Cobalt 2026: the DRC chokepoint, the Indonesian flood, and a price floor that has not held: Deluair Consultancy

Cobalt entered 2026 as the most concentrated battery metal in the world and the worst priced. The Democratic Republic of the Congo produced roughly 70 percent of mined supply in 2024, China refined about three quarters of the global total, and prices fell from above 80,000 US dollars per tonne in 2022 to a 24,000 to 30,000 corridor across 2024 and 2025. CMOC overtook Glencore as the largest miner after Tenke Fungurume and Kisanfu ramped, Indonesian high pressure acid leach byproduct flooded the market, and lithium iron phosphate cathodes erased entry level cobalt demand even as total battery deployment kept growing. This brief maps the producer stack, the policy stack, and where the next chokepoint forms. Athena, Argus, and Promethean carry the analytics.

One country, one chokepoint, two stages #

The Democratic Republic of the Congo produced approximately 220,000 tonnes of mined cobalt in 2024, roughly 70 percent of global mine output of around 290,000 tonnes, per USGS Mineral Commodity Summaries 2025. Indonesia was a distant second at roughly 28,000 tonnes, Russia at 8,700 tonnes, Australia at 6,300 tonnes, and the Philippines at 3,800 tonnes. No other producer cracked five percent. The Cobalt Institute puts mined supply growth at about 22 percent year on year, almost all from the DRC and Indonesia.

Refining sits in a different geography. China processed roughly 75 percent of refined cobalt in 2024 per BloombergNEF and the Cobalt Institute, with Huayou, GEM, CNGR, and Zhejiang Hongli running the largest sulfate and tetroxide circuits. Finland, through Umicore Kokkola and Freeport Cobalt, and Canada, through Vale Long Harbour and Sherritt Fort Saskatchewan, account for most non Chinese refined output. The DRC has effectively zero cobalt sulfate capacity: feed leaves Lualaba and Haut Katanga as hydroxide and metal, then reaches battery grade chemistry in Zhejiang, Hunan, or Jiangsu. Two stages, two countries, one combined chokepoint.

Country	2024 mined cobalt (tonnes)	Share of global mine output	Refined cobalt share
DRC	Approximately 220,000	Approximately 70 percent	Negligible refined
Indonesia	Approximately 28,000	Approximately 10 percent	Approximately 4 percent
Russia	Approximately 8,700	Approximately 3 percent	Approximately 4 percent
Australia	Approximately 6,300	Approximately 2 percent	Negligible
Philippines	Approximately 3,800	Approximately 1.3 percent	Negligible
China	Approximately 2,200	Less than 1 percent	Approximately 75 percent

Cobalt mine output and refining share, 2024. Mine figures from USGS Mineral Commodity Summaries 2025. Refining shares from BloombergNEF Battery Metals and Cobalt Institute Annual Report 2024.

The 2024 to 2025 price collapse: HPAL byproducts and the LFP lid #

Cobalt metal traded above 80,000 US dollars per tonne at the 2018 peak and again in early 2022, when supply tightness during the post pandemic restock and the Russia Ukraine shock pushed Fastmarkets MB standard grade above 82,000 US dollars. By the end of 2024 the same benchmark sat near 24,000 US dollars, and across 2025 it traded in a 24,000 to 30,000 US dollar corridor per Fastmarkets and the LME cobalt cash settlement. Cobalt hydroxide payables collapsed in parallel, falling from above 80 percent of the metal price in 2022 to a 60 to 70 percent range across 2024 and 2025, which compressed DRC mine netbacks faster than the headline metal print suggested.

Two structural forces did the damage. First, Indonesian HPAL nickel projects, Sumitomo at Pomalaa, Vale at Sorowako, Eramet at Weda Bay through the Sonic Bay JV with BASF, and Huayou at Halmahera and Sulawesi, scaled mixed hydroxide precipitate production from negligible in 2020 to roughly 28,000 tonnes of contained cobalt by 2024 per the Cobalt Institute. HPAL cobalt is a nickel byproduct with no independent supply discipline: it ships when the nickel project ships, regardless of cobalt price. Second, the cathode mix shifted. Lithium iron phosphate, cobalt free, rose from about 17 percent of global EV cathode in 2020 to above 45 percent in 2024 per BloombergNEF and Benchmark, and nickel rich chemistries moved from NMC622 toward NMC811 and NCA, cutting cobalt intensity per kilowatt hour by roughly 60 percent versus 2018. Total cobalt demand still grew, from about 130,000 tonnes in 2020 to roughly 220,000 tonnes in 2024, because absolute battery deployment outpaced intensity reductions. Demand grew. Price fell. Both are true.

