Taiwan is the world’s key hub for advanced semiconductor production. Every major discussion in Washington begins from this fact. Even under conservative estimates, close to 90 percent of leading-edge chips are made on the island.
This statistic has given rise to a central policy and industry question: if a political or military shock were to occur in the Taiwan Strait, how would the global IT, automotive, and battery industries withstand it?
This structural dependence means that the risk is not limited to a temporary price spike. Instead, it could reverberate across the entire value chain—from chip design and manufacturing, through packaging and testing, to final assembly and finished products.
Washington’s Three-Pronged Response
1. Expanding Domestic Production
The first response is to build up America’s own production capacity. With CHIPS Act subsidies moving forward quickly, TSMC’s Arizona plant began 4nm mass production in January 2025, and plans are already in motion for a third fab by 2030. The U.S. government has offered as much as $6.6 billion in direct support, and TSMC has laid out a roadmap to bring 2nm (including A16) production into U.S. territory.
This track is widely regarded as the most practical way to diversify risk away from Taiwan.
2. Tightening Export Controls
The second measure is the strengthening of export controls. In January 2025, the U.S. issued new rules that more tightly regulate the flow of AI chips and accelerators, extending restrictions beyond China to multiple countries. By September, Washington had revoked the Validated End-User (VEU) status of TSMC’s Nanjing plant, requiring every new shipment of equipment to undergo licensing.
The signal is clear: the U.S. intends not only to slow China’s push into advanced and mid-tier process nodes but also to apply stricter oversight to allied firms operating inside China.
3. Tariffs and Production Incentives
The third response is tariffs combined with production incentives. In the 2024 Section 301 review, the U.S. decided to raise tariffs on Chinese semiconductors to 50 percent starting in 2025. By mid-2025, White House messages openly floated even tougher options, such as imposing 100 percent tariffs on imported chips that lack U.S. production or production plans. By late September, reports described proposals for a “1:1 obligation”—for every imported chip, one must be produced domestically.
The point is not simple protectionism, but rather the creation of a domestic safety net to guard against a Taiwan-driven shock.
Parallel Moves in the Battery Sector
A similar pattern is unfolding in batteries. The Inflation Reduction Act (IRA) EV tax credit (Section 30D) began in 2024 to exclude components and critical minerals from “foreign entities of concern” (FEOC). Final rules issued by the Treasury and Department of Energy through 2024–2025 refined these requirements.
At the same time, DOE announced up to $725 million in additional investment in battery materials, components, and manufacturing capacity in 2025, driving deeper supply chain integration in North America.
Graphite is a prime example. Starting in 2026, tariffs on natural graphite imported from China will rise to 25 percent, part of a broader effort to reduce dependence on Chinese anode and cathode supply.
Automakers are responding by signing long-term contracts for synthetic graphite within North America, reducing exposure to concentrated risk.
The Common Direction
All of these changes point in the same direction: the greater the uncertainty around the Taiwan Strait, the more the U.S. will expand domestic manufacturing and thicken “non-adversarial” supply chains with allies. Export controls act as the brake on China’s technological climb, while tariffs and subsidies serve as the accelerator pulling investment into U.S. and allied soil. The contentious policy debates of 2025 are likely to materialize as actual fabs, production lines, and jobs by 2030.
Chart 1: Global Semiconductor Supply (2025–2030)
The first chart tells this story through supply dynamics.
The assumption is straightforward: if Taiwan’s production faces a 30 percent disruption, global semiconductor supply would fall by roughly 25 percent.
This figure reflects Taiwan’s outsized share in leading-edge chips. But what matters is the slope of the curves. The normal trajectory rises steadily with new fabs in the U.S., Taiwan, Korea, Japan, and Europe. By contrast, the crisis curve drops sharply in a given year, creating a supply shortage that could last for months—or even one to two years.
For companies, the gap between those two lines represents the “risk delta” that must be filled with larger inventory buffers, multi-sourcing, and design diversification to ensure continuity of supply.
Chart 1: Global Semiconductor Supply (2025–2030)
This chart shows two possible futures for the global supply of semiconductors.
