November 19, 2026
A Sectoral Agreement on Semiconductors
Executive Summary
To reinforce U.S. and allied leadership in the critical semiconductor sector amid rapid technological and geopolitical change, the United States and close security partners should negotiate a sectoral trade agreement on semiconductors that advances joint economic and national security objectives.
This sectoral agreement would be a preferential framework through which members extend a defined set of economic benefits to one another in return for aligning on common trade, investment, and economic and national security rules for the chip sector. Unlike existing forums focused on information exchange, this new sectoral agreement would consist of binding commitments intended to create a durable legal and policy framework that can facilitate long-term investments in the sector’s security and resiliency.
This new sectoral agreement would consist of binding commitments intended to create a durable legal and policy framework that can facilitate long-term investments in the sector’s security and resiliency.
This paper articulates the objectives for a sectoral agreement on semiconductors and outlines the types of obligations that participating economies should negotiate. It contemplates the various political pathways that could over the long term lead to such an agreement, while also putting forth specific steps that could be taken in the interim to advance the objectives of a sectoral agreement.
Objectives of the Sectoral Agreement
The sectoral agreement on semiconductors should have three overarching objectives:
- Secure supply: Members have access to a reliable, diversified, and resilient supply—free from coercion or disruption—of a full range of chips to enable the digital economy, critical infrastructure, and defense and national security platforms.
- Chip integrity: Chips deployed in sensitive applications, including related to critical infrastructure, defense, and national security, are trusted and secure.
- Technological leadership: The United States and its allies continue to lead the development and diffusion of next-generation technologies, including AI.
Approaches to Achieve the Objectives
Negotiating economies should prioritize four key approaches that, collectively, constitute a gameplan for achieving the three objectives of the agreement.
- Build scale among trusted suppliers: The heart of the sectoral agreement should be fostering a secure and integrated market for all types of chips among participating members, enabling members to achieve the scale necessary to develop and produce competitive, innovative, and affordable chips.
- Support demand for chips that members produce: The sectoral agreement can play an important role in building demand for derisked chip supply chains by expanding market opportunities for chip companies that produce in member economies and for original equipment manufacturers that procure chips from member economies.
- Complement domestic reindustrialization with a network of trusted partners: The sectoral agreement should complement and reinforce members’ domestic reindustrialization strategies on semiconductors, providing members a mechanism to secure from trusted foreign markets those chips that are not practical or strategic to produce at home, based on each member’s assessment of its own national needs.
- Strengthen aligned defensive policies: The sectoral agreement should alleviate the increasing risk of policy fragmentation from defensive policies implemented to address discrete economic and national security risks in the semiconductor sector. It can support a strengthened and aligned posture through member commitments to establish a common toolkit of defensive policies, paired with commitments to minimize the impact of defensive policies on other agreement members.
Translating Objectives and Approaches into Agreement Commitments
Participating economies should negotiate a set of binding commitments that implement the four approaches in detail and in aggregate support the pursuit of the agreement’s overall objectives. The binding commitments of the sectoral agreement should include general obligations related to institutional alignment and preferential treatment; specific measures related to implementation of trade, investment, and economic and national security policies in the chip sector; and defined processes for implementing and enforcing commitments.
General obligations: Members must commit to establish and maintain in their domestic systems a common set of legal authorities and institutional frameworks relevant to the chip sector, including:
- border measures that can be implemented to defend against unfair trade practices, counter non-market policies and practices, and respond to economic coercion;
- government procurement rules that enable or require the consideration of security objectives and other non-price criteria in awarding government procurement contracts, including to exclude untrusted chips;
- trusted technology rules that enable the exclusion of untrusted chips from commercial markets; and
- export control, investment security, and sanction authorities that address national security risks.
Additionally, members must commit to provide preferential treatment to other members in each substantive area covered by the agreement, including:
- preferential treatment for border measures;
- preferential market access to government procurement opportunities;
- national treatment for investment market access;
- preferential access to incentive programs and notifications of new incentive programs; and
- fast-track reviews, licensing, or other expedited regulatory processes related to export controls, investment security, and sanctions (and exemptions where appropriate).
The form of preferential treatment will likely vary for different types of obligations and could include a range of options. For example, certain obligations may include binary “in or out” forms of preferential treatment, whereas others may adopt a tiered approach of preferencing domestic firms first, then firms from agreement members, then firms from outside the agreement so long as they exclude adversary countries.
Specific measures: While the general obligations create the shared legal and institutional framework, alignment on specific policy measures is critical to achieve the objectives of a sectoral agreement. Such measures might include, for example, agreeing on common external tariff schedules, establishing export controls on specific dual-use products, or coordinating restrictions on the import of chips from particular suppliers under trusted technology rules.
Three sets of specific measures are critical to meeting the agreement’s objectives and should be negotiated by members as prerequisites for participation in the agreement:
- trade defense measures to address risks associated with China’s growing global market share of chip production, including but not limited to foundational chips;
- agreement among members to procure chips for the defense and/or public sectors exclusively from other signatories to the agreement; and
- export controls on advanced chips and related equipment.
The agreement should also include procedures to incorporate additional specific measures agreed to by members on a regular basis.
Enforcement and institutional arrangements: The sectoral agreement should include a standing committee, comprised of senior government officials from the members, tasked with overseeing a state-to-state dispute settlement mechanism and coordinating the negotiation of additional specific measures, among other duties as mutually agreed.
Key Scoping Decisions
While many decisions will need to be made about the agreement’s structure, which economies are a part of the agreement and which products are covered will be key.
Participating economies: The sectoral agreement should include as members those economies that play a major role in global semiconductor value chains and that are close security partners. The core group should include the G7 economies plus Taiwan, South Korea, and the Netherlands.
Covered products: The sectoral agreement should cover the full range of semiconductors—logic (both advanced and foundational), memory, and discrete, analog, and other (DAO) chips—as well as semiconductor manufacturing equipment (SME). While the agreement’s general obligations should in principle apply across all products, the implementation of specific measures—such as a particular export control or tariff rate—should be appropriately tailored to a defined set or subset of chips or SME based on the policy objective. Wherever feasible, specific measures should be designed to also apply to chips that enter the domestic markets of agreement members embedded in other components or end products.
G7 members should endorse the establishment of a limited membership forum on semiconductors, to be hosted at and facilitated by the Organisation for Economic Co-operation and Development.
A core definitional issue for the agreement will be what products are considered “trusted and secure” for the purposes of receiving preferential treatment. As a general matter, chips produced in agreement member jurisdictions should be considered “trusted and secure” by default. In some instances, chips produced outside the club may also be considered “trusted and secure” if they meet certain technical, geographic, or other criteria tied to the members’ economic and national security objectives.
Next Steps and the U.S. G7 Presidency
The U.S. G7 presidency in 2027 offers a political window to make progress on elements of a sectoral agreement on semiconductors. G7 members should endorse the establishment of a limited membership forum on semiconductors, to be hosted at and facilitated by the Organisation for Economic Co-operation and Development, with a mandate to monitor unfair trade practices and market distortions, including the potential for imbalances in the global semiconductor trade; coordinate near-term joint responses to such practices and distortions; and create the terms of reference for the negotiation of a sectoral agreement on semiconductors. These concrete steps will address an urgent policy priority for G7 economies while also laying a solid institutional foundation for longer-term work toward a binding sectoral agreement.
The Case for a Sectoral Agreement on Semiconductors
Semiconductors are the backbone of the global economy, driving advances in AI and functioning as essential inputs for the automobile, telecommunications, industrial, and consumer technology sectors. They are critical for national security, not only for use in weapons systems but also because almost all critical infrastructure relies in one way or another on semiconductors. Both projecting power abroad and defending the homeland require a secure supply of trusted chips.
