Who has to Redistrict, and What Population Data to Use

By Tony Sissons, Research Advisory Services, Inc., Phoenix, for The Arizona Guardian

The once-a-decade process of redrawing election district boundaries to restore population equality after the Census involves elected officials, jurisdiction staff, the public, interest groups, and sometimes participants from other levels of government. For most local governments, the active part of the process takes at least four months – longer for larger jurisdictions – and the required review by the U. S. Department of Justice under Section 5 of the Voting Rights Act adds two to four months (or longer).

Most likely, no one would go through a redistricting process unless they had to. So let’s look first at who is required to and who is not.

All jurisdictions that elect their governing officials on a public ballot from districts or wards within their jurisdiction will probably have to redistrict. In Arizona, here are some of the affected jurisdictions with examples:

1. State government – legislative and congressional districts.
2. County governments – supervisorial districts and community college districts.
3. Cities and Towns – ward or council districts (for the districted jurisdictions of Buckeye, Glendale, Mesa, Peoria, Phoenix, Surprise, Tombstone, and Tucson).
4. Public School Districts – Phoenix Union High School District.
5. Joint Technical Education Districts – governing board election districts. (In the past, some JTED administrators and elected officials have believed that the “alternative election system” described in JTED enabling legislation shielded them from having to redistrict. However, Arizona Attorney General Opinion I08- 013 clarifies the circumstance under which that exemption would apply, determines that it does not apply to any of the current JTEDs, and concludes that “single-member districts from which a JTED elects its governing board members must be periodically redistricted to ensure that their populations are as nearly equal as practicable and that no vote is diluted”.)

One possible exception to any jurisdiction’s redistricting requirement would be the unlikely circumstance that its districts have remained in population balance throughout the decade. That exception has two caveats, though:

1. If a minority population in the jurisdiction has become sufficiently large, concentrated, and politically cohesive, the jurisdiction will have to examine whether conditions exist that trigger the need to create a minority-majority district under Section 2 of the Voting Rights Act. If such a remedy were to be required, new districts would have to be drawn. (We will discuss the Voting Rights Act and its administration next month.)
2. The population-equality standard for legislative and local government districts has become more strict during the decade. Prior to some recent federal cases, an overall district population variation below 10 percent was sufficient. However, that ‘safe harbor’ has now been called into question. (More about population standards later in this column.)

Clearly, if a jurisdiction conducts at-large elections for its governing officials, there are no districts to redraw. But, at-large cities with sizeable and growing minority populations could, during the coming decade, find themselves being required to institute districted elections by court order, or by a determination of the U. S. Department of Justice, as a result of a successful minority voting-discrimination claim.

Judicial districts (mainly justice of the peace precincts in Arizona) are exempt from redistricting to achieve population equality. The size of a judicial district is an administrative determination related to judicial workload. The U. S. Supreme Court, in Chisom v. Roemer (501 U.S. 380 (1991)), reaffirmed that the one person, one vote standard does not apply to judicial elections. However, the Court has also determined that both Section 2 (protection of voter rights) and Section 5 (preclearance review) of the Voting Rights Act apply to the conduct of elections in judicial districts.

So, now that we know who has to redistrict, let’s look at some of the data to be used.

Representative government based on population has its foundation in Article 1, Section 2 of the U. S. Constitution, along with the requirement for a national census every ten years. As the population of the various states changed, so did the geographies of representation. From those early days, the responsibility for redrawing district boundaries to reflect population shifts fell upon state legislatures. When they bothered to do it, they weren’t particularly consistent in their approaches.

In 1962, the U. S. Supreme Court held, in Baker v. Carr (369 U.S. 186 (1962)), that federal courts had jurisdiction to bring order and consistency to redistricting. One year later, in Gray v. Sanders (372 U.S. 368, 381 (1963)), Justice William O. Douglas wrote the term we respect today: “The conception of political equality from the Declaration of Independence, to Lincoln’s Gettysburg Address, to the 15th, 17th and 19th Amendments can mean only one thing – one person, one vote”.

The population equality standard for congressional districts is very strict – generally mandating a population difference of no more than two persons between any two districts within a state.

For legislative and local government districts, a greater population deviation is allowable, but the ‘plus-or-minus five percent’ standard of ten years ago may no longer suffice. In Larios v. Cox (300 F. Supp.2nd 1320 (N.D. Ga. 2004), aff’d, 542 U.S. 947 (2004)), the District Court and the Supreme Court objected to the strategic overpopulating of some districts, under-populating of others, to favor Democratic candidates in a legislative plan having a population deviation of 9.98 percent. The plan drafters asserted that the plus-orminus 5-percent standard gave them a safe harbor; instead, the courts found a violation of the Equal Protection Clause.

The unit of measurement is total population – all people, not just registered voters, and not just persons of voting age. Even though children, persons who have not registered, some convicted criminals, and non-citizens cannot vote, each of them counts as a person to be represented.

As in the past, there are currently calls for the Census Bureau to adjust population figures for various reasons – to statistically compensate for undercounts, or to exclude or count elsewhere such subgroups as prisoners, overseas military, non-citizens, or students in college dormitories. If these efforts are not partisan to begin with, they quickly turn so, with the political parties choosing sides to try to knock out population groups whose presence in a district they think might damage their candidates’ electoral successes.

In response to those requests, the U. S. Census Bureau has committed to an earlier-thanusual release of special tabulations of prisoners, non-citizens and other population subgroups, so that the data is available during redistricting season. The Bureau cautions, however, that those tabulations will be from the ongoing American Community Survey (ACS), an approximately one-in-eight sample survey of households that will include survey responses from 2005 through 2009. No results will be available below the level of Census Tract, and the reporting geographies will be those from the 2000 Census.

During a session on Census adjustments at the National Conference of State Legislatures (NCSL) Redistricting Law 2010 Seminar in Austin, Texas (March 25-28, 2010), much of the discussion centered on the data mismatches between the 2010 Census (actual headcount, with race and ethnicity detail, reported at the Census Block level) and the ACS data (a 12.5% sample, time-centered in 2007, reported at the Tract level). An audience member asked the panel whether using adjusted data was worth the effort, considering the data problems and the possible perception of partisan purpose. The advice from the panel: If you are going to make the adjustments, have plenty of money available for consultants and expert witnesses, because, in the politically-charged arena of redistricting, population adjustments (much like every other aspect of redistricting) are court challenges waiting to happen.

Tony Sissons is a political demographer, expert witness and redistricting consultant. His enjoyment of the NCSL conference was enhanced by seeing t-shirts that read “Keep Austin Weird”.

(Originally published in The Arizona Guardian, April 15, 2010)

Specializing in the analysis of data about geographic areas – Census blocks and tracts, Zip Codes®, voting precincts, land parcels, and traffic analysis zones