I do know that the way to design complicated, successful systems is to start with simple, successful systems and build on them: starting to design a complicated system from scratch seems like a recipe for much bigger problems, and that's what you'd have to do after "burning it all down". That's even assuming that "build something new" was the second step implied after "burn it all down", though you'll note that was not stated in the original comment.
I think we can agree that "burn it all down, and that's it" is the worst possible solution.
Adding problems disguised as solutions is exactly the issue I'm speaking to. You don't actually satisfy needs in this model, you just end up chastising your subjects with "look at what we've already given you, shut up and take it because there are others who need help too". This incites division and factionalism because you've constructed a market which forces people to compete against each other for charity.
Charity is injustice, as Chris Hedges says, because charity demands a cost from the recipient. Beyond this competition, we also can be confident acknowledging that charity develops dependence, which is again the opposite of sustainable self-determinism.
Painstakingly, methodically, improve it, taking care at each step that you're making it better. :)