How to Win Friends and Choke China’s Chip Supply
On Oct. 7, the Biden administration took a risky gamble, introducing a sweeping set of new export controls that will reach deep into China’s advanced semiconductor production, supercomputing, and artificial intelligence ecosystems. These novel and ambitious restrictions usher in a new phase of technological competition with China.
But they have been implemented unilaterally, without the support of any other country. A major South Korean firm has called the new controls “painful” and European officials have rebuked the United States for dictating export control policy. If Washington seeks to keep China as far behind as possible in critical technologies, it needs to persuade key partners in Europe and Asia to implement similar controls. The United States has bet that impairing China’s advanced technology capabilities now exceeds the benefits of retaining leverage over China that can be used in the future. This bet is predicated on the assessment that the controls are so sweeping, China cannot recover from them by developing equivalent capabilities on its own. This is undoubtedly true in the short-term, as the United States dominates the “tooling” or manufacturing necessary for advanced semiconductors, as well as the design of graphic processing unit chips critical for AI. But the long-term success of the U.S. strategy is on much shakier ground. U.S. technology chokepoints will become vulnerable over time. And if other countries do not implement similar controls, foreign firms will have a strong commercial incentive to backfill U.S. technology that can no longer be sold to the Chinese market.
To its credit, the Biden administration recognizes the need for coordination and has been engaged in a concerted effort to win over key partners. However, early assessments of how quickly the United States could obtain agreements have proven wildly optimistic. The very aspects of the new U.S. rules that make them so effective today will pose challenges in building a consensus approach that can endure over time. The difficult task for U.S. export control officials will be to persuade key producers — namely the Netherlands, Japan, South Korea, and Taiwan — to fundamentally rethink their export controls strategy. This means convincing allied governments that slowing Chinese advances in supercomputing, AI, and chips, even in purely commercial areas, is urgently necessary to prevent China’s military modernization and human rights abuses. This in turn requires overcoming fundamental divergences with European and Asian allies about how hard and fast to pursue a strategic decoupling from China in advanced technology sectors.
To forge a consensus, the United States should remain laser focused on the national security justification for the new controls and offer foreign partners a clear picture of how far it intends to extend its aggressive new approach, specifically in sectors other than chips and supercomputing. Washington should proactively work with partners to mitigate the economic effects of the new controls, providing clarity and predictability for industry participants with investment timelines that stretch over a decade. Guidance on future licensing policy, as well as exemptions from the extraterritorial aspects of the new rules for those countries that implement substantially similar controls, should feature prominently in the ongoing negotiations. The ultimate objective should not be to have all partners implement identical controls. Instead, U.S. officials should prioritize aligning controls with other major producer nations to target the most significant technology chokepoints and the areas where the risk of backfilling U.S. technologies is highest.
A Profound Impact
Washington’s Oct. 7 rules impose multiple layers and types of restrictions that are intended to have a swarming effect and foreclose any legal loopholes Beijing may seek to exploit. U.S. export controls will severely restrict the flow of nearly all chips and related technologies produced globally into the high end of China’s supercomputing ecosystem, including advanced graphic processing unit chips destined for any purpose in China. They will also prevent China from using foreign chip foundries or “fabs” to attempt to manufacture chips it can no longer buy on the commercial market. New restrictions on U.S. semiconductor manufacturing equipment will punch a hole in China’s production lines as every fab producing advanced logic or memory chips depends on U.S. tools. Moreover, by prohibiting Americans from engaging in key parts of China’s semiconductor sector, the rules will starve China of U.S. expertise necessary to replicate technologies that it can no longer buy.
The core principle of U.S. export control policy towards China has for decades been to prohibit the export of any U.S. item to China that could be used for military, intelligence, or space-related purposes. For commercial technologies, the United States historically based licensing decisions on a sliding scale, accepting that China’s indigenous technology capabilities will advance and seeking to maintain a roughly two-generation advantage in the most cutting-edge commercial technologies. In contrast, the new U.S. strategy takes a broad ecosystem-based approach. This means it seeks to inhibit any progress in supercomputing, AI, and advanced chips production writ large, due to the enabling effects — rather than direct links — these technologies will have on China’s military and surveillance capabilities.
The broader implications of the new U.S. approach are profound, as it declares the advancement of key sectors of the Chinese economy as a de facto national security risk. While the United States has been careful to justify the new strategy on national security grounds, the strategy will unquestionably have significant economic impacts on China, which will compound over time as China’s capabilities are frozen in place and lag further and further behind other global producers. It also remains unclear how far the U.S. strategy will extend beyond chips. The administration has identified artificial intelligence, quantum information systems, biotechnology and biomanufacturing, and advanced clean energy technologies as fundamental to U.S. national security. If the United States imposed restrictions across these technology ecosystems comparable to what has just been done for advanced chips production and supercomputing, it would almost certainly lead to a broad technological decoupling from China.
