2020 Texas Redistricting thread

[jimrtex]

While everyone has been looking at federal redistricting, given that State Senate districts are actually larger than federal ones, opinion of this map?

https://districtr.org/edit/3296







This map should be a safe 19R-12D map. All R districts are at 58.5% Trump or more. All Dem districts are also at 58% Clinton or more.

Granted, I guess with trends and what not it could end up as a dummymander? I also think several of those districts might be illegal because of the VRA?

Senators will be drawing their own map.

The reason why there is a an Amarillo Midland-Odessa district is to keep a senator in Lubbock. When there was a vacancy in 2004 on the Amarillo Midland-Odessa district, there was a special election (open primary) which resulted in a runoff between two Republicans (the lone Democrat in the special election finished fifth of seven (R 36%, R 21%, R 16%, R 14%, D 8%, R 4%, D 2%). In the runoff, the Northern candidate received 83% in Potter, and 85% in Randall, while the southern candidate received 90% in Ector, and 77% in Midland. Only Cochran and Yoakum in the middle were vaguely competitive.

Senators don't want to be challenged either in a primary or general election. They certainly don't want to be paired.

The 2012 redistricting was an incumbent-protection plan, except for carving up the Fort Worth district of Wendy Davis. It had strong bipartisan support (i.e. nobody liked Davis). When the districts were challenged in court by Democrats, they didn't challenge the senate plan. Davis did, and eventually won. The Fort Worth district was restored.

After each redistricting, all senators are up for re-election. Senators elected in 2010 had their 4-year term truncated. When they meet, they draw for terms. 16 get a 4-year term ending in 2016, and 15 get a 2-year term ending in 2014. Ideally, a senator would either be elected for terms of 4, 4, and 2 years; or for terms of 2, 4, and 4 years.

Davis drew a 2-year term, and knowing she would likely lose in 2014 ran for governor. She has since moved from Fort Worth to Austin.

This incumbent protection results in some rather odd looking districts.

So a likely plan will start with the current districts and adjust for population changes, and make sure that senators bases are not attacked. And make sure you maintain the current district numbers.

[jimrtex]

Yes, there is a lot going on here.

1. In the event the lower house of the state legislature flips, the GOP would still control state legislative redistricting through the backup commission set up by the state constitution.  The commission is the LG, AG, Comptroller,  Land Commissioner and Speaker of the State House.  It would be 4R/1D in this scenario and can approve a map with 3 votes.  the state senate is unlikely to be competitive and in any event would be icing on the cake for Dems and irrelevant to this process unless Dems somehow held a 2/3rds majority in both chambers.

That is what happened in 2002 when there was still split control. The plans weren't passed out of the Senate, and the commission drew the maps.

I would actually let the commission draw the maps to begin with. A lot of gerrymandering is to accommodate individual legislators. The 2002 plans were quite reasonable, and the only plans to last the decade without being overturned since the 1960s.

2. There are strict rules in the state constitution on how lower house districts can be drawn, particularly focused on containing entire districts within counties whenever possible, and there are 150 districts so most of them are truly constrained by these requirements.  It isn't really possible to do a hard gerrymander of the lower house.  Indeed, this is part of the reason why the Republican majority there is precarious in 2020.

In 2012, the population of Republican-held districts would have actually been entitled to four additional districts. People prefer to live in districts with a Republican representative. All but one of the pairings were R vs. R, but the replacement districts would be drawn further out in the suburbs.

The reason that the districts in Dallas are so awful was that in pairing R's they wanted to make it a fair fight to determine who would be elected.

With Section 5 gone by, there is not a concern about retrogression, and it may be easier to draw districts representing communities of interest based on race or ethnicity.

3.  By contrast, the state senate is a free for all.  It has only 31 districts, so they are bigger than US House seats, and they can be gerrymandered as aggressively as the majority wants.

You can't get a bill before the full Senate without a 60% vote, and there are ways to filibuster the senate. It is simpler to draw an incumbent protection plan. Senators are not selfless individuals who will give up their supporters for the greater good of their party.

There are always incumbents knocked off in the House primary of the '2' year, because they may facing a 60% or more turnover in constituents, and faces a challenge from a prominent local who they have heard of.

4.  In the scenario where the lower house is Dem-controlled, or has a de facto coalition government with a narrow R majority (it arguably has a coalition government already because of the mechanics of how the speaker is chosen, but this would be even more pronounced if it were down to say, a 77R/73D split), all of the Republicans on the backup commission who would need to sign off on a gerrymander are facing statewide election in 2022, potentially in another Trump midterm.  They may have more reservations about drawing the map too aggressively given that they would have to answer for it in a statewide vote the very next cycle and would not personally benefit from the gerrymandered districts.

That is a naive supposition about voters.

5. If there is divided government, the backup commission has no authority over the congressional map, so there would either have to be a legislative compromise approved by the governor, or it would be drawn by a federal court for the 2022 cycle.  However, mid-decade redistricting is legal in Texas, so Democrats would either have to maintain control of the lower house despite the Republican-drawn map from the backup commission or win the governorship in 2022 to block a Republican gerrymander of the congressional map for future election cycles.

The US Constitution reserves to the state legislatures the drawing of congressional districts. The SCOTUS requires federal courts when drawing boundaries to respect the lines drawn by the legislature. In 2001, the legislature failed to draw congressional district lines. The federal court added in three new Republican districts, and preserved much of 1990's era awful Democrat gerrymander.

The 2003 legislature had a moral obligation to perform the function that they had failed to do in 2001.

6.  There is also the matter of trying to draw districts based on eligible voters or even registered voters instead of total population.  This would advantage Republicans, but they would first have to change existing state law to do this, and then it would go to SCOTUS.  It would have to get through the lower house, and it is likely to be more controversial than just doing a gerrymander in the traditional way.  It also introduces additional risk in that if SCOTUS blocks the eligible voters standard,  Republicans may not control the governorship anymore to pass a new gerrymander, like what happened in VA this decade.

The Constitution used to provide that senate districts be based on eligible voters. Legal guidance was that you couldn't use registered voters, and that you would have to prove the CVAP population, such as having a state census (this is the real reason Democrats don't want a citizenship question on the census, since they are enabling illegal immigration). When the constitution was cleaned up removing obsolete provisions, they eliminated the eligible voter provision along with another provision that said counties could not be split in drawing senate districts, which had been ruled unconstitutional after Reynolds v Sims. The current constitution fixes the number of senatorial districts, but  has no other requirements.

