Issue 02 · January 2026Biweekly
The Switch Stack
Where the numbers flip the narrative
Share on X Share on LinkedIn Copy Link
The Investigation · 12 min read

The Green Transition’s Invisible Workforce

The marketing says “zero emissions.” The supply chain says an estimated 40,000 children in artisanal cobalt mines in the DRC, working 12-hour shifts for $1–2 a day. When an EV manufacturer publishes its ESG report, the children in the mine are not in the spreadsheet.

The Tunnel

In Kolwezi, in the Lualaba province of the Democratic Republic of the Congo, a child wakes before dawn. He is somewhere between seven and ten years old—his exact age is uncertain because no one recorded his birth. By the time the sun clears the ridge of excavated earth that rings his village, he will be underground. The tunnel he enters is hand-dug, roughly a metre wide, and descends into darkness. Some tunnels in this region go down 100 metres. He carries no safety equipment. He has no helmet, no lamp that meets any industrial standard, no harness. He has his hands.

The mineral he is digging for is cobalt—a blue-grey metal that is essential to the lithium-ion batteries that power smartphones, laptops, and electric vehicles. In 2024, the Democratic Republic of the Congo produced 78% of the world’s cobalt supply. Roughly 15–30% of DRC cobalt comes from artisanal and small-scale mining operations like the one this child works in. [PARTIALLY VERIFIED — artisanal share varies by source]

The child earns between one and two US dollars per day. Amnesty International documented these wages in 2016. In 2024, researchers from the University of Bath confirmed that artisanal miners still earn poverty wages. The child has never heard of Tesla, Apple, or the Paris Agreement. He will not live to see its targets met.

· · ·

Follow the Cobalt

The cobalt the child mines follows a supply chain with at least six links before it reaches your dashboard. From the artisanal mine, it goes to a local trader—a négociant—who buys it by weight without asking where it came from. The trader sells to a depot. The depot sells to a processing plant, many of which are Chinese-owned. The processed cobalt is exported to a cathode manufacturer, typically in China, South Korea, or Japan. The cathode goes into a battery cell. The battery cell goes into a car.

At each link, the provenance becomes harder to trace and easier to deny. By the time the cobalt reaches a battery factory in Ningde or Ulsan, it has been washed, refined, and blended with cobalt from dozens of sources. The original mine—and the child who worked it—is invisible.

No company wants to concede that the rechargeable batteries used to power smartphones, tablets, laptops, and electric vehicles contain cobalt mined by peasants and children in hazardous conditions. — Siddharth Kara, Cobalt Red (2023)
· · ·

The Health Ledger

In February 2024, researchers published the most comprehensive health survey of DRC cobalt miners to date, based on fieldwork in Kolwezi and Likasi. The results were published in African Arguments and reviewed by the Centre for Development Studies at the University of Bath.

DRC Cobalt Miner Health Survey (RAID / University of Bath, 2024)

82% of surveyed miners reported skin problems
85.4% reported chronic breathing issues
52.7% reported a major injury in the past 12 months
937 miners said someone close to them died in the past 2 years from mining-related illness, injury, or mine collapse

In November 2024, Mongabay reported on preliminary findings linking cobalt mining to reproductive health damage in women living near mines. Researchers found evidence that mining toxicity can “disrupt gametogenesis”—the formation of sex cells—raising the possibility that the green energy transition is creating a reproductive health catastrophe in the communities that supply its raw materials.

RAID UK independently tested the water near mining operations. Of five rivers assessed in the Kolwezi area, all were contaminated by acidified industrial pollution. Two—the Katapula and Kalenge—were classified as “hyper-acidic.” Two more—the Dipeta and Dilala—were “very acidic.” None of the four can support aquatic life. None are safe for human consumption. The communities that live beside them have no alternative water source.

The most fortunate tunnel diggers in Kasulo earn around $3,000 per year. The CEOs of the companies that buy the cobalt they mine earn $3,000 in an hour.

· · ·

The ESG Mirage

The 2026 Lead the Charge Leaderboard ranks 18 automakers on supply chain sustainability. Tesla holds the top position for the second consecutive year with an overall score of 49%. Ford, Volvo, Mercedes-Benz, and Volkswagen follow.

Consider what that top score means: even the most transparent automaker in the world fails more than half the criteria assessed by the most comprehensive independent supply chain audit. Tesla discloses that more than 55% of its cobalt is sourced directly from mines or smelters with binding social and environmental requirements. The other 45% passes through intermediaries with limited visibility.

Tesla’s MSCI ESG rating is BBB—average. Its Sustainalytics risk rating is 24.7—medium risk. These are not damning scores. They are mediocre scores for a company that markets itself as the vanguard of the sustainable future.

In September 2024, the US Department of Labor expanded its List of Goods Produced by Child or Forced Labour to include cobalt from DRC industrial mines—not just artisanal operations. This means the US government now recognises that forced labour exists at the industrial mining level, not only in the informal sector that companies find easy to distance themselves from.

· · ·

The Lawsuit

On December 15, 2019, International Rights Advocates filed a federal lawsuit against Apple, Alphabet (Google), Dell, Microsoft, and Tesla. The suit was brought on behalf of fourteen Congolese families whose children had been killed or severely injured in cobalt mine collapses. The named plaintiffs included children who had been paralysed, who had lost limbs, and whose families had received no compensation of any kind.

The complaint alleged that the companies “knowingly benefited from and aided and abetted the cruel and brutal use of young children” in DRC mines. It was filed by Siddharth Kara, an anti-slavery economist affiliated with Harvard’s T.H. Chan School of Public Health.

