China Jushi: The Invisible Backbone of Modern Industry

I. Introduction & Episode Roadmap

On June 26, 2026, shares of China Jushi closed at a record RMB 77.20, capping a 357% surge from the start of the year.[^1] Three weeks later, on July 16, the stock dropped by the single-session limit of 10% to RMB 47.71 after the company announced a RMB 2.4 billion capacity expansion plan.[^1] Investors had spent six months bidding up the stock as an artificial-intelligence supply-chain play. Management then reminded the market of the company's core identity: a heavy-industrial manufacturer that responds to rising prices by constructing more furnaces.

That whipsaw illustrates the central tension within 中国巨石 China Jushi Co., Ltd. (600176.SS, Shanghai Stock Exchange). Jushi manufactures glass fiber—molten glass drawn through platinum bushings into filaments thinner than a human hair, then bundled into yarn, chopped into strands, or woven into fabric. The material reinforces wind turbine blades exceeding the wingspan of a Boeing 747, stiffens automotive body panels, strengthens sewer pipes and shipping containers, and provides thermal insulation. In its highest-value application—ultra-thin woven 电子布 electronic glass fabric—it serves as the insulating substrate onto which copper circuitry is printed for printed circuit boards, including accelerator cards in artificial-intelligence data centers.

It is an indispensable industrial material operating largely out of public view.

The enterprise grew from a near-bankrupt township cloth factory in Tongxiang, Zhejiang. Today, Jushi operates manufacturing bases in Tongxiang, Jiujiang, Chengdu, Huai'an, Egypt, and South Carolina. In 2025, it sold a record 3.2026 million tonnes of roving and related products alongside a record 1.062 billion meters of electronic fabric.¹² With total output roughly equivalent to the world's second-, third-, and fourth-largest producers combined, Jushi controls approximately 30% of global fiberglass capacity.³

The core question this article tests: Jushi's investment thesis rests on two primary claims: first, that decades of furnace engineering provide a structural unit-cost advantage competitors cannot easily match; second, that its electronic fabric business offers a high-margin shield against commodity market cycles. While operational evidence supports the cost leadership claim, the electronic fabric narrative remains largely forward-looking. In June 2026, management acknowledged that its low-dielectric specialty electronic fabrics—the specific products tied to artificial-intelligence demand—had "not formed orders or revenue."⁴ This disclosure serves as the necessary baseline for evaluating the company's strategy.

Here is how the story unfolds:

1. The township-enterprise origin and 张毓强 Zhang Yuqiang's bet on a furnace technology China had never built.
2. The hybrid alliance with 中国建材集团 China National Building Material Group (CNBM) that supplied the balance sheet for global scale.
3. The overseas playbook — Egypt and the United States — and what it actually delivered versus what it promised.
4. The unit economics: where the cost advantage genuinely comes from, and where it stops working.
5. The 2022–2024 overcapacity crash, and the electronic fabric pivot that followed.
6. Management, capital allocation, and a governance promise that has now been extended for nearly a decade.
7. The playbook, the risks, and the explicit bull-versus-bear spine.

---

↑ Back to Top

II. The Town-Enterprise Spark: From Tongxiang Workshop to National Champion (1984–2003)

The founding scene was not a corporate boardroom; it was a delivery truck.

In 1972, 张毓强 Zhang Yuqiang, a young man from Tongxiang, traveled to the Jiujiang Glass Fiber Factory in Jiangxi and returned with a crude drawing machine, installing it at the Shimen Dongfeng cloth factory in his home township—an operation then close to shutting down.⁵ That was the seed: not a grand corporate strategy or market study, but one piece of borrowed equipment brought back to keep a struggling township collective alive.

For two decades, that operation manufactured glass fiber using the method common across China: 球法拉丝 crucible drawing. The process resembles melting glass marbles in a small pot, one batch at a time, and pulling filaments out through the bottom. While cheap to set up, it was brutally inefficient to run. Each crucible operated as a discrete batch, leaking thermal energy, producing variable quality, and consuming vast amounts of manual labor per tonne. It was the technique of a capital-starved economy, yielding low-grade yarn that foreign buyers generally rejected for heavy industrial use.

The alternative was the 池窑拉丝 direct-melt tank furnace. Instead of small pots, workers construct a massive refractory-lined basin, fill it with molten glass, and run raw batch in one end while continuously drawing fiber out the other—24 hours a day for furnace campaigns lasting years. It represents the shift from small batch cooking to continuous refining. Energy consumption per tonne drops sharply, material consistency improves by an order of magnitude, and unit costs collapse. However, the capital requirements are steep. The structure must withstand years of continuous corrosion from molten glass, and the drawing bushings require 铂铑合金 platinum-rhodium alloy—precious metals permanently tied up as working capital in the hot zone.

欧文斯科宁 Owens Corning had pioneered direct-melt manufacturing in the West, but by the early 1990s, no domestic firm in China had successfully built one.

The 3·18 Project

On March 18, 1993, Zhang Yuqiang led Tongxiang Zhenshi and four other units to establish Tongxiang Jushi Glass Fiber Co., Ltd., becoming the first enterprise to move into the Tongxiang Economic Development Zone.⁵ The venture committed RMB 75 million—an extraordinary sum for a township enterprise at the time—to build the first tank furnace drawing line designed and constructed by Chinese engineers.⁵ Known internally as the "3·18 Project" after its founding date, Zhang lit the first fire in November 1994.⁵

The risk profile of that decision was stark. As a township collective without access to public equity markets, foreign technology licenses, or domestic precedents, Jushi committed its entire balance sheet to a continuous-process technology with catastrophic failure modes. If a tank furnace fails, it cannot be fixed over a weekend. Refractory erosion, thermal cracking, or flawed bushing design can strand the entire capital investment. Zhang's decision stemmed not from a belief that building a furnace was easy, but from the conviction that crucible drawing had no commercial future and that companies relying on it would not survive the decade.

That initial bet established the engineering culture that continues to define the company. Rather than licensing turn-key technology and making marginal adjustments, Jushi designed, built, tested, and rebuilt its own hardware. The operational metrics that management emphasizes today—throughput per furnace, energy consumed per tonne, bushing hole density, and refractory campaign life—originated from a team forced to learn furnace engineering from first principles.

What the furnace bought

The operational payoff emerged rapidly. By 1996, the operation had claimed a set of domestic firsts across production scale, product quality, product range, export earnings, management, and economic returns—a notable shift given what it replaced.⁵ Jushi Group was formed that same year, consolidating the operating businesses around the furnace.⁶

This transition highlights a structural mechanism that recurs throughout Jushi's history: continuous melting did more than cut costs—it unlocked new markets. Crucible-drawn yarn suffered from inconsistency that restricted its addressable market to domestic, low-specification applications where producers lacked pricing power. Tank-furnace yarn provided the consistency required for international export. Building the furnace served simultaneously as a cost-reduction project and a market-access strategy, a dual advantage management recognized early by continuing to expand furnace scale long after initial cost gains had been harvested.

At the same time, Jushi's early success benefited significantly from favorable macroeconomic conditions. The company operated in a protected, rapidly expanding domestic market supported by low local costs for energy, land, and labor, alongside government support for industrial import substitution. Achieving first-mover status in that environment provided a decade of operational learning unavailable to domestic rivals. What separates Jushi from industrial peers that squandered similar head starts is what followed: it consistently reinvested its cash flow back into furnace scale rather than diversifying into real estate or non-core ventures.

