US House Redistricting: Iowa

[muon2]

I can get a very compact 10 county district around Des Moines that deviates by less than 200 persons from the ideal.

In the map below only whole counties are used with the 2010 census data. The maximum deviation is 264 and the range is 466 persons. I can estimate the NS/EW compactness ratio using Paint, but the total perimeter compactness is harder to get since it requires counting all the boundary pixels in Paint. IA uses both measures to determine compactness.

 

CD 1 (beige) 761,419; NS/EW = 1.04
CD 2 (slate) 761,853; NS/EW = 1.21
CD 3 (forest) 761,696; NS/EW = 0.71 (but a very small perimeter)
CD 4 (pink) 761,387; NS/EW = 1.07

[muon2]

I'd guess the Democrats would be happy with that map; it looks like a 2D-1R map with a Dem-leaning IA-03.

Yup. Here are the 2004 numbers, assuming I coded all the counties right. I tried!

                                                            51.25%                                      
               Bush       Kerry        Total        Bush % 2004 GOP PVI
IA-01   161,235   193,385   354,620   45.47%   -5.78%
IA-02   192,150   212,034   404,184   47.54%   -3.71%
IA-03   179,870   181,875   361,745   49.72%   -1.53%
IA-04   218,702   154,604   373,306   58.59%    7.34%

I'm not sure that the Dem delegation would be pleased. Latham is in CD 3 with Boswell and would present a strong challenge there. Loebsack and Braley are both in CD 2, so presumably one would run in CD 1 and probably relocate as IA congressmen often do after redistricting.

[muon2]

So, if the GOP members of the legislature don't like the first map, they can ask for another one by voting down the first. They can demand a smaller deviation, so this is what they might get. The maximum deviation is 230 with a range of 439. It also provides more even NS/EW dimensions between the districts though it increases the perimeters. My guess is this would suit the GOP better.



CD 1 (beige) 761,769; NS/EW = 0.90
CD 2 (slate) 761,380; NS/EW = 0.91
CD 3 (forest) 761,387; NS/EW = 1.15
CD 4 (pink) 761,819; NS/EW = 0.85

[muon2]

...And the Dems would not be happy with that map.

Can the Dems then veto map 2, and we move on to map 3? LOL.

Either chamber or the governor can reject the first or second map. In doing so, they may specify reasons consistent with the standards for IA redistricting. The next map will be prepared based on any reasons given for rejection. However, after the third map the matter goes to the courts. Presumably, the Dems could ask for even less population variation, though I don't think they have to give a reason. If there is a third map, it can be amended like any other bill.

[muon2]

...And the Dems would not be happy with that map.

Can the Dems then veto map 2, and we move on to map 3? LOL.

Either chamber or the governor can reject the first or second map. In doing so, they may specify reasons consistent with the standards for IA redistricting. The next map will be prepared based on any reasons given for rejection. However, after the third map the matter goes to the courts. Presumably, the Dems could ask for even less population variation, though I don't think they have to give a reason. If there is a third map, it can be amended like any other bill.

In that event, the Dems will indeed drop your map 2 in the waste basket. The last thing they want is to put a reasonably safe Dem seat into play. So what is map 3, assuming the Dems pick the "reason" the fix for which just happens to be the most favorable to them (e.g. requiring that Johnson County be put back into IA-01, and Marion County be excised from IA-03)?  And do the courts have to follow the same law, and beyond following it (assuming it has to), are empowered to draw any map that they desire, without any further guidelines?

OK, so I now assume that map 2 is sent back. The Sen Dems can't really ask for a political result, but they can say that they want better equality. Perhaps they ask that this map improves upon the current one in terms of population equality.

So I consult the history from 10 years ago ...

http://www.legis.state.ia.us/Redist/Redist.html

The first rejected plan:

http://www.legis.state.ia.us/Redist/firstplan.htm

The first (rejected) plan had deviation from 182 to -301 (483 total) , with a mean absolute deviation of 130.2

See Senate resolution rejecting plan at second link.

The second plan description (include some deconstruction of the redistricting law)"

http://www.legis.state.ia.us/Redist/June2001report.htm

The second had a mean absolute deviation of 47, and a range of 40 to -94 (134 total).

Do you know if they did anything about the nesting requirement for senate districts?  50/4 -= 12.5.

... and I improve on those numbers.



CD 1 (beige) 761,534; NS/EW = 1.23
CD 2 (slate) 761,667; NS/EW = 0.55
CD 3 (forest) 761,574; NS/EW = 1.11
CD 4 (pink) 761,580; NS/EW = 0.58

The mean absolute deviation is 39, with a total range of 133 (-55 to +78).

The compactness does suffer for CD 2 and 4, but that could be the price to get near population equality. That's often a more important consideration for the courts in any case.

[muon2]

I was reading up on the IA process and found this presentation from their chief legal counsel. This quote on page 61 was interesting:

You must be logged in to read this quote.

As it turned out there was no need to create any other plan, since the first was accepted. But I went back to some of my notes from before the plan was released and I saw I had a plan with two of the districts more ideal than the accepted plan, but the other two had considerably more deviation.

