Run by Jim Henson, director of the Texas Politics Project, the department hosted a post-election roundtable November 7, featuring Henson, professor Daron Shaw, Richard Murray from the University of Houston, and Mark Jones from Rice University’s Baker Institute for Public Policy.
Henson and Josh Blank, the Texas Politics Project manager of polling and research, wrote the following analysis in the wake of the election:
The inevitable post-election finger-pointing among Democrats and chest-thumping among Republicans is unlikely to subside anytime soon given the margins of the GOP’s victories in Texas last week.
Amid all the noise, a deceptively simple central question underlies the results: Was the Republicans’ boat in Texas floated higher by the national tide that swamped Democrats almost everywhere else in the country, or was there something particular to Texas — partisan change, a particularly strong GOP campaign and/or candidates, an especially disastrous Democratic effort — that produced such large wins for Texas Republicans, and such favorable margins among important sub-groups?
A fair-eyed look at the numbers — set apart from the fault-finding and crowing of campaign advisers and consultants — suggests that 2014 was defined by relatively modest Republican success at increasing turnout coupled with Democratic failures that, while notable, don’t portend any large-scale shift in the underlying bases of Democratic support. While campaigns mattered, of course, Texas Democrats outperforming their 2010 effort would have been a far bigger surprise than their eventual 20-point losses — especially in light of Democratic troubles this year in places far more favorable to them than Texas, like Colorado, Virginia and even Massachusetts.
Low turnout combined with an electoral environment that favored Republicans defined the election in Texas, as it did most everywhere else. Turnout decreased across the state by about 4 points from 2010, according to early reports. A broad explanation for the Democrats’ decline in fortunes is based on the well-founded observation that low-turnout electorates tend to favor GOP candidates because of the unevenness with which different groups within the Democratic and Republican coalitions vote. Simply put, a decrease in turnout is likely to be concentrated in the Democratic Party, especially among the groups most likely to vote Democratic: young people, minorities and unmarried women. Another common structural observation is also based on years of political science research: Midterm elections always favor the party that doesn’t hold the presidency, and the sixth year of a president’s tenure is roundly known to be bad for his party.
Most post-game analyses of the 2014 Texas elections have understandably focused on the drubbing suffered by the Democrats and the GOP’s relative success among women and Hispanic voters. To recap, according to exit polling, Republican Greg Abbott won 54 percent of the female vote while Democrat Wendy Davis won 55 percent of the Hispanic vote. The female vote margin remained essentially unchanged from 2010, but the change in the Hispanic vote seems to represent a major backslide for Texas Democrats, who won that group by 23 points four years ago (compared with 9 points this year).
Texas Democrats’ poor performance is reflected in the numbers no matter how you slice them, and to such a broad extent that it’s fair to consider the extent to which Texas Democrats added some mass of their own to the gravity that pulled Democrats down nationwide. Democrats increased their 2010 vote count in only 14 of Texas’ 254 counties, and eight of those were counties where fewer than 11,000 votes were cast. (In Travis County, of their lone bright spots, Democrats increased their vote total by about 26,000 votes.) Compared with 2010, when Rick Perry beat Bill White, Democrats this year lost 1,080 votes per county on average and 274,000 votes overall. Most notably, they lost 76,000 votes in Harris County, 12,000 in Dallas and 3,000 in Bexar (their top three vote-getting counties in 2010). The number of Democratic voters in the counties with the top 10 vote counts in 2010 fell by an average of about 5,700 votes, meaning that on a percentage basis, the raw vote count in these counties was just on the negative side of stagnant. They also failed to increase their vote count in any of the top 10 counties that saw the most Hispanic growth between 2010 and 2014. Not exactly the stuff “destiny” is made of.
But what of the triumphalist accounts of the Republican victory? Looking at the same measures, Republicans saw an average increase in vote count in their top 10 counties of about 3.5 percent, or 1,882 votes per county. Their results were also dragged down by dismal turnout in Harris County, where the Republican vote count decreased by over 30,000. But their vote count increased noticeably in Republican strongholds like Tarrant (9 percent), Collin (15 percent) and Denton (11 percent) counties. More broadly, the GOP saw modest (206 votes on average) though widespread (149 counties) gains across all counties, picking up roughly 54,000 votes more than their 2010 totals. This added up to a good night for a hegemonic party with all the advantages, but still suggests that much of the work was done by plummeting Democratic turnout.
Falling turnout also provides context for the ongoing discussion about where Hispanic voters landed. To begin with, for all the drama surrounding whether Abbott would win more than 40 percent of the Hispanic vote, Republicans’ improved performance among the demographic — 38 percent voted for Perry in 2010, while 44 percent voted for Abbott — shouldn’t elicit too much surprise. Much of the pre-election polling suggested that the Hispanic share of the likely voter pool was not breaking strongly toward the Democrats. But any decisive reading of the result should be tempered by the large margin of error associated with the exit polling of subgroups, particularly Hispanics.
The vote count also encourages tempering any interpretation of these results as a decisive shift in the party allegiance of Hispanics. Of the top 10 counties with the most Hispanic growth between 2000 and 2010, Democrats lost roughly 10,000 votes from their 2010 haul — but Republicans also lost roughly 4,000. These numbers are more indicative of continued low Hispanic turnout than of any significant shift in partisan preferences among Hispanics, including one toward Republicans. The ambiguity of these results certainly agree with a large body of polling data suggesting that the attitudes of Hispanics as a group often put them between the two parties, depending on the issue set, rather than squarely in one camp.
The GOP’s 20-point victory margins on Nov. 4 are not the stuff of heroic epics (“how the Republicans triumphed!”) or of high tragedy (“oh, woeful be the intertwined fates of Battleground Texas and Wendy Davis!”). In fact, they’re not even mystery material. Republican candidates entered this election season with significant and deeply rooted advantages in partisanship, organization and resources, which they ably and predictably exploited. Democratic efforts to overcome these advantages faced long odds from the outset, and were likely hindered by strategies informed by an excess of optimism in the face of a grim national environment and grimmer fundamentals within the state.
All of this was relatively predictable — much less so, as it turns out, than the impact of these election results on the looming legislative session.