The listed vehicle

The capital markets piece arrived separately. The entity that trades today as 600176.SS was incorporated in April 1999 and listed on the Shanghai Stock Exchange main board that same month, initially under the name 中国化建 China Chemical Building Materials, later renamed 中国玻纤 China Fiberglass and finally, in the mid-2010s, 中国巨石 China Jushi.⁶ The fiberglass assets were progressively concentrated into that platform rather than listed directly out of the township. This detail matters: from early on, the operating business and the listed vehicle maintained different histories and different sponsors, establishing the hybrid structure that later produced both state-backed financing advantages and corporate governance complications.

What an investor takes from the early period: The durable asset created between 1993 and 2003 was not merely physical plant infrastructure, which competitors could eventually replicate. It was an engineering organization that had mastered the physics of continuous glass melting before its domestic peers, paired with a public listing to finance expansion. Everything Jushi accomplished over the following two decades built upon those two foundations. What it did not yet have was a balance sheet capable of building at world scale—and that constraint is what drove the company into the arms of the state.

↑ Back to Top

III. The SOE Hybrid Alliance: CNBM Partnership & The Scale Playbook (2004–2011)

By the early 2000s, Jushi faced a challenge common among rapidly growing industrial manufacturers: it had mastered furnace engineering better than its domestic peers, but lacked the capital required to build at scale.

The global fiberglass industry at the time operated as a Western oligopoly dominated by Owens Corning, Saint-Gobain's Vetrotex, PPG Industries, and Johns Manville. These established players controlled proprietary technology, customer qualifications, and access to low-cost capital. Building a modern production base required an indivisible, multi-hundred-million-dollar commitment made years before shipping the first tonne—financed against volatile commodity prices and with expensive precious metals permanently tied up in furnace hardware. As a private enterprise in Zhejiang, Jushi could not secure financing at rates that made large-scale expansion economical.

The deal

The solution arrived through 混合所有制 mixed ownership. Between 2004 and 2007, the state-owned 中国建材集团 CNBM system acquired control of the listed platform, while founder Zhang Yuqiang and the private 振石控股集团 Zhenshi Holding Group retained a substantial minority stake and operational control. In January 2007, the listed company disclosed a capital increase at Jushi Group and the entry of strategic investors, formalizing the ownership structure.⁷

That ownership structure remains in place. CNBM Co., Ltd. holds 29.37% of China Jushi after completing a share-purchase program that increased its holding to approximately 1.176 billion shares by May 2026.⁸ Zhenshi Group serves as the primary private anchor, adding 51.27 million shares—about 1.28% of share capital, valued at roughly RMB 1.1 billion—through a purchasing program initiated in late November 2025.⁹ The State-owned Assets Supervision and Administration Commission acts as the ultimate controller.⁶

Why the hybrid actually worked

The economic mechanism of state backing was straightforward: it lowered the cost of capital and extended debt maturities. State-backed financing allowed Jushi to commit to five-year furnace construction cycles despite 40% peak-to-trough output price swings. Private competitors facing similar cyclicality with commercial credit were forced to build smaller, delay investment, or accept equity dilution.

State capital, however, carries well-documented risks, often leading to overinvestment in unproductive assets or delayed restructuring due to policy priorities. Jushi mitigated this by preserving private management with significant equity stakes, maintaining commercial discipline over furnace site selection, timing, and retirement. Management described this reform as a balance between state balance-sheet support and private operational autonomy.

Yet this arrangement came with trade-offs. The hybrid structure introduced enduring conflicts of interest that persisted across subsequent governance decisions, demonstrating that state-backed capital efficiency carries structural governance costs.

The scale playbook

Supported by CNBM's balance sheet, Jushi executed a systematic expansion strategy over the following decade: constructing mega-production bases and increasing furnace capacities. Domestic production hubs expanded from Tongxiang to Jiujiang in Jiangxi and Chengdu in Sichuan. Single-line annual capacity grew from tens of thousands of tonnes past 100,000 tonnes. The company completed its Jiujiang 200,000-tonne line ahead of schedule and upgraded its headquarters line from 120,000 to 200,000 tonnes during cold repair.¹⁰

The unit economics of furnace expansion underpin Jushi's cost advantage. A tank furnace's fixed operating costs grow much slower than its volumetric output. Doubling output does not double wall thermal losses, required operating personnel, or supervisory engineering staff. Consequently, larger furnaces lower unit costs for energy, labor, and depreciation—provided the furnace operates reliably at scale without structural failure. This operational execution converted Jushi's earlier engineering experience into a quantitative financial moat.

Why those three cities

Site selection reflected strict supply-chain and freight logistics logic. Tongxiang served as the operational origin, holding engineering expertise, long-standing supplier relationships, and proximity to Yangtze River Delta export infrastructure. Jiujiang provided direct access to deep-water shipping on the Yangtze River, shifting bulk raw material imports and finished yarn exports from truck to lower-cost barge transport—a critical saving for a heavy industrial product where freight represents a major component of delivered cost. Chengdu anchored western expansion, offering proximity to inland industrial demand, lower historical energy and labor costs, and regional development subsidies.

Together, these locations established an integrated logistics network. Because glass fiber is dense and costly to transport relative to its unit value, saving RMB 100 per tonne on raw material delivery and finished product freight preserves a significant portion of operating margin. While furnace engineering provided production efficiency, logistics optimization yielded comparable cost benefits.

Meanwhile, the export channel enabled by continuous-melt manufacturing expanded steadily. By 2024, China's glass fiber exports reached approximately 2.12 million tonnes—a 13.1% year-over-year increase valued at roughly $3.03 billion, representing about a quarter of national output.¹² Jushi accounted for the largest share of these shipments, establishing its global market presence while increasingly exposing the company to international trade friction.

The 2008 test

The 2008 global financial crisis tested this operational model. As global industrial demand contracted, Western competitors idled production lines and cancelled capital expenditures. In continuous-melt manufacturing, idling a furnace is costly and often irreversible. Leveraging lower operating costs and state-backed credit access, Jushi maintained production and continued expanding. By the early 2010s, Jushi emerged as the world's largest fiberglass manufacturer by volume.

What this says about the business: Jushi's market-share expansion resulted primarily from balance-sheet support and cost execution rather than product differentiation. The company expanded when capital-constrained competitors were forced to contract. While this strategy established global volume leadership, reliance on cheap, state-backed capital represents a policy-dependent advantage. Furthermore, rapidly expanding export volumes eventually prompted overseas markets to implement trade restrictions.

↑ Back to Top

IV. Defeating Trade Protectionism: The Egypt & US Overseas Expansion (2012–2020)

There is a specific kind of problem that only afflicts winners in traded commodities: you get so good at exporting that the importing country makes it illegal.

By around 2010, European and Indian regulators had begun stacking anti-dumping and anti-subsidy duties on Chinese fiberglass. For a company deriving a substantial share of revenue from exports, this was not a tax — it was an existential question about whether the European market remained addressable at all. Tariffs on Chinese-origin fiberglass entering the EU reached levels that consumed the entire cost advantage Jushi had spent fifteen years engineering.

The strategic response was elegant and, at the time, genuinely novel for a Chinese industrial firm: if the tariff attaches to the country of origin, change the country of origin.

Egypt

In January 2012, Jushi Egypt for Fiberglass Industry was established in the China-Egypt Suez Economic and Trade Cooperation Zone — the first large-scale overseas fiberglass production base built by a Chinese enterprise.¹¹ The logic stacked neatly. Egypt had abundant local silica sand. It had cheap natural gas, which for a business whose largest variable input is heat is close to the whole ballgame. It had low labor costs. And, crucially, product manufactured in Egypt entered the European market outside the Chinese duty regime.¹¹

Jushi expanded the base repeatedly, building out to four production lines and roughly 360,000 tonnes of annual capacity, which turned Egypt from a fiberglass importer into a globally significant producer.¹² The financial results validated the thesis for over a decade. In 2024, the Egypt base generated RMB 1.702 billion of revenue — about 10.7% of the group total — at a net margin of 18.7%, materially above the company-wide average of roughly 16%.¹² Cheap gas plus tariff-free access to a high-priced market is a very good combination.