After some investigation I found that a plan with less deviation than the accepted plan does exist. This plan is much less compact and would violate the convenience definition in statute since there are pairs of counties in a district with wholly intervening counties of a different district. Yet the current CD 4 also has two counties with a wholly intervening county, so it seems that population equality can trump that statute.

So if the first plan had been rejected and the legal counsel advice and Senate resolution were followed, would this plan been offered? Note also that IA law bars a revote on the first plan. Would it go to a third, and if so what would they do?



CD 1: +17
CD 2: +6
CD 3: -15
CD 4: -9

Compare that to the approved plan: -41/+35/+23/-18.

[muon2]

That looks like a pretty optimal GOP gerrymander, actually, if you assume Latham>Boswell in CD-3.  CD-1 is a fantastic Dem vote sink with Iowa City plus most of the river cities (although Braley lives outside it, I think), King should be OK in CD-4, and CD-2 would be pretty competitive for a Cedar Rapids Republican to take on Loebsack.    

Except that IA-03 has Story County in it. It is probably more Dem than the IA-03 currently in place. And IA-03 is where the rubber meets the road. It should be a barn burner, with super high turnout, since Iowa may end up being the epicenter of the election, if Mittens can't get his Hispanic act together. Obama and Mittens will be flushing out every voter who is breathing.

Based on 2004 returns both CD2 and CD3 are close in that map. Bush carried CD2 by 51-49 in the two party vote. Kerry won CD3 by about 400 votes.

[muon2]

That looks like a pretty optimal GOP gerrymander, actually, if you assume Latham>Boswell in CD-3.  CD-1 is a fantastic Dem vote sink with Iowa City plus most of the river cities (although Braley lives outside it, I think), King should be OK in CD-4, and CD-2 would be pretty competitive for a Cedar Rapids Republican to take on Loebsack.    

Except that IA-03 has Story County in it. It is probably more Dem than the IA-03 currently in place. And IA-03 is where the rubber meets the road. It should be a barn burner, with super high turnout, since Iowa may end up being the epicenter of the election, if Mittens can't get his Hispanic act together. Obama and Mittens will be flushing out every voter who is breathing.

And a few hundred who aren't.

Come now. IL is that state east of the Mississippi.

[muon2]

So here is another plan, and I would suggest this is the one the GOP should have held out for. It beats the approved plan in both range (51) and average deviation (13.25). Two of the CDs are within 1 of the ideal population. CD 1 is even reasonably compact and CD 2 has better length-width compactness than the other plan I drew above this.

Looking at the 2 party vote from 2004, Bush beats Kerry in CDs 2 (51.1%), 3 (52.5%), and 4 (53.4%). CD 1 is a Dem sink with Kerry getting 55.5%. As I understand the Iowa process, if this map didn't come up in the first three tries, an amendment to place this plan before the legislature would have been in order. I wonder if anyone there considered this option?



CD 1: +25
CD 2: +1
CD 3: -1
CD 4: -26

[muon2]

As the Iowa GOP has expanded their majorities in both chambers there is a good chance they could ignore the commission .







They can either go for a 4-0 map or a 3-1 map if they chose too. The GOP could probably sweep a 4-0 map in a Biden midterm although 3-1 just make everyone happier and 4-0 would probably go badly just like a pure 3-0 in NM could go a bit badly.

The commission in IA has no real role in redistricting. The body that drafts legislation (LSA) is directed by statute to create redistricting plans according to specific criteria. The legislature can vote plans up or down, but can't amend them. If they reject a plan from the LSA the legislature is required to redirect the LSA as to which factors they need to better address in the next version. If they reject three plans from the LSA the legislature can go on their own, but will have to follow the same statutory criteria. If they don't follow the criteria there is certain to be a court challenge. The current process was put in place in 1980 to avoid regular court challenges to redistricting plans.

[muon2]

Do we have 2020 census figures per county now?

The state has made them available here:
https://www.iowadatacenter.org/data/decennial/Pop2020

In 2010 the population range was 76 with an average deviation of 29.25. I was able to reduce the range to 32 with greater erosity. We did modeling of range vs mean counties per district and a range less than 100 should again be possible.

[muon2]

From the looks of it there are no easy ways to create a compact CD from only Polk County and eight counties around it that also produces very low deviation.

I've found arrangements with less than 100 deviation. One arrangement that included Polk and Dallas plus counties to the south had a deviation of 11.

[muon2]

From the looks of it there are no easy ways to create a compact CD from only Polk County and eight counties around it that also produces very low deviation.

Interesting how populous that area is now- if you added all eight counties, you’d be 100022 people over target.

The most compact/neat option is Polk, Dallas, Jasper, Marshall, Story, Boone (excluding the three at the bottom), which results in a district 2343 under.

If that’s not close enough to target, there’s a closer option. Polk, Dallas, Jasper, Story, Warren, Madison. This is 212 under, and it’s still a decent shape- the three counties not included are the northwest, northeast, and southeast corners of the eight. Wouldn’t surprise me if they go for that.

