Hispanic Population Growth in Texas over the past 20 years

In 1990, according to the U.S. Census, there were 4,339,905 Hispanics in Texas. At this time, Hispanics were 25.5% of Texas’ total 16,986,510 population. 7 of Texas’ 30 Congressional Districts (23.3%) were Hispanic Voting Rights districts.

By 2000, Texas’ Hispanic population had increased by 2,329,761 people (enough people to fill about three and a half congressional districts) to 6,669,666, growing at an incredible rate of 53.7%. Hispanics were responsible for 60.3% of Texas’ total population growth in the 1990s. Hispanics were now 32.0% of Texas’ total population. This population growth earned Texas two new Congressional Districts following the 2000 Census. However, neither of the two new districts were drawn as a Hispanic Voting Rights district. Ironically, despite the fact that Hispanic population growth had driven Texas’ population gains, the percentage of Texas Congressional Districts which were dominated by Hispanics actually decreased slightly, to only 21.9% of districts (7 of 32).

By 2010, Texas’ population has ballooned to 25,145,561, according to Census estimates. Hispanic population growth continued to drive Texas’ overall population growth, increasing by approximately another 2,791,255 million people (enough people to fill another four congressional districts). According to the 2010 census, 37.6% of Texas’ 25,145,581 people are Hispanic.

Over the 20 year period from 1990 to 2010, Texas’ Hispanic population more than doubled from 4,339,905 to, roughly, 9,460,921 million people. Texas’ Hispanic population has added 5,121,016 people to Texas’ total population. This is enough people to fill 7.3 entire Congressional Districts. Alternatively stated, the increase in Texas’ Hispanic population from 1990 to 2010 is larger than the entire population of the state of Colorado, and almost as large as the entire population of the state of Minnesota. During this 20 year period, Hispanics were responsible for about 62.8% of Texas’ total population increase. Over the whole 20 year period, Texas’ Hispanic population increased by about 118%. By contrast, Texas’ non-Hispanic population increased by only 24%.

2011 Congressional Redistricting

But statistics about anticipated Hispanic political influence in Texas’ are much less striking than statistics about Texas’ recent Hispanic population growth. Texas will gain 4 new Congressional Districts following redistricting this year, in addition to the 2 new districts gained 10 years ago. There seems to be near universal agreement that one of the new seats will be a Hispanic dominated district in Dallas-Fort Worth, drawn to rescue overextended Republican incumbents Kenny Marchent (TX-24) and Pete Sessions (TX-32).

But beyond that one seat, it is not at all certain that the legislature will draw any other additional districts in which Hispanics can expect to have any real political influence. If only one additional Hispanic majority district is drawn in Dallas-Fort Worth, only 7 of 36 Texas Congressional Districts will be Hispanic Voting Rights districts. As a percentage of the total, this is 19.4% of Texas’ U.S. House seats. This is, of course, a lower percentage than in 1990, when 20% of Texas Congressional Districts (6 of 30) were Hispanic Voting Rights districts – despite the tremendous growth of Texas’ Hispanic population within the past 20 years.

There’s some talk of creating an additional Hispanic dominated district in South Texas as well, perhaps based around the very strong population growth in Hidalgo County. But this is not certain, and may well be accompanied by creative shenanigans designed to limit Hispanic influence in TX-23 and TX-27 as much as is humanly possible, in order to protect Republican incumbents Quico Canseco and Blake Farenthold.

Beyond that, a small number of redistricting prognosticators have suggested that a third additional Hispanic dominated (or at least Hispanic influence) district might be created. Maybe it would be somewhere in Houston. Maybe a district would be drawn, once again, in an attempt to replace Lloyd Doggett with a Hispanic Representative. But the details are not really very clear. And there has been some doubt as to whether it’s even really possible to draw three (much less 4) new genuine Hispanic Voting Rights districts, given the fact that Hispanic population growth has been spread over large parts of the state.