The United States

The second move was harder. In 2016, Jushi broke ground on a greenfield plant in Richland County, South Carolina — a roughly $300 million initial commitment, the first tenant in a new county industrial park, with plans to invest up to $500 million at full build-out and employ around 400 people.¹³¹⁴ It came online just as the US-China trade war escalated, giving Jushi domestic North American production behind the Section 301 tariff wall, close to automotive and composites customers.

The strategic logic was identical to Egypt's. The results were not.

Falsification: does the "Go Global" playbook actually travel?

The claim under test: that Jushi's manufacturing footprint strategy delivers friction-free international diversification with margin profiles comparable to home.

The disconfirming record, over the longest span that bears on it:

The US plant lost money for years. In 2024 — six years after startup — the US base produced RMB 788 million of revenue, about 5.0% of group total, and a net loss of RMB 28 million.¹² It did not turn profitable until the second quarter of 2025, and management attributed that turn substantially to tariff adjustments rather than to operational improvement.¹⁰ Put bluntly: the plant built to escape tariffs became profitable because of tariffs. Chinese capital cost and productivity advantages did not transfer.

The operating record was worse than the financial one. In July 2023, South Carolina's Department of Health and Environmental Control issued an enforcement order against China Jushi USA covering 38 violations — 21 air pollution control and 17 hazardous waste — with penalties exceeding $460,000 ($410,000 for air, roughly $57,000 for hazardous waste).¹⁵ Regulators found the plant had exceeded particulate matter limits on more than 200 days between early 2020 and mid-2022, had uncontrolled emissions escaping the facility, had poorly maintained equipment, had failed to run required tests, and had employees inadequately trained to identify hazardous chemicals.¹⁵ The agency stated that emergency action was necessary to address the danger to public health and safety presented by the operations.¹⁵ These are the fines of a plant that was not being run to standard.

And in April 2026, the Egyptian workaround itself was closed. The European Union adopted anti-dumping measures on glass fibre reinforcements from Bahrain, Egypt, and Thailand — jurisdictions the EU identified as hosting Chinese-linked producers.¹⁶ Jushi Egypt received an 11.0% anti-dumping duty, which stacks on top of the anti-subsidy measures already in force since 2020, producing a combined rate of roughly 24.1%.¹⁷ Egypt had by then become the single largest source of glass fibre reinforcement imports into the EU, accounting for about 18% of the market in 2024.¹⁷ European producers and unions had campaigned for the measures for more than a decade.¹⁷ Separately, Jushi Egypt had already litigated its subsidy treatment before the EU General Court and lost.¹⁸

The verdict: the history does not reject the regionalization playbook, but it narrows it substantially. Building inside a tariff wall works — until the destination market decides that the wall should follow the ownership rather than the geography. The Egypt base remains a genuinely good asset on cost grounds; what it is no longer is a tariff-immune asset. And the US experience demonstrates that Jushi's cost system is not portable: it is built on Chinese energy prices, Chinese construction costs, Chinese labor productivity, and a supplier ecosystem that does not exist in South Carolina.

What would change the assessment: management stated on the Q2 2025 call that roughly half of capacity growth during the 15th Five-Year Plan is intended to be international, with Southeast Asia and Africa as priorities.¹⁰ The observable test is whether any new overseas base achieves domestic-comparable gross margin within three years of startup — something neither Egypt (which succeeded on the different basis of cheap gas) nor the US (which did not succeed at all until tariffs shifted) has demonstrated for a Western-cost environment.

That footprint question sits on top of a more fundamental one: how much of Jushi's cost advantage is real, and how much of it is simply being the biggest?

---

↑ Back to Top

V. Core Business Economics: Industry Structure, Cost Moat, & Competitive Rivalry

Understanding Jushi's competitive position requires examining the distinct economics of its three main product lines:

Roving and chopped strands represent the high-volume core. These thick yarns and short fibers reinforce wind turbine blades, automotive components, industrial piping, and construction materials. Operating as a pure commodity business, buyers qualify suppliers based on strict technical grades and purchase primarily on price. This segment generates the overwhelming majority of Jushi's production tonnage and historically most of its revenue.

Electronic yarn and fabric follow a different economic logic. Here, molten glass is drawn into ultra-fine filaments—often just a few microns thick—and woven into precise fabric. Board manufacturers impregnate this cloth with resin to produce copper-clad laminates, the foundational substrate for printed circuit boards. Customer qualification cycles range from 12 to 24 months, creating higher switching costs. In 2025, electronic fabric accounted for 17.77% of Jushi's operating revenue.⁴

By-products and equipment fabrication, including in-house manufacturing of platinum-rhodium alloy bushings, contribute modest direct revenue but yield high strategic value. By fabricating its own bushings, Jushi retains core engineering improvements and process secrets internally.

The demand side nobody models correctly

Two distinct end markets drive roving demand, each operating under different commercial dynamics.

Wind energy represents the most technically demanding application. Modern turbine blades function as long, hollow composite beams subjected to continuous mechanical flexing over a 25-year operational lifespan. As blade lengths expand to maximize swept area, demand shifts toward high-modulus formulations that offer greater stiffness than standard E-glass, commanding a pricing premium. Forecasts for Chinese wind installations in 2025 projected roughly 115 gigawatts of new capacity, representing about 31% annual growth.¹²

However, volume growth does not guarantee expanding margins. Wind turbine manufacturers negotiate contracts annually, locking in prices while competing in a severe domestic price war. These OEMs pass cost pressures directly upstream to fiber suppliers. Consequently, wind energy functions as a high-volume, low-margin customer base rather than a margin driver.

Construction and infrastructure roving historically provided the steady baseline volume needed to keep continuous furnaces running at full capacity. That demand anchor weakened alongside China's property market contraction, with no single replacement market offering equivalent scale. While automotive lightweighting and thermoplastic applications have expanded, they operate from a smaller baseline and face direct competition from alternative materials.

As a result, Jushi increasingly sells bulk volume to concentrated buyers with substantial bargaining power and rigorous product specifications. In this environment, maintaining structural cost leadership represents the primary defensible advantage.

The industry structure

The global fiberglass industry exhibits high supply concentration: the top producers control roughly three-quarters of global capacity, with the three largest accounting for nearly half.¹⁹ Jushi leads the industry with total annual capacity exceeding 3.5 million tonnes. As of February 2025, this footprint included approximately 2.89 million tonnes of nameplate roving capacity (with 2.59 million active in production) alongside 270,000 tonnes of electronic yarn capacity.¹²³ 泰山玻纤 Taishan Fiberglass, operated under 中材科技 Sinoma Science & Technology, ranks second globally with more than 1.4 million tonnes of annual capacity.¹⁹ 重庆国际复合材料 CPIC serves as the third major domestic producer, while international peers such as 日本電気硝子 Nippon Electric Glass and Johns Manville target specialized high-end and regional markets.

A major industry shift occurred in early 2025 when Owens Corning—the pioneer of commercial glass fiber—decided to exit the reinforcement business entirely. In February 2025, Owens Corning agreed to sell substantially all of its global glass reinforcements segment to affiliates of the Praana Group at an enterprise value of $755 million.²⁰ The parties amended the purchase agreement in April 2026, lowering the enterprise valuation to $645 million.²¹ The transaction closed on May 1, 2026, generating approximately $280 million in net after-tax cash proceeds.²² Chief Executive Officer Brian Chambers noted that the divestiture allowed the company to focus capital on its higher-margin branded residential building products business.²²

This transaction underscored the shifting competitive landscape: the industry's legacy Western producer concluded that a capital-intensive, highly cyclical reinforcement business offered lower risk-adjusted returns than its consumer-facing building brands, accepting a 15% valuation haircut to finalize its exit.

