Using Urban County Clusters To Guide Redistricting

[muon2]

I have never had any problem with the notion that the VRA will next to insist on creating a majority-minority CD which otherwise takes in a reasonably compact and contiguous population, that involves shaving off a whole county from an urban cluster. That will probably happen. It is when it comes to more chops, non contiguous and more erose, that the field becomes more opaque, until at some point I next to not see it at all.

So on this one, to the extent of the above, and also that that 40% figure of yours looks like a reasonably horizon spot to seek out such minority clusters (at least to the extent reasonably contiguous, and not too erose), we are on the same page -  for once! 

So let me set up a scenario for Justice Torie. I'll start with the minority cluster in AL using the 40% threshold - the two darker shades in this map.



Next I want to see if I can get a majority BVAP district from this cluster. It is underpopulated, so I need to add some adjacent counties to make it whole. I do that in the following map and I add in part of Dothan in Houston county in the SE corner to bring the total BVAP to 50.2%. From this I conclude that a geographically compact (given all the rural counties) BVAP-majority district is possible, so it should be required to draw one in AL.



Of course the problem with the exercise so far is that the CD I drew cuts off the southern part of the state. It's got about 190K too many people for one CD and no contiguous path to link the south to the north. Now I could draw a narrow sliver along the Chattahoochie river that takes in nobody (I could use the river parcels on the AL side of the state line) and use that to link the south to the north. Now that would be a geographic monstrosity for sure.

So instead one of my state objectives is preserving county integrity and minimizing chops. That means I'd like a nice corridor of counties along one side of the state to link north and south. Let's try the east side corridor. Cutting out Phenix City and Eufaula drops the black population quite a bit. Running up to Tuscaloosa in whole precincts leaves the CD just short of 50% BVAP (I'll use your offering and ignore the possible play of individual blocks.)

But since I have to go somewhere to make up for the population in the corridor, I see no restriction as long as I follow generally accepted redistricting principles. For me that means only one chop and a relatively compact shape. Suppose I submit this as my district to meet the VRA need I identified at my earlier discussion of the cluster. Does it satisfy that need even though it hops into Mobile? BTW I cut off my district just north of the causeway across Mobile bay so there is no problem with connectivity.

[Torie]

I went through sort of the same intellectual process Mike.  You have to create the eastern escape route, and thus it is not possible to draw a black CD that meets the criteria. I understand the notion that if one can draw a black contiguous reasonably non erose CD (yours is close to the line with the Montco choke point, but if one is going to have a choke point, the centrally located population center for the minority belt is the place to do it) without doing chops, but can't because of state geography (here the black belt crossing the state), the notion that then, and only then, one is unleashed to cross a white bridge to stab grab a black urban population, I just don't think will fly.  Indeed, the case for Mobile is weaker than for Tuscaloosa because Mobile has that white blockage.

So I am with your approach, until your step of cashing in your credits for having a drawn a black CD that would and should be drawn, except that it can't because it screws up the balance of the map, to do an urban stab grab, particularly a stab grab that has to cut through a white epidemis. If you can't draw a black CD that meets what the VRA would probably insist upon, or at least more likely than not, then you don't draw it. KISS.

[muon2]

I went through sort of the same intellectual process Mike.  You have to create the eastern escape route, and thus it is not possible to draw a black CD that meets the criteria. I understand the notion that if one can draw a black contiguous reasonably non erose CD (yours is close to the line with the Montco choke point, but if one is going to have a choke point, the centrally located population center for the minority belt is the place to do it) without doing chops, but can't because of state geography (here the black belt crossing the state), the notion that then, and only then, one is unleashed to cross a white bridge to stab grab a black urban population, I just don't think will fly.  Indeed, the case for Mobile is weaker than for Tuscaloosa because Mobile has that white blockage.

So I am with your approach, until your step of cashing in your credits for having a drawn a black CD that would and should be drawn, except that it can't because it screws up the balance of the map, to do an urban stab grab, particularly a stab grab that has to cut through a white epidemis. If you can't draw a black CD that meets what the VRA would probably insist upon, or at least more likely than not, then you don't draw it. KISS.

But I think my point is that I can draw one that is compact, but it would result in an incredibly erose district that remains linking a narrow corridor along one or the other side. Is there anything in the VRA cases that constrains the shape of non-VRA districts? If not I'm left believing that a black-majority CD must be drawn in AL. Is there anything that specifically forbids crossing a white area to create a VRA district? I read the cases as forbidding thin tendrils and other geographically unpleasant shapes to link disparate areas, but not forbidding a geographically acceptable shape that crosses through a white area. That seems to be consistent with the dictum that race can be a factor, just not the only factor.

In my Mobile offering I chop the eastern end of the cluster by more than a thin path to satisfy my compelling interest in minimizing chops and erosity. I don't read the cases as that interest excusing me from otherwise drawing the black-majority district. It looks to me that as the mapper the VRA gives me some flexibility as to how to replace the chop on the east. It could be Tuscaloosa or Mobile.

[Torie]

No case has ever demanded that bridges be built over white areas, and the case law rule is that the minority must be compact, and bridges over the river white mean it's not compact, or at least not contiguous. I am extrapolating, but I think in the only way that makes sense. I also have trouble believing that the VRA would require a black CD to be drawn, that forces other CD's into a mess, chopping and traveling and so forth. Your map using Tuscaloosa is a closer case however, because the black stab there is more contiguous with the balance of the black zone in the state.

We are in agreement however, that no state rule should get in the way of a black CD that does meet it taking in a black area that is reasonably contiguous and compact. And your mapping out the zone with your 50% and 40% test is a useful tool. And you came close - but alas so cigar in my judgment.

How many CD's are we talking about that are at risk here over our issue? One in LA, one in Alabama. Can you think of any others? One in Florida perhaps?  I do know the black population in St. Louis is remarkably contiguous, so that CD is a go. I assume that you don't think the VRA demanded the black CD in Ohio do you?

[jimrtex]

Edit: No change to definition, one more correction to map.

This is a corrected version of the metro map for the Southeastern USA.   There is no change in definitions - I had miscolored the map.



