Latino America: How America's Most Dynamic Population is Poised to Transform the Politics of the Nation
Page 13
TABLE 7.1The National Popular Vote Share in NEP Exit Polls, by Race and Ethnicity, 2004–2012
Source: NEP exit polls, as tabulated by CNN.
Note: Numbers are percentages. The rows do not report votes for third parties and hence may total less than 100. The columns may not add to 100 owing to rounding error.
a. Owing to the particular characteristics of the NEP Latino sample in 2004, and as we have documented elsewhere, Latino Decisions, like most other observers, is deeply skeptical of this number. A better estimate of the Latino vote percentage for Republicans in 2004 is approximately 40%. See Barreto et al. (2005).
THE 2010 RESULTS IN CONTEXT
So, with the passage of SB 1070 in Arizona and the Senate’s consideration—and ultimate rejection—of the DREAM Act, how did Latino voters end up voting in 2010? Despite the description of that election as a dramatic GOP victory, the underlying dynamics of the election were very similar to elections before and since. That is, like all American elections, the local elections were won on the margins.
Table 7.1 illustrates the vote by group as reported in the National Election Pool (NEP) exit polls over the last decade of elections. In 2010 the GOP did marginally better among all groups than it did in 2006, largely owing to changes in the composition of the electorate: midterm elections turn out fewer voters of lower income and lower levels of education, resulting in a significant drop in minority turnout. In 2010 non-Hispanic whites constituted 78% of the electorate, compared with 74% two years earlier and 72% two years later. More importantly, those who tend to fall off in midterm years are disproportionately Democratic voters.
In 2010 a substantial majority of whites voted Republican, as they have in every election since 1964. For all other racial and ethnic groups, even with the decline in turnout disproportionately affecting Democratic voters, majorities voted Democratic.
HOW THE WEST WAS WON, 2010 EDITION
Despite the fact that the year started with significant disappointments and frustration with the Obama administration, the passage of SB 1070 by the Arizona GOP and Republican Senate unity in blocking the DREAM Act were sufficient to restore Latino enthusiasm for electoral participation. Latinos were approximately 8% of the electorate in 2010, the same as in 2006 when nationwide immigration marches generated substantial electoral enthusiasm. And though their support for Democrats was diminished compared with 2008—by 3%, the smallest decline in any demographic—in exit polls Democrats still outpolled Republicans among Latinos by nearly two to one. And that was in the exit polls that, as we have argued elsewhere, significantly underestimated the Latino Democratic vote.1 The Latino Decisions estimate, based on our 2010 election eve poll, was a 76% Democratic vote share in the two-party House vote.
So, despite national political trends and earlier disappointments, Latinos voted heavily Democratic in 2010, either two to one or three to one. But did they make a difference?
In four elections—the gubernatorial elections in California and Colorado and the senatorial elections in Nevada and Colorado—Latinos made a critical difference to the outcome, either in terms of actual votes cast on election day or in how the race took shape rhetorically, and usually in both ways.
This is not to say that Latinos did not matter elsewhere. In Illinois, Pat Quinn’s election as governor was a squeaker—he prevailed by 0.3%. Solid Latino turnout and an approximate 83% vote share for Quinn among Latinos contributed a net margin of 4.2%, which was far larger than his actual win. (Latinos were not enough to save Alexi Giannoulias, who lost his Senate race to Republican Mark Kirk). Also, Kamala Harris’s election as attorney general of California would not have been possible without an extremely strong Latino vote.
The Colorado Gubernatorial Race John Hickenlooper was elected governor of Colorado in 2010 with only 51.01% of the vote in a three-way race with GOP nominee Dan Maes and Congressman Tom Tancredo, who ran on the “Constitution Party” ticket. Tancredo, a former GOP member of Congress, has made anti-immigrant politics a hallmark of his political career, and it remains the raison d’être for his career. He bolted from the GOP in that cycle, ostensibly because he believed that neither primary candidate had the political strength to win the general election. Though Hickenlooper received more than 50% of the vote, including a net 6.3% from the state’s Latino electorate, the division of conservative forces no doubt played a significant role in his victory. Nevertheless, the third-party candidacy of the nation’s most outspoken opponent of undocumented immigrants, coupled with the Latino vote share, signaled a critical role for Latinos in the state’s politics.