Where the cost advantage actually comes from

Jushi's cost advantage relies on three distinct operational pillars:

Scale economics. As detailed in earlier furnace expansions, fixed overhead, engineering costs, and depreciation divide across larger production volumes with each successive furnace generation.

Procurement leverage. Jushi leverages its total volume to negotiate centralized long-term supply contracts. Historical procurement data demonstrates that Jushi secured 叶蜡石 pyrophyllite—its primary mineral raw material—at prices significantly below those paid by Taishan Fiberglass, alongside lower natural gas tariffs.²³ Applied across more than three million tonnes of annual output, these unit savings generate substantial cost reductions.

Proprietary process knowledge. Operational efficiencies derive from unpatented trade secrets accumulated over decades, including oxy-fuel combustion tuning, burner positioning, bushing hole density, refractory lining maintenance, and specialized 浸润剂 sizing agent chemical formulations.

These factors create a measurable cost gap. In 2023, Jushi's average unit production cost ran at approximately RMB 3,883 per tonne, compared to roughly RMB 4,777 per tonne for Taishan Fiberglass—a cost differential of nearly RMB 900 per tonne, equivalent to a ten-percentage-point gross margin advantage.²³

However, financial disclosures also highlight a key structural trade-off. In 2023, Jushi's average selling price stood at roughly RMB 5,361 per tonne, compared to Sinoma's average selling price of approximately RMB 6,173 per tonne.²³ Jushi's competitive moat is built on unit-cost reduction rather than premium product pricing.

Seven Powers and Five Forces, honestly applied

Evaluating Jushi through standard competitive frameworks clarifies its market dynamics:

Under Hamilton Helmer's Seven Powers framework, Jushi demonstrates clear Scale Economies and strong Process Power. Claims of a Cornered Resource based on vertical integration into raw materials or proprietary high-modulus formulations are less compelling; pyrophyllite is not globally scarce, and procurement advantages stem primarily from contract scale and logistics. Commodity roving lacks meaningful Switching Costs or Branding power. In electronic fabric, qualification requirements create switching costs, though these historically protected incumbent Japanese suppliers rather than new market entrants.

Applying Michael Porter's Five Forces model reveals similar structural realities: - Threat of new entrants: Low, driven by heavy initial capital requirements and the substantial working capital tied up in precious-metal furnace bushings. - Buyer power: Moderate to high in roving due to annual OEM contract pricing, but lower in qualified electronic substrate markets. - Supplier power: Low for silica sand and moderate for natural gas. However, supplier power remains high for specialized Japanese high-precision looms required to weave ultra-thin electronic fabric, where industry analysts expect supply constraints to persist over a one-to-two-year horizon.²⁴ - Rivalry: Intense. In continuous-melt manufacturing, marginal operating costs sit far below full unit production costs, incentivizing producers to maintain furnace output even during market downturns.

Falsification: does cost leadership prevent margin compression?

The claim under test: Jushi's cost advantage insulates the business from structural margin damage during severe industry downturns.

The disconfirming record: Operational evidence demonstrates that cost leadership cushions but does not prevent margin compression. Between June 2022 and the market trough, domestic spot prices for benchmark 2400tex alkali-free direct roving dropped over 40%, falling from RMB 6,450 per tonne to a low of RMB 3,760 per tonne—a level near cash production costs.²⁵ Reflecting this pricing collapse, Jushi's net profit fell 53.94% in 2023 to RMB 3.044 billion.²⁵ Net profit declined an additional 19.70% in 2024 to RMB 2.445 billion, even as annual revenue expanded by 6.59%, illustrating a scenario where volume growth was sustained through price concessions.²⁶

The verdict: Jushi's cost position ensures positive cash flow and operational continuity when higher-cost competitors incur losses, but it does not protect net earnings against severe industry overcapacity.

Key forward metric: Net profit per tonne of roving. Management cited a baseline of approximately RMB 900 per tonne on its Q2 2025 earnings call, defining RMB 900 to RMB 1,000 per tonne as the target range for current operating scale.¹⁰ This metric serves as the primary indicator for assessing unit profitability in upcoming quarters.

↑ Back to Top

VI. The 2022–2024 Down-Cycle & The High-End Electronic Fabric Pivot

The setup for the crash was the oldest story in heavy industry: everyone read the same forecast at the same time.

Post-COVID, Chinese producers — Jushi very much included — added capacity into an outlook built on wind power buildout, automotive lightweighting, and infrastructure. Then Chinese real estate construction slowed sharply, wind installations went through a lull, and export markets absorbed less than hoped. The new furnaces arrived anyway, because furnaces ordered three years earlier arrive whether or not you still want them. Industry inventories spiked to historic highs. Discounting became universal.

The mechanism worth understanding is why this industry overshoots so violently. A tank furnace, once lit, runs continuously for years. Shutting it early means writing off the remaining campaign and paying to rebuild the refractory. So when prices fall, producers do not cut output — they cut price, because any price above the marginal cost of the gas and sand still contributes to the fixed cost you are paying regardless. Supply is nearly perfectly inelastic downward. The result is that prices fall until they hit somebody's cash cost, and the somebody in question is always the highest-cost producer, not the market-clearing one.

Management's response

Jushi's answer had three parts, and they were more disciplined than the industry average.

First, accelerated 冷修 cold repair: taking older furnaces down early for rebuild rather than running them to the end of campaign. This is counterintuitive but correct — if the market does not want your tonnes, the cheapest time to lose production is when production is worthless, and you come back with a larger, more efficient line. The headquarters line went down at 120,000 tonnes and came back at 200,000.¹⁰

Second, mix migration: shifting allocation away from low-margin construction roving toward wind energy high-modulus fibers, thermoplastics, and electronics.

Third, and most consequentially, doubling down on electronic fabric.

The electronic fabric bet

Here the technical explanation matters, because the AI connection is not obvious.

A printed circuit board is, physically, layers of woven glass cloth impregnated with resin, with copper traces printed on them. At low signal frequencies, the cloth is just a structural filler. At the frequencies AI accelerators and high-speed switches operate at, the glass itself becomes an electrical component — its Dk dielectric constant and Df dielectric loss determine how much of the signal survives the trip across the board, and its thermal expansion determines whether the board warps when a 700-watt chip heats it. So the industry needs cloth that is simultaneously thinner, flatter, more dimensionally stable, and electrically quieter than anything previously required. That is Low-DK fabric, and it is a genuinely different product from the standard 7628-grade cloth that has gone into consumer electronics for thirty years.

Jushi's push into this culminated on March 18, 2026 — thirty-three years to the day after the 3·18 Project — with the ignition at Huai'an, Jiangsu of a line producing 390 million meters of electronic fabric and 100,000 tonnes of electronic yarn annually, described as the world's largest single production line for the material, with thin and ultra-thin grades comprising about 30% of the mix and the line alone representing roughly 9% of global capacity.²⁷

The financial results have been dramatic. Electronic fabric prices climbed to around RMB 5.8 per meter by March 2026, up 38% year on year, and were reported at RMB 6.5–6.8 per meter by mid-May, with industry inventories below one week of sales.²⁴ In the first half of 2026, Jushi sold 544 million meters of electronic fabric and 1.70 million tonnes of roving, generating revenue of RMB 11.16 billion (up 22.50%) and net profit attributable to shareholders of RMB 2.93 billion — up 73.87%.²⁸ Operating cash flow rose 68.23% to RMB 2.42 billion, and weighted average return on equity improved to 9.06% for the half from 5.49% a year earlier.²⁸ The first quarter alone delivered RMB 1.267 billion of net profit on RMB 5.282 billion of revenue, at a 39.64% gross margin.²⁹

That the profit growth ran at more than three times the revenue growth is the single most informative fact in the half-year. It means the improvement came almost entirely from price and mix, not volume — operating costs rose only 4.56%.²⁸ This is operating leverage in a fixed-cost business working in the favorable direction, and it tells you exactly how violently the same leverage would work in reverse.

