Instances of "locally complete" nesting in states with non-integer H/A:S ratios
       |           

Welcome, Guest. Please login or register.
Did you miss your activation email?
March 29, 2024, 03:45:45 AM
News: Election Simulator 2.0 Released. Senate/Gubernatorial maps, proportional electoral votes, and more - Read more

  Talk Elections
  General Politics
  Political Geography & Demographics (Moderators: muon2, 15 Down, 35 To Go)
  Instances of "locally complete" nesting in states with non-integer H/A:S ratios
« previous next »
Pages: [1]
Author Topic: Instances of "locally complete" nesting in states with non-integer H/A:S ratios  (Read 1035 times)
Kevinstat
Jr. Member
***
Posts: 1,823


Show only this user's posts in this thread
« on: February 08, 2014, 11:32:01 PM »
« edited: February 09, 2014, 08:00:12 PM by Kevinstat »

One instance: Maine has 35 Senators (it could alternatively have 31 or 33 under the present Maine Constitution) and 151 full-fledged (caveat immaterial to this discussion) Represenatives, all elected from single-member districts.  Under last year's redistricting plan (Maine waited until 2013 to redraw it's legislative districts, but future redistricting will be done the year after the census) all districts are with 5% of the ideal population as of and according to the 2010 census.  But the city of Lewiston, with 0.9641/35 and 4.1596/151 of Maine's population, will have one whole Senate district (same as in the past decade) and four whole House districts with no partial districts (it had a partial fifth district in the past decade; it had something like 4.3/151 of Maine's population as of and according to the 2000 census).

The only other instances I'm aware of are in Hawaii, at least in the plan following the 2000 census, where seats were apportioned among the counties rather than statewide and deviations only calculated from the county average.  (So Kauai's one Senate district didn't very at all from the Kauai average!  Shocker!  Wink )  See Jimrtex's thorough analysis at https://uselectionatlas.org/FORUM/index.php?topic=131671.msg2805710#msg2805710 .

Does anyone know of any other instances where a number of whole House/Assembly/[whatever the lower house of the Legislature is called] districts (including multi-member districts) are contained within a single Senate district* with no partial [lower house] districts within that Senate district even though the ratio of lower house members to upper house members in that state is not an integer?

*Preferably a single-member Senate district but I'll allow for multi-member Senate districts as long as the number of Senators in that district isn't a multiple of the denominator in the fractional [lower house member]Sadupper house member] ratio, like not, for example a 2-member, 4-member, 6-member etc. Senate district in a state with a 5:2 House member:Senate member ratio.
Logged
jimrtex
Atlas Icon
*****
Posts: 11,828
Marshall Islands


Show only this user's posts in this thread
« Reply #1 on: February 10, 2014, 12:19:14 PM »

One instance: Maine has 35 Senators (it could alternatively have 31 or 33 under the present Maine Constitution) and 151 full-fledged (caveat immaterial to this discussion) Represenatives, all elected from single-member districts.  Under last year's redistricting plan (Maine waited until 2013 to redraw it's legislative districts, but future redistricting will be done the year after the census) all districts are with 5% of the ideal population as of and according to the 2010 census.  But the city of Lewiston, with 0.9641/35 and 4.1596/151 of Maine's population, will have one whole Senate district (same as in the past decade) and four whole House districts with no partial districts (it had a partial fifth district in the past decade; it had something like 4.3/151 of Maine's population as of and according to the 2000 census).

The only other instances I'm aware of are in Hawaii, at least in the plan following the 2000 census, where seats were apportioned among the counties rather than statewide and deviations only calculated from the county average.  (So Kauai's one Senate district didn't very at all from the Kauai average!  Shocker!  Wink )  See Jimrtex's thorough analysis at https://uselectionatlas.org/FORUM/index.php?topic=131671.msg2805710#msg2805710 .

Does anyone know of any other instances where a number of whole House/Assembly/[whatever the lower house of the Legislature is called] districts (including multi-member districts) are contained within a single Senate district* with no partial [lower house] districts within that Senate district even though the ratio of lower house members to upper house members in that state is not an integer?

