Britain Desperately Needs Semiconductor Capability

Open any device in your home. Crack open the casing of your car's engine management system, prise apart your television, peer inside your gas boiler's control unit. Somewhere within each one sits a small, dark rectangle of silicon—a semiconductor chip—quietly running the show.

These chips are not components in the way a spark plug or a light bulb is a component. They are the decision-makers. They interpret, calculate, regulate, and command. Without them, the device is an inert sculpture of plastic and metal. With them, it thinks.

The modern world runs on semiconductors the way the Victorian world ran on coal and the twentieth century ran on oil. Every sector of the British economy—transport, healthcare, energy, finance, defence, communications—depends absolutely on a continuous supply of these tiny etched wafers. They are in the MRI scanners photographing tumours, the signalling systems preventing train collisions, the encryption hardware protecting bank transfers, the guidance systems steering missiles. They power the servers holding your pension records and the chips validating your contactless payment.

And Britain manufactures almost none of them.

This is not a minor strategic oversight. It is a slow-motion national emergency, one made worse by the fact almost nobody in public life seems to understand its gravity.

A Country Running on Borrowed Infrastructure

Consider what dependency actually means in practical terms. When a chip shortage struck global supply chains during the pandemic years, car factories across Europe shut their doors for weeks at a stretch. Hospitals reported delays in imaging equipment maintenance. Consumer electronics became scarce. Prices rose. Waiting times stretched.

Those disruptions stemmed from a few months of logistical chaos—ships in the wrong ports, lockdowns closing factories, demand patterns shifting faster than production could follow. They were inconvenient, expensive, and temporary.

Now imagine something more deliberate.

Imagine a major power—one controlling critical nodes in the semiconductor supply chain—deciding to restrict exports. Not through dramatic embargoes announced on the evening news, but through quieter means: licensing requirements, end-use restrictions, priority allocations, elongated approval processes. Imagine your suppliers being told they cannot ship certain components to certain destinations, or that they must seek government permission first. Imagine waiting six months for parts that once arrived in six weeks.

This is not speculation. It is already happening. The United States has constructed an elaborate architecture of export controls targeting advanced chips, manufacturing equipment, and the tools needed to produce both. These controls apply extraterritorially—they reach beyond American borders to constrain companies in allied nations whose products contain American technology or intellectual property. The logic is strategic: deny adversaries the computing power needed for advanced weapons and artificial intelligence.

Britain is not the target of these restrictions. But Britain exists within the same global supply chain, subject to the same licensing regimes, dependent on the same concentrated production nodes. In a serious geopolitical confrontation, the question is not whether we would face formal sanctions. The question is whether we could guarantee access to the components our economy requires when every other nation is scrambling to secure its own supply.

The answer, at present, is no.

How Taiwan Became the Centre of the World

To understand how precarious Britain's position has become, one must understand how concentrated semiconductor production has grown.

Taiwan Semiconductor Manufacturing Company—TSMC—produces over sixty per cent of all chips sold globally. For the most advanced processors, the ones powering cutting-edge smartphones, artificial intelligence systems, and military hardware, the figure rises above ninety per cent. No other company comes close. Not Intel in America, not Samsung in South Korea, not any consortium in Europe.

This concentration did not happen by accident. Taiwan made a strategic decision decades ago to become indispensable. The island nation invested relentlessly in fabrication technology, trained generations of process engineers, built ecosystems of suppliers and toolmakers, and welcomed the contracts that other nations considered insufficiently profitable. While Western firms chased quarterly earnings and outsourced manufacturing to wherever labour was cheapest, Taiwan built.

The result is a single point of failure for modern civilisation.

TSMC's fabrication plants sit on an island of twenty-four million people, separated from mainland China by a hundred-mile strait. Beijing considers Taiwan a breakaway province to be reunified, by force if necessary. The United States considers Taiwan's semiconductor capacity so vital to American economic and military power war planning explicitly includes its defence. American warships patrol the Taiwan Strait. American officials state publicly a Chinese invasion would trigger conflict.

This is not geopolitical theatre. It is industrial survival made manifest. The world's great powers have understood what most British policymakers have not: control over semiconductor production is control over the twenty-first-century economy.

