The Redistricting Folder: Competitive Districts

By Tony Sissons

It is easier to draw competitive districts than it is to avoid doing so. The political parties do not want voters to know this.

This column will not be about why competitive districts are better or worse than ‘safe’ districts. Most readers of this column have probably already formed an opinion.

Instead, this discourse begins with the acknowledgement that having a preference for competitive districts was written into the Arizona Constitution by the passage of Proposition 106 on the 2000 General Election ballot.

We will look at how competitiveness is measured; how the competitiveness principle fits with other district design criteria; at what stage in the process the new Independent Redistricting Commission must consider political competition; and finally, how many competitive districts can be drawn.

Measuring competitiveness. A competitive district is one in which candidates of any political party stand a reasonable chance of winning if they can appeal to more voters than contestants of other parties.

In the past, district competitiveness has been measured in relation to the two major political parties. While Independent voters may outnumber partisan voters in some parts of the state, they are not likely to be part of the competitiveness calculation until a much greater number of Independent candidates appear on the ballot. Currently, the Independent voting bloc votes almost exclusively for Republican or Democratic candidates.

The most intuitive measure of district competitiveness is simply the registration ratio between the parties. If one party’s registration is less than three (five? seven?) percentage points higher or lower than the other party, the district is considered competitive. However, the usefulness of this calculation is clouded by the substantial number of registered voters who do not vote.

In an attempt to have a measure that was based on actual vote results, the previous Arizona Independent Redistricting Commission (IRC) created AQD (Arizona Quick & Dirty), which used the ratio of total votes received by Republican and Democratic candidates for seats on the Arizona Corporation Commission in 1998 and 2000. That statewide contest was chosen because it was thought that its candidates were less likely to be known to voters, who would then rely on party labels to make their selections.

AQD happened to work well for the IRC. It could be calculated quickly, and the ballot contests on which it was based were sufficiently ordinary that it gave an acceptable measure of each district’s partisan propensity. It also tracked well with a more sophisticated statistical measure, described next. A downside to AQD, and a reason not to use it this time, is its circumstantial volatility. Recent Corporation Commission elections have become better funded, individual candidates are better known to voters, and partisan spending has not been evenly balanced –
all factors that detract from the measure’s reliability.

The third competitiveness measure used by the IRC was JudgeIt, a statistical model that predicts two‐party election outcomes based on multiple prior election cycles, developed by Harvard professors Andrew Gelman and Gary King. The model was operated by the IRC’s competitiveness expert, Dr. Michael P. McDonald, an associate professor at George Mason University in Virginia.

JudgeIt was acknowledged by the IRC as having superior accuracy, but it was not able to produce quick results. The computer model results were often not available until the following day. Most of the time, JudgeIt found one more district to be competitive than did AQD for any given proposed plan.

Competitiveness vs. other district design principles. Participants in the process and in the ensuing litigation recognized that it is impossible to apply all of the Arizona Constitution’s district design criteria to the fullest extent, simultaneously, without some criteria impinging on and detracting from full implementation of other criteria. All criteria, of equal importance, are likely to have to give way a little.

Throughout most of the previous process, the IRC claimed that the Constitution’s wording gives competitiveness a lesser importance than other listed criteria. As the perpetual least‐important factor, by the IRC’s reading of the Constitution, competitiveness was not considered until late in the process, and then was asserted to cause too much damage to districts already drawn based on the other criteria.

The Arizona Supreme Court disagreed, and in its 2009 opinion, ruled that the wording of the Arizona Constitution gives competitiveness equal standing with all other non‐federal requirements listed. From Section 35 of the Opinion:

“The direction that competitiveness should be favored unless one of two conditions occurs does not, contrary to the Commission’s assertion, mean that the competitiveness goal is less mandatory than the other goals, can be ignored, or should be relegated to a secondary role. The constitutional language means what it says: The Commission should favor creating more competitive districts to the extent practicable when doing so does not cause significant detriment to the other goals.”

Practicable means ‘can be put into practice’. It does not mean ‘practical’ or ‘if we choose to’.

When must the redistricting commission consider competitiveness? In a largely concurring minority opinion, Arizona Supreme Court Justice Andrew Hurwitz expanded upon the Redistricting Commission’s obligation by saying that the commission must consider competitiveness, in concert with the other constitutionally prescribed principles, at the first moment that it considers any of the principles. That means, immediately after publishing the ‘grid map’ that starts the redistricting process. In Section 53 of the Opinion, Justice Hurwitz writes that:

“(Delaying consideration) inevitably threatens to relegate the competitiveness goal to precisely the “secondary role” that the Court correctly abjures.”

How many competitive districts can be drawn? In the previous state redistricting process, the commissioners said that the creation of a sufficient number of minority districts to meet the requirements of the Voting Rights Act absorbed so many Democrats that there were not enough of them left over to make many competitive districts.

It seems like a convincing argument, but it is false in the totality of circumstances.

In Arizona, there are six or seven districts‐worth of highly Republican residential areas in which any district drawn cannot be made more competitive because everywhere one looks along a district edge to find Democrats to balance competition, one just finds more Republicans.

These districts cover affluent housing areas in northeast Phoenix, Paradise Valley and Scottsdale; in the Foothills areas north of Tucson; and in very conservative neighborhoods throughout the East Valley portion of the Phoenix metropolitan area.

These 6 or 7 heavily Republican districts somewhat ‘balance’ the 7 or 8 heavily Democratic voting rights districts, leaving the remaining area of state, 14 to 16 districts‐worth, with a partisan balance similar to the statewide ratio (about a 5% Republican advantage in 2000, and about 6% today).

In that remaining area, it is possible to draw as many as twelve competitive districts without conflicting with any of the other redistricting goals listed in the Arizona Constitution.

As the Arizona Supreme Court has said, the choice of how many competitive districts to draw is entirely the prerogative of the redistricting commission. It might be useful for the new commissioners to start with the knowledge of what is possible.

Tony Sissons is a political demographer, expert witness and redistricting consultant.  This post also appeared in the Arizona Guardian.