This article of mine is now live on the Columbia Law Review website. Here’s the abstract:

The law of redistricting is built on the assumption that tradeoffs among line-drawing criteria are pervasive. This view helps explain crucial elements of partisan gerrymandering, racial vote dilution, and racial gerrymandering doctrine. This Article is the first to rigorously analyze the existence and extent of redistricting tradeoffs. The Article relies on ensembles of billions of district maps generated randomly by cutting-edge computer algorithms. These ensembles cover all electoral levels for seven priority states as well as congressional maps for all states with two or more U.S. House districts.

The Article finds that, contrary to the conventional wisdom of courts and scholars, redistricting tradeoffs are generally weak to nonexistent. In most cases, progress along one dimension (like compactness, partisan fairness, or minority representation) requires no regression along another axis. This conclusion has sweeping implications for redistricting law and policy. Legally, it bolsters plaintiffs alleging partisan gerrymandering or racial vote dilution, because their objectives can typically be achieved without sacrificing other goals. In policy terms, the usual absence of tradeoffs means that line-drawers can often have it all—maps that simultaneously comply with traditional criteria, treat the major parties fairly, lead to competitive elections, and properly represent minority voters.