​​​

24, September, 2026

​​

The Slowest Asset

African material stocks, and what they can be made to fund.

​​

--

 

Congo has proved that an African resource holder can force a price.

 

It has also proved why winning this way loses. The premium being harvested is drawn from a demand base that engineers are actively removing, and harvesting it accelerates the removal. That is not a warning about the distant future. It is the mechanism now running.

 

The trade worked

 

In February 2025 the Democratic Republic of Congo suspended cobalt hydroxide exports. The DRC holds roughly 72% of global mine supply. The suspension ran eight months. In October it was replaced by a quota system: 87,000 tonnes for 2026, plus a 9,600-tonne strategic allocation at the state's discretion — together around half of 2024 export levels.

 

Cobalt opened 2025 at $24,343 a tonne, a nine-year low. It closed the year at $53,005. Hydroxide rose 328%. Sulphate rose 266%. Global stocks fell materially.

 

Most commentary either ignored this or disapproved of it. It was a competent, well-executed exercise of market power. The design was the clever part: the throttle sits at the export gate, not the mine. Most Congolese cobalt comes out of the ground as a copper byproduct, and copper economics were never disturbed. Production continued. Only shipment was rationed. A state that had been told for years it had no pricing power found the one lever that did not require it to stop mining, and pulled it.

 

The money is real. The question is what it is money for.

 

Why winning this way loses

 

LFP chemistry accounted for more than 55% of global EV battery deployment in 2025, up from nearly 50% the year before. In emerging markets it powers two-thirds of electric car sales. LFP contains no cobalt at all.

 

Inside the shrinking cobalt-bearing remainder, thrifting has already happened. NMC721, NMC811 and NCA — formulations at roughly 10-15% cobalt in metal content — made up about 80% of 2025 deployment of cobalt-containing chemistries. The high-cobalt cathode is already a legacy product.

 

The structural response is slower and larger. Sodium-ion entered mass production and stands at about 1% of lithium-ion cell capacity, with announced 2030 projects reaching roughly 7% of committed lithium-ion capacity for that year. Small. Also directional, and aimed precisely at the low end of the market where volume lives.

 

Now the part that matters. Fastmarkets' Oliver Masson, on the current price: the longer prices remain elevated, the more likely EV manufacturers move to low-cobalt or cobalt-free chemistries.

 

This is the iron law of the business. High prices fund the engineering that removes the demand. Congo's success is the signal a cathode designer needed. Every month of quota strengthens the case for the chemistry that does not require Congo at all.

 

The asymmetry underneath

 

A mine is a fifteen-to-twenty-year asset. A refinery is three to five years. A cathode reformulation is two to three. Materials-efficiency work is continuous.

 

The continent holds the slowest asset in the chain and negotiates with the fastest decision-makers in it. That is the whole structure, and it does not depend on anyone's intentions. Slow assets held against fast revision lose over enough cycles, regardless of how strong the position looks at any single moment.

 

There is a second asymmetry compounding the first. What the West is currently paying for is not scarcity. It is jurisdictional diversification — non-Chinese supply, bid up by governments who let a concentration happen and now fear it. That is a political attribute, not a geological one. It reprices on a political clock, and it expires the moment Washington and Beijing reach an accommodation.

 

Congo is not selling cobalt at $53,000. It is selling a hedge against Chinese concentration, at a price set by how frightened the buyer is this year.

 

What sophisticated capital is actually doing

 

The clearest evidence is a divergence almost nobody has remarked on.

 

Critical mineral prices rebounded strongly in 2025 and early 2026 — lithium more than doubled, cobalt rose around 130%. Over the same period, critical mineral investment fell 9%. Battery-metal investment fell more than 20%. Lithium companies cut investment by around 40%.

 

Prices up, capital expenditure down. Capital does not behave that way in front of a structural shortage. It behaves that way in front of a price it believes is policy-made and temporary. The people who commit fifteen-year money to fifteen-year assets are declining to extrapolate this premium, and they are the best-informed readers in the market.

 

Anyone building a fiscal plan on these prices is taking the other side of that trade.

 

The beneficiation case, reviewed honestly

 

Every resource-holding state on the continent now has value addition in its development plan. Do not export the ore. Process it here. Capture the chain.

