Zangge Mining: A Copper Associate It Does Not Control Now Drives Four-Fifths of Profit, and CNY 76.15 Already Pays for It

The task card's ownership starting point has also moved. Zijin originally acquired 50.1% of Julong in 2020, but Zangge's current project page states that a Zijin entity now holds 57.35% and Zangge holds 30.78%. Zijin also controls 26.01% of Zangge directly as of June 2026. On an economic look-through basis, Zijin's direct 57.35% plus 26.01% of Zangge's 30.78% works out to about 65.35% of Julong. Its effective governance influence is stronger still because Zijin controls the Zangge board and therefore the listed company's exercise of its Julong shareholder rights, subject to directors' duties and related-party rules. The roughly 58% economic interest in the task card is stale.

That same overlap creates the most important governance tension. Zijin controls Zangge and is the majority owner and operator of Julong, the asset producing nearly four-fifths of Zangge's profit. The acquisition documentation is unusually explicit about protections: Zijin undertook to avoid unfair related-party transactions, not to occupy Zangge's funds, to abstain on related-party shareholder votes, to preserve Zangge's independence, and to resolve existing lithium-related competition within 60 months after obtaining control. New potash or lithium opportunities that could compete with Zangge are supposed to be offered to Zangge on a priority basis. Those undertakings improve the formal framework, but they cannot give Zangge minorities operational control over Julong's mine plan, dividends or phase-three capital allocation.

The transfer itself was negotiated in January 2025 at CNY 35.00 per share for 392.25m shares, an aggregate CNY 13.729bn. Zijin already held roughly 0.18% through controlled entities, and subsequently purchased another 6.84m shares in February and 9.06m shares in April at prices between CNY 30.09 and CNY 34.00. By the April revised ownership report, Zijin International itself held 408.15m shares; the June 2026 interim report shows that holding at 26.01%. The control-transfer shares and other shares held by the acquirer/concert parties are subject to an 18-month statutory acquisition lock-up. The former controlling shareholder also waived voting rights over 79.02m shares for 18 months following closing and undertook not to retake control while Zijin remains controller.

Control formally passed in spring 2025. Zijin's subsequent disclosures place the share-transfer closing on April 30 and the board/senior-management reorganisation in May. The governance agreement gives Zijin the chairman, a board majority, the executive vice-president and CFO nominations, while the former controlling shareholder side nominates the general manager and two non-independent directors. The parties agreed that this basic arrangement should remain for six years while their relative shareholdings remain substantially unchanged. That structure combines Zijin's capital discipline and mining systems with continuity from the old operating team, but it also means Zangge is not simply a wholly absorbed Zijin subsidiary.

Potash is the cleaner business. In H1 2026, Zangge produced 510,300 tonnes of potassium chloride and sold 525,300 tonnes. The average realised price including VAT was CNY 3,055.25 per tonne, up 7.4%, while average sales cost fell 2.02% to CNY 975.75 per tonne. The resulting 65.19% gross margin was up 3.35 percentage points. Zangge's Qarhan mining-right area is reported at 724.35 square kilometres and its installed potash capacity at around 1.2mt per year; 2026 production and sales plans are 1.0mt and 1.04mt.

Lithium is where capacity headlines are most likely to mislead. At Qarhan, Zangge's own 2026 production and sales target is 11,000 tonnes of lithium carbonate. H1 output was 5,400 tonnes and sales 3,990 tonnes. Average realised price including VAT was CNY 164,900 per tonne and average sales cost CNY 42,200, producing a 71.08% gross margin. At Mami Cuo, the project-level 2026 output target is 20,000–25,000 tonnes, while Zangge's economic entitlement is only about 5,000–6,000 tonnes because its look-through holding is approximately 26.95%. The project completed commissioning and obtained trial-production approval by H1. This resolves much of the conflicting guidance: 11,000 tonnes is consolidated Qarhan output; 20,000–25,000 tonnes is Mami project's own 2026 target; roughly 5,000–6,000 tonnes is Zangge's economic share of Mami, not consolidated tonnes.

The market has already repriced all of this dramatically. The September 4, 2026 close of CNY 76.15 is more than twice the CNY 35 control-transfer reference price. The current 52-week range is about CNY 52.61–97.28. September 4 was also the ex-dividend date for a CNY 1.00 per-share interim cash dividend, so part of that day's reported 3.52% decline is mechanical.

The share price now carries a materially different narrative from early 2025. Then, investors were buying a governance reset and a future Julong phase-two ramp. Today, phase two is operating, copper earnings have already more than doubled Zangge's first-half profit, lithium pricing has recovered sharply from 2025 levels, and Mami is in trial production. The remaining upside increasingly requires more than successful construction: it requires sustained copper margins, reliable Julong cash distributions, genuine Mami ramp-up, and eventually phase-three value.

The bull/bear disagreement follows directly. Bulls see an extraordinary copper asset whose phase two has barely reached full utilisation, a controller with a demonstrated ability to construct high-altitude mines, two low-cost salt-lake businesses and a balance sheet with only a 7.77% liability ratio at June 2026. Bears see a CNY 119.47bn equity that earns nearly four-fifths of its profit from a 30.78% stake in a mine controlled by the same party that controls the listed company, while the stock already discounts much of phase two and some phase-three success.

Qualitative portrait: company in transition. The operating business is moving from founder-era potash producer to a Zijin-controlled listed mining platform. Its highest-value asset is already outside the consolidation perimeter; its most visible lithium project is also partly outside it. The transformation can improve asset quality and capital allocation without ever making Zangge a conventional integrated miner. That distinction should remain central to valuation.

Company vertical history

To read Zangge's history, separate the legal stock-market vehicle from the economic business now inside it. The Shenzhen-listed corporate shell dates to the 1990s. Today's investment identity was created much later, through a major restructuring that injected the Zangge potash operation into the listed company. CSRC approval in January 2016 allowed the predecessor listed company to acquire 99.22% of Zangge Potash through an asset-and-share restructuring. From that point, salt-lake resources rather than the listed shell's earlier activities became the economic centre of the equity.

The potash injection created the modern company

The 2016 restructuring solved two problems simultaneously. The listed predecessor had weak legacy economics and going-concern uncertainty, while the private Zangge salt-lake business needed a listed capital-market platform. The reverse-merger-style transaction matters more for today's investors than the original 1990s listing story. It placed a very high-margin Qarhan potash producer inside an A-share vehicle, but it also imported a concentrated ownership structure and the governance culture of the former controlling shareholder.

The early post-restructuring thesis was straightforward: low-cost brine extraction, scarce potassium resources and high gross margins. Company comparisons at the time showed potassium chloride gross margins in the same broad range as the major Qinghai salt-lake producer. The problem was that high operating margins did not translate into institutional quality. What came next permanently changed how the historical accounts should be read.

The governance failure was economically real, not cosmetic

In 2019, the Qinghai CSRC found that the company had used fictitious trading transactions during 2017–2018 to inflate revenue and profit. The regulator identified CNY 0.132bn of overstated 2017 revenue and CNY 0.468bn in 2018, with overstated 2018 profit equal to 29.9% of the reported total. It also found fictitious receivables/prepayments and CNY 2.214bn of non-operating fund occupation by the controlling shareholder and related parties, of which CNY 2.164bn remained outstanding at June 30, 2019.

The CSRC imposed a five-year securities-market ban on then-controller and chairman Xiao Yongming and three years on another executive. The five-year period from the November 2019 decision has now elapsed. The event remains essential to any historical quality assessment. Reported 2017–2018 revenue, profit and working-capital trends cannot be treated as clean observations of the underlying business. The case also explains why Zangge should not receive a full governance premium merely because the mine assets themselves are attractive.

The clean-up extended into subsequent years. In January 2022 the company cancelled 390.49m shares for a total CNY 1 consideration as compensation for unfulfilled restructuring performance commitments, reducing total shares from about 1.971bn to 1.580bn. The same annual-report period also disclosed a criminal judgment involving Xiao Yongming. Per-share comparisons that span this period therefore have to reckon with the enormous change in the share count as well as the accounting restatements.

Julong changed the source of profit

The strategic turn came when Zangge became a 30.78% shareholder in Julong Copper. Julong's phase-one project entered production in late 2021, and from 2022 onward investment income became increasingly important. The company that investors had understood primarily as a potash/lithium producer acquired exposure to a world-scale porphyry copper deposit without becoming its operator. Zijin, after acquiring control of Julong, brought large-scale mine-design, construction and operating capability that Zangge itself had not demonstrated at that scale.

