The Great Offshoring: How the World's Factory Moved East

Here is a number that should stop you cold: in 1990, the G7 nations produced roughly two-thirds of everything manufactured on Earth. By 2024, they produce less than two-fifths. In the span of a single generation — within the working life of a factory hand who started on the floor in Coventry or Cleveland — the industrial centre of gravity shifted from the Atlantic to the Pacific. It was the largest peaceful transfer of productive capacity in human history, and most of the people it happened to never voted for it, were never consulted about it, and still do not fully understand it.

The West did not lose its factories to war, revolution, or natural disaster. It gave them away. It gave them away voluntarily, enthusiastically, and for what seemed, at the time, like very good reasons. The logic was simple and seductive: why pay a worker in Detroit $25 an hour when a worker in Shenzhen would do the same job for 50 cents? The spreadsheets were irresistible. The consequences were not on the spreadsheets.

This is the story of the Great Offshoring — how the world’s factory moved East, what it cost, and why the bill is still arriving.

The First Wave: Trade Opens, Capital Flows (1990–2000)

The offshoring revolution began not with a bang, but with a fax machine. In the early 1990s, the simultaneous liberalisation of trade barriers, the opening of former Soviet economies, and the spread of container shipping created conditions that no corporate strategist could resist. A new pool of roughly 1.5 billion workers had suddenly entered the global labour market. Labour was cheap. Regulation was lighter. And the shipping container — that unglamorous steel box — had made it almost as cheap to move goods 8,000 miles across the Pacific as 80 miles across the Midlands.

The first wave was cautious, almost tentative. American companies moved operations to Mexico, emboldened by NAFTA in 1994. Western European firms looked to Poland, the Czech Republic, and Hungary, where skilled workforces could be had for a fraction of German or French wages. This was “near-shoring” — keeping production close to home, just across a convenient border.

But the real giant was already stirring. China, under the reforms initiated by Deng Xiaoping in the late 1970s, had established Special Economic Zones — patches of capitalism stitched onto the fabric of a nominally communist state. By the early 1990s, Shenzhen had grown from a fishing village of 30,000 to a manufacturing city of over 3 million. It was not just offering cheap labour. It was offering a complete package: government subsidies, purpose-built infrastructure, lax environmental enforcement, and a workforce that would work twelve-hour shifts six days a week for wages that, while meagre by Western standards, were a fortune compared to subsistence rice farming in Hunan province.

The economists called it “comparative advantage.” The workers in Sunderland and Youngstown, who had not been consulted, called it something rather less polite.

Between 1990 and 2000, China’s share of global manufacturing value added doubled, from approximately 3% to 7%. It was still modest. The G7 nations still commanded over 60% of the world’s industrial output. But the trendlines were already diverging, and anyone reading the data with clear eyes could see where they were heading.

The Pivot Point: 11 December 2001

If there is a single date that marks the death of the old manufacturing order, it is 11 December 2001 — the day China formally acceded to the World Trade Organisation.

Before WTO membership, Western companies considering Chinese production faced genuine uncertainty. Tariffs could change without warning. Intellectual property protections were notional at best. The rules of the game were subject to the whims of provincial officials. Many companies dipped a toe in but kept their core production at home.

WTO accession changed the calculus overnight. China now operated under a rules-based trading system with predictable tariffs, dispute resolution mechanisms, and — crucially — Most Favoured Nation status with virtually every major economy. The risk premium evaporated. The floodgates opened.

The numbers that followed are staggering. Between 2000 and 2010, the United States lost 5.8 million manufacturing jobs — a rate of decline faster than during the Great Depression. In the UK, manufacturing employment fell from 4.5 million in 1990 to 2.5 million by 2010. In the American Midwest, entire towns that had been built around a single factory or steel plant found themselves without an economic reason to exist.

The economists David Autor, David Dorn, and Gordon Hanson coined the term “the China Shock” to describe this phenomenon. Their research demonstrated that the regions most exposed to Chinese import competition experienced not just job losses, but cascading social consequences: higher unemployment, lower wages, increased disability claims, higher rates of divorce, drug abuse, and mortality. The factory did not just provide a wage. It provided a structure — a community, an identity, a reason to get up in the morning. When the factory left, the structure collapsed.

Yet here is the paradox that the data reveals with brutal clarity: Western manufacturing output did not collapse. It continued to rise. The United States, the UK, and Germany all produce more manufactured goods by value today than they did in 1990. What collapsed was manufacturing employment. The factories that remained became hyper-automated, producing more with fewer people. The workers who lost their jobs were not replaced by Chinese workers so much as by robots and software — but the political narrative of offshoring was simpler and angrier than the reality of automation, so it stuck.

The Smile Curve Trap

Western policymakers had a theory for why none of this mattered. It was called the Smile Curve, and it was first proposed by Stan Shih, the founder of Acer, in 1992.

