When the Walls Come Tumbling Down: Texas Data Points from the Week in Politics (September 3, 2020)

President Trump took the discussions of mail-in voting through the looking glass when he urged supporters to vote by mail and to vote in person, too. While Attorney General Bill Barr testily and ineffectually tried to clean up Trump’s nihilistic weirdness (by suggesting that what he really meant was that Republicans’ voting by mail “have to go and check their vote by going to the poll and voting that way, because if it tabulates then they won't be able to do that”), Texas Republicans from both the executive and judicial branches were doing their best to stifle the attempted expansion of voting by mail in Harris County. As Trump’s latest election play unfolds in the choppy wake of the COVID-19 pandemic, Governor Greg Abbott floated the idea a relaxation of the statewide containment measures after the Labor Day holiday, seemingly not quite taking into account the lag effects in accounting for the community spread we saw after Memorial Day and the Fourth of July weekends earlier this summer, and even as pandemic data collection in Texas continue to leave a lot to be desired. Speaking of data, Comptroller Glen Hegar provided his offices’s regularly scheduled state revenue update, which enables us to look not only at quality monthly data but also provides a look at FY 2019. Much less useful was a recent release of Texas presidential polling, which got us on our soapbox about poll disclosure (sorry, though not a lot). Finally, a Texas Tribune/ProPublica report on the unsurprising news that a section of the border wall paid for as part of the ALLEGED “We Build the Wall” grift is likely to come tumbling down made us recall results from the UT/Texas Tribune Poll back when the wall was a thing. Find polling and other data on these topics and a couple of other treats below. Have a great Labor Day Weekend – but let’s try not to screw things up.

1. Straight from the irony desk: You really just can’t make this up, vote-by-mail edition. Donald Trump seems to have finally listened to the concerns of his allies over his own efforts to discourage mail in voting, this week encouraging his supporters at a North Carolina campaign stop to send in a mail-in ballot — then suggesting that they also show up to vote, too. Per Maggie Haberman and Stephanie Saul in The New York Times:  “Let them send it in and let them go vote, and if their system’s as good as they say it is, then obviously they won’t be able to vote,” the president said. “If it isn’t tabulated, they’ll be able to vote.” Funny thing: In the October 2016 UT/Texas Tribune Poll, amidst Trump’s dark warnings about the election being rigged against him and non-citizens voting — and infinitely documented Russian efforts to help Trump — we asked Texans how serious a problem people voting multiple times would be in the upcoming election. A sizable majority of Texans who planned to vote for Trump —  64% — said people voting multiple times would be not just a problem, but AN EXTREMELY SERIOUS PROBLEM. Most of the coverage was quick to point out that knowingly voting twice is a felony in North Carolina — and if you’re wondering, it’s also a felony in Texas...and almost everywhere (h/t NCSL). In June 2019 polling, 30% of Texas Republicans said that voters knowingly break election laws “frequently”, with another 39% saying this happens sometimes — now we wonder who they were talking about. Shows you not to make assumptions.

Loading chart...
categoryNot supporting Donald TrumpSupporting Donald Trump
Extremely serious18%64%
Somewhat serious10%25%
Not too serious17%8%
Not serious at all51%3%
Don't know4%1%

Loading chart...
categoryDemocratIndependentRepublican
Never12%6%2%
Rarely46%23%16%
Sometimes17%30%39%
Frequently9%19%30%
Don’t know/No opinion17%22%12%