Bringing Others on Board with a Clear Vision
It is the potential implications of Washington’s new strategy that have precluded support from U.S. partners, rather than any dovishness on their part. Europe’s views on China have been hardening. Beijing’s complicity in the Russian invasion of Ukraine has only accelerated an ongoing trend in Europe toward viewing China as a rival, though this may be tempered by a wide range of views within Europe and the economic importance of China for certain member states. Washington’s Asian partners are also not naïve about the threats posed by China, though many still seek to manage rising geopolitical tensions without drawing the ire of their largest trading partner. Growing skepticism towards China, however, will not automatically translate into agreement with the United States on specific measures, and Washington is still far ahead of the pack when it comes to a willingness to sever its economic ties with Beijing. Key partners may be amenable to stepping up certain export controls but are unlikely to agree to an approach that deems large portions of the Chinese economy as off limits.
To build greater strategic alignment with its foreign partners, the United States should be candid about the economic implications of its new measures but steadfast in defending their urgency from a national security standpoint. Washington is not seeking to provoke China but rather respond to the country’s economic and political system, in which the line between commercial and military endeavors has been erased. Therefore, while the United States is not seeking to fully decouple from China, it does not want its own technology — nor that of its allies — to fuel China’s military modernization and weapons of mass destruction development.
The United States should offer a clear vision for which other technologies it might similarly target with an aggressive ecosystem approach and, conversely, which areas of economic engagement with China can continue unimpeded. If the Oct. 7 controls prove effective, it will be through leveraging U.S. chokepoints in critical nodes of the semiconductor supply chain. But these chokepoints are unique and not necessarily replicable in other areas. In biotechnology or quantum information systems, for example, capabilities are more diffuse, giving the United States less ability to use export controls. In these areas, a coordinated approach with partners is all the more essential, as unilateral U.S. export controls will be insufficient or potentially counterproductive. U.S. restraint in considering new export controls in these types of technologies would be prudent, both to assuage concerns of partners and because of the difficulty of imposing unilateral controls. If the European and Asian allies consider expanding export controls on advanced chips production and supercomputing, it is fair for them to push the United States to define the outer bounds of its new export control policy.
European and Asian partners will also want to consider the scope of technologies to target. Washington need not push for directly replicating U.S. controls. Rather, the goal should be for each country to focus on the technologies for which their companies are dominant producers and on which China is most reliant. In tooling, for example, the United States created new list-based controls for chip-making tools that are exclusively available from U.S. suppliers. Having Japan and the Netherlands implement identical controls in this realm will be critical over the long term to ensure that firms from these countries cannot simply backfill U.S. technologies. But it is equally important that Japan and the Netherlands control technologies available exclusively from Tokyo Electron and Advanced Semiconductor Materials Lithography.
Determining the appropriate scope of controls is a two-fold challenge, requiring agreement on both the technological thresholds for controls as well as whether to make controls static or dynamic. The U.S. approach is to freeze China at the 14 nanometer threshold for logic chips as well as comparable thresholds for memory chips. These static thresholds may well be seen as overly protectionist by U.S. partners, depending on how credible they view recent reports indicating that China’s existing capabilities already exceed them. Similarly, other countries may face political difficulties in replicating the most severe aspects of U.S. controls, such as those that prevent U.S. firms from shipping even benign items such as pencils to designated facilities. Washington should adopt a flexible approach aimed at securing the maximum amount of controls politically possible today while preserving good will to continue coordination towards more stringent controls in the future. A deal with the Dutch and Japanese governments that restricts a broader range of advanced tooling equipment while permitting the sale of pencils, for example, is still a clear win.
A further challenge will be for partners to determine how to implement substantially similar controls within their existing legal frameworks. While the United States has broad and flexible authority to impose creative new export controls, other countries’ existing legal authorities and policy practices limit them to only implementing controls agreed to through multilateral export control regimes. Countries like Japan, the Netherlands, and South Korea could potentially circumvent this by using “catch-all” authorities that allow them to control technologies with a weapons of mass destruction application. But these authorities have been used infrequently and primarily to address ad hoc issues. They have not been used to publicly list large numbers of commercial technologies that present national security concerns due to their potential for enabling military or surveillance applications. To use the catch-all authorities, partner nations would have to be willing to take an extraordinary interpretation of their existing legal framework — something they would only consider in extraordinary circumstances. Indeed, the majority of chip-producing countries were willing to do this when Russia invaded Ukraine. The major challenge now for the United States is to convince them that competition with China is equally urgent.
