Abbott and the Latino vote

The Trib drops a number on us.

I guess I need to find a new Abbott avatar

Along with his 20-point margin of victory, Gov.-elect Greg Abbott accomplished something on Election Day that many naysayers doubted the Republican could: He took 44 percent of the Hispanic vote.

For Texas conservatives, Abbott’s performance indicated that Republicans are making headway among this increasingly crucial voting bloc, which tends to lean Democratic. But upon taking office, Abbott will find himself in turbulent political waters.

[…]

But election results show that despite Republican outreach efforts, Abbott does not have a strong hold on areas of the state where most of the population is Hispanic, particularly the border counties Abbott repeatedly visited during his campaign.

In Cameron County, which Abbott had set out to win, he garnered 42 percent of the vote while Davis took 55 percent. He fared worse in Hidalgo County, with only 35 percent of the vote to Davis’s 63 percent.

The results could prove troublesome for a party looking to hone its outreach efforts as the state’s Hispanic population swells. Although they make up less than a third of eligible voters in the state, Hispanics are expected to make up a plurality of Texas’ population by 2020.

Abbott outpaced his predecessors in winning support among Hispanics, but navigating the crosscurrents of appealing to a far-right base and conservative Hispanics continues to prove difficult for Republicans when it comes to immigration.

The article is about how Abbott is going to try to balance his madrina-friendly image with the ugly xenophobia of his party. I’m not going to prognosticate about that – lots of people have been opining about what the Abbott-Dan Patrick dynamic is going to be like – but I am going to focus on those numbers. I presume that 44% figure comes from the exit polls we were promised. I know they were done and I’m aware of some complaints about their methodology, but I’ve seen basically no reporting or other analysis on them. Be that as it may, I’m going to do three things: Check the actual results to see if they line up with the 44% figure given, compare Abbott to Rick Perry in 2010, and I’ll hold the third one back till I’m ready to show you the numbers.

Comparing Latino voting performances is always a bit dicey, since the best we can do at this level is use county and State Rep district data, which is a reasonable enough rough approximation, but which can be distorted by the presence of non-Latino voters, especially if Latino turnout is lower than expected. But it’s what we’ve got, and we can at least draw some broad conclusions. A full comparison to Rick Perry in 2010 won’t be possible until all the legislative district data is published by the TLC in early 2015, but we’ll use what we do have. Here’s a look at county comparisons:

County      Perry   Abbott   White   Davis
==========================================
Cameron    40.82%   42.01%  57.30%  55.46%
El Paso    36.76%   37.25%  61.29%  60.32%
Hidalgo    31.75%   34.79%  66.82%  62.70%
Maverick   26.83%   26.27%  71.86%  70.27%
Webb       22.92%   28.86%  75.60%  68.03%

So yes, Abbott did improve on Rick Perry, but not by that much. In Cameron County, which as the Trib story notes Abbott was claiming he wanted to win, he beat Perry by a bit more than one point. He did do three points better in Hidalgo and six points better in Webb, but only a half point better in El Paso and a half point worse in Maverick. Again, this is incomplete data – the State Rep district data will tell a better story – but if Rick Perry was scoring in the low thirties in 2010, it’s hard for me to say that Abbott did any better than the mid-to-upper thirties. It’s an improvement, and he gets credit for it, but I don’t see how you get to 44% from there.

I do have State Rep district data for Harris County, so let’s take a look at that:

Dist        Perry   Abbott   White   Davis   Dewhurst    LCT
============================================================
HD140       27.9%    32.2%   70.7%   66.3%      31.6%  65.9%
HD143       29.6%    35.0%   68.9%   63.7%      33.4%  63.9%
HD144       45.2%    51.7%   52.7%   46.3%      50.8%  46.0%
HD145       36.3%    40.8%   62.0%   57.2%      41.6%  54.8%
HD148       36.3%    39.1%   61.6%   58.7%      45.0%  50.8%

The caveat here is that the Hispanic Citizen Voting Age Populations (Hispanic CVAPs) are lower in these districts than in many other Latino districts. HD140 is the most Latino, at 60.6%; by comparison, the lowest CVAP in the six El Paso districts is 59.4%, with the other five all being greater than 70% and three of the six topping 80%. Be that as it may, Abbott clearly beat Perry here, by four to six points. That also comes with an asterisk, however, since as we know Bill White outperformed the rest of the Democratic ticket on his home turf by about six points. I included the David Dewhurst/Linda Chavez-Thompson numbers as well here to serve as a further point of comparison. Add it all up, and Abbott got 39.6% of the vote in Latino State Rep districts in Harris County. That’s impressive and a number Democrats will have to reckon with, but it’s still a pretty good distance from 44%.

I’ll revisit this question later, once the TLC has put out its data. In the meantime, there’s one more dimension to consider: How well Greg Abbott did in 2010 versus how well he did in 2014:

County     Abb 10   Abb 14
==========================
Cameron    48.21%   42.01%
El Paso    42.43%   37.25%
Hidalgo    37.72%   34.79%
Maverick   26.31%   26.27%
Webb       29.12%   28.86%

Dist       Abb 10   Abb 14
==========================
HD140       35.1%    32.2%
HD143       37.2%    35.0%
HD144       54.0%    51.7%
HD145       46.4%    40.8%
HD148       48.6%    39.1%

Now of course this isn’t a real apples-to-apples comparison. Abbott was running for Attorney General in 2010 against a candidate who had no money and a self-described “funny name”. That’s a formula for him to do better. Of course, one could say that voters in these places liked him more when he had a lower profile. The more they heard about him, the less likely they were to vote for him. Make of that what you will.

Related Posts:

• UT/Trib: Cruz 41, O’Rourke 36, part 2
• Precinct analysis: State House district changes by demography
• Precinct analysis: Beto versus the spread