EG's State Senate Thread

[muon2]

As always, my question is what criteria are you applying?

[muon2]

As always, my question is what criteria are you applying?

For the numbers (the 27/3, for instance)? The rankings, which, are sometimes hard to judge, but do much better than just '08 numbers.

Otherwise I'm not sure what your talking about. I'm just drawing state senate districts.

By criteria I include compactness, erosity, respect for county and/or municipal boundaries, level of population inequality, competitiveness, partisan skew (or lack thereof), VRA, and lots of others that have been used on maps in the past. Fair maps are based on criteria, but there are lots of options out there. Without criteria there's little to prevent gerrymandering, which can be subtle.

[muon2]

As always, my question is what criteria are you applying?

For the numbers (the 27/3, for instance)? The rankings, which, are sometimes hard to judge, but do much better than just '08 numbers.

Otherwise I'm not sure what your talking about. I'm just drawing state senate districts.

By criteria I include compactness, erosity, respect for county and/or municipal boundaries, level of population inequality, competitiveness, partisan skew (or lack thereof), VRA, and lots of others that have been used on maps in the past. Fair maps are based on criteria, but there are lots of options out there. Without criteria there's little to prevent gerrymandering, which can be subtle.

OK, by your description, I'll be doing it with pretty loose standards or criteria. I'll try to take municipal and county lines into consideration, but I'm not going to take a long time just to figure out how to perfectly draw districts. I'll try to keep deviation as low as possible (below 1,000 most of the time), and I'll also try to make it fair for the VRA. But I don't really care for compactness, erosity, or anything else. I'll be trying to make the maps as fair and non-partisan as possible.

But without strong criteria to test the proposals, how should we judge if they are fair?

[muon2]

It shows how useless the Obama '08 numbers can be in legislative races. The 2012 election in VT was 22 D, 8 R. That's why I like to look at other criteria.

[muon2]

You probably need to split District 4 and 9 on a vertical axis instead of a horizontal one.

Also, maybe make the Houston district more Hispanic?

Dists 3 and 4 both are packed with too many Hispanics to avoid a VRA challenge. Those two plus dist 9 would be better if rearranged to provide similar HVAPs.

[muon2]

I'm trying to challenge myself and make an R gerrymander of the NY Senate, or at least draw a composition that resembles the current Senate, but by the looks of things, the state is going to be at least 2-1 D.  Do state Republicans just naturally do better in D-friendly territory here or something?

State Senate Republicans overperform on Long Island and in other places that seem like D-friendly territory.  The NYC suburbs aren't as Democratic as they look on the national level. 

Plus, under the current gerrymander, Senate districts are far from uniform size.  NYC districts are overpopulated relative to the mean.  Upstate districts are underpopulated.  The deviation is pretty large - 10%.

Obama numbers are a very poor measure of political performance in the suburbs of the NE and Great Lakes cities. Consider that Roskam was winning his IL-6 suburban district 58-42 in 2008 while Obama was winning it 56-42, 3 points better than his national result, and Roskam was only running for a second term which is typically a vulnerable election for incumbents. Local suburban Pubs can still make the connections that were true when the GOP dominated the northern 'burbs in decades past.

Legislative districts have a looser legal standard for redistricting at the federal level. In general districts must only adhere to a 10% range standard. Some states impose tighter population standards in their constitutions or by statute.

[muon2]

How did your IA map end up more D than the neutral map actually adopted? The current numbers are 26 D - 24 R.

[muon2]

How did your IA map end up more D than the neutral map actually adopted? The current numbers are 26 D - 24 R.

I don't know, its just the way it turned out, I guess. I don't take sides when drawing these, but I try to group together areas and counties that are similar to each other.

I suspect it's some of the same effect seen elsewhere in the Midwest using 2008 numbers. A 52% Obama district is really lean R, a 53% Obama district is a tossup, and a 54% Obama district is lean D. That would move your 37, 43, and 45 from tossup to lean R, and 7 from lean D to lean R, and 15 from lean D to tossup. That is a shift from 29.5 D to 26.5 D which is a close match to the current composition.

[muon2]

What's the number of black-majority districts in the LA map?