Corrections:

Little Rock, AR: I had colored Prairie rather than Perry.
Gainesville, FL: I did not color Gilchrist.
Lafayette, LA: Miscolored surrounding counties with urban clusters.
Greensboro-High Point, NC: I had miscolored Randolph and Rockingham.
Oklahoma City, OK: I did not color Grady,
Spartanburg, SC: I did not color Union.
Austin-Round Rock, TX: I did not color Bastrop and Caldwell.

In most of these cases these were potential micropolitan central counties, which became outlying counties of metropolitan areas.   Most fail to qualify as part of the urban county cluster, because they have little or no urbanized area population.

[muon2]

No case has ever demanded that bridges be built over white areas, and the case law rule is that the minority must be compact, and bridges over the river white mean it's not compact, or at least not contiguous. I am extrapolating, but I think in the only way that makes sense. I also have trouble believing that the VRA would require a black CD to be drawn, that forces other CD's into a mess, chopping and traveling and so forth. Your map using Tuscaloosa is a closer case however, because the black stab there is more contiguous with the balance of the black zone in the state.

Perhaps we will have to disagree here. In my mind just because no case has demanded inclusion of a significantly white area does not mean it can't be used in lieu of a contiguous black area. The IL Dems successfully did that on their 2011 legislative map and that map was upheld in federal court. They actually ignored an area that would keep a district over 50% BVAP and attached a white area showing that the resulting district was still likely to elect the candidate of choice of the minority at 46% BVAP. I think the court would find that once a district can be drawn it must be drawn and a few extra whites isn't going to bother them, to the contrary they might find that carefully excising all the whites has the potential to become a forbidden racial gerrymander using race as the only factor.

You must be logged in to read this quote.

Here are two maps that both use whole county bridges from the north to the south which I don't think the court would object to and meets my assumed requirement that a 50% BVAP CD is required. The first uses the Mobile split and has only two chops and one microchop. If you still think you can duck the requirement my second map has only three chops and only splits Tuscaloosa as a city to the extent DRA allows to make CD 2 at 50.3% BVAP. If that second plan could be required I contend either map would meet the VRA requirement as well as our rules using the urban and minority clusters.

You must be logged in to read this quote.

I think that SC and NC may also present cases of combined rural-urban or highly erose CDs. We also will have to address HVAP where I think the 40/50% VAP categories become 50/60% VAP to estimate CVAP.

As to OH, I don't know the answer, but I do know that the OH Pubs (and their lawyers) believed that it was necessary and that their plan was unsuccessfully challenged by the Dems. If the Dems could have beaten the Cleve-Akron CD as a racial gerrymander I'm sure they would have.

[Torie]

What a court would object to as a racial gerrymander, and what the VRA requires, are two different things. What we are arguing about is what a court would demand, both knowing the answer is uncertain, but with quite different takes.

Anyway, the black CD I think would be most likely to trigger the VRA is the one below, which I think Train already drew. It hews to good redistricting principles, more or less, with but the contiguous stab into Tuscaloosa. Do you disagree?

[\

[jimrtex]

No case has ever demanded that bridges be built over white areas, and the case law rule is that the minority must be compact, and bridges over the river white mean it's not compact, or at least not contiguous. I am extrapolating, but I think in the only way that makes sense. I also have trouble believing that the VRA would require a black CD to be drawn, that forces other CD's into a mess, chopping and traveling and so forth. Your map using Tuscaloosa is a closer case however, because the black stab there is more contiguous with the balance of the black zone in the state.

Perhaps we will have to disagree here. In my mind just because no case has demanded inclusion of a significantly white area does not mean it can't be used in lieu of a contiguous black area. The IL Dems successfully did that on their 2011 legislative map and that map was upheld in federal court. They actually ignored an area that would keep a district over 50% BVAP and attached a white area showing that the resulting district was still likely to elect the candidate of choice of the minority at 46% BVAP. I think the court would find that once a district can be drawn it must be drawn and a few extra whites isn't going to bother them, to the contrary they might find that carefully excising all the whites has the potential to become a forbidden racial gerrymander using race as the only factor.

You must be logged in to read this quote.

Here are two maps that both use whole county bridges from the north to the south which I don't think the court would object to and meets my assumed requirement that a 50% BVAP CD is required. The first uses the Mobile split and has only two chops and one microchop. If you still think you can duck the requirement my second map has only three chops and only splits Tuscaloosa as a city to the extent DRA allows to make CD 2 at 50.3% BVAP. If that second plan could be required I contend either map would meet the VRA requirement as well as our rules using the urban and minority clusters.

You must be logged in to read this quote.

I think that SC and NC may also present cases of combined rural-urban or highly erose CDs. We also will have to address HVAP where I think the 40/50% VAP categories become 50/60% VAP to estimate CVAP.

As to OH, I don't know the answer, but I do know that the OH Pubs (and their lawyers) believed that it was necessary and that their plan was unsuccessfully challenged by the Dems. If the Dems could have beaten the Cleve-Akron CD as a racial gerrymander I'm sure they would have.

I think that the current configuration for a majority-minority district.  This map shows the black voting age population relative to that needed to comprise 50% of a district.   There are 3,647,277 adults in Alabama, or an average of 521,039 per district.  50% of that is 260,520.  Jefferson County has 200,129 black adults, which is 0.768 of that needed for 1/2 of the electorate of a district.  The counties in white have less than 25% BVAP, which means going into those counties you are picking up at least 3 whites for every black.

Greene is 79% BVAP, but it has only 2% of the blacks needed to get to 50% (ie 1% of the total electorate).



I don't see that you are respecting the UCC, you are splitting off Autauga and Elmore from Montgomery, and when you go into Houston, Tuscaloosa, or Mobile, you are also splitting a UCC.

Why do districts have to be contiguous?

[jimrtex]

What a court would object to as a racial gerrymander, and what the VRA requires, are too different things. What we are arguing about is what a court would demand, both knowing the answer is uncertain, but with quite different takes.

Anyway, the black CD I think would be most likely to trigger the VRA is the one below, which I think Train already drew. It hews to good redistricting principles, more or less, with but the contiguous stab into Tuscaloosa. Do you disagree?

