Skip to main content

View Diary: The Texas Trend - Red Not Blue? (39 comments)

Comment Preferences

  •  Great diary, lots of good stuff here! (5+ / 0-)

    Now I am going to proceed to disagree with most of it.

    Before I go on, you might want to check here for my overall take on TX voting trends.

    Now to take your points in order:

    Regarding your first historical note: Agree that Texas has been on a long-term trend, though I think it bottomed out with the Bush elections. Obama, as you note, did worse than Kerry in a PVI sense. But a) he did much better in an absolute sense, and if (like me) you think the country as a whole is trending Dem at the presidential level, it's possible for Texas to trend R from one election to the next in terms of PVI, yet trend D in terms of straight propensity to vote Democratic, at least in principle. Having said that, though, I think the trend from Kerry to Obama was essentially flat; whatever movement there was was marginal. Also b) Obama was a pretty poor candidate for Texas, just as he was for every other southern state that had less than about a 20% African American population. I think Clinton would have done better.

    Regarding your "complicating factors":

    1. These points are true, but the increased polarization of the white vote does not entail that the white vote hasn't become more heterogeneous. I wouldn't be surprised, in fact, if the declining Dem share of rural whites has been partially offset by an increasing performance among some urban whites - albeit not enough to entirely make up the difference.

    2. If the white Dem share is indeed bottoming out (about which: see below), then any increase in minority vote share will increase overall Dem performance. What's more, the minority vote going forward will be increasing at an accelerating rate. This is because the minority vote skews young, and with each passing year more of them age into voting eligibility.

    But there's another factor which enhances this effect, namely: the over-18 Hispanic population has much lower citizenship rates than the under-18s. For instance, according to this, the citizenship rate for over-18 Hispanics is 47% in Houston, whereas it's 87% for under-18s. Statewide, 92% of under-18 Hispanics are citizens.

    What this means is that a very large portion of the demographic change in Texas hasn't been factored in to the electorate - even more than you would think just by looking at the age demographics alone, or the citizenship rates alone. So, though a lot of this change hasn't happened yet, it is imminent, and will begin to accelerate especially as we approach the end of this decade.

    3. I think we are, at least, very close to rock bottom with whites. If you look at west Texas, the ancestrally Dem region between Wichita Falls and Abilene still gives somewhat non-embarrassing vote shares to Democrats. But the population of this region is trivial. Elsewhere in the ancestrally republican parts of west Texas, from the Panhandle to the Hill Country, Dems really can't go much lower: they get Alabama-style 10% of the white vote out there.

    As I mention in my diary linked at the top of this comment, I do think Dems have a bit farther to fall in East Texas, which I would define as everything east of I-35 and north of I-10, and outside of the Houston and DFW metros.

    But let's put that in context. One way to think of Texas political geography is as comprised of four regions: the Border counties (about 2.5 million people); West Texas (2.5 million people); East Texas (3 million people); and the Texas Triangle, i.e., Houston/DFW/Austin/San Antonio (about 18 million people). So I am conceding that there may be some loss of Dem performance among whites in East Texas, but that's such a small population compared to the cities. Harris County alone has 4 million people. And I think it's reasonable to expect the white populations in the cities to grow more Dem-friendly, given the ways the economies are evolving in those cities, at least quickly enough to compensate for the decline in East Texas.

    4. Is there any evidence at all that TX Hispanics have been growing more amenable to Republicans? If so, I haven't seen it. Certainly Bush made significant inroads, as he seemed to do with Hispanics across the country. But then they reverted right back to their historical norm of about 60-65% performance for Dems with Obama and Bill White and other statewide elections.

    It seems possible that there will be cases where, e.g., individual state house districts that are majority-Hispanic will trend R, just because the Hispanic population is very large and doesn't vote monolithically like the African American population. But no exit polls I've seen point towards a slide in Dem performance among Hispanics (nor do election results in heavily Hispanic counties).

    That is not to say that this won't change. It seems possible to me that Hispanics will drift slightly towards republicans over the course of the next decade as the population becomes increasingly integrated. On the other hand, it seems at least as likely that they'll move more solidly into the Democratic column, given the increasingly nativist, Tea Party-infused sensibilities of the current GOP. We'll just have to see what happens.

    5. Again, I'd like to see more evidence of this. From everything I've seen of actual election results, especially in the border region but also in urban Houston and elsewhere, Hispanics remain a solidly Democratic constituency, and the only things keeping the political effect of this muted are low CVAP and low turnout rates.