*Preferably a single-member Senate district but I'll allow for multi-member Senate districts as long as the number of Senators in that district isn't a multiple of the denominator in the fractional [lower house member]Sadupper house member] ratio, like not, for example a 2-member, 4-member, 6-member etc. Senate district in a state with a 5:2 House member:Senate member ratio.
Hawaii's constitution provides for both senators and house members to be apportioned among island groups using the Huntington-Hill method of equal proportions (current method used for US House).  Moreover, its constitution requires that legislative apportionment be based on the permanent resident population.   The US Census measures the usually resident population, excluding vacationers and employees who maintain a residence elsewhere.

The Hawaii redistricting commission initially excluded military personnel living in on-base housing, as this was the only information readily available from the census that could be attributed to census blocks.  The initial plan was overturned by the Hawaii Supreme Court that said that the apportionment commission had not used the permanent resident population.

The second version excluded just over 100,000 persons, about 8% of the census population.  98% of the excluded population is on Oahu.  This number was based on military pay records, which include a state for purposes of state taxation, as well as information about dependents - which doesn't necessarily indicate co-residency.  They also excluded non-resident college students, based on whether they were paying out-of-state tuition at public universities, or had an out-of-state residence for private universities.   This information was then attributed to census blocks for doing the apportionment and redistricting.

Both the island unit apportionment and permanent resident provisions were challenged in federal court, and upheld (the US Supreme Court just recently denied certiorari).  The challenge was on both equal protection and due process grounds (the adjustments for permanent residence are somewhat fuzzy).

As a result, Hawaii has a 44.2% deviation among senate districts, and 21.6% among house districts.

The apportionment is interesting.  Based on the original plan:

Hawaii           3.43(3)     6.99(7)
Kauai, etc.     1.25(1)     2.55(3)
Maui,etc.        2.88(3)     5.88(6)
Oahu           17.44(18)  35.58(35)

Note that Oahu received one additional senate district, and one fewer house district than it would based on independent rounding.  The senate fractions add to 2, but only one is greater than 0.5.  Such situations generally favor larger units (Oahu could have had 17.33 and still outpaced Hawaii for the final seat). 

On the house side, the fractions add to 3, but all 4 units have a fraction greater than 0.5.  This situation is disfavorable to large units, as is seen by Oahu having a larger fraction than Kauai, but Kauai receiving the final district.

After the population was adjusted:

Hawaii           3.67(4)      7.48(7)
Kauai, etc.     1.33(1)      2.72(3)
Maui,etc.        3.09(3)      6.30(6)
Oahu           16.91(17)  34.50(35)

The entitlements for the neighbor islands increased by about 7%, and caused a shift of a senate district from Oahu to Hawaii.  Curiously, Oahu's raw entitlement declined by 1.08, but no change in the final apportionment, while its senate entitlement declined by 0.53, but a loss of a whole senate seat.

If Hawaii were to gain about 0.02 of a house district (around 500 persons, or even fewer if they moved from Oahu) it would be apportioned an 8th house seat, and nesting could be used for all districts (ie Hawaii 4:8, Kauai 1:3, Maui 3:6, Oahu 17:34)

The Washington constitution would permit similar nesting, and has in the past, but it is generally believed that it would be impossible under a modern (post-Reynolds v Sims) interpretation.

The Maui districts are nested, two house districts for each of the 3 senate districts.  Curiously, there is a fairly large variation in the population among the senate districts (9.9%), but very little between their nested pairs of house districts.

Deviation among district populations is very much calculated and shown based on the island groups, with the deviation from statewide ratios, presented in a secondary manner.

And a curious recommendation from the reapportionment commission to the legislature:

"3. The 2011 Reapportionment Commission recommends that the legislature initiate changes in law to clarify whether or not a state senate election held to fill a vacancy created when an incumbent resigns is a “regular election” for the purpose of computing Senate staggered terms."

