downloadGroupGroupnoun_press release_995423_000000 copyGroupnoun_Feed_96767_000000Group 19noun_pictures_1817522_000000Member company iconResource item iconStore item iconGroup 19Group 19noun_Photo_2085192_000000 Copynoun_presentation_2096081_000000Group 19Group Copy 7noun_webinar_692730_000000Path
Skip to main content
Default Banner Image

Jan-Bart Smits and David Harap

Every time a transistor switches, it generates heat. Pack enough transistors together and you hit a wall: the chip melts before it computes. This thermal ceiling is why Splunk notes that "as physical and economic limitations are reached, the pace predicted by Moore's Law is slowing."Light solves this problem. Photons carry information without generating heat. Semiconductor Engineering details how heat dissipation and bandwidth bottlenecks make optical solutions the only viable path forward.But photonics introduces a different problem. Silicon has an indirect bandgap, which means it cannot emit light efficiently enough to produce lasers. Building photonic systems requires III-V compound semiconductors like indium phosphide (InP) and gallium arsenide (GaAs). These materials come with manufacturing constraints: InP substrates remain limited to 150mm, and GaAs wafers top out at 150mm, while silicon runs at standard industrial diameters of 200mm or 300mm. You cannot build a complete photonic system on silicon alone, so heterogeneous integration becomes mandatory. The result is that chief technology officers (CTOs) now manage two incompatible material systems, doubling technical complexity and supply chain risk.How Different Regions Are RespondingUnited States Intel has shipped 8 million photonic chips with 32 million integrated lasers. But the move that matters most is NVIDIA adopting TSMC-Broadcom co-packaged optics in its 2025 GB300 chips. When the dominant AI hardware company makes an architectural choice, competitors either follow or lose relevance.EuropeEuropean companies are solving their scale problem through consolidation. The market grew from €124.6 billion (2022) to a projected €175 billion (2027). Between January and June 2025, EPIC recorded 125 transactions worldwide, with European companies leading 50 of them. ZEISS established a new strategic business unit with €200 million in annual revenue across 6 countries. The strategy is to build on existing strengths in materials science and precision manufacturing.ChinaChina is building a parallel system designed for self-sufficiency. CHIPX produces 6-inch lithium niobate wafers with 110 GHz bandwidth, built despite U.S. export controls. This aligns with national policy: Xi Jinping chaired a February 2023 Politburo session focused on "basic research for self-reliance in science and technology." Optics Valley now hosts 5,000+ high-tech companies, targeting self-sufficiency within 4 years.Asia-PacificJapan, Taiwan, and India are combining strengths rather than building everything domestically. Japan committed $25.7 billion to semiconductor development between 2022 and 2025, and TSMC opened its first overseas R D facility there. India offers up to 50% capital support for photonics fabs and contributes 20% of global chip designers.The Next DecadeMarket projections vary wildly because the category spans everything from mature LED lightbulbs to emerging quantum computing systems. Mordor Intelligence projects growth from $1.75 trillion in 2025 to $2.39 trillion by 2030, while MarketsandMarkets forecasts $1.09 trillion to $1.48 trillion. This uncertainty matters because executives must commit billions in capital to technologies with decade-long development cycles.2025-2026The near-term focus is power efficiency. Traditional pluggable optical modules create 22 decibels of signal loss, requiring 30W per port to compensate. Co-packaged optics cuts power consumption by 3.5x. Ayar Labs' TeraPHY will deliver 8 Tb/s using UCIe standard packaging. In automotive, entry-level LiDAR drops to $200.2027-2032Quantum photonics moves from laboratory to commercial deployment. The market grows from $850 million in 2025 to $3.78 billion by 2030, with PsiQuantum partnering with GlobalFoundries to develop million-qubit systems by 2027. Unlike superconducting qubits requiring near-absolute-zero cooling, photonic qubits function at room temperature.2032-2035+The quantum market reaches $17.4 billion by 2035. Architectures combining analog, digital, quantum, photonic, and neuromorphic computing will require new transducer technologies, which means CTOs can no longer specialize in a single computing paradigm.Energy demand accelerates all of this. Data center electricity consumption will reach 945 TWh by 2030, and photonics can reduce that by over 50% by 2035.What This Means for LeadershipEach executive role faces a distinct version of the same problem: making decisions now about technologies that won't mature for years.Chief Executive OfficersCEOs face timing decisions with no clear answer. Adopt co-packaged