Yet despite the economic and national security importance of semiconductors, U.S. and allied leadership in semiconductor supply chains is under threat. U.S. and allied producers have important advantages in key points along the semiconductor value chain (e.g., design, electronic design automation software, semiconductor manufacturing equipment [SME]) but have gaps in other areas (e.g., fabrication, packaging, materials and inputs, final assembly). U.S. and allied producers are dominant in advanced chips and have a strong position in memory chips but are increasingly challenged in the foundational market, where non-market economies have rapidly increased production capacity in recent years. As the United States and allies seek to rebuild resiliency in chip supply chains, they must contend with the fact that these supply chains are complex, interdependent, and highly dispersed across countries. No one country can build resiliency independently.
Figure 1: Semiconductor Industry Value Added by Activity and Region 2024 (%)1
Source: Semiconductor Industry Association. Reprinted with permission.
Trends and Challenges
Recent technological, market, and geopolitical developments are fundamentally shifting the global chip landscape. Governments’ semiconductor strategies will need to contend with four key trends.
Surging Chip Demand from the AI Boom
The AI boom has led to surging demand for chips, with annual revenue for chips deployed in AI data centers projected to increase to $1.2 trillion annually, a nearly tenfold increase from 2024 through 2028, according to one industry estimate.2 Highly advanced AI accelerators, which enable frontier AI model capabilities, are often the focus of policy debates. However, an AI data center can require an estimated 45 million chips, including a wide range of foundational, analog, discrete, and memory chips.3 These chips can account for more than 80 percent of chips in an AI data center by volume.4 Global leadership in AI will require a reliable and growing supply of a broad range of chips, not just the high-profile AI accelerators.
An Allied Manufacturing Gap
Looking across the chip sector value chain holistically, the United States generates about half of all global revenue, and chips are among the largest export sectors for the United States.5 Integration with global markets has been an essential driver of U.S. leadership in the chips sector. At the same time, however, consumption of chips in the United States and allied countries is outpacing the ability of these countries to fabricate chips domestically, and the manufacturing gap presents potential economic and national security vulnerabilities.
The United States accounts for 25 percent of global chip consumption, when accounting for chips incorporated into final end goods such as autos and electronics, but fabricates only 10 percent of all chips made globally.6 Europe consumes 20 percent of chips by final end use but accounts for only 10 percent of global market production share.7 Japan, South Korea, and Taiwan have significant fabrication capabilities, but these may be either concentrated in one segment (e.g., memory chips in South Korea) or subject to unique geopolitical risk (e.g., concentration of cutting edge fabrication in Taiwan). Governments in the United States and allied countries have pursued subsidy and other programs to boost chip manufacturing, yet gaps remain across both front-end fabrication and back-end packaging.
China’s growing capabilities and capacity across the chip sector complicate the ability of the United States and allies to promote competitiveness and resiliency in their domestic chip markets. While China continues to lag the United States and allied producers at the leading edge, in large part due to U.S. and allied export controls, it is becoming a dominant producer in other segments of the chip market. Overall, China now has the largest wafer production capacity among all countries.8 It has a large share in nearly every segment other than the leading edge, including the leading shares in power, analog, and mature logic, and is growing capability in both specialty and commodity memory. Its growth in foundational chips sparked concerns in Washington and European capitals, with China’s fabrication capacity doubling since 2015 and now accounting for more than half of planned production expansion globally.9
Figure 2: Five Largest Economies by Wafer Capacity per Chip Type10
Source: Organisation for Economic Co-operation and Development. Reprinted with permission.
Each panel shows an economy’s total wafer capacity for different types of chips (based on different process technologies), measured in million (M) WSPM in 8” equivalents. The dark shade corresponds to the capacity accounted for by the in-production fabs while the light shade reflects upcoming fabs in each of the economy.
Fragile Allied Lead at the Cutting Edge
The United States and allies maintain a global leadership position at the cutting edge of the chip sector, including capabilities in design (e.g., design of AI chips accelerators and memory), SME required for fabricating cutting-edge chips (e.g., extreme ultraviolet lithography), and fabrication (e.g., Taiwan Semiconductor Manufacturing Company [TSMC]). Export controls and related policies have maintained this lead, with important cooperation between the United States, Europe, and Japan to align policies on the most critical capabilities for advanced chip making.11 However, China continues to invest heavily in catching up at the leading edge and has shown an ability to innovate despite the constraints of the allied export controls. As importantly, its strengths in other areas of the chip value chain, whether that is rare earths at the input stage or its growing capacity in foundational chips, increasingly provide it with the means to effectively retaliate, and potentially deter, further allied export controls. After years of continual expansion of U.S. and allied export controls to slow China’s chip and AI capabilities, the current era of U.S.-China “strategic stability” may hamper allied efforts to update and expand controls to keep pace with China’s technological advances.
At the same time, the national security implications of AI have gone from theoretical to real-world stakes. AI has been integrated into aspects of recent U.S. military operations in Venezuela and Iran. Frontier AI models have demonstrated a step change increase in cyber capabilities. Open-weight Chinese models are fast followers to U.S. frontier capabilities and present a serious market challenge by being good enough and far cheaper than U.S. closed-weight models. Maintaining allied leadership across the AI technology stack, including the compute chips that enable innovation at the frontier, is thus critical for ensuring allies’ national security edge.
Risk of Policy Fragmentation
Maintaining the U.S. and allied lead in the chip sector (or in some cases, regaining it) requires scale: scale for companies to make the large investments necessary to keep pace with surging demand for chips and scale of market opportunities around the globe to fund these investments. Yet fragmentation in the policy environment is on track to erode this scale, as the United States and allies pursue divergent policies that create frictions between their markets. An emphasis on reshoring manufacturing capabilities, political desires to enhance technological sovereignty, and novel and aggressive uses of export controls and other national security tools are all persistent trends. Existing multilateral institutions, such as the World Trade Organization (WTO), were not designed to manage this type of security-driven fragmentation; as a result, there is no locus for coordinating policies or managing disputes between allies beyond ad hoc diplomacy and consultations.
Deep supply chain integration within the group requires a foundation of trust among members, which in turn depends on assurances that members will act against common strategic threats.
The challenges facing the United States and its allies in the chip sector are likely to be exacerbated in the years ahead. Absent deliberate efforts to change course, the current trajectory will lead to increasingly segmented allied chip markets while Chinese competitors encroach on the allies’ technological lead on advanced chips and flood global markets with inexpensive foundational chips. This is not in the interest of semiconductor companies in the United States or allied countries, who would come under growing pressure from outside competitors while facing unclear long-term demand signals in fragmented Western markets. It is also ultimately not in the interest of chip-buying industries in the United States or allied countries, whose long-term supply of chips would become unreliable and potentially subject to coercion.
The need for allied coordination to promote shared leadership in the chip sector has never been greater, as the United States and allies seek to strengthen the resiliency of their supply chains, leverage the potential benefits of AI, and drive digital prosperity across their economies. Coordination must include integrating markets alongside a clear-eyed view on the need to build trust and resiliency in those markets. Economic scale must go hand in hand with economic and national security alignment.
Overcoming Collective Action Challenges
Collective action problems, among governments and in industry, make it difficult for any one actor to develop an effective strategy to counter the trends threatening U.S. and allied leadership. If one government imposes tariffs on underpriced Chinese chips, the export surge will be diverted to other markets, and global market pricing dynamics will still be distorted. Similarly, a single company may choose to pay a premium to source chips from a more diversified pool of supplier countries, but the company will still be exposed to the risk of broader market disruption if overly concentrated supply chains result in a supply shock. A sectoral agreement is designed to overcome these coordination challenges, creating a durable, aligned government strategy and a more predictable operating environment for the private sector.
