfirming evidence. Polysilicon spot prices dropped from more than ¥300 per kilogram to the low thirties. Even though Tongwei maintained the lowest cash production cost in the industry, it reported a net loss of ¥7.04 billion in FY2024 and ¥9.55 billion in FY2025, followed by an additional ¥5.12 billion loss in the first half of 2026 as revenue contracted another 15.19%.526 The sequential pattern is revealing: the second full year of the downturn was worse than the first, and the first half of 2026 deteriorated against the prior-year period even after polysilicon prices had begun to recover from cyclical lows.
Verdict: narrowed, not rejected. Cost leadership functions effectively as a balance-sheet shock absorber. Tongwei maintained positive operating cash flow throughout, generating ¥1.38 billion in cash from operations in FY2025 and returning to positive operational cash generation in the first half of 2026, while second-tier competitors idled facilities and entered court-led restructurings.211 Market-share gains achieved during deliberate volume cutbacks confirm a relative operational edge. Yet the thesis that cost leadership guarantees cyclical resilience in accounting earnings is decisively falsified. In a commodity market burdened by twice as much manufacturing capacity as end-market demand, the lowest-cost producer commands no pricing power; it simply burns cash more slowly than its peers. The defensible framing is that Tongwei's cost curve protects enterprise solvency and market share, but translates into net profit only when industry capacity utilization recovers. The essential metric to track is polysilicon cash production cost per kilogram excluding depreciation against realized selling prices: the investment thesis turns only when realized market prices sustainably exceed total fully absorbed manufacturing costs.
Claim 2: "Vertical integration into modules strengthens the business."
Assessing this claim requires looking at the operational outcome. Tongwei entered commercial module manufacturing in August 2022 with an aggressive bid of ¥1.942 per watt in the China Resources Power tender.7 Annual module shipments expanded to 45.71 gigawatts in FY2024 before contracting to 43.25 gigawatts in FY2025.53 Concurrently, its long-standing merchant cell partnerships with LONGi, Trina Solar, and JinkoSolar—historically Tongwei's most profitable, differentiated customer base—were structurally fractured as those buyers accelerated investments in captive cell capacity.
Verdict: largely falsified as a driver of economic value creation. Integration expanded absolute volume and established direct access to project developers, which has proven strategically valuable in international markets where Tongwei controls its own distribution channels and brand. Yet module assembly remains an inherently low-margin business burdened by channel overhead, extended working capital cycles, and twenty-five-year warranty liabilities, whereas the merchant cell business it cannibalized enjoyed superior structural economics. The primary counter-argument—that peers were already internalizing cell fabrication, rendering merchant neutrality obsolete—is plausible, but describes defensive damage limitation rather than proactive value creation. Investors should treat management's framing of integration as an unqualified strength with skepticism, and watch overseas module volume as the critical test: if international shipments at self-determined pricing become the majority of module sales, downstream integration will have justified its capital expenditure. At approximately 40% of module revenue in the first half of 2026, that threshold has not yet been crossed.11
Claim 3: "Next-generation cell technology creates a technological monopoly."
Tongwei maintains an active research and development portfolio: a 650-watt TOPCon module, a 780-watt mass-production route for heterojunction (HJT) cells with silver consumption below 4 milligrams per watt, an 802.43-watt module record at 25.83% efficiency, and an automated perovskite-silicon tandem pilot line achieving 34.94% small-area cell efficiency.2911
Verdict: technical capability is verified; commercial monetization remains unproven. HJT has been described as the imminent industry standard for the better part of a decade, yet mass adoption remains constrained by silver consumption, indium scarcity, and elevated capital equipment costs. TOPCon comprehensively captured the 2023–2026 commercial cycle, and Tongwei's operating footprint is standardized on TOPCon. The historical base rate in photovoltaic manufacturing is instructive: laboratory efficiency milestones and third-party certifications do not constitute economic profits. The sector's history is filled with technological leaders whose performance records never translated into cash flow because commercial adoption is governed by unit manufacturing cost rather than peak conversion efficiency. Advanced cell technologies function as essential insurance against technological obsolescence rather than as engines of durable pricing power. The catalyst that would revise this assessment is an HJT or perovskite tandem production line operating at multi-gigawatt scale with audited unit costs per watt below prevailing TOPCon baselines.