optics in 2025-2026 and risk immature technology. Wait until 2028 and watch competitors capture market share. Japan's $25.7 billion commitment means smaller firms now compete against sovereign capital.Chief Technology OfficersCTOs must hold technical depth across incompatible domains. Silicon photonics, III-V materials, and thin-film lithium niobate each require different knowledge bases and supply chains. Most engineers specialize in one; photonics CTOs need working knowledge of all three while balancing 15-year development cycles against 2-year product roadmaps.Chief Financial OfficersCFOs must model returns on infrastructure that doesn't exist yet. The 50% power reduction from photonics changes total cost of ownership calculations, but boards need convincing before savings materialize.Corporate Boards Boards face a knowledge gap that affects governance quality. Most members don't understand why quantum-neuromorphic-photonic convergence matters at the business level. Leadership transitions signal consolidation is underway: IPG Photonics replaced its CEO in June 2024, Lumentum in February 2025.The Leadership ProblemFinding people who can run photonics companies is difficult because the field barely existed a decade ago. The technical knowledge lives in research labs. The business experience lives in traditional semiconductors. The people who combine both are rare.The broader market reflects this scarcity: over 330 R D vacancies appeared in the first half of 2025 alone. When technical roles are that hard to fill, executive searches require a global reach that most firms lack. In our searches, we regularly build single leadership teams by recruiting across China, Romania, Russia, the U.S., Germany, France, the UK, and India.The companies that figure out leadership first will have an advantage that compounds over years.About the AuthorsJan-Bart Smits is a Managing Partner at Stanton Chase Amsterdam. He serves as Global Subsector Leader for the Semiconductor industry and holds an M.Sc. in Astrophysics from Leiden University. David Harap is a Managing Director at Stanton Chase Austin with over 25 years of executive search experience. A Cornell University graduate and Father Kelly Scholar, he lectures at the University of Texas at Austin.
Read More
Political leaders worldwide are investing hundreds of billions to reduce semiconductor dependencies and secure their positions in this nearly $630 billion market, according to the World Semiconductor Trade Statistics (WSTS). Yet the extreme specialization and geographic concentration of the semiconductor supply chain makes complete self-sufficiency economically impractical and strategically questionable.After decades building an intricate global production network optimized for cost and innovation, the industry now faces pressure to splinter into regional blocks. But this raises important questions: Can any nation truly achieve chip self-sufficiency? And would disconnecting from the global ecosystem ultimately hurt competitiveness more than help security?The Independence IllusionThe global semiconductor industry has carved itself into specialized kingdoms. The United States dominates chip design and certain equipment categories, representing about 50% of global revenue. Taiwan controls roughly 67% of global foundry capacity through TSMC—so much so that semiconductors represent one-sixth of Taiwan's total GDP. Europe's strength lies in ASML's advanced EUV lithography technologies, the machines everyone needs but only one company currently makes.China plays an interesting double role too: it's both the largest semiconductor consumer at 50% of the global market and an important producer, holding 31% of total global foundry capacity in 2023.So far, every "independence" initiative has deepened interdependence. The US needs the Netherlands for lithography equipment. Europe needs Asia for high-end chip production. China develops much of its own equipment but remains dependent in key areas. The House of Cards ProblemFor decades, the semiconductor industry perfected making incredibly complex products cheaper every year through extreme specialization. Each company focused on one slice of the supply chain and became world-class at it. But nobody talked about what we built: a house of cards. The entire global economy now depends on a supply chain so specialized that losing even one supplier can shut down entire industries.The COVID pandemic exposed what industry insiders had warned about for years: the chip supply chain works brilliantly until it doesn't. When it fails, it fails spectacularly. The automotive industry alone lost $210 billion in 2021, and some manufacturers still haven't fully recovered.This 2021 chip shortage wasn't just a pandemic problem. Currently, rising geopolitical tensions are changing a supply crisis into a strategic nightmare, forcing countries to rethink their entire approach to semiconductors and their production.The Barriers to IndependenceThe