Falsification: is electronic fabric really an escape from the cycle?

The claim under test: that electronic fabric is a secular, high-margin growth engine insulated from industrial cyclicality — the "AI supply chain" framing that took the stock up 357% in six months.

The disconfirming evidence, from three directions:

First, the company itself. In June 2026, after the shares hit limit-up twice in three sessions on PCB-concept enthusiasm, China Jushi issued a clarification: its electronic fabric products are primarily standard and thin fabric series, and its low-dielectric and other specialty electronic fabric products had not formed orders or revenue.⁴ The company explicitly warned investors to be aware of the risks of hot-concept speculation and to make rational decisions.⁴ It repeated the point the following week.³⁰ When a company is publicly talking its own stock down, that is worth more than any analyst note.

Second, the competitive position in the actual high-end. The Low-DK segment is not an open field. As of 2025, 日東紡 Nittobo held approximately 59% and 旭化成 Asahi Kasei approximately 18% of the global Low-DK market — the two Japanese firms together holding over 80%.³¹ Among Chinese producers, Sinoma's Taishan Fiberglass has covered both first- and second-generation low-dielectric fabrics with certification completed and batch supply running to leading domestic and international customers, and 光远新材 Guangyuan has been supplying low-dielectric specialty cloth in volume.³²³⁰ Jushi's own disclosure in March 2026 described its low-dielectric yarn and fabric development, certification, and sampling as "proceeding in an orderly manner" — with its self-developed low-dielectric yarn having passed Nvidia certification and certifications with customers including 台光電子 Elite Material and Panasonic.³³ Certification is real progress. It is not revenue. And on this specific product, Jushi is a challenger, not the incumbent it is in roving.

Third, the historical base rate for this exact product. Electronic yarn and fabric have already had a violent boom-bust inside living memory. Heavy capacity additions across the Chinese industry in 2018 — taking domestic electronic-grade yarn capacity toward roughly 682,000 tonnes — collided with US-China trade friction and produced a severe 2019 price collapse, with producers cutting price to hold share.³⁴ The same industry then went through a further consumer-electronics inventory correction in 2022. Electronic fabric is tied to semiconductor and hardware inventory bullwhips, which are if anything more violent than construction cycles, not less.

The verdict: the claim survives in a narrowed form. Jushi genuinely is the volume leader in conventional and thin electronic cloth — it sold 1.062 billion meters in 2025 against roughly 960 million meters of nameplate capacity, holding a 35–40% share of the Chinese market.²³⁵ That business is enjoying a real, cash-generating up-cycle driven by AI-adjacent demand for standard and thin grades. What is not established is the higher-margin claim: that Jushi will capture the specialty Low-DK segment where the Japanese incumbents sit and where its domestic rival is already shipping. Management's own five-year target — roughly 15% share of high-end electronic cloth, with deliberate pacing to avoid excess capacity — is a statement about 2030, not 2026.¹⁰

The KPI that resolves it: the first reported revenue line for low-dielectric and specialty electronic fabric, and its realized price per meter relative to standard cloth. Until that number exists, the Low-DK story is optionality, not earnings.

And there is a capacity question sitting underneath all of it. In May 2026, Jushi committed RMB 4.431 billion to a new Huai'an line adding 50,000 tonnes of electronic yarn and 320 million meters of electronic fabric over eighteen months, at a projected pre-tax return near 11%.³⁶³⁷ In July, it committed a further RMB 2.405 billion for 250 million meters at Tongxiang, at a projected 13.3% pre-tax return.[^1]³⁸ Together that is 570 million meters of new annual electronic fabric capacity announced in nine weeks by a company that sold just over a billion meters in the prior full year. The market's reaction to the second announcement — an immediate limit-down — was not irrational. It was the market recognizing that the industry's oldest reflex had returned.

---

↑ Back to Top

VII. Management Assessment, Capital Allocation, & Governance Stress Test

For a company with such consistent operational performance, China Jushi's boardroom has experienced noticeable executive turnover.

The sequence unfolded rapidly: Chairman 常张利 Chang Zhangli resigned, with director 杨国明 Yang Guoming designated to serve as acting chairman in the interim.³⁹ In January 2025, 刘燕 Liu Yan was appointed chairman and director, ending Yang's interim role.⁴⁰ Yang had previously succeeded founder 张毓强 Zhang Yuqiang as general manager in March 2024, when Zhang stepped back from day-to-day executive duties. By the release of the 2025 annual report, Yang became chairman himself, replacing Liu after Liu reached statutory retirement age.¹

Having three chairmen in roughly two years is not inherently a red flag, as statutory retirement rules across Chinese state-owned enterprises generate mechanical executive shifts. However, it leaves the current chairman with a brief tenure at the top, meaning corporate continuity relies less on the chairman's office than on the experienced operational management team Yang rose through, alongside Zhang Yuqiang's ongoing influence.

The people

杨国明 Yang Guoming is a career Jushi operator whose background lies in furnace engineering and cost management rather than corporate finance or state administration. That operational focus directly influences management behavior. On the Q2 2025 earnings call, Yang maintained the pragmatic tone of a plant manager. Regarding the electronic fabric expansion, he noted that if the company pursued the buildout, it had to execute effectively and maintain low costs.¹⁰ He declined to provide interim updates on the four electronic fabric lines under development—covering first-generation, second-generation, low thermal expansion, and specialty grades—stating that management preferred to report completed metrics rather than discuss work in progress.¹⁰ While equity analysts typically push for greater transparency, refusing to pre-announce targets signals operational discipline, a posture reinforced by the company's cautious June 2026 disclosure.

Founder 张毓强 Zhang Yuqiang serves as vice chairman of China Jushi and remains the controlling figure at Zhenshi Group. He was the highest-paid director on the board, receiving RMB 4.75 million in compensation, compared to Yang Guoming's RMB 3.91 million.⁴¹⁴⁰ Over three decades after founding the business, Zhang continues to hold a board position while leading Zhenshi, one of Jushi's two primary anchor shareholders. Zhenshi has actively backed this commitment, purchasing additional shares through a program launched in November 2025.⁹ Controlling shareholder CNBM has similarly expanded its holding.⁸ That both major insider shareholders increased their financial exposure during 2025 and 2026 provides a clear alignment signal, though both buying programs began before the stock's major price run and do not represent independent valuation opinions.

Capital allocation: the record and the pressure point

Jushi's capital allocation track record has historically remained disciplined. The company funded its extensive furnace construction program primarily through operating cash flows and domestic debt rather than dilution via equity offerings, while maintaining consistent shareholder distributions. For 2025, Jushi declared a final dividend of RMB 1.9 per 10 shares alongside an interim payout of RMB 1.7 per 10 shares, returning a total of RMB 1.435 billion, or 43.66% of annual net profit.¹ It followed with an interim dividend of RMB 3.30 per 10 shares—totaling RMB 1.32 billion—for the first half of 2026.²⁸ Reinvesting cash into expansion while distributing over 40% of earnings reflects a tangible commitment to capital return in a capital-intensive cyclical industry.