The timing was fortuitous. Potash and lithium both enjoyed extraordinary commodity conditions in 2022, pushing Zangge's annual revenue to roughly CNY 8.2bn and attributable profit to about CNY 5.66bn. Those earnings then normalised as lithium and potash prices retreated: revenue declined to CNY 5.226bn in 2023 and CNY 3.251bn in 2024, while attributable profit fell to CNY 3.420bn and CNY 2.580bn. The decline in consolidated commodity earnings was partly cushioned by the growing Julong contribution. The audited 2025 numbers then turned upward again: revenue CNY 3.577bn and attributable profit CNY 3.852bn.

This period also exposes why revenue growth is a poor headline indicator for Zangge. Between 2024 and 2025 revenue increased only 10.0%, but attributable profit rose 49.3%. By the first nine months of 2025, Julong had produced 142,500 tonnes of copper and contributed CNY 1.950bn of investment income, 70.89% of Zangge's net profit. The earnings model was already becoming more copper-heavy before phase two began.

The Zijin control transfer was the second corporate founding

On January 16, 2025, Zijin International agreed to buy 392.25m Zangge shares at CNY 35 each for CNY 13.729bn. Including Zijin-controlled entities' pre-existing 0.18% stake, the initial transaction was designed to take the Zijin side to 25%. The revised April filing, after share cancellation and additional secondary-market purchases, showed Zijin International itself at 408.15m shares and the Zijin concert group at about 26.18%.

The transaction was deliberately a control deal despite a relatively modest absolute ownership percentage. The former controller agreed to waive votes on 79.02m shares for 18 months; the sellers undertook not to seek control while Zijin remains controller; and Zijin was allotted a board majority. It also nominated the chairman, executive deputy manager and CFO. The former controller side retained the right to nominate the general manager and two non-independent directors, preserving operational continuity.

This was a genuine change of corporate regime. Zijin did more than buy a passive stake: Zangge became a consolidated subsidiary of Zijin at group level, while remaining separately listed. The share transfer closed at the end of April 2025, and the governance reorganisation followed in May. For Zangge minority holders, the consequence is paradoxical. The company gained a stronger mining controller, yet the controller also became even more economically dominant over Julong, the main source of Zangge's profit.

The formal safeguards are extensive. Zijin pledged arm's-length pricing for unavoidable related-party transactions, no improper fund occupation or guarantees, abstention where appropriate, and organisational/financial independence. It also acknowledged lithium-business overlap and promised a solution within 60 months, potentially through asset restructuring, business delineation or entrusted management. New competitive lithium or potash opportunities are supposed to be offered preferentially to Zangge.

That undertaking is potentially a future source of asset injections, but any such conclusion today is speculative. There is no disclosed commitment to inject a particular Zijin lithium asset, no agreed valuation, no timetable beyond the 60-month competition-remedy window, and no evidence of a planned take-private or minority squeeze-out. The six-year governance arrangement and 18-month lock-up are more consistent with preserving Zangge as a separately listed vehicle in the near term.

Phase two turned the copper option into current earnings

Julong's second phase was approved with about CNY 17.46bn of estimated investment. Construction lifted processing capacity from 150,000 tonnes per day to 350,000 tonnes per day. The approved project was to be funded by Julong itself, which reduced the risk that Zangge would need to commit a large amount of consolidated cash simply to preserve its 30.78% holding.

Phase two officially entered production on January 23, 2026. Julong produced 60,400 tonnes of copper in Q1 alone and 134,000 tonnes in H1. Zangge recognised CNY 1.314bn of Julong investment income in Q1 and CNY 2.839bn for H1. The acceleration was so large that an earnings stream still classified as “investment income” became the dominant economic business of the listed company.

The next concept, phase three, would push Julong toward roughly 200mt of annual ore throughput and around 600,000 tonnes of annual copper if approvals, engineering and economics all support it. That would make the mine one of the world's largest by throughput. Keep phase two and phase three sharply apart: phase two is built and producing; phase three is a study and strategic plan, not an approved producing asset.

Lithium has moved from narrative to execution test

Zangge's salt-lake lithium operation at Qarhan is small compared with Julong but has exceptionally low reported unit costs. The more important growth project is Mami Cuo in Tibet. The company has an indirect economic interest of about 26.95%; the project contains reported LiCl resources equivalent to about 2.18mt of lithium-carbonate equivalent and is planned initially around 50,000 tonnes per year of lithium-carbonate capacity, with a larger long-term concept.

Mami's 2026 milestone is trial production rather than design capacity. The H1 report says commissioning had been completed and trial-production approval obtained. Management's 2026 output expectation of 20,000–25,000 tonnes is already less than half of phase-one nameplate capacity. Zangge's economic share is only roughly 5,000–6,000 tonnes. That gap between nameplate, project output and economic entitlement is the right way to track the asset.

Qarhan itself provided a warning in 2025 that resource rights matter as much as extraction technology. The lithium operation temporarily suspended production amid mining-right and permit issues before the company advanced the relevant licensing process. At a salt lake, the plant is replaceable; the legal right to extract the brine and associated minerals is the real scarce asset. That experience is why I assign no value to an assumption that mining and water permissions renew automatically forever.

The vertical story therefore compresses into four economic stages: a 2016 potash reverse merger; a 2017–2019 governance failure; a 2020–2024 transition toward copper and lithium; and a 2025–2026 Zijin-controlled phase in which Julong phase two has made copper the main earnings driver. Each stage left something durable: salt-lake cost advantage, a governance discount, a minority stake in a world-scale copper mine and, now, a much stronger controller.

Financial vertical review

The audited numbers show a highly cyclical consolidated business whose earnings quality changed structurally when Julong became material. The five-year revenue arc is more useful than a ten-year uninterrupted series because 2017–2018 accounts were affected by the CSRC findings and the share base was subsequently restructured.

2023–2025 audited figures are from the 2025 annual-report comparison table; the earlier figures are shown rounded because the restructuring and historical-accounting context make false precision unhelpful.

The 2022 peak came from price as much as volume. Zangge did not suddenly multiply its potash plant several times over; the commodity environment allowed roughly one million tonnes of potash and about ten thousand tonnes of lithium carbonate to earn extraordinary margins. That matters because a conventional growth multiple on 2022 profit would have capitalised a commodity windfall. By 2024, lower lithium and potash prices had compressed consolidated revenue dramatically even though the physical resource position had not deteriorated.

The 2025 recovery came from a different source. Revenue increased from CNY 3.251bn to CNY 3.577bn, only 10%, while attributable profit increased from CNY 2.580bn to CNY 3.852bn, 49.3%. Julong's equity-method earnings explain much of that divergence. The first nine months alone produced CNY 1.950bn of Julong investment income.

The first half of 2026 amplified the change. Revenue of CNY 2.065bn rose 23.05%; attributable net profit of CNY 3.638bn rose 102.09%; ex-nonrecurring profit of CNY 3.706bn rose 104.92%. Operating cash flow was CNY 0.984bn, up 17.99%. Assets were CNY 18.602bn and attributable equity CNY 17.197bn, leaving the reported liability ratio at just 7.77%.

The balance sheet is one of the strongest parts of the investment case. At June 2026, liabilities were only about CNY 1.4bn against CNY 18.6bn of assets. It carries nothing resembling the economic capital intensity of Julong because Julong's mine and phase-two funding sit inside the associate, not Zangge's consolidated balance sheet. This is both an advantage and an accounting warning: Zangge shareholders receive a share of Julong earnings without showing the mine's debt/capex gross on Zangge's statements, but they also lack direct control of those financing decisions.

Cash conversion has to be read differently here. In 2025, quarterly operating cash-flow figures sum to approximately CNY 2.10bn versus CNY 3.852bn of net income, a 54.5% ratio. H1 2026 operating cash flow of CNY 0.984bn was only 27.0% of attributable net profit. A superficial reading would call that poor cash conversion. The accounting reason is that Julong dividends are not the same thing as Zangge's consolidated operating cash flow: equity-method profit is booked in earnings while cash distributions from the associate are recorded separately in the cash-flow statement.

The cash is not entirely trapped. Zangge disclosed CNY 1.539bn of cash dividends received from Julong on March 24, 2026, and a further CNY 0.616bn payment was disclosed in August. Those two specifically identified distributions alone total CNY 2.155bn. The timing does not exactly match the H1 accounting period, so they should not be mechanically divided by H1 investment income, but they prove that Julong's equity earnings have been accompanied by substantial cash passthrough rather than being purely accounting accruals.

