Hey, I appreciate your insight. Especially your observation that when the underlying assumptions are wrong/broken then the model produces less reliable results.

Like you, I also don't know what would work better, nor do I believe any one individual can know.

But I do have some ideas for what would make a good framework for the evaluation?

If the idea is to allocate resources in a way that provides the most benefit to the most people, where most feel they are getting a 'fair deal' or something...

and we have social institutions that convert 'resources' to value (in quotes because time, attention, etc are 'resources'. The key principle is organizing human behavior over time to produce something humans value)...

Companies Religion Sports Government

then think about what value each creates, how it is delivered, how it is captured, ... recognizing that each offers some unique strengths and unique limitations.