However, rapid expansion has created balance-sheet pressure points. Headline leverage remains moderate, with a debt-to-asset ratio of 39.87% at mid-2026, down slightly year over year, and interest coverage improving to 27.23 times from 14.85 times a year earlier.²⁸ Yet working capital trends show expanding risk. Accounts receivable climbed 63.52% year over year to RMB 3.09 billion by mid-year, significantly outstripping 22.50% revenue growth over the same period.²⁸ A similar divergence emerged in the first quarter, when receivables grew 66.20%.²⁹ Sharply rising receivables during a price upswing can reflect higher volume delivery timing, but it can also point to looser credit terms offered to move inventory. Concurrently, Q1 2026 financial expenses more than doubled to RMB 131 million, up from RMB 52 million a year earlier, driven largely by foreign exchange losses.²⁹ Against RMB 6.8 billion in newly announced electronic fabric capital commitments, Jushi held approximately RMB 2.87 billion in cash at the end of the first quarter, set against RMB 3.63 billion in short-term debt, RMB 4.72 billion in long-term borrowings, and RMB 1.14 billion in non-current liabilities due within twelve months.²⁴

These figures do not indicate solvency stress, but they mean the coming two years will test whether electronic fabric earnings ramp fast enough to cover the capital expenditure required to build the lines.²⁴ Announcing two major capacity additions within nine weeks at the peak of a pricing cycle raises an essential capital discipline question: whether management is strategically allocating capital or aggressively expanding into a temporary price rally.

The governance overhang

A key structural issue for institutional investors remains the unresolved governance overlap within its state parent.

CNBM controls two of the world's largest fiberglass manufacturers: China Jushi and Taishan Fiberglass, the latter held through Sinoma Science & Technology. Both subsidiaries compete in identical product categories for the same customer base under a single ultimate parent. This structural overlap constitutes 同业竞争 horizontal competition, which Chinese listing regulations require controlling shareholders to eliminate.

CNBM formally committed to resolve this competitive conflict in 2017.⁴² However, the commitment remains unfulfilled.

In December 2024, CNBM notified China Jushi that it was extending the resolution window by two years.⁴² On May 13, 2026, China Jushi's board voted to approve a further extension.⁴³ Responding to regulatory inquiries in January 2026 regarding Sinoma's private placement, CNBM and its listed entity outlined potential structural remedies—including entrusted management, asset restructuring, equity swaps, and business realignments—promising to advance integration before December 2026.⁴⁴

Evaluating that track record demonstrates a clear pattern: nine years of formal commitments, multiple deadline extensions, and a resolution framework that still presents four theoretical options rather than a single binding plan. The underlying friction—valuation disparities between two listed entities, tax liabilities, and minority shareholder voting requirements—presents real structural hurdles. Yet the analytical takeaway is plain: corporate pledges have been repeatedly postponed, leaving final resolution subject to future regulatory approvals. Investors banking on a rapid consolidation have seen those expectations delayed for nearly a decade.

This governance delay carries practical market consequences. Without a mechanism to coordinate capital deployment, the world's two largest fiberglass producers under common control can independently add capacity into industry downturns. That lack of coordination amplified the 2022–2024 overcapacity crash and presents a similar risk for the ongoing electronic fabric expansion.

On operational grounds, management maintains strong credibility, having preserved cost leadership through a sharp downturn while publicly cautioning investors when market speculative enthusiasm outpaced fundamental order books. On corporate governance, however, the controlling shareholder's record of fulfillment remains poor. Investors must evaluate Jushi under the premise that horizontal competition will persist until a concrete, regulatory-approved transaction is executed, rather than relying on projected deadlines.

↑ Back to Top

VIII. Playbook: Business & Investing Lessons

1. Process power beats patent power in heavy manufacturing. Jushi's core advantage is not protected by intellectual property in an enforceable sense. It is protected by decades of operational experience—furnace geometry, burner tuning, refractory campaign management, bushing design, and sizing chemistry learned by doing, at scale, by an engineering team that remains in place. Competitors buying identical equipment will still register higher unit costs. For investors, this operational moat remains invisible on a balance sheet, surfacing primarily in a persistent margin gap against peers—as demonstrated by Jushi's roughly RMB 900 per tonne production cost advantage over Taishan Fiberglass.

2. The hybrid ownership edge carries a structural cost. State backing provided Jushi with the low cost of capital required to expand through severe industry downturns that forced Western competitors to retreat. However, that same structure created an unresolved nine-year conflict with a sister company under common state control. Capital markets must price both sides of the arrangement: patient state-backed capital offers significant value in a capital-intensive, cyclical sector, but a controlling shareholder managing competing assets while repeatedly extending compliance deadlines represents a persistent governance drag.

3. In continuous-process commodities, cost leadership buys survival rather than stability. Jushi remained cash-generative and continued building while weaker rivals sustained losses, yet its net profit still plummeted roughly 60% from peak to trough during the down-cycle. Cost leadership guarantees survival and market share gains during downturns, but it does not insulate earnings from severe industry overcapacity. Treating low-cost status as a defense against macro cyclicality misinterprets the nature of commodity manufacturing.

4. Tariff mitigation through overseas relocation is a decaying asset. Constructing production lines in Egypt secured over a decade of duty-free European market access—a creative move that pre-dated the international expansion of most Chinese peers. That advantage eroded in April 2026 when the European Union applied anti-dumping duties based on corporate ownership rather than plant geography. Any trade arbitrage reliant on regulatory definitions will eventually be narrowed by policymakers. Sustainable international expansion requires genuine local cost advantages—such as Egypt's low-cost natural gas—rather than pure tariff avoidance, which failed to yield profits in South Carolina.

5. Technical qualification does not equal commercial revenue. Securing Nvidia certification for low-dielectric yarn represents an important technical milestone. However, when stock speculation pushed share prices to limit-up levels in June 2026, Jushi disclosed to shareholders that its specialty low-dielectric fabrics had generated no commercial orders or revenue. Investors should evaluate speculative product lines based on current financial contributions and model the transition from customer qualification to realized sales conservatively. Given the industry's 2018–2019 electronic yarn downturn, technical achievements do not automatically translate into sustained pricing power.

6. Corporate actions reveal more than strategic statements. The most instructive signal in the fiberglass sector came not from Jushi, but from industry pioneer Owens Corning. After spending eight decades building the commercial glass reinforcement business, Owens Corning opted to divest the segment and subsequently accepted a lower valuation to expedite its exit. An incumbent abandoning the market it created provides a clear assessment of long-term returns on capital in heavy reinforcements. That exit validates Jushi's cost-based market dominance while highlighting a fundamental question: if through-cycle economics were compelling, why did the industry's legacy leader choose to reallocate capital elsewhere?

7. Operational and supply-chain efficiencies compound into moats. In low-value-per-tonne industrial commodities, competitive differentiation rarely stems from a single breakthrough. Instead, it accumulates through small, systemic advantages: direct access to river barge transport, long-term raw material supply contracts, lower natural gas tariffs, and in-house fabrication of precious-metal bushings. While individual logistics and procurement details rarely capture market attention, together they produce a unit-cost gap wide enough to reshape global market share.

---

↑ Back to Top

IX. Analysis, Risk Radar, & Bull vs. Bear Case

The three KPIs that matter

Evaluating Jushi's operational outlook and investment thesis requires monitoring three essential metrics:

1. Net profit per tonne of roving (RMB/tonne). This metric provides the cleanest indicator of whether Jushi's cost advantage remains intact. By isolating unit profitability from total sales volume, it reflects both price realization and operating cost discipline. It also aligns directly with management's target of RMB 900 to RMB 1,000 per tonne, following the roughly RMB 900 per tonne generated in the second quarter of 2025.¹⁰ If broader industry prices soften and Jushi's unit profit drops faster than market benchmarks, its cost buffer against Taishan Fiberglass and CPIC is narrowing.