A sectoral agreement on semiconductors would extend a defined set of economic benefits to participating members in return for their alignment on a common toolkit to defend their shared interests relative to competitors outside the agreement.12 The agreement would combine elements of traditional trade and investment agreements—such as lowering certain tariffs and facilitating cross-border investment—with deeper cooperation on economic and national security priorities of supply chain resiliency, export controls, investment security, and countering threats of foreign adversaries dominating strategic industries.13 These promote-and-protect measures should be understood as a complementary package: Deep supply chain integration within the group requires a foundation of trust among members, which in turn depends on assurances that members will act against common strategic threats.
The sectoral agreement should be designed to address the range of economic and national security risks that the United States and allies are seeking to manage. These risks will vary based on the type of chip, market structure, and geopolitical actors involved across the full value chain from inputs to final assembly. For example, at the cutting edge, risks are likely to be concentrated around technology transfer concerns and maintaining allied advantage in the overall balance of military power. In foundational chips, risks may arise more acutely in the economic competition space (e.g., unfair trade practices that undercut domestic production) but also present broader concerns related to economic coercion or resiliency of supply chains. Appropriate policy responses for these different types of risks will necessarily vary, and the agreement can establish the overall framework, process, and political will to align policy responses.
Existing Forums are Insufficient
While existing coordination efforts and international agreements touch on some similar topics as this proposed sectoral agreement, none combines the criteria-based membership, binding commitments, and holistic approach of integrating traditional trade and investment tools with economic security measures. For instance, governments have previously established coordination platforms in the sector, including Pax Silica, the G7’s semiconductor point of contact group, and the Organisation for Economic Co-operation and Development’s (OECD) Semiconductor Informal Exchange Network. Such efforts have been useful forums for exchanging information and building policymaker networks across countries. They do not, however, provide a binding legal and policy framework to influence long-term investment decisions, as the sectoral agreement would. The sectoral agreement would not displace these existing forums, as both information exchange and negotiation of binding commitments can play complementary roles in achieving shared objectives for the chip sector.
Sectoral Agreement vs. Existing Forums
The proposed sectoral agreement differs from existing forums to coordinate action on economic and security policies on semiconductors because it features:
- criteria-based membership, offering distinct economic benefits and preferential treatment to economies that agree to meet a negotiated set of economic and national security obligations in the chip sector;
- binding commitments to create a durable and predictable legal and policy framework that can facilitate long-term investments in the chips sector; and
- a holistic approach that integrates the deployment of trade and investment tools with economic and national security measures.
The proposed sectoral agreement would also differ from existing trade efforts. For instance, while the WTO’s Information Technology Agreement (ITA) covers some of the same goods as the proposed sectoral agreement, the ITA’s broad membership and structure as a WTO agreement make it an inappropriate venue for coordinating economic and national security policies. Additionally, unlike the ITA, benefits extended under this proposed agreement would be preferentially applied only to other members, avoiding free-riding concerns.
The Agreement on Trade in Critical Minerals (ATCM) proposed by the Trump administration has some parallels with this proposed sectoral agreement. However, given differences in the market structures and particular risks between the critical minerals and semiconductor industries, the specific set of policy instruments to include in the agreement will differ; for instance, price floors will likely be more effective for commodities priced on open global markets (like many critical minerals) rather than for highly differentiated goods like semiconductors. These two efforts should, however, be seen as complementary, and progress on the ATCM would certainly help build momentum for a sectoral agreement on semiconductors.
Absent deliberate efforts to change course, the current trajectory will lead to increasingly segmented allied chip markets while Chinese competitors encroach on the allies’ technological lead on advanced chips and flood global markets with inexpensive foundational chips.
Finally, this proposal differs from other efforts to align national security policies on semiconductors. Multilateral export control regimes, such as the Wassenaar Arrangement on dual-use export controls, face challenges due to membership (i.e., inclusion of Russia, which blocks consensus) and mandate (i.e., voluntary coordination of narrowly defined non-proliferation controls), which limit their utility in pushing for deeper alignment and preferential treatment among a closer group of likeminded allies. Rather than relying on the multilateral regimes, recent coordination on semiconductor and SME export controls has taken place through ad hoc, informal diplomatic negotiations, most notably among the United States, Netherlands, and Japan. Such talks have been slow and contentious; partly as a result, the United States is now considering new unilateral, extraterritorial approaches under the MATCH Act, which would compel American allies to align with U.S. controls. Creating a formal institutionalized process to negotiate controls, as the proposed sectoral agreement calls for, will not magically resolve differing opinions among allied governments on the appropriate use and restrictiveness of export controls in the semiconductor sector. But a more institutionalized process can help align expectations and identify opportunities for governments to trade off policy concessions to reach mutually agreeable, negotiated outcomes.
Objectives and Approaches
The sectoral agreement on semiconductors should be grounded in a clear vision of what the agreement seeks to achieve. From there, participating member economies should develop a set of agreed approaches to achieve these objectives.
Objectives
The sectoral agreement on semiconductors should have three overarching objectives:
- Secure supply: Members have a reliable, diversified, and resilient supply—free from coercion or disruption—of a full range of chips to enable the digital economy, critical infrastructure, and defense and national security platforms.
- Chip integrity: Chips deployed in sensitive applications, including related to critical infrastructure, defense, and national security, are trusted and secure.
- Technological leadership: The United States and its allies continue to lead the development and diffusion of next-generation technologies, including AI.
These objectives are intentionally crafted to articulate what the sectoral agreement is for, rather than what it is against. Recognizing that concerns about China’s role in the chip sector will drive many aspects of the policy conversation, these objectives nonetheless anticipate a world in which China continues to be an important player in global chip markets. They are intended to serve as a pragmatic yet still ambitious guidepost for how agreement members can advance their own economic and security interests in the context of China’s increasing capabilities in the chip sector, while also addressing the myriad other challenges that threaten the stability and resiliency of chip supply chains.
Approaches
To meet the objectives of secure supply, chip integrity, and technological leadership, members should pursue the following approaches:
Build Scale Among Trusted Suppliers
The heart of a sectoral agreement should be fostering a secure and integrated market for all types of chips among participating members. A larger, more integrated market with minimal internal barriers will allow for greater specialization, while driving competition to increase economic efficiency and innovation. It is also likely the only way to build adequate scale to create a commercially viable alternative to Chinese-produced chips in those segments of the market where China is gaining a large share.
Central to this approach will be a commitment to preferential treatment for those economies participating in the agreement, including on cross-border measures (such as tariffs), access to incentives and government procurement opportunities, and exemptions from security restrictions imposed to target geopolitical competitors. In the context of increasing barriers to trade and investment imposed in the name of economic or national security, this commitment to preferential treatment among members is critical to create a set of clearly defined economic and strategic benefits for participation in the agreement.
Support Demand for Chips that Members Produce
A priority for policymakers in the United States and allied countries is to close the chip manufacturing gap, and the last several years have seen a flurry of government interventions to support domestic semiconductor ecosystems. The United States, European Union (EU), Japan, South Korea, and Taiwan are each providing tens of billions of dollars in financial incentives for domestic semiconductor production.14 To date, government interventions have been structured to shift the geographic location of chip supply, primarily through subsidies to support the construction of new fabs in their jurisdictions.
This focus on supply, while understandable, has neglected an equally important question of demand and whether chips produced in these subsidized facilities can compete in global markets, especially when going head-to-head with Chinese-produced chips that enjoy a cost advantage and even greater subsidization.15 All else equal, chips produced in the United States or other advanced market economies will be more expensive, and original equipment manufacturers (OEMs) will be incentivized to continue buying from their traditional sources rather than shift to derisked chips supply chains. The sectoral agreement can play an important role in building demand for derisked chip supply chains by expanding market opportunities for chip companies that produce within member economies and for OEMs that procure chips from member economies.