This is likely the reason that Evenwal v Abbott was brought as a challenge to the Texas senate districts. If one interprets the principle of Reynolds v Sims somewhat literally (no one would seriously argue that the phrase "One Man, One Vote" excludes females), but to exclude those who may not legally vote do to age or citizenship, that is:
"One Voter, One Vote" then voters in districts where there are an excess of voters are being denied equal protection.

If the senate was apportioned on the basis of citizenship numbers from the ACS, opponents would have to argue that the ACS was bogus for purposes of determining the population of districts, while at the same time arguing it must be used in applying the Gingles test.

7.  SCOTUS has imposed a very strict standard for how South Texas needs to be drawn based on the VRA.  If SCOTUS backs off of VRA section 2 redistricting standards next decade, expect TX-23 to go from mildly Dem leaning to mildly GOP leaning, or they could use it to pick up more of San Antonio so that TX-35 becomes an all-Travis CD.

This was based on Section 5, barring retrogression. When the legislature drew a McAllen to Austin district, the SCOTUS ruled that it was not a district that afforded Hispanic voters an opportunity to elect a candidate of choice. Since there was no community of interest between the RGV and Austin, the district was suspect as a racial gerrymander, assigning voters to a district on the basis of race. You can no more assign black voters to a black district, than you can assign black students to Booker T Washington High School, and white students to Robert E Lee High School.

Since the RGV-Austin district was not a minority-opportunity district, TX-23 had to be redrawn in order to restore the Hispanic district quota.

Section 2 under the Gingles Test requires compact communities. No one can argue that fajita strips are compact.

8. Texas also has elected state courts, using the same statewide partisan election system as North Carolina for their highest court.  Republicans cannot do much to change the mechanics of judicial elections in Texas without passing a constitutional amendment,  which would require the cooperation of some state legislative Dems just to get it on the ballot and then the amendment would have to pass a statewide vote.  The state supreme court is currently 9R/0D, but it would likely throw out gerrymandered maps passed in 2021 if it flipped to Dem control at any point during the decade, as recently happened in NC and PA.  It could also rule that Texas state law does not permit using an eligible voters standard instead of total population, or reimpose stricter requirements for how South Texas is drawn using state law.

There is currently a court challenge to at-large election of state courts.

[jimrtex]

Can't the Texas Senate remove the 60% threshold by majority vote at the beginning of a session?  Similar to what the US Senate has done for nominee confirmation votes?  How was it lowered from 2/3rds to 60%?  If so, I don't see it as being particularly significant for the redistricting process going forward.  They will just do what the majority wants to do. 

Also, the statewide judicial elections challenge to force majority minority districts has been going on since 2016 and the plaintiffs have gotten nowhere.  Against a backdrop of federal courts relaxing VRA standards applied to redistricting in general, and one of the conservative SCOTUS justices (Thomas) being opposed to any consideration of race in redistricting whatsoever, it's a real stretch.

They could lower the 60% threshold. It used to be 2/3.

The Texas Senate does not have a calendars committee. Regular order is to take bills up in the order they are reported by committee. Traditionally, the first bill on the calendar is called the rosebush bill. The 60% threshold is to suspend regular order and take up another bill.

You can think of it as the crowd lined up to get into Club 31. The line never moves, but a women who flashes a little extra skin will be moved up and admitted by the bouncers.

In practice, the Lieutenant Governor will not recognize a motion to take up a bill out of regular order unless he has a signed sheet of paper with the 19 members who support the order. There also has to be notice given that a senator intends to take a bill up out of order (24 or 48? hours, IIRC).

If you watch the senate, there is usually nothing happening, and then a senator will be recognized and make a motion to take up a bill out of regular order. There may be a short debate on that motion, then the Secretary of the Senate will call the roll in alphabetical order, starting out slow, A, B, C, and speeding up so it sounds like LMNOP.

But I just realized something interesting. The Texas Constitution requires reapportionment of the legislature to happen in the regular session, otherwise it goes to the commission (Article 3, Section 28). That amendment was made in 1948 after the legislature had failed to apportion after the 1930 and 1940 censuses. The legislature then apportioned after the 1950 and 1960 censuses and generally every census thereafter.

The Constitution limits the regular session to 140 days. Statute sets the meeting date to the second Tuesday in January (Jan 12, 2021). The 140th day will be May 31, 2021.

The Commerce Department and the Bureau of the Census have requested a delay in delivering state apportionment until April 2021 (The December 2020 deadline is in statute). They would then deliver the PL 94-171 data in the summer of 2021. Statute currently requires delivery of the data by April 1, 2021. Traditionally, the Census Bureau tries to deliver data to states with earlier deadlines by February, while year round states or states with later 2022 filing deadlines get their data earlier.

Congress will go along with the delay, because Democrats will be concerned about an undercount, and Republicans will want a full enumeration.

It will be impossible for the regular session of the legislature to redistrict. The Constitution can't be changed, and the meeting date can't be changed without a legislative change, which would have to be made in a special session. Even if there were a special session this year, they are unlikely to delay the 2021 regular session, since that among other things will mess up the budget for the biennium that begins July 1, 2021. Texas requires a balanced budget.

If adjustments need to be made for the 2019-2021 biennium they can be made without the legislature meeting.

Theoretically, the LRB might be more willing to reapportion based on CVAP. They don't have to do the horsetrading that happens in the legislature. This happened in 2001, where they drew cleaner maps because they didn't have to pander to different legislators. The SCOTUS would likely rule that using CVAP was discretionary.

[jimrtex]

Doesn't a state law first have to be changed (legislature plus gov signature) to allow CVAP redistricting? 

The legislature passes a law specifying the district boundaries (or in this case, the LRB acting in lieu of the legislature failing to act).

The redistricting committee has been holding hearings in the interim. I'll check to see if CVAP redistricting comes up.

[jimrtex]

In TX the most relevant case is LULAC v Perry (2006). In that case Dems were gaining in the 1990's in TX-23 (Rio Grande Valley) and in 2001 the Pubs substituted a new TX-25, drawn to neutralize the growing Dem power in TX-23 and create an opportunity district elsewhere that they would also likely win based on partisan voter turnout. It seems not unlike your item 4.

