Stephen Wolf of Daily Kos Elections takes a crack at drawing a remedial Congressional map for Texas.
Just how effective is GOP gerrymandering in Texas, and what might a redrawn map look like in 2018 as a consequence of a favorable court ruling? To answer these questions, we’ll analyze a hypothetical fully nonpartisan congressional map below as part of our ongoing series on how Republican congressional gerrymandering affected the 2016 elections. We drew this map by balancing traditional nonpartisan redistricting criteria such as preserving communities of interest, minimizing city and county divisions, respect for the Voting Rights Act, and geographic compactness, while ignoring factors like where incumbents live.
To ensure that our hypothetical nonpartisan congressional map complies with the Voting Rights Act and past Supreme Court precedents, we have estimated the Citizen Voting Age Population (CVAP) according to the 2008-2012 American Community Survey in addition to the official 2010 census population figures. Since Texas has a large and disproportionately Latino non-citizen population, all demographic figures given below refer to CVAP unless noted. We have additionally calculated results by district for every statewide partisan race from 2016 back to 1996 using the Texas Legislative Council’s redistricting data sets, and you can find all of those demographic and election statistics here.
Before we delve into the map, we’ll start with a quick note about what the Voting Rights Act requires. The VRA protects racial or ethnic minority groups in certain districts where there is 1) racially polarized voting, 2) a compact minority population, and 3) a majority population that would otherwise vote as a bloc to defeat candidates chosen by minorities. The VRA does not require that these districts elect a representative who belongs to the protected racial or ethnic group, just that the group can elect its chosen candidates, who may happen to be white.
As the Supreme Court has emphasized in recent racial gerrymandering rulings, a single racial minority group does not actually need to comprise an absolute majority of a protected district’s population so long as the group can reliably elect its candidate choice in that district. Consequently, black VRA districts often do not need to be majority black, while Latino VRA seats sometimes need to be considerably more than 50 percent Latino due to low turnout rates.
With those VRA requirements in mind, here is our proposed nonpartisan Texas congressional map.
[…]
As shown below, our fully nonpartisan congressional map likely would have given Texas Democrats four or five extra House seats in 2016. Those districts include the 2nd in west Houston, the 6th in Ft. Worth, the 10th in central Austin, and the 23rd in San Antonio and El Paso, while the 25th in suburban Austin could’ve gone either way. Additionally, the GOP-held 32nd District in northern Dallas becomes slightly bluer, meaning this map’s impact could grow in future elections.
As we explained above, even if the court strikes down the GOP’s gerrymander and orders the state to draw new districts, it’s likely that Republicans will be able to draw a new gerrymander under additional constraints. Such a scenario would likely see Democrats and Latinos gain at least two seats between South Texas and Austin.
However, it’s an open question whether the court would require a new seat in Dallas-Ft. Worth that would likely elect a third extra Latino Democrat at the expense of a white Republican. The GOP would likely still get to gerrymander in Austin, Houston, and northern Dallas, but two-to-three extra safe seats would be a big deal for Democratic hopes of a House majority in 2018.
Conversely, if Texas Republicans for some reason do not get the opportunity to draw a new map and the court does it for them, the GOP really could be facing the “Armageddon” scenario that it fears. Regardless, we have demonstrated how Republican gerrymandering produces a monumental difference in the Lone Star State’s congressional delegation, and it likely cost Democrats more seats in 2016 than in any other state.
Go read the full writeup, which is very detailed. A 21R/15D split, which this map would produce if the swing CD25 stayed Republican, would be pretty representative of statewide voting patterns, basically giving Republicans 58.3% of the Congressional seats. That’s in line with my own calculations, though of course that will be a moving target over time and across Presidential/non-Presidential years. One local effect of this map would be that the gaggle of contenders in CD07 would need to refile in CD02, if they wanted a winnable race. If nothing else, this particular map is a model of compactness – there are no districts that look like they fell out of a Salvador Dali painting. The trial is now over, so this is more of an academic exercise than anything else; I don’t know if it would have been possible to file something like this as an amicus brief for the trial, but it might have been interesting to have done so. Anyway, take a look and see what you think.