Complement Domestic Reindustrialization With a Network of Trusted Partners
The sectoral agreement should be a complement to, not a substitute for, participating economies’ existing efforts to strengthen their own domestic chip capabilities. As the United States and other governments pursue reindustrialization strategies for semiconductors, such strategies may determine that some chips, or some stages of the chip supply chain, are not practical or strategic to produce at home. In those instances, governments should still seek to enhance the resiliency and competitiveness of their chip supply chains by working with trusted foreign partners. The sectoral agreement will allow participating members to augment their domestic reindustrialization efforts by linking into a network of trusted allies. This will enable governments to focus their reindustrialization strategies on those areas most critical to economic and national security, based on each member’s assessment of its own national needs, while still pursuing comprehensive economic security across the full chip supply chain.
Strengthen Aligned Defensive Measures
Defensive policies, such as export controls, carry a high risk of policy fragmentation and, consequently, market fragmentation. When aimed at geopolitical competitors, defensive policies that are not aligned can create an unlevel playing field and undermine shared strategic goals. When aimed at each other, these policies can fracture the operating environment and trust necessary to build allied scale. The sectoral agreement should facilitate the alignment of defensive policies by establishing a common toolkit to address shared economic and national security risks in the chip sector, while also committing to minimize the impact of defensive policies on other members.
Translating Objectives and Approaches into Agreement Commitments
A criteria-based, binding, and holistic sectoral agreement should include commitments at three levels. First, members should agree on a set of fixed general obligations that every member would need to commit to in order to accede to the agreement, which would include establishing the relevant domestic institutions and legal authorities as well as committing to provide preferential treatment to other members. Second, the operation of the agreement will depend on members going beyond general obligations to implement coordinated specific measures, such as agreeing on specific tariff rates or export controls on particular products. Some of these specific measures should be negotiated during the initial formation of the agreement, but others may be introduced and adapted over time by members. Third, the agreement should include institutional processes and committees for coordinating new policies or measures, sharing best practices and other information among members, and managing disputes related to the operation of the agreement.
General Obligations
Membership in the agreement will be conditioned on governments adopting a common legal and institutional framework of trade, investment, and economic security authorities related to the chip sector, including the authorities detailed below. Members will implement their general obligations through their own domestic policy processes, which in some instances may require changes to their legal frameworks or regulatory structures. Additionally, across each of the authorities included in the policy toolkit, members will commit to provide preferential treatment to other participating members. Note that preferential treatment may not necessarily mean the same thing as national treatment, i.e., in some instances, members may agree to a tiered approach of preferencing domestic firms first, then firms from agreement members, then firms from outside the agreement so long as they exclude adversary countries. The key point is that firms from participating countries receive an economically meaningful benefit relative to those from outside the agreement to incentivize firms to favor working within the agreement in their trade, investment, and supply chain sourcing decisions.
Firms from participating countries receive an economically meaningful benefit relative to those from outside the agreement to incentivize firms to favor working within the agreement in their trade, investment, and supply chain sourcing decisions.
General obligations that should be part of the sectoral agreement include:
Border Measures
Border measures aim to alter the flow of imports into an economy, including through imposing additional costs to disincentivize imports (e.g., tariffs) or setting limits on the overall volume of imports permitted (e.g., quotas). Border measures encompass a range of mechanisms, including simple tariffs on end products or more complex measures, such as dynamic tariffs, tariff rate quotas, and component or derivative tariffs.
The agreement should include a general obligation to establish and maintain domestic authorities to impose border measures to defend against unfair trade practices, counter non-market policies and practices, and respond to economic coercion in the chip sector. Additionally, members should agree to provide each other with exemptions or other preferential treatment when imposing new border measures in the chip sector when those measures have been coordinated as specific measures within the agreement. Members should also take the agreement into account in the implementation of their respective trade remedy authorities, such that anti-dumping or countervailing duty investigations (including those initiated by industry petitions) assessing the practices of another agreement member face a higher bar to result in tariffs. Recognizing that closing the manufacturing gap is a priority, members should also commit to exempt from border measures the critical inputs for chip manufacturing, such as SME, when importing from other members.
Border measures are already in widespread use in the United States for a range of policy objectives, including ongoing investigations that indicate the potential for tariffs or other restrictions in the chip sector. Other potential agreement members, such as the EU, are considering expanding their own use, including as part of an effort to defend key sectors from nonmarket practices. Including border measures in the sectoral agreement would encourage alignment between these emerging toolkits, in terms of the structure of the tools as well as their intended purpose, while also setting guardrails and expectations around the use of border measures toward other members.
Investment Market Access
An integrated chip market should include policies to encourage cross-border investments among agreement members, which can contribute to diversifying supply chains and building resiliency. There are already substantial cross-border investments underway in the United States and other potential members; for instance, the Taiwanese giant TSMC is expanding operations in the United States and Japan to diversify its global footprint.16
To encourage such investments, the agreement should include a general obligation to provide national treatment to foreign investors from other members. Investment market access for other participating economies should not be conditioned on foreign member firms transferring equity stakes to governments, sharing proprietary technologies and intellectual property, or other non-commercial demands that exceed requirements for similarly positioned domestic firms.
Incentives
Government incentives for the chip sector have expanded in recent years, and all major chip-producing countries have some sort of incentive program related to chip manufacturing.17 While government support can be an important driver of chip supply chain resiliency, an uncoordinated approach can lead to a suboptimal distribution of government resources, with multiple governments bidding up subsidy prices to compete for the same companies. This can result in important gaps in collective supply chain resiliency, for example, if all governments seek to support cutting-edge fabs while they continue to collectively underinvest in foundational chip production.
The sectoral agreement should include a requirement to notify other members of planned chip incentive programs in advance of their deployment, to encourage informal coordination of investment strategies and dampen the potential for distortionary incentives. Members should also commit to allow companies headquartered in other members to access their incentive programs, critically including R&D incentives, and ensure such companies are preferentially evaluated relative to other foreign applicants in scoring incentive program applications. Finally, members should also agree that their incentive programs will include appropriate guardrails that restrict companies accepting incentives from substantially increasing their production capacity in non-market economies, as is currently the case for the U.S. CHIPS Act.
Government Procurement
Government procurement tools can create demand signals for derisked supply chains and address security risks associated with the government’s use of chips from non-trusted sources. The United States is already leveraging its government procurement power, including through development of a list of entities that are prohibited from accessing the defense procurement market. Overly aggressive government procurement programs, however, can undermine the goal of allied scale, if opportunities are limited to only domestic producers. A sectoral agreement commitment on open procurement could counter broader trends toward closed government procurement markets, enabling members to scale demand through mutually open government procurement.
The agreement should include a general obligation to establish government procurement rules that ensure the government can consider security objectives and other non-price criteria in awarding government procurement contracts. Such criteria can be used by participating members both to exclude untrusted chips from procurement and to provide preferential treatment to chips from other agreement members. The result would be a tiered structure for access to government procurement contracts: For instance, certain contracts might initially be open only to domestic firms; then, if domestic options are unavailable, the contract would be open to firms from economies participating in the agreement; then, if there are no options from any member economy, the contract would be open to a broader range of foreign bidders but with restrictions in place to prevent the procurement of chips from China or other countries of concern. Procurement preferences for chips might also be used across the full range of government purchasing, such that bids for auto fleets, advanced machinery, or computers and other electronics are all viewed more favorably in bid evaluation schemes if the products include only chips produced within member economies.
Government procurement will itself not be a sufficient mechanism to achieve the agreement’s goals. Direct government purchases account for approximately 1 percent of the overall chip market, and even the combined purchasing power of all potential agreement members may not be sufficient to alter global trends.18 However, government procurement commitments send an important signal to market participants, and companies that serve both commercial and government customers may seek efficiencies by applying government procurement requirements across their supply chains regardless of end customer.
