Which market will dominate the semiconductor industry in the next decade?

“While China is approaching self-sufficiency, America’s ambitions are restrained by structural limits and Europe’s regulatory clarity is undermined by strategic ambiguity. Bruno Sergi and Mark Esposito argue that the 2030s will not be won using export controls but by whoever trains the most engineers, builds the most resilient supply chains and runs the most productive university–industry partnerships. Western technology policy has a denominator problem. According to the Semiconductor Industry Association , the semiconductor market reached $630.5 billion in 2024 and is on track to hit $1 trillion by 2030. The share of that prize the West captures will be decided by what gets built and trained this decade, not by what gets blocked at the border. Yet the loudest instrument in Washington and Brussels is the export control: a denial of access to leading-edge lithography, accelerators and design software. But while denial slows adversaries on the margin it does not produce a single additional fabrication plant (fab), engineer or research programme. America’s manufacturing ambition has structural limits America’s share of global chip manufacturing fell from approximately 37 per cent in the early 1990s to around 12 per cent today, even as American firms retain commanding dominance in chip design and sales. In 2024 the country accounted for 50.4 percent of global semiconductor revenues. The CHIPS and Science Act , signed in 2022, authorises $39 billion for manufacturing incentives and $13 billion for research and development (R&D) and workforce development. As of July 2025, the Department of Commerce had awarded $30.9 billion across 40 projects. The Government Accountability Office estimates that one facility in Arizona run by the Taiwan Semiconductor Manufacturing Company (TSMC), will push America’s leading-edge logic manufacturing share from 0 per cent in 2022 to 20 percent by 2030. Total announced private investment across the ecosystem has exceeded $640 billion across 140 projects in 30 states. Yet federal R&D as a share of GDP has slipped below 1 percent, and the fabs being subsidised today won’t come fully online until 2033, by which point the technology nodes they were designed for will be a generation old. Capital without a research base and a workforce buys catch-up, not leadership. China is patient, state-led, and increasingly self-sufficient China’s approach is co-ordinated, patient and state-directed. Through its National Integrated Circuit Industry Investment Fund (totalling over $138 billion) and provincial incentives and preferential procurement policies the government in Beijing has built a domestic capacity that is advancing rapidly under export pressure rather than retreating from it. The government’s 15th Five-Year Plan (2026–2030) calls for “decisive breakthroughs” across the semiconductor supply chain and aims for 80 per cent self-sufficiency and stable production of 14-nanometer chips, a common sort, using domestically built equipment. China’s production capacity share is projected by SEMI, an industry body, to grow from 25 per cent in 2024 to 42 per cent by 2028. In AI chips, Huawei’s Ascend roadmap and SMIC’s DUV multi-patterning techniques have enabled 7-nanometer-class production without access to extreme ultraviolet (EUV) lithography. Goldman Sachs estimates that China’s domestic semiconductor self-sufficiency by value will rise from 14 per cent in 2024 to around 37 per cent by 2030 , while self-sufficiency in AI graphics processing units is projected to reach 76 per cent. These efforts are structurally resilient against export restrictions. But with China facing a shortage of more than 300,000 skilled workers, the challenge is not speed of ambition but depth of engineering talent. The European Union shows regulatory clarity, but strategic ambiguity The European Union has translated policy momentum into real investment. But it risks confusing governance architecture for industrial strategy. The EU Chips Act , which entered into force in September 2023, mobilised over €80 billion ($93 billion) in commitments , nearly double the €43 billion initially targeted. Total investment is expected to reach €100 billion by 2030. Flagship projects include the ESMC joint venture in Dresden, which combines TSMC, Bosch, Infineon and NXP in a €10 billion facility. Yet Europe’s global production share has remained stubbornly near 10 per cent. In September 2025 all 27 member states signed a declaration calling for a “bold and co-ordinated” follow-on, with a Chips Act 2.0 evaluation expected from the European Commission in 2026. The EU has genuine comparative advantages in EUV equipment through ASML , in automotive-grade chips through Infineon and NXP , and in research infrastructure through CEA-Leti , a French government-funded technological research organisation and Fraunhofer , a German research institute, and imec , an R&D hub in Belgium, are considerable. And the EU writes the rules others build under. But translating these into commercial scale for the 2030s, rather than another round of governance frameworks, is the test Chips Act 2.0 will be judged on. Strategic comparison between the United States, China and the European Union Table 1 summarises the principal strategic variables across the three jurisdictions as of early 2026. Source: SIA (2025); GAO (December 2025); Korea Times (February 2026); Electronics Weekly (March 2026); Science|Business (April 2026); European Commission (2025). Geopolitical volatility as a structural variable Chips are no longer commercial components. They are strategic assets. The risk now is that security framing crowds out industrial substance. An AI strategy announced at a summit only becomes a strategy when it is wired into interagency co-ordination and public-private advisory structures with real seats for industry, academia and labour, as well as hard metrics for R&D intensity, graduate output and capacity milestones. The human infrastructure imperative must be met by universities and private industry No industrial strategy is credible without investment in the people who will operate it. The gap is already wide. SEMI projects that the global semiconductor industry will need more than one million additional workers by 2030, while a joint SIA-Oxford Economics study estimates a shortfall of 67,000 technicians and engineers in America alone. Europe requires an additional 400,000 professionals to reach its Chips Act targets, and China’s worker deficit in chip engineering exceeded 300,000 even before its current expansion drive. Private companies and universities are essential connective tissue between policy aspiration and operational reality. TSMC’s Arizona operations have allocated $50 million for local workforce training, while Intel’s CHIPS Act award dedicated $65 million to semiconductor workforce development . Micron has convened a network of over twenty universities, including Harvard and MIT, under the Northeast University Semiconductor Network . Arizona State University and Ohio State University have restructured curricula around chip design, fabricating student-designed circuits at TSMC fabs and embedding industry certifications into degrees. In Europe, Eindhoven University of Technology has co-developed summer schools with TSMC and ASML, while imec, CEA-Leti and Fraunhofer anchor the Chips JU pilot lines that bridges research and industrial deployment. These are not supplementary initiatives. They are the foundational infrastructure for a technologically competitive 2030s, and they demand the same sustained commitment as fab construction itself. Taiwan shows that co-development works when companies and universities design training programmes together, share equipment and build pathways from apprenticeships to doctoral research. A two-year apprenticeship at GlobalFoundries , an American firm, illustrates this model at scale. What remains missing in America and Europe is the institutional architecture to co-ordinate these efforts across borders, match programmes to the specific node technologies and sustain them through political and budget cycles that typically operate on shorter horizons than semiconductor manufacturing itself. A strategy for the decade ahead The 2030s will not be won by whichever bloc has the longest entity list. They will be won by whichever one trains the most engineers, builds the most resilient supply chains, runs the most productive university–industry partnerships and sustains all three through election cycles and budget fights. The West has the talent base, equipment monopolies and the capital to lead. What it lacks – and what the 2020s will be remembered for if this does not change – is the patience to spend a decade building rather than a quarter announcing. This article gives the views of the author, not the position of LSE Business Review or the London School of Economics. You are agreeing with our comment policy when you leave a comment. Image credit: Shutterstock.AI The post Which market will dominate the semiconductor industry in the next decade? first appeared on LSE Business Review .

Summary generated from the RSS feed of LSE Business Review. All article rights belong to the original publisher. Click through to read the full piece on blogs.lse.ac.uk/businessreview.