There's confluence of factors that naturally results in power law (or pareto's law) distribution. The direct consequence is "stuff largely controlled by small

There's confluence of factors that naturally results in power law (or pareto's law) distribution. The direct consequence is "stuff largely controlled by small group". How to mitigate that? It is IMO understudied area of economics because economists prefer thinking of "value" of anything as a scalar, not as random variable. In cases where it isn't possible they just conveniently assume normal/gaussian random distribution. (See @nntaleb)