In the opening of LULAC, J Kennedy laid out a number of points, one of which was "Another relevant consideration is whether the number of districts in which the minority group forms an effective majority is roughly proportional to its share of the population in the relevant area. Johnson v. De Grandy, 512 U. S. 997 ." Note that the minority group must form an effective majority to be considered as performing. A 50% HCVAP district that is not performing would presumably not count towards the rough proportionality of VRA districts. Cooper allows that a crossover or coalition district can be performing even if the minority is not over 50% VAP.

However, it seems that LULAC would support what you drew in Dallas county, as the facts seem similar to LULAC and they found for the state in the Dallas area. The key is if there is a performing Hispanic VRA CD, don't mess with it, but if there is at best a crossover, coalition, or opportunity district where the Gingles test finds no majority those lines are fair game for political gerrymandering. Note that if the new apportionment dictates that rough proportionality is better served by increasing the number of performing Hispanic CDs, then the burden would be on the state to show why they didn't create one, if a plaintiff shows that it was possible. As in LULAC performing means electing a Dem supported by the majority of Hispanics in the district.

Shouldn't the compact area of minority CVAP predominance be identified before the districts are drawn?

[jimrtex]

One last thing I didn't mention in the first post, the folks drawing the map aren't the state party leaders. It's not a top-down process to draw a redistricting map, but a bottom-up one. The members of the party all give their input, demands, wants, needs, etc, and the map is drawn from there. If the members agree, then there may be some broader goal to the map (such as the NC redistricting plan among Republican members), but otherwise the incumbents and their allies are the ones who draw the map.

I can't speak to TX, but in many states the map is drawn from the top down. The state party and legislative leaders decide what their objectives are and the hire an expert to draw in a way to meet those goals. Rank and file members are shown the draft product and can provide input as to adjustments they wish, but they rarely get a say at the start of the process.

I recall in the 2000s cycle that the chair of the senate redistricting committee bemoaned the fact that the congressional Republicans had not provided input. The target then was more either to knock off Democratic incumbents in Republican-leaning seats or induce Democratic incumbents to switch parties (give a Democrat enough Republicans so they can comfortably be re-elected as a Republican). I don't know that there any Democrats that would work for now. It would be difficult to draw a Republican-leaning seat for Henry Cuellar.

The 2010s senate redistricting plan was a bipartisan redistricting plan, other than drawing Wendy Davis out of her district. It shows the (lack of) respect of her colleagues for her that the plan was passed on a bipartisan vote, and it was not challenged under the VRA by civil rights organizations.

The 2010s House plan was mainly an incumbent protection plan. Where Republicans were paired, there was either an agreement for one to retire or to draw a a fair primary race. Some of the more ugly districts were the results of these practices. Republicans were able to place new seats in the suburbs or shore up their vote share.

Because the Texas legislature is part-time, there may be less of a reason to hold on to a legislative seat. If you are trying to run a business, there may be less incentive to hold onto a $7600/year position.

[jimrtex]

Mirror, mirror on the wall, bet on a high SES white snap back, or the Hispanic/Asian Pub trends, what is the best Pubmander strategy of them all? This map reflects the diversified approach. A little of this, a little of that, but not too much of either. The blue to red hue map is the NYT precinct swing map, which depicts the rather wild and quite surprising shifts. One would never guess that Harris County trended blue if one was unaware that the turnout among high SES WCVAP's is massively higher in the region than that of low SES persons of color, particularly HCVAP's. Will that continue going forward? It did pretty much from 2016 to 2020. The turnout percentage increase seems sluggish with that cohort of eligible voters - up but by no means dramatically.

Clairvoyance is in short supply, at least in my brain. One wild card is to what degree inflation will ramp up. It seems to be ready to reignite in a substantial way at least in some sectors. In fact, it already has in some sectors at least. The cost of shipping and construction materials has close to doubled in the last few months it seems. Whether a resolution of the supply chain choke points will resolve that or not I don’t know, given all the “free” cash floating around financed by government debt. The nexus of macroeconomics and Pubmandering writ large.

What is your crystal ball revealing to you, or did you misplace it?



I hope with some help to have this map with the 2020 POTUS election figures in due course. In the meantime, my metrics are as follows:

I. Minimize VRA risk

1. If a majority compact BCVAP or HCVAP CD can be drawn that is "compact," then a performing CD for that minority must be drawn, per Gingles.

2. I use a broad definition of what is "compact" for purposes of the Gingles trigger test.

3. Minority CD's should not look gerrymandered except to the extent necessary to make them performing.

4. For Hispanic performing CD's, in Texas for Demcroactic Hispanic CD's, that in practice means keeping the BCVAP percentage down. The metric I use is that the HCVAP must be at least 45%, and at least twice as much as the BCVAP, although if the HCVAP is over 50%, there might be a bit of wiggle room on the at least twice as high as the BCVAP percentage metric. I do not count on a 50%+ HCVAP being a safe harbor even if the CD is not in fact a performing Hispanic CD due to higher percentages of HCVAP's not voting in a Dem primary, either because they do not vote at all, or vote in GOP primaries in higher percentages.

II. Avoid county and municipal splits to the extent possible, without materially degrading the efficacy of the gerrymander, and in general make the map look as "attractive as possible consistent with such efficacy.

III. Mimimize dummymander potential.

This one is tricky in Texas. Will higher income and educated urban whites continue trending to the Dems, or even snap back, and will the Hispanic trend to the GOP hold or continue, and how about the turnout among HCVAP's? If the more upscale white trends continue, as opposed to stopping, much less a snap back, while the Hispanic trends are but an ephemeral fling, or they start voting in much higher percentages, and not for the GOP, then or course the map will have some dummymander potential. Ultimately, one cannot plan on worse case scenarios. If the Pubs lose their gains with persons of color, and keep tanking with Pubs, or former Pubs, who disdain Trump, then the Pubs are going to cease being competitive anyway in general. Black turnout in many places in Texas in 2020 was tepid too btw.

The idea is to diversify the risk, and not put the eggs all in one ethnic basket, and try to pad CD's with more Pubs where the trends look most ominous. A lot will depend on events of course, that are hard to predict. A ramp up in inflation I think would tend to facilitate a snap back, since I suspect many of the more upscale white voters tend to be fiscally cautious, and don't like unpleasant surprises, which among other things can be bad for business. But nobody knows for sure what is in store on the macro-economic level. I at least certainly don't.

IV. Minimize CD's that have such a diversity in Pub ranks, that the risk is that they will nominate someone who is an underpeformer, either due to incompetence, or because it is impossible within the Pub coalition within a CD to avoid annoying one group or the other.