That makes the five-year “OCF/net income” statistic less informative here than for an ordinary manufacturer. I would not value Zangge by consolidated OCF alone because it would effectively value Julong's dividends at zero. The more relevant owner-earnings concept is consolidated operating cash generation plus sustainable cash received from Julong, less maintenance capex and any recurring corporate cash leakage. The company does not disclose the exact maintenance/growth capex split; in the valuation model I use approximately CNY 0.25bn a year of maintenance capex for the consolidated mature operations as an analytical assumption and value growth projects separately.

Capital returns are substantial. The 2025 annual distribution was CNY 1.50 per share, and the 2026 interim distribution was CNY 1.00 per share. The latter used a dividend base of 1.5639bn shares and went ex-dividend on September 4. The company also authorised a CNY 0.2–0.4bn share repurchase in March 2026 at a ceiling of CNY 85.38 per share, although those shares were intended for an employee ownership/incentive plan rather than immediate cancellation, so I do not treat the full repurchase as equivalent to a permanent buyback yield.

Returns on equity look exceptional, but part of the reason is the Julong associate. ROE was 27.62% in 2023, 19.40% in 2024 and 24.71% in 2025; H1 2026 weighted ROE was already 21.19%. Zangge can earn a return on its historical Julong investment without funding the full mine asset base on its own balance sheet. That is economically attractive, yet it means ROIC comparisons with fully consolidated miners can exaggerate Zangge's apparent capital efficiency.

Price and valuation history

Zangge's share-price history reads as a sequence of changing narratives rather than as a commodity chart.

The first modern rerating followed the 2016 potash-asset injection. The market increasingly treated the stock as a scarce A-share salt-lake resource company. The 2019 enforcement case interrupted that narrative; its findings went to the credibility of the accounts themselves. The subsequent clean-up, performance-compensation share cancellation and Julong acquisition rebuilt the equity story around tangible mining assets rather than the old listed shell.

The 2021–2022 surge was a classic commodity rerating. Potash and lithium prices climbed, Julong phase one began production and Zangge's earnings rose to a cyclical peak. The fall through 2023–2024 corresponded to the collapse in lithium economics and lower consolidated revenue. The company's market capitalisation history shows a sharp 2021 expansion, a large 2022 decline and then renewed acceleration into 2025.

The January 2025 Zijin transaction reset the valuation centre. CNY 35 was more than a financial purchase price: it created a control reference point and gave minority shareholders a signal that Zijin saw enough value in Zangge's resource package, Julong interest and strategic optionality to pay CNY 13.729bn for the control block. The market then started valuing Zangge partly through Zijin's expected ability to operate Julong and discipline capital allocation.

Phase-two progress drove the next leg. By September 2025 the Julong expansion was already commissioning major processing systems; it officially entered production on January 23, 2026. The stock's 52-week high of CNY 97.28 coincided with the period when the market was capitalising a much larger future copper earnings stream.

Lithium helped the narrative in 2026 but was not the primary earnings driver. During the first nine months of 2025, Zangge's average lithium-carbonate selling price was only CNY 67,306 per tonne and its lithium gross margin was 31.2%; H1 2026 average realised price had recovered to CNY 164,900 and gross margin to 71.08%. That is a dramatic cyclical recovery. Yet first-half lithium revenue was only CNY 0.582bn, while Julong investment income was CNY 2.839bn. Attribute the share-price response mainly to the copper/ramp combination, with lithium as a secondary option.

At CNY 76.15, Zangge trades at about CNY 119.47bn of market value. Trailing-twelve-month attributable profit, calculated as 2025 profit plus H1 2026 profit minus H1 2025 profit, is about CNY 5.69bn. That gives a headline trailing P/E of about 21.0 times. Annualising H1 2026 profit gives approximately CNY 7.28bn and a 16.4-times run-rate P/E. Those numbers look moderate until one remembers that CNY 2.839bn of H1 profit came from a 30.78% minority associate.

The current price/book ratio is approximately 6.9 times June 2026 attributable equity. That is high for a conventional resource producer but not necessarily irrational for a company whose associate interest is carried at accounting book value far below a reasonable economic NAV. P/B tells us more about the inadequacy of book value for Julong than about cheapness or expensiveness.

Try a stress test: remove potash and lithium entirely. Julong's H1 investment income annualises to about CNY 5.68bn. The current CNY 119.47bn market value is still roughly 21 times that annualised equity-method income. The market is not paying a low multiple merely because copper dominates earnings; it is paying for Julong's growth, phase-three optionality and the separate salt-lake assets as well. Conversely, if Julong disappeared, the remaining potash/lithium operations could not justify the current market value on current earnings. Copper is the earnings dependency that matters.

Reading the daily move around the September 4 close also requires an ex-dividend adjustment. Yahoo Finance records the CNY 76.15 close and notes that September 4 was the ex-date for the CNY 1.00 dividend. The day's reported 3.52% decline overstates the change in underlying economic value by roughly the dividend amount.

Business model and moat

Zangge's business model contains three forms of ownership, and each one carries its own economics.

Qarhan potash is the classic operating business. Zangge controls the producing subsidiary, extracts brine, processes potassium chloride and sells the product. Volume is relatively stable, realised potash price and unit cost drive margins, and the key asset is the mining/brine right rather than a sophisticated downstream brand. H1 2026 output of 510,300 tonnes and cost of CNY 975.75 per tonne demonstrate the scale and cost advantage.

Qarhan lithium uses the same broad salt-lake resource logic but with more processing complexity and greater commodity-price volatility. At 5,400 tonnes of H1 production it is not a large lithium producer in global terms. The economic quality comes from unit cost: H1 average sales cost was CNY 42,200 per tonne against a realised price of CNY 164,900. When lithium prices collapse, that cost position protects the business; when prices rebound, the operating leverage is severe.

Mami is an equity option on future salt-lake production. A project-level nameplate of 50,000 tonnes should not be added to Zangge's consolidated capacity as though the listed company owns the whole plant. Its look-through interest is roughly 26.95%, and the 2026 project-level production target is only 20,000–25,000 tonnes. The right economic exposure for 2026 is about 5,000–6,000 tonnes before considering any ramp shortfall.

Julong is different again. Zangge owns 30.78% of the company but does not operate it. H1 2026 Julong revenue, net profit and mine output belong to the associate; Zangge recognises only its proportional profit. So segment revenue tables cannot compare the importance of the copper business with potash and lithium. The copper business is nearly absent from consolidated revenue and dominant in net income at the same time.

Cost structure and operating leverage

Potash has strong operating leverage to price but relatively stable physical economics. Using H1 2026 numbers, the realised price including VAT was about three times unit sales cost. Even a sizeable potash-price correction leaves a material contribution margin before fixed corporate costs. The larger operational risks are production disruptions, brine/resource constraints and regulatory limitations, rather than an inability to cover cash extraction costs at ordinary potash prices.

Lithium is far more convex. A change of CNY 50,000 per tonne in lithium price against roughly 11,000 tonnes of annual production changes pre-tax segment economics by several hundred million yuan, while the cost base is much less volatile in the short run. The 2025–2026 experience shows the effect: nine-month 2025 average selling price was only CNY 67,306 per tonne; H1 2026 realised CNY 164,900. Gross margin moved from 31.2% to 71.08%.

Julong has the largest earnings torque because phase two simultaneously increases volume and exposes the expanded mine to copper prices. H1 net profit of CNY 9.223bn on CNY 15.004bn of revenue is an exceptional 61.5% net margin at the associate level, helped by copper plus gold, silver and molybdenum by-products. A lower copper price will hit both the absolute profit pool and the multiple investors are willing to pay for the minority stake.

The real moats

The strongest moat is geological and legal access to low-cost brines and an extraordinary copper orebody, not brand or customer lock-in. Qarhan provides a scarce potash resource with established infrastructure, large-scale evaporation/extraction systems and unit economics that have survived multiple commodity cycles. Julong's reported 25.88mt of identified copper resources make it the largest recorded copper resource base in China, while phase two has shown that the orebody can be processed at enormous scale despite altitude and low grades.