I don't know what the circumstances were, but Hawaii elects its senators from staggered 4-year terms, with the staggers maintained based on district continuity across redistricting.  A similar system in California has resulted in 10% of Californians with no senate representation since 2012.
Logged
Kevinstat
Jr. Member
***
Posts: 1,823


Show only this user's posts in this thread
« Reply #2 on: February 11, 2014, 10:42:40 PM »
« Edited: February 11, 2014, 10:50:14 PM by Kevinstat »

New Hampshire has just one Senate district that consists entirely of whole House districts: Senate District 13, consisting of 6 of Nashua's 9 wards.  Eash Nashua ward is a 3-member House district, and there are no floterial districts covering any of that city.  So that's 18:1 Representative:Senator nesting in a state with a 16_2/3:1 Representative:Senator ratio.  I haven't looked at the numbers but if all districts involved are within 5% of the ideal then the math is fairly tight, as six three-member House districts (or 18 single-member House districts) averaging -3.85% would make for a Senate district at +3.85%.  Plausible though as wards in a city, if they have been redrawn in time to be used as the building blocks for House districts, should be close to equal in population.

Three other Senate districts are entirely covered by House districts that are entirely within that Senate district, but they each contain part, but not all, of at least one floterial House district.  And one of those floterials elects three members so it's not like, "Oh, that hardly counts."  It definitely counts.  The only "violation" in one of those three districts (Senate District 4) is a single-member floterial that is probably about 2/3 in that Senate district based on the number of Representatives elected by the two single-member districts contained within it.  But Senate District 3 (Carol County and friends) probably has a greater claim to "near-nesting" than Senate District 4, as its only violation is its inclusion of Winterville Valley (pop. 247 as of and according to the 2010 census), the only town in Grafton County in Senate District 3.  Of course, 247 is around 7 or 8 percent of a population of a single-member New Hampshire House district with no floterial (which Winterville Valley is in) so not as negligible as it might seem at first glance to people in states with much larger populations per State Representative.
Logged
jimrtex
Atlas Icon
*****
Posts: 11,828
Marshall Islands


Show only this user's posts in this thread
« Reply #3 on: February 13, 2014, 01:00:46 AM »

In 2000, El Paso County (TX) had less than 5/150 of the population, but it was close enough to require 5 districts.  Meanwhile, it had more than 1/31 of the population, but close enough that it could have been a single senate district.  Nonetheless it was split, with a district stretching from San Antonio, similar to TX-23 (the previous decade's district stretched from Lubbock, looping east of Midland-Odessa through San Angelo).  Midland-Odessa was in a district with Amarillo, like it still is.

One of the judges in the redistricting case (the same judges considered both legislative and congressional redistricting) said that he thought El Paso should have been a single senate district.  Most redistricting decisions are subjective and political, particularly when it comes to intent.  So plaintiffs enter all kinds of facts as evidence, and judges can include them in their opinion so as to justify their "reasoning".

El Paso is growing faster than the country, but slower than the the state.  In 2010, the senate district was extend outward by a few counties, and by 2020 will probably reach Del Rio.  Meanwhile the San Antonio-based 23rd was stretched further into El Paso which probably contributed to the election of Pete Gallego of Alpine, defeating San Antonio candidates in both the primary and general election.
Logged
Kevinstat
Jr. Member
***
Posts: 1,823


Show only this user's posts in this thread
« Reply #4 on: February 15, 2014, 06:13:01 PM »
« Edited: February 15, 2014, 06:20:29 PM by Kevinstat »

El Paso is growing faster than the country, but slower than the the state.  In 2010, the senate district was extend outward by a few counties, and by 2020 will probably reach Del Rio.  Meanwhile the San Antonio-based 23rd was stretched further into El Paso which probably contributed to the election of Pete Gallego of Alpine, defeating San Antonio candidates in both the primary and general election.

Yet El Paso County still has five whole House districts.  They were all within 5% of the ideal House district population as of 2010 (but all over 4% on the small side), and the Senate district is only +0.68% (it would have been -1.29% without the four small counties next the east) but it seems odd to me for a single-member upper house district in a state with less than an n:1 lower house member:upper house member ratio to consist of all the territory of n lower house districts plus some more territory.

Does Texas's "county line rule" (the prevailing interpretation of the juxtaposition of requirements in the Texas constitution with the federal "one person, one vote" requirements, last I knew) prevent a county that can have a certain number of whole House districts but they would average less than the ideal (like El Paso County), does it prevent it from having territory outside the county added to that last House district to make the numbers closer?

