The Long Game, Lost

Why Bosch and ZF are failing, Zeiss is not — and what Mondragón's Fagor disaster teaches us

The Idea

The foundation-owned company is a German invention — and one of the more interesting answers to the question of who a company should belong to. Not the founder. Not the shareholders. Not the state. But a purpose.

Ernst Abbe formulated this idea first. The physicist and co-founder of Carl Zeiss came from humble origins and had a precise sense of the injustice embedded in capital ownership: why should the person who happens to own something determine everything — and receive everything? In 1889, Abbe transferred his shares in Carl Zeiss and Schott to a newly founded foundation. The statutes he carefully drafted established two rules: the shares may never be sold, and the foundation's purpose is the long-term security of the companies and the promotion of science.

Robert Bosch followed in his footsteps, for different reasons. He was not an academic but an entrepreneur, and his motivation was not theory but character. His 1917 will contained a phrase that said everything: "Half-starved rather than fully humiliated." The company should not be sold after his death, not broken apart, not handed to the first bidder. In 1964, after his death, the Robert Bosch Foundation was established. It holds 92% of the capital shares. Voting rights are exercised by a separate industrial trust — to prevent the foundation from being drawn into operational decisions.

ZF Friedrichshafen has a different, less idealistic origin. The Zahnradfabrik was founded in 1921 at the initiative of the Graf Zeppelin Foundation — as an industrial operation designed to generate dividend income for the foundation and the city of Friedrichshafen. 94% of ZF's shares are held today by the Zeppelin Foundation, which belongs to the city. This is not Abbe and not Bosch — this is municipal economic policy dressed in foundation clothing.

Mahle GmbH — third in the trio of Germany's major automotive supplier foundations — was founded in 1920 and converted to a similar structure in 1964.

The Promise

What do these structures promise? The long game. No quarterly pressure, no activist shareholders demanding short-term returns. No hostile takeover. No sale to a private equity firm that extracts value and disappears. Profits largely remain in the company — for investment, for reserves, for the foundation.

The promise was real. It held for decades. Bosch survived wars, currency reforms, and economic crises without ever seriously wavering. Zeiss survived the division of Germany into two states, reunification, and decades of political turbulence — and kept delivering. Foundation companies were regarded as bedrock in the German economic miracle. Their employees trusted them. Society trusted them.

Three Patterns of Failure

Then came the crisis. Not the crisis of the foundation model — the structure remains intact. What came was a crisis of strategy. And here the paths of Germany's foundation companies diverge dramatically.

Pattern one: Monoculture

Bosch generates nearly two thirds of its 90 billion euros in revenues from the automobile. ZF generates almost everything from it. So does Mahle. For decades this was no problem — the automotive market grew, margins held, market positions were secure. Why diversify when the core business runs?

Now it does not run. The electric vehicle transition makes classic core products — diesel injection pumps, transmission oil, combustion control units, conventional gearboxes — progressively obsolete. Chinese suppliers are entering the market with dramatically lower prices. Unit volumes fall, factories run below capacity. Bosch is cutting 22,000 jobs by 2030. ZF plans 14,000 layoffs, likely more. The EBIT margin in Bosch's mobility division stood at 3.8% in 2024 — the 7% target is far out of reach.

Betting everything on one technology means losing everything when that technology shifts. This is not a property of the foundation model. It is a strategic decision — made over decades and never corrected.

Pattern two: Debt-financed expansion

ZF provides the sharpest lesson. In 2015, the Friedrichshafen foundation company acquired the American supplier TRW for roughly 7 billion euros — at the time the largest acquisition in the German automotive supplier industry. In 2020, Wabco followed for another 7 billion. Both times: debt-financed.

A foundation company cannot issue equity. It has no access to the capital market. The solution for growth through acquisition is therefore: credit. Today, ZF carries liabilities exceeding 10 billion euros. Interest payments and amortization are pushing the company into loss territory — both in 2024 and in the first half of 2025. Revenue fell 10% in the first six months of 2025.

The paradox is clear: exactly what is supposed to protect the foundation company — independence from capital markets — became its undoing. Without equity financing, debt was the only path to acquisitive growth. In good times, manageable. In crisis, a death spiral.

Pattern three: Wrong technological bets

Bosch invested billions in hydrogen technology and fuel cells — as a strategic answer for its former diesel production sites. Demand did not materialise. The hydrogen hype faded. Infrastructure was missing. Costs remained too high. The investments were not recovered.

This is not a moral failure. It is a strategic one — and, more deeply, a structural one. When a company has no external owner demanding accountability, it often lacks anyone to raise uncomfortable questions in time. The foundation model protects against activist shareholders. In doing so, it also protects against critical ones.

Zeiss: the Counter-Model

Carl Zeiss is also a foundation company. The same structure: no stock market listing, no external shareholders, profits remain in the company. The Carl Zeiss Foundation — the same one Abbe established in 1889 — holds 100% of the shares.

