Beyond the Border Wars: Representation, Scale and Multi-Member Districts

Keywords: redistricting, gerrymandering
Abstract Type: Paper Abstract

Abstract

The clustering of partisans creates a tension between traditional districting principles and the metrics that judge fairness to parties. Attention to compactness will tend to pack clustered partisans and lead to undesirable outcomes in measures like the efficiency gap. Maps can be drawn to enhance or diminish this effect, but the tension is a constant of redistricting battles in the U.S. Built into this framework for redistricting is an implicit assumption that a ‘fair’ map will offer proportional representation such that votes for one party translate into seats for that party in the legislature. While partisans may seek more or less power for their party, the principle of proportionality is accepted unproblematically across most non-partisan commentators on redistricting without much consideration of whether parties offer their voters representation.
Significantly, geographers typically understand districting as a two-part problem that includes both zoning (where to draw the boundaries) and scale (how many sub-units to form). Redistricting is typically understood as a zoning problem, but by increasing the size of sub-units we can make districts more heterogeneous and reduce the significance of partisan clustering. By adjusting scale strategically, we can not only limit the potential for maps to be drawn for partisan advantage we can also create voting scenarios with multi-member districts that incentivize representation not just proportionality. This paper proposes three-member districts for Pennsylvania with some form of Ranked Choice Voting as a possible model for addressing both gerrymandering and representation at the same time.