On the 2018 Districting Process

It seems that my opponent and I remember the 2018 Districting process a little differently. Actually, “remember” is a bit of a misnomer, because my “memory” involves quite a bit of going back to my notes and the presentations given during those meetings. I will link files as I refer to them, because I do save every one.

Contrary to my opponent’s statements, one of the concerns the city demographer had was indeed keeping then-current city council members in separate districts. This was a common topic of discussion from city council when coming up with district criteria to give the court, and it was a common point brought up by the demographer herself when discussing the maps that were submitted. You can see her references to this in slide 7 of her presentation to the districting committee on 3 July 2018 and again when considering the final maps on 5 July 2018 (see slide 2). The separation of existing council members into their own districts is a reason map option 3 was accepted. For my opponent to say that this was not a consideration is not only false, but would have been against the council’s — and the court’s — criterion.

I went to every districting committee meeting, as well as city council meetings relating to districting and Measure A. I was deeply involved, and participated in every public event and exercise that was made available. I spoke out on many occasions and presented to both committees and council. I gave presentations on districts and provided an analysis of the map options at the Santa Clara Citizens Advisory Committee. I attended every court meeting during the CVRA lawsuit in San Jose from before the verdict through the end of the remediation phase (in fact, I sat next to the city demographer for every meeting she attended).

There were many issues with the districting process itself, from the makeup of the committee to the process created by the demographer the city hired, Jeanne Gobalet. We verified problems with the committee members when at least one was recycled to be on a subsequent Charter Review committee, again addressing the issue of districts, but seemingly without any of the knowledge gained from previous experience. The problems with some of the process I detailed in my submissions to the city, some of which you can see here: 20180705-DistrictingSubmission-KevinPark. I will sum up my points below.

The “tools” Gobalet provided for the public were rudimentary at best: a spreadsheet that few people could understand, and one that contained a number of errors when initially presented. In fact, there were only three residents who submitted maps — note that all residents were from technical fields — which was a disappointment because there were a few other residents who would have liked to have submitted maps if they could have understood the process better. This was a distinct departure from other cities, like Menlo Park (in English and Spanish! [local copy]), which had easier-to-comprehend online tools provided to residents. You can get a copy of Gobalet’s original 6-district spreadsheet, or my modified spreadsheet for 7 districts.

It was obvious that the pieces, and the tool, were designed for the 2-district maps used for Measure A earlier in 2018 and modified for 6 districts. The fact that minimal changes were made to both pieces and tools contributed to the initial errors in the tool and the inadequacy of the pieces when going to 6 districts. From my submission:

There seems to be some work done to create additional pieces since the last districting committee meetings earlier this year (there are an additional five pieces created by splitting some of the original 31 pieces, for a total of 36 pieces in the most recent Pieces map). It would have been nice to have also spent time dividing the most populous areas into smaller pieces as well, especially since the definition of more districts essentially requires an increase in the “resolution” of each area with respect to population. Perhaps a guideline stating that each piece should contain no more than some number (on the order of (total_population / number_of_districts) / pieces_desired_for_each_district, where pieces_desired_for_each_district is at least 3​) people/CVAP.

This may be largely due to the time frames involved (short, as there was an election in just a few months time and a nomination period that was already in progress), but the process, and push, certainly did not give the residents of this city their due. Why was improper preparation of pieces such a big issue? From my submission:

While the “pieces” data may have seemed sufficient to split the city into two parts, we can see with just a little bit of work that they are not sufficient when considering six and especially seven districts. If we try to make seven districts with the pieces Dr. Gobalet defined, Piece #17 has to stand on its own and adjacent pieces are often too populated to be combined. This is similarly true with six districts. I tried to work off of the block data, but it is unlikely that I will finish or that Dr. Gobalet can validate such a submission within the timeframe.

Even with six districts, the size of the pieces obviates several groupings once a single district is created. When creating an initial group based on the districting criteria (namely: Topography; Geography; Cohesiveness; contiguity, integrity, & compactness of territory; and “Communities of interest”), the other districts tend to define themselves once you try to meet the population deviation guidelines. This is why so many attempts look almost exactly, if not exactly, like Dr. Gobalet’s original drafts.

Consider Piece #17 alone and you will see that there are only a few options with six districts — and, as mentioned above, only one possible option with seven districts.​ This is largely true of Piece #28 as well.​ In fact, Piece #17’s population alone deviates more than 2100 between the 2010 census and 2017 estimates.

While there seems to be some freedom with the large pieces that contain fewer people (see District 2 in Draft Plan1 and District 1 in Draft Plan 2), that is a false hope that does not help balance the numbers or improve representation.

Gobalet’s admission that she, as a professional demographer, could not think of how to break up Piece 17 in any way other than along main thoroughfares — and did not see this as an issue when coming up with the 6-district tool — was disappointing to say the least, and doubly so when we were able to meet additional demographers and experts in court and take a look at their maps. Their maps more-closely resembled the heat maps that defined minority areas, and it was clear how their districts tried to keep communities intact. Gobalet’s maps looked like they could be 6-piece toddler puzzles. Her biggest problem with the plaintiff maps? The plaintiff’s maps looked “gerrymandered”. When asked to explain the gerrymandering she saw:

“I don’t know what gerrymandering is, but I know it when I see it.”

Jeanne Gobalet, court testimony during the 2018 CVRA lawsuit

It is unfortunate that we were not allowed to record the court proceedings, as there was quite a bit of eye-opening discussion. But we can see and hear what the city demographer said in council and committee meetings. In addition to not being able to create a good basis for the maps the public was supposed to create, she was not entirely truthful as to the splitting of precincts in court, which her pieces also did to her own admission in a council meeting: https://www.youtube.com/watch?v=FHrE5v1bAlQ&t=7450

We deserve better districts for better representation in Santa Clara. We deserve council members who will do more than give lip service, but actively take notes on major issues before them and make changes that are consistent with what they say. That is why we need change.