Gingles’ 3 Prongs and Section 2 of the Voting Rights Act

This doubt is misplaced. It is very much possible not only to create 3 new Hispanic Voting Rights districts, but to create 4 new Hispanic Voting Rights districts. What’s more, these are not contorted racial gerrymanders, but rather reasonably compact districts. Nor are they districts with nominal 50% Hispanic majorities, where Hispanics are not a majority of the Voting Age population. With the possible exception of the second Dallas-Fort Worth Hispanic Voting Rights Act district, we’re talking about compact districts with Hispanic population shares well into the 60s, 70s, and even 80s (in the case of a few South Texas districts).

* More or less everyone seems to concede that the time has come for 1 Hispanic Voting Rights district in the Dallas-Fort Worth Metroplex. Well, it turns out that the Dallas-Fort Worth area is actually on the verge of having not just 1 Hispanic Voting Rights district, but 2 Hispanic Voting Rights District.

* In addition, an additional compact Hispanic Voting Rights Act district (about 2/3 Hispanic) can be created in Houston, while preserving the existing TX-29.

* Finally, Hispanic population growth has been strong enough in South Texas to draw an additional Hispanic Voting Rights district, even while keeping all South Texas districts with overwhelming Hispanic majorities (more than 70% or even 80% Hispanic). Frankly, the interesting question along the Rio Grande is not whether one more Hispanic Voting Rights district can be added, but rather whether it might be possible to add another 2 Hispanic Voting Rights districts.

And if it is possible to draw additional compact Hispanic Voting Rights districts, there is a strong case that Texas is legally obligated to do so by Section 2 of the Voting Rights Act. In the case of Thornberg v. Gingles, the Supreme Court layed out a three basic threshold requirements for finding a minority vote dilution violation of Section 2 of the Voting Rights Act in redistricting cases. They are:

1) The minority group “is sufficiently large and geographically compact to constitute a majority in a single-member district.”

2) The minority group is “politically cohesive.”

3) “That the white majority votes sufficiently as a bloc to enable it … usually to defeat the minority’s preferred candidate.”

All three of these requirements apply in the case of Texas:

1) As I will show below, Hispanics in Texas do form a sufficiently large and compact population in Texas to constitute a majority in a single member district. Indeed, they form a sufficiently large and compact population to form not just bare majorities, but very large majorities in multiple new districts.

2) Texas Hispanics are politically cohesive. They tend to vote as a bloc for candidates (usually Democrats), and tend to vote especially strongly for Hispanic candidates. This is abundantly clear from sources such as exit polls and past election results.

3) In Texas, white voters vote overwhelmingly Republican as a bloc, in opposition to Hispanics. Voting in Texas occurs to large degrees along racial lines. For the most part, the only place where this may not be the case quite as much is Austin.

To be sure, redistricting and the Voting Rights Act are unpredictable areas of Supreme Court jurisprudence, and almost anything could happen in the coming flurry of lawsuits that will be filed across the nation as redistricting begins. But if the Texas legislature, as anticipated, limits the number of new Hispanic Voting Rights districts drawn, and if Hispanics in Texas file Section 2 vote dilution lawsuits, they will have strong claims and a good chance at success.

Now, to the districts!

Overall Map

First of all, this map is drawn with 2010 census data (data available for download here) using 2010 precincts (shapefiles available for download here). Because this map is drawn on the precinct level, it misses the level of detail you could get by drawing on the block group or block level. Also, I drew it quickly and without dedicated redistricting software (just plain old GIS software), and that means that I probably did a pretty sloppy job. For those reasons, it is almost certainly possible to improve on this map, making the districts more compact and/or increasing the Hispanic percentages in the new Hispanic Voting Rights districts. My purpose was not to draw the best possible map adding 4 new Hispanic Voting Rights districts, but rather to simply show that it is very much possible to draw 4 new Hispanic Voting Rights districts.

I also did not bother to draw the non-Voting Rights Act districts. Those are not the interesting part of Texas redistricting; it goes without saying that the State Legislature will draw all of them as safe Republican districts to protect their incumbents. So I ignored those districts.