Why You Cannot Simply Build More Factories

Some will ask the obvious question: if Taiwan is so risky, why not build factories elsewhere? The answer reveals why semiconductors are unlike any other industrial product humans manufacture.

A single advanced fabrication plant—a "fab"—costs between fifteen and thirty billion pounds. Construction takes five to ten years. The facility requires foundations engineered to absorb vibrations from traffic miles away, air cleaner than an operating theatre, water purer than anything used in pharmaceutical manufacturing, and power supplies so stable a millisecond flicker can ruin an entire production batch.

Inside, extreme ultraviolet light—generated by vapourising droplets of molten tin with high-powered lasers—etches patterns onto silicon wafers at scales measured in atoms. A modern chip contains tens of billions of transistors, each smaller than a virus, arranged with tolerances which would make a Swiss watchmaker weep. A single speck of dust, invisible to the naked eye, can destroy the chip.

This is not engineering. It is controlled sorcery at the edge of what physics permits.

The machines required to perform this sorcery come from a handful of companies, most critically ASML in the Netherlands, which holds a near-monopoly on extreme ultraviolet lithography systems. These machines cost hundreds of millions of pounds each, require years to build, and are themselves subject to export controls.

You cannot simply decide to manufacture advanced chips and then do so. You need the machines, the chemicals, the ultra-pure materials, the trained engineers, the supplier networks, the institutional knowledge accumulated over decades of continuous production. Lose any link in that chain and rebuilding takes a generation.

Britain lost its links thirty years ago.

The Betrayal of an Industry We Invented

Britain did not lose its semiconductor industry because it lacked the talent or the technology. Britain helped invent semiconductors.

British companies—Ferranti, Plessey, GEC, Mullard—manufactured transistors and integrated circuits when Silicon Valley was still orchards. British engineers pioneered concepts which underpin modern computing.

In 1978, the British government established Inmos, a national champion intended to compete directly with Intel. The company produced the Transputer, a parallel-processing chip so advanced its architectural concepts would not become mainstream for another three decades. The Transputer's approach to on-chip communication and scalable computing anticipated the design principles now used in the graphics processors powering artificial intelligence.

Inmos built a fabrication plant in Newport, Wales. For a brief moment, Britain possessed genuinely world-class semiconductor manufacturing capability.

Then came the Treasury.

Whitehall demanded commercial returns on political timescales. Long-term industrial strategy gave way to short-term accountancy. Inmos was sold to SGS-Thomson (now STMicroelectronics). The Newport fab changed hands repeatedly before eventually being acquired by a Chinese-backed firm, prompting belated government intervention under national security powers.

Meanwhile, Taiwan, South Korea, and Japan took the opposite approach. Their governments provided patient capital, long-term subsidies, and strategic coordination. They treated semiconductor manufacturing not as a commercial venture subject to market forces but as national infrastructure essential to sovereignty.

Britain chose ideological purity. The market would decide.

The market decided—and moved offshore.

What Remains: Design Without Manufacture

Out of the wreckage of British semiconductor manufacturing emerged one genuine success: ARM.

ARM designs the processor architectures used in virtually every smartphone on Earth. Its intellectual property sits at the heart of billions of devices. ARM's achievement is real and remarkable—a triumph of British engineering and business innovation.

But ARM does not manufacture anything.

ARM's designs are fabricated in Taiwan, South Korea, and increasingly in American plants funded by massive government subsidies. The value created by British ingenuity flows to factories controlled by other nations. If those nations choose to restrict access, British-designed chips become unavailable to British buyers.

In 2016, ARM was sold to SoftBank, a Japanese conglomerate. The British government approved the sale with minimal conditions other than promises for the firm to remain in Cambridge and increase headcount. Britain surrendered control of its most strategically significant technology company in exchange for assurances about job numbers that have since proved hollow.

Five years later in 2021, NVIDIA wanted it. That sale collapsed due to a Public Interest Intervention Notice (PIIN) because of national security concerns.

Does all this sound depressingly familiar?

This is the pattern: design capability retained, manufacturing capability abandoned, ownership dispersed. Intellectual brilliance married to industrial impotence.

The Real Meaning of Dependency: Coercion

Critics will object that Britain has survived perfectly well by importing chips. Supply chains adapt. Markets clear. Shortages resolve themselves.