    6 (a). Democrats are relying less on African Americans - but only because their overall base of support is growing. In fact, the AA percentage of Texas population is actually getting larger, which is sort of remarkable considering that Hispanic growth ought to make that hard to do. So consider that the presidential electorate in TX in 2008 was 63% white and 13% black. By 2024 (again, according to my diary) this ought to be more like 50% white but still 13% black. One way to think of this is that the black vote will "cancel out" an increasing share of the white vote with each passing election.

    The growing black population in Texas is an important demographic variable that shouldn't be overlooked.

    (b). Yes, this is an institutional problem. But I would hope and expect that the national party would invest in Texas if and when it does start becoming competitive. (Though really they ought to be laying the groundwork for that now.)

    (c). White liberals. As I've alluded to above, I think there is room for growth here. For one thing, there are white liberals in Austin not "only because of UT and Texas State," but because of changes in the nature of the economy, especially an increase in high-tech. That's most true of Austin, but it's increasingly true of DFW too, and maybe Houston to a lesser extent. I'm not too familiar with Dallas, but Houston at least does have an Anglo Dem region of about 150-200,000 people which will probably continue to grow. I suspect something similar is true of Dallas.

    It's also true that Texas college towns are inordinately conservative. But Obama, at least, made some inroads in College Station and Lubbock, and if Dems can continue to increase performance among young voters then this is a very low floor from which we can grow.

    Also: I have heard Ruy Teixeira mention in passing that the suburbs in Houston and DFW are starting to shift a bit Demwards in terms of white people's voting habits. I haven't been able to track down any more extended comments from him about this, but it's an intriguing remark.

    (d). Yeah, maybe. But so far at least, none of these Republican voter ID laws have had really large effects, and the worst of them (including Texas' own) have been blocked in the courts. This isn't a problem to which Texas is especially prone, compared to other states.

    7. The bench may be thin now, but that's a short-term problem, and no one is really claiming Democrats will really be competitive until the '20s anyway (no one realistic, at any rate). You never know if the next mayor of Houston or Dallas or Fort Worth might be a wunderkind - or whether the next Wendy Davis might show up in the state senate, for that matter.

    But also, I hardly think being a Hispanic is disqualifying for statewide office in Texas. Ted Cruz's republican primary electorate would certainly beg to differ. And (though I'm having trouble finding old exit poll data right now) I recall Tony Sanchez getting about 30% of the white vote in 2002, and it's not like he had preternatural political skills.

    8. I don't assume any significant net effect from in-migration. But I do think it might help, and at any rate it certainly can't hurt. First of all, it's worth noting that in-migrators are themselves disproportionately minorities, and that is part of what's helping minority growth in the state (especially growth in the black population). But as for white in-migrators, so long as they vote less than 3-1 for republicans, their effect will be to increase the Dem share of the white vote. Personally I think it's reasonable to suppose they're close to the national average, and vote about 40% for Democrats - especially considering they're coming from states like California and Illinois.

    I've heard this argument before that the people moving to Texas are doing so (at least in part) because they're fleeing liberal policies in blue states - Orange County republican refugees or some such. I find this argument pretty implausible. For one thing, like I say, domestic in-migrators to TX are disproportionately minorities, so I very much doubt this scenario applies to them. But more to the point: does anyone actually do this - decide where to live based on the state-level political scene? People decide where to live based on where they can get a job. The low cost of living, and especially housing, in a place like Texas may be a factor. And, very much as a secondary concern perhaps, people might re-locate based on their feelings about the culture of a place. But I just find it implausible that large numbers of people are moving to Texas because the unions are weaker there.

    Conclusion: I just don't see Dem performance sliding any farther than it already has. 2016 might see a bit of slippage due to lower enthusiasm and vote share among blacks (a problem we may face all over the country). Or the increasing Hispanic vote share might cancel that out. Or Hillary might be able to outperform Obama among conservadems and run up the score with Hispanics (both reasonable possibilities) and accelerate the Dem trend.

    Extrapolating from my diary, I think a reasonable expectation is that the 2024 electorate will be 50% Anglo, 29% Hispanic, 14% black, and 7% other. That is with some fairly conservative assumptions. If the Dem gets 27% of the white vote, 63% of the hispanic vote, 90% of the black vote, and 60% of the other vote, the election will end up 51.4-48.6 for the republican. If we push these numbers to 2028 we might expect an electorate that is 44/34/14/8 (white/hispanic/black/other). Given the same vote share assumptions, the result would be a Dem victory by 50.7-49.3. It looks like you would still expect the Republican to win by about 10 points in 2028. I would like to know, then, where you think my numbers are off. Do you think Texas whites will be voting like Mississippi whites by then? Do you really think Hispanics are going to vote increasingly Republican? What would be your concrete prediction for the composition and vote shares of the 2028 electorate?

