Gerrymandering has a bad name because the party in control of a state typically divides the state into House districts that give it more than its fair share of seats in Congress.
Surprisingly, there is a simple solution to this problem: Give the majority party the right to divide the state into two parts according to the proportion of votes it and the minority party received in the last congressional election.
For example, if a state has 10 congressional districts and the majority party wins 60 percent of the statewide vote, the majority party would be able to divide the state 60-40 in population. Its part would be the larger part with six districts, and the minority’s part would have four districts.
Each party would then be able to gerrymander its part. This is “fair gerrymandering” because each party can gerrymander only the districts in its part.
To compare this with how gerrymandering works today, consider the 2014 congressional elections in Pennsylvania in which Republicans won 55.5 percent of the statewide congressional vote.
Because they controlled the state legislature, they were able to gerrymander Pennsylvania’s 18 congressional districts so that they won 13 districts (72.2 percent) and the Democrats won only five in 2016.
With fair gerrymandering, the Republicans, as the majority party, would have been able to divide the state in two, with one part comprising 10 districts and the other 8, which approximates the 2014 vote split.
The Republican gerrymander of its part probably would have garnered it about seven of its 10 districts, and the Democrats about six of its eight districts. Overall, each party probably would have won about its proportion of the statewide vote — a very different result from the 13-5 split favoring Republicans in 2016.
Ideally, of course, it would be preferable to eliminate gerrymandering entirely by having an independent commission draw the district lines.
Laws have been enacted in six states to do this, but it is highly unlikely that the other 44 states will move quickly to reform the districting process. (In Pennsylvania, however, a court challenge by the Democrats led to the Pennsylvania Supreme Court apportioning the state into mostly competitive districts for the upcoming 2018 elections.)
This past June, the U.S. Supreme Court refused to consider partisan gerrymandering cases from two states, Maryland and Wisconsin. It may revisit these or other cases to try to settle the constitutionality of partisan gerrymandering, but in the meantime, gerrymandering remains a fact of political life in most states.
If fair gerrymandering is not a perfect solution, it is a pragmatic one that puts a constraint on the degree to which gerrymandering can be exploited by one party. Details remain to be worked out, such as what flexibility should be given to the majority party in dividing a state in two.
It seems desirable that the majority party should be restricted to drawing a more-or-less straight line that cuts the state in two, proportional to the number of seats to which each party is entitled.
It should, however, be allowed to respect traditional boundaries that do not arbitrarily split jurisdictions or communities. What is arbitrary will need to be spelled out.
Another question concerns the entitlements of minor parties that qualify for one of more congressional seats in larger states. I suggest that after the two major parties receive their parts of a state reflecting their vote shares, other parties would then be permitted to divide the remainder of the state according to their shares.
This is not entirely fair because the Green Party, for example, is unlikely to win in any congressional districts. But “possessing” a district may better enable it to push the major parties toward its positions.
Fair gerrymandering will work better the larger a state is. If a state has only two or three districts, the majority party may be able to divide it so as to win them all. But with four or more districts, the minority party will have more and more ability to gerrymander its part so that it wins a proportional share of districts in the state.
A more sophisticated version of fair gerrymandering, which would require two cuts, applies the principle of “I cut, you choose.” The larger party would make a cut that gives it the extra districts to which it is entitled (beyond 50 percent).
In Pennsylvania, this would be a part that contains two districts (10 – 8), which it would presumably be able to gerrymander so that it wins both.
The smaller party would then be able to divide the remaining districts into two equal parts, which in Pennsylvania would each contain eight districts.
The minority party (Democrats in Pennsylvania) would then choose the part it prefers, and the majority party (Republicans) would get the other part. This would give each party a good chance of winning most of the districts in its eight-district part.
The fact that one party will predominate in its part of the state is not entirely fair to voters of the other party who live in that part. But their votes won’t be entirely wasted because they will count toward how the state is split into districts two years later.
Steven J. Brams is professor of politics at New York University and author of, “Mathematics and Democracy: Designing Better Voting and Fair-Division Procedures,” (Princeton University Press, 2008) and, with Alan D. Taylor, “Fair Division: From Cake-Cutting to Dispute Resolution,” (Cambridge University Press, 1996).