The Washington PostDemocracy Dies in Darkness

Opinion We have the data to make voting fair. Let’s use it.

A voter takes a sticker during 2014 elections in St. Clair Shores, Mich. (Joshua Lott/Getty Images)

Each week, In Theory takes on a big idea in the news and explores it from a range of perspectives. This week we’re talking about “one person, one vote.” Need a primer? Catch up here.

Peter A. Morrison is an applied demographer with experience constructing and evaluating local districting plans to comply with Voting Rights Act requirements. He is lead author ofBrief of Demographers as Amici Curiae in Support of Appellants in Evenwel v. Abbott.”

The “one person, one vote” principle is meant to ensure that voters in different districts hold equally weighted ballots and requires equalizing population across districts to do so. But what does that really mean? The Supreme Court has never defined the relevant population that must be equalized in order to protect the principle.

The plaintiffs in the upcoming case of Evenwel v. Abbott have asked the court to hold that “one person, one vote” requires electoral equality — that is, equalization across districts of persons eligible to vote so that every eligible voter has approximately equal influence. The Obama administration has entered the case and asked the court to hold that the rule requires representational equality — that is, equalization of districts’ total population so that each elected representative represents approximately as many people as any other representative. The state of Texas has asked the court to allow states the discretion to choose whichever population they want to equalize — eligible voters or total population. What is needed is judicial guidance on how to proceed.

I believe that judicial guidance should recognize and effectively protect the principle of electoral equality. The lack of any such protection has expanded the universe of potential ways to gerrymander voting districts or draw them with other purposes in mind.

[Other perspectives: ‘Equal representation’ should include non-citizens’]

Imagine that you are one of the 70,000 residents in a city that elects seven council members by district. Representational equality means there should be 10,000 residents in each district, and in most cases, this is a reasonable way to draw the lines.

But suppose a district has a large concentration of individuals who are ineligible to vote. For instance, your district of 10,000 residents holds 10,000 eligible voters, but mine has only 2,000 eligible voters. That means my vote is five times stronger than yours when it comes to electing our respective councilmembers. If the “one person, one vote” rule requires electoral equality, then this seemingly unfair distribution of voting-age citizens would be unconstitutional.

I favor using Census Bureau-endorsed demographic methods to effectively protect the principle of electoral equality in drawing voting districts. The data needed to do so are available: Demographers can draw voting districts using both or either one of two official Census Bureau data sets: the complete count of all persons (the “total population”) from the latest 2010 decennial census or the estimated share of that total population who are eligible voters (the “citizen voting-age population,” or CVAP) from the American Community Survey.

Protecting electoral equality is feasible, contrary to what Nathaniel Persily and others claim. The official CVAP estimates are routinely used to prove and settle inequalities in voting, including voting rights actions by the Justice Department itself. Applied demographers can use those data to distribute voting power fairly among state citizens and their local governments, with sufficient reliability to afford a statistical “safe harbor.”

The Evenwel case and “one person, one vote” have gained importance in recent years as the principle of electoral equality has been subordinated in many instances to the principle of representational equality. The common denominator in voting districts where the issue arises — towns ranging from Chelsea, Mass., to Yakima, Wash., in my experience — is a substantial presence of people who are ineligible to vote under state laws: typically, non-citizen immigrants or convicted felons.

Yet these two principles — electoral equality and representational equality — guide applied demographers like myself in crafting election districts that conform to the law. We are able to draw districts that adhere to one principle exclusively or that harmonize both.

State and local governments can and should avoid constructing districts of flagrantly disproportionate voting strength. They have reliable data at their fingertips to reduce those imbalances. If states wish to equalize total population in order to advance representational equality, they should be permitted to do so — but only as long as doing so does not compromise electoral equality.

Explore these other perspectives:

Richard H. Pildes: ‘Equal representation’ should include non-citizens

Nathaniel Persily: ‘One person, one vote’ isn’t broken, and the Supreme Court shouldn’t fix it

Ilya Shapiro: Why Texas is wrong in the ‘one person, one vote’ case

Aaron Blake: Thought immigration reform was unlikely? ‘One person, one vote’ could make things worse.

Nina Perales: The Supreme Court should seize the chance to strike down voter discrimination

Rob Richie: Let’s move beyond winner-take-all elections