【SMM Analysis】 2026 H1 Consumer & Cobalt Market Review and Outlook — Can We Still Look Forward to the Future?

Key Takeaways

• Recycling supply outpaces expectations. China's recycled cobalt supply is projected to reach 45,000–50,000 tonnes of contained cobalt in 2026, meeting approximately 35% of domestic refined cobalt consumption and emerging as the largest marginal increment on the supply side in H1.

• LCO is being rapidly displaced by NCM cathode materials. Some smartphone batteries have adopted an LCO-NCM blend (8:2 ratio), and the advancement of silicon-carbon anodes is closing the energy density gap — leaving mid-to-high-end consumer electronics as potentially the last remaining stronghold for lithium cobalt oxide (LCO).

• Severe inventory overhang in the consumer supply chain. Inventories across the consumer value chain (from cobalt salts to battery cells) exceed the normal restocking cycle by roughly 11,000 tonnes of contained cobalt. This "dammed lake" effect is the primary factor suppressing prices.

• End-market demand squeezed by dual cost pressures. Chip price hikes are raising finished-device costs, while cobalt and lithium price increases are elevating battery costs.

• No inflection point expected in H2 2026; 2027 hinges on foreign miners' shipment strategies. The oversupply regime will persist in H2 2026, keeping prices under pressure. In 2027, if foreign mining companies proactively curtail shipment volumes, a price turnaround window may emerge between late 2026 and early 2027.

Part I: H1 2026 Retrospective

Part II: H1 2026 Segment-by-Segment Review

Part III: H2 2026 Segment-by-Segment Outlook

Part IV: Supply-Demand and Price Forecast

I. H1 2026 Retrospective

Entering late 2025, market sentiment was uniformly bullish — cobalt prices and consumer electronics demand were both at cyclical highs. In Q1 2026, as DRC intermediate product exports remained persistently sluggish, confidence in further price appreciation ran high. However, by March and April, inventory across the value chain failed to decline as expected; instead, stocks remained stubbornly high despite destocking efforts, and prices began to show early signs of softening.

Looking back today, two structural shifts have reshaped the market: recycling has expanded supply, while substitution has eroded demand — and both ultimately manifest in inventory accumulation. Several key developments were not fully anticipated at the time:

1. The Surge in Recycled Cobalt Supply

In 2025, China's recycled cobalt supply stood at just over 20,000 tonnes of contained cobalt. In H1 2026 alone, recycled cobalt output reached 21,000 tonnes, with full-year projections now at 45,000–50,000 tonnes — enough to satisfy 35% of China's total cobalt demand. This is an extraordinarily large increment. Although recycled cobalt salt production had already reached 3,200 tonnes of contained cobalt by December 2025, the market did not pay sufficient attention at the time. In hindsight, that was only the beginning.

Two factors underpin this faster-than-expected recycling ramp-up:

• Policy drivers: In October 2025, China liberalized the import and export of black mass (battery powder). Early on, market participants were still navigating the new regulatory framework, and no one could confidently forecast how much material would enter the system or how much recycled output it would generate.
• Internal waste recycling loops: Approximately 70% of current black mass originates from production scrap generated during cell manufacturing. A significant portion of this material does not circulate in the open market; instead, it is consumed internally through direct partnerships between cell manufacturers and upstream recyclers.

The combination of these two factors has driven recycled cobalt supply far beyond initial expectations.

2. LCO-to-NCM Substitution

In 2025, LCO demand totaled 124,000 tonnes. Even under relatively optimistic assumptions, we now forecast only 96,000 tonnes for 2026 — representing a decline of 17,000 tonnes of contained cobalt demand. This contraction reflects both sluggish end-market consumer electronics demand and the accelerating substitution by NCM cathode materials.

Previously, I estimated that for a 20Wh battery, LCO costs roughly RMB 5 more than NCM — a seemingly trivial difference. Based on that figure, I believed 3C electronics would remain the core demand base for LCO, and that such a small cost gap would not meaningfully displace LCO in consumer applications. At the time, I only downgraded LCO demand projections for PCs and tablets.

However, the pressure to reduce battery costs has pushed all manufacturers to seriously evaluate NCM alternatives. In H1 2026, several smartphone OEMs launched pure NCM battery programs, and a growing number of devices adopted LCO-NCM blended chemistries (80% LCO / 20% NCM) to cut costs.

One might ask: what about energy density? NCM intrinsically offers lower energy density than LCO — but silicon-carbon anode technology is advancing rapidly. Mature commercial solutions now achieve silicon content above 30%, delivering up to 40% higher energy density, while adding less than RMB 2 to the anode cost per cell. Silicon-carbon anode progress has effectively closed the energy density gap that once favored LCO.