Here’s the overall statewide map, with county/precinct lines taken out so you can see the shape of the districts:

And here’s the same map, with county/precinct lines visible:

And here’s a table showing the demographic makeup of all the districts, including total population and VAP data by Race and Hispanic origin. Note that for African Americans and Asian Americans, I used the census data for Non-Hispanic Single Race African Americans and Asian Americans. So if you were to include multiple race African Americans and Asian Americans, the actual numbers would be a bit higher for those two groups.

South Texas

There are four districts based in South Texas. In addition, I am lumping in TX-16, which is in El Paso, as part of “South Texas.”

TX-27: 21.3% Anglo, 74.6% Hispanic, 2.1% African American, 1.2% Asian (Hispanic VAP – 70.7%)

TX-15: 20.3% Anglo, 76.9% Hispanic, 1.7% African American, .6% Asian American (Hispanic VAP – 72.4%)

TX-35: 15.0% Anglo, 82.9% Hispanic, .6% African American, 1.0% Asian American (Hispanic VAP – 79.9%)

TX-28: 18.4% Anglo, 79.2% Hispanic, 1.1% African American, .5% Asian American (Hispanic VAP – 75.5%)

TX-16: 14.6% Anglo, 80.2% Hispanic, 2.9% African American, 1.1% Asian American, (Hispanic VAP – 77.7%)

In South Texas, TX-27 is preserved largely as is, but has to drop its part of San Patricio County because of population growth. This is bad news for Blake Farenthold, because the district becomes more heavily Hispanic. TX-15 remains similar to its previous incarnation, but now picks up some additional counties to the north, including Victoria. TX-35 is added as a new seat based in Hidalgo County. Like TX-15, it heads way to the north, almost but not quite to San Antonio. TX-35 ended up as an 82.9% Hispanic district – there were simply more Hispanics in South Texas than I could figure out what to do with, even after adding a new Congressional district. TX-28 shifts to the west. It’s still based in Laredo, but stretches all the way to El Paso, taking a lot of territory that was formerly in TX-23. In addition, San Angelo in Tom Green County is added to the district. It pulls entirely out of Hidalgo County and out of the San Antonio area. And speaking of El Paso, TX-16 remains pretty much exactly the same.

Clearly, there’s no difficulty at all with fitting 5 full districts along the US-Mexico border. There’s no need for any of these districts to enter San Antonio, Austin, or Houston: all 5 districts easily have 70% or more Hispanic population, while staying outside of major non-border metro areas. There are also many other counties with large minority (and even majority) Hispanic populations nearby, leading me to wonder what the map might look like would happen if one tried to add not just one, but two new South Texas Hispanic Voting Rights Districts.

You can make an argument that many of these districts are not really compact. For example, the new 35th district starts in McAllen and, reaches up to the San Antonio’s exurbs nearly 300 miles away. And in truth, you could draw more compact districts by simply drawing a district entirely within Hidalgo County. But currently and in the past, districts in South Texas have been drawn with appendages reaching to the north, into counties with significant but not monolithic Hispanic populations. South Texas districts are drawn this way for a reason. All the counties directly along the Rio Grande are overwhelmingly Hispanic. Hidalgo County, for example, is now 90.6% Hispanic, and has enough population for 1.1 Congressional Districts. If districts were drawn solely within counties along the border, those districts would be almost entirely made up of Hispanics, thus packing Hispanics into a small number of overwhelmingly Hispanic districts. At the same time, counties a bit further north such as Victoria County, which are a bit less than outright Hispanic majority, would end up in districts which might have significant Hispanic populations, but not high enough Hispanic populations to form genuine Hispanic Voting Rights districts. This issue was considered in redistricting in the 1980s, and it was determined that packing Hispanics into one or two 90% Hispanic districts along the border would be a violation of Hispanic voting rights, and would violate the Voting Rights Act. Neither the facts nor the conclusion has changed since then.

San Antonio (and Austin)

In the South Texas section above, I distinguished San Antonio (Bexar County) from South Texas. I did this for a reason: it is now possible to draw two compact 2/3 Hispanic Congressional Districts entirely within Bexar County. Hispanic population growth in San Antonio has been strong, and TX-23 can contract entirely within the county.

