I’d add to the other replies by saying that this isn’t just pure inefficiency - it’s the ability to try things and get them wrong without dying. Often several things. If one of them works out, maybe they justify the rest - just like investing in startups, ironically.

Second, it’s also about redundancy, having more employees covers you from key man syndrome where your operations could be adversely affected by an employee leaving. Even if it means you technically have more employees than you need at a bare minimum.

Third, I’d argue that the level of “efficiency” required by a startup simply isn’t sustainable in the long run, unless you want everyone to burn out. Successful companies likely span a spectrum of efficiency, but none of them need to be on the far efficient end like startups do, and that’s better for everyone working there.

Once a company becomes successful, it attracts the second kind of people in Iron Law of Bureaucracy.

https://www.jerrypournelle.com/reports/jerryp/iron.html

It takes effort, requires goodwill, has damaging implications to internal politics and ultimately is not the best way to maximize tenure at C-level roles.

Executives are solving for their job, not for the company's success.

I think the main driver is a lot more simple than other people have said: when you don’t need that efficiency you can focus on resilience instead. As the pandemic showed with JIT supply chains, efficiency and resilience are often at odds. And the bigger you are the more resilience you want to protect your core business and let you just turn the fly wheel that’s making you profitable. That’s also why you see companies start to engage in monopolistic and anticompetitive practices - now that they have the weight they can waste resources on protecting their business instead of just growing it by being more efficient.

This is 100% bad from a country/market level but is what you get when you let local optimizations trump capitalism. In some sense that local optimization also isn’t bad - it’s not uncommon that that behavior is also what lets efficiencies be extracted.

In other words, in many ways all of this is inevitable behavior as things grow and a lot of it are themselves (perhaps inevitable) symptoms of the mechanics at play rather than the mechanics themselves.

In other words, once “founder mode” is documented, you’ll have everyone claiming they’re doing founder mode same as how everyone says they’re now doing agile or whatever other fad you want to pick and you’ll be back at square one until someone then invents “strategic founder mode” or “growth founder mode” and claims to have figured out how you can spot that vs what everyone is just pretending to do right now.

An upper-level employee wants many employees under them. Then they can argue for a higher salary and have a good line on their resume.

Maintaining a good structure is a cost on its own. A good analogy is how most many engineers think "if I spend two days writing documentation then I'll save one hour figuring out things later, resulting in net gain of negative fifteen work hours, great job".

> What are the disincentives to doing so (beyond the need for requiring more from people)?

Working out how to motivate people whose individual impact now has near zero impact on the stock price.