Let us run the numbers. LCO requires 1,400–1,800 tonnes per GWh, depending on operating voltage. High-voltage NCM 6-series requires approximately 1,700 tonnes per GWh — so the mass per GWh is comparable. The difference in cobalt content, however, is dramatic: LCO contains 60% cobalt, while NCM 6-series contains only 6.9%. This means every unit of LCO displaced by NCM reduces cobalt demand by a multiple. When cost reduction and performance preservation can be simultaneously achieved, mid-to-high-end consumer products may be all that remains of LCO's core market.

3. Inventory Dynamics

The inventory issue must be examined on two levels: the pace of aggregate destocking, and the structural distribution of inventories across segments.

On the aggregate level, new supply additions coincide with contracting demand, continuously slowing the pace of destocking across the value chain. The market had expected effective inventory digestion by Q1, but in practice, stocks were never meaningfully reduced.

What has exerted sustained downward pressure on prices, however, is the structural dimension. While inventory levels in other end-markets remain broadly within reasonable ranges, consumer-segment inventories are clearly elevated. Across the entire consumer value chain — from cobalt salts to cells — normal stocking-cycle inventory would be approximately 14,000 tonnes of contained cobalt; current estimates place it closer to 25,000 tonnes.

In Q1, sentiment remained relatively optimistic, and concerns over DRC intermediate product export disruptions created a "supply shortage" narrative that partially offset the high-inventory overhang. Prices held steady and market liquidity remained adequate. Entering Q2, however, the reality of weak end-market demand became increasingly clear, and forward expectations were steadily revised downward. At that point, high inventory ceased to function as a buffer and became an inventory overhang (the "dammed lake" effect) — downstream buyers slowed procurement, caution prevailed, market liquidity contracted rapidly, and prices weakened noticeably.

Inventory itself is not the problem. The danger lies in high inventory coinciding with a shift in sentiment from optimism to pessimism — at which point every tonne of stock becomes additional weight on prices.

II. H1 2026 Segment-by-Segment Review

1. Consumer Market: Price Hikes Suppress Demand; Production Contracts in Tandem

1) Smartphones

From a production standpoint, the smartphone segment has performed relatively well within consumer electronics this year — even better than many had anticipated. In H1 2026, China's cumulative mobile phone output (including smartphones and feature phones) reached 689 million units, a decline of approximately 1.4% from 699 million units in H1 2025, marking the entry of production into contraction territory.

Rising chip prices have steadily pushed up finished-device costs, forcing OEMs to raise end prices and further dampening consumer replacement willingness. The Q2 trajectory tells the story: Q1 managed a modest 1.7% increase on a low base, but Q2 dropped to -4.5%, with the decline accelerating sharply within a single quarter. Weak end-market sell-through has transmitted upstream to production scheduling faster than expected.

Structurally, smartphones continued to grow, reaching 578 million units in H1, up approximately 4.0% year-on-year. Feature phones, by contrast, contracted sharply to 110 million units, a decline of over 22%. Smartphone share rose from 79.6% in H1 2025 to 83.9%, reflecting ongoing industry consolidation toward smartphones. However, this growth is driven primarily by the continued conversion of feature-phone users and proactive product-mix adjustments by OEMs — not by expansion of overall demand.

In summary, H1 data confirms the smartphone industry has entered a downward trajectory, with the Q2 acceleration a notable warning signal. Under the dual pressure of rising costs and weak end demand, H2 output will likely continue to contract year-on-year.

2) Microcomputers (PCs & Tablets)

If the smartphone market can be described as "mildly impacted," the computer market tells a different story entirely.

In H1 2026, China's cumulative microcomputer (PC and tablet) output reached 150 million units, a decline of approximately 9.4% from 166 million units in H1 2025 — significantly steeper than the 1.4% drop in mobile phones. PCs and tablets already have long replacement cycles; chip-driven cost increases have further reduced upgrade incentives, leaving output with nowhere to go but down.

Moreover, the contraction has been remarkably uniform: both Q1 and Q2 registered -9.4% declines. This consistent, sustained contraction points to a structural step-down in demand rather than seasonal fluctuation. What is more concerning is that the sector was still growing in H1 2025 (+5.8%) before reversing so sharply this year — suggesting the industry is not merely "adjusting" but retreating. Against the broader backdrop of soft 3C consumption, H2 output is expected to remain in negative territory, with the full-year contraction widening further.