I don't think they will be concerned about those specific eight. They will try for 4 CDs that minimize deviation while remaining reasonably compact. Population equality is the priority, though it is not absolute. As I noted above they settled for a range of 76, since lower deviations were considerably less compact.

[muon2]

Here is my sample map using the new county data.


The district deviations are blue +26, yellow +55, red +11 and green -91 (rounded). The range is 146 with an average deviation of 45.62.

State law establishes the following criteria for both congressional and state legislative districts:[32]

   Districts must be "convenient and contiguous."
   Districts must "preserve the integrity of political subdivisions like counties and cities."
   Districts must "to the extent consistent with other requirements, [be] reasonably compact–defined in terms of regular polygons, comparisons of length and width, and overall boundary perimeter."

The comparison of length and width rule favors districts that fit in boxes of equal height and length, which is generally accomplished by all four districts. The shape of the yellow district doesn't look great, but the others are quite nice. The boundary between the yellow and blue districts is what creates the poor shape, but it isn't extreme in terms of its length. Since the plan is judged by the total perimeter of all districts, one weak district/boundary usually doesn't hurt the metric much.

I'm sure this isn't the optimal plan since it was only my second stab at it today. I'm curious to see other plans in the same or better population equality ranges.

[muon2]

Here's a IA-3 with a deviation of 7, lol.

I think the IA-4 in this map is near perfect.   IA-1 and IA-2 are both pretty boxy too.

I don't know if the deviation on 1 and 2 is too high though.



https://davesredistricting.org/join/a91c340d-5ea3-4a40-bd8b-c830f11874c2

I think this is what I'd probably draw though -



https://davesredistricting.org/join/c35f45e2-2a4a-47c6-8115-eb98389e0b1f

I like it. I hadn't started playing with a split of Polk and Dallas. West Des Moines and Urbandale are large suburbs and substantially in both counties. The team that follows state law (avoid city splits) to draw the maps may not want to go in that direction.

What are the deviations for all the CDs on both plans?

[muon2]

Here's a IA-3 with a deviation of 7, lol.

I think the IA-4 in this map is near perfect.   IA-1 and IA-2 are both pretty boxy too.

I don't know if the deviation on 1 and 2 is too high though.



https://davesredistricting.org/join/a91c340d-5ea3-4a40-bd8b-c830f11874c2

I think this is what I'd probably draw though -



https://davesredistricting.org/join/c35f45e2-2a4a-47c6-8115-eb98389e0b1f

I like it. I hadn't started playing with a split of Polk and Dallas. West Des Moines and Urbandale are large suburbs and substantially in both counties. The team that follows state law (avoid city splits) to draw the maps may not want to go in that direction.

What are the deviations for all the CDs on both plans?

Hmm...That's interesting about the Municipality splits.   Does this entail that Polk, Dallas, Warren, and Madison all end up in the same district?

First map deviations-
1
-1,900

2   
1,095

3   
 7

4   
799

Second Map-
1
196

2   
171

3   
344

4   
-710

I don't think minor overlaps are as much of a concern as major ones. Keep in mind that each CD will have 25 nested House districts and they are paired to form Senate districts such that only two Senate districts will span CDs - one per pair of adjoining CDs. The minimize county and city split rules apply to the legislative districts, too, and the mappers build all the plans at once.

The legislature doesn't draw the maps, but they can vote them down with instruction as to the statutory criteria they felt weren't adequately met. So if the Pub legislature doesn't like the Polk - Dallas split they can send it back to the mappers with specific instructions to maintain city integrity.

I'm sure all the deviations will be considered too high. Last cycle the deviations were all less than 100. I did an analysis of Atlas submissions back in 2013 and found that one can reasonably predict the deviation one should be able to achieve in a whole county plan. I excerpted from the thread and put it in the Atlas WIKI. For 99 counties and 4 CDs that number is under 100, which is why I expect that there are more precise plans than the one I put up. I may play with your first plan to see how it might be improved for pop equality.

[muon2]

I kinda think aiming for zero population deviation is a rather useless effort.   It doesn't actually change the district in any meaningful way to add/subtract a few hundred people.

But that aside,  here's what I came up with,  I don't know if the IA-1 district's eastern arm would work though.





https://davesredistricting.org/join/c35f45e2-2a4a-47c6-8115-eb98389e0b1f

IA-1 with deviation of 1 and IA-3 with deviation of 2 isn't all that bad, lol.

The issue is that SCOTUS allows for 0 population deviation for CDs. If there are other considerations then the population deviations must be lowest achievable given those considerations. I agree that deviations as large as a 1% range would be fine, especially since the population shifts between Apr 2020 and Nov 2022 are comparable to a few tenths of a percent deviation in most CDs. SCOTUS hasn't been convinced yet, but there are those observers who think that the current composition may be open to loosening the equality requirement.

BTW I found a nice safe compact Dem cluster (includes Johnson, Linn, Black Hawk and Dubuque) with a deviation of only 13! I put it with one of your earlier Polk clusters to get this plan. I wasn't able to get a sub 100 deviation when I put it with your new and improved Polk cluster.



The deviations are blue -13, yellow -48, red +7, green +55. The avg dev is 30.75 and the range is 103.