Semiconductor Supply Chains Are Being Rebuilt in Public

For thirty years the semiconductor industry optimised for one thing: efficiency. Chips were designed in one country, fabricated in a second, packaged in a third and shipped just in time to factories everywhere. The system produced astonishingly cheap computing. It also produced a fragility that became impossible to ignore when shortages idled car plants and delayed everything from washing machines to medical devices.

Governments responded with industrial policy on a scale not seen in decades: subsidies for fabrication plants, export controls on advanced equipment, and strategic stockpiles. The rebuilding now under way will not recreate the old map, but it is redrawing it, plant by expensive plant.

Why chips are so hard to make anywhere else

A leading-edge fabrication plant costs more than an aircraft carrier and takes years to build, staff and qualify. The equipment inside it comes from a handful of specialist suppliers, the chemicals must be purified to extraordinary standards, and the workforce combines physics, chemistry and software at a depth few regions can assemble quickly. Announcing a fab is the easy part; yielding good chips at volume is the achievement.

That reality has tempered early ambitions. Projects have slipped schedules, costs have risen, and some subsidies have been renegotiated. Analysts tracked by the BBC and trade press note that the industry’s centre of gravity is shifting only gradually, because the ecosystems around existing clusters, from suppliers to universities, took generations to build.

Diversification, not decoupling

The emerging strategy is better described as diversification than decoupling. Companies want a second source in a second region for critical components, friendlier jurisdictions for new capacity, and larger buffer stocks than just-in-time orthodoxy allowed. Efficiency has not been abandoned; it has been repriced, with resilience now appearing as a line item in procurement decisions.

For buyers of chips, which increasingly means every manufacturer, the practical effect is longer planning horizons and closer relationships with foundries. Car makers that once ordered through tiers of suppliers now negotiate directly for capacity years ahead. More of this industrial reshaping is covered in our business coverage.

What it means for prices and products

Duplicated capacity in higher-cost regions will not make chips cheaper in the short run, and some of the subsidy bill ultimately lands with taxpayers and consumers. Set against that is the cost of the alternative: a single disruption, geopolitical or natural, halting production lines across continents. Insurers and finance directors increasingly treat that comparison as the real calculation.

The next decade will therefore feature a hybrid system: global where it is safe, regional where it is strategic, and stockpiled where it is critical. The chip, a component most consumers never see, has become the clearest example of how supply chains everywhere are being rebuilt with security priced in.

None of this will be quick, and some of it will be wasteful, as industrial policy usually is. But the direction is set and the money is committed. The world’s most important supply chain is being rebuilt in public, one foundation pour at a time. Follow developments via our homepage.

Packaging and testing, the less glamorous end of the chain, are attracting investment just as quickly.

The skills shortage, more than money, is now the binding constraint on new plants.

The evidence base is still growing, and researchers are careful to separate what is demonstrated from what is merely plausible at this stage.

For readers trying to judge the headlines, the useful question is always the same: what changed, who measured it, and how large is the effect in ordinary terms?

Policy tends to move more slowly than research, which is why the findings that matter most are usually the ones replicated across several countries and methods.

There is also a distributional question running through the data, because averages hide the groups who gain most and the groups left carrying the costs.