2. Lithium Cobalt Oxide (LCO) Cathode Materials

Upstream materials have held up somewhat better than end products. In H1 2026, domestic LCO cumulative output reached approximately 49,200 tonnes, down about 2.3% from 50,300 tonnes in H1 2025 — a decline notably smaller than that of smartphones and computers.

This is not necessarily good news.

Quarterly breakdown: Q1 output totaled approximately 25,900 tonnes, down a marginal 1.7% year-on-year and essentially flat with the prior year. Q2 output fell to approximately 23,200 tonnes, with the year-on-year decline widening to roughly 10.4%. Entering Q2, the pressure of weak end demand began to transmit clearly upstream, and manufacturers adopted more cautious production scheduling.

Industry operating rates tell the same story: the monthly average operating rate in H1 2026 was approximately 52%, down from about 56% in H1 2025. Q2 monthly operating rates hovered between 48% and 52%, compared with 57%–68% in the same period of 2025 — a gap of nearly 15 percentage points in operating rates, representing genuine production cuts.

The competitive landscape is also shifting. The CR3 concentration ratio for LCO cathode materials rose from 73.2% in H1 2025 to 77.0% in H1 2026, with the leading producer's share expanding from 51.2% to 57.8%. Amid contracting downstream demand, leading players — leveraging technological advantages and deep industrial partnerships — are squeezing smaller manufacturers, reinforcing industry consolidation.

Overall, the LCO market in H1 was characterized by both supply and demand remaining subdued. Output declines reflect proactive supply-side contraction in response to soft downstream demand, while rising concentration indicates that industry leaders are accelerating market share gains during this adjustment cycle.

3. Cobalt Tetroxide (Co₃O₄)

If LCO's performance was merely "not great," cobalt tetroxide (tetra-cobalt) has been the worst-performing segment.

In H1 2026, domestic cobalt tetroxide cumulative output reached approximately 42,900 tonnes, down 21.3% from 54,500 tonnes in H1 2025. Why such a steep drop? The problem is not with tetra-cobalt producers themselves — it is demand.

Downstream players began stockpiling late last year, and by H1 2026 many held more inventory than they needed. As tetra-cobalt accumulated downstream, cathode material plants eventually stopped taking delivery entirely.

Quarterly breakdown: Q1 output totaled approximately 22,800 tonnes, down 8.8% year-on-year, with material producers still carrying high inventories and procurement appetite already weakening. Entering Q2, material plants accelerated destocking, and tetra-cobalt output plummeted to approximately 18,100 tonnes — a year-on-year decline of 39.4%.

Industry concentration has also intensified. CR3 rose from 67.4% in H1 2025 to 72.8% in H1 2026, with CNGR (Zhongwei) capturing roughly one-third of the market and cementing its leadership position. As downstream procurement contracts sharply, orders are concentrating among fewer producers, and the divergence between leaders and second-tier players points to ongoing industry reshuffling.

4. Cobalt Chloride

Compared with the "cliff drop" in tetra-cobalt, cobalt chloride has fared considerably better. In H1 2026, domestic cobalt chloride cumulative output reached approximately 29,500 tonnes, down 14.2% from 34,400 tonnes in H1 2025.

Two factors explain this relative resilience:

• Diversified downstream demand: Cobalt chloride feeds not only cobalt tetroxide but also cobalt powder, additives, and chemical applications — making it less vulnerable to a single product's demand collapse.
• Integrated production layouts: Most cobalt chloride producers are vertically integrated, allowing them to flexibly adjust output based on downstream scheduling. Production cuts and destocking can proceed in tandem, preventing inventory pressure from accumulating excessively at any single node.

Quarterly breakdown: Q1 output totaled approximately 16,300 tonnes, up 3.2% from 15,800 tonnes in Q1 2025, essentially maintaining prior-year levels. In Q2, however, simultaneous production cuts and destocking by tetra-cobalt and cathode material producers caused cobalt chloride demand to contract abruptly. Q2 output fell to approximately 13,200 tonnes, down 28.9% from 18,600 tonnes in Q2 2025 — following the same "stable first half, declining second half" pattern as tetra-cobalt.

The most noteworthy development in cobalt chloride, however, is the shift toward recycled feedstock.

In H1 2025, recycled materials accounted for less than 7% of total cobalt chloride output on average, peaking at just over 10%. By H1 2026, this share had jumped dramatically — reaching 54.4% in June, when recycled output surpassed primary (virgin ore) output for the first time. In the span of one year, recycled material has moved from a supporting role to the leading source.

