Majority-Win Percentages: An Approach to the Votes-Seats Relationship in Light of Davis v. Bandemer
Richard Niemi and Stephen G. Wright
Initial reactions to the Supreme Court's ruling in Davis v. Bandemer (106 S. Ct. 2797 ( 1986)) suggested tortured logic and indecision. Witness, for example, the Toles' cartoon, which showed a snakelike sentence proclaiming that "gerrymandering is unconstitutional, in some cases, even when districts meet one-person one-vote requirements," and so on for nearly a dozen more clauses ( Buffalo News, July 6, 1986). In fact, the decision and accompanying arguments may be less convoluted than they first appeared to be. As justice White noted ( Davis v. Bandemer, p. 2805), the 1962 case of Baker v. Carr only determined justiciability and certainly did not settle all the questions of how to measure population equality and what deviations from exact equality were permissible. When we consider the greater complexity of the gerrymandering issue, it is not surprising that much was left undecided by the Bandemer ruling. Moreover, there are strong clues in the plurality opinion as to what is needed to demonstrate an impermissible gerrymander.
In this paper we draw on the plurality opinion written by justice White to suggest the kind of work needed to show (an expectation of) the "continued frustration of the will of a majority of the voters" ( Davis v. Bandemer, p. 2811). Our approach is analogous to what in the political science literature has been called swing ratio analysis, but it is designed specifically to address major points made by the plurality.