I want more companies to try something: take a random guy from your existing non-management workers, pay him a nice but not exorbitant wage ($100k?), and make h

I want more companies to try something: take a random guy from your existing non-management workers, pay him a nice but not exorbitant wage ($100k?), and make him CEO. Then see if he can really do any more damage, or make any poorer decision, than one of these hot-shot CEOs. Seriously, what do they have to lose? Some CEOs seem to screw up everything imaginable and leave in 6 months; isn't 6 months of $100k (i.e. $50k) worth the experiment instead?