Anaheim’s 25.25 Maps, Part 3: Varying Approaches to Creating Latino Plurality Districts

.

.

.

The final stretch of the judges’ Advisory Committee on Electoral Districts portion of the Anaheim’s Districting process kicks off tonight at 6:30 in City Hall.  The public is invited to attend.  My grand plan to complete all of the analysis of the 25¼ maps under consideration (in whole or part) by tonight did not succeed due to various factors (including preparation for CATER’s big fundraiser tomorrow night at Servite H.S.), but we can still finish enough of it to give us plenty to discuss.

To allow latecomers to catch up: Part 1 of this series simply set forth all of the maps under consideration as well as their attending statistics. (It has been getting a decent amount of hits despite a lack of comments, so OJB is happy to have provided a public service).  Part 2 of this series introduced the spreadsheet that can be used to compare those maps on various grounds, noting the number of Latino majority and plurality districts (using various criteria for “plurality”) that each creates.

As an aside: this is probably a good place to note yet again that while this process was set into motion by the desire to increase Latino representation, the system can be beaten by those who want to prevent it — or at least impede the election the “wrong kind of Latino” who is not beholden to the wealthier interests in the city.  Under the at-large electoral system, this has been going on for years: unknown candidates (sometimes with well-known names) like Jennifer Rivera were recruited to split the votes that would otherwise be likely to go to other Latinos.  The most audacious of these may be what happened in the latest election, when perennial candidate Joe Moreno redesignated himself as “Jose ‘Joe’ Moreno” to run against School Board Trustee Dr. Jose F. Moreno for school board — and then switched races and names, running for City Council as “Jose Moreno,” when Dr. Moreno ran for that office instead.  We’ll see more of that in this new system, although because it will be more prevalent it will probably also be more obvious and irritating.

Because of the prospect of intentionally “splitting the Latino vote,” the basic tool that is used to create Latino-oriented districts is readily blunted.  But still — that is the only tool that one has, and to some extent we are legally required to use it, such as to create at least one Latino-majority district.  This piece therefore goes back to the spreadsheet to explore how the different maps try to create Latino majority districts — and in the already trite expression of the 2010s Internet, the results may surprise you.

Here’s the spreadsheet we’ll be working from, now in living color. (Note that corrections to the categorization of the Gallegos and Kim maps are corrected in the graphic displayed later on below; just ignore those for now):

The maps represent lots of different approaches to maximizing Latino representation — or in some cases, to avoid doing so.

What you see to the left is the same representation of 25 maps — I’ve left out the Ponderosa single-district map (see the previous posts if you don’t know what that means) — with some color added.  In each unshaded column, you’ll see the proportion of voters from a given district that are Latino “Citizens of Voting Age Population” — known as “CVAP” — which is essentially the pool of eligible voters.  In the shaded column that each has to its right, you’ll see the margin by which the proportion of Latino voters either exceeds or trails the number of non-Latino White voters.  The district that has the CVAP most favorable to Latino voters has its numbers in hot pink.  The second most favorable district’s numbers are in bright red.  And the third most favorable of the six districts is in orange.

To the right, you’ll see that I have reordered those numbers to the left to form three sets of columns: each map’s districts with the 1st, 2nd, and 3rd highest Latino CVAP.  From this, we can see the strategy that each map uses — intentionally or not — to determine the number of districts that will have Latino CVAP pluralities or majorities.  I’ve color-coded these strategies in the columns in between the parts of the map with uncolored backgrounds.  This turns out to be a pretty good way to identify an important feature of each map that would generally not be perceived at first glance.

Here’s some background: Anaheim still has a higher White than Latino CVAP overall , despite its raw population skewing substantially Latino, for two reasons: (1) many of its residents are not citizens (which, please, generally means that they reside here legally, but as permanent residents or visa holders) and (2) many of its resident citizens are not yet of voting age.  (That’s why “CVAP” is narrowed down by including both a “C” and a “VA.”)

However, 1/6 of Anaheim’s population is in Anaheim Hills, east of the bottleneck around Tustin Avenue, and with six districts it is very difficult to make it anything other than its own district.  That easternmost district will have a very strong white majority — 63% to 16% is the most likely, with the balance being Asian, Black, or Other.  Anaheim’s five other districts will average a Latino over White plurality of 41% to 36%.  But again, at least one of those must be at least 50% Latino CVAP — this is known as a “minority majority” district.  So the question is: of the remaining four districts, how many will be plurality or majority Latino, and how many plurality white?

Note: I tried hard to draw a plurality Asian district.  I came closest in West Anaheim, but I concluded that it can’t (yet) be done.  My sense from doing a lot of maps is that it would also not be possible to draw a second majority White district outside of Anaheim Hills, although I didn’t really try.

Approaches to fostering Latino representation in Anaheim essentially fall into two broad categories.  (These are based on my judgment that a district with a margin of more than 12 is a “strong” or “high” margin for Latinos and one with a margin of under 6 is “weak” or “low.”  Other definitions of these terms would yield different results.)

• Some will try to create TWO districts with strong Latino pluralities or majorities, at the cost of using up the Latino votes and virtually guaranteeing that only two of Anaheim’s five flatland districts — or two of five overall — will be Latino majority.  Despite the “vote-splitting” caveat included above, this is probably the best way to promote the possibility of Latino representation in districts — but unless two of the other five seats (including the Mayor’s) are also Latino-friendly, it puts the Latino representatives on City Council into a minority.  (And if none are Latino-friendly, it puts the non-Latinos into an over 2/3 supermajority.).