[muon2]

Illinois:



3: 51.4% McCain, 46.6% Obama = Safe R
4: 53.6% McCain, 44.8% Obama = Safe R
5: 54.5% McCain, 43.5% Obama = Safe R
6: 51.6% Obama, 46.8% McCain = Likely R
7: 56.3% Obama, 41.8% McCain = Toss-Up
8: 52.1% McCain, 46.3% Obama = Safe R
9: 51.0% McCain, 47.3% Obama = Safe R
10: 53.7% McCain, 44.5% Obama = Safe R
11: 54.4% McCain, 44.0% Obama = Safe R
12: 56.7% Obama, 41.7% McCain = Lean D
13: 53.7% McCain, 44.5% Obama = Safe R
14: 50.9% McCain, 47.5% Obama = Safe R
15: 50.5% McCain, 47.8% Obama = Safe R
16: 58.6% Obama, 40.1% McCain = Likely D
18: 49.3% Obama, 48.9% McCain = Likely R
19: 54.9% Obama, 43.5% McCain = Toss-Up
58: 55.8% Obama, 42.5% McCain = Toss-Up

12 is Peoria county and is a good bellwether, Toss-up.
16 was just won by a Pub in 2014, Lean-D is a better call.
18 this is safe R.
19 & 58 are likely R. The Dems couldn't do anything to gerrymander a lean D seat out of this area or they would have.

The Dems gerrymandered a seat out of 3 and 5 with a perfect conservative Dem to hold it. They did the same with parts of 11 and 13, though both house seats are Pub and the incumbent is not running next year. They also manufactured a seat out of 6, 8, and 10. They linked 7 to Danville to keep it out of the toss-up category.

The St Louis suburbs (Metro East as it's known here) is reasonable.

Can you post the VAP% for the inner and outer Chicago seats?

[muon2]

That 91% black SD is going to be a target for litigation as illegal packing, and I only count 6 SD that are majority black. Currently there are 9 Chicago area seats that elect black Senators. Most observers would say that anything less than 8 SDs where the black population can elect a candidate of choice is going to run into challenges.

You only have 3 majority and 1 plurality Hispanic SDs. There are 4 current seats, though many think a fifth should have been drawn. 59% HVAP is seen as a minimum to avoid dilution charges, though the fifth seat could possibly be under that.

[muon2]

That 91% black SD is going to be a target for litigation as illegal packing, and I only count 6 SD that are majority black. Currently there are 9 Chicago area seats that elect black Senators. Most observers would say that anything less than 8 SDs where the black population can elect a candidate of choice is going to run into challenges.

You only have 3 majority and 1 plurality Hispanic SDs. There are 4 current seats, though many think a fifth should have been drawn. 59% HVAP is seen as a minimum to avoid dilution charges, though the fifth seat could possibly be under that.

I don't get it. Some of the downstate districts are over 90% white, is that "illegal packing" too? Black people happen to live around other black people, there's nothing I can do about that and I want the districts to look neat with no obscure gerrymanders. And the whole idea of a democracy is for everyone to elect a candidate of their choice, not just a particular race or group. People can elect whoever they want, I think these quotas just make drawing districts harder than it needs to be.

If it was boxed in by other minority districts you might be able to make a case for it, but the district to the west is white, so it is possible to make two districts that will both elect the black population's candidate of choice. Congressional IL-4 has survived decades of challenge despite the bizarre shape because it is constrained by the black CDs. Without that constraint it would be forced to a far more compact shape.

[muon2]

There are cases where sound principles like county lines were overruled because they amounted to cracking the black minority vote. I don't know why they wouldn't do the same for a case of packing. Here's SCOTUS in Johnson v DeGrandy (1994).

You must be logged in to read this quote.

In DeGrandy SCOTUS found that one does not need to maximize the number of minority districts. They held that

You must be logged in to read this quote.

Most plans involving areas with multiple minority districts look at the proportionality question. What you are raising is whether the court would except less than rough proportionality if there are other redistricting principles followed. In this case, it looks the principles are more one of following one's eye to nice shapes. I'm not convinced that would stand up.

