Siemens: How Breaking Itself Apart Made the Conglomerate More Valuable

Siemens is the most successful conglomerate dismantling in European industrial history, and it was executed over twenty years rather than in a single dramatic transaction. Understanding how it was sequenced is more useful than admiring the outcome, because most portfolio simplification programmes fail at exactly the points Siemens navigated. This case study opens the industrial pillar of the Germany Company Stories hub and connects to the hidden champions model that dominates the tier below it.

What was the conglomerate problem Siemens was solving?

A discount. A group operating in semiconductors, lighting, telephony, power generation, trains and medical imaging is valued by the market at a blended multiple that is lower than any of its good businesses would achieve alone, because investors cannot allocate to the part they want.

The discount is not irrational. A diversified industrial group forces investors to accept exposure they did not choose, and internal capital allocation across unrelated divisions rarely beats the market at deciding where money should go. Cash generated by a strong division is frequently used to defend a weak one, which destroys value twice.

There was also a management bandwidth problem. Businesses as different as semiconductor fabrication and rail signalling have almost nothing in common operationally, and a single executive board cannot hold deep domain judgment in all of them simultaneously.

What made Siemens unusual is that it acted on the diagnosis. Most conglomerates defend the structure with synergy arguments until a crisis forces the issue, by which point the assets are worth less and the seller has no leverage.

Why give up control of Healthineers rather than sell the stake?

Because selling a stake of that size would be both tax-inefficient and market-destructive. Siemens announced a plan to transfer roughly thirty per cent of Healthineers shares directly to its own shareholders, reducing its holding from around two thirds to a significant minority, and signalled an intention to reduce further toward a financial holding over the medium term.

The mechanics matter. Distributing shares as a dividend in kind would trigger a substantial tax charge; a direct spin-off structured as a corporate separation does not, and Siemens explicitly identified the spin-off as the preferable route. Selling tens of billions of euros of stock into the market would also depress the price of exactly the asset being realised.

The strategic case is capital allocation flexibility. A controlling stake in a medical imaging company consumes an enormous amount of Siemens's balance sheet while offering limited operational synergy with factory automation software.

The secondary benefit is governance. Deconsolidation removes a large minority interest line from the accounts, simplifies reporting and lets each company pursue its own capital structure and acquisition strategy without reference to the other.

What is Siemens actually trying to become?

An industrial technology company where software carries the margin and hardware carries the installed base. The strategic thesis is that the value in automation is migrating from the physical controller to the software layer that designs, simulates and operates the plant.

This is why the acquisitions moved in the opposite direction to the divestments. Siemens has spent heavily on industrial software, simulation and engineering design capability, building a portfolio intended to cover the full lifecycle from product design through production execution.

The demand case rests on two trends: electrification of infrastructure and industrial artificial intelligence applied to real production environments. Both favour a company with deep domain data from installed equipment, which is an asset that pure software firms cannot easily replicate.

The risk is execution and multiple. Guidance for the 2026 financial year pointed to comparable sales growth in the mid-to-high single digits alongside currency headwinds and industrial market uncertainty, which fell short of expectations after the shares had already re-rated on artificial intelligence optimism.

Did the separated companies actually do better alone?

Mixed, and the pattern is instructive. Infineon became a substantial independent semiconductor company. Healthineers has performed strongly since its listing. Osram struggled independently and was eventually acquired. Siemens Energy went through a severe crisis over wind turbine quality problems before recovering sharply on grid demand.

The common factor in the successes is that the separated business had a defensible position in its own market and enough scale to fund its own development. The failures were businesses that had been sustained by group balance sheet rather than by competitive position, and separation exposed that immediately.

This is the honest caveat on conglomerate breakups. Separation does not create value in a weak business; it reveals the weakness faster. The parent benefits either way, because it stops funding the problem, but shareholders who receive the spun-off shares do not.

For an investor receiving spin-off shares, the practical question is whether the business was separated because it was good enough to stand alone or because the parent wanted it off the balance sheet. The two look identical in the announcement.

How does codetermination shape a breakup of this size?

It slows it and makes it more durable. Every separation at Siemens required negotiation with works councils and supervisory board employee representatives, since spinning off a division changes employer identity, collective agreements and pension arrangements for tens of thousands of people.

The practical effect is that separations are structured with employment guarantees, site commitments and transition periods attached. That raises the cost and reduces the speed relative to what a purely financially driven board would achieve.

It also raises the completion rate. A transaction negotiated with labour representatives in advance rarely collapses at the point of execution, and it does not generate the industrial conflict that has derailed restructuring elsewhere in German industry, as the Volkswagen case demonstrates.

The governance mechanics behind this are examined in the codetermination and governance pillar of this hub.

What should a diversified group take from the Siemens playbook?

Three things. Separate on the basis of customer and technology logic rather than on financial performance, sequence the separations so each one funds and de-risks the next, and use spin-offs rather than sales when the asset is genuinely good.

The sequencing point is the least appreciated. Siemens did not announce a twenty-year plan; it made one separation, absorbed the organisational cost, and then made the next. Each transaction taught the company how to carve out shared services, disentangle information systems and negotiate pension splits, and those capabilities compounded.

The customer logic point matters because the standard test for divestment is margin, which is the wrong test. A low-margin business that shares customers, channel and technology with the core may be worth keeping; a high-margin business that shares nothing is a financial holding pretending to be a division.

For mid-sized industrial groups, the transferable version is narrower: audit which divisions genuinely share engineering, purchasing or customers, and treat the rest as a portfolio to be exited over years rather than defended indefinitely.

What did the Altair acquisition and the software strategy actually buy?

Simulation. Siemens has assembled a portfolio covering computer-aided design, product lifecycle management, electronic design and, with the Altair acquisition, physics simulation and high-performance computing, which together allow a customer to design, test and optimise a product digitally before anything is built.

The strategic logic is that simulation is where industrial artificial intelligence becomes economically useful. Applying machine learning to a factory requires a physically accurate model of that factory, and the model is worth more than the algorithm because it cannot be recreated from public data.

This positions Siemens against enterprise software firms with far larger platforms but no domain physics, and against engineering software specialists with domain depth but no installed industrial base. The claim is that owning both is the defensible position.

The financial question is margin mix. Software carries higher margin and higher multiples than automation hardware, so the more revenue that shifts to software and recurring licences, the more the group's valuation should converge on technology rather than industrial comparables. That re-rating is the implicit target of the whole strategy.

What are the risks in the remaining Siemens?

Cyclicality and concentration. Automation and smart infrastructure are both capital expenditure businesses, and short-cycle automation orders are among the earliest indicators of an industrial downturn, which the 2026 guidance for currency headwinds and industrial market uncertainty reflected.

The second risk is that industrial software growth does not compensate. Software is a smaller revenue base growing from a smaller number, and it takes considerable time for a mid-single-digit software growth contribution to offset a decline in the hardware business that funds it.

The third is that the group is now more exposed to a small number of secular themes. Electrification, data centre build-out and factory automation are all currently strong, and all are capital cycles that will eventually turn together.

Against that, the balance sheet is stronger and simpler than at any point in the company's modern history, which is the point of the whole exercise: a focused business with a clean balance sheet can absorb a downturn without being forced into a defensive divestment.

Does the breakup make Siemens an acquisition target?

Not realistically, on size alone. What it does change is the character of the shareholder base: a focused industrial technology company attracts investors with a different mandate than a diversified conglomerate did, and those investors are more demanding about capital allocation.

That is the intended consequence. A shareholder register that understands the business exerts useful pressure on strategy, whereas a register holding the stock for diversification exerts none.

Frequently Asked Questions