In Ludwigshafen, BASF operates the world's largest integrated chemical site. 39,000 people work there. At the same time, BASF is building the world's third-largest Verbund (integrated production) site in Zhanjiang, southern China, for €8.7 billion — the largest single investment in the company's history. Finished by 2030. Without delays. Without cost overruns. In Germany, BASF is closing plants in Frankfurt and Knapsack, cutting 3,200 jobs at its headquarters site, launching savings programmes.
This is not news about a company. This is news about the direction of travel.
In Duisburg-Hochfeld, the unemployment rate in March 2025 stands at 13.4 per cent. In the federal election, the AfD wins up to 36 per cent in districts such as Marxloh and Neumühl. The son of the Anatolian Thyssen worker who came to Duisburg in 1975 today works in a plant that is cutting thousands of jobs. He votes AfD — because of energy prices, he says. What he actually means is: the system has forgotten me.
These two images — Hyderabad and Hochfeld — describe the same phenomenon from different sides. One shows where the capacity is going. The other shows what is left behind.
I. What deindustrialisation means
Deindustrialisation is not a cyclical phenomenon. The business cycle is ups and downs — the same economy, different levels of capacity utilisation. Deindustrialisation is structural loss: capacities that disappear, knowledge that dissolves, networks that tear apart. What is gone does not come back — not because no one wants it to, but because industrial ecosystems take decades to emerge and cannot be reconstructed by decree.
Germany has achieved no real economic growth in the last seven years — since 2018. The metal and electrical industry is in its tenth quarter of recession. In 2025 alone, 124,000 jobs were lost in industry, and more than 111,000 in the automotive industry since 2019. The chemical industry association VCI reports average plant utilisation of 70 per cent — permanently below the profitability threshold. This is not a dent. This is a change of direction.
At the same time, silent business closures — closures that appear in no insolvency statistics — rose by 50 per cent in 2025. Datev counts a rate of 2.43 per cent among its client mandates with planned or unplanned closures and relocations — three times the formal insolvency rate. According to KfW, 231,000 companies are thinking of shutting down. Most without a successor: 114,000 closures a year until 2029, because no one is taking over and no one is expected to. The average age of SME owners is 54, and 39 per cent are over 60. The demographic erosion of the Mittelstand (Germany's backbone of small and medium-sized, often family-owned firms) is a silent catastrophe that makes no headlines because it has no dramatic moment — only a thousand unnoticed endings.
II. The energy trap
The chemist Henrik Follmann from East Westphalia put it clearly in 2022: his own combined heat and power plant costs him fifteen times as much as he can afford in order to produce competitively on the world market. In the USA, industrial electricity costs four cents per kilowatt-hour. In Germany, several times that. This price difference is not the business cycle. It is a political decision — accumulated over two decades of energy policy that wanted to exit coal, nuclear power and gas all at once without building up the alternatives in time.
The result: energy-intensive industries can no longer produce economically in Germany. Aluminium, steel, basic chemicals, foundries — sectors that cannot simply switch to renewable energy because they need gas not only as an energy source but as a raw material, because their processes depend on continuous high temperatures, because the conversion costs cannot be shouldered in an already strained margin situation. They throttle back, relocate, close.
The Megamachine dimension: the energy path was chosen politically — through subsidies for particular technologies, through the nuclear phase-out, through CO2 pricing without adequate compensation for competitiveness, through the primacy of climate policy over industrial location policy. That was not a single wrong decision. It was the result of decades of interest-group politics in which various actors of the Megamachine pushed through their goals — energy companies, environmental associations, party strategists, the bureaucracies of the European Commission — without anyone drawing up the overall bill for industry. And now the Megamachine can no longer reverse the decision. The operating licences have expired. The know-how is gone. The supply chains have reorganised themselves.
III. Zhanjiang
In November 2025, BASF begins production at its new Verbund site in Zhanjiang in the southern Chinese province of Guangdong. An investment of €8.7 billion — the largest single investment in the company's history. Four square kilometres of site. A steam cracker producing one million tonnes of ethylene a year. 100 per cent renewable energy. Completed according to plan, without cost overruns, and by 2030 the third-largest BASF site worldwide after Ludwigshafen and Antwerp.
At the same time: closure of plants in Frankfurt and Knapsack. A savings programme in Ludwigshafen. 3,200 jobs gone at the headquarters site.
BASF chief executive Kamieth explains it soberly: China today has a 50 per cent share of the global chemicals market. 80 per cent of growth up to 2035 will be concentrated in Asia-Pacific. In Germany, the German business is posting losses running into billions — because of energy costs, bureaucracy, weak demand. In China, the market is growing. So that is where one invests. This is not bad management. It is a rational response to incentive structures that the Megamachine has set.