The Smile Curve argued that the value in any product lies at the two ends of the production chain — in design and R&D at one end, and in marketing, branding, and retail at the other. The middle of the chain — the actual manufacturing — was the lowest-value activity. On a graph, the value chain formed a U-shape: high at both ends, low in the middle. It looked like a smile.

The implication was reassuring: let the East have the dirty, low-margin business of bending metal and sewing fabric. The West would keep the lucrative work of designing products and building brands. Apple would design the iPhone in Cupertino, capture 60% of the profit margin, and let Foxconn assemble it in Zhengzhou for pennies. Everybody wins.

For a while, this narrative held. Silicon Valley boomed. The City of London boomed. The service economy expanded. GDP kept growing. If you looked only at the aggregate numbers and squinted at the distributional effects, all was well.

But the Smile Curve had a fatal flaw, and it took two Harvard professors to identify it. In 2009, Gary Pisano and Willy Shih published a landmark paper arguing that manufacturing capability and innovation capability are not separate things — they are deeply intertwined. When you offshore the production of a technology, you eventually offshore the ability to innovate in that technology. The knowledge of how to make things is not abstract. It lives on the factory floor, in the tacit expertise of engineers and technicians who solve problems at the point of production.

The evidence is now overwhelming. The United States invented the lithium-ion battery (at Exxon, in the 1970s). It lost the ability to manufacture batteries at scale when production moved to Asia. By 2024, China controlled 77% of global lithium-ion battery cell manufacturing capacity. China does not merely assemble electric vehicles. It dominates the entire supply chain, from mining the lithium in the Congo and Chile, to processing the cathode materials in Fujian, to manufacturing the cells in CATL’s megafactories, to writing the battery management software. The country that makes the batteries is the country that innovates in battery technology. America, which invented the underlying science, now imports the finished product.

The same pattern has repeated in solar panels, telecommunications equipment, advanced displays, shipbuilding, and increasingly in semiconductors. The Smile Curve turned out to be a smirk.

Deindustrialisation by the Numbers

The raw data on deindustrialisation reads like an autopsy report. Manufacturing as a share of GDP in the United Kingdom fell from 16.5% in 1990 to approximately 8.5% in 2024. The United States saw a comparable decline, from 16.6% to roughly 10.4%. Germany, which retained a more explicit industrial policy and a stronger apprenticeship system, fared better — but still fell from 26% to 19%.

Meanwhile, China’s manufacturing share has barely moved, hovering between 27% and 33% for three decades. And a new generation of manufacturing nations has emerged in China’s wake. Vietnam’s manufacturing share of GDP surged from 12% in 1990 to approximately 25% in 2024, as companies pursuing a “China Plus One” strategy diversified their supply chains into Southeast Asia. Bangladesh’s garment sector now employs over 4 million workers and accounts for 85% of the country’s export earnings.

The geographic redistribution is even starker when measured by manufacturing value added in absolute terms. In 1990, the G7 nations accounted for approximately $3.5 trillion of the world’s $5.3 trillion in manufacturing output. By 2024, the global total had risen to roughly $16 trillion — but the G7’s share had fallen to around $6.4 trillion, while China alone accounted for approximately $5 trillion. One country, which produced almost nothing of global consequence forty years ago, now manufactures nearly as much as the entire G7 combined.

The New Map: Fragmentation and the China Plus One Era (2010–Present)

The offshoring story entered a new phase after 2010, driven by three forces: rising Chinese wages, geopolitical anxiety, and pandemic-induced supply chain chaos.

Chinese factory wages, which had been the engine of the entire offshoring movement, began rising sharply. Average manufacturing wages in China increased roughly fivefold between 2005 and 2020. Shenzhen, once the epitome of cheap labour, now has wages comparable to parts of Southern Europe. The arbitrage that launched the Great Offshoring was closing.

At the same time, geopolitics intruded. The US-China trade war, initiated under Donald Trump in 2018 and largely continued under Joe Biden, imposed tariffs and export controls that made Chinese supply chains politically risky. The COVID-19 pandemic in 2020 then delivered a visceral demonstration of what concentration risk actually means: when Chinese factories shut down, Western companies could not get components for anything from automobiles to antibiotics.

The response has been “China Plus One” — not the abandonment of China, but the addition of alternative suppliers in Vietnam, India, Mexico, and Bangladesh. Vietnam’s exports to the United States tripled between 2018 and 2023. Mexico is experiencing a “nearshoring” boom, with foreign direct investment in manufacturing hitting record levels in 2023. India, under its “Make in India” initiative, is courting the factories that might otherwise have gone to China.

Yet the deeper structural shift has not reversed. East Asia’s share of global manufacturing value added continues to climb. The G7’s share continues to fall. The Western “reshoring” initiatives — the US CHIPS and Science Act, the EU’s Net-Zero Industry Act — are real, and they are large. The CHIPS Act alone commits $280 billion in federal spending. But they are investments in specific sectors (semiconductors, green energy), not a reversal of the broader trend. Building a TSMC fab in Arizona does not bring back the textile mills, the furniture factories, or the consumer electronics assembly lines. Those are gone, and they are not coming back.