Trusted Technology Imports
Similar to government procurement, trusted technology import programs support demand for derisked chip supply chains while also directly addressing security risks associated with the import of chips from unreliable or nontrusted suppliers. Trusted technology import programs, however, potentially have broader reach among multiple end users—not just government end users—across the economy. In the United States, the Commerce Department’s Information and Communications Technologies and Services program and the Federal Communications Commission’s (FCC) Covered List program allow the U.S. government to impose economy-wide restrictions on technology imports that pose a threat to U.S. national security.
In the sectoral agreement, members should commit to establish and maintain authorities to regulate the import of technology goods and services for the purposes of national security. They should also agree to not use these authorities to place undue restrictions on the imports of chips from other members of the agreement.
Export Controls/FDI Screening/Outbound Investment/Sanctions
Export controls, foreign direct investment (FDI) screening, outbound investment controls, and sanctions are all mechanisms that can address national security risks arising from specific trade or investment transactions. The United States and many of its allies already have robust mechanisms for export controls, FDI screening, and sanctions, though important differences and gaps remain. The United States has the most advanced program to regulate outbound investments, focused on U.S. investments in chip and other advanced technology sectors in China.
Agreement members should commit to establish and maintain export control, FDI screening, outbound investment, and sanctions authorities that meet a defined set of criteria.19 As with other elements of the agreement, there should be a commitment to provide preferential treatment to other members, including that such authorities will be used narrowly toward other agreement members and only for legitimate national security concerns and through upfront exemptions or expedited processes for regulatory review, as consistent with members’ own assessments of their national security interests.
Specific Measures
The general obligations outlined above, each carried out through member economies’ own domestic processes, would constitute the core shared institutional framework of the proposed sectoral agreement. With these general obligations in place, members of the agreement could then agree to coordinate specific policy measures to advance their collective economic and security interests in the chip sector. Thus, for instance, it will be a general obligation—a condition a member economy must agree to meet to sign on to the agreement—to have the authority to enact an import ban on chips that pose a national security risk through a trusted technology import program. Participating economies may then further agree to coordinate action on a relevant specific measure, such as coordinated bans on a specific chip or category of chip from a particular producer or country. Specific measures should be appropriately tailored, informed by the type of chip, the overall market conditions for that market segment, and the types of economic or national security risk that members seek to jointly address. Wherever possible, members should coordinate these specific measures to ensure the competitiveness and integrity of an integrated market among participating members.
A key question will be striking the right balance between incentivizing members to agree and implement common specific measures while recognizing that members will want to maintain sovereign discretion in such matters.
Certain specific measures are so foundational to the agreement’s objectives that members should negotiate them up front during the initial creation of the agreement. This would deepen alignment and address outstanding sources of tension among members. Three in particular stand out.
The first is crafting a joint policy response to surging production of foundational chips in China. This might include a variety of instruments, such as procurement commitments to generate demand for specific classes of derisked foundational chips, as well as common border measures (tariffs or quotas) on particular chips. Second, governments should also build on the general commitments on procurement to include a specific measure to procure all chips for the defense and/or public sectors from other members, with exceptions or extended transition periods in instances where no viable sources are available from members. Third, specific export controls on advanced AI chips, as well as the equipment necessary to produce AI chips, should also be negotiated up front during the initial agreement creation phase.
The agreement should also have established processes for the nomination, negotiation, and review of additional specific measures. These could occur on an ad hoc basis as market or security developments require, as well as through a regular cadence of scheduled opportunities, similar to how existing export control regimes meet annually to determine additions to technology control lists. The agreement’s standing committee, noted in the following section on enforcement and institutional arrangements, should be responsible for coordinating the nomination, negotiation, and review processes for additional specific measures.
A key question will be striking the right balance between incentivizing members to agree and implement common specific measures while recognizing that members will want to maintain sovereign discretion in such matters, particularly as they relate to national security. Where members agree on specific measures, they should establish processes to assess compliance and enforcement of such measures, as well as monitoring and evaluation processes to determine whether the measures are effective and to update or remove them when warranted. Where members do not reach agreement to impose a common specific measure but one member nonetheless moves forward with its own defensive action, members should adopt a “no undercut” norm that they will not seek to capitalize on any market opportunities this may present.
Enforcement and Institutional Arrangements
The proposed sectoral agreement would integrate traditional trade and investment authorities with export control, investment screening, and other national security authorities, raising complex questions about enforcement of the agreement. Typically, U.S. trade agreements (and those of major trading partners) would include an essential security exception that permits deviation from agreed trade rules if a country determines that such deviation is necessary to guard its essential security interests. The United States has long held that invocation of essential security interests is self-judging and that the validity of the underlying security interests cannot be adjudicated by independent arbitration panels under a trade agreement’s dispute settlement procedures.20 National security matters have been held entirely separate from the operation of a traditional trade agreement. In the case of the proposed sectoral agreement, however, matters of economic and national security are core to the animating logic of the effort. A traditional essential security exception would exempt the core rather than the deviation, undermining the ability of the agreement to drive policy alignment between members.
Agreement members will rightly want to preserve discretion to implement economic and national security measures according to their own sovereign judgment. However, fully omitting any consideration of enforcement in the agreement risks leaving the agreement without real bite. Binding commitments with some level of enforceability will be important to make the agreement a durable architecture that goes beyond a talk shop.
To balance the need for enforcement with the need for national sovereignty, members should center any dispute management system on state-to-state consultations—that is, a political process for resolving disputes that will inevitably arise. Typical forms of independent arbitration or dispute settlement procedures found in traditional trade agreements are inappropriate for many of the more sensitive commitments included in the sectoral agreement. Instead, members should establish a standing committee, comprised of government officials from each member, that is tasked with managing disputes and providing a regularized forum for consultation.
Members could also institute the right for any member to revoke preferential benefits that it provides under the agreement to another member if that member is not fully meeting one of the general obligations of the agreement. For example, if a member fails to establish and effectively utilize one of the core authorities required under the agreement (e.g., FDI screening), then other members could revoke the associated preferential treatment (e.g., fast-track licensing of investments) that it may extend under its own authorities to other members. Or preferential treatment in another area of the agreement could be revoked (e.g., tariff exemptions are revoked for failure to implement FDI screening) if members agree to cross-issue linkage. Cross-issue linkage would enhance enforcement options but may be more politically complex to negotiate.
Fully omitting any consideration of enforcement in the agreement risks leaving the agreement without real bite.
For specific measures, if a member has agreed to join a particular coordinated action (such as implementing tariffs on certain chips from outside the agreement) but then fails to follow through, other members could revoke benefits tied to that coordinated action (such as exemptions from those tariffs). Under no circumstance would the agreement provide a right to challenge or overturn a specific measure that any other member takes (or chooses not to take) in its own domestic system. For example, there would be no independent arbitral process that could result in a ruling that a member had to revoke a specific measure to maintain consistency with the agreement. Instead, enforcement would be centered on the availability—or revocation—of benefits under the agreement, supported by the state-to-state political consultation process.21
The agreement’s standing committee, in addition to managing the state-to-state dispute settlement process, should also have a mandate to ensure the agreement remains a living institution that is regularly updated to incorporate new policy priorities. The committee should meet at least annually and identify areas of mutual interest for deepening cross-border cooperation, tasking specific work plans to relevant agencies within member governments. Areas of cooperation could include advancing R&D, workforce development programs, or anti-smuggling and export control evasion.
Key Scoping Decisions
In addition to determining what commitments should be included in the sectoral agreement, negotiators will need to decide two foundational scope questions: What economies should participate in the agreement, and what products should the agreement apply to? This section considers both of those questions.