The Zhanjiang paradox: the company that has been rooted in Ludwigshafen since 1865 and operates the world's largest integrated chemical site there — is building its investment in the future in China. Not in order to leave Germany. But because Germany can no longer offer the conditions that make this growth possible: affordable electricity, fast permits, a growing home market. What BASF built in Zhanjiang in five years would have required fifteen years of permit procedures, citizens' protests and energy debates in Germany — and might never have been built at all.
Kamieth says: "Imagine what a BASF without China, half of the world market, would look like." That is true. But the counter-question also arises: what does Germany look like when its largest industrial companies systematically make their growth investments elsewhere?
IV. The value chain as a domino effect
What the public debate on deindustrialisation systematically underestimates is interconnection. Industry is not a set of individual companies standing side by side — it is an ecosystem of value chains in which every link depends on the others.
An example from Nordenham: in 2022 the Glencore zinc smelter throttles production because of high electricity prices. Its neighbour Kronos Titan — the main customer for the sulphuric acid produced in smelting — has for its part throttled production. So there is no longer any use for the sulphuric acid. So the zinc smelter throttles back further. Two companies driving each other into crisis — not because either of them is badly managed, but because the system that connects them collapses under external pressure.
The same applies on a large scale to the automotive industry and its suppliers. VW, BMW and Mercedes are struggling with the shift to electric mobility and with Chinese competition — so order volumes for suppliers shrink. The suppliers — automotive suppliers account for 30 per cent of the industrial employees affected by insolvencies — come under pressure. When a medium-sized automotive supplier closes that was the only domestic supplier of a particular component, that knowledge also disappears from the system. The OEM then has to import the component or develop it anew — both more expensive, slower, more vulnerable.
What the Megamachine has produced is an industry that is highly specialised, highly interconnected and therefore highly fragile. Decades of efficiency optimisation have bred out the redundancies. Just-in-time, lean production, single sourcing — those were the principles that secured competitiveness. Now they are the principles that maximise vulnerability.
IV. What the Ruhr shows
The Ruhr is the oldest example and therefore the most instructive. The coal and steel industry was dismantled here over decades — with enormous subsidies, with socially cushioned job cuts, with structural policy programmes meant to attract new industries. The result after 40 years: Duisburg has an unemployment rate of 13.4 per cent. The AfD wins up to 36 per cent in the former working-class districts. The grandson of the Ruhr steelworker sits in the park and explains why he votes AfD.
This is not the failure of structural change. It is the result of a structural change that succeeded — in its own way. Industry really was dismantled. The workers really were given severance pay and retraining. New sectors came — logistics, the creative industries, universities. But they did not replace what had disappeared: not the wages, not the sense of identity, not the dignity of a region that had defined itself through its industrial achievement.
What happened in the Ruhr over four decades is now happening faster and across the board. The automotive industry in Baden-Württemberg. The chemical industry along the Rhine. Mechanical engineering in Bavaria. The glass industry in Saxony. Foundries everywhere. These are not four regional crises — this is a national structural shift.
VI. What the Megamachine cannot do
The Megamachine has responded to deindustrialisation with what it always does: subsidies, funding programmes, commissions, special funds. €500 billion of infrastructure investment. Debates about an industrial electricity price. Promises to cut red tape. Summits on Germany as an industrial location.
What it cannot do: reverse the structural decisions that fixed the energy path. The nuclear power stations have been shut down — the knowledge that made operation and maintenance possible has dispersed, the skilled workers are in other sectors, and there is hardly any public acceptance for a return. What it cannot do: reconstruct the value chains that have dissolved. When a specialised foundry supplier closes and the knowledge of its skilled workers scatters, that cannot be brought back by a funding programme. What it cannot do: reverse the demographics that leave 231,000 businesses without a successor.
That is the real Megamachine insight: the system has manoeuvred itself into a situation from which it structurally can no longer escape — because every decision it has made has created facts that constrain the next decision. The path dependency of deindustrialisation is not an accident. It is the result of decades in which short-term interests determined long-term structures.
When the capacities are gone, they are gone. When the know-how is gone, it is gone. When the networks tear, it takes decades to weave them anew. The Megamachine can do a great deal — but it cannot write industrial history backwards.
VII. Duisburg, Hyderabad and Zhanjiang
Back to the three images. BASF opens a digital hub in Hyderabad. BASF builds the world's third-largest chemical site in Zhanjiang for €8.7 billion. And in Duisburg-Hochfeld the unemployment rate stands at 13.4 per cent.
All three decisions are rational. None is evil. Together they describe a Megamachine that optimises capital and know-how globally and externalises the social costs of this optimisation regionally.
The son of the Anatolian Thyssen worker in Hochfeld votes AfD — because of energy prices, he says. What he means is: the system has forgotten me. The Megamachine that chooses Hyderabad, builds Zhanjiang and leaves Hochfeld behind cannot give him an answer. Nor can a politics that does not change the Megamachine but merely administers it.
This is not an indictment. This is a description. The Megamachine acts according to its rules. The question is whether anyone is capable of setting different rules — before the lights go out completely.