WASHINGTON -- The Supreme Court this term could change how states meet the basic democratic goal of "one person, one vote." Ironically, a victory for the conservative plaintiffs who brought the case may turn on a national survey that Republicans have tried to eliminate.
In the Supreme Court case of Evenwel v. Abbott, the plaintiffs argue that the votes of eligible voters -- like themselves -- are unconstitutionally diluted because Texas counts non-voters when drawing its legislative districts. Specifically, Texas uses "total population" data, which include such non-voters as children, inmates, former felons who haven't had their voting rights restored and non-citizen immigrants.
The plaintiffs, backed by the activist nonprofit Project on Fair Representation, want the Supreme Court to rule that states must draw their legislative districts based instead on the number of voting-age citizens or registered voters. The likely outcome of that ruling would be to shift power away from cities -- which tend to have more children, non-citizen immigrants and Democrats -- and toward rural and suburban areas -- which skew older, whiter, richer and Republican.
The Supreme Court itself has never specifically addressed whether state legislative districts should have roughly the same number of people or the same number of eligible voters. The 14th Amendment to the Constitution says that the Census must count "the whole number of persons in each State" every 10 years for the purpose of apportioning seats in the U.S. House of Representatives. But it does not explicitly cover state legislative districts. As The New York Times noted, other federal courts have repeatedly ruled that using total population to draw state legislative districts is acceptable.
Texas is not an outlier in doing this: Almost every state currently uses total population numbers drawn from the U.S. Census. So a ruling in favor of the Evenwel plaintiffs could radically upend how state legislators are chosen and people are represented across the country.
This is where the American Community Survey comes into the picture.
The U.S. Census, which surveys every household in the United States every 10 years, does not ask recipients about their citizenship status. The American Community Survey, which is also conducted by the Census Bureau, is sent annually to a random sample of 2.5 percent of U.S. households. It does inquire about citizenship.
The ACS gathers reams of data on such matters as employment status, health insurance, languages spoken at home, flush toilets and disability status, and that information helps determine how billions of dollars in state and federal funding are spent each year. Still, many conservatives and libertarians are highly suspicious of the survey, claiming that it's intrusive and a waste of government resources. They've pushed to defund the survey or at least make it voluntary.
For instance, Andrew Coulson, a senior fellow of education policy at the libertarian Cato Institute, wrote in 2012 that the "prying … probing" ACS "should not be mandatory. It probably shouldn't exist at all."
But had congressional Republicans succeeded earlier in eliminating the ACS, the Evenwel plaintiffs and their supporters wouldn't be able to argue now that ACS data should be used to draw state legislative districts. Cato submitted a brief in the Supreme Court case that cites the ACS's data on voting-age citizens in Texas to argue that voters in more rural parts of the state are underrepresented in the state legislature compared to voters in urban areas.
Voting rights groups almost universally support using total population as the basis for drawing legislative districts. They say that just because some people can't vote doesn't mean they shouldn't be represented in the legislature. Many of those ineligible to vote pay taxes, and lawmakers make decisions about their lives. Children are affected by choices on school funding, and non-citizen immigrants have been drafted into war.
Voting rights advocates also point out that shifting to citizenship or voter registration numbers as a basis for redistricting isn't a new idea. After waves of European immigrants came to the United States in the late 1800s and early 1900s, congressional lawmakers repeatedly attempted to change the way districts were drawn. One Colorado lawmaker expressed the fear that "alien elements" could control who was elected to Congress, as the Brennan Center for Justice pointed out in its Supreme Court brief supporting Texas.
For periods of time, "politically dominant groups of the day were able to deny equal representation to parts of states where African Americans in the South, Irish-Americans in Massachusetts, Mormons and Chinese in the Nation's western territories, and immigrants in New York City lived," attorneys for the center wrote.
"That troubling history teaches why for 200 years the overwhelming practice, and the now settled constitutional tradition, is apportionment based on population," the center wrote. Eventually, the argument that there should be "no taxation without representation" won out at both the congressional and state levels.
But the Evenwel plaintiffs and their supporters don't agree. They argue that equalizing the number of voting-age citizens in each district should be more of a concern than representing the interests of non-citizens or others who are ineligible to vote.
"It cannot be that voter equality is subordinate to representational interests," the plaintiffs wrote in their brief to the Supreme Court. "Otherwise, non-voters would have standing to bring a one-person, one-vote challenge to an apportionment plan that equally distributes eligible voters."
And that, the Evenwel brief warns, is "a radical position."
Still, constitutional rights are not just theoretical exercises; they have to be implemented in the real world.
Five respected political scientists in the fields of election law and redistricting contend in their Supreme Court brief that the plaintiffs' case "is radical not only in its theoretical underpinnings, but also in its real, practical implications for the redistricting process."
The professors, including Nathaniel Persily of Stanford and Charles Stewart of MIT, say that the ACS data on citizens isn't a viable substitute for the decennial Census numbers. They say that the ACS estimates on numbers of voting-age citizens are not current or detailed enough at the census block or precinct level, which is the unit at which redistricting begins. The ACS's annual estimate is released for areas that are usually in excess of 65,000 people -- which is much bigger than a census block -- and the ACS's best data are not released in as timely a fashion as the Census data are.
In other words, the set of data that the Evenwel plaintiffs say could be used to draw districts based on citizenship does not actually exist.
"An interpretation of the Fourteenth Amendment that would prohibit the use of the most accurate and only constitutionally mandated population dataset and, in effect, mandate the creation of some new count of eligible voters would be both unprecedented and incredibly destabilizing to the U.S. Census and redistricting process," the political scientists wrote.
What about adding a question about citizenship to the Census? A group of former Census directors warn against that.
"Directly inquiring about citizenship status as part of the short form Census is not a solution to the data problem posed by Appellants' legal theory," the former officials wrote in their brief supporting Texas. "Doing so would likely exacerbate privacy concerns and lead to inaccurate responses from non-citizens worried about a government record of their immigration status."
The Evenwel plaintiffs downplayed the practical problems. "This appeal need not resolve every implementation issue," they told the Supreme Court.
One way the justices might resolve Evenwel v. Abbott is to explicitly leave it up to the states to choose a redistricting basis other than total population. Voting rights attorneys fear that such an outcome is conservatives' real goal. If states were given the go-ahead to abandon total population, those with Republican-controlled legislatures and large populations of non-citizen immigrants, like Texas or Arizona, might leap at the chance. The voices of the less popular and less powerful would fade even further.
The Supreme Court, which has not yet heard oral argument in Evenwel v. Abbott, will likely decide the case by June 2016.