Also, does the "county line rule" apply to the Senate districts at all?  I see at least three partial districts in Dallas County, so I'm thinking no, but that could have been required by some other situation such as VRA requirements.
Logged
jimrtex
Atlas Icon
*****
Posts: 11,828
Marshall Islands


Show only this user's posts in this thread
« Reply #5 on: February 17, 2014, 08:47:27 PM »

El Paso is growing faster than the country, but slower than the the state.  In 2010, the senate district was extend outward by a few counties, and by 2020 will probably reach Del Rio.  Meanwhile the San Antonio-based 23rd was stretched further into El Paso which probably contributed to the election of Pete Gallego of Alpine, defeating San Antonio candidates in both the primary and general election.

Yet El Paso County still has five whole House districts.  They were all within 5% of the ideal House district population as of 2010 (but all over 4% on the small side), and the Senate district is only +0.68% (it would have been -1.29% without the four small counties next the east) but it seems odd to me for a single-member upper house district in a state with less than an n:1 lower house member:upper house member ratio to consist of all the territory of n lower house districts plus some more territory.

Does Texas's "county line rule" (the prevailing interpretation of the juxtaposition of requirements in the Texas constitution with the federal "one person, one vote" requirements, last I knew) prevent a county that can have a certain number of whole House districts but they would average less than the ideal (like El Paso County), does it prevent it from having territory outside the county added to that last House district to make the numbers closer?

Also, does the "county line rule" apply to the Senate districts at all?  I see at least three partial districts in Dallas County, so I'm thinking no, but that could have been required by some other situation such as VRA requirements.
Yes, the current harmonization of the Texas and US Constitution is that you may not add additional territory outside of El Paso, even though it would make for greater population equality.  The current apportionment of five representatives is "good enough" without violating the Texas constitution.  "Greater" equality is not preferred if it means violating or even stretching the Texas constitution.

The Texas Constitution provides for 3 types of house districts:

(1) Multi-county single member districts.
(2) Single-County multi-member districts, including single-county single-member districts.
(3) Single member districts apportioned to several counties, based on either the whole county population and the surplus population from one or more type (2) counties.

The Texas Supreme Court has ruled that type (2) counties may be subdivided into single subdistricts for purposes of electing representatives, and the US Constitution (equal protection of racial minorities) mandates it.  When Texas first started doing this, they often would designate these districts with a number and a letter suffix (57A, 57B, ... 57K).

Type (3) districts are constitutional for apportionment purpose, but not electoral purposes.  If the pre-OMOV interpretation  were used, El Paso would elect 4 representatives at large, and then El Paso plus a few small counties would elect a 5th representative.  But even though the small counties were providing about 1/4 of the population to apportion the 5th representative, they would have only 4 or 5% of the electorate.

So the Texas Supreme Court has said that you can create a district using an area of the larger county containing the surplus population along with smaller counties.   So you could conceivably have a district with El Paso's surplus population beyond 4 districts plus the smaller counties.  But the Texas Supreme Court has said that this may only be done if necessary to comply with equal protection (OMOV). 

So in practice there are three types of districts.

(1) Multi-county single member districts.
(2) Single member districts wholly within a county.
(3) Single member districts containing areas with surpluses from one or more large counties, plus all of zero or more small counties.

Type 3 districts don't precisely conform with the Texas Constitution, but are consistent with the underlying principle.

Henderson County is split under the current plan.  The triggering cause is that Ellis County has not enough population for its own district, but too much population to be paired with a neighbor - and the belief that it can not be combined with Dallas or Tarrant counties.  Since these counties are large enough that you can always create whole districts within 5% deviation, there is a belief that they can not have a partial district.

I think this is wrong.  Dallas could have 14 whole districts, plus a small portion attached to Ellis County.  This would preserve equal protection, and not completely ignore the Texas Constitution.

This would create a hierarchy for comparing possible plans.