Membership
The agreement should include the key jurisdictions that (a) occupy a significant role in semiconductor supply chains as either producers or consumers and (b) are U.S. security partners with economic and national security interests aligned with those of the United States.
A good starting point for the negotiations would be the G7 nations (the United States, Canada, Japan, the United Kingdom, Germany, France, and Italy, plus the EU) alongside South Korea, Taiwan, and the Netherlands. The agreement should be designed so each of these countries sees that its interests will be advanced by being inside the agreement rather than outside and therefore are incentivized to participate in the negotiations. The G7, as the existing policy coordinating mechanism of advanced industrialized democracies, is well-placed to kick-start the process of negotiating a sectoral agreement on semiconductors. Between them, G7 members are home to key research institutes, SME manufacturers, and chip design and fabrication companies. They also collectively have large enough domestic markets to achieve demand scale.
A good starting point for the negotiations would be the G7 nations alongside South Korea, Taiwan, and the Netherlands.
In addition to G7 members, there are a few other countries whose participation would be critical for the agreement’s success, given their crucial roles in global chip supply chains. Taiwan dominates production of leading-edge chips and plays a significant role in the production of foundational chips. Taiwan’s participation would not be without complications. The agreement’s objective to enhance the resiliency and diversity of chip supply chains would be advanced by reducing current dependencies on Taiwanese production, in part by working with TSMC to transfer some of its advanced technologies to other jurisdictions. Some in Taiwan may worry that such shifts will diminish the island’s strategic importance, although TSMC itself appears to have already bought into a geographic diversification strategy, recognizing the value of broadening its footprint to maintain its centrality in global supply chains (while keeping its most cutting-edge chips in Taiwan). South Korea, home to leading memory chipmakers, should also be included. South Korea will face some trade-offs in joining the agreement, as several of its chip companies are still deeply invested in China, and it may resist efforts to unwind these business relationships.
Finally, because the agreement would cover both economic and national security issues, it would require the involvement of both the EU (which has authority over trade policy and issues related to the common European market) and key member states (which have authority over national security). In addition to the European member states already part of the G7, the Netherlands should also be invited to join as a founding member of the agreement, given its authority over export controls impacting ASML.
This proposed set of countries is relatively narrow. In principle, negotiations could also be opened to other countries that play a role in global chip supply chains, such as those that specialize in assembly, test, and packaging, or Gulf countries that are emerging as major purchasers of AI chips and whose sovereign wealth funds are heavily invested in chip companies around the world. Yet there is a trade-off in expanding the membership in such a way, as negotiations among this broader set of countries, with more diverging economic and security interests, would likely be protracted. A more focused set of countries, which still account for sufficient market scale to allow for commercially viable production, is more likely to reach consensus on a sectoral agreement. After the agreement is launched, participating members may want to leave the door open for additional countries that are willing and able to meet the criteria to accede to the agreement.
It is important to underline, however, that the agreement is not designed to exclude non-member countries from global chip supply chains. Indeed, many non-members may benefit from the creation of the agreement even if they are formally outside the framework. For instance, the agreement should not seek to reshore all chip packaging to the participating member countries, which would be economically impractical, but rather to encourage a diverse and resilient packaging ecosystem spread across many countries. Countries such as Costa Rica and Mexico, which are seeking to expand their chip packaging capabilities, could stand to gain from such efforts. Agreement members could negotiate definitions of “trusted and secure” chips to encompass these parts of the value chain or could consider certification programs or other mechanisms for such countries to benefit from the ecosystem created by the agreement, even if they are not positioned to take on full obligations.
Product Scope
The general obligations of the agreement should in principle cover the full range of semiconductors—logic (both advanced and foundational), memory, and discrete, analog, and other (DAO) chips—as well as SME. The implementation of specific measures should be appropriately tailored to certain types of chips or SME based on the policy objective of the measure. For instance, specific export control measures should focus on cutting-edge AI chips as well as the SME necessary to produce them, while specific trade protection measures will generally be most important for foundational chips.
In many instances, the product scope for specific measures will need to be purposefully defined at very detailed levels. For example, if members intend to put tariffs on chips “from China,” they will need to write regulations that account for a wide range of real-world scenarios, including chips that are designed by a Chinese company, use a wafer produced in a Chinese fab, are packaged at a back-end facility in China—or any of the above. Given that both market dynamics and relevant security risks vary across the stages of the chip production cycle, product scope definitions will have a significant impact on the operation of the agreement, with real potential for unintended consequences and market disruptions if not crafted carefully.
Defining what qualifies as a “trusted and secure” chip will be central to how the agreement accords preferential treatment, and it may vary based on particular end uses and associated risks. For example, governments may require that chips going into classified missile programs are sourced only from a select list of manufacturing facilities where all workers have necessary security clearances. As a general rule, however, for most purposes, chips produced within any member economy should be considered “trusted and secure.” Other chips could also be considered “trusted and secure” for certain purposes, based on technical, geographic, or other criteria, with scalable benefits provided to these chips based on these criteria. For example, a chip produced in a jurisdiction outside the agreement but in a facility owned and operated by a company headquartered in a member economy may merit preferential treatment above a chip that has no link to a member economy.
Product scope definitions will have a significant impact on the operation of the agreement, with real potential for unintended consequences and market disruptions if not crafted carefully.
Another key product scope question is how to treat chips that enter the domestic markets of agreement members embedded in other components or end products. Indeed, the majority of chips that are imported into members’ markets are imported not as chips but embedded in other products. Thus, for instance, a tariff on Chinese chips would have little direct impact because most prospective members import few chips directly from China but instead import consumer goods, auto parts, and other goods that contain Chinese-produced chips. Similarly, governments procure relatively few chips directly but procure many more products that contain chips. The commitments in the agreement will be far more effective if they apply to chips embedded in other products. In principle, this should apply across all sectors, so chips used throughout the economy are treated on a level playing field; if only certain final goods face additional tariffs on their embedded chips, for instance, producers of such goods will be disadvantaged relative to other sectors.
As a practical matter, however, implementing these various obligations to cover embedded chips will be challenging. The existing tariff system, for instance, is designed to apply a certain tariff rate based on the product classification of the complete product crossing a border, not variable tariff rates based on every embedded component. Moreover, many of the firms selling final products further down the supply chain—who would need to comply with the obligations of the agreement—have imperfect and incomplete visibility into their full supply chains and may not be aware where their tier 3, tier 4, tier 5, and so on suppliers source their chips.
Despite these challenges, however, the agreement should seek to extend obligations to apply to embedded chips wherever practically feasible. This would align with a broader shift in U.S. policy, as in recent years various policy instruments have begun to create obligations requiring more extensive supply chain due diligence to identify where chips are sourced. For instance, under Section 5949 of the fiscal 2023 National Defense Authorization Act, the federal government will not procure any products that contain chips produced by the Chinese chip companies CXMT, YMTC, and SMIC.22 The U.S. Department of War has released a new policy calling for expansive supply chain illumination requirements in its supply chain, and it has also expanded the list of Chinese companies prohibited from selling to the Pentagon.23 The FCC recently implemented new regulations to close the “component part loophole” on its Covered List, such that products containing logic-bearing hardware components produced by an entity on the Covered List will also be blocked from the market; the agency has since imposed restrictions on two products that contain chips produced by HiSilicon, a Huawei subsidiary on the Covered List.24 The experience in implementing such measures should provide useful lessons for designing specific measures for the proposed semiconductor sectoral agreement that would capture embedded chips.