    And more generally: what made you turn so pessimistic? Not long ago you were expecting Democrats to take over the state house by 2024, which was more optimistic than I was. What changed?

    •  Great comment and great diary (0+ / 0-)

      I pretty much agree with everything Cachy said. Contrary to the polyanna's Texas probably won't be competitive until into the 2020s. Contrary to the (well written and reasoned) points made in this diary, Texas will necessarily be competitive by the end of the 2030s so long as Republicans do not actually succeed in reversing their decades-long record of poor performance among hispanics. Texas is a fascinating state.

      26, originally OK-1, currently NY-10. Former swingnut.

      by okiedem on Thu Sep 06, 2012 at 02:35:00 PM PDT

      [ Parent ]

    •  On Texas whites: (0+ / 0-)

      I think that Texas whites voting like Mississippi whites is more plausible than you think. I, mean, who in 1990 would have predicted Mississippi whites going 90% Republican less than two decades later? Who in 1990 would have predicted Texas whites going 74% Republican?

      What happens if you decrease the Democratic vote amongst whites to 20%, or even 15%? If it goes down to 20%, then Democrats end up with 47.6% of the vote in 2028. If it goes down to 15%, then Democrats end up with 45.2% of the vote in 2028 - and we're back where we started.

      by Inoljt on Thu Sep 06, 2012 at 03:09:03 PM PDT

      [ Parent ]

      •  Texas is vastly, vastly more urban (3+ / 0-)
        Recommended by:
        MattTX, KingofSpades, nimh

        and cosmopolitan than Mississippi. You think Democrats might win only 15-20% of the white vote in a state where three-quarters of the population lives in a top-35 metro? And half the population lives in a top-5 metro? Not a chance.

        As for this:

        I, mean, who in 1990 would have predicted Mississippi whites going 90% Republican less than two decades later? Who in 1990 would have predicted Texas whites going 74% Republican?
        Well, Clinton only got 22% of whites in MS in 1992, and 27% in TX, albeit with a lot of votes going to Perot. But it goes to show that the great realignment of southern ways to the republicans had largely been accomplished by the 90s.

        Or take the TX senate races in '94 and '96: the Dem candidate got 27 and 30 percent of whites in those races respectively. And that's just about where Dems have been in statewide races ever since.

      •  The difference between TX and MS Whites (1+ / 0-)
        Recommended by:

        MS Whites are basically all rural. MS does not have any really major cities. It has Jackson, but Jackson is not than Corpus Christi. Basically all white voters in MS are rural/small town white voters.

        What MS doesn't have is large Metropolises. There is simply no equivalent to Dallas/Houston/Austin. And that is why TX white voters will not vote like MS white voters.

        The thing is, rural/small town white voters in TX don't vote all that differently from rural/small town white voters in MS (maybe a couple points more Democratic, but not much more than that).

        That is simply something that MS doesn't have, because it does not really have urban/suburban white voters in the first place.

        So if white voters OVERALL in TX were to vote like white voters in TX, this would require the GOP to do just as well with white voters in places like Dallas and Houston as they already do with white voters in places like the Panhandle and East Texas.

        The problem with that is that it would require the GOP to win over genuine liberal Democratic base voters. We're not talking about rural Consevadems.

      •  TX is not MS (0+ / 0-)

        MS has no post-industrial, white-collar metro areas that tend to attract (or retain) white Dems. TX has Austin, which is nearly as big a containment area for white Dems as the triangle area of NC. Dallas and Houston also seem to be headed in that direction on a smaller scale. East Texas is a good comp for Mississippi. Texas as a whole is not.

        SSP poster. 43, new CA-6, -0.25/-3.90

        by sacman701 on Thu Sep 06, 2012 at 03:54:19 PM PDT

        [ Parent ]

      •  Some quick back of the envelope (1+ / 0-)
        Recommended by:


        If you give the Democrat 50% of the white vote in Travis County (Austin), and 33% in Harris, Dallas, Tarrant, and Bexar counties, then Dems would only need to win about 22% of the white vote in all other counties combined to hit 27% statewide, which is about what they've averaged in recent statewide elections.

        That's 22% in all other counties, including urban counties like El Paso, and big and blue-trending (if still red) suburban counties like Williamson, Collin, Denton, Fort Bend, etc. I don't see that number going much lower. (Consider that Obama got 29% of the white vote in Oklahoma.) On the other hand, it's not hard to imagine the numbers in the 5 big urban counties going higher.

