Greg's Opinion Greg's big blog of whatnot


House Redistricting Hearing: Liveblog

The House Redistricting Committee is holding a hearing with invited testimony only today. They're about to kick off and I'll be adding some notes as I listen in while digging my way out of some research at work.

The Committee will hear from the Texas Legislative Council and the Office of State Demographer on the 2010 Census data.

Dr. Lloyd Potter, the state demographer, is first up. Interesting stat for the day: International migration accounts for somewhere around 25-30% of population growth. Of international migrants, 50% are Hispanic, 24% are Anglo, 18% other, 8% African-American. That's probably a lower share of Hispanic immigration than I suspect a lot of people think exists.

Potter: "It's not a question of whether or not there were any undercounts: there were undercounts." Can't help but wonder if that argument doesn't lend some credence to whether or not Houston actually had 2.1 million in city limits last year.

Rep. Mike Villareal asks Lloyd Potter if there's any analysis of growth in Hispanic population in the suburban areas - essentially something that shows the diffusion of growth in the Hispanic population. Guess I now have another map project to pile onto my to-do list. Short answer: yes, there is Rep. Villareal. I will have something started on that this week.

David Hanna of the Texas Legislative Council is now up. He's talking about the County Line Rule. This is fun stuff. Wish I had some popcorn. The rule comes from Sec. 26, Article III of the Texas Constitution and governs the way county lines can be cut by State Rep districts. This is an area where "One Person/One Vote" rulings have clouded the state law. Hanna also notes the Larios case that busted a Georgia map that was within the 10% population deviation. It's an odd ruling that may or may not be anything to worry about. But lots of states are now wondering whether the 10% pop. deviation is sufficient.

State Supreme Court cases involving County Line Rule and One Persons/One Vote rulings:
- Smith v Craddick Texas Supreme Court - 1971
- Clements v Valles, Texas Supreme Court - 1981

One thing that Hanna notes, and I haven't thought of before: the 10% population deviation isn't necessarily +/-5% ... you can have +8/-2 or something else. As long as it's in the 10% range. And granted that Larios doesn't come back to bite that allowance in the rear end.

Some great examples that Hanna offers: Hays County's population warrants 93.7% of a full district, based on the ideal population size. If you combine it with Blanco County, you get within 33 people of an ideal district population. I think he just redistricted HD45. Ellis County needs to add population, but by adding any contiguous county, you go over the 10% deviation. Therefore, you need to cut any county added. He suggests that Dallas or Tarrant are not sacrosanct from being carved out to add to Ellis. That could be VERY interesting to see.

Jeff Archer, the chief legislative counsel at the Texas Legislative Council is now up. Talking about how the Census data deals specifically with minority voting rights.

Archer gets into CVAP information from his legal angle, but the Reps bring back David Hanna to ask about the logistics of having CVAP data by smaller levels. In this case, he's suggesting that it wouldn't be useful to focus on the individual tract-level data on CVAP, but that TLC may report on CVAP by district. He specifically mentions that for smaller city council districts, CVAP may offer too broad of a margin of error. It's sounding like CVAP data may not even be put into the state's RedApple software system. That's a pretty big admission of the complexity in dealing with that data.

Part of the issue here is that each demographic within each tract has a margin of error. Getting an overall MOE isn't a simple matter of adding up the individual MOEs - it's also a function of the overall sample size. In other words, the bigger the district, the smaller that MOE gets. The smaller a district, the bigger that MOE gets. But it's never just the sum of individual MOEs. I now feel greatly justified for not databasing that bit of information.

Hearing's concluded. Some good useful backgrounder info. I'll see about slicing a clip tomorrow if time permits.

Related Posts:

Comments (10) Trackbacks (0)
  1. I doubt any part of Dallas would be added to Ellis in a house district though I could see Deuell’s SD2 pick up Ellis County. John Carona needs population in SD16and may want some of his old territory in Garland, Rowlett, and Sachse that was taken away from him to beat David Cain ten years ago.

  2. I perked up when Hanna claimed Dallas county was no more “sacrosanct” than any other county when it came to being split for purposes of joining with Ellis. Although the Constitution is not clear about what the order of precedence would be when a county WAS split, I don’t remember that option ever being considered in other congested areas like around Harris county when it would be easy to join part of Harris to Galveston, Brazoria, Fort Bend, or Montgomery surpluses.

    The operative language from Article 3, Section 26 is “…when any one county has more than sufficient population to be entitled to one or more Representatives, such Representative or Representatives shall be apportioned to such county”

  3. I live in Hays county. Hays county was redistricted the instant they released the census numbers. I had been looking at population estimates (from the Census Bureau and the State Data Center) leading up to the release of the census. The only question had been whether Hays county would be big enough to stand on its own, but if not, Blanco county was right next door with about 6% of a district’s population. Easy.

  4. I’d still be stunned if they were to consider cutting into Dallas or Tarrant, but suspect that the members on the committee were equally attentive to that point. If they were to present a map that cut into Dallas or Tarrant, it would possibly create multiple enemies of the plan instead of one or two that might have wished Ellis were made whole with another configuration.

  5. There’s an easy solution to the Ellis county situation. Just to the east is Kaufman county which is currently joined to Henderson county to form HD4. The combination of Kaufman/Henderson is slightly too big now to form a district within range, so by splitting out a small piece of Kaufman (or less likely Henderson) and joining it to Ellis, the two districts are now in balance with minimal disruption.