The second moat is cost position. Potash H1 unit sales cost below CNY 1,000 per tonne and lithium cost around CNY 42,200 per tonne create room to remain profitable through prices that would stress higher-cost marginal suppliers. Cost advantage is particularly valuable in lithium, where the commodity itself is homogeneous and producers cannot rely on brand pricing.

The third moat, though it belongs primarily to the controller rather than Zangge itself, is high-altitude project execution. Julong phase two was built in a severe Tibetan operating environment and reached production with a CNY 17.46bn project that more than doubled processing capacity. This capability reduces development risk for Julong and may help Mami, but Zangge minority shareholders should be precise about who owns the capability: Zijin and the Julong operating organisation do.

Technology is useful but not a standalone impregnable moat. Zangge has practical salt-lake extraction know-how and has improved lithium recovery processes, yet other Chinese salt-lake developers also possess adsorption, membrane and extraction technologies. The durable advantage is the combination of low-cost brine chemistry, rights, infrastructure and operating experience.

Brand, network effects and customer switching costs are essentially irrelevant. Potassium chloride, lithium carbonate and copper concentrate are commodities. A buyer chooses on specification, price, reliability and logistics. Zangge's economics therefore depend much more on the left side of the cost curve than on downstream pricing power.

Management and governance after the transfer

A Zijin nominee chairs the new board, while the former controlling shareholder side retains representation and the right to nominate the general manager under the control-transfer arrangement. The split was meant to keep the local operating organisation in place while giving Zijin board control, finance oversight and strategic authority.

That is preferable to a sudden purge of local expertise, but management credibility now has to be assessed in two pieces. Zijin has already demonstrated project execution at Julong. Zangge's own post-transfer listed-company capital allocation record is only about a year old. The legacy company's 2019 enforcement history prevents me from assigning high credibility to the corporate institution based on assets alone.

Capital allocation has several positives. Julong became a transformative investment; phase two was financed within Julong rather than by a large Zangge equity call; Zangge has distributed substantial cash dividends; Mami financing is partly contained in an investment structure. The main future test is whether Zijin's related-party position routes attractive projects and cash flows fairly between the parent, Zangge and Julong.

The competition undertaking is especially important. Zijin acknowledged existing or potential lithium overlap and promised a remedy within five years. It also promised that newly arising potash or lithium opportunities that could compete with Zangge would be offered preferentially to the listed company. That could turn Zangge into Zijin's designated listed platform for some salt-lake assets; it could also generate complex related-party asset transactions whose valuation matters as much as the strategic logic.

Industry and cycle

Zangge is exposed to three commodity cycles that share very little except that each involves a mine or brine resource.

Potash is fundamentally an agricultural-input cycle. Crop nutrient requirements and food-security policy support demand, while supply concentrates in a limited number of geological basins and producing countries. Zangge's economic edge comes from domestic brine resources in a country that structurally values potash security. Its H1 2026 realised price of CNY 3,055 per tonne and 65.19% gross margin indicate that the domestic market remained favourable at the base date.

Lithium is an inventory-and-capacity cycle layered on top of electric-vehicle and energy-storage demand. It has much faster supply responses and more violent price swings than potash. Zangge's own numbers capture the cycle better than any narrative: average lithium selling price of CNY 67,306 per tonne in the first nine months of 2025 versus CNY 164,900 in H1 2026. Guangzhou Futures Exchange data around the base date showed the September 2026 lithium-carbonate contract around CNY 155,000 per tonne, broadly consistent with the recovery evident in Zangge's realised pricing.

A much larger global industrial cycle drives copper: construction and manufacturing in the near term, grid investment, electrification, data centres and energy infrastructure over longer periods, and a slow mine-supply response. Julong's phase-two expansion entered production into a fundamentally different cycle from Zangge's lithium rebound. Shanghai Futures Exchange data confirm active domestic copper pricing in CNY per tonne on the September 4 base date; the valuation below deliberately uses a normalised copper deck rather than extrapolating a single spot print.

Dated commodity decks

The following are valuation assumptions, not forecasts disguised as facts.

The potash base case is close to H1 2026 realised price of CNY 3,055.25. The lithium base case is slightly below H1 realised CNY 164,900 and close to September GFEX futures levels around CNY 155,000. The copper deck is intentionally normalised rather than tied to the highest 2026 quotations because Julong is a multi-decade asset.

At one million tonnes of potash sales and roughly CNY 1,000 per tonne unit cost, those price assumptions produce approximate pre-corporate gross profit of CNY 1.39bn, CNY 1.80bn and CNY 2.21bn. At 11,000 tonnes of Qarhan lithium and CNY 42,200 per tonne cost, the lithium deck produces approximate gross profit of CNY 0.51bn, CNY 1.00bn and CNY 1.48bn. These calculations normalise VAT and are sensitivity illustrations, not management guidance. The scale difference explains why lithium can matter to incremental profit but still not match Julong's current earnings contribution.

For Julong, the same exercise is necessarily rougher because copper comes with gold, silver and molybdenum, and mine grade/cost changes with the ore plan. At a normalised 325,000 tonnes of annual copper, I model Julong net profit around CNY 14bn, CNY 20bn and CNY 26bn under the three decks; Zangge's 30.78% economic share would be roughly CNY 4.31bn, CNY 6.16bn and CNY 8.00bn. H1 2026 actual Julong net profit of CNY 9.223bn on only 134,000 tonnes of copper shows that these are not aggressive at the middle of the range if 2026 metals pricing persists.

Regulation, altitude and resource tenure

Qinghai and Tibet convert permitting from a generic risk factor into an operating variable. Qarhan's value rests on mining and brine-extraction rights over its reported 724.35-square-kilometre area, not merely on processing equipment. The company suspended lithium production temporarily in 2025 while licensing issues were addressed, which proves that a low-cost plant cannot produce without the legal right to extract the relevant mineral-bearing brine.

I could verify the mining-right area and the subsequent licensing progress but not, from the primary materials accessible in this research session, one unambiguous current expiration date and complete water-allocation schedule covering all Qarhan rights. That is a genuine information gap rather than something to fill with an assumed renewal date. The valuation therefore treats the rights as continuing but applies a resource-tenure discount in the conservative case.

Mami has a different constraint. The project is at more than 4,000 metres in Tibet, with weather, construction seasons, power, logistics and environmental permitting all constraining ramp speed. H1 2026 brought approval for trial production. That removes one hurdle but does not establish steady-state recovery, product quality or 50,000-tonne annual output.

The 20,000–25,000-tonne 2026 Mami target itself is a useful reality check against capacity marketing. Management is not assuming immediate utilisation of the phase-one design. A valuation that capitalises 50,000 tonnes from day one would be analytically wrong.

The company's Laos potash option illustrates the same point outside China. Zangge plans a 2mt-per-year Vientiane potash project, but H1 2026 said feasibility work, environmental assessment and licence procedures were still under way and that a Lao National Assembly Standing Committee resolution had complicated rights processing. I assign very little base-case value to the project until those rights are resolved.

Horizontal competitor analysis

There is no single direct comparable company because no conventional peer reproduces the ownership architecture. A pure potash producer prices Qarhan but misses Julong. A lithium producer prices the salt-lake growth option but ignores potash and copper. A copper miner benchmarks Julong, then has nothing to say about Zangge's minority ownership or salt-lake assets. This is a Scenario A competitive landscape: several useful partial comparators, no true whole-company peer.

Salt Lake Co. (000792.SHE) is the natural Chinese operating comparator for Qarhan. Both derive economics from Qinghai salt-lake brines and sell potash while developing lithium. Salt Lake's relevance comes from scale and direct operation: it is the benchmark for judging Zangge's unusually low potash costs and resource rights. What separates Zangge is that the copper associate has become more valuable to earnings than its own salt-lake revenue, something a pure Qarhan comparison cannot capture. Zangge itself describes its approximately 1.2mt potash capacity as China's second-largest domestic position.

Asia-Potash International (000893.SHE) is a better reference for the growth side of potash. Its investor narrative centres on adding new potash tonnes outside China's mature Qinghai basin. Zangge's proposed 2mt Laos project could eventually move it toward that model, but as of H1 2026 Zangge is still at the feasibility, environmental and mineral-right stage. An investor should value Zangge's existing Qarhan output on proven economics and Laos as an option, rather than use a peer's producing-capacity multiple on unpermitted tonnes.