In addition, it’s now possible to draw a majority Hispanic district in the Austin area (without reaching into South Texas, San Antonio).

TX-20: 22.9% Anglo, 65.6% Hispanic, 8.3% African American, 1.6% Asian American (Hispanic VAP – 62.1%)

TX-23: 24.3% Anglo, 65.6% Hispanic, 5.9% African American, 2.4% Asian American (Hispanic VAP – 61.6%)

TX-25: 31.9% Anglo, 53.4% Hispanic, 10.6% African American, 2.3% Asian American (Hispanic VAP – 47.9%)

I simply divided TX-20 and TX-23 so that both districts would have the same Hispanic population %. Of course, the State Legislature would probably try to draw the lines between the two districts differently, to make TX-23 as Republican as possible and to give Quico Canseco the best chance of winning.

TX-25 is a purely Central Texas Hispanic Majority District. But while TX-25 has an overall 53.4% Hispanic majority, Hispanics are only 47.9% of the VAP in this district, so I am not counting it as one of the 4 new Hispanic Voting Rights Districts. It might be possible to exceed 50% of the VAP by drawing on the block level, but it looks like the 50% hurdle may be just slightly out of reach. The legislature is probably unlikely to draw a district like this – either they will attempt to outright eliminate Lloyd Doggett (perhaps at their own peril) or they will simply give him the most Democratic district possible in the Austin Area. And this is not the most Democratic district possible in the Austin area, but rather (roughly) the most heavily Hispanic district possible in the Austin area.

Houston

It VERY easy to draw a second compact Hispanic Voting Rights district in Houston – it seems like the only way Houston will not get a second Hispanic Congressional District is if the Supreme Court overturns Gingles. Hispanic population growth in Southwest Harris County has been strong enough that it is no longer necessary for TX-29 to extend an arm into the north side of Houston. This means that the north side can be combined with increasingly heavily Hispanic portions of western Houston to form a new Hispanic Voting Rights district. In addition to drawing two Hispanic districts, I drew TX-9 and TX-18, which are currently dominated by African American voters, in order to demonstrate that African American voting rights are not harmed by the creation of a new Hispanic district.

TX-9: 12.2% Anglo, 38.2% Hispanic, 36.3% African American, 12.9% Asian American (African American VAP – 35.9%; Hispanic VAP – 35.2%)

TX-18: 28.7% Anglo, 26.0% Hispanic, 38.8% African American, 4.7% Asian American

TX-29: 18.4% Anglo, 67.7% Hispanic, 10.3% African American, 2.6% Asian American (Hispanic VAP – 63.3%)

TX-36: 15.6% Anglo, 64.2% Hispanic, 14.7% African American, 4.4% Asian American (Hispanic VAP – 60.0%)

Admittedly, the western part of the new TX-36 has somewhat non-compact lines. However, this is because the precincts in this area are very large and there has been a lot of population growth in this area. There was an excess of 35-45% Hispanic precincts that could have been added to this district – my problem was deciding which ones to include. Ultimately I picked 45% Hispanic precincts over 40% Hispanic precincts, which made the western part of the district choppy looking. This is one area where using block group and block level data could make a big difference in drawing a more compact looking district with a higher Hispanic population. There is also a delicate balancing act in deciding which precincts should go in TX-18, vs. which should go in TX-29 and TX-36, because a lot of precincts have high Hispanic and high African American populations. I did not do an optimal job of this, and I am very sure that a more careful drawing could make these districts better looking and/or could increase the Hispanic percentages in TX-29 and TX-36. But even with my sloppy drawing, I was easily able to draw a district which is nearly 65% Hispanic. There are also some heavily Hispanic parts of South-west Houston around the intersection of the Southwest Freeway and the Westpark Tollway which I left out of the new TX-36. By including those areas in the district, the Hispanic population could probably be increased while also making the new TX-36 more compact looking. However, it would require cutting through the Galleria area (and possibly separating

TX-29 speaks for itself – impressive Hispanic population growth makes a much more compact South-West Harris County district possible.