V.  Take cognizance of the needs of Pub incumbents, and avoid a situation where their CD has a group of voters in it where they will substantially underperform, to the point of putting the CD in play over time. That is a problem in the Austin area in particular. Sessions and Cloud need to be kept out of the zone for example.

VI. Number CD's so that Pub incumbents keep their same CD number if reasonably possible.

Any VRA litigation will take into account whether you are drawing an incumbent out of Their district, whether you are removing Their district office from Their district. and are removing significant economic activity from Their district.

[jimrtex]

Apparently Loving County came in at far below its estimate and a drop from 2010? Could there have been fraud involved in the inflated estimates?

There are several man camps in the county. There are a lot of temporary housing in the oil fields. People are reluctant to build houses which might go vacant in a few years. There are a lot of RV parks. Someone who is working 7x12 shifts doesn't need much beyond a place to sleep and to eat.

If they have a wife and kids back in Louisiana, they might get counted there by the Census. while estimates show them elsewhere.

I seem to recall the first birth in the county in several decades.

Though it used to be true that more votes were cast in Loving than people living there. Loving is actually quite wealthy (lots of petro$$$$ per capita), so people can live elsewhere on their royalty checks, and vote absentee.

After this was publicized they cleaned up their voter rolls. In 2000, 156 of 211 registered voters voted.

In 2004 it was 80 of 108. In 2020 it was 66 of 111.

Supposedly county elections were hotly contested, with the two parties being akin to the Hatfield and the McCoys, though I seem to recall it was more like the Hatfields and the Hatfields. All counties in Texas have 4 county commissioners so there were like 16 voters per commisioners precincts.

[jimrtex]

And the Supreme Court has held since Wesberry v. Sanders that congressional districts have to be of as close to equal population as possible. Any map which tries to overpopulate Democratic districts is being instantly struck down. Texas Republicans are evil, but they aren't stupid.

3. The constitutional requirement in Art. I, § 2,that Representatives be chosen "by the People of the several States" means that, as nearly as is practicable, one person's vote in a congressional election is to be worth as much as another's.

Representatives are not literally chosen by "the People" but rather by qualified electors.

The House of Representatives shall be composed of Members chosen every second Year by the People of the several States, and the Electors in each State shall have the Qualifications requisite for Electors of the most numerous Branch of the State Legislature.

[jimrtex]

https://twitter.com/scottbraddock/status/1438257150439174153
I imagine this is a strong sign that Texas will deliver for Republicans in congressional redistricting

Let's be honest, how much further can they go? I see net R+1 as the best case scenario for the GOP as they move to take two RGV seats and cede a Dem seat in Austin, but they're pretty close to maxed out and this to me, just indicates they're stupid.

Not really it's really easy to do a 21-10 map. I'll get you started with the D sinks
2 Dallas
3 Houston-Fort Bend
1 Austin-Williamson
1 Austin-Hays
1 San Antonio
1 El Paso
1 RGV

Then you get 19 R seats that are at least Trump+20 and the final 2 are Trump+16 in the RGV

There are 9 districts that are overpopulated by 5% or more. 8 of these are Republican held.

There are 10 districts that are underpopulated by 5% or more. 6 of these are Democratic held.

If you have a Republican seat next to a Democratic seat, what are the likely alignment of the areas along the boundary? They will be more Democrat than the district as a whole.

It makes no sense for a Laredo seat to wrap around into Austin. People in Maverick, Zavala, and Dimmit would prefer to be represented by a senator from Laredo. So we add Guadalupe and Caldwell to SD-19. SD-10 should take more of northern Fort Worth.

[jimrtex]

State Senate map proposal -

https://data.capitol.texas.gov/dataset/plans2101  

https://dvr.capitol.texas.gov/Senate/14/PLANS2101

Looks like 2 D seats in Dallas, 3 in Houston, 1 in Austin, 2 in San Antonio, 1 in El Paso, and I "think" 3 RGV fajita strips?

12 D seats total,  but a lot of seats in Houston and Dallas along with SD-25 won't last the decade.

There is not enough population along the border for 5 districts.

[jimrtex]

This has data on the plans including proposed amendments.

https://data.capitol.texas.gov/topic/redistricting

[jimrtex]

I expect this to get a rather unhappy reception in the courts.

It would be beautiful if this massively backfires and the courts draw a dem majority map

Who would do that?  The crappy right wing Texas courts or the Crappy right wing 5th Cir./Crappy right wing SCOTUS?

Stealing Judicial seats did pay off for Cocaine Mitch..  It's definitely bought his nearly extinct party some time.

If it ends up as a federal case, it would be some random District Court judge (probably in the Western District of Texas, since that's where Austin is) who does the actual drawing, which could be anyone. The Supremes/5th Circuit wouldn't get involved in actually drawing the map.

Why wouldn't the 5th Circ. just overturn the random judge and return it to the legislature's map?  And then the SCOTUS refuse to hear any appeal?  That seems like the likely outcome if it even got that far.  

Well, we're assuming there's a new map at all; obviously if the maps are upheld then it doesn't matter who is hearing the case.

Also, even within the 5th Circuit you could luck out and get liberal judges. I doubt redistricting would be viewed as so important as to result in an en banc rehearing.

Redistricting cases are heard as a three-judge district court panel. The plaintiffs generally forum shop (since a plaintiff is typically a resident of one of the four federal district courts in Texas, they choose one where they hope to get a more favorable judge (e.g. 1990s Southern District in Houston; 2000s Eastern District in Tyler; 2010s Western District in San Antonio).

In the past, the federal cases have been filed as soon as the apportionment numbers have been announced in December. The complaint says that Texas does not have "36" Congressional districts and on information and belief the nefarious legislature will not redistrict or if they do it will be illegal. The judge will tell them to cool their jets, but all cases will be consolidated.

The 2010s case may still be ongoing, or perhaps suit will be filed as soon as the governor signs the bills.

All redistricting cases are directly appealed to the SCOTUS, which does not have discretion to hear them. In 2011, the SCOTUS first heard the Texas redistricting case, they asked when the case would be back before them, and someone jokingly (or not) said "two weeks".

[jimrtex]

I expect this to get a rather unhappy reception in the courts.

It would be beautiful if this massively backfires and the courts draw a dem majority map

Who would do that?  The crappy right wing Texas courts or the Crappy right wing 5th Cir./Crappy right wing SCOTUS?

