US President Donald Trump signed an executive order imposing a 15% tariff on imports of polysilicon and related products, effective December, following a national security investigation. Polysilicon is a critical material in both semiconductor manufacturing and solar panel production. The order also sets minimum import prices on polysilicon imports. Trump’s order cited the sharp decline in US polysilicon production — from approximately 50% of global output in 2005 to less than 2% by 2024 — as a national security risk, noting China now holds a near-monopoly on global production. China’s embassy in Washington said the measure “seriously disrupts” bilateral trade and that Beijing would act to protect its companies, accusing Washington of “abusing state power to go after Chinese businesses.” Separately, China announced tighter export controls on drones and launched a national security review of imported printers and copiers — both part of a broadening countermeasures package.
The received wisdom
The free-trade objection to this tariff is principled and deserves a fair hearing. The case runs like this: polysilicon has fallen in price dramatically because of Chinese manufacturing scale and efficiency, which has benefited American solar panel installers, electronics manufacturers, and ultimately consumers. Imposing tariffs raises costs across the entire supply chain — including for American companies building out semiconductor fabs or solar installations. The beneficiaries are concentrated: primarily Hemlock Semiconductor and Wacker Chemie, the two main US polysilicon producers. The costs are diffuse, spread across every downstream user of the material. This is a classic protectionist transfer: taxing the many to subsidise the few. Moreover, the argument continues, the US cannot realistically rebuild polysilicon manufacturing capacity in a few years; China’s cost advantage is structural. The tariff, on this reading, will raise prices without actually restoring the strategic supply.
There is also a legitimate concern about retaliation spirals. China has already responded to previous US tech restrictions with countermeasures on drones and rare-earth processing materials. Each escalation raises costs for both sides and risks fragmenting global supply chains in ways that reduce efficiency system-wide.
A different read
These are real costs. But the received wisdom of free-trade maximalism has a poor track record when applied to sectors where China’s competitive advantage derives not from comparative advantage in the conventional sense but from decades of state-directed industrial policy, subsidised inputs, and captive domestic markets that foreign competitors cannot access on reciprocal terms.
The US share of global polysilicon production falling from 50% to under 2% over two decades is not a triumph of market efficiency. It is the result of a deliberate Chinese industrial policy that used state-owned enterprises, directed bank lending, and below-market energy prices to capture the global market for a material deemed strategic. The Trump order itself notes that America “allowed foreign firms to weaken United States producers” — which, while a somewhat self-serving framing, accurately describes a policy passivity that has been bipartisan. The Obama, Trump, Biden, and Trump administrations have all, at various points, recognised the semiconductor supply chain as a national security vulnerability; the policy responses have been episodic rather than systematic.
The historical precedent worth studying is the response to Japanese semiconductor dominance in the 1980s. By the mid-1980s, Japanese manufacturers had captured roughly 80% of the global DRAM memory chip market, partly through state-directed industrial policy. The Reagan administration negotiated the 1986 US-Japan Semiconductor Trade Agreement and used a combination of tariffs and antidumping measures. The results were mixed: Japanese DRAM dominance eventually waned, but partly for reasons — the rise of South Korea’s Samsung and Hynix, subsidised by Seoul — that had nothing to do with US trade policy. The lesson is not that intervention works reliably; it is that strategic industries in which one nation achieves monopoly or near-monopoly control through state direction represent a genuine national security problem that pure market logic cannot resolve.
The polysilicon tariff sits within a broader framework that also includes the CHIPS Act investments in domestic semiconductor fabrication, export controls on advanced chip technology to China, and — as the Guardian reported last week — escalating tension over Chinese EV and clean-energy supply chains in Europe. The cumulative effect is a partial decoupling of high-technology supply chains between the US-aligned bloc and China. The economic cost of that decoupling is real. The strategic alternative — continued dependence on Chinese production for materials critical to military electronics — is, arguably, more costly still.
The caveat is execution. A 15% tariff, effective in December, is an incentive signal. It will not on its own rebuild US polysilicon manufacturing capacity, which requires capital investment, technical workforce development, and stable long-term policy commitments that survive administration changes. The Trump administration’s record on delivering that kind of patient, consistent industrial policy — as opposed to dramatic tariff announcements that subsequently get modified in trade negotiations — is, to put it charitably, uneven. The incentive is right; the staying power is unproven.
What to watch
- Whether Congress moves to complement the tariff with direct investment incentives for US polysilicon producers, parallel to the CHIPS Act model for chip fabrication
- China’s promised countermeasures: which specific US exports or sectors Beijing chooses to target will reveal its strategic prioritisation
- The December implementation timeline: whether the tariff survives any interim US-China trade negotiations, which have a pattern of pausing individual measures in exchange for broader commitments
- The reaction of US solar panel installers and consumer electronics manufacturers — the downstream buyers of polysilicon who bear the cost increase — and whether they organise effective political opposition
— J