2. But wait — how will Texas Republicans vote twice? Well, given the partisan age breakdown, maybe this isn’t such an impediment. In the least surprising news of the week, Texas Attorney General Ken Paxton followed through on threats by the Secretary of State (there’s a phrase you don’t see a lot) and sued the Harris County Clerk, per the official petition, “to prevent him from sending over two million applications for mail ballots to every registered voter in Harris County, irrespective of whether any given voter requested an application or even qualifies to vote by mail.” The OAG piled on to a lawsuit originally filed by Steven Hotze and a Republican judicial candidate in Harris County and subequently joined by the Harris County GOP, per Zach Despart’s coverage in the Houston Chronicle. The Texas Supreme Court on Wednesday, in response to the Hotze/GOP suit, issued a temporary stay ordering Hollins to refrain from sending applications for absentee ballots to voters under 65 years of age, pending further action in the state’s suit against Hollins, per Alexa Ura’s very attentive coverage in The Texas Tribune (with support from Daily Texan alum Sami Sparber - hook ‘em). So for the moment, Hotze, the GOP, and the OAG have forestalled the county’s efforts to get applications in the hands of most voters. The partisan line up in the legal battles reflects public attitudes among voters toward expanding voting by mail as a result of the pandemic, as well as how partisans say they would vote were that expansion to take place. As is frequently the case with voting and election fights, partisans seem to be taking the cues from partisan elites, which are plentiful, especially given this issue’s entanglement with the COVID-19 pandemic. A national CNN poll found similar results, per the write-up by their polling director Jennifer Agiesta, to wit: “Biden's backers remain more likely to say they plan to vote before Election Day (49% by mail, 29% early, 21% election day) than are Trump's supporters (10% by mail, 21% early, 68% election day).” (And yes, we get the ritual satisfaction the tradition-inclined get from voting on Election Day, but (1) we doubt that’s the only disposition being expressed in these results, and (2) we wonder how voting twice squares with the tradition.)

Loading chart...
CategoryDemocratIndependentRepublican
Favor88%45%21%
Oppose7%32%72%
Don't know/No opinion5%23%7%

Loading chart...
CategoryDemocratIndependentRepublican
In-person early29%41%54%
In-person on election day14%15%28%
By mail52%34%13%
Don't intend to vote0%2%1%
Don't know/Unsure6%8%4%

3. Like a Groundhog Day remake minus the learning. (And yes, the reference is low-hanging fruit.) This week, Governor Abbott is floating trial balloons on opening Texas up a bit after Labor Day weekend (well, some learning, see: July 4), citing the decline in hospitalization rates, even as Texas continues to see new daily cases comparable to mid-June, before the peak, but still significantly higher than last Spring, per Nicole Cobler's story in the Austin American Statesman, among other coverage of the Governor's Twitter feed.

Source: Texas 2036 Covid Tracker, accessed 9/3/20 

The confidence is admirable as K-12 schools in Texans lurch towards various stages of re-opening, and while the college experiment is still in  its earliest stage. At UT-Austin, where classes began on August 26, this nifty dashboard shows that 72 students have already tested positive since opening day, including 42 on September 2 alone. On the other hand, a model built by UT predicted that between 82 and 185 students would show up to campus already infected — so, glass half full?

Speaking of learning, let’s wait and see if the Governor opens up the bars, or if the recent relaxation of the 51% rule to allow food trucks to help out is about as big a risk as he’s willing to take at this point with bars, and the focus is on restaurants. (The Abbott Tweet that launched this speculation was an RT with comment to someone from the restauraunt business.)

4. 87th Texas Legislature gloom check. The Texas Comptroller of Public Accounts released their “monthly state revenue watch,” which reports revenue collected in August and also gives us a look at collections for the fiscal year. As the legislature and associated trades gird themselves for the legislative session, the toplines from Comptroller Hegar: “General Revenue-related revenue for fiscal 2020 totaled $56.98 billion, down 1.5 percent from fiscal 2019. All Funds tax collections were $57.38 billion, down 3.4 percent from fiscal 2019.” As Congress and the White House continue to make a hash of efforts to get another relief bill passed, the trend in federal income is of continuing interest. The year-to-year increase in federal income has been a major boost to revenue in the last few months as most sources of revenue have fallen since the pandemic’s effects took hold in March. The major injection of federal support from the first CARES Act meant a 260% increase in federal income in the April report, and federal income was still up a little more than 39% in August over the previous year. Per the chart below, for a state that made an art of thumbing its nose at the federal government while worshipping at the altar of self-reliance in our humblest moments, federal income makes up a big share of revenue – the state needs it now more than ever.