Stealing Judicial seats did pay off for Cocaine Mitch..  It's definitely bought his nearly extinct party some time.

If it ends up as a federal case, it would be some random District Court judge (probably in the Western District of Texas, since that's where Austin is) who does the actual drawing, which could be anyone. The Supremes/5th Circuit wouldn't get involved in actually drawing the map.

Why wouldn't the 5th Circ. just overturn the random judge and return it to the legislature's map?  And then the SCOTUS refuse to hear any appeal?  That seems like the likely outcome if it even got that far.  

Well, we're assuming there's a new map at all; obviously if the maps are upheld then it doesn't matter who is hearing the case.

Also, even within the 5th Circuit you could luck out and get liberal judges. I doubt redistricting would be viewed as so important as to result in an en banc rehearing.

Redistricting cases are heard as a three-judge district court panel. The plaintiffs generally forum shop (since a plaintiff is typically a resident of one of the four federal district courts in Texas, they choose one where they hope to get a more favorable judge (e.g. 1990s Southern District in Houston; 2000s Eastern District in Tyler; 2010s Western District in San Antonio).

In the past, the federal cases have been filed as soon as the apportionment numbers have been announced in December. The complaint says that Texas does not have "36" Congressional districts and on information and belief the nefarious legislature will not redistrict or if they do it will be illegal. The judge will tell them to cool their jets, but all cases will be consolidated.

The 2010s case may still be ongoing, or perhaps suit will be filed as soon as the governor signs the bills.

All redistricting cases are directly appealed to the SCOTUS, which does not have discretion to hear them. In 2011, the SCOTUS first heard the Texas redistricting case, they asked when the case would be back before them, and someone jokingly (or not) said "two weeks".

The following redistricting cases do not qualify for the three-judge panel procedure, and instead proceed through the federal courts as do nearly all other cases:

A challenge to a congressional districting plan under the Voting Rights Act.
A challenge to a legislative districting plan under the Voting Rights Act.

The complaint argues that the current (2011) maps are malapportioned. Though it includes some complaints under the VRA, it is a challenge to a statewide districting plan.

[jimrtex]

https://twitter.com/alexazura/status/1450201263535271943

Currently there are 33 districts with a majority HCVAP.

28 are underpopulated by a total of 558K, equivalent to almost three districts.

Five are overpopulated by 144K. 81K of this in in HD-117 on the far west side of San Antonio. HD-124 in San Antonio is also slightly overpopulated by 3K. The other five HCVAP majority districts in San Antonio are underpopulated by 61K. HD-117 is stripped of the northern part of the district increasing the HCVAP by nine points. The most populous Bexar districts under the plan are HD-121 and HD-122 in the northern part of the county (and not majority HCVAP). An argument that majority HCVAP districts were overpopulated deliberately or otherwise does not hold water.

Only one of the 13 districts in South Texas is overpopulated (HD-40 in Edinburg). South Texas should lose a district, but instead the districts are underpopulated in the House plan. This is a systemic underpopulation of HCVAP majority districts.

HD-32 is dropped from 50.6% HCVAP to 42.0% because it is partially pushed out of Nueces County.

HD-76 is eliminated in El Paso because El Paso is not entitled to five representatives.

To the extent that some majority HCVAP districts are overpopulated it is because the legislature failed to comply with Texas Constitution and all districts in the county are overpopulated, but the districts are not out of line relative to other districts in the county.

[jimrtex]

https://twitter.com/alexazura/status/1450201263535271943

Currently there are 33 districts with a majority HCVAP.

28 are underpopulated by a total of 558K, equivalent to almost three districts.

Five are overpopulated by 144K. 81K of this in in HD-117 on the far west side of San Antonio. HD-124 in San Antonio is also slightly overpopulated by 3K. The other five HCVAP majority districts in San Antonio are underpopulated by 61K. HD-117 is stripped of the northern part of the district increasing the HCVAP by nine points. The most populous Bexar districts under the plan are HD-121 and HD-122 in the northern part of the county (and not majority HCVAP). An argument that majority HCVAP districts were overpopulated deliberately or otherwise does not hold water.

Only one of the 13 districts in South Texas is overpopulated (HD-40 in Edinburg). South Texas should lose a district, but instead the districts are underpopulated in the House plan. This is a systemic underpopulation of HCVAP majority districts.

HD-32 is dropped from 50.6% HCVAP to 42.0% because it is partially pushed out of Nueces County.

HD-76 is eliminated in El Paso because El Paso is not entitled to five representatives.

To the extent that some majority HCVAP districts are overpopulated it is because the legislature failed to comply with Texas Constitution and all districts in the county are overpopulated, but the districts are not out of line relative to other districts in the county.

There seem to be a lot of Pub unforced errors here. Dumbasses.

I don't know whether you read the complaint or not. It claims that HCVAP majority districts were deliberately overpopulated. That is simply untrue. Within counties with the multiple seats, the HCVAP majority seats were not overpopulated relative to other districts within the county.

Generally speaking (28 of 33) the current majority HCVAP districts are underpopulated. This is because they were areas that did not grow as fast as Texas as a whole (15.9%). In areas along the border where ALL districts have a majority HCVAP population, this should mean a loss of districts.

HCVAP majority districts in inner cities tend to not grow for a number of reasons. (1) There is no room to add houses. If there was a Hispanic family with 6 children in 2010, three of the children are now adults. One is in the Army. One is the first HS and college graduate for the family and now teaching bilingual education in the suburbs, and the third decided to  go into construction like his father, and lives in the suburbs where the houses are being built and he can afford a house with a garage where he can park his truck and not worry about the tools being stolen. The districts may also be gentrifying with high rises for singles who work downtown replacing cottages.

The border area is where the vast majority of majority HCVAP districts exist. There should be a loss of a district. Where they went off the rails is trying to preserve existing districts by underpopulating them. And then for El Paso County which was entitled to less than 4.5 was to extend it outside the county, but not enough for the extra 1/2 representative, but so they could get to the home of the existing representative for Hudspeth County, who happens to live in Eagle Pass (Maverick County).

It is the equivalent of underpopulating congressional districts in New York and California so that none are lost and overpopulating those in Florida and Texas so none are gained.

Shouldn't part of the Gingles Test be that if the minority-majority district would be a violation of Brown v Board of Education if it were designed to be an attendance zone for Lorenzo de Zavala or Booker T Washington High School?

