The Invisible Infrastructure of Modern Technology
Every smartphone, laptop, car, appliance, medical device, and networked object in modern life depends on semiconductors — the chips that process information, store data, enable communication, and control physical systems. The geographic concentration of semiconductor manufacturing, the specialized nature of the equipment that makes chip fabrication possible, and the complexity of the supply chains that bring chips from raw silicon to finished electronics create a vulnerability that became dramatically visible during the 2021–2023 global chip shortage and remains a central concern of both industrial policy and technology supply chain management.
Understanding the semiconductor supply chain isn’t necessary for using technology. It is necessary for understanding why a drought in Taiwan, a political dispute over trade, or a single factory fire can create ripple effects across multiple industries — why your new car’s delivery was delayed by a shortage of $10 chips, why gaming GPU prices tripled during a period of pandemic demand, and why governments are spending hundreds of billions of dollars building domestic chip manufacturing capacity.
The Concentration Problem: TSMC and the Taiwan Situation
Taiwan Semiconductor Manufacturing Company (TSMC) manufactures approximately 90% of the world’s most advanced semiconductor chips — the cutting-edge process nodes that power the latest smartphones, AI processors, and high-performance computing systems. This concentration is the result of TSMC’s technological leadership (its manufacturing processes are ahead of all competitors), decades of investment in specialized expertise and equipment, and the economics of semiconductor fabrication that favor concentration (a chip fab costs $10–$20 billion to build and requires years to reach full production yield).
The geopolitical dimension of this concentration is explicitly discussed in US, EU, and Japanese defense and economic policy: Taiwan’s semiconductor production capacity is considered a strategic vulnerability because of the political tension between Taiwan and mainland China. A disruption to TSMC’s production — whether from conflict, natural disaster, or political crisis — would affect global semiconductor supply in ways that would ripple through every industry that depends on advanced chips.
The CHIPS Act and Global Fab Building
The US CHIPS and Science Act (2022) provided $52.7 billion in subsidies for semiconductor manufacturing and research in the United States, with the explicit goal of reducing dependence on concentrated overseas manufacturing and building domestic production capacity. TSMC, Samsung, Intel, and others have received funding and are building or expanding US fabrication facilities. Similar policies have been implemented in the EU (EU Chips Act) and Japan.
The reality check: building semiconductor fabrication capacity takes 3–5 years from groundbreaking to first production wafers, and the specialized expertise required (thousands of highly trained engineers, equipment technicians, and process chemists) doesn’t materialize as quickly as physical infrastructure. The chips made in new US and EU fabs won’t be at leading-edge process nodes when they first open — they’ll be at process nodes that are 1–3 generations behind the current state of the art. Closing the technology gap requires sustained investment over decades, not just construction of new facilities.
The Equipment Layer: ASML and the Bottleneck Above the Fabs
Even if every country with ambitions to manufacture chips could build the factories, all of them would need equipment from a handful of specialized manufacturers. ASML, a Dutch company, is the sole manufacturer of the Extreme Ultraviolet (EUV) lithography machines required to manufacture the most advanced semiconductor process nodes. Each ASML EUV machine takes a year to build, costs $380 million, contains more than 100,000 components, and requires the company’s own engineers to install and calibrate. ASML makes approximately 60 of these per year.
The US has used export controls to prevent ASML from selling EUV machines to China — a measure designed to prevent China’s semiconductor industry from reaching leading-edge capability. The equipment layer of the semiconductor supply chain is as geographically concentrated and as difficult to replicate quickly as the fabrication layer itself, and it represents a different category of supply chain risk.
What This Means for Technology Consumers
For everyday technology consumers, the semiconductor supply chain affects product availability, pricing, and the pace of technology advancement. The 2021–2023 chip shortage demonstrated how dramatically these effects can manifest: new vehicle delivery delays of 12+ months, GPU prices 2–3x above MSRP, and consumer electronics shortages across multiple categories. The supply chain investments being made now are intended to prevent recurrence, but the full effect of new fab construction won’t be felt until the mid-2030s.
The longer-term consumer implication is positive: competition between multiple advanced chip manufacturers — TSMC, Samsung, Intel Foundry, and potentially others — is better for technology pricing and advancement pace than effective monopoly or duopoly. The policy investments in geographic diversification of semiconductor manufacturing, while expensive and slow, create the supply chain resilience that prevents single-point disruptions from having decade-scale impacts on technology availability.