[\

If you put Elmore and Autauga into the majority-minority district, in exchange for Monroe, Conecuh, Crenshaw, Pike, how low does that drop the BVAP?  That would make it less of a hook for the SE district, since it would have more population below the Black Belt, and it would be more that the majority-minority district was just extending out to pick Macon and Bullock, rather than trying to make a corridor along the border.

[jimrtex]

Edit: Correction to map only.



Explanation:

BIRMINGHAM-HOOVER, AL 854/1004 (2/4): Jefferson County (Birmingham, AL 593)  658  593  90% SUA;  Shelby County (Birmingham, AL 148)  195  148  76% SUA;  St. Clair County (Birmingham, AL 9)  84  9  11% NSUA;  Walker County ()  67  0  0% NSUA.

The Birmingham-Hoover UCC has 854K population in its final configuration, 1,004K before the final cut.  Two of the final 4 were retained.  Jefferson County has 593K in the Birmingham Urbanized Area, and a total population of 658K, with 593K or 90% in urbanized areas.   Since 593K is greater than 25K, and 90% is greater than 5%, Jefferson County is retained.   Similarly  Shelby County, 148K > 25K and 76% > 25% is also retained.  St.Clair and Walker counties fail the test, and are excluded, as indicated by red.  They are shown in a lighter tint on the map.

Counties shown in green (see Elmore County in the Montgomery, AL UCC) have between 25K and 50K in urbanized areas, and between 25% and 50% of the population in urbanized area.  If the thresholds were increased to 50K and 50% it would be excluded.


Alabama

BIRMINGHAM-HOOVER, AL 854/1004 (2/4): Jefferson County (Birmingham, AL 593)  658  593  90% SUA;  Shelby County (Birmingham, AL 148)  195  148  76% SUA;  St. Clair County (Birmingham, AL 9)  84  9  11% NSUA;  Walker County ()  67  0  0% NSUA.

HUNTSVILLE, AL 335/418 (1/2): Madison County (Huntsville, AL 277)  335  277  83% SUA;  Limestone County (Huntsville, AL 10)  83  10  12% NSUA.

MOBILE, AL 413/413 (1/1): Mobile County (Mobile, AL 326)  413  326  79% SUA.

MONTGOMERY, AL 363/363 (3/3): Montgomery County (Montgomery, AL 205)  229  205  90% SUA;  Elmore County (Montgomery, AL 27)  79  27  34% SUA;  Autauga County (Montgomery, AL 32)  55  32  58% SUA.

TUSCALOOSA, AL 195/195 (1/1): Tuscaloosa County (Tuscaloosa, AL 139)  195  139  71% SUA.

DAPHNE-FAIRHOPE-FOLEY, AL 182/182 (1/1): Baldwin County (Daphne--Fairhope, AL 57; Pensacola, FL--AL 6)  182  64  35% SUA.

DECATUR, AL 119/119 (1/1): Morgan County (Decatur, AL 70)  119  70  59% SUA.

FLORENCE-MUSCLE SHOALS, AL 147/147 (2/2): Lauderdale County (Florence, AL 47)  93  47  51% SUA;  Colbert County (Florence, AL 30)  54  30  55% SUA.

DOTHAN, AL 102/102 (1/1): Houston County (Dothan, AL 66)  102  66  65% SUA.

AUBURN-OPELIKA, AL 140/140 (1/1): Lee County (Auburn, AL 75; Columbus, GA--AL 27)  140  102  73% SUA.

ANNISTON-OXFORD-JACKSONVILLE, AL 119/119 (1/1): Calhoun County (Anniston--Oxford, AL 75; Gadsden, AL 0)  119  75  63% SUA.

GADSDEN, AL 104/104 (1/1): Etowah County (Gadsden, AL 64)  104  64  61% SUA.

COLUMBUS, GA-AL 53/53 (1/1): Russell County (Columbus, GA--AL 34)  53  34  65% SUA.


Arkansas

LITTLE ROCK-NORTH LITTLE ROCK-CONWAY, AR 671/671 (4/4): Pulaski County (Little Rock, AR 336)  383  336  88% SUA;  Faulkner County (Conway, AR 65)  113  65  58% SUA;  Saline County (Little Rock, AR 65)  107  65  60% SUA;  Lonoke County (Little Rock, AR 31)  68  31  45% SUA.

FAYETTEVILLE-SPRINGDALE-ROGERS, AR-MO 424/424 (2/2): Benton County (Fayetteville--Springdale--Rogers, AR--MO 147)  221  147  66% SUA;  Washington County (Fayetteville--Springdale--Rogers, AR--MO 148)  203  148  73% SUA.

FORT SMITH, AR-OK 188/188 (2/2): Sebastian County (Fort Smith, AR--OK 91)  126  91  72% SUA;  Crawford County (Fort Smith, AR--OK 30)  62  30  48% SUA.

JONESBORO, AR 96/96 (1/1): Craighead County (Jonesboro, AR 65)  96  65  68% SUA.

PINE BLUFF, AR 77/77 (1/1): Jefferson County (Pine Bluff, AR 53)  77  53  69% SUA.

HOT SPRINGS, AR 96/96 (1/1): Garland County (Hot Springs, AR 55)  96  55  57% SUA.

TEXARKANA, TX-AR 43/43 (1/1): Miller County (Texarkana--Texarkana, TX--AR 26)  43  26  60% SUA.

MEMPHIS, TN-MS-AR 51/51 (1/1): Crittenden County (Memphis, TN--MS--AR 40)  51  40  79% SUA.


Florida

MIAMI-FORT LAUDERDALE-WEST PALM BEACH, FL 5565/5565 (3/3): Miami-Dade County (Miami, FL 2486)  2496  2486  100% SUA;  Broward County (Miami, FL 1748)  1748  1748  100% SUA;  Palm Beach County (Miami, FL 1263)  1320  1263  96% SUA.

TAMPA-ST. PETERSBURG-CLEARWATER, FL 2783/2783 (4/4): Hillsborough County (Tampa--St. Petersburg, FL 1186; Lakeland, FL 0)  1229  1186  96% SUA;  Pinellas County (Tampa--St. Petersburg, FL 914)  917  914  100% SUA;  Pasco County (Tampa--St. Petersburg, FL 342; Zephyrhills, FL 67; Spring Hill, FL 12)  465  421  91% SUA;  Hernando County (Spring Hill, FL 136)  173  136  79% SUA.