2. Electronic fabric revenue mix and realized price per meter, with specialty grades disclosed separately. Electronic fabric represented 17.77% of total revenue in 2025.⁴ The product mix highlights the progress of high-margin reallocation, while realized price per meter reveals whether revenue gains stem from value-added product upgrades or standard fabric sold into a temporary price surge. The critical milestone—which has not yet appeared in corporate disclosures—remains initial commercial revenue from low-dielectric and specialty fabrics.

3. Overseas revenue share and the margin spread between domestic and international manufacturing bases. Overseas operations generated 37.4% of total revenue in 2024.¹² The profitability gap between regions indicates whether global expansion builds structural value or merely bypasses trade barriers—a contrast illustrated in 2024 by Egypt's 18.7% net margin alongside losses at the U.S. facility.¹² With European Union anti-dumping duties now applying to Egyptian exports and roughly half of upcoming capacity expansion targeted abroad, international margin performance will test execution across foreign markets.

Risk radar

Domestic supply discipline. Industry efforts to limit destructive competition (反内卷 anti-involution) have moderated rival expansion, with management noting during the second-quarter 2025 call that key competitors had scaled back capacity plans.¹⁰ However, Jushi itself committed RMB 6.8 billion to new electronic fabric projects within nine weeks in mid-2026, while Sinoma Science & Technology continues to advance major expansion projects. The structural overcapacity that depressed margins from 2022 through 2024 has not disappeared; higher market prices have simply paused its effects.

Input costs and energy volatility. Natural gas represents the primary variable expense in continuous glass melting. Sudden gas price increases compress overall profit margins even if Jushi maintains its relative unit-cost advantage over peers. Additionally, fluctuations in platinum and rhodium prices directly impact the upfront capital expenditures required for new production lines due to precious-metal bushing requirements.

Trade restrictions and carbon regulations. Beyond the April 2026 European Union anti-dumping duties, the EU Carbon Border Adjustment Mechanism presents an enduring structural challenge. By taxing the embedded carbon emissions of gas-fueled melting furnaces regardless of manufacturing origin, this mechanism cannot be bypassed through production relocation. Jushi's renewable energy deployment in Huai'an—which generated 562 million kilowatt-hours in 2025—offers only a partial offset.¹ Meanwhile, shipping disruptions near the Suez Canal represent an ongoing operational risk for exports from the Egyptian production base.

Working capital expansion and financing needs. Balance-sheet risks are accumulating as accounts receivable expand at nearly three times the rate of top-line revenue growth, financial expenses double due to foreign exchange losses, and major capital commitments hit the balance sheet concurrently.²⁸²⁹²⁴

Corporate governance and structural overlap. Strategic coordination remains hampered by unresolved horizontal competition under parent company CNBM, compounded by the operational complexities of sharing an ultimate owner with direct competitor Sinoma.

Technology substitution and qualification hurdles. High-cost carbon fiber continues to capture top-tier structural applications, though glass fiber retains the vast majority of the broad composite market. The primary near-term threat stems from customer qualification barriers, as established Japanese suppliers leverage long-standing technical certifications to restrict market entry in high-end electronic substrates.

The bull case

Jushi enters late 2026 as the lowest-cost producer in a consolidating global industry, following Owens Corning's exit from the glass reinforcements business. Its unit-cost advantage over domestic peers remained resilient throughout the 2022–2024 downturn. Strong demand across wind energy, automotive lightweighting, and electronic substrates enabled first-half 2026 net profit to grow more than three times faster than revenue, supported by a modest 4.56% increase in operating costs.²⁸ The Huai'an facility operates the world's largest single electronic fabric line at a time when industry inventories sit below one week of sales and electronic fabric prices have increased approximately 38% year-over-year.²⁴²⁷ Continued demand for artificial-intelligence hardware substrates provides substantial upward operating leverage, supporting both capital expenditure requirements and a dividend payout ratio exceeding 40%, while both CNBM and Zhenshi Group have expanded their equity stakes.

The bear case

Conversely, current earnings gains remain tied to commodity market cycles, and management is expanding aggressively into a temporary price surge. Structural demand for domestic construction roving may never return to historical peaks, permanently shrinking Jushi's main volume anchor. Meanwhile, the Egyptian tariff advantage ended in April 2026 following European Union trade measures that imposed a combined duty near 24%.¹⁷ Operations at the U.S. facility showed that Jushi's domestic cost structure cannot be replicated easily in Western jurisdictions. Crucially, the high-end artificial-intelligence narrative that fueled the stock's rally relies on specialty low-dielectric products that have generated no commercial revenue, in a market segment where two Japanese incumbents control over 80% of global market share and domestic rival Taishan Fiberglass is already delivering certified products.⁴³¹³² Despite past warnings regarding industry overcapacity, management approved 570 million meters of new electronic fabric capacity across nine weeks in mid-2026, relying partly on debt financing and assuming current peak prices persist. Finally, parent company CNBM's nine-year delay in resolving horizontal competition prevents coordinated supply management across state-owned assets.

Reconciling the two

Reconciling the bull and bear investment theses requires separating product segments and time horizons.

In glass fiber roving, the bull case is established while the bear case reflects broader commodity cycles rather than company-specific weakness. Jushi maintains a proven unit-cost advantage, ensuring operational survival through downturns. The central uncertainty is the timing and severity of the next industry oversupply cycle.

In electronic fabric, the dynamics are reversed. Current profit expansion reflects strong cyclical demand for standard and thin fabric series rather than a structural technological breakthrough. The broader claim that Jushi will evolve into a high-margin specialty substrate supplier with economics matching Japanese incumbents remains unproven. That narrative is tempered by management's June 2026 disclosure confirming a lack of commercial specialty orders, alongside earlier market entry by domestic competitor Taishan Fiberglass.

Consequently, characterizing Jushi's strategic shift as an immediate business transformation overstates current disclosures. The empirical evidence demonstrates a highly efficient commodity producer capitalizing on a cyclical market recovery while deploying capital into a debt-funded, technically viable, but commercially unproven high-end product expansion. The ultimate success of that expansion will depend on future revenue disclosures for specialty low-dielectric products.

↑ Back to Top

X. Epilogue & Strategic Outlook

Thirty-three years after Zhang Yuqiang and his partners registered a venture in Tongxiang on March 18, 1993, to build China's first domestic tank furnace, China Jushi lit the world's largest single electronic fiberglass production line in Huai'an on March 18, 2026—a facility designed around a smart, low-carbon manufacturing template supported by dedicated renewable power generation.²⁷¹

Across those three decades, a township collective drawing yarn from crucibles expanded into a global industrial manufacturer controlling roughly 30% of world fiberglass output—a scale and cost discipline that contributed to Owens Corning's decision to exit the reinforcements business entirely.

What remains unresolved is whether the 2026 Huai'an line signifies the same structural leap as the 1993 furnace. The original furnace represented a bet on process engineering that Jushi mastered to achieve global cost leadership. The 2026 expansion represents an entry into a market segment where Jushi operates as a challenger rather than an incumbent: technical qualification gatekeepers remain Japanese, corporate sibling Sinoma achieved earlier customer integration, and management has confirmed that its specialty low-dielectric products have not yet generated commercial orders or revenue.

What to watch from here

Three developments will clarify Jushi's strategic trajectory over the next 18 to 24 months:

1. Resolution of horizontal competition. The primary test is whether parent company CNBM converts its commitment to resolve horizontal competition into a binding, board-approved transaction before December 2026. A concrete structural resolution would eliminate a nine-year governance overhang and enable coordinated domestic supply management. Conversely, another deadline extension would signal persistent governance inertia.

