A power index for multistage and multiagent decision systems

An index of power is proposed, based on the concept of the entropy of information, which measures the distribution of power among decision makers who commit themselves at distinctive stages in the hierarchical decision systems. The index is tested and is shown to work well with a model of public decision-making in an indirect democracy. Moreover, it is shown that the entropy decreases monotonically in hierarchical and sequential decision systems. This property indicates support for the applicability of the index.