A practical sourcing strategy article on where China still has structural advantages and how buyers should really decide.
The lazy version of this debate asks whether a company should “leave China.” The useful version asks something harder: what kind of product is this, how complex is the bill of materials, how sensitive is the category to quality drift, and how exposed is it to logistics or compliance shocks? China was still the world’s largest merchandise exporter in 2024, and global trade still pushed toward record levels rather than collapsing. So the real sourcing problem is rarely whether China disappeared. It is whether your category can be rebuilt somewhere else without losing speed, control, or resilience.
That distinction matters more than most headlines admit. In many categories, buyers do not actually move the chain. They move the final operation. Kearney’s 2024 reshoring work showed a clear correlation between rising US imports from alternative Asian locations and those same countries’ imports from mainland China. The Dallas Fed makes a similar point in North America: rising imports from Mexico and other emerging economies often sit alongside continued Chinese and East Asian intermediate input flows. A new country label does not automatically mean a new supplier base, a new risk profile, or a genuinely new cost structure.
The behavior of procurement teams already reflects that shift. Gartner found that 73% of companies had added or removed production locations in the previous two years, with risk management overtaking pure cost as the main driver. But that does not translate into a clean exodus from China. In the European Chamber’s 2024 China survey, only 1% said they were completely shifting current China supply chains to other markets, while 12% were building alternative supply chains outside China and keeping current China chains in place. Serious buyers are not choosing slogans. They are building options.
If a process is already heavily automated, lower wages elsewhere do not automatically create a lower landed cost. IFR reported that China had 2.027 million industrial robots operating in factories in 2024, with annual installations hitting 295,000 units that year, representing 54% of global demand. In other words, many Chinese factories are no longer competing mainly on cheap labor. They are competing on process capability, throughput, repeatability, and the ability to absorb engineering changes without blowing up yield.
China’s advantage is often cluster density, not just factory count. MERICS notes that the heavy concentration of export industries in places such as the Pearl River Delta and Yangtze River Delta created unusual production speed and flexibility, supported by sophisticated supplier networks, efficient logistics infrastructure, and a mature customs regime. For buyers, that translates into fewer handoffs, faster sampling, quicker rework, easier component substitution, and less time lost solving small operational problems that never show up in a formal quotation.
NBER’s work on industrial clusters in China helps explain why a thin alternative ecosystem often feels expensive even when nominal unit prices look attractive. The study highlights characteristics buyers routinely underestimate: vertical disintegration across stages of production, strong buyer-seller networks, trade credit, shared tools and information, and proximity to markets for intermediate goods. Together, those features reduce inventory needs and working-capital strain while making it easier to coordinate across multiple stages of production. When a buyer leaves that environment, the replacement is rarely one factory. It is an entire local operating system.
China is not automatically the best answer when freight dominates economics. For bulky, low-value goods, or for products with thin margins and high seasonality, geography can matter more than ecosystem depth. The 2024 Red Sea disruption made that obvious. OECD/ITF found that transits through the Bab-el-Mandeb Strait fell about 55%, containership transits fell 75%, and rerouting via the Cape of Good Hope added roughly 10 days to a typical Asia-Europe voyage. UNCTAD also estimated that the China Containerized Freight Index rose by about 120% from October 2023 to June 2024. In those categories, region-fit can matter more than factory density.
Tariff-sensitive sourcing decisions should be made at SKU or category level, not at country level. USTR’s 2024 Section 301 update raised tariffs sharply in selected sectors, including electric vehicles to 100% in 2024 and several other categories to materially higher rates over 2024–2026. The EU also adopted definitive countervailing duties on Chinese BEVs, applicable from 30 October 2024. That does not prove China is no longer competitive. It proves that the same sourcing country can make sense for one product and fail completely for another depending on the end market, the customs treatment, and the policy regime around that category.
This is where many buyers still fool themselves. The EU’s corporate sustainability due diligence rules entered into force in July 2024, and the EU forced-labour regulation entered into force in December 2024 and becomes applicable on 14 December 2027 across all companies and all products, including exports from the EU market. Meanwhile, DHS reported in July 2024 that CBP had reviewed more than 9,000 shipments valued at over $3.4 billion under UFLPA-related enforcement since June 2022. Moving final assembly does not solve that problem if the same risky inputs, subcontracting layers, or traceability gaps remain in place.
A useful sourcing comparison begins with the chain itself. Before comparing China with Vietnam, India, Mexico, or anywhere else, score the product on four dimensions: BOM complexity, required supplier density, logistics sensitivity, and compliance exposure. A simple finished product with stable specifications is one decision. A high-mix electronics product with tight tolerances, frequent engineering changes, and multiple sourced subcomponents is another. Buyers make bad country decisions when they compare geographies first and only later discover that the category needed a denser ecosystem than the chosen country could provide.
The first architecture is China-centric: keep China when automation, supplier density, and process maturity are genuinely difficult to replace. The second is China+1: build a second source or move selected operations while keeping critical capabilities or inputs in China. The third is nearshore or region-fit: move closer to the end market when freight, tariffs, or regional industrial policy change the economics. This is the central strategic shift the article should leave with the reader. The right answer is not “China yes” or “China no.” The right answer is matching architecture to category.
Before moving anything, calculate the costs that quotation sheets usually hide: scrap during ramp-up, longer engineering loops, extra safety stock, more expediting, more audit load, origin documentation, and the management drag of operating inside a thinner ecosystem. The Dallas Fed’s Mexico analysis is useful here because it shows that even where nearshoring is real, it often coexists with continued East Asian intermediate imports rather than a clean break from China. Buyers who ignore that usually budget for one supply chain and end up managing two.
The first practical move is not a country announcement. It is a category assessment and supplier map. That is exactly where structured China sourcing and export support should help: identify which parts of the chain are truly local, which inputs are still Chinese, where the compliance burden sits, and whether the product belongs in a China-centric, China+1, or nearshore model. Done properly, that step saves more money than it costs because it kills bad assumptions before they become contracts, tooling, or inventory.
If China remains part of the answer, a real factory audit in China becomes more important, not less. The right audit is not just a license check. It should test whether the factory actually owns the process, how much it subcontracts, what its quality controls look like, whether traceability exists at batch level, and how critical inputs are managed. That matters even more in categories exposed to due diligence, forced-labour scrutiny, or tariff risk, where paperwork alone is not a sufficient risk-control tool.
And if the decision depends on comparing clusters, sub-suppliers, or alternative manufacturing nodes in real time, procurement trip support in China can be faster and more truthful than another round of remote calls. A site visit can reveal whether the “factory” is a real production node, whether the surrounding ecosystem is dense enough for your category, and how quickly engineering or sourcing changes can actually be absorbed. When you are ready to test your own BOM, target market, and risk profile against this logic, the natural next step is to submit a sourcing brief and assess the chain before you move it.
