Introduction
Taiwan Semiconductor Manufacturing Company (TSMC), the world’s largest contract chipmaker, is showing a tale of two realities in 2025 and U.S. fab in Arizona has become profit in record time, while its Japanese venture in Kumamoto is facing mounting losses and project delays.
This stark contrast reflects more than just financial performance—it reveals how market demand, government policy, and technology adoption shape the success or failure of global semiconductor manufacturing.
techovedas.com/tsmc-arizona-fab-hits-just-7-output-is-u-s-chip-strategy-failing
Key Highlights
TSMC reported NT$398.27 billion net profit in Q2 2025, with Arizona contributing NT$6.447 billion in investment income.
Arizona’s first fab, producing 30,000 wafers per month at 4nm, is fully booked by Apple, AMD, and others.
TSMC invested US$165 billion in U.S. fabs, with a second Arizona fab ready for tool move-in by 2026.
Kumamoto’s JASM fab lost NT$4.52 billion in H1 2025, running at just 50% utilization.
Delays could push Kumamoto’s second fab to 2029, signaling weak demand and limited adoption of advanced chips in Japan.
techovedas.com/tsmc-global-expansion-strengthening-semiconductor-leadership-in-the-u-s-and-taiwan
Arizona’s Fast Track to Profitability
TSMC’s Arizona fab has quickly become a symbol of U.S. semiconductor revival. In Q2 2025, the site generated NT$4.232 billion in net income and added NT$6.447 billion in recognized investment income to TSMC’s consolidated results.
Though modest compared to the company’s overall earnings, the milestone proves that Arizona fabs are not only strategic but also financially viable.
The success comes down to one critical factor: capacity utilization rate (UTR). With a monthly output of 30,000 4nm wafers, the fab is already at full capacity. Apple and AMD have locked in production slots, ensuring that the fab operates at nearly 100% utilization—a condition necessary for profitability in chipmaking.
Join Our WhatsApp News for real time information on semiconductors & AI
TSMC Arizona Fabs Thrive with CHIPS Act Support in 2025
Arizona’s achievement also reflects the impact of the CHIPS and Science Act, which incentivizes companies to localize semiconductor production in the U.S. For TSMC, Arizona offers both financial returns and geopolitical security, reducing risks tied to over-reliance on Taiwan.
With US$165 billion already invested, TSMC is deepening its American footprint. The second Arizona fab, finished in structure, is scheduled to begin tool move-in by Q3 2026. It is expected to support next-generation nodes, further cementing Arizona as a hub for cutting-edge chips.
Kumamoto’s Struggles With Low Utilization
In sharp contrast, TSMC’s Japanese venture—Japan Advanced Semiconductor Manufacturing (JASM) in Kumamoto—has been struggling. The fab recorded a net loss of NT$4.52 billion in the first half of 2025, with utilization stuck at around 50%.
The challenges are multi-layered:
- Mature process competition: Kumamoto focuses on specialty and mature nodes, but this segment is already oversupplied and highly competitive.
- Weak automotive and consumer demand: Japan’s traditional strength in automotive chips has not translated into steady demand, as global carmakers are sourcing advanced chips elsewhere.
- Lagging adoption of cutting-edge nodes: Unlike the U.S., where Apple and AMD push demand for advanced processes, Japan lacks local champions adopting next-generation designs.
Although the Japanese government has offered subsidies to support JASM, these measures are insufficient without customer demand to keep the fab running at profitable levels.
/techovedas.com/tsmc-opens-its-first-plant-in-japan-as-part-of-global-expansion
Delays in Kumamoto’s Second Fab
The situation is further complicated by delays in Kumamoto’s second fab. Reports from MoneyDJ and Nikkan Kogyo Shimbun suggest the launch may be postponed by up to 18 months, pushing operations into the first half of 2029.
Two reasons stand out:
- Uncertain demand outlook in the automotive and consumer electronics sectors, leaving capacity bookings low.
- Shift in global automotive semiconductor leadership, as U.S. and European companies now dominate advanced autonomous driving chip designs—an area where Japan was once strong.
This delay reflects a broader structural problem: without global customers for advanced nodes, Kumamoto may remain underutilized despite government incentives.
What This Means for TSMC’s Global Strategy
The contrasting performances of Arizona and Kumamoto highlight important lessons for TSMC and the semiconductor industry:
- Anchor customers drive profitability: Arizona succeeded because Apple and AMD committed early. Kumamoto lacks similar global anchors.
- Subsidies can’t replace demand: Japan’s financial support helped launch JASM, but without utilization, losses pile up.
- Geopolitical alignment matters: Arizona’s success is tied to U.S. policy support and strategic customer interest.
- Technology leadership fuels margins: Fabs focused on advanced nodes (4nm and below) secure better profits than those limited to mature nodes.
techovedas.com/why-tsmc-delayed-kumamoto-fab-2-to-2029-u-s-strategy-shifts-focus
Conclusion
In 2025, TSMC global expansion tells two very different stories and Arizona in profit, riding on strong U.S. customer demand, policy incentives, and high utilization. Kumamoto, however, is struggling, weighed down by weak demand, low utilization, and delayed expansion.
This split underscores a core truth: building fabs is not enough. Sustainable profitability comes from aligning technology, customers, and market demand with government support. For TSMC, Arizona may become a model for global expansion. While Kumamoto serves as a warning about the limits of subsidy-driven semiconductor projects.
For expert insights and strategies in the semiconductor space, trust Techovedas to keep you ahead of the curve!
