Suchergebnisse

In one of the earliest attempts to examine the effect of a priori voting power on actual political phenomena, Riker (1959) looked at changes in party affiliation in the French National Assembly in 1953–54, and used these data to test the hypothesis that deputies who switched parties were seeking thereby to increase their a priori voting power. His findings were negative, or at best inconclusive.

We explain the meaning of a priori voting power and outline how it is measured. We distinguish two intuitive notions as to what voting power means, leading to two approaches to measuring it. One conception, I-power, focuses on a voter's potential influence over the outcome of decisions by a voting body. The second conception, P-power, focuses on voters' payoff, their expected share of a fixed winning 'prize'. We discuss and rebut some philosophical and pragmatic objections, according to which a priori (as distinct from actual) voting power is worthless or inapplicable.

These are comments on the Symposium `Power Indices and the European Union' in the July 1999 issue of this Journal. We point out several common inter-connected confusions and errors concerning the meaning of voting power. We stress the vital distinction between two different intuitive notions of voting power. We emphasize the need for a unified approach to the study of a priori and actual voting power. We show that the family of `strategic' measures proposed by some of the participants in the Symposium are a natural generalization of the Banzhaf measure