Ganfeng Lithium (002460.SHE) is useful for the opposite reason. It represents a broad lithium portfolio whose valuation rises and falls with lithium price expectations and project ramp economics. Zangge has far less lithium volume, but its salt-lake unit costs can be lower than many hard-rock routes when the brine chemistry works. At today's earnings mix, however, valuing Zangge as a lithium equity would overstate lithium's importance: H1 lithium revenue was only CNY 0.582bn, while Julong contributed CNY 2.839bn of investment income.

Zijin Mining (601899.SHG) is the most important capital-market comparator because it is both Zangge's controller and Julong's operator. Zijin owns 57.35% of Julong directly according to Zangge's current project page and 26.01% of Zangge directly at June 2026. It controls the mine, sets the operating culture and participates directly in phase-three decisions. Zangge offers a much more concentrated claim on Julong per unit of corporate complexity, but it inserts an extra minority/holding-company layer between the investor and the mine.

That makes Zangge neither a simple substitute for Zijin nor an obviously leveraged version of Zijin. Zijin shareholders own a globally diversified operator with direct mine-control rights. Zangge shareholders own 30.78% of Julong plus the potash/lithium businesses, but their cash from Julong depends on distributions decided inside an entity controlled by the same parent that controls Zangge. Concentration gives Zangge more Julong sensitivity; governance layering makes every CNY of look-through Julong NAV worth less than a CNY of directly controlled NAV.

Western Mining (601168.SHG) is a useful regional copper comparator because its Tibetan/Qinghai mining footprint shows that high-altitude Chinese copper operations can have substantial strategic value without being priced like global diversified majors. The more important contrast is control: Western Mining consolidates its key controlled mining operations, while Zangge's most important mine remains an associate.

Financial comparables therefore have to work segment by segment rather than through one blended P/E. Potash should be compared with mature low-cost potash cash flows; Qarhan/Mami with lithium producers and project NAVs; Julong with quality copper mines; and then a discount applied for Zangge's minority ownership and controller overlap.

Which company became what

Salt Lake became a domestic salt-lake scale operator. Asia-Potash became a capacity-growth potash story. Ganfeng became a diversified lithium-cycle vehicle. Zijin became a global mining operator and capital allocator. Zangge became something stranger: a low-cost salt-lake producer whose valuation is dominated by an associate mine operated by its own controller.

Customers do not choose Zangge because of a differentiated consumer proposition. Potash buyers care about delivered cost and reliable nutrient supply; battery-material buyers look at lithium-carbonate specification and price; copper concentrate goes into commodity and smelting markets. Competitive advantage sits in geology, extraction cost, logistics, recoveries and licences. In the capital market, ownership structure does the differentiating, not the product.

This horizontal view argues for a valuation discount rather than a premium to a controlled copper miner. Julong may deserve a premium asset multiple because of resource scale and growth. Zangge's 30.78% stake deserves less than 30.78% of that undiluted NAV because Zangge cannot unilaterally decide the mine's dividend, phase-three budget or related-party transactions. I use a 15–25% minority/holding-company discount in the optimistic/base cases and a larger effective discount in the conservative case.

Current fundamentals and bull/bear divergence

The latest operating picture is materially stronger than 2025, but the composition matters more than the headline growth rate.

H1 2026 revenue increased 23.05% to CNY 2.065bn. Attributable net profit climbed 102.09% to CNY 3.638bn and ex-items profit 104.92% to CNY 3.706bn. Operating cash flow rose only 17.99% to CNY 0.984bn, reflecting both the equity-method accounting structure and ordinary working-capital timing.

Potash was steady rather than explosive. Production reached 510,300 tonnes, almost exactly half the 1.0mt annual plan, and sales 525,300 tonnes. Pricing was 7.4% higher and cost 2.02% lower, lifting gross margin 3.35 points. Unless potash prices move sharply, this business looks like a reliable CNY 1bn-plus annual gross-profit contributor rather than the source of the next earnings doubling.

Lithium was the strongest consolidated cyclical recovery. H1 output of 5,400 tonnes was close to half the 11,000-tonne annual target. The average selling price rose to CNY 164,900 per tonne and cost fell to CNY 42,200, producing CNY 0.582bn of revenue and a 71.08% gross margin. Q1 alone had produced only 1,765 tonnes, which implies a large Q2 production improvement as the operation normalised.

Julong was the earnings engine. H1 copper output was 134,000 tonnes, sales 133,500; gold output and sales were 423kg; silver output was 80.35 tonnes; molybdenum output was 4,711 tonnes. Net profit was CNY 9.223bn, of which Zangge recognised CNY 2.839bn. Against Julong's stated 2026 copper target around 300,000 tonnes, first-half output leaves the second half needing roughly 166,000 tonnes to hit that level. That is the single most important near-term operating test.

The market is trading the phase-two earnings run-rate first, copper price second, and Mami/lithium optionality third. A low headline forward P/E is not sufficient evidence of cheapness because the earnings denominator already assumes a very profitable Julong ramp.

The bull case begins with the ramp. Phase two is physically complete, has already lifted H1 copper production materially, and was funded within Julong. At 300,000–350,000 tonnes of annual copper, the mine could produce substantially more earnings than the phase-one asset did, while phase three remains a further option rather than a requirement for the existing investment case.

The second bull point is cash passthrough. Minority stakes deserve discounts when profits are permanently trapped. Zangge has actually received large Julong dividends, including CNY 1.539bn in March and another CNY 0.616bn in August. The discount should therefore be meaningful but not punitive enough to assume zero distributions.

The third is cost position. Potash cost below CNY 1,000 per tonne and lithium cost around CNY 42,200 give Zangge resilience against commodity corrections. Mami's trial production could add another low-cost lithium stream without requiring Zangge to consolidate the entire project capex.

The bear case begins with control. Zijin is the majority owner/operator of Julong and controller of Zangge. The legal related-party protections are real, but they cannot make Zangge the decision-maker over Julong. A future phase-three programme could retain years of Julong cash rather than distribute it. Zangge would still book its share of profit while cash passthrough deteriorated.

The second bear point is expectations. At CNY 76.15, the stock is more than twice the control-transfer reference price and roughly 21 times trailing attributable earnings. Phase two is no longer hidden value. The market already knows the 300,000–350,000-tonne target, the Mami trial-production milestone and the lithium-price recovery. Incremental rerating increasingly requires better-than-base outcomes.

The third is commodity asymmetry. H1 Julong net margin was over 60%, an extraordinary profitability level. Such margins create large downside torque if copper falls. At the same time, a lithium recovery from CNY 67,000 to CNY 165,000 per tonne has already occurred in Zangge's realised price. The investor is no longer buying either commodity near the depressed point reflected in 2024–2025 financials.

Valuation analysis

A blended P/E is the wrong primary method. I value potash, Qarhan lithium, Mami and Julong separately, then add net financial assets/options and apply a minority/holding-company discount where appropriate.

Historical and headline valuation

At the September 4 close, trailing P/E is about 21.0 times and annualised-H1 P/E about 16.4 times. P/B is about 6.9 times June 2026 attributable equity. The lower “forward” P/E comes almost entirely from the Julong phase-two earnings jump, which is exactly why it should not be treated as equivalent to a 16-times controlled operating business.

The CNY 35 transfer price is a useful but imperfect historical anchor. Against 2024 EPS of CNY 1.64, it represented roughly 21 times trailing earnings, and it was a control transaction, not a minority-market bargain. Current trailing P/E is coincidentally close to that level because both the price and earnings have approximately doubled. The business quality improved; the market capitalised much of that improvement at the same time.

The historical multiple has shifted for a legitimate reason: Zangge now has a stronger controller and a phase-two Julong earnings stream. The question is how much additional phase-three and lithium success is already embedded.

Cash-flow passthrough before valuation

2025 operating cash flow of approximately CNY 2.10bn was 54.5% of CNY 3.852bn net income. H1 2026 operating cash flow was CNY 0.984bn versus CNY 3.638bn net income, or 27.0%. Those ratios look weak, but they exclude the economic importance of cash dividends received from the equity-accounted Julong stake.

For valuation, I therefore define owner earnings as:

consolidated after-tax operating cash generation + sustainable Julong cash distributions − maintenance capex.

The filings do not separately label maintenance and growth capex, so I use CNY 0.25bn annual maintenance capex for the mature consolidated salt-lake assets as an explicit analytical assumption. Growth investment in Mami, Laos or additional projects is valued separately rather than deducted twice.

