Clay Wilson of Alexandria, Va., stands in front of the U.S. Supreme Court while the justices hear arguments on gerrymandering on Oct. 3, 2017. (Bill O’Leary/The Washington Post)

Last spring, in Gill v. Whitford, the Supreme Court decided not to strike down Wisconsin’s state legislative map as an unconstitutional partisan gerrymander. The court failed to decide whether it is against the Constitution for legislatures to draw maps that favor one party and instead requested that the plaintiffs have to show that individual Wisconsin voters were harmed at the district level.

But could an entirely different standard both improve the quality of representation in legislative districts and pass muster with the Supreme Court? Our new research suggests the answer is yes.

There are many ways to evaluate the quality of a legislative map, including what political scientists call the “efficiency gap,” which the Wisconsin plaintiffs used, and the “seats-votes curve.” We argue the quality of a map should be judged by how well it enables citizens and representatives to communicate and reach out to one another. Redistricting often does the opposite, making it more difficult for members to efficiently communicate with their constituents and more difficult for constituents to properly identify their representatives.

A solution: Use Zip codes

A simple way to improve redistricting is to have district boundaries overlap with Zip code boundaries. Zip codes were designed to allow the U.S. Postal Service to efficiently deliver mail to all residents from a central post office within a single day. They are smaller than the usual political units that states seek to preserve in redistricting, such as counties and towns. Each Zip code has an average population of 3,000, while a typical U.S. House district has a population of 710,000.

Towns or counties’ boundaries are rarely adjusted as populations shift and change. That’s not true for Zip codes, which the post office adjusts to account for new housing units. As a result, Zip codes can be easily allocated into districts — while simultaneously upholding the Supreme Court’s dictum that all districts have the same number of residents. As an independent agency, the USPS holds a patent on Zip codes, cannot be dictated to by Congress and can be sued for delays in mail delivery. It is more immune to political manipulation than political units like precincts.

What happens to representation when Zip codes are split up?

Most citizens use their Zip codes to look up their member of Congress. When Zip codes are split among different representatives, many Americans are stymied. For example, in North Carolina, Zip code 27713 is split among four districts. When you go to the Congress’s “find your member” page and enter that Zip code, you see four potential representatives, as you can see in the image here.


­­­

It is possible to then search by address, but many people do not. In total, 28 congressional websites explicitly report receiving mail from out of district because of split Zip codes. Citizens might see a message like the one found on Democratic Rep. Sheila Jackson Lee of Texas’s page, which reads:

There are multiple Representatives who share the same 5-digit zip code that was entered. Due to the large volume of US mail, email and faxes I receive, I am only able to accept messages from residents of the 18th congressional District of TEXAS. A 9-digit zip code including the zip+4 extension is required to identify your correct Representative.

This has real consequences for voters. In our research, we found that citizens living in Zip codes split into more than one U.S. House district are less likely to know the name of their member of Congress and less likely to contact him or her. What’s more, citizens living in split Zip codes also perceive greater ideological distance between themselves and their members — even when they are of the same party and racial group.

How to draw a map that respects Zip codes

Is it possible to draw a map that does not divide a single Zip code, while still ensuring that districts meet other standards? The answer appears to be yes, as you can see in the figures below.

The original map, as you can see here, was heavily gerrymandered by a Republican state legislative majority and split 10 to 3 in Republicans’ favor — even though in recent presidential elections, the state was almost evenly balanced between Democratic and Republican voters.


We redrew the 2013 North Carolina congressional district map without splitting a single Zip code, as you can see here — adhering as closely to the original 2013 Republican gerrymander while preserving Zip codes and population equality.


Unlike the current map, our map creates four Democratic districts, four competitive districts and five Republican districts, which more accurately reflects the state’s partisan makeup by average presidential vote share and enables competition in a way not possible under the congressional map. Our map meets other criteria as well. The districts have equal populations and, in line with the Voting Rights Act’s mandate, there is one majority-minority district and four minority-influence districts in which African Americans could conceivably elect a candidate of their choosing.

What’s more, we found that once you respect Zip code boundaries, it is effectively impossible to re-create a map as biased toward the Republican Party as the 2013 map.

Could a Zip code standard pass muster with the Supreme Court?

In Davis v. Bandemer (1986), the first Supreme Court ruling on partisan gerrymandering, the majority ruled that election outcomes alone cannot demonstrate that elected representatives will not respond to their entire constituencies. But we show, however, that violating Zip codes boundaries in redistricting does damage representation. Constituents who do not know their representatives and thus cannot contact them and feel alienated from them are not being effectively represented.

Congressional incumbents and challengers who must write off split Zip codes when mailing and campaigning due to costs and confusion cannot represent constituencies well, because of the districts’ boundaries. Splitting Zip codes among multiple congressional districts makes it harder for representatives and their constituents to communicate even more than when counties and towns are split among several districts.

Thus splitting Zip codes harms constituents in a way that can be measured at the district level, as the Supreme Court requested. Evaluating maps by this criterion might be an effective way to identify unconstitutional partisan gerrymanders.

Tyler Steelman and John Curiel are PhD candidates in political science at the University of North Carolina at Chapel Hill and placed third in the 2018 Common Cause partisan gerrymandering competition.