ORLANDO-KISSIMMEE-SANFORD, FL 2134/2134 (4/4): Orange County (Orlando, FL 1011; Kissimmee, FL 106)  1146  1117  97% SUA;  Seminole County (Orlando, FL 409)  423  409  97% SUA;  Lake County (Leesburg--Eustis--Tavares, FL 130; Orlando, FL 82; Lady Lake--The Villages, FL 17)  297  229  77% SUA;  Osceola County (Kissimmee, FL 208; Orlando, FL 8)  269  216  80% SUA.

JACKSONVILLE, FL 1245/1346 (3/5): Duval County (Jacksonville, FL 839)  864  839  97% SUA;  Clay County (Jacksonville, FL 157)  191  157  82% SUA;  St. Johns County (St. Augustine, FL 69; Jacksonville, FL 69)  190  138  73% SUA;  Nassau County ()  73  0  0% NSUA;  Baker County ()  27  0  0% NSUA.

NORTH PORT-SARASOTA-BRADENTON, FL 702/702 (2/2): Sarasota County (Sarasota--Bradenton, FL 304; North Port--Port Charlotte, FL 59)  379  363  96% SUA;  Manatee County (Sarasota--Bradenton, FL 304)  323  304  94% SUA.

CAPE CORAL-FORT MYERS, FL 619/619 (1/1): Lee County (Cape Coral, FL 530; Bonita Springs, FL 51)  619  581  94% SUA.

LAKELAND-WINTER HAVEN, FL 602/602 (1/1): Polk County (Lakeland, FL 262; Winter Haven, FL 201; Tampa--St. Petersburg, FL 0)  602  463  77% SUA.

DELTONA-DAYTONA BEACH-ORMOND BEACH, FL 590/590 (2/2): Volusia County (Palm Coast--Daytona Beach--Port Orange, FL 263; Deltona, FL 182)  495  445  90% SUA;  Flagler County (Palm Coast--Daytona Beach--Port Orange, FL 86)  96  86  90% SUA.

PALM BAY-MELBOURNE-TITUSVILLE, FL 543/543 (1/1): Brevard County (Palm Bay--Melbourne, FL 453; Titusville, FL 54; Sebastian--Vero Beach South--Florida Ridge, FL 9)  543  516  95% SUA.

PENSACOLA-FERRY PASS-BRENT, FL 449/449 (2/2): Escambia County (Pensacola, FL--AL 270)  298  270  91% SUA;  Santa Rosa County (Pensacola, FL--AL 64; Fort Walton Beach--Navarre--Wright, FL 56)  151  120  79% SUA.

PORT ST. LUCIE, FL 424/424 (2/2): St. Lucie County (Port St. Lucie, FL 254; Sebastian--Vero Beach South--Florida Ridge, FL 15)  278  268  97% SUA;  Martin County (Port St. Lucie, FL 123; Miami, FL 5)  146  127  87% SUA.

TALLAHASSEE, FL 275/275 (1/1): Leon County (Tallahassee, FL 240)  275  240  87% SUA.

OCALA, FL 331/331 (1/1): Marion County (Ocala, FL 157; Lady Lake--The Villages, FL 44; Homosassa Springs--Beverly Hills--Citrus Springs, FL 2)  331  203  61% SUA.

NAPLES-IMMOKALEE-MARCO ISLAND, FL 322/322 (1/1): Collier County (Bonita Springs, FL 259)  322  259  81% SUA.

GAINESVILLE, FL 247/247 (1/1): Alachua County (Gainesville, FL 188)  247  188  76% SUA.

CRESTVIEW-FORT WALTON BEACH-DESTIN, FL 181/236 (1/2): Okaloosa County (Fort Walton Beach--Navarre--Wright, FL 127)  181  127  70% SUA;  Walton County (Fort Walton Beach--Navarre--Wright, FL 9; Panama City, FL 1)  55  10  18% NSUA.

PANAMA CITY, FL 169/169 (1/1): Bay County (Panama City, FL 143)  169  143  85% SUA.

PUNTA GORDA, FL 160/160 (1/1): Charlotte County (North Port--Port Charlotte, FL 109; Sarasota--Bradenton, FL 35; Cape Coral, FL 0)  160  145  90% SUA.

HOMOSASSA SPRINGS, FL 141/141 (1/1): Citrus County (Homosassa Springs--Beverly Hills--Citrus Springs, FL 79)  141  79  56% SUA.

SEBASTIAN-VERO BEACH, FL 138/138 (1/1): Indian River County (Sebastian--Vero Beach South--Florida Ridge, FL 126)  138  126  91% SUA.

SEBRING, FL 99/99 (1/1): Highlands County (Sebring--Avon Park, FL 62)  99  62  62% SUA.

THE VILLAGES, FL 93/93 (1/1): Sumter County (Lady Lake--The Villages, FL 52; Leesburg--Eustis--Tavares, FL 2)  93  54  58% SUA.

[muon2]

I don't see that you are respecting the UCC, you are splitting off Autauga and Elmore from Montgomery, and when you go into Houston, Tuscaloosa, or Mobile, you are also splitting a UCC.

Why do districts have to be contiguous?

The UCC includes a cluster of counties. A single county urbanized core that is chopped gets penalized for the chop like any other county, and there is no other penalty. There's no good reason to double penalize a county split just because its urbanized. That's why we've been asking for a map with just the multi-county clusters within a state. A cluster with less than the population for one CD gets penalized for a chop even if counties are whole as in my split of the Montgomery cluster. Larger clusters get penalties for chops beyond the minimum needed.

Contiguity is a fundamental requirement of the courts (and most state constitutions). There are rare exceptions permitted, but those relate to political units that have a small discontiguous fragment. An example would be the fragment of West Feliciana county cut off by the river and separated by Pointe Coupee. This type of split has been upheld in state legislative maps that require whole political units. However, even when permitted that type of split cannot be used to bridge to other units that would be otherwise disconnected.

