The Hunt for Blue November: Why Battleground Texas Needs White Women
SOURCE: CENTER FOR AMERICAN PROGRESS. *ESTIMATES BASED OFF TABULATION AND EXTRAPOLATION OF CHANGE IN ELEGIBLE VOTERS FOUND IN 2008 AND 2011 AMERICAN COMMUNITY SURVEY
Published: July 24, 2013
Teasing electoral trends from demographic data requires juggling dozens of variables culled from imperfect raw numbers. Strategizing around those trends amounts to calling out audibles from a political vacuum. Simulating results for 2016 is like predicting a horse race without knowing the horses or track conditions.
Most observers tend to respond by simplifying. Consider, for example, an April 11 article by BusinessWeek entitled, “Texas, Election Battleground: Democrats Aim to Mobilize the Hispanic Vote.” The author concludes by saying: “The bottom line: In 2012, 2.2 million Texas Hispanics who were eligible to vote didn’t cast a ballot. If half had voted for Obama, he would have won the state.”
Actually, Republican candidate Mitt Romney corralled his 38 Texas electoral college votes by a hefty margin of 1,261,719 ballots, according to Texas Secretary of State records. All other demographic variables being the same, half of 2.2 million voting for President Barack Obama would not, in fact, have overcome that deficit.
But more importantly, it is fallacious to assume that if all 2.2 million additional eligible Latinos had hit the voting booths during last November’s presidential contest, the entirety of them would have opted to reelect the incumbent Dem. Nationally, Latinos favored Obama 71 percent to 27 percent. If that same blue bias held true in Texas during a hypothetical full-count Hispanic turnout, Romney still would have cleared the popular tally in the Lone Star by almost 300,000 votes.
As an additional wrinkle, however, Texan Latinos hover a bit more to the right than they do nationally. Exit polls by the National Election Pool were not conducted in Texas in 2012, but in the previous presidential contest, Latinos here voted for Obama over then-contender John McCain by a 63-percent-to-35-percent margin. If the same support levels in 2008 repeated in 2012, Romney would have outstripped Obama by 646,000 votes, even with a full house of 2.2 million additional Latinos hitting the polling stations.
This effect is so profound that, if the 2008 proportion held true in 2012, in excess of 4.5 million additional Latino voters would have needed to cast ballots before tipping the scales toward the Democratic candidate. So it’s not half of eligible non-voting Latinos to cover the spread. It’s actually more than double. And it’s more than the total number of eligible Latino voters in Texas that year.
Fast forward through eight years of demographic change and the figures converge much closer to a tipping point by 2016. Utilizing 2008 Texas exit polls (not available for the state in 2012) crossed with U.S. Census Bureau voter turnout figures from 2012, and applying those ratios to a projection of the number of 2016 eligible voters by ethnicity as calculated by the Center for American Progress yields a Republican victory in in the next contest.
Even under an assumption that Battleground Texas successfully mobilizes target voters, elevating Hispanic and other minority turnout to levels achieved by whites last year (which is actually a slight decrease for African-Americans, the group posting the highest 2012 turnout rate), the state retains its crimson hue by a five-point 52-47 margin. However, given those same parameters, tweaking up Democratic support by white women from 28 percent (from 2008 exit polls) to, say, a lowly 35 percent, blues the result.