And yet: in fiscal year 2024/25, revenues rose 9% to 11.9 billion euros. The EBIT margin stands at a constant 13%. 15% of revenue flows into research and development — a proportion that would have been compressed by quarterly targets long ago if Zeiss were publicly listed. Zeiss grows while Bosch and ZF shrink.

What is different? Not the model. The strategy. Zeiss is the global market leader in precision optics for semiconductor manufacturing. Its Semiconductor Manufacturing Technology division grew 23% in 2024/25 and crossed the 5 billion euro revenue threshold for the first time. The AI boom drives chip demand — and therefore demand for Zeiss lithography optics — to record levels. Whoever wants to manufacture EUV chips cannot avoid Zeiss.

But this was not a fortuitous accident. Zeiss consistently invested in niches where technological depth is harder to replicate than volume production. The foundation model made this possible — no shareholder would have had that patience. Zeiss demonstrates what the foundation model can be. Bosch and ZF demonstrate what it does not automatically guarantee.

A further advantage: the Carl Zeiss Foundation holds two companies — Zeiss and Schott. It is not hostage to a single business model. If Zeiss were to encounter serious difficulties, Schott exists. The dependence on a single technology cycle is at least halved.

Fagor: the Cooperative Parallel

What is happening to Germany's foundation companies today, Mondragón already experienced in 2013. With Fagor Electrodomésticos.

Fagor was not just any company. It was the first — the original cooperative, founded in 1956 by five graduates of Arizmendiarrieta's school, the starting point of the entire Mondragón network. From Fagor came Caja Laboral, Ikerlan, the Mondragón University. Fagor was the nucleus from which everything else grew.

And Fagor made exactly the same mistake as ZF: debt-financed international expansion into a saturated mass market. In 2005, Fagor acquired the French household appliance manufacturer Brandt for 162.5 million euros. Revenue nearly doubled. Market share in France rose to 14%, in Spain to 16%. Fagor became the fifth-largest household appliance manufacturer in Europe. Everything appeared to be working.

Then came 2008. The financial and property crisis hit Spain harder than almost any other country in Europe. Consumer demand collapsed. In the first half of 2013, Fagor's revenue fell 19% to 491 million euros — with a loss of 60 million euros. Debt stood at 800 million euros. On 17 October 2013, Fagor filed for insolvency.

"The company does not correspond sufficiently to the needs of the market. Even if the money were available, it would not guarantee Fagor's future." — Mondragón Corporation, October 2013

That was the decision. Mondragón had paid tens of millions to support Fagor — then stopped. The reasoning was measured and correct: it was not the model that had failed, but a company that had abandoned its own logic. Pouring in capital that would not return risked damaging other cooperatives.

Most Fagor employees were absorbed into other Mondragón cooperatives. The solidarity fund — to which all cooperatives contribute 2% of their profits — cushioned the transition. It was painful. It was the first time Mondragón could not entirely avoid real layoffs. But the network survived, continued growing, and is today stronger than before the Fagor crisis.

Mondragón had the capacity to say no. And it said no — to the oldest cooperative in the group.

That is the distinction on which everything turns.

Governance Without Accountability

The foundation company has a blind spot. It lies not in the ownership structure but in the control structure.

In a publicly listed company, there exists — despite all its shortcomings — a mechanism of external accountability: the market punishes bad decisions through falling share prices. Activist investors ask uncomfortable questions. The supervisory board is answerable to shareholders. A general assembly can vote out management.

At Bosch, none of this exists. The foundation holds the shares. The supervisory board of Robert Bosch GmbH partly consists of the same individuals who sit in the industrial trust — the body that exercises the foundation's voting rights. Six trust members sit simultaneously on the supervisory board. Management is overseen by a body whose key members share the same institutional interests as management itself.

This is not malice. It is structure. And the structure asks: who controls whom?

When a company spends 7.4 billion euros on an acquisition while simultaneously cutting 22,000 jobs — who reviewed those decisions? Who asked: is this the right priority? Is this a sound use of capital? Who demanded accountability when the 7% return target has been missed for years running?

At a public company: the market, the general assembly, the supervisory board. At Bosch: no one outside the system that the system was supposed to control.

The Sentimental Illusion

Bosch workers have carried placards reading "Robert, come back!" At vigils outside the Feuerbach plant the message read: "Here lie the Bosch values." Drawings showed Robert Bosch confronting today's CEO Stefan Hartung: "What have you done with my company?"

This is humanly understandable. It is analytically wrong.

Robert Bosch was not a socialist. He was an entrepreneur who treated his workers fairly — because he understood that this was better for the company in the long run than exploitation. He introduced the eight-hour day, holiday pay, and social benefits long before they were legally required. But he did this to secure the company — not out of altruism that ranked above corporate interest.