The Montgomery urban cluster overlaps the Black Belt minority cluster and since the overlap runs across the entire state counties in the belt will have to be removed from either the east or west side. The two plans I put up show both examples. The best I could get with whole counties and the full MCC and UCC was 46.4% BVAP. That led me to conclude that I would sacrifice the UCC chop since the VRA district takes precedence. A CD that links Birmingham to the Black Belt could preserve the Montgomery UCC, but would involve an unneeded chop of the Black Belt.

[muon2]

What a court would object to as a racial gerrymander, and what the VRA requires, are too different things. What we are arguing about is what a court would demand, both knowing the answer is uncertain, but with quite different takes.

Anyway, the black CD I think would be most likely to trigger the VRA is the one below, which I think Train already drew. It hews to good redistricting principles, more or less, with but the contiguous stab into Tuscaloosa. Do you disagree?

[\

I think this does as does my second plan with the western corridor. Where we disagree is that I believe that once a VRA district is required the court has less concern about how you construct the complying district. That is to say that there is a difference between the proof of principle district and the complying district. The proof of principle district is complying, but the complying district may not be proof of principle.

Consider a minority population that meets the Gingles test and one can draw a district to show that it meets the test. If the surrounding white population is sufficiently Dem an alternate district can be drawn that is 46% BVAP yet is demonstrably able to elect the candidate of choice of the black population by recognized statistical modelling of election results. That 46% district will be upheld by the court as satisfying the VRA even though it would not meet the Gingles test by itself. This happened in this last redistricting cycle so it is not speculation. Dems in OH were also prepared to do that in Cuyahoga had they won the map, basing it on the Cleveland-Akron potential CD.

Based on that it seems to me that if a district is shown to be necessary the mapper can substitute an alternate district that might not have triggered the VRA yet complies with the VRA. I don't see why it shouldn't be acceptable in these rules. My Mobile-Montgomery district certainly fits this case. We can debate whether it would or wouldn't trigger the VRA or be required by the court, but that isn't the question once we have Train's map or my plan B to trigger section 2.

[jimrtex]

Edit: Correction to map only.




Georgia

ATLANTA-SANDY SPRINGS-ROSWELL, GA 4901/5081 (17/19): Fulton County (Atlanta, GA 911)  921  911  99% SUA;  Gwinnett County (Atlanta, GA 801; Gainesville, GA 0)  805  801  100% SUA;  DeKalb County (Atlanta, GA 690)  692  690  100% SUA;  Cobb County (Atlanta, GA 686)  688  686  100% SUA;  Clayton County (Atlanta, GA 257)  259  257  99% SUA;  Cherokee County (Atlanta, GA 176)  214  176  82% SUA;  Henry County (Atlanta, GA 176)  204  176  86% SUA;  Forsyth County (Atlanta, GA 155; Gainesville, GA 3)  176  158  90% SUA;  Paulding County (Atlanta, GA 110)  142  110  78% SUA;  Douglas County (Atlanta, GA 112)  132  112  84% SUA;  Coweta County (Atlanta, GA 83)  127  83  65% SUA;  Carroll County (Atlanta, GA 21)  111  21  19% NSUA;  Fayette County (Atlanta, GA 87)  107  87  82% SUA;  Bartow County (Cartersville, GA 52; Atlanta, GA 8)  100  60  60% SUA;  Newton County (Atlanta, GA 69)  100  69  69% SUA;  Rockdale County (Atlanta, GA 72)  85  72  85% SUA;  Walton County (Atlanta, GA 28)  84  28  33% SUA;  Barrow County (Atlanta, GA 12)  69  12  17% NSUA;  Spalding County (Atlanta, GA 37)  64  37  58% SUA.

AUGUSTA-RICHMOND COUNTY, GA-SC 325/325 (2/2): Richmond County (Augusta-Richmond County, GA--SC 182)  201  182  91% SUA;  Columbia County (Augusta-Richmond County, GA--SC 101)  124  101  82% SUA.

SAVANNAH, GA 295/348 (2/3): Chatham County (Savannah, GA 250)  265  250  94% SUA;  Effingham County (Savannah, GA 1)  52  1  3% NSUA;  Bryan County (Savannah, GA 9)  30  9  31% SUA.

COLUMBUS, GA-AL 201/201 (2/2): Muscogee County (Columbus, GA--AL 184)  190  184  97% SUA;  Chattahoochee County (Columbus, GA--AL 8)  11  8  70% SUA.

MACON, GA 156/156 (1/1): Bibb County (Macon, GA 132; Warner Robins, GA 1)  156  133  86% SUA.

ATHENS-CLARKE COUNTY, GA 150/150 (2/2): Clarke County (Athens-Clarke County, GA 110)  117  110  94% SUA;  Oconee County (Athens-Clarke County, GA 16)  33  16  50% SUA.

GAINESVILLE, GA 180/180 (1/1): Hall County (Gainesville, GA 127; Atlanta, GA 13)  180  140  78% SUA.

WARNER ROBINS, GA 140/168 (1/2): Houston County (Warner Robins, GA 126)  140  126  90% SUA;  Peach County (Warner Robins, GA 7)  28  7  23% NSUA.

ALBANY, GA 123/123 (2/2): Dougherty County (Albany, GA 81)  95  81  86% SUA;  Lee County (Albany, GA 14)  28  14  51% SUA.

CHATTANOOGA, TN-GA 133/133 (2/2): Walker County (Chattanooga, TN--GA 31)  69  31  45% SUA;  Catoosa County (Chattanooga, TN--GA 45; Dalton, GA 1)  64  46  72% SUA.

DALTON, GA 142/142 (2/2): Whitfield County (Dalton, GA 73)  103  73  71% SUA;  Murray County (Dalton, GA 12)  40  12  30% SUA.

VALDOSTA, GA 109/109 (1/1): Lowndes County (Valdosta, GA 76)  109  76  70% SUA.

BRUNSWICK, GA 80/80 (1/1): Glynn County (Brunswick, GA 51)  80  51  64% SUA.

ROME, GA 96/96 (1/1): Floyd County (Rome, GA 61)  96  61  63% SUA.

HINESVILLE, GA 63/63 (1/1): Liberty County (Hinesville, GA 49)  63  49  77% SUA.


