by Dr. Andy Jackson
Director of the Civitas Center for Public Integrity, John Locke Foundation
Part 1 of this series examined criteria related to political boundaries (counties, municipalities, and precincts). Part 2 looked at criteria related to geography other than political boundaries (contiguity, compactness, and equal population). This final part reviews what map drawers may not consider (election data, racial data) along with nonmandatory considerations (member residence, communities of interest).
Each section will begin with language taken directly from the criteria adopted by the joint redistricting committees.
Here are the criteria concerning election data:
Election Data. Partisan considerations and election results data shall not be used in the drawing of districts in the 2021 Congressional, House, and Senate plans.
An assessment of election data in drawing districts was at the heart of two successful 2019 lawsuits, Common Cause v. Lewisand Harper v. Lewis, that overturned state legislative and congressional districts, respectively.
In Common Cause v. Lewis, the court noted that Rep. David Lewis, the chair of the House Redistricting Committee, proposed that “political consideration[s] and election results data may be used in drawing up legislative districts in the 2017 House and Senate plans” (page 16). They also noted that Dr. Thomas Hofeller, employed by Republicans to conduct redistricting analysis, used “elections results and partisanship formulas” (page 17) to help Republicans win more seats in the General Assembly. Partisan considerations are data sources other than election results that help map drawers determine the partisan lean of districts. They include voter registration data.
In response, the court ordered several legislative districts be redrawn and banned the General Assembly from using “partisan considerations and election results data” (page 348) when drawing the new districts. The resulting redistricting process won praise from both parties. The ban on using political data this time around indicates that legislators want to both repeat that 2019 experience and try to better protect their maps from lawsuits.
Here are the criteria concerning racial data:
Racial Data. Data identifying the race of individuals or voters shall not be used in the construction or consideration of districts in the 2021 Congressional, House, and Senate plans. The Committees will draw districts that comply with the Voting Rights Act.
Excluding racial data will help protect legislators from charges that they used such data to gerrymander districts racially. Section 2 of the Voting Rights Act (VRA) requires states to draw districts in which racial minorities have opportunities to “elect candidates of their choice,” however. So how can we be sure that a redistricting plan complies with the VRA?
We have seen that it is possible to draw VRA-compliant districts without using racial data. For example, in Common Cause v. Lewis (2019), a three-judge panel noted on pages 283 and 284 that districts drawn by Dr. Jowei Chen, an expert witness for the plaintiffs, using a process “which ignored racial data,” produced districts that had “higher or much higher BVAP [black voting-age population]” than districts defendants said were drawn to comply with the Voting Rights Act.
The redistricting committees adopted two other criteria that will help create VRA-compliant districts without resorting to racial data. Compact districts drawn in urban areas and the “black belt” counties of Eastern North Carolina will likely produce districts in which minority voters can elect candidates of their choice. While communities of interest is a vague consideration, it often coincides with race, meaning that consideration of communities of interest may also help create districts in which minority voters can elect candidates of their choice.
North Carolina’s political geography makes it possible to draw VRA-compliant districts without using racial data.
Here are the criteria concerning member residence:
Member Residence. Member residence may be considered in the formation of legislative and congressional districts.
The idea behind this consideration is to avoid “double bunking,” placing two incumbents in the same redrawn district. The idea is that incumbents, who have ongoing relationships with constituents, should not be written out of their districts or forced to compete against other incumbents, especially in primaries.
North Carolina senators and representatives are legally required to live in their districts. While members of Congress are not required to live in their districts, living outside the district is politically disadvantageous.
Considering member residencies can complicate redistricting. For example, the Charlotte metro area has the residencies of four incumbent members of Congress in close proximity (see the map below). Considering incumbent residencies could prevent map drawers from following redistricting best practices such as maximizing compactness and respecting political boundaries. That is why I have advocated against considering the addresses of incumbents when drawing districts.
Current North Carolina congressional districts with homes of incumbents marked. Source: North Carolina General Assembly.
Member residence is a “may” rather than a “shall” consideration. If legislators do not make avoiding double bunking the primary factor when drawing districts, their inclusion in the redistricting criteria should not be too disruptive.
Here are the criteria concerning community consideration:
Community Consideration. So long as a plan complies with the foregoing criteria, local knowledge of the character of communities and connections between communities may be considered in the formation of legislative and congressional districts.
Community of interest is a nebulous concept. Stephenson v. Bartlett, which has a large influence on how state legislative districts are drawn in North Carolina, simply states “communities of interest should be considered in the formation of compact and contiguous electoral districts” (pages 44-45). The court provided little guidance about what they meant by communities of interest other than stating that there was “clear identity and common interests that counties provide” (page 15).
The result of that uncertainty over what constitutes a community of interest is that those seeking to influence the redistricting process can define almost any grouping of people as a community of interest. If unchecked, that could produce a detrimental effect on redistricting. For example, while considering factors that coincide with race (such as geography or economic interests) can help create Voting Rights Act–compliant districts, an overreliance on those factors could lead to racial gerrymandering.
So it is appropriate that the adopted criteria say that map drawers “may” consider communities of interests only “so long as a plan complies” with the other criteria. Aside from being a good policy, it also follows the guidance in Stephenson, which places communities of interest as a “nonmandatory criteria” (page 68).