The appeal to the founder is a flight into mythology. The founder becomes an imaginary protective authority: as long as his spirit presides, nothing bad can befall the company. That is sentimentality, not argument. Robert Bosch, faced with the same economic pressures today, would likely have made similar decisions — probably faster and more decisively.

What the workers are mourning is not a betrayal of the founder. It is the absence of a governance structure that makes bad decisions visible — and corrects them before they become irreversible.

ZF: the Municipal Hostage

At ZF the constellation is more complicated and politically more explicit. The Zeppelin Foundation, which owns 94% of ZF's shares, belongs to the city of Friedrichshafen. Dividends flow into the municipal budget — funding swimming pools, cultural institutions, social services. Friedrichshafen is, for a city of its size, extraordinarily well resourced. It owes this to ZF.

This creates a dependency that distorts strategic decisions. When ZF should be building capital reserves, that capital is missing from the city budget. When ZF finances acquisitions with debt because it cannot issue equity, the city bears the consequences. When ZF needs restructuring, the political costs for local municipal government are immense.

No mayor of Friedrichshafen steps before residents and says: "I have decided today that ZF will pay less in dividends so the company can build up equity — therefore we are closing the municipal pool." That is not politically feasible. So it is not done. So ZF borrows instead.

This is not a management failure. It is a structural conflict of interest, built into the ownership arrangement. The foundation that is supposed to secure the company is itself a political institution — with interests that can directly contradict the company's interests.

The Fagor Paradox: Whoever Can Say No, Survives

Mondragón let Fagor fall. That took courage. It was correct. And it reveals precisely what German foundation companies lack.

If Bosch were to face genuine existential crisis — who would say no? Who would decide: this division, this plant, this subsidiary — we release it, so the rest can survive? Who would have the institutional authority to hold that decision against all political resistance?

At Bosch: the foundation has no operational decision-making power. The industrial trust is not a crisis body. The supervisory board is — as shown — structurally not independent. The union is interested in job preservation, not corporate restructuring. The federal government would face overwhelming political pressure to intervene.

At Mondragón: the structure is built so that no single company is too big to fail — not even the oldest one. The solidarity fund absorbs the human cost. The annual assembly of all cooperatives can make collective decisions about capital allocation. And when money is insufficient to rescue a company without endangering others — it is let fall. Not cruelty. Rationality in service of the whole.

"If the group itself does not believe in a future, the Basque government will not do so either." — Arantza Tapia, Basque Minister of Economy, November 2013

The Fagor paradox is this: Mondragón let its founding mother die — and grew stronger for it. German foundation companies are not capable of an analogous act. The institutions are not built for it. The political environment does not permit it. The sentimentality problem compounds the rest.

What Is Actually Missing

The debate about foundation companies usually revolves around capital market access. Do Bosch and ZF need the stock exchange? Can a partial listing mobilise at least some capital? These are legitimate questions. They are not the decisive ones.

What is actually missing is three things.

First: genuine external accountability. Not shareholder meetings with short-term return demands — but also not the pseudo-oversight of a supervisory board whose members share the same interests as management. An independent body that makes strategic errors visible before they become irreversible. This could be an external foundation audit, an independent expert panel, a mandatory peer review by companies in the same sector — anything that is not part of the system it is supposed to control.

Second: clear decision rules for emergencies. Mondragón has such rules — embedded in individual cooperatives and across the network. At what level of losses is restructuring triggered? At what point is a division divested? Who decides, with what mandate, against what opposition? At German foundation companies all of this is undefined — and in emergencies it is overridden by political pressure.

Third: the capacity for orderly failure. This is the most uncomfortable point. Companies can fail. Foundation companies can fail. Even if the founder was admirable and the employees are decent people. The question is not whether this can happen — it is whether the system is built to limit the damage when it does. Mondragón proved this: Fagor is gone, Mondragón lives. The solidarity fund, the flexibility of worker allocation, the clear decision structure — these limited the damage.

German foundation companies have none of these three elements. They have the idea — and that is much. But ideas without mechanisms are appeals. And appeals do not change the world.

The Open Question

The foundation company is not a bad model. Zeiss demonstrates this every day. The underlying idea — a company that belongs to no one except its own purpose — is correct at its core.

But the idea does not protect against monoculture. It does not protect against debt-financed acquisitions. It does not protect against wrong technological bets. It does not protect against a governance structure that makes errors visible only once 22,000 jobs are gone.

Mondragón lost Fagor. That was painful. But Mondragón learned from the pain, strengthened its structures, and in 2024 recorded record earnings.

The question is whether Bosch, ZF and Mahle have the capacity to do the same. Not: summon the founder's ghost. But: rebuild their own structures so that they protect what the founder actually wanted — not the name, not the sentiment, but the principle.

"Half-starved rather than fully humiliated." — Robert Bosch, 1917

Humiliation is not job cuts. Humiliation would be: disguising the inability to self-criticise as loyalty to the founder — and watching the company slowly die in the name of a man who would not have tolerated that for a moment.