In a high trust society people aren't trying to con you but they're still subject to structural constraints. When Alice has a good idea but Bob has a monopoly, Alice isn't allowed to implement it and Bob doesn't understand it because he's not smart enough to rather than because he's crooked. When a monopolist has a stable source of revenue their management is living in a world where money is available and they start thinking up ways to "help" by spending it, which raises costs even when it's well intentioned and then harms the poor who now have to pay a premium to receive a necessity.
Structural constraints aren't exclusively a problem because they enable corruption. There can be incompetence without malice. All it takes is for someone to be well-intentioned but incorrect, or for the system to allocate rulemaking authority to someone without the information necessary to make the best decision.
The classic example is a large hierarchical organization where the line workers have knowledge of the process but no management competence and correspondingly don't know what information is even relevant to pass up the chain, and the top brass have management training but have been put in charge of so many people that they don't have the bandwidth to know what's actually happening on the ground.
I am aware and mostly agree with the dogma about how decision-making power should be delegated to those with the most relevant information to that decision. In fact, that to me seems a bit tautological and beside the point.
My point is essentially that there are situations where a monopoly is the best system available especially if you can rely on those involved to act in good faith. For example, I think the government holding a monopoly on policing power is the best available system. Within that monopoly power can be delegated a whole manner of ways, but ultimately the public good of policing is distributed through a manner of sub-institutions all answerable to a single institution.
I think the real delta in our views is I see the values of the people within a system as of equal or more importance to the structure of the system itself. It seems to me that you reason about this topic in an almost purely abstract manner, attempting to posit a "perfect" system which could work no matter who was actually implementing it, and indeed would be the best no matter who was actually implementing it. I used to think this way, but came to the conclusion that it was impossible.
Actually it very literally does.