• Others will try to create ONE district with a Latino majority and TWO districts with moderate or weaker Latino pluralities.  This tends to be a more forward-looking strategy, given the expectation that more Latino citizens will age into becoming voters, so that a five-point plurality now will probably become a larger one by 2020.  On the other hand, having a smaller Latino advantage means that each district is less likely to tilt Latino — especially given the vote-splitting tricks described above — meaning that even after this reform Latinos who want to vote on ethnic lines could easily be in a 6 to 1 minority on the Council.

In that color-coded table, the “two strong Latino districts” maps use “warm” colors — pink, red, orange, and yellow — each ranked from pink (high) to yellow (low) in order of how strong the third-best Latino district is.  The “one strong Latino-majority district” maps use “cool” colors — green, blue, and purple — ranked in order of how strong the CVAP margin is in the two non-majority districts.

Here’s a version of only the right half of the above table, which will be easier to read:

Some specific examples from the table will make this analysis more clear:

The LULAC 2 map (pink background) actually manages to create 3 Latino majority districts — but as we’ll see, it comes at a cost.  (More than one, in fact.)  The Mills 2 map creates one Latino majority district and two strong plurality districts: while this too comes at a cost, if one’s sole interest is in maximizing Latino representation, the Mills 2 map is probably the best of the lot.  The orange-background maps — including two I largely like, the Brown and Warner maps — create two reasonably strong Latino districts and one plurality districts where Latinos are probably not yet competitive, but may be by 2018.  The yellow background maps create two strong majority districts and no other Latino plurality districts at all.  Districts created by these maps would probably not be competitive even by 2020.

Note: I apologize to Consultant Justin Levitt, who has been very helpful to me and others during this process, if it seems like I’m sandbagging him here with what comes next.  That is not my intention.  I had not noticed this next result before doing this calculation while preparing this post — so it really is “breaking news” to me.  I couldn’t raise it before now.

It comes through very clearly in this table that three of the four maps by the Consultant fall into this category of creating two, and probably only two even by 2020, Latino plurality districts.  Only three of the other 16 other map-submitting entities do so.  (Kim and Gagne, who are politically quite different, both have two strong Latino districts.  A third, from Dresser, manages to be even worse from the perspective of Latino representation.)

This disturbs me, as it suggests that the Consultant has been committed from the outset two a “two and only two Latino districts” model that seems discrepant from that preferred by the vast majority of people submitting maps.  If the Consultant’s maps seem diverse, we can see that at this very basic level he seems to have been working with a goal in mind that is at best questionable.  I don’t know if that reflects his own judgment as to what’s best for Anaheimers or for Latinos — or if it reflects someone’s instructions.  Either way, it suggests that the judges may want to submit at least one alternative map.

Of the maps that create only one district with a strong Latino majority, about half create two moderate-strength Latino Districts (in green) and half create one moderate plurality and one weak one (in cyan blue.)  The Dresser and Bengochea maps do worse.

It is not clear whether the “warm” approach (two strong majority or plurality districts) or the “cool” approach (one strong and at least one moderate plurality districts, plus a third plurality) better serves to facilitate the greater Latino representation that spurred the lawsuit demanding redistricting.  I could put on my “political scientist” had and analyze it for weeks and would still not likely come to a definitive conclusion.  But in looking at this table, I think that I can come to another useful conclusion.  Within each approach, there is better and worse.

Within the “warm” approach, I think that Anaheim should reject the maps that do not even generate three Latino plurality districts out of six.  I’m speaking less as a former practicing political scientist (addressing voting behavior) here than as an activist that cares about the Latino community: creating four plurality White districts out of six is simply an insult that we don’t need.  The City’s CVAP by now is probably very close to even between Latinos and Non-Latino Whites; in the flatlands, the Latino advantage is certainly even higher than the 2010-2013 figures we have here.  Drawing four White-majority districts out of six just seems wrong.  I don’t mean to impugn anyone’s motives here; I was paying attention, and it obviously took me a lot of analysis before getting a sense of what the lay of the land was.

I therefore can’t support the “yellow” maps.  (I reject the LULAC 2 map for reasons that will become clear later, partly because it creates three districts where Latinos are so deeply in the hole that they wouldn’t grown out of it by 2022 and maybe beyond, as those majority-minority borders may get locked in.)  That leaves six “warm” maps worth considering:  Consultant 2, Brown, Gallegos, LULAC 1, Mills 2, and Warner.  (The Brown, Gallegos, and Warner maps, in which the third districts have a Latino plurality of 1, are on the bubble.) Of the “cool” maps, I’d consider only the green and the cyan maps, of which there are 11.  Adelekan, Chuchua 4, XTRA 1 — the latter two of which are mine! — and probably the Duron 2 and Henninger maps, where the third-best district has a margin of 2, are on the bubble as well.)

Now that doesn’t mean that the other maps (except Bengochea, Dresser, and LULAC 2) should be out of the overall discussion.  This is just one of several criteria by which to evaluate them — and this doesn’t yet even involve looking at them.  But achieving better Latino representation is an important criterion, and we should at least avoid maps that convey to the public a desire to discourage it — even if it’s just based on the symbolic difference between the Latino margin in the third-best district being at “1” versus “-1.”

In Part 4, we get to talk about the “equal population” criterion.  After spending that quality time with the spreadsheet, we’ll go back to looking at the maps themselves!