I think the big problem with designing human systems is that as systems evolve there is a parallel unpredictable cultural evolution. I guess the only reasonable way to design human systems is through a slow process of evolution and iterative design, at each point trying to modify rules to adapt to the culture that develops around how people use the system as it exists. When a system is fully formed, along with the system rules there are also a raft of social norms that determine how it functions within society. Moral hazards create situations where individuals are incentivised to break the social norms that align with the goals of the system. Once norms get eroded, that behaviour can become commonplace, and you end up with broken systems. Unfortunately, the meta-systems we have in place restrict what changes are possible/incentivised. Also, people tend to have a limited imagination when it comes to solving broken systems usually wanting either a) more rules, b) harsher punishments or c) to get rid of the system entirely. So, if there's a public outcry, it's usually calling attention to a real problem, but calling for an impractical solution.

Personally I don't think the current meta-systems we have in Western capitalist democracies are optimal. I'm sure that better can be done. However, I'm also very sure that we can't do better by building on idealistic principles - because of the massive changes that would entail and the corresponding unpredictability of the results. It's interesting that libertarians and neo-marxists basically want a really extreme change in two directions that almost never happen. The rich tend to always get richer, the government tends to always get bigger in size and scope. I guess this is kind of like the "get rid of it entirely" mentality. Perhaps it would be better to seek out changes to our current system that would allow iterations towards reducing income disparity and shrinking of the government.