Louisiana

NEW ORLEANS-METAIRIE, LA 1168/1190 (7/8): Jefferson Parish (New Orleans, LA 425)  433  425  98% SUA;  Orleans Parish (New Orleans, LA 342)  344  342  99% SUA;  St. Tammany Parish (Slidell, LA 91; Mandeville--Covington, LA 88)  234  179  77% SUA;  St. Charles Parish (New Orleans, LA 47)  53  47  89% SUA;  St. John the Baptist Parish (New Orleans, LA 37)  46  37  80% SUA;  St. Bernard Parish (New Orleans, LA 34)  36  34  96% SUA;  Plaquemines Parish (New Orleans, LA 15)  23  15  64% SUA;  St. James Parish ()  22  0  0% NSUA.

BATON ROUGE, LA 733/733 (5/5): East Baton Rouge Parish (Baton Rouge, LA 410)  440  410  93% SUA;  Livingston Parish (Baton Rouge, LA 73; Hammond, LA 3)  128  75  59% SUA;  Ascension Parish (Baton Rouge, LA 84)  107  84  78% SUA;  Iberville Parish (Baton Rouge, LA 11)  33  11  34% SUA;  West Baton Rouge Parish (Baton Rouge, LA 17)  24  17  70% SUA.

LAFAYETTE, LA 295/467 (2/5): Lafayette Parish (Lafayette, LA 203)  222  203  92% SUA;  Iberia Parish (Lafayette, LA 46)  73  46  63% SUA;  Acadia Parish (Lafayette, LA 0)  62  0  0% NSUA;  Vermilion Parish (Lafayette, LA 2)  58  2  4% NSUA;  St. Martin Parish (Lafayette, LA 0)  52  0  1% NSUA.

SHREVEPORT-BOSSIER CITY, LA 372/413 (2/3): Caddo Parish (Shreveport, LA 215)  255  215  84% SUA;  Bossier Parish (Shreveport, LA 83)  117  83  71% SUA;  Webster Parish ()  41  0  0% NSUA.

HOUMA-THIBODAUX, LA 208/208 (2/2): Terrebonne Parish (Houma, LA 89)  112  89  79% SUA;  Lafourche Parish (Houma, LA 51; New Orleans, LA 0)  96  52  54% SUA.

LAKE CHARLES, LA 193/193 (1/1): Calcasieu Parish (Lake Charles, LA 143)  193  143  74% SUA.

MONROE, LA 154/154 (1/1): Ouachita Parish (Monroe, LA 117)  154  117  76% SUA.

ALEXANDRIA, LA 132/132 (1/1): Rapides Parish (Alexandria, LA 80)  132  80  60% SUA.

HAMMOND, LA 121/121 (1/1): Tangipahoa Parish (Hammond, LA 65; Mandeville--Covington, LA 1)  121  65  54% SUA.


Mississippi

JACKSON, MS 482/510 (3/4): Hinds County (Jackson, MS 205)  245  205  83% SUA;  Rankin County (Jackson, MS 94)  142  94  66% SUA;  Madison County (Jackson, MS 53)  95  53  56% SUA;  Yazoo County ()  28  0  0% NSUA.

GULFPORT-BILOXI-PASCAGOULA, MS 371/371 (3/3): Harrison County (Gulfport, MS 144)  187  144  77% SUA;  Jackson County (Pascagoula, MS 50; Gulfport, MS 47)  140  98  70% SUA;  Hancock County (Gulfport, MS 17)  44  17  40% SUA.

MEMPHIS, TN-MS-AR 161/161 (1/1): DeSoto County (Memphis, TN--MS--AR 128)  161  128  80% SUA.

HATTIESBURG, MS 131/131 (2/2): Forrest County (Hattiesburg, MS 53)  75  53  70% SUA;  Lamar County (Hattiesburg, MS 28)  56  28  50% SUA.


North Carolina

CHARLOTTE-CONCORD-GASTONIA, NC-SC 1803/1881 (6/7): Mecklenburg County (Charlotte, NC--SC 910)  920  910  99% SUA;  Gaston County (Gastonia, NC--SC 158; Charlotte, NC--SC 1)  206  159  77% SUA;  Union County (Charlotte, NC--SC 146)  201  146  73% SUA;  Cabarrus County (Concord, NC 130; Charlotte, NC--SC 14)  178  144  81% SUA;  Iredell County (Charlotte, NC--SC 99)  159  99  62% SUA;  Rowan County (Concord, NC 85)  138  85  61% SUA;  Lincoln County (Charlotte, NC--SC 11; Gastonia, NC--SC 0)  78  11  14% NSUA.

RALEIGH, NC 1070/1070 (2/2): Wake County (Raleigh, NC 833; Durham, NC 0)  901  833  92% SUA;  Johnston County (Raleigh, NC 37)  169  37  22% SUA.

GREENSBORO-HIGH POINT, NC 488/724 (1/3): Guilford County (Greensboro, NC 312; High Point, NC 107; Burlington, NC 7; Winston-Salem, NC 0)  488  426  87% SUA;  Randolph County (High Point, NC 21)  142  21  15% NSUA;  Rockingham County ()  94  0  0% NSUA.

WINSTON-SALEM, NC 514/602 (2/4): Forsyth County (Winston-Salem, NC 324; High Point, NC 1)  351  325  93% SUA;  Davidson County (Winston-Salem, NC 48; High Point, NC 37; Concord, NC 0)  163  86  53% SUA;  Stokes County (Winston-Salem, NC 12)  47  12  24% NSUA;  Davie County (Winston-Salem, NC 7)  41  7  17% NSUA.

DURHAM-CHAPEL HILL, NC 401/401 (2/2): Durham County (Durham, NC 251; Raleigh, NC 1)  268  253  94% SUA;  Orange County (Durham, NC 91; Burlington, NC 5)  134  96  71% SUA.

ASHEVILLE, NC 404/404 (3/3): Buncombe County (Asheville, NC 181)  238  181  76% SUA;  Henderson County (Asheville, NC 71)  107  71  67% SUA;  Haywood County (Asheville, NC 26)  59  26  45% SUA.

FAYETTEVILLE, NC 366/366 (2/2): Cumberland County (Fayetteville, NC 277)  319  277  87% SUA;  Hoke County (Fayetteville, NC 27)  47  27  57% SUA.

