After a week of dragging their feet, the Tennessee General Assembly has finally released the statutory text of their Congressional proposal. Setting aside that this is perhaps the least accessible form, SSP Labs (a division of Daily Kos Elections) was able to find quite a few, er, points of interest.
Using this statutory description, I created a block equivalency file (a "blockeq" file lists the Congressional District to which each and every census block - all 240,116 - is assigned) and was able to create our customary Google Maps overlay from Census Bureau TIGER/LINE files, as well as our standard redistribution analysis. Based on the summary statistics I ran using my blockeq file, I believe that my file is correct. For example, consider the redistribution analysis:
If you note the total column at the far right, this ties out to the minimal deviation target for which these plans typically aim. Additionally, the Google Maps overlay I created also closely resembles the map posted by the General Assembly.
Let's start with the easy stuff. The General Assembly posted a "narrative" regarding the new map. Digs at the 2002 map aside, this narrative contains several errors:
- The First District fails to include Johnson County in its entirety.
- The Third District fails to include Polk County in its entirety.
- The Eighth District fails to include Crockett County, but double counts Chester County (which is part of the Seventh).
But of course, that's not the statutory language that would get enacted into law. (N.B.: I am not a lawyer and do not purport to be. But, the plain reading of the following statutes would seem to agree with my interpretation.)
Before diving in further, two important facts should be established:
First, contiguity was explicitly cited as an objective by the legislature in Section 2 of HB1558, the Congressional redistricting bill:
It is the legislative intent that all congressional districts be contiguous and contiguity by water is sufficient, and, toward that end, if any voting district or other geographical entity designated as a portion of a district is found to be noncontiguous with the larger portion of such district, it shall be constituted a portion of the district smallest in population to which it is contiguous.
Second, boundaries as defined in the 2010 TIGER/LINE system provided by the Census Bureau are the relevant ones that are to be considered here. Tennessee Code Annotated, Section 2-16-103(a)(1), as amended by Section 1 of HB1558, would read as follows (strikeout and emphasis mine):
All census descriptions, counties, voting districts (VTDs), tracts, blocks, census delineations, census district lines and other census designations are those established for or by the United States department of commerce, bureau of the census, for taking the 2000 2010 federal decennial census in Tennessee as contained in the automated geographic database, known as the TIGER (Topologically Integrated Geographic Encoding and Referencing) System.
Now that it has been established that Census Bureau Geographies are the relevant ones and that contiguity was an intent specifically cited by the legislature, let's get to the really fun stuff: the non-contiguousness of the districts proposed in HB1558. There are two examples: the first is weaker, in Tipton County in the proposed 8th district; the second is much more egregious, in Loudon County between the proposed 2nd and 3rd districts.
Tipton County and the proposed 8th district
What is shown here is a non-contiguous portion of Tipton County – a part across the Mississippi River that contains the unincorporated community of Corona, Tennessee resulting from a seismic shift after the New Madrid earthquake of 1812. This section is composed of Tipton County, Census Tract 401, Block Group 1, Block 1029 through Block 1046, inclusive and contains 9 residents and is assigned to the proposed 8th district (in blue). It, however, is still connected to the rest of Tennessee, but, importantly only to Shelby County.
But, the adjoining part of Shelby County is located within the proposed 9th District (in purple) – meaning that this area would be non-contiguous. Water contiguity (which was designated by the legislature as acceptable) is insufficient here, as even the water areas that would connect the two non-contiguous parts of Tipton County are either a part of Shelby County (and the proposed 9th District) or the State of Arkansas. Therefore, the proposed 8th district is non-contiguous.
Under the 2002 map, this was not an issue, as the entirety of Tipton County and the connecting part of Shelby County were both located within the 8th district.
Loudon County and the proposed 2nd and 3rd districts
What is shown here are two exclaves of Loudon County (in purple), the eastern one of which is surrounded entirely by Monroe County and the western one of which is sandwiched between McMinn and Monroe Counties. While I'm uncertain of how this odd boundary arose, I'll note that it's not unusual for counties and county equivalents to be non-contiguous. The most evident example, of course, is Saint Martin Parish, Louisiana.
Indeed, the first exclave is composed of Loudon County, Census Tract 607, Block Group 1, Block 1150 through Block 1152, inclusive, and contains 30 residents; the second of Loudon County, Census Tract 607, Block Group 2, Block 2154 through Block 2157, inclusive and contains 15 residents. (While the permissibility touch-point contiguity is a matter of intense debate, I assume arguendo that touch-point contiguity is acceptable and won't further discuss the three Census blocks that are touch-point contiguous.)
Loudon County, in its entirety, is assigned to the proposed 2nd district (in purple). McMinn and Monroe Counties, in their entireties, are assigned to the proposed 3rd district (in green). Therefore, these two exclaves of Loudon County are also exclaves of the proposed 2nd district – which also renders the proposed 2nd district non-contiguous.
Again, this was not an issue in the 2002 map, as the entirety of Loudon, McMinn, and Monroe Counties were assigned to the 2nd district.
So What?
The impact of these two non-contiguities is as follows. Under Section 2 of HB1558, these areas would be assigned to the least populous district to which they are connected. Since both non-contiguities only adjoin one district each (the proposed 9th for Tipton; the proposed 3rd for Loudoun), the population in each area would have to be assigned the corresponding district.
As a result, the proposed 2nd district would have a deviation of -45 from the ideal population of 705,123; the proposed 3rd district would have a deviation of +44. Similarly, the proposed 8th district would have a deviation of -10 and the proposed 9th district +9.
Sure, that's not a huge deal in the scheme of things. (I'll reiterate here that I'm not a lawyer.) But, this could easily be insufficient to qualify as "a good faith effort to achieve population equality" as specified by the Supreme Court in 1983 in Karcher v. Daggett, and consequently, could easily be unconstitutional.
Tennessee General Assembly, the ball's in your court.