I'm also concerned that the chartreuse district takes in a heavily Latino area. The 7th circuits has approved plans with strange shapes in that area to separate the black and Latino areas. If a plan is not providing for both minorities to have sufficient representation, then that can be a problem, too.

[muon2]

You are the lawyer, but every legal expert in redistricting I've interacted with has advised that unless you have concurrence from the minority groups in question then one should aim for rough proportionality or be prepared to show why it is impossible. The key is whether accepted neutral mapping principles can limit a minority group's ability to fairly participate in the political process. Principles that have the effect of denying fair participation I think would be held unconstitutional based on 20 years of the aforementioned advice.

Here's what I would offer for south and west Cook for EG. All districts are within 1000 of the quota. I follow political lines to the extent possible and seek reasonable divisions elsewhere. There are 8 BVAP majority SDs (1-8) and 5 HVAP majority SDs (9-13). The crossover into SE DuPage could be avoided with some ugly gerrymandering of SD 7 into the southside along I-90/94 to keep it over 50% BVAP.

SD 11 is a lower HVAP at 55.8% than I would like, but it is completely outside Chicago while SD 9 and 10 which stay within the city limits. If challenged it would be easy to split the city line and bring all three over 59%. SD 13 is also under 59% (55.7%) but a second NW side Latino district at 59% requires extreme erosity including a finger into NE DuPage.

[muon2]

Where is the case that says, absent minority concurrence, it needs to be "impossible?" Your map has lots and lots of chops. Sure, it's legal, but is it legally required? That, again, is the point. Absent any case law, that addresses the precise points I am making, color me skeptical. Sure, the "experts" might say to be legally safe, because the law is unclear, do x and y, but that is entirely a different matter. To suggest without any legal authority, that a certain path is probably legally dictated, needs to be questioned. I am questioning it. Maybe you should print out this thread, and ask one of your legal experts, if he has any case law, preferably in the form of a memo, to suggest that my point of view is not well taken.

One other point. It is one thing, if a map follows no neutral principles, to question it, if minorities don't get the max, quite another to challenge a map which does. Illinois is a cesspool, so I can see a court saying since you have no credibility at all, don't screw with minorities. If you chopped and erosed for a bunch of white Dem hacks, you had better chop and erose for non white ones as well, in other words.

Make sense?

I'm not sure it does make sense. The DeGrandy case made it clear that minorities do not get the max and that there was no section 2 violation because rough proportionality existed in the plan. It's true that DeGrandy does not mandate rough proportionality, but what you are not addressing is the edict that one must determine "whether minority voters have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice." I don't see how one can sweep that away with neutral mapping principles.

Actually the Dem plan in IL arguably did mess with minorities. The Latinos were shorted and the blacks were given extra seats but at sub 50% BVAP. However since this was all to gain Dem seats, when the Pubs challenged the plan and the Court saw the major minority advocacy groups were either with the Dems or on the sidelines, they took that as a factor that minority voters had adequate opportunity with the plan.

[muon2]

You quoted nothing in the case that one must violate neutral mapping principles to get there, assuming the map in its totality hews to neutral mapping principles. If a map is gerrymandered, I can see a court finding that minorities must get their share of the spoils. If a map does follow neutral principles, to claim a minority is discriminated against is a huge stretch.  

We will then have to disagree on the meaning of "determining whether minority voters have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice," as the standard for a section 2 violation when viewing the plan in its totality. For me if a neutrally drawn map fails to give minorities "their share of the spoils," then they have been denied the opportunity required by the VRA. Congress specifically amended the VRA in 1982 to respond to the Mobile v Bolden decision (1980) and prohibit practices that had discriminatory effect, not just discriminatory intent.

[muon2]

While you were writing I added a note to my post about discriminatory effect in the VRA, though I'm not sure that would sway you.

[muon2]

In response to your codicil, is there anything in the amended statute suggesting that one can find discriminatory effect if neutral line drawing principles are followed? Has that ever been litigated? I keep asking that question. My point of view, my instinct, is that absent something more on point, with facts more on point (what really matters to lawyers is rulings versus fact patterns, not sweeping statements of a general rule, that can be interpreted a host of ways, and was written not really thinking about other fact patterns), if there are multiple maps choices following neutral principles, than one needs to go with the version that maxes minority districts ( at least until the spoils get proportional to population percentages, however that is defined). But that is the extent of it. I would love to argue this before SCOTUS as the rule they should adopt, to get rid of all of this confusion and gaming. It's just awful. It's time to end it.