The majority HCVAP district in Fort Worth is hideous. It was expanded enough to get the population back up to par. The HCVAP population dipped slightly below 50%.

I suspect that the plaintiffs will claim that majority HCVAP districts must be underpopulated. Let's say that the quota was 100,000; and it is impossible to draw a majority HCVAP district with that population. But it was possible to draw a district with 80,000 persons, that was majority HCVAP, that the district might not only be drawn, but must be drawn.

[jimrtex]

I just had one little thought. If the Dems gain the state legislature and governorship by the middle of the decade, they could try a mid-decade redistricting, given it should be easy enough to make a 24-14 or so map in the other direction.

Revenge for 2003...

The legislature did not redistrict in 2001.

[jimrtex]

Here we go again ig

Predictions?

Predictions for what?

Redistricting; they’re taking it up again and are gonna make some changes.

For congress, or just the legislative maps?

I think for both, but can’t confirm. There’s a few articles on it but the information is vague.

The main concerns seem to be that they passed the maps too late in the previous session cause of delayed census, and some legal concerns around the maps.

They can’t honestly do much more in terms of maximizing their gerrymanders.

For congressional, they could theoretically crack heavily Hispanic TX-29 in Houston but that could cuss a VRA legal issue. They might also try to optimize the RGV a bit more. Idk.

My understanding is the Texas constitution requires state legislative redistricting to pass in the first *regular* session after the new census data is received.  Because the 2020 census data wasn't received until after the regular session that year ended, maps had to be passed in the special session that would presumably be valid only for 2022 and then replaced for the decade by whatever the 2023 session draws.

There should be no constitutional obligation to redo the congressional map, but there's nothing prohibiting it either in Texas. 

Back in 1950 or so, the legislature was failing to redistrict after the Census. An amendment was passed that said if the legislature failed to redistrict in the first regular season after the census data is received, the Legislative Redistricting Board (LRB) will take over. The LRB is made up of state officers (Lieutenant Governor, AG, Comptroller, House Speaker, and Land Commissioner).

The legislature is likely to produce an incumbent protection plan - in order to be able to pass the plan. The LRB is likely to produce a more radical plan because they are not worried about incumbents and they are a smaller body. They can just let one person, likely the AG, draw the map. The normal procedure is for each body to draw their own map. Then when the bills go over to the opposite chamber, they station doormen at the doors of the chambers which are at opposite ends of the Capitol. They then gavel the bills passed simultaneously.

In 2021 the census data was received after the regular session which is tightly restricted in Texas to end around Memorial Day. It is pretty well recognized law in Texas that the legislature may redistrict at any time (they typically do this to ratify any litigation), or depending when the decision comes down the court will let the plan be used for MCD2, and order the legislature to prepare a new plan for MCD4.

There was litigation in state court that the legislature did not have authority to redistrict last fall. I think the Democrats either wanted a court to draw a map (i.e. if the legislature could not draw a map then nobody could, and a court would have to intervene for equal protection grounds that the districts were not equal population, or maybe they wanted the old districts to be used). In any event the Texas Supreme Court agreed that the redistricting was constitutional (with respect to process).

Since it was understood that the legislature would be drawing new districts, the federal court stayed any litigation. I think this also covered congressional districts.

The district court probably wanted to see how the SCOTUS will rule on the Alabama and Louisiana cases, knowing that the SCOTUS would probably stay any district court ruling.

A curiosity is that senate terms are staggered, but the terms are drawn after all 31 senators are elected. Half will draw an initial 2-year term, and so the district will elect in 2022, 2024, and 2028. The other districts will elect in 2022, 2026, and 2030, with the last election for a 2-year term.

So after this newest redistricting, all 31 senators will face election in 2024. In 2025, there will be a drawing for terms.

There was redistricting before the 1992, 1994, and 1996 elections. But rather than have a drawing in 1997, the Senate consents to a court order, rather actually pass a bill. AG Dan Morales, before he went to prison, said this was OK.

[jimrtex]

Ya'll wanna work on fair maps to submit to the committee.

Not that it'll affect anything, but better to do smtg with your interests ig.

My House plan from 2021

[jimrtex]

Ya'll wanna work on fair maps to submit to the committee.

Not that it'll affect anything, but better to do smtg with your interests ig.

My House plan from 2021

FF map, ig my only thing is don’t Harris and Dallas have to be their own “cluster”?

That is based on an erroneous interpretation of the Texas Constitution.

Dallas is entitled to 13.451 representatives, Harris to 24.349, and Bexar to 10.341. According to the text of the constitution, Dallas is apportioned to 13 representatives, and has a surplus of 0.451. The surplus of 0.451 can be combined with Rockwall which is entitled to 0.555. Collectively, Dallas and Rockwall have a population of 14.006 representatives. We can draw 14 districts with an error of 0.04%. This is in perfect comportment with the directive that representatives be apportioned among the counties as "as nearly as may be."

The The Texas Constitution also specifies that Dallas has 13 representatives. And that Dallas and Rockwall should have one representative. It does not say whether the 13 representatives be represented by district, by position, by plurality (Top 13) or some proportional method.

But the floterial district combining Dallas and Rockwall clearly violates equal protection. Rockwall is supplying 55% of the population, but has only 4.0% of the electorate (more or less depending on the distribution of citizen adults).

The Texas Supreme harmonized this conflict, by saying that you can combine an area containing a population equal to the surplus with adjacent counties (whole or an area containing their surplus population). That is what I did.

Equal protection (and the VRA) forbid election of 13 representatives countywide, at least when there are distinct racial areas that have distinct political opinions, within a county that can be formed into districts. It might be constitutional to use a proportional election method - but that seems inconsistent with other areas electing only one representative.

Combining Dallas and Rockwall is correct for apportionment purposes. It is not correct for electoral purposes if one representative is assigned to the whole of the two counties, and 13 representatives to the whole of Dallas County.

It is nonsensical to say that Lubbock, El Paso, Webb, Hidalgo, Cameron, Nueces, Hays, Travis, McLennan, Denton, Collin, Smith, Jefferson, Galveston, Fort Bend, Montgomery, and Brazos should be combined with adjacent counties in a particular way, and Harris, Dallas, and Bexar may not even though they do have large surpluses.

