determine the optimal level of centralization for a particular variety of intervention
I don't see how you can get to the single "optimal level" without assuming a single underlying system of values. Optimality is a function of your evaluation of outcomes and your evaluation of outcomes is a function of your value system.
Who will determine that "optimal level" and what happens when there is significant disagreement about optimality?
A post from Gregory Cochran's and Henry Harpending's excellent blog West Hunter.
The commenter Ron Pavellas adds:
The Wasserman Test.