I think my logic (and those I have listened to) derives from Mobile v Bolden and its aftermath. At that time Mobile elected its city commission at-large. Blacks sued under the VRA at that time saying that they could not elect a candidate of choice. SCOTUS found that blacks could now register and vote without hindrance and the at-large system was racially neutral. They held that the lack of a discriminatory purpose meant that the statute was not a violation of section 2. In response to the decision Congress amended the VRA to make discriminatory effect the standard. The amendment was written to bar facially neutral laws that had a demonstrable impact on a protected minority. Subsequently Mobile changed to single member districts, as did many other jurisdictions where there were significant minority populations.

Your hypothetical is whether this would apply to a system of neutral redistricting rules like the muon2 rules. My sense is that if the plaintiff's can show a disparate effect, such as having significantly fewer opportunities than the minority share of the CVAP population overall, then the ball goes to the state to defend against strict scrutiny. The fact that the rules are neutral isn't a shield under the 1982 VRA.

In some sense this played out in the IL congressional challenge in 2011. The Pubs asserted that there was no opportunity for Latinos to win more than IL-4 despite their growth to be the largest minority in the state. The Court (panel of the 7th circuit) found that there was an effect, but Dems claimed that a plan with 2 CDs at 59% HVAP wasn't feasible. The Pubs didn't submit a map to show that one was possible (they submitted one with just under 50% HVAP) and lost on their claim of a section 2 violation.

[muon2]

An at large voting system is not analogous in my view to neutral line drawing. Neutral line drawing allows for a variety of maps. Pick the one that helps the minorities the most. What you are essentially asserting is that if minority locations mean that reasonable line drawing must be thrown in the trash, I doubt SCOTUS will go there, particularly in this day and age. In reality, neutral line drawing will typically not have much of an adverse effect on minorities, since they tend to be concentrated in inner cities. The toughest case is where, which will be atypical, a minority is on both sides of a line, and to create a minority district, one must do an extra chop. The case is clearest for the first district. We may  have to wait a long time for a case however, generated by the Pubs doing what they tried to do in Illinois. Would a court have demanded another Hispanic CD that was able to elect an Hispanic, but was hideously erose, chopped all over the place, and involved non contiguous Hispanic nodes? That would be the question before the court.

In the case the Pubs lost in Illinois, did the court just focus on 50% HVAP not being sufficient - rather one needed 50% HCVAP, or even a figure higher than 50% HCVAP (if 59% HVAP was higher than 50% HCVAP), or did it also mention the chops and erosity? I do think within a jurisdiction, with respect to a contiguous minority population, one probably does need to go erose to create more minority districts, unless really grotesque perhaps.  That is because, at least in part, as we well know, measuring erosity is subject to a variety of interpretations and ways to measure.

We may have to wait a long time for a case, alas, because in many places, the Dems not maxing minority seats has no partisan impact, or political realities force them to max such seats within the parameters we are arguing about. An example of the former is Maryland, where the Steny Hoyer seat should be black under your standard, but isn't (except maybe not, since 2 black seats is probably proportional to the black population in Maryland), and an example of the latter is New York. However, in most places, with respect to blacks, they tend to be within the big city. It will also be rare, because since one has some choice as to where to chop, typically with neutral criteria, it will be possible to select the chop locations that max the number of minority seats.

I really cannot think of an instance when it comes to CD's anywhere, where there is likely to be a partisan battle over this issue at the moment, other than your example in Illinois, where it appears that the trigger was not in play, because the Hispanic population of the second CD was too low. But for legislative seats, the ground for this type of litigation might be more fertile.

The magic number in IL federal suits is 59.2% HVAP. The court notes that was the value in the 1990 map and Gutierrez won and continues to win in that seat. The Dem plan had one CD at 65.9% HVAP and the Pub plan offered CDs of 59.4% and 46.5% HVAP. Even the Pub expert admitted that the 46.5% HVAP was not going to reach the 59% level by the next Census. A Pub map with an erose second Latino CD at 60% HVAP was available, but the Pubs did not use it in part because it didn't help their claim of political gerrymandering, which also lost.