On a normalised basis, I estimate sustainable owner earnings around CNY 5.0–6.0bn under the base commodity deck. At CNY 119.47bn of equity value, this implies an owner-earnings yield of roughly 4.2–5.0%, or an effective P/owner-earnings multiple around 20–24 times. That is materially less attractive than the 16.4-times annualised headline P/E. Because the gap exceeds 30% at the upper end, the SOTP below is anchored to distributable cash and asset value rather than accounting EPS.

Segment sensitivity

The base case closely reflects current realised economics; the conservative case still leaves a profitable potash operation because Zangge sits low on the domestic cost curve. H1 operating data underpin the volume and cost assumptions.

The lithium business has far more percentage earnings elasticity than potash but a much smaller tonnage base. H1 realised price was CNY 164,900 and reported cost CNY 42,200.

These are my scenario estimates, not company guidance. The anchor is Julong's actual H1 2026 CNY 9.223bn net profit on 134,000 tonnes of copper; actual future profit will also depend on grade, recoveries, gold/silver/molybdenum by-products and costs.

Sum-of-the-parts

I value the Julong stake on normalised earnings/NAV, not Zangge's reported P/E. I then apply a 25% discount in the conservative case, about 20% in the base case and 15% in the optimistic case for minority status, cash-distribution uncertainty and the controller overlap. The discount narrows as proven dividends and phase-two execution reduce uncertainty; it never disappears because Zangge still lacks operating control.

I value potash as a mature low-cost commodity cash generator. Qarhan lithium gets a lower-cycle multiple and commodity-price sensitivity. Mami gets project NAV only for Zangge's economic share, discounted for trial-production/ramp risk. Phase three I treat as an option: almost zero value in the conservative case, modest probability-weighted value in the base case and material value only in the optimistic case.

† The conservative residual includes net financial assets and a modest value for existing non-Julóng interests; no material phase-three value is assumed. § The optimistic residual includes substantially more phase-three and development-option value; ranges are rounded, so the component bounds need not add mechanically to the per-share endpoints.

The unusual-looking residual is there because this is a probabilistic NAV, not a liquidation balance sheet. Julong's phase-three option, the difference between accounting and economic cash balances, Mami's funding structure and the Laos/other mineral rights cannot be valued with the same confidence as producing potash.

Scenario analysis

This is valuation-scenario analysis within a research framework, not investment advice.

Expectation gap

The market appears to be pricing a successful phase-two ramp and assigning non-zero value to phase three and lithium growth. Current price sits almost exactly in my base-value region rather than near conservative value. So the next expectation gap is unlikely to come from “phase two exists”; investors already know that. It will come from the slope of the ramp and the quality of cash conversion.

The most important next data are Julong H2 copper output, realised profitability per tonne and cash dividends. A full-year result around 300,000 tonnes with strong distributions supports the base case. Output materially below that level, or strong accounting profit accompanied by sharply weaker dividends, would undermine it.

For lithium, investors should ignore Mami capacity headlines and watch actual saleable output. A 50,000-tonne plant producing 20,000–25,000 tonnes in its first year is normal ramp behaviour; a project still producing well below 20,000 tonnes into 2027 would indicate that commissioning, brine balance or product-quality issues are more persistent.

Margin-of-safety recheck

Current price is above the CNY 58–62 conservative intrinsic-value range. On that discipline, the margin of safety is zero.

The base case's most fragile assumption is the combination of sustained Julong profitability and a manageable minority discount, not potash or even lithium. If the value I assign to Julong in the base case is cut to 70%, base SOTP falls by roughly CNY 19–21bn, or about CNY 12–13 per share. Base value would fall from CNY 72–82 toward roughly CNY 59–69.

If earnings are flat for three years and the valuation multiple does not expand, shareholder return depends principally on dividends. A sustainable payout around CNY 2 per share gives only about a 2.6% current cash yield; CNY 3 gives about 3.9%. That can exceed a low sovereign bond yield, but it is far too narrow an excess return for commodity, minority-stake and governance risk. Flat earnings do not create a persuasive margin of safety at CNY 76.15.

This is not a “bad company at any price” case. It is a set of good assets whose current price already recognises most of the base-case improvement.

Margin-of-safety sufficiency verdict: none.

Risk analysis

The first permanent-loss risk is a Julong profit reset. I assign medium probability and high impact. H1 Julong net margin above 60% and CNY 9.223bn of profit were exceptional. A combination of copper returning toward CNY 70,000 per tonne, lower by-product credits and phase-two grades below early ramp expectations could cut Julong profit toward CNY 14bn or lower on a full-year basis. Zangge's equity income would fall, and the market could simultaneously remove the growth multiple attached to phase three. The observable indicators are Julong quarterly copper output, realised earnings per tonne and Zangge's investment income.

The second is cash-control risk at Julong. Probability is medium and impact high. Phase two was self-funded, which was good for avoiding a Zangge capital call, but the same structure proves that Julong can retain its own cash for growth. If phase three proceeds with a very large budget, Zijin may rationally favour reinvestment over dividends. Zangge could continue reporting equity-method earnings while cash receipts fall. The indicator is Julong cash dividends relative to Zangge's recognised Julong income over rolling 12–24 months. The risk becomes serious if distributions remain below roughly half of recognised profit for two consecutive years without a clearly accretive, funded project explanation.

The third is controller/minority conflict. Probability is medium, impact medium-to-high. Formal commitments are strong, but Zijin occupies both sides of the central economic relationship: it controls Zangge and directly controls/operates Julong. It also has its own lithium assets. An asset transfer, service agreement, capital increase or business delineation can be economically fair or unfair depending on price and terms. The observable indicators are related-party transaction volumes, independent-director opinions, valuations in any asset injection, Julong capital actions and whether the 60-month competition undertaking produces transactions on transparent terms.

The 18-month voting waiver is a near-term governance checkpoint. It runs from the 2025 closing and is therefore approaching expiry in late 2026. The former controller promised that after expiry it would determine whatever continuing waiver is needed to keep its combined voting influence below Zijin's. The market should verify the actual post-expiry arrangement rather than assuming today's voting structure continues unchanged.

The fourth is lithium execution/permitting. Probability is medium and impact medium. Qarhan's 2025 temporary suspension showed that mineral rights can interrupt a profitable operation. Mami is only at trial production and is located in a high-altitude region where construction seasons, energy, brine engineering and environmental approvals matter. The alert is Mami output below 20,000 tonnes in 2026 or failure to approach the 50,000-tonne design rate on a credible timetable during 2027.

The fifth is valuation compression. Probability is medium and impact high because the balance sheet itself is unlikely to cause insolvency. At roughly 21 times trailing profit and almost 7 times book value, a commodity downturn can reduce both earnings and the multiple. A move from CNY 5–6bn of owner earnings toward CNY 4bn, accompanied by a 12–14-times owner-earnings multiple, produces equity values around half the current market capitalisation.

Legacy governance is now a lower-probability direct risk but remains relevant to valuation. The 2019 CSRC case documented false trading, inflated accounts and CNY 2.214bn of controlling-shareholder fund occupation. Control has changed and the new controller's formal undertakings explicitly prohibit these behaviours. A recurrence under the new regime would therefore be much more damaging than an ordinary compliance mistake because it would destroy the central premise of the 2025 governance rerating.

Catalysts and tracking indicators

The most important positive catalyst over the next twelve months is evidence that Julong phase two can run at the intended level with good margins. Full-year copper close to or above 300,000 tonnes, followed by a visible path toward 300,000–350,000 tonnes of steady-state annual production, would reduce the execution discount. Continued large cash distributions would reduce the holding-company discount at the same time.

Mami's catalyst is actual output, not another capacity announcement. Production approaching the 20,000–25,000-tonne 2026 target with battery-grade product and evidence of a 50,000-tonne run-rate would justify moving the project from option value toward operating NAV.

Potash offers a steadier catalyst: realised pricing above CNY 3,000 per tonne while unit cost stays close to CNY 1,000. Because the installed volume is mature, price-cost spread rather than new tonnes is what moves earnings.

Phase-three approval would be a powerful stock catalyst but ambiguous for cash flow. A technically attractive 600,000-tonne long-term plan creates NAV; a capital-intensive project that absorbs dividends for years can reduce near-term owner earnings. The market should not automatically treat a larger capex announcement as positive.

Negative catalysts are the mirror image: Julong H2 output below the level required to reach the 300,000-tonne annual plan; a copper-price correction; Mami production substantially below guidance; a return of Qarhan permitting constraints; related-party transactions with weak minority safeguards; or a post-voting-waiver governance arrangement that increases uncertainty.