Normal Scenario (solid line): If there are no major disruptions, global chip supply grows steadily year after year. New factories in the U.S., Korea, Japan, and Europe gradually increase overall capacity, so the supply line slopes gently upward.
Crisis Scenario (dashed line): If a shock occurs—such as a 30% disruption in Taiwan’s chip production—the global supply suddenly falls by about 25%. In the chart, you can see this as a sharp drop, like a step downward. After that, supply slowly recovers as new facilities are built and production ramps back up, but it takes several years to return to normal growth.
How to read it: The gap between the two lines represents the risk that companies face. That gap shows how much supply could be “missing” during a crisis. Companies must decide how to bridge this gap, for example by holding extra inventory, using multiple suppliers, or designing products that can work with chips made in different factories.
Chart 2:
Note on Pricing Assumptions
The $10 average chip price used in this article is a simplified assumption for scenario analysis, not an actual market-wide ASP. In reality, global semiconductor ASPs are much lower (around $0.60 across all devices), while advanced logic chips and processors can cost $10 or far more. The figures here are meant to illustrate the scale of impact under different scenarios, not to reflect precise current market prices.
The second chart focuses on pricing. With the average chip price set at $10 in 2025, a 20 percent tariff on Taiwanese and Korean imports would push it up to $12 immediately.
The key issue is how much of that increase downstream industries can pass on. Smartphones, servers, and automotive electronics each have different demand elasticities.
On average, finished products would face a 5–10 percent price increase, distributed unevenly across segments.
If the U.S. goes further and implements the 100 percent tariff option floated after 2025, average chip prices could rise even higher. At the same time, chips produced inside the U.S. would effectively enjoy a “tariff-free” advantage, amplifying the incentive to localize production. Ultimately, this is a question of where to make chips, and with whom.
This chart focuses on price rather than quantity.
No Tariff Scenario (solid line): Without new tariffs, the average price of a chip is about $10 in 2025. Over time, prices rise slowly and predictably, reaching about $11.20 by 2030.
20% Tariff Scenario (dashed line): If the U.S. adds a 20% tariff on chips from Taiwan and Korea, the average price jumps immediately to $12 in 2025. Prices then continue to rise on a similar slope, ending above $13 by 2030.
How to read it: The space between the two lines shows the cost burden of tariffs. For downstream industries—like smartphones, servers, and electric vehicles—this extra cost does not stop at the chip itself. It flows through the supply chain, meaning final product prices could rise by 5–10%.
What Companies Should Do
The steps market stakeholders need to take are not overly complex.
Reduce the weight of Taiwan and China in procurement portfolios, building multi-route supply networks with the U.S., Japan, Korea, and Southeast Asia.
Factor post-2026 tariff schedules and FEOC rules into product roadmaps and BOM cost models.
Standardize dual design strategies that allow the same function to be produced in different processes and regions.
Increase safety stocks of mission-critical chips for AI, servers, and automotive systems to 1.5–2 times current levels, while also strengthening North American and European packaging and testing capacity to cut lead times.
In batteries, gradually increase the share of graphite, lithium, and nickel sourced within North America, locking in synthetic graphite and recycled feedstock through long-term contracts.
Taken together, these actions allow companies to shrink the gap between the supply curve and the price curve in times of crisis.
So, what?
In 2025, the U.S. is weaving tariffs, subsidies, and export controls into a single framework aimed at absorbing the Taiwan risk through domestic and allied supply chains. The horizon to 2030 is not far away. The two charts presented here—supply scenarios and price trajectories—are not distant hypotheticals.
Each number and each curve represents a signal that will shape investment, contracting, and design priorities. What remains is for companies to interpret these signals in the language of their own business and act on them.
References
Reuters
台灣積體電路製造公司 (TSMC)
United States Trade Representative
U.S. Department of Energy (Energy.gov)
U.S. Department of the Treasury
Access This CoreBrief
The full CoreBrief is available exclusively to AI Strategica clients and subscribers.
➡Request Access
Contact us at Contact@AIStrategica.com to receive pricing, subscription options, and a sample excerpt.
Discover more from AI Strategica
Subscribe to get the latest posts sent to your email.