 

Indonesia is the case everyone cites, and it genuinely worked on its own terms. After the raw nickel export ban, nickel exports went from around $1 billion a year to roughly $15 billion by 2023. Indonesia took more than 50% of global output. Foreign direct investment surged into Sulawesi and North Maluku.

 

Look at what it cost and who collected.

 

The plants were financed by Chinese capital — Tsingshan alongside China Development Bank, China Exim, Bank of China, ICBC. The result, in the assessment of researchers reviewing the policy, was an oligopsony in which Chinese-Indonesian smelting consortia set the price paid to miners. Output expanded until nickel traded at roughly half its 2022 peak on oversupply, at which point refiners cut production and laid off workers. Mining accounts for about 3.6% of the regional workforce; agriculture and services employ over 76%, with minimal formal employment gain. More than 75,000 hectares of forest were cleared between 2007 and 2022. Between 2015 and 2024, incidents at nickel facilities caused 101 deaths.

 

And the enabling input was not policy resolve. It was electricity: 80% of Indonesia's captive coal capacity was built after 2014, concentrated in exactly those two provinces.

​

So the honest interpretation of the success case is that a state can win the export-value argument and lose the development argument in the same decade. The hard currency arrived. The employment, the price-setting power and the industrial depth did not.

 

The constraint that actually binds

 

Refining is where the West says it wants diversification. Concentration went the other way: the average share of the top refining country rose to 72% in 2025 from 70% in 2023.

 

The reason is not policy failure. Refining is capital-heavy, energy-heavy, cyclical and thin-margined. It is dominated by whoever will run it at low margin on subsidised power. That is a description of an electricity advantage, not a minerals advantage.

 

Which locates the binding constraint on African industrial development exactly where the development plans do not put it.

 

Roughly 565 million people in sub-Saharan Africa have no access to electricity — about 85% of the global deficit. The access statistic understates the industrial problem, because a household receiving four hours a day is counted as connected, as is a clinic that cannot hold a vaccine cold chain and a workshop that cannot run machinery consistently. Productive activity needs the upper tiers of supply. Most of what has been added is in the lowest. For contrast: Central and Southern Asia reduced their access deficit from 414 million in 2010 to 27 million by 2023.

 

No beneficiation policy survives contact with this. A smelter is an electricity consumer with a mineral hobby. Without firm industrial power at a competitive tariff, value addition is a communiqué.

 

What is genuinely scarce

 

Not every African material stock sits on the wrong side of the substitution clock.

 

South Africa supplies 70-80% of mined platinum and holds around 91% of global reserves; South Africa, Russia and Zimbabwe together account for roughly 90% of primary PGE supply. The revealing number is the supply response: South African primary production fell 26% between 2006 and 2025, across multiple price cycles above $2,000 an ounce. Platinum is entering a fourth consecutive annual deficit, with above-ground inventory drawn down roughly 42% since 2023. That is a genuinely inelastic position of a kind Congolese cobalt is not.

 

Arable land is the larger prize and attracts a fraction of the attention. Nobody is engineering a substitute for food. The demand curve cannot be thrifted by a materials scientist.

 

And the demographic position is the only major one in the world moving in the right direction while everywhere else ages — an asset that cannot be replicated by policy, since the states that tried to move their own labour stock deliberately discovered they could reduce it far more easily than restore it.

 

What we're watching right now

​

If cobalt sustains prices near current levels through 2028 while LFP's share of global battery deployment stalls or reverses, then substitution is slower than we claim and export controls are a durable source of rent rather than a temporary one.

 

If critical-minerals investment turns up sharply while prices hold, sophisticated capital has changed its mind about the premium being policy-made, and our reading of that divergence is wrong.

 

If a significant non-Chinese refining industry is established and sustained in a jurisdiction without cheap firm power, the electricity constraint is not binding in the way we have described and the beneficiation case is stronger than we allow.

​

We will revisit each.

 

What the money is for

​

The window is real. It is a window.

 

Sell the hedge while it is bid. Take the premium in cash, on the shortest tenor available, and price it as the expiring political asset it is rather than the permanent endowment it is being described as.

​

Then spend it on what does not reprice. Firm power, because every industrial ambition on the continent reduces to it. Logistics, because deliverability rather than endowment is what a buyer actually purchases. And the unglamorous institutional work that turns an entity into a counterparty — because no book in the world buys a country, and the operators who can be verified will be funded long after the premium has gone.

 

The minerals will be in the ground either way. That is the one thing about them that has never been in question.