This objection misunderstands the nature of the risk.

A shortage is a temporary mismatch between supply and demand. Prices rise, alternatives emerge, equilibrium returns. Shortages are inconvenient but manageable.

Dependency is different. Dependency means another party can impose costs on you—or deny you resources entirely—as an instrument of policy. Dependency transforms commercial relationships into potential vectors of coercion.

Consider how this works in practice.

The United States has demonstrated repeatedly it will use export controls strategically. American restrictions on semiconductor equipment to China are not commercial decisions but geopolitical weapons, designed explicitly to retard Chinese technological development. The logic is openly stated: advanced computing power is a strategic asset, and adversaries must be denied access.

If Britain matches fully with American policy in a future confrontation, Chinese counter-leverage may target upstream materials and components where China retains market power. If Britain diverges from American policy, the architecture of US export controls—applied extraterritorially through the supply chain—can still constrain what British companies receive.

This is the trap of dependency: you lose freedom of action in both directions. Allyship and independence both carry costs you cannot control.

The scenario requiring a dramatic embargo is not necessary. Quiet restrictions suffice. Licensing delays. Allocation priorities favouring domestic customers. Informal guidance to suppliers. The effect is the same: a country dependent on imported chips finds its options narrowing.

When the Chips Stop Coming

Abstract arguments about strategic autonomy can seem remote. The consequences are not.

In the first weeks of a serious supply disruption, vehicle production halts. Modern cars contain thousands of chips—engine management, braking systems, airbag sensors, navigation, entertainment. Without a continuous flow of replacements, assembly lines stop. Service garages cannot obtain parts for repairs.

Within months, deeper systems begin failing. Railway signalling equipment requires semiconductor components for maintenance. Power grid control systems need regular replacement parts. Hospital imaging machines—CT scanners, MRI systems, ultrasound equipment—depend on chips for processing and control. When components fail and cannot be replaced, the machines fall silent.

Telecommunications infrastructure degrades as network equipment cannot be serviced. Data centres, the backbone of modern finance and commerce, face the same constraints. Banking systems, which rely on continuous hardware refreshes to maintain security and capacity, begin encountering bottlenecks.

Defence systems, supposedly ringfenced from civilian concerns, face their own vulnerabilities. Modern military hardware runs on semiconductors like everything else. Secure communications equipment, guidance systems, surveillance platforms, logistics software—all require chips. Stockpiles help, but stockpiles deplete.

The National Health Service, already stretched beyond design capacity, would find equipment maintenance impossible. The choice would become stark: cannibalise working machines for parts, ration access to diagnostic equipment, or watch imaging capabilities collapse.

This is not a war scenario. This is what happens when a supply chain under political pressure fails to deliver.

The American Response

The United States has recognised the danger. The CHIPS and Science Act, passed in 2022, allocated fifty-two billion dollars in direct subsidies for domestic semiconductor manufacturing, alongside hundreds of billions more in tax incentives and research funding. Intel, TSMC, and Samsung are all building new fabrication plants on American soil—in Arizona, Texas, Ohio, New York.

The strategic logic is explicit. Semiconductors are too important to outsource. America will pay whatever it costs to rebuild domestic capacity, even if such capacity is more expensive than Asian alternatives. Efficiency is secondary; security is primary.

This represents a fundamental shift in American economic philosophy. For decades, the orthodoxy held global supply chains optimised for cost were inherently desirable. Concentration of production in low-cost regions was efficient. Markets allocated resources better than governments.

The CHIPS Act repudiates that orthodoxy. It declares, in legislative form, some industries are too strategically vital for market logic alone. America will intervene, subsidise, and protect—not because doing so is profitable, but because doing so is necessary.

Britain's Response: Memos and Compound Semiconductors

Against America's fifty-two billion dollars in chip manufacturing subsidies, Britain's twenty-year semiconductor strategy launched under Sunak promised up to one billion pounds over a decade.

The money is real, but the scale is laughable. One billion pounds would not cover the cost of a single advanced fab. It would not fund the supplier ecosystems, the training programmes, the research facilities required to rebuild genuine manufacturing capability. It is a rounding error on American and Asian investment.