TX-18 remains basically similar to its current form. However, to reach the northern part of the district (the area around George Bush Continental Airport), it cuts through the north-east rather than through the north-west, in order to make room for TX-36. I also added some white Democratic areas (Montrose), which makes John Culberson in TX-7 safer. My precinct selection was pretty sloppy though, and I included some white Republican areas as well. In any case, a strong

TX-9 actually ended up with a slight Hispanic plurality (but with an African American VAP plurality). I was not very careful at all about precinct selection here, and I am pretty sure I left some heavily Democratic precincts out, which would go in either TX-7 or TX-22. In drawing this district, I imagine that the legislature will be interested in doing whatever it possibly can do to shore up TX-7 and TX-22.

Dallas Fort-Worth

It’s pretty much a given that Dallas-Fort Worth will get its first Hispanic Voting Rights district. What is not appreciated is that the Metroplex is actually on the verge of being able to have not just one, but two Hispanic Voting Rights districts. These two new Hispanic Voting Rights districts can be drawn without decreasing the African American population in TX-30. Actually, even in my sloppy drawing, the African American population percentage goes up 5% to 44.3%, as a result of removing heavily Hispanic areas which are currently packed into the district.

TX-30: 27.1% Anglo, 22.5% Hispanic, 44.3% African American, 4.1% Asian American

TX-33: 22.4% Anglo, 58.8% Hispanic, 14.6% African American, 2.8% Asian American (Hispanic VAP – 52.8%)

TX-34: 20% Anglo, 60.7% Hispanic, 15.4% African American, 2.5% Asian American (Hispanic VAP – 54.9%)

TX-34 is the most non-compact looking of all the districts I drew. It could benefit greatly from a more careful drawing, and from using block group and block level data rather than precinct level data. I probably did not arrange Metroplex Hispanic TX-33 and TX-34 in the most efficient way between the two districts, and I ended up including a number of North-Arlington precincts with very low Hispanic populations, in order to increase the compactness of the district a little bit. These could be taken out and the Fort Worth area could be cleaned up. This map is not optimally drawn, either in terms of compactness or Hispanic population %.  Ultimately, the question of whether or not Dallas-Fort Worth gets two Hispanic Voting Rights districts may come down to what exactly a court considers to be a “compact” enough district for the purposes of Gingles’ second prong. It’s a pretty good bet that groups like MALDEF and LULAC will try to draw a map in which both TX-33 and TX-34 are as compact as possible and have as high Hispanic populations as possible, and that they will present their maps to a court in a Section 5 lawsuit.

There is less to say about TX-33. It is more compact looking than TX-33 (except maybe for the thin connection through downtown Dallas), but has a slightly lower Hispanic population percentage. There may be some question about whether the Hispanic population is really large enough in this district. There are some heavily Hispanic areas in north Dallas which I left out of the district. It might be possible to increase the Hispanic percentage by linking the eastern and western parts of this district through the LBJ Freeway (I-635) rather than through downtown, but I did not do this because it would have disrupted TX-32 (possibly even throwing Pete Sessions into TX-30!!!). Obviously, that is something that the legislature will avoid when they are drawing the map.

Conclusion

So there we have it, Texas with 4 new Hispanic Voting Rights districts (plus with a quasi-Hispanic Voting Rights district in the Austin area).

These are obviously not districts that the Republican controlled legislature will want to draw, but these are districts that clearly can be drawn, given the magnitude and concentration of Texas’ Hispanic population growth. We can expect a large legal brawl over these four districts.

But at the end of the day, even if all 4 new districts are drawn as new Hispanic Voting Rights districts, only 11 of Texas’ 36 congressional districts (30.6%) will be Hispanic Voting Rights districts. Hispanics will still be under-represented when compared to their share of the Texas population (37.6%) and compared to their share of the Texas Voting Age Population (33.6%).

And again, my map is simply a crude initial drawing, outlining some of the things that are possible in Texas. With time, software, block group/block level data, and care, it is undoubtedly possible to draw more compact looking districts with higher Hispanic population percentages than I achieved. It may even be possible to add more Hispanic Voting Rights districts than 4. It will be interesting to see what people come up with.