HICKORY-LENOIR-MORGANTON, NC 328/328 (3/3): Catawba County (Hickory, NC 101; Charlotte, NC--SC 0)  154  101  65% SUA;  Burke County (Hickory, NC 52)  91  52  57% SUA;  Caldwell County (Hickory, NC 54)  83  54  66% SUA.

WILMINGTON, NC 203/255 (1/2): New Hanover County (Wilmington, NC 198)  203  198  98% SUA;  Pender County (Wilmington, NC 2)  52  2  4% NSUA.

JACKSONVILLE, NC 178/178 (1/1): Onslow County (Jacksonville, NC 105)  178  105  59% SUA.

GREENVILLE, NC 168/168 (1/1): Pitt County (Greenville, NC 118)  168  118  70% SUA.

ROCKY MOUNT, NC 152/152 (2/2): Nash County (Rocky Mount, NC 50)  96  50  52% SUA;  Edgecombe County (Rocky Mount, NC 17)  57  17  31% SUA.

BURLINGTON, NC 151/151 (1/1): Alamance County (Burlington, NC 108)  151  108  71% SUA.

NEW BERN, NC 104/104 (1/1): Craven County (New Bern, NC 51)  104  51  49% SUA.

GOLDSBORO, NC 123/123 (1/1): Wayne County (Goldsboro, NC 61)  123  61  50% SUA.

MYRTLE BEACH-CONWAY-NORTH MYRTLE BEACH, SC-NC 107/107 (1/1): Brunswick County (Myrtle Beach--Socastee, SC--NC 20; Wilmington, NC 20)  107  40  37% SUA.

VIRGINIA BEACH-NORFOLK-NEWPORT NEWS, VA-NC 0/0 (0/0): None.

[muon2]

This is a change from your previous map that was just posted as corrected earlier in the day. What is the change here, and which should we be using?

This is a corrected version of the metro map for the Southeastern USA.   There is no change in definitions - I had miscolored the map.



Corrections:

Little Rock, AR: I had colored Prairie rather than Perry.
Gainesville, FL: I did not color Gilchrist.
Greensboro-High Point, NC: I had miscolored Randolph and Rockingham.
Oklahoma City, OK: I did not color Grady,
Spartanburg, SC: I did not color Union.
Austin-Round Rock, TX: I did not color Bastrop and Caldwell.

In most of these cases these were potential micropolitan central counties, which became outlying counties of metropolitan areas.   Most fail to qualify as part of the urban county cluster, because they have little or no urbanized area population.

[Torie]

Yes, I agree that if the VRA is triggered, it may be possible to substitute a less than 50% BVAP CD based on the evidence, of less than impermeable racial block voting. What we are arguing about is what effects the trigger. Maybe the CD above does, despite have to do the prong chop into Tuscaloosa.  I have elucidated what I think are most reasonably likely to be the factors. What the odds are, are guesstimates. I just think that the odds are less that the VRA will be triggered if one had only Mobile or Birmingham to go to, than more minority contiguous Tuscaloosa. I also think the trend is towards leashing the VRA vis a vis minority CD creation that violates other redistricting principles, rather than the reverse.

Some of this politically might be moot, if the parties can dicker and veto from a set of maps. Both parties will want a majority minority CD (the Dems because the blacks will insist upon it, the Pubs to stack the Dems).

[muon2]

Yes, I agree that if the VRA is triggered, it may be possible to substitute a less than 50% BVAP CD based on the evidence, of less than impermeable racial block voting. What we are arguing about is what effects the trigger. Maybe the CD above does, despite have to do the prong chop into Tuscaloosa.  I have elucidated what I think are most reasonably likely to be the factors. What the odds are, are guesstimates. I just think that the odds are less that the VRA will be triggered if one had only Mobile or Birmingham to go to, than more minority contiguous Tuscaloosa. I also think the trend is towards leashing the VRA vis a vis minority CD creation that violates other redistricting principles, rather than the reverse.

Some of this politically might be moot, if the parties can dicker and veto from a set of maps. Both parties will want a majority minority CD (the Dems because the blacks will insist upon it, the Pubs to stack the Dems).

So it sounds like we agree that if the Gingles test is met a substitute district may be used that provides for the ability to elect the candidate of choice and hews to generally acceptable redistricting principles - ie our rules and scoring. I think we agree that is the situation in AL, and we should look at the other states for their minority clusters. I'm OK with playing the political card since this is a political process.

Did you have any thoughts about the modified percents for HVAP areas?

[Torie]

What is the HVAP issue? I can't readily find a post above about that. If this is a HCVAP matter, that is circuit specific under the VRA if I recall correctly.

[muon2]

What is the HVAP issue? I can't readily find a post above about that. If this is a HCVAP matter, that is circuit specific under the VRA if I recall correctly.

The issue is to extend the idea of the minority county clusters to Hispanic populations. Because of citizenship the 40%/50% categories don't really work and HCVAP is not available. For the purposes of clustering I suggested that 40%/50% BVAP be replaced by 50%/60% HVAP for clustering. That would be a reasonable proxy for HCVAP or whatever circuit-based value is needed to test for the ability to elect a candidate of choice.

[Torie]

What is the HVAP issue? I can't readily find a post above about that. If this is a HCVAP matter, that is circuit specific under the VRA if I recall correctly.

The issue is to extend the idea of the minority county clusters to Hispanic populations. Because of citizenship the 40%/50% categories don't really work and HCVAP is not available. For the purposes of clustering I suggested that 40%/50% BVAP be replaced by 50%/60% HVAP for clustering. That would be a reasonable proxy for HCVAP or whatever circuit-based value is needed to test for the ability to elect a candidate of choice.

Yes, that makes sense.

On another issue, I know you get penalized or banned (we argue about that), which respect to an excess chop into an urban cluster, but how about one out?  Can a chop in be made larger, thereby forcing a chop out from an urban cluster?  I asked because I can get microchops with the map below, with what presumably is a cut out to the north from the Birmingham urban cluster, rather than taking a bite of the county to the south (assuming the county to the north is not in the cluster, and the county to the south is).

[muon2]

What is the HVAP issue? I can't readily find a post above about that. If this is a HCVAP matter, that is circuit specific under the VRA if I recall correctly.