Two forces are driving this structural shift:

• Primary feedstock contraction: Constrained by tight intermediate product supplies, primary ore-based output fell from 5,192 tonnes in January to 1,990 tonnes in June — a decline of over 60%.
• Recycled feedstock expansion: With relatively stable supply and cost advantages, recycled materials are rapidly filling the gap left by declining primary production.

On the inventory front, both upstream and downstream stocks have been drawn down continuously. Total inventories fell from 9,356 tonnes in January 2026 to 6,180 tonnes in May — a decline of approximately 34%. Upstream smelter inventories dropped from 5,815 tonnes to 3,745 tonnes, while downstream material plant inventories fell from 3,541 tonnes to 2,434 tonnes.

Unlike cobalt tetroxide's structural mismatch — where cathode plants accumulated excess inventory while tetra-cobalt producers were forced to cut output — cobalt chloride has seen synchronized destocking across the value chain, reflecting the proactive adjustment capability of integrated production models.

III. H2 2026 Segment-by-Segment Outlook

The following outlook is based on the current market consensus: no significant seasonal recovery in end-market demand during H2, and the current pace of destocking continuing broadly unchanged.

1. Lithium Cobalt Oxide (LCO)

For H2 2026 (July–December), cumulative LCO output is projected at approximately 47,000 tonnes, down about 36% from 74,000 tonnes in H2 2025. Monthly output is expected to range between 7,500 and 8,000 tonnes — well below the 10,000–13,000 tonne monthly range seen in H2 2025. The second half of 2025 was a period of aggressive restocking by cathode plants, with monthly output averaging 12,000 tonnes; this year, even 8,000 tonnes per month will be difficult to sustain. A lackluster peak season will define the H2 LCO market.

Two core constraints are at play:

• End-market level: Persistently rising chip prices are inflating 3C electronics finished-device costs. Smartphone and PC OEMs have been forced to raise prices, further suppressing already-weak consumer demand. The device replacement cycle has lengthened to nearly 38 months — a historical high. Weak end-market sell-through inevitably transmits upstream.
• Materials level: LCO cells themselves face mounting cost pressure. While cobalt and lithium raw material prices have retreated from their early-year highs, absolute levels remain elevated by historical standards. Meanwhile, the cost disadvantage of alternative chemistries is narrowing — pushing more end-device OEMs to seriously consider NCM-blended solutions.

When end demand is contracting and substitution is accelerating, downstream strategy becomes straightforward: procure as little as possible, and draw down existing inventory wherever possible. Cathode plants are unlikely to restore procurement appetite until orders pick up meaningfully and inventories return to reasonable levels — leaving LCO output with little upward momentum.

For full-year 2026, cumulative LCO output is forecast at approximately 97,000 tonnes, down about 22% from 124,000 tonnes in 2025. H2 output will be lower than H1 — a reversal of the normal seasonal pattern where H2 exceeds H1. Cost-driven price increases suppressing end demand, combined with accelerating NCM substitution on the materials side — both forces will continue to weigh on LCO production through H2.

2. Cobalt Tetroxide (Co₃O₄)

Assuming no meaningful seasonal recovery in end demand and destocking proceeding at the current pace, inventory normalization may not arrive until year-end.

For H2 2026 (July–December), cumulative cobalt tetroxide output is projected at approximately 32,800 tonnes, down about 52.1% from 68,400 tonnes in H2 2025. Monthly output will remain in a low range of 5,200–5,900 tonnes. Even during the traditional peak season of September–December, output is forecast at only 5,200–5,900 tonnes per month — far below the 10,900–11,850 tonne range seen in the same period of 2025.

For full-year 2026, cumulative cobalt tetroxide output is forecast at approximately 75,700 tonnes, down about 38.4% from roughly 122,900 tonnes in 2025. The year traces a pattern of front-loaded stability followed by a steep decline: Q1 offered brief support, output trended lower from Q2 onward, and H2 will linger at depressed levels for an extended period.

The contraction effect transmits from end-market consumer electronics to cathode materials, and is ultimately amplified dramatically at the precursor stage — with inventory mismatches further magnifying the impact. The cobalt tetroxide sector will remain in a slump until industry inventories return to reasonable levels.

3. Cobalt Chloride

For H2 2026 (July–December), cumulative cobalt chloride output is projected at approximately 31,000 tonnes, down about 31.4% from roughly 45,200 tonnes in H2 2025. Monthly output will stabilize in the 4,300–5,300 tonne range — a modest recovery from the H2 trough but still well below the 6,000–7,600 tonne levels of H2 2025.

The recycled-material share warrants continued monitoring: recycled and primary feedstock will remain roughly balanced through H2, with the recycled share fluctuating between 50% and 55% and no further significant increase expected. Primary ore-based output will gradually recover from 1,960 tonnes in July to 2,400 tonnes in December, driven primarily by improved raw material availability as DRC intermediate products arrive at Chinese ports in H2.