2. Return on electronic fabric capital deployment. A second indicator is whether the RMB 6.8 billion in electronic fabric capital commitments announced in May and July 2026 delivers projected returns. Underwritten at estimated pre-tax returns of 11% to 13.3% during a sharp pricing rally, these projects require 18 months of construction. Historical precedent—such as the 2018 electronic yarn expansion that depressed 2019 pricing and the post-COVID roving additions that triggered the 2022 downturn—underscores the risk of adding significant capacity into temporary price spikes.

3. Commercial revenue from specialty products. The critical milestone remains the formal disclosure of commercial sales from low-dielectric and specialty electronic fabrics. Reclassifying Jushi from a commodity manufacturer into a high-margin artificial-intelligence hardware supplier depends directly on the appearance and growth of this revenue line.

These factors highlight two concurrent realities: China Jushi remains an exceptionally efficient heavy-industrial producer, but its high-end technology narrative currently outpaces verified financial results. The company spent 33 years proving it can manufacture commodity glass fiber at a lower unit cost than any global competitor. It has not yet proven it can manufacture specialized high-end fabrics superior to the Japanese incumbents that dominate that segment. Commodity cost control and specialized material innovation require distinct operational capabilities. Until specialty product sales appear as a distinct line item in financial disclosures, Jushi's investment thesis rests on operational cost leadership rather than structural business transformation.

---

References

1. 中国巨石股份有限公司2025年年度报告摘要 — 上海证券报, 2026-03-20 ↩↩↩↩↩

2. 中国巨石：2025年净利润同比增长34% 玻纤粗纱与电子布销量再创历史新高 — 财联社, 2026-03 ↩↩

3. AI算力带火高端玻纤！国产电子布迎来黄金发展期 — 证券时报, 2025 ↩↩

4. 3天2板中国巨石：公司电子布产品主要是普通布、薄布系列 低介电等特种电子布产品未形成订单及收入 — 每日经济新闻, 2026-06-17 ↩↩↩↩↩↩

5. 初心在身边丨中国巨石：嘉兴飞出世界玻纤领头雁 — 浙江在线, 2018-06-20 ↩↩↩↩↩

6. 中国巨石(SH:600176)：全球玻纤工业领导者——公开信息梳理与行业维度观察 — ZAKER新闻 ↩↩↩

7. 中国玻纤股份有限公司关于巨石集团有限公司增资并引入战略投资者的公告 — 巨潮资讯网 CNINFO, 2007-01-06 ↩

8. 中国巨石股东中国建材增持590.2万股 — 新浪财经, 2026-05-16 ↩↩

9. 中国巨石：获股东振石集团增持5127万股，金额达11亿元 — 网易财经 ↩↩

10. Earnings call transcript: China Jushi Q2 2025 sees strong profit growth — Investing.com, 2025 ↩↩↩↩↩↩↩↩↩↩

11. China's fiberglass company Jushi plans to raise investment, production in Egypt — China Daily, 2016-01-19 ↩↩

12. 中国巨石研究报告：走出周期底部，迎新需求周期 — 搜狐财经, 2025 ↩↩↩↩↩↩↩↩

13. China Jushi bringing 80,000-ton fiberglass production line to Richland County — South Carolina Department of Commerce ↩

14. SC fiberglass plant, one of state's largest projects, could fall victim to tariffs — The Post and Courier, 2018 ↩

15. Chinese industrial plant posed 'danger' to SC's health and safety; huge fine levied — The State via Yahoo News, 2023 ↩↩↩

16. EU Imposes new anti-dumping duties on glass fibre imports — Glass International, 2026 ↩

17. EU Anti-Dumping Duties on Glass Fibre from Egypt, Bahrain, Thailand — IndexBox, 2026-04-16 ↩↩↩↩

18. Jushi Egypt for Fiberglass Industry v Commission, Case T-540/20 — EUR-Lex, Court of Justice of the European Union ↩

19. Worldwide Fiberglass Industry — Featuring China Jushi, Owens Corning & Taishan Fiberglass Among Others — PR Newswire ↩↩

20. Owens Corning to Sell Glass Reinforcements Business to Praana Group — Owens Corning Newsroom, 2025-02-13 ↩

21. Owens Corning Enters Into Amended Agreement with Praana Group for Sale of Glass Reinforcements Business — Owens Corning Newsroom, 2026-04-14 ↩

22. Owens Corning Completes Sale of Glass Reinforcements Business to Praana Group — Owens Corning Newsroom, 2026-05-01 ↩↩

23. 2024年建材行业专题报告：玻纤全成本分析框架 — 未来智库, 2024-04-10 ↩↩↩

24. 中国巨石再掷44亿建电子布产线，杠杆与利润的一场赛跑 — 界面新闻, 2026-05 ↩↩↩↩↩↩

25. 【中泰建材&化工】中国巨石2024年年报点评：玻纤价格持续修复，龙头竞争优势凸显 — 新浪财经, 2025-03-21 ↩↩

26. 中国巨石2024年报解读：营收微增6.59%，净利润下滑19.70% — 新浪财经, 2025-03-19 ↩

27. 中国巨石的高端电子布已经量产，2025年就开始批量供货 — 东方财富网财富号, 2026-06-26 ↩↩↩

28. China Jushi H1 net profit surges over 70% to 2.93 billion yuan, electronic fabric volume and price gains stand out — BigGo Finance, 2026-08 ↩↩↩↩↩↩↩↩

29. 中国巨石2026年一季报解读：归母净利增73.48% 财务费用翻倍增长 — 搜狐财经, 2026-04 ↩↩↩↩

30. 中国巨石：公司低介电等特种电子布产品未形成订单及收入 — 每日经济新闻, 2026-06-24 ↩↩

31. 2026年中国电子布行业市场现状及趋势分析：高端电子布需求旺盛 — 华经情报网, 2026 ↩↩

32. 电子布持续提价，中国巨石们的拐点来了？ — OFweek新材料网, 2025-06 ↩↩

33. 中国巨石：公司低介电系列玻纤及电子布产品研发、认证及送样工作正在有序推进 — 新浪财经, 2026-03-23 ↩

34. 电子级玻纤行业发展现状分析 谨防电子玻纤纱产能建设过度 — 前瞻产业研究院 ↩

35. China Jushi to Invest USD653 Million in Expanding PCB Raw Materials Output — Yicai Global, 2026-05-14 ↩

36. 44亿加码电子布！千亿巨头年内股价翻倍，关注AI驱动业绩弹性 — 21世纪经济报道, 2026-05-14 ↩

37. 中国巨石股份有限公司第七届董事会第三十二次会议决议公告 — 上海证券报, 2026-05-30 ↩

38. 中国巨石拟24亿元加码电子布产能 — 中国金融信息网, 2026-07-16 ↩

39. 中国巨石：董事长常张利辞职，杨国明代行董事长职责 — 界面新闻 ↩

40. 中国巨石聘任60岁刘燕为董事长及董事 — 新浪财经, 2025-01-25 ↩↩

41. 中国巨石聘任53岁商德颖为董事，副董事长张毓强年薪最高475万 — 新浪财经, 2025-01-25 ↩

42. 中国建材拟再次延期履行避免同业竞争承诺两年 — 中国水泥网 ↩↩

43. 中国巨石股份有限公司关于公司控股股东及实际控制人解决避免同业竞争承诺延期履行的公告 — 证券日报, 2026-05-14 ↩

44. 中材科技就第二轮审核问询函回复 明确同业竞争解决方案及2026年底前整合安排 — 新浪财经, 2026-01-16 ↩