Mercedes-Benz and the Protection Machine

The same letter took something away. About 90,000 of the company's roughly 108,000 German employees had been due a special payment in July worth 18.4 percent of a month's base pay, and the board pushed it into next year, keeping the right to cancel it. The profit share had already fallen from up to 7,300 euros for 2023 to 3,139 euros for 2025.

A week later around 20,000 people stood in the yard at Sindelfingen, the company's biggest plant, south-west of Stuttgart, and chanted "Ola raus," Ola out.

Companies ask their staff for concessions all the time, and Mercedes has real cost problems that a better brand campaign won't fix, so the demand itself didn't surprise me. What stayed with me is the diagnosis underneath it, the idea that the productivity and cost of each hour are what's broken, because Mercedes has already tested that diagnosis on itself for seven years.

Seven years is a full trial. When you've pulled the same lever for that long and the number you wanted to fix has lost two thirds of its value, one more pull won't change the result, and I'd bet most of the people who signed the June letter know it.

They pulled it again anyway, because it's the lever their seats at the table entitle them to pull, and it feels like the only one they know how to use.

Much of German industry takes its strategy from consultants and top-school graduates who all learned the same manual, the one in which cost comes first because cost is what a spreadsheet can model. The people I'd want at that table are the ones who have watched a playbook stop working in their own business, and were willing to try what it didn't cover, or to write a new playbook themselves.

Whatever the car division is missing, you won't find it on the payroll.

The last ten years produced real firsts too. In 2022 Mercedes became the first carmaker to get a Level 3 system approved in Germany, a car that drives itself in slow motorway traffic while the driver is legally allowed to look away, and in January 2023 it certified the first one for the American market, in Nevada. The EQXX prototype drove more than a thousand kilometers on a single charge. Both were serious engineering, and neither has changed what a buyer expects when he walks into a showroom.

The best proof of what Stuttgart can still do is on sale now. The electric CLA carries much of what the EQXX taught onto the road, and in the Ecotest of the ADAC, Germany's largest motoring club, the 250+ version achieved a range of 740 kilometers on a charge, further than any other electric car the club had measured. It used 13.4 kilowatt-hours per hundred kilometers, and the next best, Hyundai's Ioniq 6, needed 15.5.

It's built in Rastatt, one of the German plants whose models Kecskemét is now set up to take over.

So Mercedes leads at something again, and the lead is real. It's a lead in a race whose rules were written elsewhere, though. Porsche and Hyundai had 800-volt charging on the road years earlier, and in March 2025 BYD showed cars that charge at up to 1,000 kilowatts, about three times the CLA's peak, and add roughly 400 kilometers in five minutes. The same testers who praise the Mercedes mark it down for a cramped back seat, driver assistance that drops out and a price of almost 56,000 euros.

The CLA is the best version yet of a car buyers had already learned to want. That's a serious achievement, and it belongs to a different category from ABS in 1978, which gave buyers something they hadn't known to ask for.

Since 2010 Mercedes has sold itself with a line it traces back to Gottlieb Daimler, one of its founders: "The best or nothing." It's a good line, and it holds only as long as the buyer can feel the best for himself, in the car, every time he drives it. The star on the hood carries what the brand promised for a century. Once the driver stops feeling that promise, the star goes dark, however carefully it's polished.

Even the line has started to wander. At Rétromobile in Paris this January, where Mercedes was celebrating 140 years of the automobile, it advertised with "Don't dream it. Drive it." Jaguar used those four words on and off for decades. A company with 140 years of its own history went to a competitor's drawer for its birthday line.

Nine months later, outside Waco in Texas, Tesla showed a Roadster that hops over the rest of its lineup on compressed-air thrusters borrowed from SpaceX. It's a stunt. The car has been promised since 2017, production is meant to start in 2028, and Musk himself says it won't move revenue much. Mercedes doesn't need to become Tesla, and a brand that sells calm and safety shouldn't try. A stunt is still a bet that people will talk about you for something new, though, and Mercedes spent its birthday on somebody else's old line.

A brand can stay profitable for years after it has stopped setting the direction. The balance sheet reports on decisions made a decade earlier, and by the time it catches up, the category has already rearranged itself around somebody else's ideas.

It works this way because of who sits at the top. Mercedes has a supervisory board of twenty, ten seats for the shareholders and ten for the employees, and every employee seat is held by a member of IG Metall, Germany's metalworkers' union. The chairman, always from the shareholder side, gets a second vote to break a tie. Below the board sit works councils at every plant and a collective agreement that turns any change in working hours into a negotiation with the union.

