Take the Money. Keep the Proof.
Washington’s new minerals program cannot pour concrete. That is the interesting part: it pays for proof — and proof is the layer that decides where every plant gets built.
The week the State Department published its U.S.-Africa Strategic Investment Program — up to $500 million, roughly ten awards, critical minerals at the center — the tempting readings arrived from both directions. Enthusiasts saw Washington finally writing checks for African mining. Skeptics noticed that construction is excluded and concluded the opposite: half a billion dollars that cannot build so much as a warehouse, another program of studies and workshops. Read the exclusion properly and it is the most revealing clause in the document. Washington is not offering to buy plants. It is offering to buy certainty — geological data, feasibility work, pilot processing, transaction support. And certainty about minerals is not an abstraction. It is a file.
No bank finances rock. Banks finance evidence: what the ore actually contains, whether recovery holds outside a laboratory, what the process demands in power and water, what happens to the tailings, which buyer will take the output and at what specification. Until that file exists, a deposit is geology. The moment it exists, the deposit is collateral. Whoever assembles the file decides, in practice, where the plant gets built, in whose currency the work is invoiced, and which ventures graduate from prospect to operator.
For a century that file has been assembled somewhere else. African drill core flies to reference laboratories headquartered in Geneva, Brisbane, Paris and London; flowsheets are drawn in Denver and Perth; the conclusions come home as consultants’ PDFs, priced in dollars and owned by the firms that wrote them. The branch office draws the sample. The headquarters holds the standard. Africa is the world’s great exporter of samples and one of its smallest owners of conclusions — about its own ground. That, not the absence of smelters, is the dependency underneath the dependency: the plant is downstream of the proof.
What this moment offers is rare: a counterparty willing to pay for the proof and structurally indifferent to where it is produced. The program’s money is grant money — non-dilutive, no equity taken, no lien registered. Washington’s interest is that the file exist, because its supply chains need the certainty. Nothing in the program requires the assay to run in Denver rather than Kitwe, or the pilot plant to sit in a consultancy’s yard rather than an hour from the pit in Kolwezi. The Madagascar rare-earths commitment shows the shape: $4.48 million of early money wrapped around a proposed $150 million project — not to replace the financing, but to manufacture the evidence that lets financing arrive.
Run the program the usual way and it will end the usual way: feasibility contracts to the familiar firms, samples on planes, corridor studies on a server in Washington — and when the appropriation lapses, America will be more secure and Africa will be precisely where it started, rich in ore and poor in proof. The enclave model does not need fences to work; laboratories will do. Run it the other way — testwork in Copperbelt labs, flowsheet intellectual property registered to companies in Lubumbashi and Boké, technicians trained on equipment that stays — and the program’s expiry date stops mattering. Programs keep political time. Capability keeps generational time.
So measure this one by different numbers than the press releases will offer. Not dollars announced, not memoranda signed. Where do the assays run? Who owns the flowsheet when the pilot ends? Can the company that proved one deposit walk the next one through the same file — and sell that service to three buyers rather than one patron? A firm that can prove processing works serves every mine in its corridor. That is a more durable asset than any single deposit it certifies.
The full dispatch maps where those firms should sit — the Copperbelt junctions, the corridor gateways, the port ends of Lobito, TAZARA, Walvis Bay, Nacala, Beira and the Guinea lines. The commentary’s point is shorter. The proof layer of the minerals economy will be built this decade, because the buyers’ anxiety demands it. The only open question is the address. Take the money. Keep the proof.