Federal income as share of tax collections and net state revenue in Texas, FY2003-FY2019

Loading chart...
categoryFed as share of tax collectionsFed as share of net revenue
200380.28%35.97%
200478.44%34.81%
200576.44%34.66%
200673.71%34.14%
200765.96%31.58%
200863.44%30.16%
200981.59%36.61%
2010104.21%42.19%
201198.90%40.77%
201274.69%34.79%
201368.08%32.85%
201467.20%32.65%
201567.20%31.31%
201681.43%35.47%
201777.28%34.50%
201871.28%32.97%
201970.57%32.75%

Texas tax collections, federal income and total net revenue (in thousands), FY2003-FY2019

Loading chart...
Tax collectionsFederal incomeTotal net revenue
200326126675.0020975687.0058309957.00
200427913002.0021895532.0062893862.00
200529838278.0022809751.0065810167.00
200633544498.0024726454.0072420950.00
200736955630.0024376053.0077189291.00
200841357929.0026238328.0086983025.00
200937822453.0030859931.0084285463.00
201035368901.0036856627.0087357158.00
201138856176.0038430476.0094270890.00
201244079119.0032922040.0094641053.00
201347781046.0032530326.0099036382.00
201450992562.0034266043.00104942290.00
201550992562.0034266043.00109428258.00
201648476226.0039473835.00111280871.00
201749643422.0038365630.00111195221.00
201855584775.0039618568.00120165619.00
201959380722.0041904474.00127941402.00

5. Data, good; data in a vacuum, not so much. We love data, but only if it comes with at least the most minimal of explanation. Morning Consult released the results of surveys this week in 11 states — including Texas — in an attempt to assess whether or not the conventions had any immediate impact on the presidential contest. Most states looked relatively unchanged, or at least, within the margin of error, with Arizona standing as the outlier, where Trump was beating Biden by 2 before the conventions, and now trails by 10, per the data. In Texas, the polling shows Trump ahead by 1-point, the same lead he held before the convention in the last Morning Consult poll of Texas. So what does this tell us? As professionals who conduct and examine surveys regularly: we don’t know. And the reason we don’t know is because Morning Consult provided only minimal methodological information for all but the Arizona result out of the 11 states surveyed. We know the survey was conducted August 28 to the 30th, and that’s about it. We don’t know how many people were surveyed, therefore, we also don’t know the margin of error. We have no idea how Morning Consult defines a “likely voter” (those surveyed, apparently), and we have no idea what other questions were asked on the survey besides the trial ballot item. Ultimately, all of this falls under the rubric of “disclosure.” The basic point here is that you can’t, and in fact, no one can, make a judgement about the quality and accuracy of a survey without some of these basic bits of information about how a survey was conducted (and yes, we reached out to morning consult for clarification and have not received a response). This isn’t to beat on one pollster – everyone should disclose more information about how their poll was conducted. While on the subject: As we get closer to Election Day, we will continue to see journalists reporting on campaign polls that provide even less disclosure than Morning Consult’s, which was presumably unbiased by a consideration of electoral strategy — unlike most every single campaign poll that gets handed to a reporter. (If anyone reading this can find out some of this basic information about the Texas sample, we’d be happy to update or correct our Morning Consult read and add the information to our Texas Presidential Poll Tracker.)