Tracking dashboard

† The October 16 date is a market-data estimate rather than a company-announced statutory date; it should be rechecked when Zangge publishes the formal Q3 reporting schedule.

I built the dashboard around variables that can actually falsify the thesis. Copper output and dividends show whether Julong's accounting earnings are becoming cash. The potash price-cost spread shows whether the consolidated floor remains intact. Mami output tells us whether lithium “capacity” has become production, and related-party disclosures and voting arrangements tell us whether the governance rerating remains justified.

Cross-synthesis summary

Looking vertically, Zangge has proven one capability beyond doubt: it has accumulated economically valuable mineral interests. Qarhan potash survived the company's governance troubles because the underlying resource and cost position were real. The Julong stake became extraordinarily valuable because the orebody was real and Zijin turned it into an operating mine. Salt-lake lithium has shown credible low-cost economics, though not yet large scale. The corporation has been better at owning resources than at proving an uninterrupted history of first-class governance.

That distinction explains the company's unusual journey. Its biggest historical success factors were geology, commodity cycles and capital ownership. Management execution mattered at Qarhan, but the old controlling regime also produced one of the clearest reasons not to equate high margins with high business quality: the CSRC found fictitious trades, inflated profit and CNY 2.214bn of controlling-shareholder fund occupation.

The 2025 control transfer changed more than the name on the shareholder register. It imported a mining operator whose own economic interest in Julong is larger than Zangge's and whose construction record includes bringing a CNY 17.46bn, 200,000-tonne-per-day expansion into production at extreme altitude. The market correctly assigned value to that change.

Yet the same transaction made Zangge's governance paradox sharper. The party capable of maximising Julong's operating NAV is the party that controls both Julong and Zangge. That alignment is excellent when the question is whether the mine will run well. It is less obviously aligned when the question is whether Julong should pay dividends, retain CNY 20bn for phase three, buy services from related entities or restructure assets between Zijin vehicles. Legal undertakings protect process. They do not eliminate economic conflicts.

Horizontally, Zangge's own operating advantage over a generic resource producer is its salt-lake cost base. Potash cost below CNY 1,000 per tonne against realised pricing above CNY 3,000 is hard to replicate without equivalent brine geology and rights. Lithium cost around CNY 42,200 provides similar asymmetry. But neither business currently determines most of the share price.

Julong does. Compare Zangge with Zijin before comparing it with a lithium producer. Zangge gives the shareholder a more concentrated economic claim on Julong plus two salt-lake assets. Zijin gives the shareholder control of Julong and a much broader global mining portfolio. Zangge is higher-beta to Julong's economics but lower-quality in governance terms at the asset interface. It is a concentrated minority proxy, not a superior substitute for owning the operator.

The current price also tells us that the market understands more than the trailing financial statements suggest. CNY 76.15 is 118% above the CNY 35 control-transfer reference price. H1 attributable profit has doubled, Julong phase two is producing, lithium realised prices have more than doubled from their 2025 nine-month average, and Mami has entered trial production. The stock no longer asks an investor to believe that these things might happen; it asks the investor to believe that their economics remain strong enough to support a CNY 119bn valuation.

The market's most likely misjudgment is to treat phase-two accounting earnings and phase-three NAV as though both can be capitalised in full while also assuming a generous near-term dividend stream. Those claims compete for the same Julong cash. If phase three is funded heavily from retained earnings, its NAV may rise while Zangge's cash yield falls. A proper SOTP needs to avoid counting both at 100%.

The opposite market mistake is possible too. A simple holding-company discount can become too punitive when the underlying associate pays real cash. Julong has already distributed substantial amounts to Zangge. If it establishes a recurring distribution record even while funding growth, a 30–40% minority discount would be excessive. That is why my base discount is around 20%, rather than treating the stake as trapped.

Over the next twelve months, copper output is the key variable. Julong produced 134,000 tonnes in H1. Reaching 300,000 tonnes for the year requires roughly 166,000 tonnes in H2. If that happens with margins close to current levels, Zangge can report another very strong earnings period. If the mine instead stalls materially below the target, the phase-two rerating will have run ahead of physical delivery.

The second one-year variable is cash. Another year in which Julong dividends broadly track a meaningful fraction of Zangge's investment income would make the earnings much more valuable. A sharp divergence, especially if phase-three spending begins, would justify expanding the holding-company discount.

The third variable is Mami output. The project does not need to hit 50,000 tonnes immediately. It does need to show that a 20,000–25,000-tonne 2026 target can become a reliable path toward nameplate capacity. Until then, the correct valuation is probability-weighted project NAV rather than capacity times lithium price.

At three years, phase three becomes the key strategic question. A permitted, engineered, economically funded Julong producing toward 600,000 tonnes could justify a much larger NAV even with a minority discount. But the investor has to care about funding. Zangge's 30.78% ownership should not be diluted, and the mine's growth spending should not consume essentially all distributions indefinitely, unless the incremental return comfortably exceeds the value of cash returned to shareholders.

At five years, the company itself may look structurally different. Zijin's lithium-overlap remedy is due within the 60-month commitment period. Mami should either have proved itself or disappointed. Laos potash may have obtained rights and moved into construction, or remained an option. Zangge could emerge as Zijin's listed Chinese salt-lake platform around a large Julong stake. That is plausible, but it is not yet a contractual endpoint.

The conditions that would make Zangge a materially better investment are clear. The price can fall while the operating thesis stays intact; Julong can prove a stable 300,000–350,000-tonne run-rate; cash dividends can remain substantial; Mami can ramp toward 50,000 tonnes; and related-party transactions can remain demonstrably arm's length. A stock price in the high CNY 40s with those facts intact would create a very different expected-return distribution from today's CNY 76.15.

The thesis should be overturned in the other direction if Julong cannot ramp despite phase-two completion, if distributions remain structurally low because the controller prioritises other uses of Julong cash, if phase three requires dilution or unattractive capital calls, or if the new governance regime shows anything resembling the old controlling shareholder's approach to related-party value transfer. The latter would deserve an immediate and severe governance discount because the entire 2025 rerating rests partly on institutional improvement.

Bull and bear reasons

Core bull reasons:

• Julong phase two has already raised H1 2026 copper output to 134,000 tonnes and Zangge's Julong investment income to CNY 2.839bn, up 124.57%.
• The CNY 17.46bn phase-two expansion was funded inside Julong, avoiding a disclosed pro-rata Zangge capital call while more than doubling processing capacity.
• Qarhan potash and lithium have H1 unit costs of CNY 975.75/t and CNY 42,200/t, giving the consolidated businesses substantial commodity-downside resilience.
• Julong has paid meaningful cash dividends, showing that equity-method earnings are not automatically trapped.
• Mami entered trial production in H1 2026, creating a tangible path from project NAV to lithium cash flow.

Core bear reasons:

• 78.03% of H1 profit came from a 30.78% associate that Zangge does not operate.
• Zijin simultaneously controls Zangge and the majority of Julong, creating unavoidable economic conflicts around dividends, phase-three capex and related-party dealings despite formal safeguards.
• At CNY 76.15 the stock is already more than twice the CNY 35 control-transfer reference price and approximately 21 times trailing earnings, so phase-two success is substantially priced.
• Lithium has already moved from a 2025 nine-month average realised price of CNY 67,306/t to CNY 164,900/t in H1 2026, reducing the probability that today's buyer is purchasing at a lithium trough.
• Mami's 20,000–25,000-tonne 2026 project target is well below 50,000-tonne design capacity, so nameplate capacity remains ahead of proven output.

Pre-mortem: where this research could be wrong

A three-year 50% loss could occur without any insolvency. Imagine copper normalises to around CNY 65,000–70,000 per tonne during 2027 while Julong's phase-two ore mix delivers lower grades than the strong early-2026 period. Julong net profit falls from the H1 2026 annualised run-rate above CNY 18bn to CNY 10–12bn. Zangge's share falls toward CNY 3.1–3.7bn. Lithium carbonate retreats toward CNY 100,000, so Qarhan's contribution compresses and Mami's ramp no longer receives a growth multiple. The market then values normalised owner earnings at 12–14 times rather than more than 20 times. An equity value around CNY 35–40 per share becomes plausible, roughly half today's price.