In March 2024, the US Court of Appeals for the District of Columbia ruled 3-0 in favour of the companies. The court held that purchasing cobalt through a supply chain constituted nothing more than an “ordinary buyer-seller transaction” and did not create legal liability for conditions at the mine. The case was dismissed.

She had been thrown to a pack of wolves by a system of such merciless calculation that it somehow managed to transform her degradation into shiny gadgets and cars sold around the world. — Siddharth Kara, describing a child named Elodie, Cobalt Red (2023)
· · ·

What the Industry Gets Right

Any honest account of this issue must acknowledge that the industry is making real changes.

Battery chemistry is shifting. LFP (lithium iron phosphate) batteries contain zero cobalt. They now represent approximately 40% of the global EV battery market, driven primarily by Chinese manufacturers. The average cobalt content per battery electric vehicle has dropped from roughly 12 kg in 2018 to 4.8 kg in 2023. Sodium-ion batteries—which contain neither cobalt nor lithium—are entering commercial production.

Regulation is tightening. The EU Battery Regulation, effective for vehicles placed on the EU market from February 2027, will mandate that automakers and battery producers demonstrate traceability of cobalt, lithium, nickel, and graphite back to the mine of origin. This is the most significant regulatory development in critical minerals governance.

Recycling is scaling. EU mandates require 90% recovery rates for cobalt, copper, and nickel by late 2025, escalating to 95% by 2031. Breakthroughs in hydrometallurgy are achieving 95% lithium recovery. Recycling is projected to supply 10–15% of lithium demand by late 2025.

These are real developments. They represent genuine progress. The question is whether they are fast enough to matter to the child in the tunnel in Kolwezi today.

· · ·

The Water Price

Cobalt is not the only mineral with a human cost. Lithium extraction—the other critical battery input—carries its own ledger.

In Chile’s Salar de Atacama, one of the driest places on Earth, lithium mining operations consume an estimated 65% of the region’s water. Producing one tonne of lithium carbonate requires roughly 500,000 to 600,000 litres of brine water to be evaporated. A 2024 peer-reviewed study found that portions of the salt flat are subsiding by up to 2 centimetres per year. In October 2024, the Council of Atacameño Peoples filed a formal complaint against the mining companies, accusing them of destroying the aquifer.

The industry disputes the causal link. A University of Massachusetts study concluded that meteorological conditions, not mining, explain changes to the water table. The indigenous communities who live there and are watching their water disappear are not persuaded.

Globally, less than 10% of lithium-ion batteries are recycled. More than 90% are landfilled. The minerals extracted at such human and environmental cost are, overwhelmingly, used once and discarded. [PARTIALLY VERIFIED — exact recycling rate varies by methodology]

· · ·

The Invisible Workforce

Electric vehicles produce approximately 40–52% fewer lifecycle greenhouse gas emissions than comparable gasoline vehicles. This is a verified scientific finding, confirmed by the IEA, the US Department of Energy, and independent lifecycle analyses. The green transition is real, and it is necessary.

But “zero emissions” is not a scientific statement. It is a marketing label. It counts the tailpipe and erases the mine. It counts the dashboard and forgets the tunnel. It publishes the ESG report and omits the river.

The estimated 40,000 children in DRC cobalt mines [PARTIALLY VERIFIED] are not a bug in the green transition. They are a feature of the supply chain that makes it possible. They are invisible not because they cannot be seen, but because the frameworks designed to measure “sustainability”—ESG scores, lifecycle analyses, carbon labels—were not built to see them.

The lawsuit was dismissed. The certification scheme was de-listed by its own industry body. The river is hyper-acidic. The child is still digging. And somewhere in Palo Alto, a software engineer feels good about driving electric.

“Colonialism hardly ever exploits the whole of a country. It contents itself with bringing to light the natural resources, which it extracts, and exports to meet the needs of the mother country’s industries.”

— Frantz Fanon, The Wretched of the Earth (1961)

Sources & Methodology

  1. 40,000 children: UNICEF estimate (~2012), cited in Amnesty International “This Is What We Die For” (2016). No updated independent census. [PARTIALLY VERIFIED]
  2. 78% DRC cobalt: Cobalt Institute Market Report 2024 (May 2025). VERIFIED.
  3. $1–2/day wages: Amnesty International (2016); African Arguments / University of Bath (Feb 2024). VERIFIED.
  4. Miner health data: RAID UK / University of Bath, published in African Arguments (Feb 2024). VERIFIED.
  5. River contamination: RAID UK, environmental report (2024). VERIFIED.
  6. Reproductive health: Mongabay (Nov 2024), citing preliminary research. VERIFIED as reported.
  7. Kalando collapse: Al Jazeera (Nov 15, 2025). VERIFIED.
  8. Tesla ESG scores: MSCI BBB; Sustainalytics 24.7; Lead the Charge 49% (2026). VERIFIED.
  9. Lawsuit dismissal: US Court of Appeals, DC Circuit (Mar 2024). CNN, JURIST. VERIFIED.
  10. US DOL designation: September 2024. VERIFIED.
  11. Cobalt per battery: Adamas Intelligence (2023–2024). VERIFIED.
  12. Water data: ScienceDirect (2024); Mongabay (Dec 2024). VERIFIED/PARTIALLY VERIFIED.
  13. Recycling rate: Industry estimates, ~5–10%. [PARTIALLY VERIFIED]
  14. EV lifecycle emissions: IEA Global EV Outlook 2024; US DOE FOTW #1357 (Aug 2024). VERIFIED.