US Bets on Madagascar Rare Earths to Loosen China's Magnet Grip

Harena Rare Earths confirmed on 22 July 2026 that it had signed a project development funding agreement with the US International Development Finance Corporation, known as the DFC. Under the agreement the US agency committed up to $4.84 million toward the development budget of the Ampasindava ionic clay rare earth project on the northwest coast of Madagascar. The London-listed company called the deal the first direct financial commitment from the US government to the asset, and said drawdown of the first tranche was already under way when the announcement was made. Several news agencies later reported the commitment as $4.48 million, while the company statement and the specialist mining press cite $4.84 million.

The funds are earmarked for permitting work, environmental and social studies, and a proof-of-concept pilot plant. That package is designed to carry the project toward a definitive feasibility study and, after that, a construction decision. The DFC has also retained the right to be considered for a larger financing or investment role as the project advances, which the company presented as a signal for its future funding pipeline.

The reason this small ticket matters to supply chain planners has little to do with its size. China controls the large majority of global rare earth mining and an even larger share of separation and refining capacity, and it has reinforced that position through export controls on magnet materials. Ampasindava is one of the few sizeable ion adsorption clay deposits under development outside China. Deposits of that type are valued because they yield heavy rare earths such as dysprosium and terbium alongside light elements, a mix that most hard-rock projects struggle to deliver at scale.

The project numbers set the ceiling on what it can contribute. Harena holds a 75 percent interest in Ampasindava, which hosts a reported resource of 699 million tonnes grading 868 parts per million total rare earth oxides. A prefeasibility study published earlier in 2026 outlined 71,000 tonnes of rare earth oxides over a 20 year mine life for an investment of about $142 million, with development work starting in 2027. Company guidance points to roughly 4,000 tonnes of rare earth oxides a year once running, including about 1,700 tonnes of high-value magnet oxides based on neodymium, praseodymium, dysprosium and terbium. Total investment for the project is generally quoted at around $150 million.

As of 2 August 2026 the project does not yet hold the exploitation permit it needs to move to construction. Executive chairman Ivan Murphy told Reuters in the week of 27 July that Harena was seeking that permit and hoped for approval within a couple of weeks. Madagascar's mines ministry did not comment publicly on the timing. No grant has been confirmed since, so permitting remains the single biggest near-term gate on the schedule. The US State Department separately framed its support as part of a wider effort to increase American and allied investment in African critical minerals.

The host jurisdiction adds a second layer of uncertainty. Madagascar has been run by a military-led transitional administration since the removal of President Andry Rajoelina in October 2025, and the African Union suspended the country shortly afterwards. That same administration lifted a 16 year freeze on new mining permits at the end of January 2026 for critical minerals including rare earths, nickel, cobalt, graphite and mineral sands, while keeping gold excluded. The reopening is what makes an Ampasindava permit possible at all, but it also means the approval will be issued by an interim government whose legal framework and timelines are still settling. Infrastructure gaps and the biodiversity profile of the Ampasindava peninsula are further constraints on both cost and environmental approval.

Mining the clay is only half of the dependency problem. Harena has said it is evaluating processing routes in the United States and Europe, and has named MP Materials, USA Rare Earth and Solvay among the parties it has discussed refining with. Without a firm non-Chinese separation agreement, concentrate from Madagascar could still end up flowing back into Chinese processing plants, which would leave the strategic exposure largely unchanged.

For buyers of magnet materials the practical conclusion is that this announcement changes nothing about supply in 2026 or 2027. Even if permitting, financing and construction all run to plan, first production is targeted for the middle of 2028, and an annual output of 4,000 tonnes is modest against a global market where Chinese mine quotas alone run into the hundreds of thousands of tonnes. Ampasindava should be read as one line in a medium-term diversification portfolio rather than as relief from current pricing and export control pressure.

The milestones worth tracking from here are the grant of the exploitation permit, the commissioning and results of the pilot plant, the timing of the definitive feasibility study, any follow-on DFC financing, and the signature of a binding refining or offtake agreement outside China. Failure or delay at any of those points would push first output beyond 2028 and extend the current concentration of magnet rare earth supply.