Myth versus reality: three consensus narratives worth checking.
"Tongwei is a counter-cyclical acquirer." Partially true, but narrower than market reputation suggests. The company executed an exceptional counter-cyclical purchase in 2013 by acquiring Hefei LDK, utilizing the parent group's balance sheet. It subsequently expanded capital expenditure pro-cyclically into the 2021–2022 peak, abandoned a distressed acquisition of Runergy in early 2025 when balance-sheet capacity was constrained, and pursued another large asset acquisition during the current trough. The historical record demonstrates opportunistic execution rather than a systematic counter-cyclical doctrine.
"The feed business makes Tongwei uniquely safe." Directionally accurate, but quantitatively bounded. The agriculture and animal husbandry division generated ¥13.45 billion in revenue in the first half of 2026, expanding 7.69% year-over-year.11 While it remains a growing, dependable, and cash-generative enterprise, its operating income cannot offset consolidated half-year losses of ¥5.12 billion.6 Agricultural cash flow sustains liquidity and buys operational time; it cannot fund the solar division's multi-billion-renminbi capital programs.
"Chinese policy will fix the overcapacity." A widely held consensus view that regulatory actions have repeatedly complicated. Government policy in 2026 pointed in two different directions within a single year: an antitrust intervention in January against coordinated industry output curbs, followed by market compliance guidance in July targeting below-cost pricing.3419 Beijing favors market consolidation and stricter energy-efficiency thresholds over coordinated producer supply pacts. While compatible in principle, these approaches operate on very different timelines: administrative efficiency standards are expected to retire uncompetitive capacity gradually through 2027, whereas producer coordination could have shifted supply balances within quarters.19
The consolidation question — and the regulator's answer.
This regulatory dynamic is the central variable that will govern the next phase of the industry cycle.
Beginning in July 2025, leading polysilicon manufacturers convened under the banner of industry self-discipline, developing a framework to pool approximately ¥50 billion to acquire and permanently retire roughly one million metric tons of polysilicon capacity—out of an estimated national total of three million metric tons—while operating remaining facilities at 60% to 70% utilization.35 Tongwei held the largest equity stake in the proposed consolidation platform at 30.35%, followed by GCL Technology at 16.79%, East Hope at 11.30%, Daqo New Energy at 11.13%, Xinte Energy at 10.12%, and Asia Silicon at 7.79%.36
On 6 January 2026, China's State Administration for Market Regulation intervened. The antitrust regulator concluded that participating producers had coordinated output, allocated market share, and pressured downstream margins through commitment letters and production adjustment pacts. The agency prohibited agreements on production volume, pricing, capacity, and market division, ordering participants to submit formal rectification plans by 20 January.34 The industry association promptly suspended its coordination sessions, and polysilicon futures declined 9.00% on 8 January and another 8.11% on 9 January, closing at ¥51,300 per metric ton.34
That regulatory ruling is the single most important development for evaluating the bull thesis, because it dismantled the private coordination mechanism that many investors were counting on. Beijing encourages industry consolidation, but actively opposes cartel-style price coordination. Navigating that distinction lengthens the industry shakeout and increases its cost.
The pivot to M&A. On 25 February 2026, Tongwei signed an intention agreement to acquire 100% of 青海丽豪 Qinghai Lihao Clean Energy through share issuance, cash consideration, and supporting fundraisings, prompting a trading suspension in its shares and convertible bonds.3738 Lihao had built approximately 150,000 metric tons of polysilicon capacity by late 2023 and was valued at ¥14 billion in January 2024; completing the acquisition would lift Tongwei's domestic polysilicon market share to roughly 30%.38 At the company's 18 May 2026 investor briefing, audit and asset appraisal procedures were still underway, with definitive terms undisclosed.29
Two implications follow. First, direct corporate acquisition provides a legally compliant route to the consolidation previously pursued through a multi-firm coalition: owning the assets outright allows Tongwei to idle or operate lines unilaterally without regulatory antitrust exposure. Second, from a governance perspective, Tongwei is proposing to fund part of the transaction through newly issued shares at a depressed equity valuation for an asset appraised during an earlier market cycle. While the Runergy termination demonstrated management's willingness to abandon unviable transactions, the 2016 Hefei transfer highlighted the valuation sensitivities inherent in large asset injections. Although Lihao is a third-party asset rather than a related-party entity, the resulting equity dilution and final valuation multiples will be the primary details for minority shareholders to scrutinize when terms are published.