It is a few days of NHS spending.

We have promised £21.8 billion in international vanity money to Ukraine:

The strategy itself leans into British strengths: chip design, compound semiconductors for specialist applications, research and development. These are not trivial areas. Compound semiconductors—using materials beyond silicon—enable high-frequency communications, power electronics, and advanced sensors. Britain has genuine expertise here.

But compound semiconductors are a niche within a niche. They do not address the core vulnerability: mass-market chips for everyday applications. The engine control units in cars, the processors in medical equipment, the controllers in power systems—these use conventional silicon chips produced in massive volumes by Asian fabs.

Britain's strategy essentially concedes mass manufacturing is beyond reach. The plan is to remain clever while others remain capable.

The False Comfort of Allied Supply Chains

Some will argue Britain need not build chips itself. Allied nations—America, Japan, South Korea, the Netherlands—can provide what Britain cannot make. We will purchase from friends.

This argument assumes allies will prioritise British needs during a crisis. History suggests otherwise.

When supply chains tightened during the pandemic, nations secured their own populations first. Medication, personal protective equipment, medical devices—all became subject to export restrictions and national stockpiling. The European Union contemplated blocking exports to Britain. America invoked defence production authorities to redirect supplies domestically.

Allies remain allies. But governments answer first to their own citizens. In a severe semiconductor shortage, American fabs would serve American customers. Korean fabs would serve Korean customers. The British would join a queue.

The market cannot solve this. Market mechanisms work beautifully when supply is abundant and demand predictable. In a crisis, markets fragment, hoard, and fail. Prices signal scarcity; they do not create supply.

Industrial Sovereignty: The Capacity to Disagree

Strip away the technical details and the argument reduces to a single principle: a nation which cannot produce what it needs cannot control its own destiny.

Industrial sovereignty does not mean autarky—the fantasy a nation can manufacture everything within its borders. Modern economies are too complex, too specialised, too interdependent for that.

But industrial sovereignty does mean maintaining domestic capability in strategic sectors. It means ensuring critical supply chains include nodes under national control. It means possessing the capacity to sustain essential systems when global markets convulse or when great powers turn hostile. In a word: resiliency.

Most fundamentally, industrial sovereignty means retaining the ability to disagree.

A nation dependent on another for critical resources cannot easily oppose that nation's policies. The dependency creates leverage—sometimes explicit, more often implicit. Decisions are shaped by supply considerations which never appear in official documents. Freedom of action erodes.

Britain's political class speaks constantly about sovereignty. The word was invoked endlessly during debates about European integration. Yet the same political class has allowed industrial dependency to deepen to the point where foreign governments can, through control of chip supplies, constrain British choices without firing a shot.

This is not sovereignty. It is the illusion of sovereignty, maintained while the foundations crumble.

What Rebuilding Actually Requires

The gap between rhetoric and reality must be closed. Doing so requires measures that will offend every orthodoxy of modern British economic policy.

1. First, scale of investment. The sums required are measured in tens of billions, not single billions. A credible programme to rebuild semiconductor manufacturing capability—even focused on mature-node chips rather than cutting-edge processes—requires capital deployment an order of magnitude larger than current plans contemplate.
2. Second, time horizons. Semiconductor fabs take years to build and more years to optimise. Workforce training requires sustained programmes running for decades. Supply chains develop over generations. This is not a project for one parliament but a commitment spanning decades.
3. Third, tolerance for commercial inefficiency. A British fab will not compete on cost with a Taiwanese or Korean facility enjoying decades of accumulated advantages. Government must underwrite the difference—not as subsidy in the pejorative sense, but as investment in strategic capability. Defence procurement already operates on this logic; industrial strategy must too.
4. Fourth, integration of design and manufacture. ARM's model—British design, foreign fabrication—delivers commercial returns but not industrial security. Future policy must link British design strengths to British (or at minimum allied) production pathways.
5. Fifth, training and retention. The engineers who operated British fabs have largely retired. Their knowledge has not been transmitted. Rebuilding requires educational programmes, apprenticeships, and career paths that make semiconductor manufacturing attractive again.
6. Sixth, political durability. Industrial strategy cannot survive changing governments cancelling predecessors' projects for partisan advantage. A national commitment, insulated from electoral cycles, is essential.

