Diversification versus reshoring

A more useful question than where to relocate

Supply-chain leaders face a practical choice: which dependencies need to change, and which alternatives could protect customer service at an acceptable cost? A new domestic factory or a supplier closer to the market may help. Location alone does not establish that the required capacity, skills, inputs and service are available.

Our interpretation of the research is to treat diversification and reshoring as options to test against a defined exposure. Neither is a universal prescription. The decision should follow the product, process stage and customer requirement.

What the research establishes

The OECD’s 2025 resilience review models substantial economic costs from broad relocalisation and finds that it does not consistently improve resilience. These are scenario results, not an observed outcome for every company or a forecast of a particular sourcing decision. The useful challenge is to test whether a proposed change actually reduces the relevant exposure. ^[1]

The EIB and European Commission’s 2025 study, drawing on a survey of around 850 EU importers conducted in 2024, describes diversification and new sourcing countries. It also reports difficulty finding EU alternatives for essential Chinese inputs. This provides European firm-level context; it cannot establish a single worldwide relocation trend. ^[2]

Real operating changes also exist. TSMC’s 2025 annual report records that its first Arizona fab entered volume production in the fourth quarter of 2024. The same report distinguishes that milestone from later facilities and expansion plans. This is company-reported evidence of local production, rather than proof that an entire semiconductor ecosystem has moved or that a specific offshore activity has returned. ^[3]

Keep the evidence stages separate

For an individual supplier or investment, we suggest tracking a sequence: announced intention, committed funding, construction, installed equipment, customer qualification, commercial output and sustained operating performance. These are SCP’s proposed assessment stages, not a statistical model of how often projects succeed.

A supplier that has reached commercial output may still be unable to meet your specification or volume. A qualified second supplier may depend on the same upstream processor, specialist tooling, port or power infrastructure as the first. Assessing those shared dependencies is part of establishing whether the alternative is meaningfully different.

Different operations need different answers

In manufacturing, start with the critical component and production stage: assembly, fabrication and testing may have different qualification requirements. In retail and consumer goods, compare replenishment speed with stock availability, capacity and cost. In logistics and distribution, assess alternative corridors alongside handling and delivery constraints. These are proposed decision lenses, not findings that one network design is best for each sector.

The selected evidence here covers institutional modelling, European importers and one semiconductor producer. It is not a global census or a representative sample of all industries. Further regional and sector evidence is needed before drawing a more specific investment conclusion.

Five questions for the next network review

• Which customer commitments are exposed, and to what failure point?
• Is the alternative operational and qualified for our actual requirement?
• Which upstream suppliers, routes or infrastructure would still be shared?
• What would switching require in tooling, capacity, inventory, cash and time?
• Which evidence and service measures must be met before increasing the commitment?

From a headline to an operating decision

A sensible first step is a bounded review of one product family or critical dependency. Map the current exposure, compare feasible alternatives on total cost and service, and define staged commitments with named owners. That creates a decision that can be examined and challenged before resources are committed.

Supply Chain Partners’ proposed Resilience and Network Strategy service follows this approach. A discovery workshop would establish the decision boundary, initial evidence needs and whether a separately scoped assessment is worthwhile.