I agree that it will be hard to find the test case you seek. All I would add is that SCOTUS viewed at-large districts as a neutral districting mechanism, which is how I think they would classify the muon2 rules. I read that as they saw no lines as a subset of lines in general.

[muon2]

At large voting is very different, because with polarized voting, it is a recipe for shutting out the minority entirely. And there is no good policy reason for it. There is a good policy reason for neutral line drawing, and the cases where it screws the minority will be rare, and even then, only at the margins when it comes to proportionality.

True, but that wasn't relevant to the SCOTUS decision. To SCOTUS it was just a neutral districting principle, and at that time they found that it was not a violation of section 2. Because of the policy implications you mention Congress acted to change the VRA, but they didn't identify whether a neutral districting principle was good policy. Congress amended the VRA to say that any policy, good or not, that had a discriminatory effect was a violation. That effect includes anything that diminishes the minority's fair participation in the political process.

[muon2]

Than any lines, no matter how gross, under that interpretation, that do not max minority seats, would be deemed to have a discriminatory effect. I doubt SCOTUS would interpret the statute as amended as abandoning any semblance of a balancing test. Nor do I think any court has demanded gross lines. Rather, I think the courts would find any approach that is grounded on good public policy, even if it does not max minority seats, is not discriminatory. Failing to max minority seats does not automatically mean it's discriminatory. It would be interesting if there were any legislative history supporting the bolded bit of your statement, that " any policy, good or not, that had a discriminatory effect ... ." It would be further interesting to ascertain if the legislative history elucidated what "discriminatory" meant.

I don't know if I've made my references clear, since I'm not trying to support the line I have bolded above. Congress has said that the effect of districting methods, not just intent, cannot deny minorities fair opportunity, and that was said specifically to address a situation of neutral districting as identified by SCOTUS. SCOTUS has clearly said that nothing compels one to max out minority seats, as maximum is not the same as fair opportunity. SCOTUS has said that rough proportionality is a factor to consider to show that there is fair opportunity.

My conclusion is that if there is significantly less than rough proportionality then it is more likely this will be the subject of a federal section 2 claim. Based on Congress' 1982 amendment of the VRA one can't have neutral districting completely trump the VRA. Plaintiffs will show that the state's plan has a measurable effect on minority participation. Plaintiffs will show a map with modest accommodations from the strict rules to show that a plan can both substantially comply with neutral districting and provide fair opportunity for minorities. That's basically how I approached my version of south Cook above reposted below. Other than possibly my SD 6 is there any of the black districts (1-8) that are particularly bad from a neutral mapping perspective? We can debate if plaintiffs would prevail, but isn't that the sort of balancing test you would advocate?

[muon2]

Well of course it's all about what is reasonable.

SD 1 (51.9% BVAP) and 2 (55.0%) follows township and muni lines
SD 3 (65.7%) stays within Chicago and goes north to include all of Hyde Park.
SD 4 (66.5%) and 5 (75.9%) were pulled west following political boundaries, main roads and railroads and maintain compact shapes.
SD 6 (59.9%) is squeezed by the lake and the need to create districts with HVAP sufficient to elect candidates of choice. It also keeps Chinatown intact in SD 9. The court has said in the case of IL CD-4 that unusual shapes can be justified to comply with the needs of two different minority groups.
SD 7 (53.2%) follows rail lines, highways and the L, and stays within Chicago. The chop to the north encompasses the whole neighborhoods of River West and Goose Island. The chop into the Austin neighborhood to the west is required to equalize population with SD 8.
SD 8 (57.9%) follows muni lines to the west of Chicago and is constrained on the north by Latino SDs.

Is that a modest accommodation from neutral principles and more reasonable than providing only 6 black majority SDs, which is demonstrably less than rough proportionality?

This is more than a hypothetical exercise. There is a petition drive under way in IL to go to a redistricting commission and respecting political boundaries is a component. In response there is an active campaign to discredit the petition drive primarily by suggesting that minorities will be disenfranchised though these neutral principles.