The Mineral Analogy: Coal, Oil, and Now Silicon

Every era has its essential resource—the material without which the economy cannot function and the military cannot fight.

For nineteenth-century Britain, it was coal. The industrial revolution ran on British coal, mined from British seams by British workers. Coal powered the factories, the railways, the ships, the entire edifice of Victorian prosperity and imperial reach. When Britain controlled its own coal supply, Britain controlled its destiny.

For twentieth-century powers, it was oil. The shift from coal to petroleum remade geopolitics. Nations with oil reserves—or with the military capacity to secure them—dominated. Nations without found themselves dependent on those who had. The history of the Middle East, of American foreign policy, of global conflict in the twentieth century, cannot be understood without understanding the strategic centrality of oil.

In the twenty-first century, the essential resource is silicon—or more precisely, the processed silicon wafers etched with billions of transistors which comprise semiconductor chips.

Control over chip production is control over the commanding heights of the modern economy. Nations who manufacture chips possess leverage over those that do not. In extremis, that leverage becomes a weapon.

Britain once understood this logic. British coal powered British industry and British power. British policymakers grasped, if imperfectly, the connection between resource security and national strength.

Somewhere along the way, such understanding was lost. Britain became a nation which assumes supply chains will always deliver, markets will always provide, dependency on foreign production carries no strategic cost.

The assumption is false. The cost is already accruing.

The Machines Stop

E.M. Forster wrote a short story in 1909 called The Machine Stops. In it, humanity lives entirely dependent on an omnipresent Machine that provides everything—light, food, communication, meaning. The Machine's workings are beyond any individual's understanding. When it begins to fail, no one knows how to repair it. Civilisation collapses.

We are closer to Forster's dystopia than most imagine.

Modern Britain depends on systems whose inner workings few understand and fewer still can repair without access to components manufactured on the other side of the world. The complexity is hidden beneath smooth interfaces and seamless services. Everything works—until it doesn't.

The pandemic offered a preview. Ships queued outside ports. Factories idled for want of parts. Supermarket shelves emptied. The infrastructure of daily life revealed its fragility.

That fragility has not been addressed. It has been covered over with the scar tissue of returning normalcy and the convenient amnesia of political discourse. The underlying vulnerability remains.

A serious disruption to semiconductor supply would make the pandemic look like a rehearsal. Not because chips are inherently more important than other goods, but because chips are embedded in everything else. Disable chips and you disable transport, power, healthcare, finance, defence, communications—simultaneously and cumulatively.

The systems do not fail gracefully. They fail together.

The Choice That Can No Longer Be Postponed

Britain faces a choice its political class has spent decades avoiding. We can continue on the present course: celebrating design excellence while manufacturing capacity withers, signing strategy documents while other nations build factories, trusting global markets will always supply what we cannot make, hoping allies will prioritise our needs when their own populations demand first claim.

Or we can rebuild.

Rebuilding is expensive. It requires investment on a scale which makes Treasury officials blanch. It requires patience electoral cycles discourage. It requires admitting market forces alone will not secure national interests—a confession which offends ideological pieties across the political spectrum.

But the alternative is worse. The alternative is permanent dependency, permanent vulnerability, permanent subordination of British choices to foreign calculations.

The chip is the new coal. A nation that cannot secure its supply cannot call itself sovereign.

For three decades Britain has acted as though this truth does not apply—as though we are somehow exempt from the strategic logic governing everyone else. We have outsourced, offshored, sold off, and run down. We have confused intellectual sophistication with industrial strength, financial services with productive capacity, consumption with capability.

The bill is coming due.

Other nations have understood. America is spending hundreds of billions to rebuild what it should never have lost. Taiwan staked its entire national strategy on semiconductor supremacy. South Korea, Japan, and the European Union are all scrambling to secure their positions.

Britain sleeps.

The question is not whether we can afford to rebuild a semiconductor industry. The question is whether we can afford not to. The answer stares from every device, every vehicle, every machine on which modern life depends.

We made these chips once. We can make them again.

If we choose to.

If we find the will.

If we remember that a nation's wealth means nothing when it cannot make the things that keep civilisation running.