Ignoring the surplus can produce a systemic bias. Imagine you were going through a cafeteria line, and the cook was dishing out mashed potatoes. They scoop and get roughly 1/2 cup for each customer. This is fair, even though it is not precisely a 1/2 cup. If they took the time to weigh each portion, and then used a teaspoon to add a bit more, or subtract a bit, the line would stall, and some students might not even be served.

Now imagine if you watched, and the cook consistently scooped deeper for students dressed in red, and shallower for those dressed in blue, or perhaps did this based on race. Even though the scoops were within +/- 5% of 1/2 cup there is a systemic bias.

Now what if every district in Dallas County averaged 3.9% less than the quota? That is a systemic bias that delivers extra representation to Dallas County.

Or what if every district in Dallas County (13 districts) averaged 3.5% more than the quota, every district in Bexar County (10 districts) averaged 3.4% more than the quota, and every district in Harris County (24 districts) averaged 1.45% more than the quota?

You would just have apportioned 47 districts to 3 counties that are entitled to 48.140 representatives. You have shorted them a whole representative that you have given to the rest of the state. But it will be impossible to give each other of 102 districts, -1.12% less. In East Texas because of the generally larger county populations you will be forced to have some districts with surpluses, this will mean that in West Texas you can consistently create underpopulated districts.

After Reynolds v. Sims, a provision in the Texas Constitution that limited representation of the largest counties was ruled to be unconstitutional. This provision limited representation to counties that were entitled to more than 7 representatives. It had been added after the 1930 Census when Harris, Dallas, and Bexar counties first would have been apportioned more than 7 representatives. It has since been deleted from the Constitution.

In addition floterial districts were outlawed, and the practice ever since has been to recognize a surplus as an area with a population equivalent to the surplus. Note that in some instances the remainder of the county elected its representatives at large.

1972 House Districts (PDF)

In this map notice that the outlying areas of Tarrant County are combined with Parker County in a single member district, while the central area (Fort Worth, Arlington, HEB, etc. are in a single district electing 9 representatives.

Notice also that HD-17 in Harris County extends into Chambers and Galveston County. In effect it is formed from the surpluses of Harris and Galveston counties and the entirety of Chambers County. Similarly, HD-45 is formed from a surplus of Bexar County and 6 other counties. Rather than being treated as irregular, they are normative.

1976 House Districts

At-large elections were outlawed, requiring the creation of single member districts in Tarrant, Travis, and throughout the state, but the treatment of the surpluses in Harris, Bexar, and Tarrant were treated the same (the area in Tarrant combined with Parker was modified, but it is still an area with the surplus population).

In the 1960's the structural problems (1) restriction of representation for largest counties, and use of floterial districts were eliminated. Because districts had to be added to Harris, Dallas, Bexar and Tarrant counties, the rest of the state had to be reapportioned.

Following the 1970 Census, the legislature completely ignored the Texas Constitution and divided the state into 150 districts often ignoring county lines. In particular they attempted to carve a young Republican from Midland County, Tom Craddick, out of his district. Craddick sued (Craddick v Smith) and won. The Texas Supreme Court required that surpluses be handled in the way they are today.

The 1972 map was drawn after this decision. That map was challenged in federal court appealed to the SCOTUS (White v Regester). That decision (1) upheld the district courts decision mandating single-member districts in Dallas and Bexar counties; (2) remanded to the district court whether other multi-member counties (e.g. Tarrant and Travis) should also be single-member districts. The district court later determined that this should be done in all counties except Hidalgo (i.e. in Hidalgo it is impossible to draw a non-majority Hispanic district). Since that time all districts have been single member, even in Hidalgo County; and (3) that the 10% range of deviation in the plan did not violate equal protection.

This has since been made a national standard by the SCOTUS. Technically it is a burden shifting threshold. If the deviation range is greater than 10%, the defendant (state or local jurisdiction must justify the deviation). If the deviation range is less than 10%, the plaintiff must prove that deviation is not justified (e.g., the deviation was biased with discriminatory intent or significantly less deviation could be achieved while following state requirements such as observing county boundaries).

You will notice in the 1972 map that Red River County is split. It is impossible to draw a district that includes Bowie and one of its neighbors that is within 5% deviation. Relative populations are Bowie (0.908), Cass (0.323), Red River (0.192), and Morris (0.169).

Bowie alone would be -9.2% underpopulated, Bowie and Morris would +7.7% overpopulated. Bowie and Morris might have also blocked formation of other districts.

I personally would have split the larger county, Bowie, putting Texarkana with Cass, and keeping Red River whole. But that is a policy decision, and neither configuration complies with the Texas Constitution. This situation has led to a believe that "you will always have to split a smaller county", rather than the correct observation that "you may have to split a smaller county". The current official map splits Henderson County because of this false belief.

You will also notice that Smith, Brazoria, Jefferson, and Hidalgo have two surplus areas (you may have to zoom in to see the latter two). These are of dubious compliance with the Texas Constitution since these intended to be conversion of floterial districts. In effect, Smith (1.301) would have had its own district, and a majority of the population of two other districts. I don't think these were necessary in 1970, though it would have required rearrangements of districts.

In the 1976 map, the second surplus area in Jefferson County was eliminated when single-member districts were drawn.

After the 1980 Census, the legislature completely forgot all they had been told in the 1970's and willy-nilly chopped up counties. The Texas Supreme Court overruled the map. It had so many errors that they also said that treating large counties that had a surplus as not having a surplus when it could be smudged around within the county.

What is a reasonable definition of when a surplus must be recognized? For moderate-sized counties, the surplus must be recognized when the average district size within a county has a absolute relative deviation per representative is greater than 5%; OR when the total deviation for the county is between 25% and 75% (i.e. roughly 1/2); OR when recognizing a surplus can avoid dividing a small county, or not having the maximum number of whole districts within a county entitled to one or more districts, or having more than one county fragment.

When Texas was mostly rural, counties in eastern Texas were entitled to one or two representatives, or somewhere in between. Counties were formed so that farmers could get to the courthouse when necessary. Farms were roughly of the same size - farmed by a single family without mechanized equipment. So rural counties had roughly had the same number of people. It is only with urbanization that the counties began to diverge.

So if a county had the population equivalent to one or two representatives could be apportioned those representatives. Then the legislature could add a floterial district when a couple of adjacent counties were entitled to about 1.5 representatives (or 1.6 and 1.4; or 1.4 and 1.4; or 1.6 and 1.3; or 1.7 and 1.4; etc.) You can see in the old bill books where amendments have been hand-written in when someone found where an additional floterial district could be created. This was a reasonable solution when elections were administered wholly at the county level. There is no reason not to continue this use practice even with the largest counties.

