This proposal is a simple, appealing way of solving the problem, but I doubt it would work well over time, because, like all efforts to control human behavior via a simple set of rules, it will end up being gamed by human beings.
Campbell's Law: "The more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor."[0]
Ultimately, I believe, we'll need human moderators to (1) detect and prevent abuse of the rules, and (2) modify the rules when they stop working.