        •  Obama White performance by County (4+ / 0-)
          Recommended by:
          Chachy, okiedem, Inoljt, MichaelNY

          I have looked at this a lot, and I'm pretty sure these are approximately the shares of the white vote that Obama got in 2008 in the counties you mention (ignoring 3rd party votes):

          Travis: 55% White voters for Obama
          Harris: 25% White voters for Obama
          Dallas: 35% White voters for Obama
          Tarrant: 25% White voters for Obama
          Bexar: 30% White voters for Obama

          Williamson: 35% White voters for Obama
          Collin: 25% White voters for Obama
          Denton: 25% White voters for Obama
          Fort Bend: 15% White voters for Obama

          And then there are places like Midland County, where Obama got (less than???!!!) 10% of the white vote...

          Also, I think Obama's statewide Hispanic vote share was really more like 70% rather than 65%, and I think his white vote share was a point or two lower than exit polls say.

          •  interesting (1+ / 0-)
            Recommended by:

            How did you work that out?

            Harris looks a little low to me, though not inconceivable I guess.

            •  Methodology & Harris County Details (6+ / 0-)

              I worked it out partly through actually running analysis to see what is statistically the best fit, but also cross checking that with common sense, the precinct results, Spanish Surname voter registration, and VAP percentages.

              In the case of Harris County, White voters are clearly more Dem in some areas than others. They are split something like 45D/55R within the inner loop (610), with some more D areas like Montrose and the Heights but also more R areas like River Oak/West University Place/. But that is counterbalanced by the suburbs/exurbs (the Northwest, The Northeast, and the Southeast). In those areas, whites generally probably voted about 20% Obama (and in some precincts less than that - it can't have been otherwise given that you can find a number of <20% Obama precincts). Overall I think it comes out to about 25% when you combine everything together. It's possible that white voters in Harris County might be a point or two more (or, frankly, less) Democratic than 25%, but I have a hard time seeing how the numbers would get as high as 30%.

              Overall in Harris County I would estimate that the electorate was approximately:

              55%-56% White: (voting 74%-76% McCain over 24%-26% Obama)
              20%-21% Black: (voting 98% Obama over 2% McCain)
              18% Hispanic [SSVR is 18.8% in Dave's Redistricting App]: (voting 67%-75% Obama over 33%-25% McCain)
              5% Asian: (voting 50%-70% Obama over 30%-50% McCain)
              1% Other: (voting 50%-70% Obama over 30%-50% McCain)

              If you do the math, you get something very close to Obama's actual 1.5% win.

              For Obama to have done better than about 25% among whites in Harris County, he would either have to have performed implausibly poorly among minorities or minority turnout would have to have been implausibly low.

              •  Wow great stuff! (1+ / 0-)
                Recommended by:

                26, originally OK-1, currently NY-10. Former swingnut.

                by okiedem on Sat Sep 08, 2012 at 12:26:48 PM PDT

                [ Parent ]

              •  I actually think your hispanic numbers are (1+ / 0-)
                Recommended by:

                probably too high. I'll try to look at this more later but from what I can tell hispanic turnout in Houston is very very low due to the fact that the population is disproportionately non-citizen.

                26, originally OK-1, currently NY-10. Former swingnut.

                by okiedem on Sat Sep 08, 2012 at 12:43:54 PM PDT

                [ Parent ]

              •  I think that your Asian numbers are too high (1+ / 0-)
                Recommended by:

                as well. Judging by the percent of the electorate they compose in exit polls (especially California) and turn-out rates amongst eligible voters (lower than Hispanics), I think that you generally divide the percent Asian overall by two to get the percent Asian in the electorate. So it would be 3% Asian, since Harris is 6% Asian.

                Anyways, fascinating stuff!


                by Inoljt on Sat Sep 08, 2012 at 03:27:08 PM PDT

                [ Parent ]

                •  Asian turnout (2+ / 0-)
                  Recommended by:
                  MichaelNY, Inoljt

                  That is a good point, and I probably didn't discount Asian turnout quite as much as I should have.

                  Asian VAP in Harris County is 6.7% though (not just 6%). The real % may more likely be 4% than the 5% that I assumed. 3% could be possible, but seems like it might be too low to me.

              •  yep that's how i calculate it (1+ / 0-)
                Recommended by:

                its best to go to precincts where a certain race is more than 75% of the population on DRA to see how they act when left to their own devices.

                RRH expat (known as AquarianLeft). Also known as freepcrusher on leip atlas forum

                by demographicarmageddon on Sat Sep 08, 2012 at 11:47:53 PM PDT

                [ Parent ]

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site