A second script does not require weak copper. Julong phase three receives approval in 2027 with a very large capital programme. Zijin chooses, rationally from the mine's perspective, to retain most Julong cash for several years. Zangge continues booking CNY 5–7bn of annual associate profit but receives far less cash. Investors stop treating accounting investment income as equivalent to owner earnings; the minority discount widens from about 20% to 35–40%. At the same time Mami remains around 20,000–25,000 tonnes rather than approaching 50,000. Even with copper strong, a holdco rerating can take the stock into the CNY 40s.

Final research conclusion

Zangge's assets are better than its headline corporate form suggests. Qarhan contains a genuine low-cost potash franchise and a low-cost lithium operation. Julong is a rare Chinese copper asset with enormous resources and a phase-two ramp that has already transformed Zangge's earnings. The balance sheet is exceptionally light. Zijin's control improves operating credibility and makes Julong's next expansions more plausible.

The current minority shareholder nevertheless pays CNY 119.47bn for a structure in which most profit comes from an associate controlled by the same parent that controls the listed company. At CNY 76.15, my SOTP places the stock inside the base fair-value region and well above conservative value. I see no margin of safety sufficient to justify initiating a position merely because the annualised headline P/E is in the mid-teens. The better reason to own Zangge would be a price that gives substantial credit for today's potash and phase-two economics while asking little for phase three and Mami.

I would change that conclusion without requiring a lower price if Julong establishes a demonstrably sustainable 350,000-tonne-plus production run-rate, maintains strong distributions while funding growth, Mami proves a fast path to 50,000 tonnes and Zijin's competition-remedy transactions demonstrably increase Zangge per-share value. At today's evidence set, those outcomes remain partly future value already being paid for.

【Company-profile scores】

• Fundamental quality: medium
• Growth: high
• Moat: strong
• Financial soundness: strong
• Management credibility: medium
• Valuation attractiveness: medium
• Risk level: high
• Suitable investor type: cyclical

【Investment rating】

• Rating: Hold
• One-line thesis: Julong phase two is transforming earnings, but CNY 76.15 already capitalises much of that growth without a conservative margin of safety.

【Ideal Buy Price】46–49 CNY

Basis: at least 20% below the CNY 58–62 conservative SOTP range, allowing for copper-cycle, associate-cash-flow and governance uncertainty.

• Acceptable hold price: CNY 66–89, approximately the ±15% region around the base SOTP centre.
• Clearly overvalued price: CNY 119 and above, at least 10% above the optimistic CNY 98–108 intrinsic-value range.
• Current-price classification: acceptable hold.
• Whether to wait for a better price: yes. For a new position I would wait for CNY 49 or below with Julong's phase-two ramp intact, no deterioration in Julong cash distributions and no adverse governance change. The opportunity cost is missing further upside if phase three is approved or copper remains unusually strong.
• Target holding horizon: 3–5 years.
• Expected annualized return: conservative about -6% to -2%; base about 3–7%; optimistic about 12–17%, including assumed cash dividends and a three-year convergence toward scenario value.
• Max-loss risk: roughly 45–55% under the pre-mortem case in which Julong profit falls toward CNY 10–12bn, lithium returns near CNY 100,000/t and the owner-earnings multiple compresses to 12–14 times.
• Reassessment-trigger signals: Julong 2026 copper below 280,000 tonnes; Julong distributions below 50% of Zangge-recognised associate profit for two consecutive years without a clearly accretive funding rationale; Mami failing to reach 20,000 tonnes project output in 2026; potash unit cost above CNY 1,200/t for two reporting periods; or a related-party transaction that materially transfers Julong/lithium economics without independent valuation and minority safeguards.

【Valuation Range】

• current: 76.15 (close as of 2026-09-04)
• bear (conservative · ideal buy zone): [46, 49]
• base (fair · acceptable hold zone): [66, 89]
• bull (optimistic · above the clearly-overvalued line): [119, 130]

Key data tables

Current operating snapshot

The table shows why revenue multiples are nearly useless. Julong's CNY 15.004bn of associate revenue is outside the consolidated top line, while Zangge's CNY 2.839bn share of profit enters earnings. A conventional “revenue growth versus P/S” screen therefore compares incompatible perimeters.

Operating assets and ownership

The ownership table corrects two easy valuation errors: counting Mami's entire project capacity as Zangge production and treating Julong as a consolidated mine. Neither is appropriate.

Control-transfer reference points

The increase from the original 392.25m block to 408.15m direct shares is not a unit discrepancy. The revised ownership filing discloses additional Zijin purchases of 6.84m shares in February 2025 and 9.06m in April at CNY 30.09–34.00 per share.

Research uncertainties

The largest blind spot is the precise current term, expiry and water-allocation conditions across the Qarhan mining/brine rights. The reported 724.35-square-kilometre mining-right area is verifiable, and the 2025 lithium-permit disruption is verifiable, but I could not establish one definitive current expiry schedule from the primary documents available in this session. This matters enough that the conservative SOTP explicitly discounts tenure risk rather than assuming perpetual renewal.

The second is phase-three financing. Zijin has disclosed the 200mt-per-year throughput concept and roughly 600,000-tonne copper ambition, but there is no final approved capex, funding structure or binding Zangge capital obligation. Phase two was self-funded by Julong; extrapolating that financing structure to phase three would be an unsupported assumption.

The third is Mami steady-state economics. Trial-production approval and the 20,000–25,000-tonne 2026 target are clear, but a long operating history of recovery, brine balance, product quality and steady-state unit cost does not yet exist. The project deserves option value, not a full mature-plant multiple.

The fourth is the maintenance-versus-growth capex split. Zangge does not provide an investor-ready decomposition suitable for direct owner-earnings calculation. My CNY 0.25bn mature-business maintenance assumption is therefore a research assumption, not a disclosed figure.

The fifth is the requested internal Zijin report 601899-2026-07-22. It was not available through the connected research sources in this session. I therefore cross-checked Julong operating scale, phase-two construction and ownership against Zijin/Zangge primary public disclosures rather than inheriting the internal report's framing or valuation.

Sources

Primary source hierarchy for this report begins with Zangge's 2026 half-year report summary, which provides the financial statements, shareholder structure, potash/lithium operating data, Mami status and Julong operating contribution.

The 2026 first-quarter report provides the phase-two start date, Q1 Julong output/investment income, the March cash dividend, potash/lithium Q1 prices and costs, and Mami construction status.

The audited 2025 annual-report summary provides CNY 3.577bn revenue, CNY 3.852bn attributable profit, CNY 16.244bn year-end equity, quarterly cash-flow data and 2023–2025 comparisons.

The 2025 third-quarter report provides the pre-phase-two baseline: 142,500 tonnes of Julong copper output, CNY 1.950bn of Zangge investment income, and the low 2025 lithium-price/margin data.

Zijin's January 2025 control-acquisition announcement provides the CNY 35/share reference price, CNY 13.729bn consideration, governance design and the detailed undertakings on related-party dealings, lithium competition, priority for new potash/lithium opportunities and corporate independence.

The revised detailed equity-change report provides the post-purchase holdings, 79.02m-share voting waiver, no-control undertakings and Zijin's additional February/April market purchases.

The CSRC Qinghai Bureau's 2019 market-ban decision is the primary source for the historical false-trading, inflated-account and controlling-shareholder fund-occupation findings.

Zangge's current Julong project page is the primary current source used for Julong's 57.35% Zijin / 30.78% Zangge ownership split and the 25.88mt copper-resource figure.

Shanghai Futures Exchange and Guangzhou Futures Exchange data are used only to anchor the 2026-09-04 commodity environment; the valuation price decks remain my normalised assumptions.

The September 4 share-price reference is Yahoo Finance's CNY 76.15 close, cross-checked against the company's issued-share count to obtain CNY 119.47bn market capitalisation.

Other tickers mentioned

• 601899.SHG: Zijin Mining, Zangge's controlling shareholder and the majority owner/operator of Julong Copper.
• 000792.SHE: Salt Lake Co., the closest domestic Qarhan potash and salt-lake lithium operating comparator.
• 000893.SHE: Asia-Potash International, a potash capacity-growth reference relevant to Zangge's prospective Laos project.
• 002460.SHE: Ganfeng Lithium, a lithium-cycle comparator for valuing Zangge's much smaller salt-lake lithium exposure.
• 601168.SHG: Western Mining, a regional high-altitude copper-mining reference for Tibetan/Qinghai asset economics.

This report is based on public information and does not constitute investment advice. Markets carry risk; invest with caution.