The data tells us something uncomfortable: the centre of industrial gravity has shifted to the Pacific, and the inertia is immense. The West is not reindustrialising. It is spending enormous sums to maintain strategic footholds in a few critical technologies while the broad base of global manufacturing continues to consolidate in East and Southeast Asia.

What Was Lost

The standard economic defence of offshoring is that it was a net positive: Western consumers got cheaper goods, Western corporations got higher profits, and the developing world got jobs and growth. All of this is true. A television that cost a month’s wages in 1990 costs a day’s wages in 2024. Global poverty fell faster between 1990 and 2020 than in any previous period in human history, largely because Chinese and Asian manufacturing employment lifted hundreds of millions out of subsistence farming.

But the defence obscures what was lost. What was lost was not merely employment — it was the economic foundation of an entire social class. The factory job in the West was never just a job. It was a gateway to the middle class: a reliable income, a pension, employer-provided healthcare (in the US), union representation, and — critically — a sense of dignity and purpose that came from making things. The service-sector jobs that replaced manufacturing work were, on average, lower-paid, less stable, and less likely to provide benefits.

The political consequences are still reverberating. The regions that lost the most manufacturing jobs are the regions that swung hardest toward populist politics — toward Brexit in the UK, toward Trump in the United States, toward the AfD in Germany, toward Le Pen in France. The correlation is not coincidental. People who feel that the economic system has abandoned them tend to abandon the political system that oversaw the abandonment.

And there is a strategic dimension that transcends economics. A nation that cannot manufacture advanced goods is a nation that depends on others for its security. The United States discovered during the pandemic that it could not produce enough N95 masks for its own hospitals. Europe discovered in 2022, when Russia invaded Ukraine, that it had outsourced most of its ammunition and artillery shell production and could not ramp up fast enough to sustain a conventional war on its own continent. The dependency is not theoretical. It is measured in hospital beds and artillery rounds.

The Lesson That History Keeps Teaching

The Great Offshoring was not the first time a dominant civilisation traded productive capacity for short-term comfort. Venice, which dominated Mediterranean trade for centuries, gradually outsourced its shipbuilding and manufacturing to cheaper competitors in the eastern Mediterranean during the sixteenth and seventeenth centuries. Venetian merchants found it more profitable to finance trade than to build the ships that carried it. By the eighteenth century, Venice was a museum — beautiful, culturally rich, and economically irrelevant.

The Netherlands experienced a similar trajectory. The Dutch Republic of the seventeenth century was the most productive economy in Europe: it built ships, processed raw materials, manufactured textiles, and innovated in finance. By the eighteenth century, Dutch capital had shifted from production to finance and speculation. The country remained wealthy — but its wealth was increasingly derived from lending money to others rather than making things itself. When the strategic crises of the Napoleonic era arrived, the Netherlands lacked the industrial base to defend its position.

The pattern is consistent: nations that shift from production to intermediation — from making things to financing the making of things — enjoy a period of extraordinary prosperity, followed by a slow decline in strategic weight. The financiers always believe they have found a way to get rich without getting their hands dirty. They are usually right, for a generation. Then the people who are getting their hands dirty turn out to have built something more durable.

The Bill Arrives

We are now thirty years into the Great Offshoring, and the bill is arriving in three envelopes.

The first envelope contains the social costs: the hollowed-out industrial towns, the opioid epidemic in America, the political disintegration of the centre in Europe, the rise of populism from São Paulo to Stockholm. These costs were not on the original spreadsheets, but they are real, and they are large.

The second envelope contains the strategic costs: dependence on Chinese manufacturing for everything from pharmaceuticals to semiconductors, a defence-industrial base that cannot produce artillery shells at the rate Ukraine consumes them, and a green energy transition that requires components — batteries, solar cells, rare earth magnets — that the West cannot make in sufficient quantity.

The third envelope has not yet been opened. It contains the innovation costs: the long-term consequences of outsourcing production on the ability to innovate. If Pisano and Shih are correct — and the evidence from batteries, solar, and telecommunications suggests they are — then the West has not merely outsourced the making of things. It has begun to outsource the thinking about things. The next generation of breakthrough technologies may emerge not from Stanford or Cambridge, but from the factory floors of Shenzhen and Seoul and Ho Chi Minh City, where the people who make the products are also the people who understand how to improve them.

The Great Offshoring was the largest economic experiment of the modern era. The spreadsheets said it would make everyone richer. It did — in aggregate. But aggregates are the favourite hiding place of economists who would rather not talk about distribution. The factory workers in Guangdong got richer. The shareholders in Connecticut got richer. The factory workers in Coventry got redundancy notices.

You cannot outsource the making of things without eventually outsourcing the power that comes with making things. That is the lesson of Venice, of the Netherlands, of every civilisation that decided it was too rich and too clever to bother with production. The West made that bet in the 1990s. The data is now in. And the East, which accepted the factories, also accepted the future.