Onshoring vs. Nearshoring: Where Advanced Semiconductor Manufacturing Actually Creates More Value
The semiconductor industry is in the middle of a geographic restructuring unlike anything it has experienced since the original offshoring wave of the 1980s and 1990s. Government subsidies, export controls, and supply chain resilience imperatives are driving billions of dollars of investment decisions about where to build the world's next generation of chip factories. The central debate — whether to onshore production back to the US and Europe, or to nearshore to Mexico, Vietnam, Malaysia, and Poland — is increasingly framed as a political choice. It is actually an economic and operational choice with measurable trade-offs that vary significantly by chip type, production stage, and time horizon. This analysis disaggregates the debate.
Defining the Terms: What Onshoring and Nearshoring Actually Mean in Semiconductors
Semiconductor manufacturing is not a single activity — it is a value chain spanning wafer fabrication (the highest-skill, capital-intensive stage), advanced packaging (increasingly critical for chiplet architecture), and assembly, testing, and packaging (ATP, the most labour-intensive stage). Onshoring and nearshoring arguments apply very differently across these three stages. The debate about building TSMC fabs in Arizona and Intel foundries in Germany is fundamentally about wafer fabrication. The nearshoring conversation in Malaysia, Vietnam, and Mexico is primarily about ATP — a different business with different economics, skill requirements, and strategic rationales.
This distinction matters because conflating the two leads to analytical errors. A US policymaker celebrating Intel's Ohio fab as supply chain resilience is discussing a different risk than a technology company celebrating its decision to move ATP operations from China to Vietnam. Both decisions reflect legitimate strategic imperatives, but they address different vulnerabilities and create different economic outcomes for the host country.
The Onshoring Economics: What the CHIPS Act Numbers Actually Show
The US CHIPS and Science Act committed USD 52.7 billion in semiconductor incentives. Intel has received USD 8.5 billion in direct grants and USD 11 billion in loans; TSMC Arizona has received USD 6.6 billion; Samsung Austin received USD 6.4 billion. The stated rationale is supply chain resilience and national security — reducing dependence on Taiwan for leading-edge logic chips. The economic calculus is more complex. TSMC Arizona's construction cost was reported at approximately three times the equivalent fab cost in Taiwan, driven by labour costs, construction productivity differentials, and supply chain ecosystem gaps. TSMC's management publicly acknowledged that Arizona wafer costs will be "a little bit more expensive" than Taiwan — a diplomatic understatement for what industry analysts estimate as a 30%–50% manufacturing cost premium at mature production volumes.
The strategic value of onshoring leading-edge logic is not zero — a genuine supply disruption in Taiwan would have catastrophic consequences for global technology production, and the option value of domestic fab capacity has real strategic worth. But the commercial economics require sustained government subsidy to be viable at current labour and construction cost structures. The US and EU are essentially choosing to pay a strategic premium for supply chain insurance — a legitimate policy choice, but one that should be evaluated as insurance spending rather than industrial competitiveness investment.
The Nearshoring Economics: Malaysia, Vietnam, Mexico Compared
Nearshoring ATP operations to Malaysia, Vietnam, and Mexico reflects a fundamentally different economic calculation — one where the labour cost differential is large enough to justify relocation even without government subsidy, and where geographic proximity provides supply chain benefits that pure offshoring to Asia cannot deliver for US and European customers respectively.
Malaysia is the most established nearshoring ATP destination — housing operations from Intel, Infineon, NXP, Renesas, and STMicroelectronics that collectively make it the world's fifth-largest semiconductor exporter by value. Malaysia's advantage is its 50-year semiconductor manufacturing talent base: the Penang corridor has a workforce with generational ATP expertise that Vietnam and Mexico are still building. The risk is concentration — Malaysia's political stability and its geographic exposure to the South China Sea shipping corridor create supply chain risks that have become more salient since 2022.
Vietnam is the fastest-growing ATP nearshoring destination — Intel's Ho Chi Minh City facility processes over 50% of its global ATP volume, and Samsung's Vietnamese operations produce over 50% of its global smartphone output. Vietnam's competitive advantage is labour cost (30%–40% below Malaysia), improving infrastructure, and a deliberate government policy of attracting foreign semiconductor investment. The nearshoring case for US customers is less compelling than for Asian OEMs — Vietnam's geographic position does not reduce shipping time to US markets meaningfully versus established Asian production hubs.
Mexico is the nearshoring case most specific to US-oriented supply chains. A chip packaged in Monterrey or Guadalajara can be truck-delivered to US customers within 24–72 hours — supply chain proximity that Malaysia and Vietnam cannot match at any cost. Mexico's semiconductor ATP sector is nascent, but the combination of USMCA duty advantages, geographic proximity logistics, and an improving engineering talent pool (graduating approximately 130,000 engineering students annually) positions it as the most compelling medium-term nearshoring destination for US-market-oriented ATP operations. The constraint is infrastructure: Mexico's industrial electricity reliability, clean room construction capacity, and DI water supply infrastructure require significant investment before large-scale ATP operations are commercially viable outside existing industrial park clusters in Monterrey and Guadalajara.
Where More Value Is Actually Created: A Framework
The honest answer to "onshoring versus nearshoring" is that the two strategies are not in competition — they address different parts of the supply chain, different risk profiles, and different time horizons. For leading-edge logic fabrication (sub-5nm), onshoring to the US and Germany creates the most strategic value because the Taiwan concentration risk is a genuine systemic vulnerability — and because no nearshoring market has the fab infrastructure, water supply, or engineering talent to host leading-edge logic production in the near term. For mature-node logic and memory, nearshoring to Malaysia, Vietnam, and Eastern Europe creates more commercial value than onshoring — the labour cost differential is real, the talent base is established, and the strategic resilience benefit of onshoring mature-node production to the US or Germany is marginal relative to cost.
For ATP — the fastest-growing strategic battleground as chiplet architectures increase packaging complexity — nearshoring to Mexico (for US market) and Poland or Czechia (for European market) creates the most value by combining the labour cost advantage of emerging markets with the supply chain proximity that advanced customers increasingly require. The companies that allocate their restructuring capital correctly across these three distinct strategic bets — onshoring leading-edge fab, nearshoring mature node and ATP selectively — will generate the best risk-adjusted returns from the semiconductor geographic restructuring of the late 2020s.