Finally, the product scope of the agreement could conceivably be extended to cover certain critical minerals, which are vital inputs for producing chips and face some similar supply chain risks, particularly related to China’s dominant role in the global market. Yet expanding the product scope in such a manner would significantly complicate the negotiation and functioning of the agreement. It would require bringing a much broader set of countries into the negotiations, as there are many countries relevant to the critical mineral supply chain that are not major chip producers or suppliers. And critical minerals also need a different set of market interventions—potentially including price floors, long-term offtake agreements, stockpiling, and pooled financing for investments—which are not considered in this proposal. For these reasons, plurilateral cooperation on critical minerals should proceed in parallel to, but distinct from, progress toward a semiconductor sectoral agreement. But it is worth underlining that the proposed agreement’s objectives will be impossible to realize without a predictable, reliable supply of critical minerals.
Conclusion: Next Steps and the U.S. G7 Presidency
This paper is intended to inform ongoing policy dialogues among allies on options to strengthen the resiliency and security of chip supply chains while advancing shared technological leadership. It presents a vision of a comprehensive and durable agreement. A benefit of negotiating a comprehensive agreement is it allows members to trade off concessions in one area for wins in another, encouraging participating economies to make tough but difficult decisions so long as, on net, each economy ultimately sees this deal as advancing its own interests. It is this sense of mutual benefit that will also support the durability of the agreement, as participating economies will be incentivized to maintain their membership in the agreement over time, amid changes in government and other temporary trends.
At the same time, the path to achieving a binding and holistic plurilateral agreement is both steep and long. There are political and technical obstacles likely to impede near-term progress. The current lack of trust among the United States and allies, as well as lack of bandwidth among trade negotiators who are juggling multiple other negotiations and tariff investigations, makes it unlikely that a sectoral agreement on semiconductors could be fully negotiated in the next two years.
There are, however, still opportunities for intermediate progress and to build momentum for collective action on secure, resilient chip supply chains. In October 2026, on the sidelines of an OECD Trade Council meeting, the United States and 14 other governments released a ministerial statement calling for new sectoral platforms to address structural excess capacity in five key strategic sectors, one of which is foundational semiconductors.25 This followed similar work in the G20, with the United States leading efforts to raise concerns related to excess capacity. As the United States prepares to assume the rotating presidency of the G7 for the 2027 calendar year, it could build on such efforts.
As former structures of the global economic order recede, the economic security-centered approach of the sectoral agreement puts forward an optimistic vision for how a new order can be built.
Specifically, the United States should prioritize securing a G7 commitment to endorse the establishment of a limited membership forum on semiconductors, to be hosted at and facilitated by the OECD, focused on trade, investment, and economic and national security issues in the semiconductor sector. The OECD has substantial analytic capacity on semiconductors that could be leveraged in support of the forum, and there is precedent of the OECD hosting similar forums successfully. The semiconductor forum could be designed in a similar fashion to the existing Global Forum on Steel Excess Capacity, which is hosted at and facilitated by the OECD but with distinct criteria for membership and a mandate set by the participating governments. The mandate given to the forum on semiconductors should be to immediately establish a mechanism to monitor unfair trade practices and market distortions, including the potential for imbalances in the global semiconductor trade, and to negotiate coordinated joint actions in response. This forum could also be given a mandate to develop terms of reference for a sectoral agreement on semiconductors as outlined in this paper, laying the institutional foundation for the longer-term goal of negotiating such an agreement.
Actions taken by potential members in the meantime may expedite or complicate pathways toward a sectoral agreement. In particular, the United States has a pending investigation related to the semiconductor sector under its Section 232 national security authorities, which may lead to the imposition of tariffs on semiconductors imported into the United States. While certain potential sectoral agreement members have already negotiated a ceiling on tariff rates that their chips could receive under any future 232 actions, the imposition of broad U.S. tariffs in the sector may nonetheless create new urgency among U.S. trading partners toward a negotiated solution to bring tariff rates down. The obligations and trade-offs considered in this paper provide a framework for what a potential negotiation could entail, should potential members choose to pursue such a path.
As they begin taking steps toward joint actions on chips, potential agreement members will need a strategy to deter and respond to likely Chinese retaliation. As a starting point, potential members should be clear that their motivation in pursuing a sectoral agreement is not to hold back China’s chip ambitions but rather to invest in the competitiveness and resiliency of their own chip sectors. Additionally, the sequencing of actions toward a sectoral agreement may matter in shaping Chinese responses. Potential agreement members should consider the relative benefits of a “boil the frog” or “reverse salami slicing” approach in which smaller steps are taken over time as a means to lessen the odds of retaliation, as opposed to a more sweeping approach that may invite greater retaliation but also takes more decisive action to address the urgent issues facing the chip sector. Ultimately, the best defense against Chinese retaliation is for allied governments to act in unity, making it costly for China to push back on multiple fronts at once.
The semiconductor sector is at a critical juncture. Its products have never been in higher demand, yet the challenges facing the sector are also increasing. Negotiating a sectoral agreement will not be an easy task, and any negotiation will have to accommodate governments’ clear preferences toward domestic solutions to challenges facing the semiconductor sector. The sectoral agreement outlined in this paper is not a reversion to old approaches of simply opening markets and hoping for the best. Instead, it is intended to harness the benefits of integration to achieve the scale necessary to compete globally, while creating tiers of preferences to allow members to securely access foreign markets when domestic solutions are insufficient. As former structures of the global economic order recede, the economic security-centered approach of the sectoral agreement puts forward an optimistic vision for how a new order can be built.
About the Authors
Emily Kilcrease is senior fellow and director of the Energy, Economics, and Security Program at the Center for a New American Security (CNAS). Her research focuses on the U.S.-China economic relationship, the alignment of national security objectives and economic policy, and the use of coercive economic statecraft. Previously, she served as the deputy assistant U.S. trade representative for investment and held positions related to economics and national security at the National Security Council and Department of Commerce.
Geoffrey Gertz is a senior fellow with the Energy, Economics, and Security Program at CNAS. His work focuses on economic tools for protecting and promoting critical technologies, digital policy and data governance, and geoeconomic competition. Previously, he served as director for international economics for the National Security Council and National Economic Council, as well as a senior advisor at the State Department’s Bureau of Cyberspace and Digital Policy.
About the Energy, Economics, and Security Program
CNAS’s Energy, Economics, and Security Program explores the changing global marketplace and implications for U.S. national security and foreign policy. In a highly interconnected global financial and trade system, leaders must increasingly leverage economic and financial assets to defend and promote U.S. national interests. The program develops practical strategies to help decision-makers understand, anticipate, and respond to these developments.
Acknowledgments
The authors are grateful to the numerous experts in Washington, Brussels, Paris, Tokyo, and elsewhere who engaged in workshops, one-on-one conversations, and spirited exchange to inform the analysis of this report. The authors thank the CNAS design and publications team for their work on this report, as well as the CNAS communications team. This report was made possible with general support to the CNAS Energy, Economics, and Security Program.
As a research and policy institution committed to the highest standards of organizational, intellectual, and personal integrity, CNAS maintains strict intellectual independence and sole editorial direction and control over its ideas, projects, publications, events, and other research activities. CNAS does not take institutional positions on policy issues, and the content of CNAS publications reflects the views of their authors alone. In keeping with its mission and values, CNAS does not engage in lobbying activity and complies fully with all applicable federal, state, and local laws. CNAS will not engage in any representational activities or advocacy on behalf of any entities or interests and, to the extent that the Center accepts funding from non-U.S. sources, its activities will be limited to bona fide scholastic, academic, and research-related activities, consistent with applicable federal law. The Center publicly acknowledges on its website annually all donors who contribute.