Mitigating Economic Impact
Mitigating the negative commercial impacts of the Oct. 7 rules, anticipating possible Chinese retaliation, and addressing the diplomatic friction caused by the use of extraterritorial measures will also help win allied support. Foreign multinationals with major manufacturing facilities in China, such as South Korean memory chip companies and Taiwan Semiconductor Manufacturing, want to know if their facilities will be commercially viable over the mid- to long-term. Under the new rules, U.S. technology cannot be exported to any fab in China that meets designated technology thresholds, including those operated by U.S. or partner country companies. SK hynix, a major South Korean memory chip firm, has noted that the rules could force it to shutter certain operations in China. It is not in Washington’s strategic interest to force these firms to exit from China without warning. Doing so could create further shocks to global chip supply and dent the commercial prospects of the exact firms that the United States should be working with to improve friendshoring. The United States has offered a short-term reprieve by granting temporary licenses for foreign multinationals, but this is a band-aid solution rather than a firm guarantee that can guide large-scale investments with a five- to 10-year time horizon. If the United States does not intend to grant licenses in perpetuity, it should coordinate with these firms to allow for an orderly and economically sustainable transition out of China. At a minimum, the United States should grant licenses for a five-year period, accompanied by a clear commitment on licensing policy after that, so that foreign multinationals can plan and, if needed, transition their operations smoothly outside of China.
European and Asian partners may also be concerned about further economic impacts stemming from Chinese retaliation. While Beijing did file a complaint against the United States at the World Trade Organization, its response to the Oct. 7 controls has otherwise been muted, consistent with China’s tendency to act with restraint when responding to U.S. measures. Smaller economies, however, have not been so lucky in the past, and China has a well-known track record of using trade coercion to respond to political disputes. To blunt any Chinese retaliation, future deals should include as many partners as possible, since a larger coalition makes retaliation more difficult and risky. For example, even if the Dutch are the key European member state to join a coalition around tooling, these controls should be supported by the European Union as a whole. The United States should also commit to providing robust political and economic support to any economy that is a target of Chinese retaliation over new export control measures.
Addressing the economic impacts on foreign multinationals is all the more important given that the same group of key allies have also raised concerns about the Inflation Reduction Act and its provisions to favor U.S. electric vehicle makers. These provisions flout U.S. trade obligations and have caused a backlash in South Korea, whose carmakers will be particularly hard hit. European and Japanese officials have also raised serious concerns. U.S. policymakers would be naïve to consider these as unrelated events. Cumulatively, they are raising doubts about Washington’s commitment to its allies and partners and cynicism toward Washington’s support for a rules-based international trading system.
Washington should also address the friction caused by the expansion of extraterritorial export controls, which partners can perceive as coercive or bullying. In the Chinese context, this type of extraterritorial rule was previously limited to Huawei, a single company. The United States is now applying extraterritorial rules across whole sectors in China. Extraterritorial rules were used expansively in the Russia export controls, but this was an extraordinary wartime measure. In contrast, the extensive use of extraterritorial rules in the Oct. 7 package signals a much greater willingness to leverage U.S. market power to impose global restrictions. What was once novel has now become commonplace.
European partners have consistently objected to extraterritorial application of U.S. authorities in other contexts. Secondary sanctions, which can be used to deter European firms from engaging in transactions with U.S.-sanctioned entities that are legal under E.U. law, became particularly controversial after the United States withdrew from the Iran nuclear deal. This, along with concerns over China’s economic coercion, helped motivate the creation of the E.U. anti-coercion instrument, which would provide the European Union with the ability to counter coercive trade practices by imposing retaliatory tariffs or other economic restrictions in response. Asian partners are equally concerned about being forced into significant new controls without their consent.
To address these concerns, the United States should exempt foreign partners from the extraterritorial aspects of the Oct. 7 rules if these partners implement similar export controls of their own. The export controls imposed on Russia in 2022 offer a template, as the United States exempted 37 countries from the sweeping set of extraterritorial export controls once these countries had committed to take equivalent steps. The United States can and should offer a similar carrot for the Oct. 7 controls while also pushing partner countries to close re-export loopholes that have become problematic in the Russian case. Extraterritorial rules become unnecessary if the United States is joined by a coalition of other countries whose own rules render them redundant. Moreover, the United States will be in a better position to monitor compliance if joined by other governments. Rather than acting alone to enforce these complex rules across the entirety of the global chip supply chain, a coalition approach allows the United States to leverage the enforcement capabilities of a range of partners, fundamentally strengthening its position.
The worst-case scenario for Washington would be if it were unable to build a consensus on more far-reaching controls and instead expanded its use of extraterritorial controls to target foreign tooling capabilities that have been exempted while negotiations are ongoing. The United States should avoid the urge to take this route. Even a mostly symbolic or partial deal with allies would be valuable, signaling Western unity and serving as the basis for future coordination. As Washington considers further restrictions across a range of economic fronts, including additional export controls and outbound investment screening, it can ill afford to alienate its natural partners in Europe and Asia.
Washington’s Oct. 7 rules are both unilateral and extraterritorial. Yet the need for a shared approach to export control policy has never been more urgent. Bringing other countries on board is vital to restricting China’s progress in advanced chips, AI, and supercomputing. If done right, this can also serve as a critical first step towards a new era of strategic trade controls that will prepare Washington and its democratic allies to better manage their strategic competition with China.
Emily Kilcrease is senior fellow and director of the Energy, Economics, and Security Program at the Center for a New American Security. She previously served in economics and security roles at the Office of the U.S. Trade Representative, the National Security Council, and the Department of Commerce.