In Q3 2024, US rare earth mine production hit an all-time high. Yet the Defense Logistics Agency's rare earth purchase orders remained flat. That divergence is a red flag. Over the same period, container ship tracking data shows a 30% increase in ore shipments from US mines to Asian processing hubs. The policy narrative was 'supply chain security.' The on-chain reality is 'token issuance without a utility.' This is not a mining problem. It is an infrastructure problem—the US built the mint but not the refinery.
Rare earth elements are the silicon of advanced militaries. They go into F-35 radar gyrometers, missile guidance systems, and submarine propulsion. In 2023, the US Department of Defense listed rare earth magnets as a critical supply chain vulnerability. Then came the Trump-backed push to ramp up domestic mining. MP Materials, the largest US producer, increased output. But here is the data point that breaks the narrative: MP Materials ships nearly all its concentrate to China for processing. According to their 2023 annual report, 100% of their rare earth concentrate was sent to China for separation. That is not a weaponized supply chain. That is a tollbooth operated by Beijing.
We can model this as a financial supply chain. The mining companies are the 'oracles' reporting supply. The processing plants in China are the 'liquidity pools' that actually convert raw ore into usable materials. The US defense industry is the 'user' trying to swap ore for magnets. But there is no direct route—every swap goes through the China pool. The US government subsidized the oracle, but the pool has zero US oversight.
Let me apply the same forensic framework I used to trace the 70,000 ETH from FTX's hot wallets to Alameda in November 2022. Instead of Ethereum addresses, track shipping manifests and customs data. Use publicly available data from the US International Trade Commission and satellite imagery of port activity.
Step 1: Trace the flow. From Mountain Pass mine in California, ore is trucked to the port of Long Beach. There, it is loaded onto container ships destined for Shanghai or Ningbo. Customs data shows the commodity code for rare earth ores (HS 253090) with US origin and China as destination. In 2023, the US exported over $200 million worth of rare earth ores to China. The US imported $400 million worth of processed rare earths from China. Net importer, but with a twist—the ore we exported is the same ore that returns as processed material. That is the classic 'round-tripping' pattern seen in wash trading. Correlation is a map, but causation is the terrain.
Step 2: Identify the latency. Between mining and separation, there is a 6-9 month lag. That means any disruption in China's processing capacity (due to environmental crackdowns, export controls, or geopolitical tensions) would not show up in US stockpiles for 6-9 months. This is analogous to the 'settlement risk' in DeFi lending pools—you think you have collateral, but it is locked in a foreign contract with a 6-month unstaking period.
Step 3: Stress-test the system. If China imposed a ban on processed rare earth exports, what happens? The US has about 6 months of strategic stockpiles for military applications, according to a 2023 GAO report. Civilian industries (EVs, wind turbines) would face immediate shortages. The mining expansion only matters if the processing capacity is also domestic. Currently, there is one rare earth processing facility under construction in the US (MP Materials in Texas), but it is not expected to be operational until 2026. Even then, its capacity covers only 10% of US demand. This is like a DeFi protocol announcing a new L2 scaling solution that won't launch for two years and will only handle 10% of current volume. Not a solution.
From my 2020 DeFi yield analysis, I learned that inflation of token supply without real yield is a ponzi. Here, the US is inflating ore supply without real processing yield. The 'yield' of supply chain security is illusory.
The contrarian angle: Perhaps the US policy is not as naive as it seems. Sending ore to China for processing takes advantage of China's cheap, efficient separation technology. It lowers costs for the US industry in the short term. The strategy could be to build up mining capacity now, then later coerce or incentivize processing plants to be built domestically by threatening export tariffs on the ore. But that assumes the US can control the ore. Data shows the ore is already sold under long-term contracts to Chinese processors. There is little flexibility. The US does not have the 'governance token' to change the rules.
Moreover, correlation between mining output and national security is not causation. Increased mining does not automatically improve security if the entire downstream chain is foreign. It could even worsen security by making the US more dependent on the ore-to-metal conversion pipeline. This is the same fallacy as judging a DeFi protocol's health by its total value locked without looking at the smart contract risk of the bridges. Supply chain is the smart contract; policy is the frontend.
Takeaway: Next week, watch for two signals. First, any announcement from the US DoD regarding funding for rare earth separation as a 'national security priority.' Second, the quarterly export data from MP Materials. If the proportion of ore sent to China remains above 80%, the policy is failing. If it drops, the supply chain is finally being audited. The market is not pricing this risk. When it does, the premium on domestic processing will spike. Follow the ore, not the press release. A mineral in the ground is potential; a processing plant is kinetic.