Each of those parties has something concrete to lose if a twenty-year bet on a new architecture fails, and none of them earns anything if it pays off. The union collects no bonus when a new platform succeeds in 2040, and it loses members if the bet fails in 2030. Nobody has to block anything. Everyone at the table simply prefers the number they can calculate today to the future they can't, which is exactly what each of them was put there to do.

The capital side reaches for the same lever. Martin Brudermüller, the chairman with the casting vote, told Handelsblatt in June that Germany should seriously go back to a 40-hour week, and that same month Seeger asked the workforce for more Gewinnermentalität, a winner's mindset. In September the board ended the remote-work agreement, with up to four office days a week from January, and on 21 September Michael Schiebe, the board member for production, said that two German plants would close if costs there didn't come down.

Every move made this year is a way of pricing the hour. What that hour should be building hasn't come up once.

I'll be the first to argue for a winner's mindset. Wanting to win, wanting to play in the Champions League instead of settling for mid-table, is the first thing I look for in a company and the first thing I try to build in one. So I understand why a board under this much pressure asks for it.

What I can't follow is the line drawn from that mindset to the home office, as if the engineer working on battery software at her kitchen table on a Tuesday had stopped wanting to win. Thirty-five hours is less than forty, of course. But if the work done in those thirty-five hours is still good work, and a van division earning 10.2 percent out of German plants suggests it is, then the hours are the wrong place to look.

I don't believe the only inventive engineers in the car business sit at Tesla or BYD. The EQXX showed in 2022 what Mercedes' own engineers can build when they're given the room. My guess, and it's only a guess, is that a good part of the innovation Mercedes is missing has existed for years, in a drawer or on a slide, and was argued down meeting after meeting by people who couldn't defend it with a number. The generals again, fighting the last war with the old manual.

If that's right, the old saying applies, in German as much as in English: the fish rots from the head. At Mercedes the head is more than one person. It's the management board, the chairman, the shareholders, and the works council members too. Where, over the years, did each of them block something that would have moved the company forward, and why? The cost of an hour has a place in that conversation.

It just shouldn't be the only item on the agenda.

ASML, the Dutch company that builds the machines used to print the world's most advanced chips, spent roughly twenty years on extreme ultraviolet lithography, a technology so hard to get working that much of the industry doubted it would ever ship. It made the bet from strength. By then it already led the market for the previous generation of machines, so it was staking a healthy, profitable business on an unproven one.

It could afford to because the risk never sat on its books alone. In 2012 Intel, TSMC and Samsung, its three biggest customers, committed 1.38 billion euros to the research and bought 23 percent of the company for another 3.85 billion. They paid because they had to: if the technology failed, their own chip roadmaps failed with it. On the supply side sat Zeiss, the optics maker from Oberkochen in southern Germany, whose lithography lenses go back to work for Philips in 1983, and Trumpf, the family-owned laser maker from near Stuttgart. Peter Leibinger of Trumpf once called the three of them a "virtually merged company."

Everyone at ASML's table won only if the bet worked.

Mercedes can't build that table. Its customers are millions of private buyers, none of whom goes under if Mercedes picks the wrong architecture, so none of them will co-fund ten years of research to find out. The parties who do share Mercedes' downside are the union, the works councils, the state of Baden-Württemberg and the suppliers around Stuttgart, and each of them is rewarded, in money or in votes, for keeping what already exists.

Every one of those protections was added for a good reason, in decades when progress meant a better version of last year's engine. Together they've become a machine for protecting what Mercedes already has.

In 2013 BMW put the i3 on sale, built in Leipzig, my hometown, around a carbon-fiber passenger cell nobody else in the industry was using, and a year later the i8 followed. Project i had its own budget, its own team and its own production line, under the same co-determination and the same union as Mercedes. A German carmaker can ring-fence a radical bet, and I won't pretend otherwise.

The experiment was allowed to exist. It was never allowed to replace the core, because the people whose consent the core business needs were never going to vote for a side project to take over the thing they're there to defend. Under German co-determination a bet like that survives as a satellite, and the satellite stays small.

In July Mercedes doubled the yearly capacity of its plant in Kecskemét, a town about an hour south-east of Budapest, to 400,000 cars, and the plant is set up to build the same models as Bremen and Rastatt. The company puts its factor costs there, wages, energy and logistics together, at 70 percent below Germany's. The new electric C-Class comes off that line, built by a workforce with none of the guarantees, none of the board seats and none of the history that Sindelfingen carries.

A thousand kilometers north-west, the people who built careers around what the star used to mean are being asked how many hours they'll give for the same pay.

Both are Mercedes, and neither is lying about what the company has become. I don't know which of them will decide what the brand means in ten years, and I doubt anyone on that supervisory board knows either. I do know it won't be settled by a letter about the price of an hour.