For full-year 2026, cumulative cobalt chloride output is forecast at approximately 60,500 tonnes, down about 24.0% from roughly 79,600 tonnes in 2025. The full-year trajectory falls into three phases: stable output in Q1, a cliff-like drop in Q2, and bottoming without meaningful rebound in H2.

Compared with cobalt tetroxide's 38.4% full-year decline, cobalt chloride's contraction is relatively mild — yet both share the same trajectory of steady start, downward drift, then bottoming out. The structural shift from primary to recycled feedstock is the overarching theme for the year.

IV. Supply-Demand and Price Forecast

Before discussing prices, let us address raw material sources. On the supply side, there are three primary feedstocks: intermediate products, recycled materials, and MHP (mixed hydroxide precipitate). On the demand side, consumption is distributed across several end markets: NCM cathode materials, LCO, electrolytic cobalt, cobalt powder, and other chemical applications.

Different demand segments have different feedstock requirements. NCM cathode materials have the most flexible feedstock options — they can use intermediates, MHP, and recycled materials, although some producers have hard requirements for intermediates (reflected in our modeling). LCO, by contrast, can only use intermediates and high-cobalt-content recycled materials. Electrolytic cobalt and cobalt powder also rely primarily on intermediates.

1. Baseline Assumptions for Intermediate Product Imports

2026:

• For the 25Q4–26Q3 intermediate product quota: Chinese-funded enterprises will ship 100% of their allocation to China; foreign-funded enterprises will ship approximately 50% to China.
• Since unused Q2 2026 quotas have been canceled and the full extent of lost volume cannot be comprehensively tallied across individual companies, we adopt the most conservative scenario — treating Q2 quota volumes as effectively zero.

2027:

• For the 26Q4–27Q3 intermediate product quota: Chinese-funded enterprises will continue to ship 100% to China; the foreign-funded shipment ratio is expected to rise to 70%.

Based on these assumptions, we have modeled the supply-demand balance over the coming quarters. Under this baseline scenario, the conclusion is clear: the oversupply regime will not change in H2 2026, and prices will remain under pressure. In 2027, as inventories are gradually digested, fundamentals will shift from loose to tight-balanced, providing price support.

There is, however, one easily overlooked variable in this projection — the shipment strategies of foreign mining companies.

2. Wildcard: Foreign Miners' Shipment Strategies

In our baseline, we assumed a flat 50% China shipment ratio for foreign miners. But foreign miners are not a monolith — strategies vary considerably. Some prioritize rapid monetization, while others prefer to control shipment pace when prices are low.

The single most critical player is the largest foreign mining company. As one of the world's largest cobalt producers, its quota accounts for approximately 60% of total foreign allocations — meaning its shipment cadence alone is sufficient to shift market balance.

Breaking out foreign shipments separately:

• H2 2026: Foreign imports into China will total approximately 10,000 tonnes of contained cobalt. At this volume — and given today's high-inventory environment — no single foreign miner's strategy will be decisive. Q3 is seasonally a demand trough, and inventories are already elevated; continued intermediate product arrivals will only intensify price pressure.

• 2027: Foreign imports into China are projected at approximately 22,000 tonnes of contained cobalt. By then, the largest miner's decision weight changes fundamentally. With nearly 60% of foreign quota share, its shipment pace will directly determine whether the market remains tight-balanced or shifts into deficit.

3. A Key Observation

If select mining companies begin adopting conservative shipment strategies from now — slowing delivery schedules and controlling spot market volumes — inventory drawdown could proceed faster than the baseline scenario, even accounting for the current domestic inventory of nearly 25,000 tonnes of contained cobalt.

Under this path:

• Domestic inventories could return to relatively normal levels before year-end.
• Entering 2027, effective intermediate product supply would shift from "loose" to "tight."
• Prices may not wait for fundamentals to actually tighten before reacting — market expectations of supply tightening typically manifest in prices 1–2 months in advance.

This suggests that the window between late 2026 and Q1 2027 could be the critical period when price sentiment shifts from "bearish" to "stable or even bullish."

This is, of course, only one possible path. The ultimate outcome depends on the interplay of two variables:

1. The actual shipment cadence of foreign miners (especially the strategy of the largest foreign producer).
2. The pace of domestic high-inventory destocking (dependent on end-demand recovery and the sustainability of recycled material supply).

At present, the uncertainty surrounding the former is greater than the latter.

Robyn Wang   +86 15927163529