The “efficiency gap” measures Gerrymandering’s insidious effect of on democracy’s aspiration of “one person, one vote”.
Quoting the WSJ article:
The metric devised by Mr. Stephanopoulos and Eric McGhee, a research fellow at the nonpartisan Public Policy Institute in California, tests whether district plans treat political parties equally by comparing the number of votes each wastes.
Wasted votes are those cast for a losing candidate as well as those cast for a winner in excess of the number needed for victory.
The efficiency gap divides the difference between the two parties’ wasted votes by the total number of votes cast. The result is a percentage that indicates the portion of seats the dominant party won because it wasted fewer votes than the other side.
The authors offer a simple example:
Imagine 1,000 voters spread across 10 election districts. Party A collects 55% of the votes, wins eight seats and wastes 150 votes. Party B collects 45% of the votes, wins two seats and wastes 350 votes.
Party A wasted 200 fewer votes than Party B, or 20% of the 1,000 votes cast, indicating that it won 20% more seats than it would have if the district plan was unbiased, or two extras seats.
The question, then, is whether that degree of bias rises to the level of an unconstitutional partisan gerrymander.
To decide, the authors propose setting a threshold above which a district plan would trigger scrutiny then testing to confirm it had no neutral justification, gave a persistent advantage to the dominant party and that its effect was intentional.
They suggest flagging congressional plans that help the majority party win two or more additional seats and state legislative plans that have an efficiency gap of at least 8%.
While the efficiency gap is better than previous attempts to measure partisan gerrymandering, it isn’t perfect. It isn’t suitable for states with a small number of congressional districts, such as Nebraska, because flipping a single seat would cause the gap to move by a large percentage. (The authors applied the formula in states with eight or more districts.
It also builds in a bonus for the majority party by awarding it twice as many seats after winning 50% than if it were simply proportional, a feature that is easier to detect when the formula is reduced to an equation:
(seat margin – 50%) – 2 x (vote margin – 50%) = efficiency gap
Or, in the authors’ simple example: (80% – 50%) – 2 x (55% – 50%) = 20%
The bonus allows the majority party to win more seats before meeting the threshold of an unconstitutional gerrymander. The authors say the formula reflects the way elections have worked out historically, but it disturbs people who don’t want that cushion locked into a legal definition of partisan gerrymandering.
“We don’t see a good argument for that factor of two,” said Moon Duchin, a Tufts University math professor who has organized five upcoming workshops on using mathematics and technology to draw nonpartisan congressional and legislative districts.
The Supreme Court is scheduled to hear arguments in the Wisconsin case in the term that begins in October.