Year	Fastmarkets MB cobalt metal (USD per tonne)	Hydroxide payable (percent of metal)	Global mine output (tonnes)
2018 peak	Approximately 82,000	Approximately 85 percent	Approximately 145,000
2020	Approximately 32,000	Approximately 65 percent	Approximately 140,000
2022 peak	Approximately 82,500	Approximately 88 percent	Approximately 198,000
2023	Approximately 33,000	Approximately 75 percent	Approximately 230,000
2024	Approximately 24,500	Approximately 65 percent	Approximately 290,000
2025 average	Approximately 26,800	Approximately 67 percent	Approximately 305,000 (est)

Cobalt metal price, hydroxide payable, and global mined supply, 2018 to 2025. Prices from Fastmarkets MB cobalt standard grade history and LME cobalt cash. Volumes from USGS Mineral Commodity Summaries 2020 to 2025 and Cobalt Institute Annual Report 2024.

The CMOC takeover of the Lualaba cluster #

CMOC, the Hong Kong and Shanghai listed entity formerly China Molybdenum, displaced Glencore as the largest cobalt miner in the world in 2023 and held that position through 2025. CMOC produced approximately 114,000 tonnes of cobalt across Tenke Fungurume Mining and Kisanfu in 2024 per its HKEX annual report, against Glencore's 38,200 tonnes from Mutanda and Katanga per the 2024 production report. CMOC bought the Tenke controlling stake from Freeport McMoRan in 2016 for 2.65 billion US dollars and Kisanfu in 2020 for 550 million. The 2022 to 2023 export blockade, when Gecamines disputed royalties on undeclared reserves at Tenke, ended in April 2023 with a settlement including an 800 million US dollar payment and a higher royalty schedule, and the export resumption anchored the 2023 to 2024 surge.

Glencore restarted Mutanda in early 2022 after a 2019 to 2021 idle period, but the restart did not restore leadership. Mutanda produced roughly 21,000 tonnes of contained cobalt in 2024, well below its 2018 peak above 27,000, and Glencore guided 2025 cobalt to a 35,000 to 40,000 tonne range across Mutanda and Katanga. Eurasian Resources Group at Metalkol, China Nonferrous at Deziwa, Zijin at Kolwezi, and smaller Chinese owned operators round out the industrial base. Of the roughly 220,000 DRC tonnes in 2024, Chinese controlled assets account for above 70 percent on a contained cobalt basis, per Cobalt Institute and company filings.

The artisanal and small scale mining segment supplies an additional 15 to 20 percent of DRC output depending on year and price, per OECD and the Responsible Minerals Initiative. ASM cobalt enters the formal market through Entreprise Generale du Cobalt, the state monopoly under the 2019 decree, and through informal traders that route material into industrial wash plants for blending. Enforcement has been partial, and the 2024 export tax review reopened how artisanal volumes are taxed and traced.

Due diligence stack: OECD, Dodd-Frank, EU CMR, CIRAF #

Compliance scaffolding around DRC cobalt is older and denser than for lithium or rare earths. OECD Due Diligence Guidance, third edition 2016, sets the five step framework. Dodd-Frank Section 1502, in force since 2012, requires SEC registrants to disclose conflict mineral sourcing for tin, tantalum, tungsten, and gold from the DRC and adjoining countries; cobalt is not directly covered, but every major buyer applies the OECD framework to it. The EU Conflict Minerals Regulation 2017 821 applied from January 2021 to 3TG importers above defined thresholds. EU Battery Regulation 2023 1542, in force from August 2023 with phased due diligence from August 2025, extends mandatory supply chain due diligence to cobalt, graphite, lithium, and nickel.

Industry layered the Cobalt Industry Responsible Assessment Framework, CIRAF, on top from 2019. CIRAF is administered by the Cobalt Institute, aligned to OECD, and used by Apple, BMW, Tesla, Volkswagen, Samsung SDI, and LG Energy Solution as the supplier audit reference. Apple's 2024 Conflict Minerals Report disclosed 100 percent of identified cobalt refiners as participating in third party audits. Tesla's 2024 Impact Report sources cobalt from CIRAF and RMAP audited facilities. BMW signed direct purchase from Managem in 2018, which operates Bou-Azzer in Morocco, the only cobalt mine in Africa outside the DRC of meaningful scale at roughly 2,000 tonnes per year.

Two unresolved frictions remain. First, the unwind of Dan Gertler's Ventora holdings, sanctioned by OFAC in 2017 and partially restructured in the 2022 settlement that returned Metalkol and Mutanda royalty streams to the DRC state, is incomplete and continues to surface in audits. Second, Norilsk Nickel, the second largest refined cobalt producer outside China at approximately 4,500 to 5,000 tonnes per year, has been excluded from most Western OEM chains since the 2022 sanctions package, but its metal still reaches refined markets through traders and the LME warrant system, which CIRAF audits cannot fully trace.

Demand split: NMC, NCA, LFP, sodium-ion, and 2030 deployment #

Passenger EV cathode chemistry split in 2024 was approximately 45 percent lithium iron phosphate, 38 percent nickel manganese cobalt, 12 percent nickel cobalt aluminum, and the remainder a mix of lithium manganese oxide and emerging chemistries, per BloombergNEF and Benchmark. LFP is structurally cobalt free. NCA, used by Tesla and Panasonic in Model 3, Model Y, and stationary storage, contains roughly 9 to 12 percent cobalt by cathode mass. NMC ranges from NMC532 at 12 percent cobalt to NMC811 at 6 percent and NMC9 0.5 0.5 below 5 percent. IEA Global EV Outlook 2024 and Critical Minerals Outlook 2024 project LFP share holds in a 40 to 50 percent corridor through 2030.