The material risk radar.
Geopolitical quarantine. Tongwei operates virtually no manufacturing infrastructure outside mainland China, leaving it effectively shut out of the United States by the Uyghur Forced Labor Prevention Act, Section 301 tariffs, and anti-circumvention duties. Competitors with operational facilities in Southeast Asia or the United States capture materially higher realized prices in the American market. The Runergy termination left this geographic exposure unaddressed, and no alternative overseas manufacturing strategy has been announced.
Process disruption. GCL Technology's fluidized bed reactor granular silicon route, if it consistently delivers electronic-grade N-type quality at scale, directly challenges the specific electrical efficiency advantage upon which Yongxiang's cost moat was constructed.
Leverage and refinancing. A debt-to-asset ratio near 74% at the end of March 2026, cash reserves down roughly one-third year-over-year, ¥17.6 billion in outstanding bond obligations, and a pending acquisition requiring capital or equity dilution all constrain balance-sheet flexibility.6303 Positive operating cash flow provides the critical offset. If operating cash flow turns negative for two consecutive quarters, the challenge will shift from cyclical endurance to balance-sheet distress.
Regulatory ambiguity. Tongwei is simultaneously the primary commercial beneficiary of industry capacity discipline and the largest target of regulatory scrutiny when coordination occurs. On 31 July 2026, market authorities issued price compliance guidance across the solar sector aimed at curbing destructive below-cost competition—an administrative policy pointing toward informal pricing floors, whereas the January antitrust ruling penalized output coordination.19 Management remains exposed to shifts in administrative instruments that it cannot influence.
The bull case, stated at its strongest.
Chinese authorities have signaled that updated energy-efficiency standards could retire 20% to 30% of substandard industry capacity by 2027.19 Under that scenario, supply contracts toward underlying demand, plant utilization recovers, and the lowest-cost producer with the largest operational scale captures the majority of the margin rebound. Market anticipation of this shift was demonstrated on 3 August 2026, when domestic polysilicon futures rose to their daily trading limit, closing at ¥35,890 per metric ton, while Tongwei's equity hit its 10% daily ceiling with ¥867 million in pending buy orders.19 In the background, the agricultural feed business continues to generate steady cash, while expanding overseas sales establish pricing channels outside China's state-dominated utility tenders.
The bear case, stated at its strongest.
Spot N-type polysilicon averaged roughly ¥31.9 per kilogram in early August 2026, trailing the futures curve and indicating that physical market fundamentals had not yet confirmed paper market sentiment.19 Capital assets in heavy manufacturing rarely vanish permanently; they are typically purchased out of distress and eventually restarted by new operators. Absorbing a 1,100-gigawatt global manufacturing base against 10% to 15% annual demand growth points to an extended multi-year rationalization rather than a rapid cyclical snapback. Furthermore, Tongwei's own financial results illustrate that vulnerability: first-half 2026 losses widened relative to the first half of 2025 despite firmer polysilicon prices, as downstream cell and module margins compressed under surging silver metallization paste costs.63 In an enduring supply glut, being the lowest-cost survivor does not prevent multi-year equity value destruction.
What would change the verdict. The balance between the bull and bear arguments hinges on an observable operational benchmark: whether industry-wide polysilicon capacity utilization rises to a level where the marginal producer covers cash costs, and remains there for two to three consecutive quarters. If it does, industry economics normalize and the lowest-cost producer captures the majority of recovered operating cash flow. If utilization languishes, Tongwei faces an ongoing war of attrition, sustained by feed earnings and working-capital discipline while leverage mounts. That clearing dynamic remains outside management's control.
The three KPIs that matter.
- Polysilicon cash production cost per kilogram, excluding depreciation, versus realized selling price. This spread reveals whether the upstream cost advantage is generating cash flow or merely minimizing accounting losses.
