Why Supply Chain Diversification Has Become a Strategic Priority

The China+1 Era Is No Longer Optional

For two decades, "Designed in California, assembled in China" was more than a line etched on the back of an iPhone it was the operating logic of global manufacturing. That logic is breaking down in real time. Apple now runs roughly 50% of its facilities in China, but of the 17 companies handling its electronics manufacturing, only two are Chinese-owned, and the company has pushed its India supplier base from just 14 in 2023 to more than 40 today, overtaking Vietnam's 35. This is not a marginal hedge. It is a rebuilding of production geography under real financial and political pressure, and it is happening across industries far beyond consumer electronics.

The trigger points are familiar by now: tariff volatility, the fragility exposed by the Red Sea and Strait of Hormuz shipping disruptions, and a hardening geopolitical line around Taiwan. What has changed is the willingness of blue-chip companies to absorb real cost penalties to reduce single-country exposure. Manufacturing costs in India and Vietnam run 5% to 10% higher than in China, largely because sensors, circuit boards and batteries still have to be imported from Chinese suppliers even when final assembly happens elsewhere. Apple is paying that premium anyway.

What the Data Actually Shows

Apple: Diversification With an Asterisk

Apple's shift to India is real but incomplete. Tim Cook told investors the majority of iPhones sold in the US would carry an Indian country-of-origin tag, with Vietnam handling nearly all US-bound iPads, Macs, Apple Watches and AirPods. Apple's India operation now spans five assembly facilities three run by Tata Group, two by Foxconn — supported by around 45 component suppliers. Yet 71% of the components inside an Indian-assembled iPhone still originate in China. Chinese firms have dropped from roughly 15-20% to under 10% of Apple's India-based supplier base, but the subcomponent dependency hasn't disappeared, it has just moved one tier back in the chain. That distinction matters for anyone building a diversification strategy: relocating final assembly is the easy 20%; relocating the component ecosystem underneath it is the hard 80%.

TSMC: Betting $165 Billion on Geographic Redundancy

Semiconductors show the same pattern at a much larger scale. TSMC has committed $165 billion to its Arizona campus six fabs, two advanced packaging facilities and an R&D center up from an initial $12 billion pledge in 2020. The economics are brutal: US construction costs run 4-5x what an identical facility costs in Taiwan, and AMD CEO Lisa Su has said Arizona-made chips cost 5% to 20% more than the Taiwan-made equivalent. TSMC is raising advanced-node prices 5-10% starting in 2026 specifically to offset the build-out. Companies are paying a diversification tax and treating it as the cost of doing business, not a discretionary expense.

The urgency is geopolitical as much as commercial. A January 2026 US-Taiwan agreement pairs $250 billion in direct Taiwanese investment with $250 billion in credit guarantees to expand US chip capacity, with Commerce Secretary Howard Lutnick stating the explicit goal of bringing 40% of Taiwan's semiconductor supply chain onto US soil. India's semiconductor push is moving in parallel: the India Semiconductor Mission has approved 10 projects, including two fabrication plants, drawing roughly $17.3 billion in commitments, with Tata Electronics and Powerchip building a $10.9 billion fab in Dholera and Micron running an assembly plant in Sanand.

Tesla and Samsung: Splitting Orders on Purpose

Tesla's approach to its next-generation AI5 chip is a clean example of deliberate dual-sourcing rather than reactive scrambling. Elon Musk confirmed the AI5 design will be manufactured jointly by Samsung's Taylor, Texas plant and TSMC's Arizona facility a split made explicitly to keep chip supply under Tesla's control regardless of demand scenario, rather than betting the company's FSD roadmap on one foundry. It is worth noting that even this diversification has a ceiling: TSMC still controls roughly 70% of the global foundry market versus Samsung's 7%, meaning true single-supplier risk still lingers behind the appearance of a split order book.

Where This Leaves Procurement and Strategy Teams

Three practical implications fall out of these examples. First, geographic diversification and supply-chain diversification are not the same thing Apple's India build-out proves that moving the factory doesn't automatically move the supplier base underneath it. Second, redundancy has a real, quantifiable price (5-10% on Indian/Vietnamese assembly, 5-20% on Arizona-made chips), and companies that treat diversification as a zero-cost insurance policy will misbudget it. Third, dual-sourcing works best when it's structured before a crisis, the way Tesla split its AI5 order, rather than assembled under pressure after a shock, the way many electronics firms are now scrambling to do in response to helium shortages tied to the Strait of Hormuz.

For leaders evaluating their own footprint, the honest question is not "can we add a second country" but "how many tiers deep does our single point of failure actually sit." Apple's experience says the answer is usually more than one.

Key Stats at a Glance