    Kaufman CAN be joined to Van Zandt and Rains to form a district, but that would mean the pairing of two Republicans (Gooden and Flynn).

  6. Just a note regarding the range of deviation discussion.

    Sure, a 10% overall range of deviation doesn’t have to be + or – 5%, but the +8/-2 example is perhaps not the best. Maybe +5.5/-4.5 would be more realistic.

    There are two circumstances which operate to pull that range away from the extremes. One is purely mathematical. The overall population is a static number, and any overage in one district must be compensated by a loss of population in another district or districts. Thus your range could never be +10/-0 because that 10% would have to come from somewhere. Now assume I have 10 districts and one is 9% over — the only way to achieve the 10% range is to make the other nine all at -1%, with zero degrees of freedom. It gets easier as you move to the 5% number.

    The other, and more relevant constraint is the claim certain counties or groups of counties would have on being made into a whole district or a whole number of districts. Take El Paso, for example. El Paso county has just enough population to merit 4.78 districts. This can be rounded to 5 whole districts if each district averages -4.48% below the ideal population. That’s a narrow deviation to work with but El Paso can lay claim to those 5 whole districts. If someone were to say the deviation will be +6/-4, El Paso could challenge that with the Constitution on their side (see the operative language from Article 3, Section 26 in a previous post) Simultaneously there will be other counties or groups of counties staking out claims on the high end of the deviation range. With the range being pulled in this way from both ends, it will tend towards the center rather than the extremes.

    One area to look at in this regard is HD 61 (represented by Phil King) west of Ft. Worth. The district currently consists of Parker and Wise counties. Using the new census numbers the district would come in at +5.02%, so that works if you just pull the bottom up to -4.98% which still accommodates El Paso. [Of course, there are other ways Parker can be formed into a district.]

  7. If my memory of Stats class is correct (and I’m not willing to claim that it’s perfectly so), the larger number of districts for State Rep seats would mean that you would have a harder time doing something like +8/-2. So the +/-5 is hard to get too far away from. A smaller number of districts strikes me as easier to do an odd +/- balance. I’ll leave it to the math majors in the blogosphere to clarify that. Bottom line, though, is that I don’t see the Lege getting overly creative with the deviation despite whatever mathematical possibilities may exist.

  8. In the 1990 state house redistricting, MALDEF/Lulac or similar groups convinced the state to undersize minority districts -7.5 as I recall, while oversizing Anglo districts +2.5. This was the subject of a legal challenge in Thomas v. Bush, but the dispute was settled before the three judge federal court in Austin ruled upon the legal issue. If the state were to do this in 2011, they likely would face a strong legal challenge (under Shaw) if they again made redistricting decisions with an excessive reliance upon race.

  9. Has anyone ever seen the actual text of ‘Clement v Valles’, and what the reasoning of the Texas Supreme Court was? I haven’t found it online.

    The Texas Constitution doesn’t say anything about single member house districts. It is all about apportionment of representatives among counties or groups of counties. If done strictly by the constitution, then you could have one district consisting solely of Dallas County with 15 representatives elected at large, and another floterial district consisting of Dallas and Ellis counties electing a single member. Both would violate the US constitution, because at large elections would violate the rights of political and and racial and ethnic minorities; and the floterial district would violate the rights of Ellis County voters since they would be so outvoted by Dallas County voters even though they are providing 90% of the population.

    So did the Texas Supreme Court allow one cheat by drawing single member districts. But is the creation of a floterial district and then instantly merging it with other multi-member districts (eg Dallas+Ellis with 16 representative) divided into single member districts considered a double cheat, and only done if necessary to meet US constitutional equality standards?

    From a modern perspective, we would tend to think of chopping Dallas up into 15 districts of 160,000 persons and that the “surplus” is an area in Wilmer or Lancaster that gets attached to Ellis County.

    But in the 2010 case, the double cheat of Ellis+Dallas for 16 representatives may be the best double cheat to get an Ellis containing district within the 10% range. It would be preferred to an Ellis-Kaufman-Henderson combine that would be a two-member district under the Texas Constitution, and I would the read the constitution of only permitting multi-member apportionment to single counties.

    The El Paso case could result in a challenge to the current county line rule interpretation. If someone wonders why the El Paso districts are 3% underpopulated, an argument that Texas was trying to avoid a floterial district doesn’t hold water when there wasn’t one, but simply a district that included eastern part of the county and some smaller counties to the east. And it could violate the Texas Constitution since an apportionment of 4 representatives to El Paso County and a 5th for the surplus and some smaller counties, would more closely comply with the requirement of an apportionment being in accord with population as “nearly as may be”.

  10. In 2001, the House (Delwin Jones) drew a House plan that would have left Wichita County as single district with just a little over 5% below the idea population, even though if Archer County were included it would just barely over. The Senate never considered the House plan, and the LRB ended up drawing the boundaries, including a Wichita-Archer district, and another district (HD-68) that wraps around the two.

    The LRB plan apportioned 25 seats to Harris County, which had more population than the area west of I-35 which the Jones plan had apportioned 25 representatives to. The problem with allowing an unbalanced distribution is that it gives leeway for a cheating. It is sort of like if your shots were hitting the left side of the rim, you move the basket instead of improving your aim. It is only because of not having clever lawyers that Butler is not national champion.

Leave a comment

Trackbacks are disabled.

Fatal error: Call to undefined function xwp_footerx() in /nfs/c01/h12/mnt/4748/domains/ on line 18