(1) Fewest split small counties.   Zero is ideal.
(2) Fewest Type 3 districts with partial surpluses.  Consider a county with a surplus of 100,000.  It might be possible to take an area with 40,000 and use it as part of a Type 3 district, and another area with 60,000 and use it as part of another Type 3 district.  That is the surplus would be split.  Zero is ideal, and so could only be used to avoid use of (1).
(3) Fewest Type 3 districts.   This gives a preference to using surpluses from multiple counties in a single district (eg Cameron+Hidalgo in current plan; vs the surplus in Cameron in one district; and the surplus in Hidalgo in another district.

The Texas Constitution used to provide that no county could be split to create a senate district.

Quote from: Restricted
You must be logged in to read this quote.

This went by the wayside after the OMOV decisions.  The constitution also provided that senate districts be apportioned on the basis of "qualified electors", which provided a somewhat different basis for the population base of the two houses.  Reapportionment guidance was that "qualified electors" could be used if the legislature could find a basis for measuring it - since the census did not.

The 59th legislature (1965) split Harris, Dallas, Bexar, and Tarrant counties, but did it in a way similar to how pseudo-floterial districts are done for the House.  Harris had 4 districts, plus a portion of a 5th;  Dallas had 3 districts, plus part of a 4th; Bexar had 2 districts, plus part of a 3rd; and Tarrant had one whole district plus part of a 2nd.  In addition Cameron was split, with a portion attached to Hidalgo, and the rest going northward to Nueces.  There might not have been a convenient way to put Cameron, Hidalgo, and Nueces in two districts, and if you put them into three, it could get ugly.

Confining districts to within the larger counties, maintains as many districts as possible for the other counties - and would give senators from those areas a shot at keeping their seats.  So they may have been preserving the sense of the constitution, or the senators may have been preserving their own opportunities.

For the 14 whole county districts, the redistricting law simply lists the counties that comprise the district.  The partial county districts are described by metes and bounds, while the mixed districts combine a list of counties, plus a split county (ie "District 9 is composed of Collin, Cooke, Denton, Grayson, Hunt, Kaufman, Rains, and Rockwall Counties and that part of Dallas County not included in Districts 8, 16 and 23.")

This pattern was followed pretty much in 1972.  In 1982, some of the splits shifted outward into suburban counties.   Traditionally, each house reapportions itself, and with only 31 districts, districts can be somewhat preserved between redistricting.  Since all senators are elected after redistricting, there is an added incentive to preserve existing districts.  It is more convenient (easier to get a bill passed) to add parts of high growth counties to several districts.

In the 1992-1994-1996-1998 redistricting the splitting of counties went wholesale.  District 25 included 8 whole counties, and parts of 9 other counties.

The 2002 redistricting was done by the Republican-dominated Legislative Redistricting Board and cleaned up many of the county splits, but added some if it were convenient to dislodge a Democrat.

The 2012 redistricting was passed on a 29-2 vote, so there are some county splits that reflect wishes of individual senators.  There is no reason to split Taylor (Abilene) as the western districts expand to the east.  But Taylor was part of a District 24 before, and so you end up with a split, and District 28 wrapping around it.

The senate redistricting provisions in the constitution were edited in 2001, as part of an amendment to "eliminate obsolete, archaic, redundant, and unnecessary provisions and to clarify, update, and harmonize certain provisions of the Texas Constitution."  If you read the text of the amendment (HJR 75 (77th RS)), you will see that Texas does not limit constitutional amendments to single subjects.

Quote from: Restricted
You must be logged in to read this quote.

Quote from: Restricted
You must be logged in to read this quote.

Notice that there is no longer a population equality standard, or even that redistricting be based on population at all in the Texas Constitution - it is implicitly based entirely on the US Constitution (equal protection).

While they could have stopped with the senate district at the El Paso county line, it doesn't make sense from a community of interest point of view.   It is 550 miles from San Antonio to El Paso.  The other district is based in San Antonio, but there may have been enough (concentration of) population in El Paso to justify a district office.  IIRC, the district lost 60,000 persons from El Paso County, and about 40,000 from the smaller counties.  It is unlikely that there would have been an office for the trans-Pecos, and the counties have a much greater affinity with El Paso than distant San Antonio.
Logged
Pages: [1]  
« previous next »
Jump to:  


Login with username, password and session length

Terms of Service - DMCA Agent and Policy - Privacy Policy and Cookies

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines

Page created in 0.05 seconds with 12 queries.