- Overseas share of module and cell shipments. A metric tracking whether Tongwei is successfully diversifying away from domestic reverse-auction utility tenders. In the first half of 2026, international sales accounted for approximately 40% of module revenue, matching management's full-year target.
- Quarterly operating cash flow against net capital expenditure. The fundamental solvency test. As long as operational inflows fund maintenance capital requirements and feed earnings support the balance sheet, Tongwei can endure the downcycle. If operating cash flow turns negative, the investment thesis changes fundamentally.
XI. Epilogue: What Tongwei Teaches Us About Chinese Industrial Champions
A common telling of Tongwei's history treats the animal feed business as a quaint prologue—the venture a founder pursued before discovering his true industrial calling. That narrative misinterprets the company's balance-sheet architecture, and the downturn since 2023 has demonstrated why.
Feed earnings sustained Yongxiang through the first solar winter between 2009 and 2012, when polysilicon absorbed corporate cash and pure-play peers slid into insolvency. That same cash engine supports the enterprise through the present downturn, expanding 7.69% to deliver dependable operating earnings while photovoltaic revenue contracted more than 26% year-over-year in the first half of 2026.11 For two decades, Liu Hanyuan's 双主业 (dual main business) model was often dismissed by public markets as unfocused conglomerate diversification. In practice, it created an industrial architecture capable of holding capital assets through an industry governed by violent four-year supply cycles with no natural exit mechanism.
The sharper lesson concerns the limits of cost leadership. Tongwei stands as the most efficient large-scale polysilicon and cell manufacturer in the global solar sector, commanding benchmark silicon consumption rates, leading electrical efficiency, the industry's largest production clusters, and a nine-year tenure as the world's top merchant cell supplier. Yet the company incurred more than ¥21 billion in net losses across the two and a half years between 2024 and mid-2026. In heavy continuous-process manufacturing, scale and cost discipline are essential for surviving a downcycle, but entirely insufficient to evade it. Mistaking factory-floor efficiency for a pricing moat in an oversupplied commodity market remains an expensive analytical error.
The strategic burden now falls to Liu Shuqi. Taking leadership at thirty-four, she inherited an enterprise with world-class operating metrics, a compromised merchant position, and mounting leverage. Her tenure has already produced one major disciplined refusal—terminating the troubled Runergy acquisition—alongside an unresolved consolidation gamble in Qinghai Lihao. Direct corporate acquisition provides the only legally compliant route to industry rationalization following the January 2026 antitrust intervention against output pacts, and Tongwei commands the scale to pursue it. Whether management can execute that transaction without overpaying, taking on unsustainable leverage, or repeating the historical valuation disparities of earlier asset injections will determine what Tongwei is worth when industry economics normalize.
Beyond Tongwei itself, the trajectory offers a broader template for evaluating Chinese industrial champions. Corporate disclosures throughout the downturn have highlighted genuine operational milestones: record feed volumes, expanding cell shipments, surpassing 400 gigawatts in cumulative cell deliveries, declining unit resource consumption, growing patent portfolios, and laboratory module efficiency records. Every metric is verifiable. Yet none prevented consecutive annual losses running into billions of renminbi. In commodity manufacturing burdened by structural overcapacity, operating excellence and equity returns can diverge for years. Reading an industrial champion's financial statements requires holding both realities in view simultaneously.
The ultimate resolution will not come from laboratory records or market-share tables. It will depend on cash cost per kilogram, the export revenue mix, and operating cash flow generation—the same disciplined operational arithmetic that lifted an eighteen-year-old technician's carp cage into an industrial empire four decades ago.