[jimrtex]

Ya'll wanna work on fair maps to submit to the committee.

Not that it'll affect anything, but better to do smtg with your interests ig.

My House plan from 2021

FF map, ig my only thing is don’t Harris and Dallas have to be their own “cluster”?

If you click on the download arrow at the top of the map, the app will generate a plan report.

[jimrtex]

The Senate Redistricting Committee has scheduled hearings beginning next week.

Senate Redistricting Committee Hearings (PDF)

[jimrtex]

Ah thank you for this thorough analysis.

Under the GOP's lineas, then it makes sense why the GOP just did 13 Dallas seats, 10 Bexar seats, and 24 Harris seats if on net they're denying Ds a seat from all these deep blue counties. This really is a case where every seat matters because of how extreme Ds geography edge is.

Ig that begs the question, on the current map, under what you're stating, would the Hidalgo-Cameron-Willacy group be problematic because that is 2 floterial districts and 35 specifically doesn't is shared between Hidalgo and Cameron without being a clear Floterial for either.

They may have just looked at the raw entitlement and determined that Dallas at 13.451 should be rounded down. This might be a case where they looked at Dallas and decided that it was a good thing.

There was a similar situation in the 2000s redistricting. Harris was intermediate between 24 and 25. The Harris delegation was supposedly united on keeping 25 districts in Harris County. Pete Laney was Speaker at the time, and the head of the Redistricting Committee was from Lubbock. When the map went before the House, Harris had been apportioned 24, and Democrats had drawn the map - with the intent of cutting out a Republican representative.

In 2001, the legislature failed to pass a redistricting bill (in part because action was blocked in the Senate). This resulted in the LRB drawing a map with 25 districts in Harris County, and also more compact boundaries.

If you compare to the US apportionment, you would have to first define multi-seat clusters, and then a super-cluster comprised of all the area in single-member districts. What you would find is that the super-cluster was being apportioned one extra seat.

In the 2020 Census there was some question on how the floterial district should be drawn in the RGV. The Democrats wanted to cut J.M.Lozano out of his seat (because he had switched parties in 2003. Ultimately, they figured out how to configure HD-35 so that it snaked around from La Joya to Harlingen.

Cameron is growing slower than the State, and by 2030 might be down to 2 districts. Willacy is only entitled to 0.100, so HD-37 is close to a full quota but not quite. It also meant that TX-35 had to take a larger share of Cameron (the natural surplus, plus the part  cut out to accommodate Willacy.

The goal of this plan was to protect Ryan Guillen who switched parties. He lives in Starr County. By stretching TX-31 north to Wilson County it made it winnable by a Republican.

I'm not sure whether shared floterials (e.g. Montgomery-Brazos, Harris-Fort Bend, or Hidalgo-Cameron) are preferred or not. I can see it particularly in the case of two large counties that share a district between themselves, since it may mean more districts comprised of whole counties.

The current map also treats Travis and El Paso badly. El Paso is entitled to 4.455 representatives. In the past it has just barely qualified for 5 representative in the county.

El Paso has a surplus of 0.455 based on the definition in the constitution. We would want to combine this with other counties with a population equivalent to 0.545. Instead 0.758 was added.

4.455 + 0.758 equals 5.213. 5.213/5 = 1.043. Three of the five districts have deviations of +4.95%, +4.88%, +4.60%. They pushed these districts to the limit.

The Democrat plan did just the opposite, adding enough to get the total above 4.75, so that the 5 districts would have close to -5.00% deviation.

Imagine you had water glasses which had a black line representing 8 ounces, and two red lines representing +/-5% more or less (7.6 and 8.4 ounces).

There are four glasses filled to the black line, and one that is half full or half empty depending on your political persuasion.

What should be the procedure is to fill the fifth glass up to the black line from a pitcher.

Instead the Republicans poured from the fifth glass into the four other glass, so they were close to the upper red line, and then were able to add more water from the pitcher into the fifth glass.

The Democrats would have poured some of the water from the full glasses into the fifth glass, so they were close to the lower red line. There was then less room in the fifth glass for additional filling.

I think that the following are clearly in violation of the Texas Constitution:

(1) Division of Henderson County.
(2) Failure to draw two districts wholly in Cameron County.
(3) Treatment of surplus of El Paso County.
(4) Treatment of surplus of Travis County.

It is subjective whether Harris, Dallas, and Bexar should be treated as having surpluses. It is clear that by doing so, that the apportionment is "as near as may be".

I think that HD-62 and HD-68 exhibit bad policy choices by connecting Delta to Franklin; and Shackleford to Eastland via quite tenuous connections.

[jimrtex]

Under the GOP's lineas, then it makes sense why the GOP just did 13 Dallas seats, 10 Bexar seats, and 24 Harris seats if on net they're denying Ds a seat from all these deep blue counties. This really is a case where every seat matters because of how extreme Ds geography edge is.

Ig that begs the question, on the current map, under what you're stating, would the Hidalgo-Cameron-Willacy group be problematic because that is 2 floterial districts and 35 specifically doesn't is shared between Hidalgo and Cameron without being a clear Floterial for either.

I remembered another side effect of the clustering, under the rules for consideration of the House Bill for redistricting, any amendment has to fit in to the rest of the map. So if you wanted to change the districts in Dallas County, you would have to accept that there are 13 districts and limit adjustments among some or all of those. You would be constrained to overpopulate most or all of the districts some.

My ideal would have county redistricting boards in each county, appointed in around the x8 year. They would be responsible for identifying communities for redistricting. If each community has less than 10% of the population of a House district, they can be assembled into a House district within 5% deviation of the ideal.

This is sort of done in Massachusetts, where all the cities and towns are required to re-ward to precincts of less than 5000, even though most don't use districts for local governments. The problem there is that communities of interest don't have equal population.

If you instead had county redistricting boards dividing counties into communities of interest in the 08 year, they could concentrate on respecting city and school district boundaries (in cities like they Houston, they would only respect the city limits to the extent that they earn respect).

After the Census, a statewide redistricting board could draw congressional and senate districts that minimize county division, and drawing clusters for House districts.

The county district boards could then redistrict the county commissioner and JP precincts, and any cities or ISDs that have districts. They could also refine any congressional or senate districts in their counties, and do house districts in their areas.