But if the founder gets hit by a bus it's going to drop to 2% or die. Professional management is like a lot of software tools, you can build a better framework yourself that is better suited to your usecase but no one else knows how to use it or can learn in a reasonable amount of time.
This isn't saying management is always the right route. A founder can lead for 30+ years, but they could also quit or get hit by a bus tomorrow so its not always irrational for shareholders to pick the "less efficient" option here