The California Gubernatorial Race The race to replace California’s termed-out governor, Arnold Schwarzenegger, was unusual, to say the least. The Democratic nominee was previous two-term Democratic governor Jerry Brown, who had returned to political life to serve as mayor of Oakland and as attorney general before running for his old office. The state’s partisan evolution since the 1990s made this an uphill race for any Republican, despite the fact that Brown had shown increasing reluctance to fund-raise, campaign, and the like.2 The conventional wisdom was that the only GOP candidate who stood a chance would be someone like Schwarzenegger—a socially moderate, fiscally cautious candidate with little or no connection to the state party establishment.
Say, someone like Meg Whitman.
At the time, Whitman was the former CEO of eBay and widely respected in corporate and Silicon Valley circles. She had a history of public-spirited action, but not as an elected official. Her business acumen looked to be an ideal selling point in a state that was suffering (at that time) serious fiscal problems. Finally, and most importantly, she had a huge bankroll and a willingness to spend a bunch of it on the election.
In the end, Meg Whitman spent over $140 million of her own money in addition to a sizable sum raised elsewhere. And she lost by 1.3 million votes, 53.8% to 40.9%. How such an attractive candidate could lose by such a large margin deserves explanation. Moreover, Whitman lost to someone with almost no campaign infrastructure and no Spanish-language website.3 Despite the absence of a Spanish-language website, Latinos turned out in droves for Jerry Brown. Latino Decisions’ 2010 election eve poll estimated that Brown received 86% of the Latino vote, which, at 18% of the electorate, meant that Latinos contributed a net 13.1% of Brown’s overall total.
Without Latinos, the race would have been a virtual tie.
There were four reasons for the Latino enthusiasm for Brown. First, several independent expenditure groups ran a sophisticated messaging campaign designed to mobilize support for Brown, even though he had little campaign infrastructure to reinforce the message. Despite the campaign’s paltry efforts, there were Latino-targeted ads in English and Spanish running in most of the state’s media markets, as well as large-scale direct mail and get-out-the-vote phone campaigns.
Second, Meg Whitman got caught using different campaign messages with different audiences. Specifically, while running ads in Spanish saying that she had no interest in Arizona-style anti-immigrant legislation, she was giving interviews to conservative talk-radio and telling a somewhat different story. This was particularly true during the primary, when Whitman was working to put away more conservative primary rivals by touting her opposition to in-state tuition and other state benefits for the undocumented. The juxtaposition of these clearly mixed messages was called out by the Brown campaign and its surrogates as dos caras, or “two-faced.” The label stuck.
Third, any credibility that Whitman had on immigration was further eroded when she appointed Pete Wilson, the former Republican governor and architect of the Prop 187 anti-immigration initiative, as a co-chair of her election campaign. Though this came sixteen years after the passage of the dreaded initiative—one credited with reshaping California politics4—a huge share of Latino voters in the state held strongly negative associations with Pete Wilson. Figure 7.5 illustrates attitudes about Wilson among Latino voters. Even among eighteen- to thirty-year-old voters, who were between the ages o
f two and fourteen when Prop 187 was passed, 86% were somewhat or very concerned that Whitman had tapped Wilson.
Finally, the salience of the immigration issue was raised, not lowered, over the course of the campaign by the revelation that Whitman had an undocumented person performing domestic labor in her home. Exacerbating the situation, Whitman had fired this person in June 2009, on the eve of her campaign for governor. In the minds of the voters, Whitman was wrong twice. Her credibility on immigration was undermined by her alleged knowledge of having an undocumented worker in her employ. And her treatment of that employee was similarly found to be both hard-hearted and self-serving.