References
-
通威股份：2025年归母净利润亏损95.53亿元 — 新浪财经 Sina Finance, 2026-04-29 ↩↩↩↩↩↩↩↩↩
-
通威股份（600438）2025年年报&2026年一季报点评：硅料&电池销量稳居行业第一，组件出货量进入全球前五 — 中国能源网 China5e, 2026-05 ↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩↩
-
Tongwei Co., Ltd. 2022 Annual Report — Shanghai Stock Exchange, 2023-06-27 ↩↩
-
Tongwei Co., Ltd. 2024 Annual Report (2024年年度报告) — Shanghai Stock Exchange, 2025-04-26 ↩↩↩↩↩↩
-
Tongwei in first place to secure 3GW module tender from China Resources Power — PV Tech, 2022-08 ↩↩↩
-
Liu Hanyuan, the man of the water and the sun — Gemway Assets, 2022-10-25 ↩
-
Tongwei Co., Ltd. 2025 Annual Report (2025年年度报告) — Shanghai Stock Exchange / cninfo, 2026-04-30 ↩↩
-
通威股份：双主业筑牢底盘，一体化龙头蓄力穿越光伏周期 — 证券市场周刊 Weekly on Stocks, 2026-08-31 ↩↩↩↩↩↩↩↩↩↩
-
Tongwei Group Acquires 100% Equity of Hefei LDK Solar for RMB 870 Million — Caixin, 2013-09-27 ↩
-
Report on Tongwei Co., Ltd. Major Asset Acquisition and Related-Party Transaction (Hefei Tongwei and Sichuan Yongxiang) — Shanghai Stock Exchange, 2016-01-08 ↩
-
Tongwei Dominates the Polysilicon Industry — Bernreuter Research ↩
-
Chinese PV Industry Brief: Tongwei scores 3 GW supply deal — pv magazine International, 2022-08-19 ↩
-
Chinese Solar Panel Makers Slump as Silicon Giant Tongwei Enters Their Turf — Yicai Global, 2022-08 ↩
-
Announcement on the Resignation of Chairman and CEO Xie Yi and Election of Liu Shuqi as Chairwoman — Shanghai Stock Exchange / Tongwei Co., Ltd., 2023-03-22 ↩↩
-
Chinese Silicon Giant Tongwei Names Founder's Daughter as New Chair, CEO — Yicai Global, 2023-03 ↩↩
-
China Solar Giant Tongwei Swings to 2024 Loss as Polysilicon Price Slump Bites — Reuters, 2025-04-26 ↩
-
Top ten solar manufacturers shipped over 500GW of modules in 2025 — PV Tech, 2026 ↩↩
-
Tongwei Reports H1 2025 Loss, Cites Persistent PV Market Pressures — TaiyangNews, 2025-09 ↩↩
-
Chinese PV Industry Brief: Longi, Tongwei announce sharp losses for fiscal 2025 — pv magazine International, 2026-01-23 ↩
-
Chinese PV Industry Brief: Tongwei, GCL and Daqo post steep H1 losses — pv magazine International, 2025-09-12 ↩
-
Announcement on Intent to Acquire Equity in Jiangsu Runergy New Energy Technology Co., Ltd. via Capital Increase and Share Transfer — Shanghai Stock Exchange / Tongwei Co., Ltd., 2024-08-14 ↩
-
Announcement on the Termination of Proposed Acquisition of Equity in Jiangsu Runergy New Energy Technology Co., Ltd. — Shanghai Stock Exchange / Tongwei Co., Ltd., 2025-02-15 ↩
-
Tongwei drops plan to acquire Runergy — pv magazine International, 2025-02-18 ↩↩
-
Tongwei Pulls Plug on $690 Million Solar Deal for Runergy — Bloomberg News, 2025-02-17 ↩
-
Tongwei and Risen Energy Release Latest Investor Briefing Minutes — EnergyTrend, 2026-05-20 ↩↩↩↩↩↩
-
Tongwei Co., Ltd. Rating Report — Lianhe Ratings Global, 2025-09-25 ↩
-
China's competition regulator halts $7 billion plan to consolidate polysilicon industry — pv magazine International, 2026-01-09 ↩↩↩
-
China's $7 billion consolidation plan may be reflected in prices in short time — pv magazine International, 2025-08-14 ↩
-
China Launches Platform for Polysilicon Consolidation; Industry Awaits Supply-Side Impact — OPIS, A Dow Jones Company ↩
-
Tongwei plans to acquire polysilicon competitor Qinghai Lihao — pv magazine Global, 2026-02-25 ↩
-
China's Tongwei Seeks to Cement Solar Dominance With Acquisition — Caixin Global, 2026-02-25 ↩↩
LinkedIn
WhatsApp
Reddit
Hacker News
Email
Kindle
RSS Feed
Spotify
Apple Podcasts
Amazon Music
Audible
YouTube