- North Carolina’s county clusters are a unique solution to the problem of balancing two constitutional requirements for redistricting
- A bill to alter how county clusters are determined runs counter to separation of powers and is unnecessary
- A better proposal would be to simply limit political considerations when drafting county clusters
The latest round in the fight over redistricting has begun in the General Assembly. However, S715, sponsored by Senators Ben Clark (D-Cumberland, Hoke) and Mike Woodard (D-Durham, Granville, Person) is not about drawing legislative districts; it is about the drafting of county clusters.
County what?
A unique solution to a constitution conundrum
So, what are county clusters and why does North Carolina have them?
The short answer, as provided on page two of S715, is that a county cluster is a “grouping of one or more counties that is capable of containing a whole number of legislative districts.”
The long answer is a bit more interesting.
County clusters are the result of two court cases, Stephenson v. Bartlett one (2002) and two (2003). In those rulings, the North Carolina Supreme Court sought to balance the state constitutional requirement that counties not be divided with the state constitutional requirement that legislators represent roughly an equal number of inhabitants. The court also rejected the use of multimember districts to balance those requirements (which is allowed by the state constitution) over concerns that they may violate the Voting Rights Act.
The rules that the court laid out in the Stephenson rulings severely limited how much legislators could gerrymander districts by creating a county clustering requirement for redistricting. It did so by requiring that legislators first group counties into clusters that contained the minimum number of districts within the minimum number of counties (ideally one district entirely within one county, but that is rarely achieved). It further required that legislative district lines could only cross county lines within those clusters and only to the minimum extent possible to make districts roughly equal in population.
So, before the legislative districts are drawn, the General Assembly must first make clusters of counties that comply as closely as possible to the one county requirement as interpreted in the Stephenson rulings. Those rulings had the effect of severely limiting how creative members of the General Assembly could be when drawing districts (Seth Warren Whitaker, Virginia Law Review, page 207):
In Stephenson I, the court reached a completely unexpected resolution to a lawsuit over the state’s legislative redistricting plans by fashioning a set of judicially created redistricting criteria. In Stephen II, the court provided further information about just how stringent it intended the criteria devised in Stephenson I to be and gave a glimpse of the very narrow range of discretion remaining for the North Carolina General Assembly in legislative redistricting.
The Stephenson requirements prevent some of the more extreme forms of gerrymandering in which the General Assembly had previously engaged, such as the 2001 House map, a detail of which is seen in figure 1.
Figure 1: A section of a pre-Stephenson map, Sutton House Plan 3, that passed the General Assembly in 2001 but was struck down in Stephenson v Bartlett 1. Two features shown here, how Edgecombe and Wilson counties are both divided between districts 23 and 24, and how district 7 occupies two non-contiguous parts of Nash County, were banned by the court decision. Graphic Source: NC General Assembly
County cluster bill is a constitutional and political nonstarter
With S715, Senators Clark and Woodard are seeking to transfer responsibility for developing potential sets of county clusters from the General Assembly to the State Board of Elections.
Yes, sets of county clusters.
A 2019 study led by Duke Mathematician Jonathan Mattingly found that the county clusters currently used for legislative districts are “optimal” in that they have the largest number of clusters while following the Stephenson requirements (implying that it maximized the number of one-county and two-county clusters). However, they also found that there could be other optimal clusters.
So S715 is the first shot in the upcoming fight over county clusters that the legislature will have to deal with before it deals with actual districts.
There are several problems with Clark and Woodard’s bill. The first is constitutional: the NC Constitution places redistricting exclusively within the purview of the legislature (the governor cannot even veto redistricting legislation), it would be inappropriate at best for the legislature to turn over part of its redistricting authority to a state agency whose board members are appointed by the governor. If legislators want someone other than the legislators themselves to draft sets of county clusters, they could propose that the Legislative Services Commission do it.
Then there is the political problem with the bill: why would the Republican-controlled General Assembly cede its authority to develop county clusters to the Democratic-controlled State Board of Elections?
The final problem with the bill is practical. Just as we know both sides develop their own set of ideal district maps during the redistricting process, both sides will also develop sets of county clusters during the 2021 redistricting process. Injecting the State Board of Elections into the process to do what legislators from both parties will do on their own anyway just adds an unnecessary step.
There is one bright spot in S715: the proposed addition of a requirement that the General Assembly not consider incumbency protection or partisan advantage when selecting county clusters. A much better bill would simply call for those requirements without all the unnecessary clutter S715 contains.
Senators Clark and Woodard should go back to the drawing board.
Find out more: To learn more about how the county clustering works, check out this explainer.