Texas Presidential Election Polling
Poll Field Dates Sample Size Sample Composition MOE Trump Biden Spread
Morning Consult 8/28-8/30 ND Likely Voters ND 48 47 R+1
Public Policy Polling (TDP) 8/21-8/22 764 Adults? +/- 4% 47 48 D+1
THPF/Rice U/ Baker Institutue 8/4-8/13 846 Registered Voters +/- 3.37% 47.5 40.5 R+7
Morning Consult 7/24-8/2 2576 Likely Voters +/- 2.0% 46 47 D+1
Morning Consult 7/17-7/26 2685 Likely Voters +/-1.9% 45 47 D+2
Quinnipiac 7/16-7/20 880 Registered Voters +/-3.3% 44 45 D+1
CBS News/YouGov 7/7-7/10 1212 Registered Voters +/-3.3% 45 45 Tie
Dallas Morning News/UT Tyler 6/29-7/7 1677 Likely Voters +/- 2.5% 43 48 D+5
UT/Texas Politics Project 6/19-6/29 1200 Registered Voters +/-2.8% 48 44 R +4

*Estimated from information in report;
ND means "No Disclosure." We will consider inclusion of polls without reasonable disclosure about sample size and methodlogy if the pollster has a track record of reliable polling.

6. If only he had gotten Mexico to build it… A privately built section of border wall, supported in part by the “We Build the Wall” effort that led to the recent indictment of former Trump advisor Steve Bannon, should be expected to fail, per two engineering reports filed in federal court this week and reported on by Jeremy Schwartz and Perla Tevizo for ProPublica and The Texas Tribune, supporting prior speculation about the instability of the wall built right on the bank of the Rio Grande (We’re not engineers ourselves, but as homeowners, we know: “Water bad”.)  While the wall isn’t the issue it used to be for Trump, given his lack of delivery on this grandiose promise — only about 307 miles of wall have been built or improved on the 2000-mile U.S. Mexico border according to Customs and Border Protection, and it’s been paid for by the taxpayers and otherwise grifted Americans rather than Mexico — and this particular stretch of wall, at 3 miles is barely a speck of sand in that border.  However, it sure does seem to be a likely candidate for a quick zing in a debate or a Lincoln Project ad — since it was both crappy, small, and allegedly part of another former Trump advisor’s misconduct. UT/Texas Tribune polling on the border wall suggest it’s a good reinforcer of Democratic attitudes, and of the partisan divisions in the suburbs that both campaigns are leaning on so hard.

Loading chart...
categoryDemocratIndependentRepublican
Strongly support4%38%76%
Somewhat support6%19%13%
Somewhat oppose10%8%3%
Strongly oppose76%30%5%
Don't know/No opinion4%5%2%

Loading chart...
categoryDemocratIndependentRepublican
Strongly support3%27%66%
Somewhat support4%19%14%
Somewhat oppose3%9%8%
Strongly oppose87%34%7%
Don't know/No opinion3%11%5%

Loading chart...
categoryUrbanSuburbanRural
Strongly support21%37%53%
Somewhat support9%10%12%
Somewhat oppose8%5%4%
Strongly oppose57%42%27%
Don't know/No opinion6%5%5%

Bits and pieces from a colorful week

This was interesting regarding proposed cuts in state funding for women’s health. Feels like the chair of the Senate Finance Committee was subtweeting the Lt. Governor. 

Personally, we’re not into dynasties, and don’t really care what the surname is of a guy who thinks primarying an incumbent Democratic U.S. Senator in Massachusetts is a good idea. But we’re glad Aman still thinks about Texas now that he works at THE New York Times

September 1 marked former Texas Governor Ann Richards’ 87th birthday. One of the Texas Politics Project’s best early days was interviewing the former governor in the Public Strategies offices in Austin. We wish we had outtakes of the off-camera conversation before the interview, when she was scarily skeptical of why we would be putting together an online textbook about Texas politics (“Oh, Texas government...ok,” if Henson’s memory serves) and recently seen movies (she liked Alexander Payne’s “About Schmidt” with Jack Nicolson; when asked if she had seen it as a conversational ploy, she said, “Oh yeah, I see everything.”). Here’s the entirety of what we did record, via the Texas Politics Project YouTube account

Subscribe to the Texas Politics Project Email List

* indicates required