FIGURE 7.5California Latino Voters’ Perceptions of Former Governor Pete Wilson, September–October 2010
The Colorado Senate Race Like the gubernatorial race, the Colorado Senate race featured at least one candidate identified with the ideological, or “Tea Party,” wing of the Republican Party. Ken Buck, the GOP nominee in Colorado, held forcefully articulated views on immigration and immigrants. He made his name in state politics as the district attorney of Weld County, a role in which he masterminded what was then the largest immigration raid in US history—a 2006 raid of a beef processing plant in Greeley. A profile in The Nation described his interest in the immigration issue as “obsessed.”5 During the course of the campaign, he accused incumbent senator Michael Bennet of favoring “amnesty.”
But Colorado’s electorate was 17% Latino in 2010 (and almost 20% today). Using our 2010 election eve poll, Latino Decisions estimated that Senator Bennet received 81% of the Latino vote, meaning that Latinos contributed a net 6.2% to Bennet’s total on election day. Since the statewide margin was only 1.7% of the vote, a more even distribution of Latino votes would have meant an easy win for Buck. Senator Bennet owes his seat to Latino voters.
The Nevada Senate Race Harry Reid is pivotal to our story in two very important ways. First, as the Senate majority leader, it was he who brought the DREAM Act to a vote in September 2010. Second, as an incumbent seeking reelection, he faced one of the most explicitly racialized campaigns of the year, run by his challenger, Sharron Angle.
Angle, a former Republican member of the state legislature, ran an insurgent campaign against the presumed nominee, Sue Lowden, a former local TV news celebrity, and two others. Angle defeated Lowden by around fourteen percentage points, in some measure because Democrats and their allies had targeted Lowden (who they perceived as the bigger threat) with ads during the primary campaign, and in part as a consequence of the Tea Party emergence in 2010.
Angle is, to put it mildly, erratic in public. She’s prone to gaffes and appears to have fringe beliefs regarding 9/11, the Department of Education, Muslims, the United Nations, and other bêtes noires. But none of her views attracted as much attention as her views on immigration, which became the centerpiece of her advertising campaign.
In a widely decried ad—a version of which other GOP nominees ran in other states—Senator Reid’s support for the inclusion of undocumented persons in several federal benefits programs was illustrated with images of apparent gang members who Reid would help go to college, frightened and frustrated (white) Americans, and a classroom full of (white) American children who would apparently be prevented from speaking English if Reid was reelected.
To be sure, Reid was aided by substantial voter registration efforts in Nevada between the 2008 and 2010 elections by organizations like Mi Familia Vota and the Hispanic Institute, among others. But there is no question that Angle’s specifically racial ads had a significant effect on Latino mobilization and vote choice.
Almost every major poll predicted an Angle victory. But on election day, Reid defeated Angle by almost six percentage points. A Latino Decisions election eve poll estimated that 90% of Latino voters chose Harry Reid (a number we have since validated with precinct-level analysis). Without Latinos, or with an even distribution of Latino votes, Reid would have lost and Angle would have won.
FIGURE 7.6Latino Voter Attitudes on the Importance of the Immigration Issue to their Vote Choices, Election Eve 2010
Although overall the 2010 midterms were a pretty thorough defeat for Democrats, in numerous elections in 2010, and particularly in elections in the West, Latinos showed that they are a critical element in the Democratic coalition—without them, Democrats lose elections. Immigration was front and center in the minds of the voters we interviewed. As we report in Figure 7.6, 60% of Latino voters said that immigration was “very” important to their choice to vote and their choice of candidate, while another 23% said that it was “somewhat” important. Immigration politics affects the vote choices and mobilization of Latinos across national-origin groups and generations, and it has become even more important to Latinos as some states have tried to regulate immigration in harsh and racially suspect ways.
