The copperbelt turns west

On 23 December 2023, a train of copper left Kolwezi and turned the wrong way. For most of the past half-century the ore dug out of the Central African copperbelt — the metal-rich seam straddling the border between the Democratic Republic of Congo and Zambia — has travelled east to Dar es Salaam or south to Durban, to the Indian Ocean and the long way round to market. This train went west, toward the Atlantic. It carried roughly 1,110 tonnes of copper concentrate, the first batch of a 10,000-tonne trial run, and it was bound for the Angolan port of Lobito along a railway that, for most of living memory, had not run at all (Ivanhoe Mines, "Ivanhoe Mines' Exports Commence from Kamoa-Kakula Copper Complex Along Lobito Atlantic Rail Corridor," 2 January 2024).

The distance tells the story. From the Kamoa-Kakula complex near Kolwezi, the run to Lobito covers 1,289 kilometres on the Angolan network plus a further 450 on the DRC's SNCC line — by Ivanhoe's own account, roughly half the distance to Durban, and a journey of about eight days against the twenty-five or so the same copper takes by road to the South African coast (Ivanhoe Mines, 2 January 2024).

A landlocked prize

What makes the direction matter is what the copperbelt has become. The DRC produced an estimated 3.2 million tonnes of copper in 2025 — up about 7% on the year, roughly 14% of world mine production, and second only to Chile — and it remains the source of an estimated 73% of the world's mined cobalt, the two metals that electrification and battery chemistry have turned into strategic commodities (US Geological Survey, "Mineral Commodity Summaries 2026: Copper," February 2026; "Mineral Commodity Summaries 2026: Cobalt," February 2026). Just across the border, Zambia is Africa's second-largest copper producer. Together they sit on one of the most concentrated deposits of the metals the energy transition needs — and both are landlocked, a thousand kilometres and more from any coast.

That single fact has governed the region's trade for a century: whoever controls the rail to the sea controls how the copperbelt reaches the world. There are only a handful of ways out, and until late 2023 every one of them faced an ocean other than the Atlantic.

The railway the war took

The western exit is not new. The Benguela railway was begun under a Portuguese concession granted to the British financier Robert Williams in 1902 and built inland from Lobito from 1903, expressly to carry the copperbelt's wealth to the Atlantic (Wikipedia, "Benguela railway"). Then Angola's civil war, from 1975 to 2002, took it apart — tracks lifted, bridges blown, all but a 34-kilometre stub near the coast abandoned. For a generation the shortest line out of the copperbelt simply did not exist.

Here is the wrinkle that the corridor's new geopolitics tends to leave out: the railway was put back together by China. Between 2006 and 2014 the China Railway Construction Corporation rebuilt the Benguela line at a cost of about $1.83 billion; trains reached Luau, on the Congolese border, in 2013, and full end-to-end operations began in 2018 (Wikipedia, "Benguela railway"). The track now carrying a Western-backed challenge to Chinese mineral logistics was laid by a Chinese state contractor barely a decade ago.

Two oceans, two patrons

What changed in the 2020s was not the steel but who runs the trains. In July 2023 a thirty-year concession to operate the corridor passed to the Lobito Atlantic Railway — a consortium of the commodity trader Trafigura, the Portuguese builder Mota-Engil, and the rail operator Vecturis (Trafigura, "Concession for railway services transferred to Lobito Atlantic Railway in Angola," July 2023). Around it, Washington built a signature piece of infrastructure diplomacy: a US Development Finance Corporation loan of $553 million toward the Angolan railway, part of a $753-million package that also drew in the Development Bank of Southern Africa, against a broader US commitment of some $4 billion and total project value near $6 billion (Trafigura, "Lobito Atlantic Railway secures USD753 million to accelerate development in Angola," 2025; Atlantic Council, "What to know about the Lobito Corridor," 20 December 2024). A planned 830-kilometre greenfield line would extend the corridor from Lucano, on Angola's Benguela line, to the Zambian network at Chingola, with groundbreaking expected by early 2026 (Africa Finance Corporation, "AFC-led Zambia Lobito Rail Project receives boost from Biden visit to Angola," 4 December 2024).

Beijing's answer runs in the opposite direction. In late 2025 China, Tanzania and Zambia signed a $1.4-billion deal to rebuild TAZARA — the 1,860-kilometre line to Dar es Salaam that China first built in the 1970s as a Cold War statement — with a foundation stone laid in Lusaka in November (China-Global South Project, "China, Zambia and Tanzania Seal $1.4 Billion Deal to Modernize Tazara Railway," 1 October 2025; TanzaniaInvest, "Tanzania and Zambia Launch Major TAZARA Railway Rehabilitation," 21 November 2025). The rehabilitation aims to lift the line's freight capacity from about 200,000 tonnes a year to 2.4 million within three years, and to cut transit times by roughly two-thirds (Railway Gazette International, "China-backed Tazara revitalisation officially launched," 4 December 2025).

So the same landlocked ore body now has two state-sponsored railways being rebuilt at once, pulling its copper toward opposite coasts and opposite patrons. The copperbelt has stopped being a place that ships its metal one way; it has become the prize in a contest over which way it ships at all.

The thesis: distance is not the same as leverage

The easy story is a race, and on the map the West is winning it: Lobito is the shorter line, it is already moving copper, and eight days beats twenty-five. But shipments are not the same as capacity, and capacity is not the same as power. The questions worth tracking are narrower than the headlines.

First, can Lobito carry weight, not just samples? A 10,000-tonne trial is a proof of concept against a copperbelt that now produces millions of tonnes a year; the corridor matters only if the Zambia extension is built and the line moves at industrial scale. Second, what does the copperbelt actually capture? A faster railway run by a foreign trader and financed by a foreign government can lower the cost of getting ore out without changing who books the margin — the recurring complaint of an extraction economy is that it earns transit fees while the value accrues downstream and abroad. Third, the contest may not have a single winner. Two functioning railways to two oceans is not redundancy the producers will resent; it is the first real bargaining position the copperbelt has ever held, the ability to play Atlantic against Indian Ocean, Washington against Beijing.

That is the falsifiable claim. If, three trading cycles from now, the Lobito line is still moving trial-scale tonnages while TAZARA's capacity comes online — or if both railways run full but the DRC and Zambia still capture nothing but freight — then the turn west will have been a change of direction without a change of position. If instead the copperbelt uses two exits to set terms it could never set with one, then the wrong-way train of December 2023 will read, in hindsight, as the moment a landlocked ore body stopped being shipped and started to choose.