Sodium-ion reached small commercial volumes in 2024 in CATL, BYD, and HiNa entry level vehicles and Chinese grid storage. Sodium-ion is cobalt free and lithium free. 2030 forecasts put it at 5 to 10 percent of global cell deployment. Cobalt Institute and BloombergNEF central cases converge on 290,000 to 330,000 tonnes of cobalt demand by 2030, against a 2024 starting point near 220,000 tonnes. Demand keeps growing in absolute terms, but the elasticity of cobalt demand to EV unit growth has fallen by more than half since 2018 because of LFP and intensity reductions inside nickel rich families.

For precursor cathode active material makers, the cobalt share of revenue in a typical NMC line fell from above 30 percent in 2018 to roughly 12 to 15 percent by 2024 at prevailing prices. BASF Schwarzheide and Harjavalta, Umicore Kokkola and Nysa, Posco Future M, LG Chem, and Samsung SDI together built non Chinese precursor capacity reaching roughly 250,000 tonnes per year by 2025. The bottleneck for that capacity is no longer cobalt: it is qualified non FEOC nickel sulfate and lithium hydroxide. Cobalt has moved from binding constraint to passenger metal.

Where the next chokepoint forms #

The first order chokepoint, DRC mine concentration, is not loosening. Indonesian HPAL cannot replace DRC at scale because nickel project economics drive HPAL siting. The Code Minier 2018 raised the cobalt royalty to 10 percent as a strategic mineral, and the 2024 export tax review proposed an additional levy on unprocessed hydroxide that would push effective government take above 20 percent of mine gate value. Lualaba and Kolwezi economics still work at 24,000 US dollar metal because CMOC, Glencore, and ERG run at unit costs below 15 dollars per pound, but new entrants and ASM compress hard at this level.

The second order chokepoint, Chinese refining concentration, is the one moving in 2026 and 2027. EU Battery Regulation due diligence from August 2025, IRA Section 30D FEOC rules tightening through 2027, and the Critical Raw Materials Act 65 percent processing self sufficiency target for 2030 push qualified non Chinese refining to expand. Umicore announced a Loyalist precursor plant in Ontario, partially de-scoped in 2024 against weak EV demand. BASF restructured its Sonic Bay HPAL JV with Eramet in 2024 and pivoted toward European refining of imported intermediates. Vale Long Harbour, Sherritt Fort Saskatchewan, and a planned Glencore expansion at Nikkelverk Norway anchor non Chinese supply, but cumulative additions through 2027 cover less than a quarter of refined demand outside China.

For OEMs, FEOC traceability now requires mine and refinery level audit, and the cheapest path at scale is long term offtake from Bou-Azzer, Vale, Sherritt, and CMOC volumes routed through Korean or Japanese refining. For refiners, greenfield non Chinese capacity does not clear at 24,000 US dollar metal without offtake premium, IRA Section 45X credits, or EU CRMA support. Argus tracks the FEOC ownership graph, Athena maintains price and payable history weekly, and Promethean simulates Section 30D, EU Battery Regulation, and DRC export tax outcomes on landed precursor cost. The next chokepoint is not in the rock. It is in the qualified refined tonne.

Sources #

Cite this brief

@misc{hossen2026cobaltdrcsupplychain2026,
  author = {Hossen, Md Deluair},
  title  = {Cobalt 2026: the DRC chokepoint, the Indonesian flood, and a price floor that has not held},
  year   = {2026},
  url    = {https://deluair.com/consultancy/insights/cobalt-drc-supply-chain-2026},
  note   = {Deluair Consultancy briefs}
}

Hossen, M. D. (2026). Cobalt 2026: the DRC chokepoint, the Indonesian flood, and a price floor that has not held. Deluair Consultancy briefs. https://deluair.com/consultancy/insights/cobalt-drc-supply-chain-2026

Hossen, Md Deluair. "Cobalt 2026: the DRC chokepoint, the Indonesian flood, and a price floor that has not held." Deluair Consultancy briefs, 2026-04-26. https://deluair.com/consultancy/insights/cobalt-drc-supply-chain-2026.

On the watchlist

Upcoming dates that bear on this brief.

See the full firm watchlist for the rest of the calendar.

June 1, 2026 Regulation

ISA Mining Code Council session

Whether Council reaches consensus on tax-and-royalty regime, environmental thresholds, and contractor liability framework.

Throughout 2026 Trade

DRC Code Minier review and royalty regime

Whether the export royalty rate moves up and the implication for CMOC, Glencore, and the Indonesian HPAL substitution flow.

Related insights

Adjacent reading.

Trade and tariff analytics

Cobalt 2026: the DRC chokepoint, the Indonesian flood, and a price floor that has not held