Read the full report
- Semiconductor Industry Association, State of the U.S. Semiconductor Industry 2025, 20, https://www.semiconductors.org/2025-state-of-the-u-s-semiconductor-industry/. Based on data from IPnext; Wolfe Research; Gartner; SEMI; BCG analysis. Reprinted with permission. ↩
- David Isaacs et al., Powering AI: The Semiconductor Ecosystem at the Foundation of Data Centers (Deloitte and Semiconductor Industry Association [SIA], June 2026), 3, https://www.semiconductors.org/wp-content/uploads/2026/06/SIA-AI-Data-Center-Report_June-2026.pdf ↩
- Powering AI, 5, 7. ↩
- Powering AI, 36. ↩
- Semiconductor Industry Association, State of the U.S. Semiconductor Industry 2026 (July 2026), 7, https://www.semiconductors.org/wp-content/uploads/2026/07/SIA_2026_State-of-the-Semiconductor-Industry.pdf. ↩
- U.S. International Trade Commission, Recent Developments in Global Semiconductor Industry (November 2023), 1, https://www.usitc.gov/publications/332/executive_briefings/ebot_recent_developments_in_global_semiconductor_industry.pdf. ↩
- “Chips Act,” European Commission, accessed August 2026, https://digital-strategy.ec.europa.eu/en/factpages/chips-act. ↩
- Jan-Peter Kleinhans and Sabhyata Jha, The Chip Landscape: Geographical Distribution of Wafer Fabrication (Organisation for Economic Co-operation and Development, December 2025), 13, https://www.oecd.org/content/dam/oecd/en/publications/reports/2025/12/the-chip-landscape_27ef5d87/02dbd028-en.pdf. ↩
- Reva Goujon et al., Thin Ice: US Pathways to Regulating China-Sourced Legacy Chips (Rhodium Group, May 13, 2024), https://rhg.com/research/thin-ice-us-pathways-to-regulating-china-sourced-legacy-chips/. ↩
- Organisation for Economic Co-operation and Development (OECD), The Chip Landscape: Geographical Distribution of Wafer Fabrication, December 2025, p18. Based on data from OECD Semiconductor Production Database, SEMI, Semico Research, TechInsights. Reprinted with permission. Note that, due to data limitations, the graph double counts fabs with multiple different process technologies and thus the graph overstates total wafer capacity. For further discussion of this point and other limitations associated with the data see OECD, The Chip Landscape, 11–12, 17. ↩
- Emily Kilcrease, Hit It With Your Best Shot: An American Doctrine of Economic Pressure (Center for a New American Security [CNAS], April 1, 2026), 38–43, https://www.cnas.org/publications/reports/hit-it-with-your-best-shot. ↩
- Some of the earlier concepts for a semiconductor sectoral agreement were developed in Emily Kilcrease and Geoffrey Gertz, “Public Comments Submitted in Response to USTR Initiation of Section 301 Investigations: Acts, Policies, and Practices of Certain Economies Relating to Structural Excess Capacity and Production in Manufacturing Sectors,” comments submitted to USTR Docket No. USTR–2026-0067, April 21, 2026, https://www.cnas.org/publications/commentary/public-comments-submitted-in-response-to-ustr-initiation-of-section-301-investigations. ↩
- For more on the logic and rationale for economic security agreements, see Geoffrey Gertz and Emily Kilcrease, “A World Safe for Prosperity,” Foreign Affairs, February 6, 2025, https://www.foreignaffairs.com/united-states/world-safe-prosperity. ↩
- Semiconductor Industry Association, State of the U.S. Semiconductor Industry 2025, (July 2025), 17, https://www.semiconductors.org/2025-state-of-the-u-s-semiconductor-industry/. ↩
- U.S. Department of Commerce, Public Report on the Use of Mature-Node Semiconductors (December 2024), 7, https://www.bis.gov/media/documents/public-report-use-mature-node-semiconductors-december-2024. ↩
- Steve Lohr, “TSMC Adds $100 Billion to Its U.S. Spending Plan,” New York Times, July 16, 2026, https://www.nytimes.com/2026/07/16/business/tsmc-chips-investment-arizona.html. ↩
- SIA, State of the U.S. Semiconductor Industry 2025, 17. ↩
- SIA, State of the U.S. Semiconductor Industry 2026, 8. ↩
- For examples of what these criteria should be, see relevant sections in Emily Kilcrease and Geoffrey Gertz, Economic Security in North America: Recommendations for the USMCA Review (CNAS, December 3, 2025), https://www.cnas.org/publications/reports/economic-security-in-north-america. ↩
- For more analysis on this point, see Emily Kilcrease and Adam Tong, Disorderly Conduct: How U.S.-China Competition Upended the International Economic Order and What the United States Can Do to Fix It (CNAS, June 2024) 9–14, https://s3.us-east-1.amazonaws.com/files.cnas.org/documents/DisorderlyConduct_EESCarnegie_2024_Final_062624.pdf. ↩
- For more specific details on how dispute settlement can be designed in the context of an economic security agreement, see Kilcrease and Gertz, Economic Security in North America. ↩
- James M. Inhofe National Defense Authorization Act for Fiscal Year 2023, Pub. L. No. 117–263, 136 Stat. 3485. ↩
- Donald J. Trump, “Securing America’s Defense Supply Chains and Ensuring Domestic Acquisition of Critical Materials,” Exec. Order No. 14415 (July 20, 2026), https://www.whitehouse.gov/presidential-actions/2026/07/securing-americas-defense-supply-chains-and-ensuring-domestic-acquisition-of-critical-materials/; U.S. Department of War, “DOW Releases List of Chinese Military Companies in Accordance With Section 1260H of the National Defense Authorization Act for Fiscal Year 2021,” press release, June 8, 2026, https://www.war.gov/News/Releases/Release/Article/4511232/dow-releases-list-of-chinese-military-companies-in-accordance-with-section-1260/. ↩
- Federal Register, “Protecting Against National Security Threats to the Communications Supply Chain Through the Equipment Authorization Program,” September 11, 2026, https://www.federalregister.gov/documents/2026/09/11/2026-18535/protecting-against-national-security-threats-to-the-communications-supply-chain-through-the; Federal Communications Commission, “FCC Blocks Two Devices Containing Foreign Surveillance Chips,” press release, October 22, 2026, https://docs.fcc.gov/public/attachments/DOC-426130A1.pdf. The researchers found no statement from Huawei in response these actions. ↩
- Trade Ministers of Argentina, Australia, Canada, the European Union, France, Germany, India, Italy, Japan, the Republic of Korea, Mexico, Poland, Türkiye, the United Kingdom, and the United States, “Joint Ministerial Statement on Addressing Structural Excess Capacity and Production,” October 2026, https://ustr.gov/sites/default/files/files/Press/Releases/2026/Joint%20Ministerial%20Statement%20Addressing%20Structural%20Excess%20Capacity%20and%20Production%20(10.2026).pdf.
↩
More from CNAS
-
Public Comments Submitted in Response to USTR Initiation of Section 301 Investigations
Executive Summary In its request for comment, the Office of the U.S. Trade Representative (USTR) solicits comments on and recommendations “regarding the acts, policies, and p...
By Emily Kilcrease & Geoffrey Gertz
-
Hit It with Your Best Shot
America needs an economic pressure doctrine. The country is using economic pressure in more novel ways and at greater scale than any other time in the postwar period....
By Emily Kilcrease
-
Economic Security in North America
Executive Summary The United States, Mexico, and Canada Agreement (USMCA) negotiated during President Donald Trump’s first term includes a novel review clause. By July 2026, t...
By Emily Kilcrease & Geoffrey Gertz
-
A World Safe for Prosperity
U.S. President Donald Trump jolted the global economy this past weekend when he announced sweeping tariffs on Canada, China, and Mexico, the United States’ three largest tradi...
By Geoffrey Gertz & Emily Kilcrease
-
Disorderly Conduct
The United States must develop a strong, pragmatic strategy for advancing its economic and security interests within the U.S.-China economic relationship, accounting for the f...
By Emily Kilcrease & Adam Tong