semiconductor industry faces serious barriers that make true independence incredibly difficult for any single nation.First, the supply chain depends on chokepoints controlled by just a few companies in specific regions. Electronic Design Automation tools—essential software for designing any chip—come mostly from three US companies: Synopsys (~31%), Cadence Design Systems (~30%), and Siemens EDA (~13%). Without these design tools, you simply cannot create modern semiconductors. Manufacturing equipment presents an even tighter bottleneck, with ASML holding 100% control of EUV lithography machines needed for advanced chips. Second, the talent shortage makes building new capabilities nearly impossible. By 2030, semiconductor companies will need 1 million additional skilled workers. Developing semiconductor expertise takes a decade of hands-on experience, and most skilled professionals already work in established industry clusters like Taiwan, South Korea, and Silicon Valley. You can't simply relocate these engineers or train new ones quickly enough to staff multiple new regional semiconductor industries.Third and finally, resource requirements exceed what most countries can realistically provide. Building advanced semiconductor chip plants costs $20-30 billion each and they take years to construct before producing a single chip. These facilities consume up to 15 million litres of ultra-pure water daily and large facilities require up to 100 megawatt-hours of power per hour. Beyond the physical infrastructure, technical complexity has made first-time silicon success rates drop to just 14%, while 40% of semiconductor demand still comes from older process nodes, requiring completely separate supply chains for different chip generations.The Trillion-Dollar Investment RaceConcerns about supply chain security have triggered government interventions worldwide. The United States committed $52.7 billion through the CHIPS Act plus additional tax credits. While President Trump initially called for eliminating the program in March 2025, he instead signed an executive order on March 31, 2025, creating the "United States Investment Accelerator" to take over CHIPS Act implementation. TSMC also announced a new $100 billion investment to build five additional chip facilities in the US.Countries across the globe are racing to establish or strengthen their semiconductor capabilities. India has entered the semiconductor competition with its $10 billion Semiconductor Mission and secured investment from Micron Technology, which is constructing a $2.75 billion assembly and test facility. Japan has intensified its semiconductor strategy by establishing Rapidus Corporation with a government support package that is estimated to reach $11.46 billion aimed at revitalizing its domestic chip industry. Meanwhile, the European Union has established a €43 billion Chips Act through 2030, China launched its third "Big Fund" phase in May 2024 with $47.5 billion, and South Korea has developed a $450 billion K-Semiconductor strategy through 2030.These initiatives are changing the semiconductor industry on a global scale. However, complete self-sufficiency would require significant additional global investment and result in 35-65% semiconductor price increases due to suboptimal scale and inefficiencies.What Comes NextThe quest for chip self-sufficiency has become a trillion-dollar global endeavor, with countries placing enormous bets on facilities that may not pay off for years. Complete semiconductor independence remains financially prohibitive for any country, but strategic resilience is achievable.The winners will be those who build the most resilient networks and manage interdependence best. Rather than chasing impossible independence, nations should focus on strengthening their existing advantages while addressing their most vulnerable dependencies. Full independence remains a fantasy, but smart interdependence offers a realistic approach to semiconductor security.Click here to read the full white paper.About the AuthorsJan-Bart Smits is a Managing Partner at Stanton Chase Amsterdam. He began his career in executive search in 1990. At Stanton Chase, he has held several leadership roles, including Chair of the Board, Global Sector Leader for Technology, and Global Sector Leader for Professional Services. He currently serves as Stanton Chase’s Global Subsector Leader for the Semiconductor industry. He holds an M.Sc. in Astrophysics from Leiden University in the Netherlands. David Harap is a Managing Director at Stanton Chase Austin, bringing over 25 years of executive search experience to his role. He has successfully placed hundreds of senior executives and functional leaders across various industries. A Cornell University graduate and Father Kelly Scholar, Harap lectures at the University of Texas at Austin. He is a certified Ambassador for Hofstede Insights, bringing unique insights on organizational culture to his work.
Read More