The issue is to extend the idea of the minority county clusters to Hispanic populations. Because of citizenship the 40%/50% categories don't really work and HCVAP is not available. For the purposes of clustering I suggested that 40%/50% BVAP be replaced by 50%/60% HVAP for clustering. That would be a reasonable proxy for HCVAP or whatever circuit-based value is needed to test for the ability to elect a candidate of choice.

Yes, that makes sense.

On another issue, I know you get penalized or banned (we argue about that), which respect to an excess chop into an urban cluster, but how about one out?  Can a chop in be made larger, thereby forcing a chop out from an urban cluster?  I asked because I can get microchops with the map below, with what presumably is a cut out to the north from the Birmingham urban cluster, rather than taking a bite of the county to the south (assuming the county to the north is not in the cluster, and the county to the south is).

One way to measure UCC chops is to count the number of districts that cover the UCC. In your case you have two for Birmingham and that's the minimum number. If you count by regions then the outward chop means that three CDs span the region, since it pulls in the chopped CD 7, which is one more than needed. Then again if we are still on jimrtex's first map from yesterday Walker is in the UCC so there are three CDs covering the cluster which is one more than needed.

[Sol]

I think, aside from the VRA issues, the biggest issue with this approach is this dilemma- Is it preferable to have a a district that swings in a circle around an urban center to maintain UCCs, or to split the cluster into neighboring rural districts to minimize erosity? A good number of these counties might be more tied to neighboring rural places, after all.

[traininthedistance]

I think, aside from the VRA issues, the biggest issue with this approach is this dilemma- Is it preferable to have a a district that swings in a circle around an urban center to maintain UCCs, or to split the cluster into neighboring rural districts to minimize erosity? A good number of these counties might be more tied to neighboring rural places, after all.

The whole point of this UCC business, as I see it, is to try and come up with an alternative, more stringent definition of "metropolitan area" such that the outlying rural and rural-ish counties in a metro area are free to go in a rural district, whereas the urban centers and well-developed suburbs remain tied together.  Basically, if we're defining our terms right, the counties in the UCC won't be more tied to neighboring rural places- they'll be most tied to the metro center, or at least some associated satellite/edge city.

[Sol]

I think, aside from the VRA issues, the biggest issue with this approach is this dilemma- Is it preferable to have a a district that swings in a circle around an urban center to maintain UCCs, or to split the cluster into neighboring rural districts to minimize erosity? A good number of these counties might be more tied to neighboring rural places, after all.

The whole point of this UCC business, as I see it, is to try and come up with an alternative, more stringent definition of "metropolitan area" such that the outlying rural and rural-ish counties in a metro area are free to go in a rural district, whereas the urban centers and well-developed suburbs remain tied together.  Basically, if we're defining our terms right, the counties in the UCC won't be more tied to neighboring rural places- they'll be most tied to the metro center, or at least some associated satellite/edge city.

I understand. The specific example I had in mind was Charlotte, where you basically have to create 2 districts- A Charlotte based one and a greater suburban one, the latter combining some areas that are quite far apart*. The new district is also quite erose. This is the point where many would question whether putting Gaston county with Union county is any better than splitting the Charlotte UCC.

*I believe the Charlotte UCC is actually greater than 2 districts in population, but I don't believe that's relevant (correct me if I'm wrong! )

[traininthedistance]

I think, aside from the VRA issues, the biggest issue with this approach is this dilemma- Is it preferable to have a a district that swings in a circle around an urban center to maintain UCCs, or to split the cluster into neighboring rural districts to minimize erosity? A good number of these counties might be more tied to neighboring rural places, after all.

The whole point of this UCC business, as I see it, is to try and come up with an alternative, more stringent definition of "metropolitan area" such that the outlying rural and rural-ish counties in a metro area are free to go in a rural district, whereas the urban centers and well-developed suburbs remain tied together.  Basically, if we're defining our terms right, the counties in the UCC won't be more tied to neighboring rural places- they'll be most tied to the metro center, or at least some associated satellite/edge city.

I understand. The specific example I had in mind was Charlotte, where you basically have to create 2 districts- A Charlotte based one and a greater suburban one, the latter combining some areas that are quite far apart*. The new district is also quite erose. This is the point where many would question whether putting Gaston county with Union county is any better than splitting the Charlotte UCC.

*I believe the Charlotte UCC is actually greater than 2 districts in population, but I don't believe that's relevant (correct me if I'm wrong! )

I personally have zero problems with a Gaston-Union district (which would also include a couple hundred thousand folks from Mecklenburg), though of course that is a case where the algorithms might blanch at using Mecklenburg as a bridge.  There is, in my mind, a pretty compelling case that "outer suburbs of Charlotte" is a quite cohesive community of interest.

[muon2]

I think, aside from the VRA issues, the biggest issue with this approach is this dilemma- Is it preferable to have a a district that swings in a circle around an urban center to maintain UCCs, or to split the cluster into neighboring rural districts to minimize erosity? A good number of these counties might be more tied to neighboring rural places, after all.

The whole point of this UCC business, as I see it, is to try and come up with an alternative, more stringent definition of "metropolitan area" such that the outlying rural and rural-ish counties in a metro area are free to go in a rural district, whereas the urban centers and well-developed suburbs remain tied together.  Basically, if we're defining our terms right, the counties in the UCC won't be more tied to neighboring rural places- they'll be most tied to the metro center, or at least some associated satellite/edge city.

I understand. The specific example I had in mind was Charlotte, where you basically have to create 2 districts- A Charlotte based one and a greater suburban one, the latter combining some areas that are quite far apart*. The new district is also quite erose. This is the point where many would question whether putting Gaston county with Union county is any better than splitting the Charlotte UCC.

*I believe the Charlotte UCC is actually greater than 2 districts in population, but I don't believe that's relevant (correct me if I'm wrong! )

The Charlotte UCC is about 2 1/2 CDs so it has to add some population. Mecklenburg has to be chopped, and there's no rule that favors keeping one CD entirely within as opposed to splitting Mecklenburg relatively equally between 2 CDs. The chop count would be one either way, so if you can make a plan less erose that would be preferred.