Chapter 8
A “DECISIVE VOTING BLOC” IN 2012
With Loren Collingwood, Justin Gross, and Francisco Pedraza
In 2012 the Latino vote made history.* For the first time ever, Latinos accounted for one in ten votes cast nationwide in the presidential election, and Obama recorded the highest ever vote total for any presidential candidate among Latinos, at 75%.1 Also for the first time ever, the Latino vote directly accounted for the margin of victory—simply put, without Latino votes, Obama would have lost the election to Romney (at least in the popular vote). Indeed, the day after the election dozens of newspaper headlines proclaimed 2012 the “Latino tide” and lamented the GOP’s undeniable “Latino problem.” As Eliseo Foley wrote on the Huffington Post site, “The margins are likely bigger than ever before, and bad news for the GOP. . . . ‘Republicans are going to have to have a real serious conversation with themselves,’ said Eliseo Medina, an immigration reform advocate and secretary-treasurer of the Service Employees International Union. ‘They need to repair their relationship with our community. . . . They can wave goodbye to us if they don’t get right with Latinos.’”2
TABLE 8.1Latino Contribution to National and State Margins for Obama, 2012
However, the performance of Latino voters in 2012 had not been guaranteed. Early in 2012, many journalists, campaign consultants, and scholars had questioned whether Obama would be able to win over Latinos. Would the struggling economy and the lack of progress on immigration reform result in millions of disaffected Latino voters?
It was not until the summer of 2012 that Obama solidified his image as a champion of immigrant rights, and at the same time Romney solidified his own image as out of touch with working-class families and, even worse, as anti-immigrant. From the summer to the fall, Obama stuck to a script of extensive, ethnic-based outreach to Latinos while Romney, in hopes of winning more conservative white votes, continued to oppose popular policies like the DREAM Act. The result, of course, was the worst showing ever for a Republican candidate among Latino voters.
Among Latino voters, Barack Obama outpaced Mitt Romney by a margin of 75% to 23% in the 2012 election—the highest rate of support ever among Latinos for any Democratic presidential candidate.3 While turnout declined nationally from 2008 to 2012 (by 2%), among Latinos there was a 28% increase in votes cast in 2012 (from 9.7 million to 12.5 million), and Obama further increased his vote share among Latinos in 2012 compared to 2008.4
However, this outcome was not a foregone conclusion: many theories circulated after 2009 suggesting that the Latino vote might be under-whelming in 2012.5 Given the high rate of Latino unemployment and the record number of immigrant deportations during Obama’s first administration, why did he do so well?6 Latinos’ historic party identification with the Democratic Party was strong evidence that Obama would win a majority of Latino votes.7 The other indicators, however, such as age, resources, and connections to politics, pointed toward lower turnout and less enthusiastic support for Obama.8 As late as September 2012, a common headline in the popular press was something along the lines of “Latinos’ Enthusiasm Gap Worries Dems,” and there was widespread concern that the
Latino vote seemed to be “fading.”9
In the end, post-election media accounts of the 2012 Latino vote suggested that Obama performed so well among Latino voters precisely because of their unique demographic characteristics: Latino voters are younger than average voters (younger voters tend to vote Democratic), they have lower-than-average incomes (historically, poorer voters side with Democrats), and, perhaps as a result, they tend to identify as Democrats.10 Others have suggested that Obama did so well among Latinos because he supported the DREAM Act and initiated an executive order (DACA, or Deferred Action for Childhood Arrivals) that authorized immigration officials to practice “prosecutorial discretion” toward undocumented Latino youth.11 Finally, some activist organizations have also suggested that Romney’s move to the right on immigration had a negative impact on his campaign among the Latino electorate.12
This chapter puts these accounts to the test. The 2012 Latino vote may be explained in large part by traditional vote-choice models, which include items such as partisanship, political ideology, gender, age, religion, presidential approval, views on the economy, and most important issues. These models have “worked” for fifty years, from The American Voter by Angus Campbell and his colleagues (1960) to The American Voter Revisited by Michael Lewis-Beck and his team of researchers more recently (2008).13 As we detailed the 2008 election in Chapter 6, however, we demonstrated that traditional models of voting don’t work quite as well for understanding Latino voting patterns. Further, as the electorate continues to diversify, scholars need to begin to ask how vote-choice models can be improved to better explain minority vote choice.